From bfe69aea09d46c9bb77ad9f9109448a08568b211 Mon Sep 17 00:00:00 2001 From: Dominik Maier Date: Sun, 29 May 2022 03:23:02 +0200 Subject: [PATCH] Format C/Cpp code in ./scripts/fmt_all.sh (#653) * format all (clang format 13) * added clang-format check * re-add missing newline * cargo doc was missing * more brackets * fixed fmt workflow * clang format * shellcheck * install clang-format-13 * update ubuntu for maximum clang-formattability * yml whitespaces * fmt * shellcheck only for .sh * oops path * ignored shellcheck warning --- .github/workflows/build_and_test.yml | 17 +- .../c_code_with_fork_executor/src/harness.c | 12 +- .../src/harness.c | 2 +- .../command_executor/src/test_command.c | 4 +- .../forkserver_executor/src/program.c | 5 +- fuzzers/forkserver_simple/src/program.c | 49 +- fuzzers/frida_libpng/harness.cc | 311 +- fuzzers/fuzzbench/fuzz.c | 3 +- fuzzers/fuzzbench_fork_qemu/libfuzzer_main.c | 4 +- fuzzers/fuzzbench_qemu/libfuzzer_main.c | 4 +- fuzzers/fuzzbench_text/fuzz.c | 3 +- fuzzers/fuzzbench_weighted/fuzz.c | 3 +- fuzzers/libfuzzer_libmozjpeg/harness.cc | 129 +- fuzzers/libfuzzer_libmozjpeg/hook_allocs.c | 37 +- fuzzers/libfuzzer_libpng/harness.cc | 104 +- .../libfuzzer_libpng_accounting/harness.cc | 104 +- fuzzers/libfuzzer_libpng_ctx/harness.cc | 95 +- fuzzers/libfuzzer_libpng_launcher/harness.cc | 95 +- fuzzers/libfuzzer_reachability/harness.cc | 95 +- fuzzers/libfuzzer_reachability/weak.c | 3 +- fuzzers/libfuzzer_stb_image/harness.c | 17 +- fuzzers/libfuzzer_stb_image/stb_image.h | 12782 ++++++++-------- .../fuzzer/harness.c | 23 +- .../fuzzer/harness_symcc.c | 12 +- .../fuzzer/stb_image.h | 12780 +++++++-------- fuzzers/libfuzzer_stb_image_sugar/harness.c | 23 +- fuzzers/libfuzzer_stb_image_sugar/stb_image.h | 12780 +++++++-------- fuzzers/python_qemu/fuzz.c | 9 +- fuzzers/qemu_launcher/harness.cc | 98 +- libafl_cc/src/afl-coverage-pass.cc | 57 +- libafl_cc/src/autotokens-pass.cc | 11 +- libafl_cc/src/cmplog-routines-pass.cc | 9 +- libafl_cc/src/common-llvm.h | 22 +- libafl_cc/src/coverage-accounting-pass.cc | 12 +- libafl_cc/src/no-link-rt.c | 41 +- libafl_frida/src/gettls.c | 6 +- libafl_qemu/libqasan/dlmalloc.c | 1595 +- libafl_qemu/libqasan/hooks.c | 101 +- libafl_qemu/libqasan/libqasan.c | 36 +- libafl_qemu/libqasan/libqasan.h | 42 +- libafl_qemu/libqasan/malloc.c | 40 +- libafl_qemu/libqasan/map_macro.h | 1 - libafl_qemu/libqasan/patch.c | 40 +- libafl_qemu/libqasan/qasan.h | 24 +- libafl_qemu/libqasan/string.c | 134 +- libafl_qemu/libqasan/uninstrument.c | 21 +- libafl_qemu/src/asan-giovese.c | 456 +- libafl_qemu/src/asan-giovese.h | 61 +- libafl_qemu/src/interval-tree/compiler.h | 14 +- .../src/interval-tree/interval_tree_generic.h | 287 +- libafl_qemu/src/interval-tree/rbtree.h | 75 +- libafl_qemu/src/interval-tree/rbtree.inl | 45 +- .../src/interval-tree/rbtree_augmented.h | 329 +- libafl_targets/src/cmplog.c | 41 +- libafl_targets/src/cmplog.h | 42 +- libafl_targets/src/common.h | 156 +- libafl_targets/src/sancov_cmp.c | 2 +- libafl_targets/src/value_profile.h | 34 +- scripts/afl-persistent-config | 4 +- scripts/autofix.sh | 2 +- scripts/clean_all.sh | 9 +- scripts/clippy.sh | 2 +- scripts/fmt_all.sh | 15 +- scripts/publish.sh | 38 +- scripts/shmem_limits_macos.sh | 1 + scripts/test_all_fuzzers.sh | 18 +- 66 files changed, 21797 insertions(+), 21629 deletions(-) diff --git a/.github/workflows/build_and_test.yml b/.github/workflows/build_and_test.yml index f05c3f8f70..a2e4479d03 100644 --- a/.github/workflows/build_and_test.yml +++ b/.github/workflows/build_and_test.yml @@ -41,11 +41,8 @@ jobs: - name: Test libafl_targets no_std run: cd libafl_targets && cargo test --no-default-features - - - ubuntu: - runs-on: ubuntu-latest + runs-on: ubuntu-22.04 steps: - uses: actions-rs/toolchain@v1 with: @@ -55,7 +52,7 @@ jobs: - name: set mold linker as default linker uses: rui314/setup-mold@v1 - name: Install deps - run: sudo apt-get install -y llvm llvm-dev clang ninja-build + run: sudo apt-get install -y llvm llvm-dev clang ninja-build clang-format-13 shellcheck - name: get clang version run: command -v llvm-config && clang -v - name: Install cargo-hack @@ -80,6 +77,10 @@ jobs: - uses: actions/checkout@v2 - name: Format run: cargo fmt -- --check + - name: Run clang-format style check for C/C++ programs. + run: clang-format-13 -n -Werror --style=file $(find . -type f \( -name '*.cpp' -o -iname '*.hpp' -o -name '*.cc' -o -name '*.cxx' -o -name '*.cc' -o -name '*.h' \) | grep -v '/target/' | grep -v 'libpng-1\.6\.37' | grep -v 'stb_image\.h' | grep -v 'dlmalloc\.c') + - name: run shellcheck + run: shellcheck ./scripts/*.sh - uses: actions/checkout@v2 - name: Build Docs run: cargo doc @@ -87,6 +88,7 @@ jobs: run: cargo +nightly test --doc --all-features - name: Run clippy run: ./scripts/clippy.sh + ubuntu-concolic: runs-on: ubuntu-latest steps: @@ -100,6 +102,7 @@ jobs: run: sudo ./libafl_concolic/test/smoke_test_ubuntu_deps.sh - name: Run smoke test run: ./libafl_concolic/test/smoke_test.sh + bindings: runs-on: ubuntu-latest steps: @@ -117,6 +120,7 @@ jobs: - uses: actions/checkout@v2 - name: Run a maturin build run: cd ./bindings/pylibafl && maturin build + fuzzers: env: CC: ccache clang # use ccache in default @@ -158,6 +162,7 @@ jobs: if: runner.os == 'macOS' # use bash v4 run: /usr/local/bin/bash ./scripts/test_all_fuzzers.sh - run: ccache --show-stats + nostd-build: runs-on: ubuntu-latest steps: @@ -177,12 +182,14 @@ jobs: run: cd ./libafl && cargo test --no-default-features - name: libafl armv6m-none-eabi (32 bit no_std) clippy run: cd ./libafl && cargo clippy --target thumbv6m-none-eabi --no-default-features + build-docker: runs-on: ubuntu-latest steps: - uses: actions/checkout@v2 - name: Build docker run: docker build -t libafl . + windows: runs-on: windows-latest steps: diff --git a/fuzzers/backtrace_baby_fuzzers/c_code_with_fork_executor/src/harness.c b/fuzzers/backtrace_baby_fuzzers/c_code_with_fork_executor/src/harness.c index 92f88fddc5..a9f2cec04b 100644 --- a/fuzzers/backtrace_baby_fuzzers/c_code_with_fork_executor/src/harness.c +++ b/fuzzers/backtrace_baby_fuzzers/c_code_with_fork_executor/src/harness.c @@ -6,13 +6,13 @@ #define MAX_INPUT_SIZE 3 #define SHMEM_COUNT 100 -int shmid; +int shmid; key_t key = 58974; int create_shmem_array() { shmid = shmget(key, SHMEM_COUNT * sizeof(uint8_t), IPC_CREAT | 0666); printf("created a shared memory segment with shmid=%d\n", shmid); - void *res = shmat(shmid, NULL, 0); + void *res = shmat(shmid, NULL, 0); uint8_t *array_ptr = (uint8_t *)res; for (int i = 0; i < SHMEM_COUNT; i++) { array_ptr[i] = 0; @@ -21,22 +21,20 @@ int create_shmem_array() { } int set_value(int i) { - void *res = shmat(shmid, NULL, 0); + void *res = shmat(shmid, NULL, 0); uint8_t *array_ptr = (uint8_t *)res; array_ptr[i] = 1; return 0; } uint8_t get_value(int i) { - void *res = shmat(shmid, NULL, 0); + void *res = shmat(shmid, NULL, 0); uint8_t *array_ptr = (uint8_t *)res; return array_ptr[i]; } int destroy_shmem(int id) { - if (-1 == shmctl(id, IPC_RMID, NULL)) { - return -1; - } + if (-1 == shmctl(id, IPC_RMID, NULL)) { return -1; } return 0; } diff --git a/fuzzers/backtrace_baby_fuzzers/c_code_with_inprocess_executor/src/harness.c b/fuzzers/backtrace_baby_fuzzers/c_code_with_inprocess_executor/src/harness.c index 0cfdb4b9a4..9be6379cba 100644 --- a/fuzzers/backtrace_baby_fuzzers/c_code_with_inprocess_executor/src/harness.c +++ b/fuzzers/backtrace_baby_fuzzers/c_code_with_inprocess_executor/src/harness.c @@ -8,7 +8,7 @@ #define MAP_SIZE 100 // uint8_t *array; -uint8_t array[MAP_SIZE]; +uint8_t array[MAP_SIZE]; uint8_t *array_ptr = &array; int init() { diff --git a/fuzzers/backtrace_baby_fuzzers/command_executor/src/test_command.c b/fuzzers/backtrace_baby_fuzzers/command_executor/src/test_command.c index 2fc58b17cc..5d9617a068 100644 --- a/fuzzers/backtrace_baby_fuzzers/command_executor/src/test_command.c +++ b/fuzzers/backtrace_baby_fuzzers/command_executor/src/test_command.c @@ -35,9 +35,7 @@ uint8_t get_value(int i) { } int destroy_shmem() { - if (-1 == shmctl(shmid, IPC_RMID, NULL)) { - return -1; - } + if (-1 == shmctl(shmid, IPC_RMID, NULL)) { return -1; } return 0; } diff --git a/fuzzers/backtrace_baby_fuzzers/forkserver_executor/src/program.c b/fuzzers/backtrace_baby_fuzzers/forkserver_executor/src/program.c index 9fcf8d7e6f..86a9459bd6 100644 --- a/fuzzers/backtrace_baby_fuzzers/forkserver_executor/src/program.c +++ b/fuzzers/backtrace_baby_fuzzers/forkserver_executor/src/program.c @@ -5,11 +5,8 @@ __AFL_FUZZ_INIT(); int main(int argc, char **argv) { - FILE *file = stdin; - if (argc > 1) { - file = fopen(argv[1], "rb"); - } + if (argc > 1) { file = fopen(argv[1], "rb"); } // The following three lines are for normal fuzzing. /* diff --git a/fuzzers/forkserver_simple/src/program.c b/fuzzers/forkserver_simple/src/program.c index de3bec4124..5987751523 100644 --- a/fuzzers/forkserver_simple/src/program.c +++ b/fuzzers/forkserver_simple/src/program.c @@ -6,39 +6,30 @@ __AFL_FUZZ_INIT(); void vuln(char *buf) { - if(strcmp(buf, "vuln") == 0) { - abort(); - } + if (strcmp(buf, "vuln") == 0) { abort(); } } +int main(int argc, char **argv) { + FILE *file = stdin; + if (argc > 1) { file = fopen(argv[1], "rb"); } -int main(int argc, char **argv){ + // The following three lines are for normal fuzzing. + /* + char buf[16]; + char* p = fgets(buf, 16, file); + buf[15] = 0; + */ - FILE* file = stdin; - if (argc > 1) { - file = fopen(argv[1], "rb"); + // The following line is also needed for shared memory testcase fuzzing + unsigned char *buf = __AFL_FUZZ_TESTCASE_BUF; + + printf("input: %s\n", buf); + if (buf[0] == 'b') { + if (buf[1] == 'a') { + if (buf[2] == 'd') { abort(); } } + } + vuln(buf); - // The following three lines are for normal fuzzing. - /* - char buf[16]; - char* p = fgets(buf, 16, file); - buf[15] = 0; - */ - - // The following line is also needed for shared memory testcase fuzzing - unsigned char *buf = __AFL_FUZZ_TESTCASE_BUF; - - - printf("input: %s\n", buf); - if(buf[0] == 'b'){ - if(buf[1] == 'a'){ - if(buf[2] == 'd'){ - abort(); - } - } - } - vuln(buf); - - return 0; + return 0; } diff --git a/fuzzers/frida_libpng/harness.cc b/fuzzers/frida_libpng/harness.cc index 94327e1561..b7fae8e3f1 100644 --- a/fuzzers/frida_libpng/harness.cc +++ b/fuzzers/frida_libpng/harness.cc @@ -24,55 +24,54 @@ #define PNG_INTERNAL #include "png.h" -#define PNG_CLEANUP \ - if(png_handler.png_ptr) \ - { \ - if (png_handler.row_ptr) \ - png_free(png_handler.png_ptr, png_handler.row_ptr); \ - if (png_handler.end_info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - &png_handler.end_info_ptr); \ - else if (png_handler.info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - nullptr); \ - else \ - png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ - png_handler.png_ptr = nullptr; \ - png_handler.row_ptr = nullptr; \ - png_handler.info_ptr = nullptr; \ - png_handler.end_info_ptr = nullptr; \ +#define PNG_CLEANUP \ + if (png_handler.png_ptr) { \ + if (png_handler.row_ptr) { \ + png_free(png_handler.png_ptr, png_handler.row_ptr); \ + } \ + if (png_handler.end_info_ptr) { \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + &png_handler.end_info_ptr); \ + } else if (png_handler.info_ptr) { \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + nullptr); \ + } else { \ + png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ + } \ + png_handler.png_ptr = nullptr; \ + png_handler.row_ptr = nullptr; \ + png_handler.info_ptr = nullptr; \ + png_handler.end_info_ptr = nullptr; \ } struct BufState { - const uint8_t* data; - size_t bytes_left; + const uint8_t *data; + size_t bytes_left; }; struct PngObjectHandler { - png_infop info_ptr = nullptr; + png_infop info_ptr = nullptr; png_structp png_ptr = nullptr; - png_infop end_info_ptr = nullptr; - png_voidp row_ptr = nullptr; - BufState* buf_state = nullptr; + png_infop end_info_ptr = nullptr; + png_voidp row_ptr = nullptr; + BufState *buf_state = nullptr; ~PngObjectHandler() { - if (row_ptr) - png_free(png_ptr, row_ptr); - if (end_info_ptr) + if (row_ptr) { png_free(png_ptr, row_ptr); } + if (end_info_ptr) { png_destroy_read_struct(&png_ptr, &info_ptr, &end_info_ptr); - else if (info_ptr) + } else if (info_ptr) { png_destroy_read_struct(&png_ptr, &info_ptr, nullptr); - else + } else { png_destroy_read_struct(&png_ptr, nullptr, nullptr); + } delete buf_state; } }; void user_read_data(png_structp png_ptr, png_bytep data, size_t length) { - BufState* buf_state = static_cast(png_get_io_ptr(png_ptr)); - if (length > buf_state->bytes_left) { - png_error(png_ptr, "read error"); - } + BufState *buf_state = static_cast(png_get_io_ptr(png_ptr)); + if (length > buf_state->bytes_left) { png_error(png_ptr, "read error"); } memcpy(data, buf_state->data, length); buf_state->bytes_left -= length; buf_state->data += length; @@ -84,163 +83,153 @@ extern "C" int afl_libfuzzer_init() { return 0; } -static char * allocation = NULL; -__attribute__((noinline)) -void func3( char * alloc) { - //printf("func3\n"); - #ifdef _WIN32 +static char *allocation = NULL; + +__attribute__((noinline)) void func3(char *alloc) { +// printf("func3\n"); +#ifdef _WIN32 if (rand() == 0) { alloc[0x1ff] = 0xde; printf("alloc[0x200]: %d\n", alloc[0x200]); } - #else +#else if (random() == 0) { alloc[0x1ff] = 0xde; printf("alloc[0x200]: %d\n", alloc[0x200]); } - #endif +#endif } -__attribute__((noinline)) -void func2() { - allocation = (char*)malloc(0xff); - //printf("func2\n"); +__attribute__((noinline)) void func2() { + allocation = (char *)malloc(0xff); + // printf("func2\n"); func3(allocation); } -__attribute__((noinline)) -void func1() { - //printf("func1\n"); +__attribute__((noinline)) void func1() { + // printf("func1\n"); func2(); } - // Export this symbol #ifdef _WIN32 -# define HARNESS_EXPORTS __declspec(dllexport) + #define HARNESS_EXPORTS __declspec(dllexport) #else -# define HARNESS_EXPORTS + #define HARNESS_EXPORTS #endif // Entry point for LibFuzzer. // Roughly follows the libpng book example: // http://www.libpng.org/pub/png/book/chapter13.html -HARNESS_EXPORTS extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - - if (size >= 8 && *(uint64_t*)data == 0xABCDEFAA8F1324AA){ - abort(); - } - if (size < kPngHeaderSize) { - return 0; - } +HARNESS_EXPORTS extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, + size_t size) { + if (size >= 8 && *(uint64_t *)data == 0xABCDEFAA8F1324AA) { abort(); } + if (size < kPngHeaderSize) { return 0; } #ifdef TEST_ASAN - func1(); + func1(); #endif - std::vector v(data, data + size); - if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { - // not a PNG. - return 0; - } + std::vector v(data, data + size); + if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { + // not a PNG. + return 0; + } - PngObjectHandler png_handler; - png_handler.png_ptr = nullptr; - png_handler.row_ptr = nullptr; - png_handler.info_ptr = nullptr; - png_handler.end_info_ptr = nullptr; + PngObjectHandler png_handler; + png_handler.png_ptr = nullptr; + png_handler.row_ptr = nullptr; + png_handler.info_ptr = nullptr; + png_handler.end_info_ptr = nullptr; - png_handler.png_ptr = png_create_read_struct - (PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); - if (!png_handler.png_ptr) { - return 0; - } - - png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); - if (!png_handler.info_ptr) { - PNG_CLEANUP - return 0; - } - - png_handler.end_info_ptr = png_create_info_struct(png_handler.png_ptr); - if (!png_handler.end_info_ptr) { - PNG_CLEANUP - return 0; - } - - png_set_crc_action(png_handler.png_ptr, PNG_CRC_QUIET_USE, PNG_CRC_QUIET_USE); -#ifdef PNG_IGNORE_ADLER32 - png_set_option(png_handler.png_ptr, PNG_IGNORE_ADLER32, PNG_OPTION_ON); -#endif - - // Setting up reading from buffer. - png_handler.buf_state = new BufState(); - png_handler.buf_state->data = data + kPngHeaderSize; - png_handler.buf_state->bytes_left = size - kPngHeaderSize; - png_set_read_fn(png_handler.png_ptr, png_handler.buf_state, user_read_data); - png_set_sig_bytes(png_handler.png_ptr, kPngHeaderSize); - - if (setjmp(png_jmpbuf(png_handler.png_ptr))) { - PNG_CLEANUP - return 0; - } - - // Reading. - png_read_info(png_handler.png_ptr, png_handler.info_ptr); - - // reset error handler to put png_deleter into scope. - if (setjmp(png_jmpbuf(png_handler.png_ptr))) { - PNG_CLEANUP - return 0; - } - - png_uint_32 width, height; - int bit_depth, color_type, interlace_type, compression_type; - int filter_type; - - if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, - &height, &bit_depth, &color_type, &interlace_type, - &compression_type, &filter_type)) { - PNG_CLEANUP - return 0; - } - - // This is going to be too slow. - if (width && height > 100000000 / width) { - PNG_CLEANUP -#ifdef HAS_DUMMY_CRASH -#ifdef __aarch64__ - asm volatile (".word 0xf7f0a000\n"); -#else - asm("ud2"); -#endif -#endif - return 0; - } - - // Set several transforms that browsers typically use: - png_set_gray_to_rgb(png_handler.png_ptr); - png_set_expand(png_handler.png_ptr); - png_set_packing(png_handler.png_ptr); - png_set_scale_16(png_handler.png_ptr); - png_set_tRNS_to_alpha(png_handler.png_ptr); - - int passes = png_set_interlace_handling(png_handler.png_ptr); - - png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); - - png_handler.row_ptr = png_malloc( - png_handler.png_ptr, png_get_rowbytes(png_handler.png_ptr, - png_handler.info_ptr)); - - for (int pass = 0; pass < passes; ++pass) { - for (png_uint_32 y = 0; y < height; ++y) { - png_read_row(png_handler.png_ptr, - static_cast(png_handler.row_ptr), nullptr); - } - } - - png_read_end(png_handler.png_ptr, png_handler.end_info_ptr); + png_handler.png_ptr = + png_create_read_struct(PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); + if (!png_handler.png_ptr) { return 0; } + png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); + if (!png_handler.info_ptr) { PNG_CLEANUP - return 0; -} + return 0; + } + png_handler.end_info_ptr = png_create_info_struct(png_handler.png_ptr); + if (!png_handler.end_info_ptr) { + PNG_CLEANUP + return 0; + } + + png_set_crc_action(png_handler.png_ptr, PNG_CRC_QUIET_USE, PNG_CRC_QUIET_USE); +#ifdef PNG_IGNORE_ADLER32 + png_set_option(png_handler.png_ptr, PNG_IGNORE_ADLER32, PNG_OPTION_ON); +#endif + + // Setting up reading from buffer. + png_handler.buf_state = new BufState(); + png_handler.buf_state->data = data + kPngHeaderSize; + png_handler.buf_state->bytes_left = size - kPngHeaderSize; + png_set_read_fn(png_handler.png_ptr, png_handler.buf_state, user_read_data); + png_set_sig_bytes(png_handler.png_ptr, kPngHeaderSize); + + if (setjmp(png_jmpbuf(png_handler.png_ptr))) { + PNG_CLEANUP + return 0; + } + + // Reading. + png_read_info(png_handler.png_ptr, png_handler.info_ptr); + + // reset error handler to put png_deleter into scope. + if (setjmp(png_jmpbuf(png_handler.png_ptr))) { + PNG_CLEANUP + return 0; + } + + png_uint_32 width, height; + int bit_depth, color_type, interlace_type, compression_type; + int filter_type; + + if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, &height, + &bit_depth, &color_type, &interlace_type, &compression_type, + &filter_type)) { + PNG_CLEANUP + return 0; + } + + // This is going to be too slow. + if (width && height > 100000000 / width) { + PNG_CLEANUP +#ifdef HAS_DUMMY_CRASH + #ifdef __aarch64__ + asm volatile(".word 0xf7f0a000\n"); + #else + asm("ud2"); + #endif +#endif + return 0; + } + + // Set several transforms that browsers typically use: + png_set_gray_to_rgb(png_handler.png_ptr); + png_set_expand(png_handler.png_ptr); + png_set_packing(png_handler.png_ptr); + png_set_scale_16(png_handler.png_ptr); + png_set_tRNS_to_alpha(png_handler.png_ptr); + + int passes = png_set_interlace_handling(png_handler.png_ptr); + + png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); + + png_handler.row_ptr = + png_malloc(png_handler.png_ptr, + png_get_rowbytes(png_handler.png_ptr, png_handler.info_ptr)); + + for (int pass = 0; pass < passes; ++pass) { + for (png_uint_32 y = 0; y < height; ++y) { + png_read_row(png_handler.png_ptr, + static_cast(png_handler.row_ptr), nullptr); + } + } + + png_read_end(png_handler.png_ptr, png_handler.end_info_ptr); + + PNG_CLEANUP + return 0; +} diff --git a/fuzzers/fuzzbench/fuzz.c b/fuzzers/fuzzbench/fuzz.c index 7eea9f3b30..56b10527b2 100644 --- a/fuzzers/fuzzbench/fuzz.c +++ b/fuzzers/fuzzbench/fuzz.c @@ -2,8 +2,7 @@ #include int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size) { - if (Size >= 8 && *(uint32_t*)Data == 0xaabbccdd) - abort(); + if (Size >= 8 && *(uint32_t *)Data == 0xaabbccdd) { abort(); } } /* diff --git a/fuzzers/fuzzbench_fork_qemu/libfuzzer_main.c b/fuzzers/fuzzbench_fork_qemu/libfuzzer_main.c index a834699f30..e785cc170e 100644 --- a/fuzzers/fuzzbench_fork_qemu/libfuzzer_main.c +++ b/fuzzers/fuzzbench_fork_qemu/libfuzzer_main.c @@ -4,8 +4,6 @@ int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size); int main() { - - char buf [10] = {0}; + char buf[10] = {0}; LLVMFuzzerTestOneInput(buf, 10); - } diff --git a/fuzzers/fuzzbench_qemu/libfuzzer_main.c b/fuzzers/fuzzbench_qemu/libfuzzer_main.c index a834699f30..e785cc170e 100644 --- a/fuzzers/fuzzbench_qemu/libfuzzer_main.c +++ b/fuzzers/fuzzbench_qemu/libfuzzer_main.c @@ -4,8 +4,6 @@ int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size); int main() { - - char buf [10] = {0}; + char buf[10] = {0}; LLVMFuzzerTestOneInput(buf, 10); - } diff --git a/fuzzers/fuzzbench_text/fuzz.c b/fuzzers/fuzzbench_text/fuzz.c index 7eea9f3b30..56b10527b2 100644 --- a/fuzzers/fuzzbench_text/fuzz.c +++ b/fuzzers/fuzzbench_text/fuzz.c @@ -2,8 +2,7 @@ #include int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size) { - if (Size >= 8 && *(uint32_t*)Data == 0xaabbccdd) - abort(); + if (Size >= 8 && *(uint32_t *)Data == 0xaabbccdd) { abort(); } } /* diff --git a/fuzzers/fuzzbench_weighted/fuzz.c b/fuzzers/fuzzbench_weighted/fuzz.c index 7eea9f3b30..56b10527b2 100644 --- a/fuzzers/fuzzbench_weighted/fuzz.c +++ b/fuzzers/fuzzbench_weighted/fuzz.c @@ -2,8 +2,7 @@ #include int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size) { - if (Size >= 8 && *(uint32_t*)Data == 0xaabbccdd) - abort(); + if (Size >= 8 && *(uint32_t *)Data == 0xaabbccdd) { abort(); } } /* diff --git a/fuzzers/libfuzzer_libmozjpeg/harness.cc b/fuzzers/libfuzzer_libmozjpeg/harness.cc index 8ce7b9c1b3..fb48ca1d42 100644 --- a/fuzzers/libfuzzer_libmozjpeg/harness.cc +++ b/fuzzers/libfuzzer_libmozjpeg/harness.cc @@ -3,11 +3,10 @@ #include #include - struct my_error_mgr { - struct jpeg_error_mgr pub; /* "public" fields */ + struct jpeg_error_mgr pub; /* "public" fields */ - jmp_buf setjmp_buffer; /* for return to caller */ + jmp_buf setjmp_buffer; /* for return to caller */ }; typedef struct my_error_mgr *my_error_ptr; @@ -16,77 +15,67 @@ typedef struct my_error_mgr *my_error_ptr; * Here's the routine that will replace the standard error_exit method: */ - METHODDEF(void) -my_error_exit(j_common_ptr cinfo) -{ - /* cinfo->err really points to a my_error_mgr struct, so coerce pointer */ - my_error_ptr myerr = (my_error_ptr)cinfo->err; +METHODDEF(void) +my_error_exit(j_common_ptr cinfo) { + /* cinfo->err really points to a my_error_mgr struct, so coerce pointer */ + my_error_ptr myerr = (my_error_ptr)cinfo->err; - /* Always display the message. */ - /* We could postpone this until after returning, if we chose. */ - (*cinfo->err->output_message) (cinfo); + /* Always display the message. */ + /* We could postpone this until after returning, if we chose. */ + (*cinfo->err->output_message)(cinfo); - /* Return control to the setjmp point */ - longjmp(myerr->setjmp_buffer, 1); + /* Return control to the setjmp point */ + longjmp(myerr->setjmp_buffer, 1); } - - - -int do_read_JPEG_file(struct jpeg_decompress_struct *cinfo, const uint8_t *input, size_t len) -{ - struct my_error_mgr jerr; - /* More stuff */ - JSAMPARRAY buffer; /* Output row buffer */ - int row_stride; /* physical row width in output buffer */ - /* Step 1: allocate and initialize JPEG decompression object */ - /* We set up the normal JPEG error routines, then override error_exit. */ - cinfo->err = jpeg_std_error(&jerr.pub); - jerr.pub.error_exit = my_error_exit; - /* Establish the setjmp return context for my_error_exit to use. */ - if (setjmp(jerr.setjmp_buffer)) { - jpeg_destroy_decompress(cinfo); - return 0; - } - /* Now we can initialize the JPEG decompression object. */ - jpeg_create_decompress(cinfo); - /* Step 2: specify data source (eg, a file) */ - jpeg_mem_src(cinfo,input, len ); - /* Step 3: read file parameters with jpeg_read_header() */ - (void)jpeg_read_header(cinfo, TRUE); - /* Step 4: set parameters for decompression */ - /* In this example, we don't need to change any of the defaults set by - * jpeg_read_header(), so we do nothing here. - */ - /* Step 5: Start decompressor */ - (void)jpeg_start_decompress(cinfo); - /* JSAMPLEs per row in output buffer */ - row_stride = cinfo->output_width * cinfo->output_components; - /* Make a one-row-high sample array that will go away when done with image */ - buffer = (*cinfo->mem->alloc_sarray) - ((j_common_ptr)cinfo, JPOOL_IMAGE, row_stride, 1); - /* Step 6: while (scan lines remain to be read) */ - /* jpeg_read_scanlines(...); */ - while (cinfo->output_scanline < cinfo->output_height) { - (void)jpeg_read_scanlines(cinfo, buffer, 1); - /* Assume put_scanline_someplace wants a pointer and sample count. */ - //put_scanline_someplace(buffer[0], row_stride); - } - /* Step 7: Finish decompression */ - (void)jpeg_finish_decompress(cinfo); - /* Step 8: Release JPEG decompression object */ - //jpeg_destroy_decompress(cinfo); - return 1; -} - - - -extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - - - struct jpeg_decompress_struct cinfo; - do_read_JPEG_file(&cinfo,data,size); +int do_read_JPEG_file(struct jpeg_decompress_struct *cinfo, + const uint8_t *input, size_t len) { + struct my_error_mgr jerr; + /* More stuff */ + JSAMPARRAY buffer; /* Output row buffer */ + int row_stride; /* physical row width in output buffer */ + /* Step 1: allocate and initialize JPEG decompression object */ + /* We set up the normal JPEG error routines, then override error_exit. */ + cinfo->err = jpeg_std_error(&jerr.pub); + jerr.pub.error_exit = my_error_exit; + /* Establish the setjmp return context for my_error_exit to use. */ + if (setjmp(jerr.setjmp_buffer)) { + jpeg_destroy_decompress(cinfo); return 0; - + } + /* Now we can initialize the JPEG decompression object. */ + jpeg_create_decompress(cinfo); + /* Step 2: specify data source (eg, a file) */ + jpeg_mem_src(cinfo, input, len); + /* Step 3: read file parameters with jpeg_read_header() */ + (void)jpeg_read_header(cinfo, TRUE); + /* Step 4: set parameters for decompression */ + /* In this example, we don't need to change any of the defaults set by + * jpeg_read_header(), so we do nothing here. + */ + /* Step 5: Start decompressor */ + (void)jpeg_start_decompress(cinfo); + /* JSAMPLEs per row in output buffer */ + row_stride = cinfo->output_width * cinfo->output_components; + /* Make a one-row-high sample array that will go away when done with image */ + buffer = (*cinfo->mem->alloc_sarray)((j_common_ptr)cinfo, JPOOL_IMAGE, + row_stride, 1); + /* Step 6: while (scan lines remain to be read) */ + /* jpeg_read_scanlines(...); */ + while (cinfo->output_scanline < cinfo->output_height) { + (void)jpeg_read_scanlines(cinfo, buffer, 1); + /* Assume put_scanline_someplace wants a pointer and sample count. */ + // put_scanline_someplace(buffer[0], row_stride); + } + /* Step 7: Finish decompression */ + (void)jpeg_finish_decompress(cinfo); + /* Step 8: Release JPEG decompression object */ + // jpeg_destroy_decompress(cinfo); + return 1; } +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + struct jpeg_decompress_struct cinfo; + do_read_JPEG_file(&cinfo, data, size); + return 0; +} diff --git a/fuzzers/libfuzzer_libmozjpeg/hook_allocs.c b/fuzzers/libfuzzer_libmozjpeg/hook_allocs.c index 62ab077570..1399261c65 100644 --- a/fuzzers/libfuzzer_libmozjpeg/hook_allocs.c +++ b/fuzzers/libfuzzer_libmozjpeg/hook_allocs.c @@ -2,32 +2,30 @@ #include #include -#define MAP_SIZE (16*1024) +#define MAP_SIZE (16 * 1024) #ifdef _WIN32 -#define posix_memalign(p, a, s) (((*(p)) = _aligned_malloc((s), (a))), *(p) ?0 :errno) -#define RETADDR (uintptr_t)_ReturnAddress() + #define posix_memalign(p, a, s) \ + (((*(p)) = _aligned_malloc((s), (a))), *(p) ? 0 : errno) + #define RETADDR (uintptr_t) _ReturnAddress() #else -#define RETADDR (uintptr_t)__builtin_return_address(0) + #define RETADDR (uintptr_t) __builtin_return_address(0) #endif #ifdef __GNUC__ -#define MAX(a, b) \ - ({ \ - \ - __typeof__(a) _a = (a); \ - __typeof__(b) _b = (b); \ - _a > _b ? _a : _b; \ - \ - }) + #define MAX(a, b) \ + ({ \ + __typeof__(a) _a = (a); \ + __typeof__(b) _b = (b); \ + _a > _b ? _a : _b; \ + }) #else -#define MAX(a, b) (((a) > (b)) ? (a) : (b)) + #define MAX(a, b) (((a) > (b)) ? (a) : (b)) #endif size_t libafl_alloc_map[MAP_SIZE]; void *malloc(size_t size) { - uintptr_t k = RETADDR; k = (k >> 4) ^ (k << 8); k &= MAP_SIZE - 1; @@ -37,15 +35,11 @@ void *malloc(size_t size) { // Hence, even realloc(NULL, size) would loop in an optimized build. // We fall back to a stricter allocation function. Fingers crossed. void *ret = NULL; - if (posix_memalign(&ret, 1<<6, size) != 0) { - return NULL; - } + if (posix_memalign(&ret, 1 << 6, size) != 0) { return NULL; } return ret; - } void *calloc(size_t nmemb, size_t size) { - size *= nmemb; uintptr_t k = RETADDR; @@ -54,9 +48,6 @@ void *calloc(size_t nmemb, size_t size) { libafl_alloc_map[k] = MAX(libafl_alloc_map[k], size); void *ret = NULL; - if (posix_memalign(&ret, 1<<6, size) != 0) { - return NULL; - }; + if (posix_memalign(&ret, 1 << 6, size) != 0) { return NULL; }; return ret; - } diff --git a/fuzzers/libfuzzer_libpng/harness.cc b/fuzzers/libfuzzer_libpng/harness.cc index 65faff685d..5c36517376 100644 --- a/fuzzers/libfuzzer_libpng/harness.cc +++ b/fuzzers/libfuzzer_libpng/harness.cc @@ -23,55 +23,54 @@ #define PNG_INTERNAL #include "png.h" -#define PNG_CLEANUP \ - if(png_handler.png_ptr) \ - { \ - if (png_handler.row_ptr) \ - png_free(png_handler.png_ptr, png_handler.row_ptr); \ - if (png_handler.end_info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - &png_handler.end_info_ptr); \ - else if (png_handler.info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - nullptr); \ - else \ - png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ - png_handler.png_ptr = nullptr; \ - png_handler.row_ptr = nullptr; \ - png_handler.info_ptr = nullptr; \ - png_handler.end_info_ptr = nullptr; \ +#define PNG_CLEANUP \ + if (png_handler.png_ptr) { \ + if (png_handler.row_ptr) { \ + png_free(png_handler.png_ptr, png_handler.row_ptr); \ + } \ + if (png_handler.end_info_ptr) { \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + &png_handler.end_info_ptr); \ + } else if (png_handler.info_ptr) { \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + nullptr); \ + } else { \ + png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ + } \ + png_handler.png_ptr = nullptr; \ + png_handler.row_ptr = nullptr; \ + png_handler.info_ptr = nullptr; \ + png_handler.end_info_ptr = nullptr; \ } struct BufState { - const uint8_t* data; - size_t bytes_left; + const uint8_t *data; + size_t bytes_left; }; struct PngObjectHandler { - png_infop info_ptr = nullptr; + png_infop info_ptr = nullptr; png_structp png_ptr = nullptr; - png_infop end_info_ptr = nullptr; - png_voidp row_ptr = nullptr; - BufState* buf_state = nullptr; + png_infop end_info_ptr = nullptr; + png_voidp row_ptr = nullptr; + BufState *buf_state = nullptr; ~PngObjectHandler() { - if (row_ptr) - png_free(png_ptr, row_ptr); - if (end_info_ptr) + if (row_ptr) { png_free(png_ptr, row_ptr); } + if (end_info_ptr) { png_destroy_read_struct(&png_ptr, &info_ptr, &end_info_ptr); - else if (info_ptr) + } else if (info_ptr) { png_destroy_read_struct(&png_ptr, &info_ptr, nullptr); - else + } else { png_destroy_read_struct(&png_ptr, nullptr, nullptr); + } delete buf_state; } }; void user_read_data(png_structp png_ptr, png_bytep data, size_t length) { - BufState* buf_state = static_cast(png_get_io_ptr(png_ptr)); - if (length > buf_state->bytes_left) { - png_error(png_ptr, "read error"); - } + BufState *buf_state = static_cast(png_get_io_ptr(png_ptr)); + if (length > buf_state->bytes_left) { png_error(png_ptr, "read error"); } memcpy(data, buf_state->data, length); buf_state->bytes_left -= length; buf_state->data += length; @@ -82,10 +81,8 @@ static const int kPngHeaderSize = 8; // Entry point for LibFuzzer. // Roughly follows the libpng book example: // http://www.libpng.org/pub/png/book/chapter13.html -extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - if (size < kPngHeaderSize) { - return 0; - } +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + if (size < kPngHeaderSize) { return 0; } std::vector v(data, data + size); if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { @@ -99,11 +96,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_handler.info_ptr = nullptr; png_handler.end_info_ptr = nullptr; - png_handler.png_ptr = png_create_read_struct - (PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); - if (!png_handler.png_ptr) { - return 0; - } + png_handler.png_ptr = + png_create_read_struct(PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); + if (!png_handler.png_ptr) { return 0; } png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); if (!png_handler.info_ptr) { @@ -144,12 +139,12 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { } png_uint_32 width, height; - int bit_depth, color_type, interlace_type, compression_type; - int filter_type; + int bit_depth, color_type, interlace_type, compression_type; + int filter_type; - if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, - &height, &bit_depth, &color_type, &interlace_type, - &compression_type, &filter_type)) { + if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, &height, + &bit_depth, &color_type, &interlace_type, &compression_type, + &filter_type)) { PNG_CLEANUP return 0; } @@ -158,11 +153,11 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { if (width && height > 100000000 / width) { PNG_CLEANUP #ifdef HAS_DUMMY_CRASH - #ifdef __aarch64__ - asm volatile (".word 0xf7f0a000\n"); - #else - asm("ud2"); - #endif + #ifdef __aarch64__ + asm volatile(".word 0xf7f0a000\n"); + #else + asm("ud2"); + #endif #endif return 0; } @@ -178,9 +173,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); - png_handler.row_ptr = png_malloc( - png_handler.png_ptr, png_get_rowbytes(png_handler.png_ptr, - png_handler.info_ptr)); + png_handler.row_ptr = + png_malloc(png_handler.png_ptr, + png_get_rowbytes(png_handler.png_ptr, png_handler.info_ptr)); for (int pass = 0; pass < passes; ++pass) { for (png_uint_32 y = 0; y < height; ++y) { @@ -194,4 +189,3 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { PNG_CLEANUP return 0; } - diff --git a/fuzzers/libfuzzer_libpng_accounting/harness.cc b/fuzzers/libfuzzer_libpng_accounting/harness.cc index 65faff685d..5c36517376 100644 --- a/fuzzers/libfuzzer_libpng_accounting/harness.cc +++ b/fuzzers/libfuzzer_libpng_accounting/harness.cc @@ -23,55 +23,54 @@ #define PNG_INTERNAL #include "png.h" -#define PNG_CLEANUP \ - if(png_handler.png_ptr) \ - { \ - if (png_handler.row_ptr) \ - png_free(png_handler.png_ptr, png_handler.row_ptr); \ - if (png_handler.end_info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - &png_handler.end_info_ptr); \ - else if (png_handler.info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - nullptr); \ - else \ - png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ - png_handler.png_ptr = nullptr; \ - png_handler.row_ptr = nullptr; \ - png_handler.info_ptr = nullptr; \ - png_handler.end_info_ptr = nullptr; \ +#define PNG_CLEANUP \ + if (png_handler.png_ptr) { \ + if (png_handler.row_ptr) { \ + png_free(png_handler.png_ptr, png_handler.row_ptr); \ + } \ + if (png_handler.end_info_ptr) { \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + &png_handler.end_info_ptr); \ + } else if (png_handler.info_ptr) { \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + nullptr); \ + } else { \ + png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ + } \ + png_handler.png_ptr = nullptr; \ + png_handler.row_ptr = nullptr; \ + png_handler.info_ptr = nullptr; \ + png_handler.end_info_ptr = nullptr; \ } struct BufState { - const uint8_t* data; - size_t bytes_left; + const uint8_t *data; + size_t bytes_left; }; struct PngObjectHandler { - png_infop info_ptr = nullptr; + png_infop info_ptr = nullptr; png_structp png_ptr = nullptr; - png_infop end_info_ptr = nullptr; - png_voidp row_ptr = nullptr; - BufState* buf_state = nullptr; + png_infop end_info_ptr = nullptr; + png_voidp row_ptr = nullptr; + BufState *buf_state = nullptr; ~PngObjectHandler() { - if (row_ptr) - png_free(png_ptr, row_ptr); - if (end_info_ptr) + if (row_ptr) { png_free(png_ptr, row_ptr); } + if (end_info_ptr) { png_destroy_read_struct(&png_ptr, &info_ptr, &end_info_ptr); - else if (info_ptr) + } else if (info_ptr) { png_destroy_read_struct(&png_ptr, &info_ptr, nullptr); - else + } else { png_destroy_read_struct(&png_ptr, nullptr, nullptr); + } delete buf_state; } }; void user_read_data(png_structp png_ptr, png_bytep data, size_t length) { - BufState* buf_state = static_cast(png_get_io_ptr(png_ptr)); - if (length > buf_state->bytes_left) { - png_error(png_ptr, "read error"); - } + BufState *buf_state = static_cast(png_get_io_ptr(png_ptr)); + if (length > buf_state->bytes_left) { png_error(png_ptr, "read error"); } memcpy(data, buf_state->data, length); buf_state->bytes_left -= length; buf_state->data += length; @@ -82,10 +81,8 @@ static const int kPngHeaderSize = 8; // Entry point for LibFuzzer. // Roughly follows the libpng book example: // http://www.libpng.org/pub/png/book/chapter13.html -extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - if (size < kPngHeaderSize) { - return 0; - } +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + if (size < kPngHeaderSize) { return 0; } std::vector v(data, data + size); if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { @@ -99,11 +96,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_handler.info_ptr = nullptr; png_handler.end_info_ptr = nullptr; - png_handler.png_ptr = png_create_read_struct - (PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); - if (!png_handler.png_ptr) { - return 0; - } + png_handler.png_ptr = + png_create_read_struct(PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); + if (!png_handler.png_ptr) { return 0; } png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); if (!png_handler.info_ptr) { @@ -144,12 +139,12 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { } png_uint_32 width, height; - int bit_depth, color_type, interlace_type, compression_type; - int filter_type; + int bit_depth, color_type, interlace_type, compression_type; + int filter_type; - if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, - &height, &bit_depth, &color_type, &interlace_type, - &compression_type, &filter_type)) { + if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, &height, + &bit_depth, &color_type, &interlace_type, &compression_type, + &filter_type)) { PNG_CLEANUP return 0; } @@ -158,11 +153,11 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { if (width && height > 100000000 / width) { PNG_CLEANUP #ifdef HAS_DUMMY_CRASH - #ifdef __aarch64__ - asm volatile (".word 0xf7f0a000\n"); - #else - asm("ud2"); - #endif + #ifdef __aarch64__ + asm volatile(".word 0xf7f0a000\n"); + #else + asm("ud2"); + #endif #endif return 0; } @@ -178,9 +173,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); - png_handler.row_ptr = png_malloc( - png_handler.png_ptr, png_get_rowbytes(png_handler.png_ptr, - png_handler.info_ptr)); + png_handler.row_ptr = + png_malloc(png_handler.png_ptr, + png_get_rowbytes(png_handler.png_ptr, png_handler.info_ptr)); for (int pass = 0; pass < passes; ++pass) { for (png_uint_32 y = 0; y < height; ++y) { @@ -194,4 +189,3 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { PNG_CLEANUP return 0; } - diff --git a/fuzzers/libfuzzer_libpng_ctx/harness.cc b/fuzzers/libfuzzer_libpng_ctx/harness.cc index 65faff685d..e26e707e17 100644 --- a/fuzzers/libfuzzer_libpng_ctx/harness.cc +++ b/fuzzers/libfuzzer_libpng_ctx/harness.cc @@ -23,40 +23,38 @@ #define PNG_INTERNAL #include "png.h" -#define PNG_CLEANUP \ - if(png_handler.png_ptr) \ - { \ - if (png_handler.row_ptr) \ - png_free(png_handler.png_ptr, png_handler.row_ptr); \ - if (png_handler.end_info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - &png_handler.end_info_ptr); \ - else if (png_handler.info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - nullptr); \ - else \ - png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ - png_handler.png_ptr = nullptr; \ - png_handler.row_ptr = nullptr; \ - png_handler.info_ptr = nullptr; \ - png_handler.end_info_ptr = nullptr; \ +#define PNG_CLEANUP \ + if (png_handler.png_ptr) { \ + if (png_handler.row_ptr) \ + png_free(png_handler.png_ptr, png_handler.row_ptr); \ + if (png_handler.end_info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + &png_handler.end_info_ptr); \ + else if (png_handler.info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + nullptr); \ + else \ + png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ + png_handler.png_ptr = nullptr; \ + png_handler.row_ptr = nullptr; \ + png_handler.info_ptr = nullptr; \ + png_handler.end_info_ptr = nullptr; \ } struct BufState { - const uint8_t* data; - size_t bytes_left; + const uint8_t *data; + size_t bytes_left; }; struct PngObjectHandler { - png_infop info_ptr = nullptr; + png_infop info_ptr = nullptr; png_structp png_ptr = nullptr; - png_infop end_info_ptr = nullptr; - png_voidp row_ptr = nullptr; - BufState* buf_state = nullptr; + png_infop end_info_ptr = nullptr; + png_voidp row_ptr = nullptr; + BufState *buf_state = nullptr; ~PngObjectHandler() { - if (row_ptr) - png_free(png_ptr, row_ptr); + if (row_ptr) { png_free(png_ptr, row_ptr); } if (end_info_ptr) png_destroy_read_struct(&png_ptr, &info_ptr, &end_info_ptr); else if (info_ptr) @@ -68,10 +66,8 @@ struct PngObjectHandler { }; void user_read_data(png_structp png_ptr, png_bytep data, size_t length) { - BufState* buf_state = static_cast(png_get_io_ptr(png_ptr)); - if (length > buf_state->bytes_left) { - png_error(png_ptr, "read error"); - } + BufState *buf_state = static_cast(png_get_io_ptr(png_ptr)); + if (length > buf_state->bytes_left) { png_error(png_ptr, "read error"); } memcpy(data, buf_state->data, length); buf_state->bytes_left -= length; buf_state->data += length; @@ -82,10 +78,8 @@ static const int kPngHeaderSize = 8; // Entry point for LibFuzzer. // Roughly follows the libpng book example: // http://www.libpng.org/pub/png/book/chapter13.html -extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - if (size < kPngHeaderSize) { - return 0; - } +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + if (size < kPngHeaderSize) { return 0; } std::vector v(data, data + size); if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { @@ -99,11 +93,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_handler.info_ptr = nullptr; png_handler.end_info_ptr = nullptr; - png_handler.png_ptr = png_create_read_struct - (PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); - if (!png_handler.png_ptr) { - return 0; - } + png_handler.png_ptr = + png_create_read_struct(PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); + if (!png_handler.png_ptr) { return 0; } png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); if (!png_handler.info_ptr) { @@ -144,12 +136,12 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { } png_uint_32 width, height; - int bit_depth, color_type, interlace_type, compression_type; - int filter_type; + int bit_depth, color_type, interlace_type, compression_type; + int filter_type; - if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, - &height, &bit_depth, &color_type, &interlace_type, - &compression_type, &filter_type)) { + if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, &height, + &bit_depth, &color_type, &interlace_type, &compression_type, + &filter_type)) { PNG_CLEANUP return 0; } @@ -158,11 +150,11 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { if (width && height > 100000000 / width) { PNG_CLEANUP #ifdef HAS_DUMMY_CRASH - #ifdef __aarch64__ - asm volatile (".word 0xf7f0a000\n"); - #else - asm("ud2"); - #endif + #ifdef __aarch64__ + asm volatile(".word 0xf7f0a000\n"); + #else + asm("ud2"); + #endif #endif return 0; } @@ -178,9 +170,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); - png_handler.row_ptr = png_malloc( - png_handler.png_ptr, png_get_rowbytes(png_handler.png_ptr, - png_handler.info_ptr)); + png_handler.row_ptr = + png_malloc(png_handler.png_ptr, + png_get_rowbytes(png_handler.png_ptr, png_handler.info_ptr)); for (int pass = 0; pass < passes; ++pass) { for (png_uint_32 y = 0; y < height; ++y) { @@ -194,4 +186,3 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { PNG_CLEANUP return 0; } - diff --git a/fuzzers/libfuzzer_libpng_launcher/harness.cc b/fuzzers/libfuzzer_libpng_launcher/harness.cc index 65faff685d..e26e707e17 100644 --- a/fuzzers/libfuzzer_libpng_launcher/harness.cc +++ b/fuzzers/libfuzzer_libpng_launcher/harness.cc @@ -23,40 +23,38 @@ #define PNG_INTERNAL #include "png.h" -#define PNG_CLEANUP \ - if(png_handler.png_ptr) \ - { \ - if (png_handler.row_ptr) \ - png_free(png_handler.png_ptr, png_handler.row_ptr); \ - if (png_handler.end_info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - &png_handler.end_info_ptr); \ - else if (png_handler.info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - nullptr); \ - else \ - png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ - png_handler.png_ptr = nullptr; \ - png_handler.row_ptr = nullptr; \ - png_handler.info_ptr = nullptr; \ - png_handler.end_info_ptr = nullptr; \ +#define PNG_CLEANUP \ + if (png_handler.png_ptr) { \ + if (png_handler.row_ptr) \ + png_free(png_handler.png_ptr, png_handler.row_ptr); \ + if (png_handler.end_info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + &png_handler.end_info_ptr); \ + else if (png_handler.info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + nullptr); \ + else \ + png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ + png_handler.png_ptr = nullptr; \ + png_handler.row_ptr = nullptr; \ + png_handler.info_ptr = nullptr; \ + png_handler.end_info_ptr = nullptr; \ } struct BufState { - const uint8_t* data; - size_t bytes_left; + const uint8_t *data; + size_t bytes_left; }; struct PngObjectHandler { - png_infop info_ptr = nullptr; + png_infop info_ptr = nullptr; png_structp png_ptr = nullptr; - png_infop end_info_ptr = nullptr; - png_voidp row_ptr = nullptr; - BufState* buf_state = nullptr; + png_infop end_info_ptr = nullptr; + png_voidp row_ptr = nullptr; + BufState *buf_state = nullptr; ~PngObjectHandler() { - if (row_ptr) - png_free(png_ptr, row_ptr); + if (row_ptr) { png_free(png_ptr, row_ptr); } if (end_info_ptr) png_destroy_read_struct(&png_ptr, &info_ptr, &end_info_ptr); else if (info_ptr) @@ -68,10 +66,8 @@ struct PngObjectHandler { }; void user_read_data(png_structp png_ptr, png_bytep data, size_t length) { - BufState* buf_state = static_cast(png_get_io_ptr(png_ptr)); - if (length > buf_state->bytes_left) { - png_error(png_ptr, "read error"); - } + BufState *buf_state = static_cast(png_get_io_ptr(png_ptr)); + if (length > buf_state->bytes_left) { png_error(png_ptr, "read error"); } memcpy(data, buf_state->data, length); buf_state->bytes_left -= length; buf_state->data += length; @@ -82,10 +78,8 @@ static const int kPngHeaderSize = 8; // Entry point for LibFuzzer. // Roughly follows the libpng book example: // http://www.libpng.org/pub/png/book/chapter13.html -extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - if (size < kPngHeaderSize) { - return 0; - } +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + if (size < kPngHeaderSize) { return 0; } std::vector v(data, data + size); if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { @@ -99,11 +93,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_handler.info_ptr = nullptr; png_handler.end_info_ptr = nullptr; - png_handler.png_ptr = png_create_read_struct - (PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); - if (!png_handler.png_ptr) { - return 0; - } + png_handler.png_ptr = + png_create_read_struct(PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); + if (!png_handler.png_ptr) { return 0; } png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); if (!png_handler.info_ptr) { @@ -144,12 +136,12 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { } png_uint_32 width, height; - int bit_depth, color_type, interlace_type, compression_type; - int filter_type; + int bit_depth, color_type, interlace_type, compression_type; + int filter_type; - if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, - &height, &bit_depth, &color_type, &interlace_type, - &compression_type, &filter_type)) { + if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, &height, + &bit_depth, &color_type, &interlace_type, &compression_type, + &filter_type)) { PNG_CLEANUP return 0; } @@ -158,11 +150,11 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { if (width && height > 100000000 / width) { PNG_CLEANUP #ifdef HAS_DUMMY_CRASH - #ifdef __aarch64__ - asm volatile (".word 0xf7f0a000\n"); - #else - asm("ud2"); - #endif + #ifdef __aarch64__ + asm volatile(".word 0xf7f0a000\n"); + #else + asm("ud2"); + #endif #endif return 0; } @@ -178,9 +170,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); - png_handler.row_ptr = png_malloc( - png_handler.png_ptr, png_get_rowbytes(png_handler.png_ptr, - png_handler.info_ptr)); + png_handler.row_ptr = + png_malloc(png_handler.png_ptr, + png_get_rowbytes(png_handler.png_ptr, png_handler.info_ptr)); for (int pass = 0; pass < passes; ++pass) { for (png_uint_32 y = 0; y < height; ++y) { @@ -194,4 +186,3 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { PNG_CLEANUP return 0; } - diff --git a/fuzzers/libfuzzer_reachability/harness.cc b/fuzzers/libfuzzer_reachability/harness.cc index 65faff685d..e26e707e17 100644 --- a/fuzzers/libfuzzer_reachability/harness.cc +++ b/fuzzers/libfuzzer_reachability/harness.cc @@ -23,40 +23,38 @@ #define PNG_INTERNAL #include "png.h" -#define PNG_CLEANUP \ - if(png_handler.png_ptr) \ - { \ - if (png_handler.row_ptr) \ - png_free(png_handler.png_ptr, png_handler.row_ptr); \ - if (png_handler.end_info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - &png_handler.end_info_ptr); \ - else if (png_handler.info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - nullptr); \ - else \ - png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ - png_handler.png_ptr = nullptr; \ - png_handler.row_ptr = nullptr; \ - png_handler.info_ptr = nullptr; \ - png_handler.end_info_ptr = nullptr; \ +#define PNG_CLEANUP \ + if (png_handler.png_ptr) { \ + if (png_handler.row_ptr) \ + png_free(png_handler.png_ptr, png_handler.row_ptr); \ + if (png_handler.end_info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + &png_handler.end_info_ptr); \ + else if (png_handler.info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + nullptr); \ + else \ + png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ + png_handler.png_ptr = nullptr; \ + png_handler.row_ptr = nullptr; \ + png_handler.info_ptr = nullptr; \ + png_handler.end_info_ptr = nullptr; \ } struct BufState { - const uint8_t* data; - size_t bytes_left; + const uint8_t *data; + size_t bytes_left; }; struct PngObjectHandler { - png_infop info_ptr = nullptr; + png_infop info_ptr = nullptr; png_structp png_ptr = nullptr; - png_infop end_info_ptr = nullptr; - png_voidp row_ptr = nullptr; - BufState* buf_state = nullptr; + png_infop end_info_ptr = nullptr; + png_voidp row_ptr = nullptr; + BufState *buf_state = nullptr; ~PngObjectHandler() { - if (row_ptr) - png_free(png_ptr, row_ptr); + if (row_ptr) { png_free(png_ptr, row_ptr); } if (end_info_ptr) png_destroy_read_struct(&png_ptr, &info_ptr, &end_info_ptr); else if (info_ptr) @@ -68,10 +66,8 @@ struct PngObjectHandler { }; void user_read_data(png_structp png_ptr, png_bytep data, size_t length) { - BufState* buf_state = static_cast(png_get_io_ptr(png_ptr)); - if (length > buf_state->bytes_left) { - png_error(png_ptr, "read error"); - } + BufState *buf_state = static_cast(png_get_io_ptr(png_ptr)); + if (length > buf_state->bytes_left) { png_error(png_ptr, "read error"); } memcpy(data, buf_state->data, length); buf_state->bytes_left -= length; buf_state->data += length; @@ -82,10 +78,8 @@ static const int kPngHeaderSize = 8; // Entry point for LibFuzzer. // Roughly follows the libpng book example: // http://www.libpng.org/pub/png/book/chapter13.html -extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - if (size < kPngHeaderSize) { - return 0; - } +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + if (size < kPngHeaderSize) { return 0; } std::vector v(data, data + size); if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { @@ -99,11 +93,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_handler.info_ptr = nullptr; png_handler.end_info_ptr = nullptr; - png_handler.png_ptr = png_create_read_struct - (PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); - if (!png_handler.png_ptr) { - return 0; - } + png_handler.png_ptr = + png_create_read_struct(PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); + if (!png_handler.png_ptr) { return 0; } png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); if (!png_handler.info_ptr) { @@ -144,12 +136,12 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { } png_uint_32 width, height; - int bit_depth, color_type, interlace_type, compression_type; - int filter_type; + int bit_depth, color_type, interlace_type, compression_type; + int filter_type; - if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, - &height, &bit_depth, &color_type, &interlace_type, - &compression_type, &filter_type)) { + if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, &height, + &bit_depth, &color_type, &interlace_type, &compression_type, + &filter_type)) { PNG_CLEANUP return 0; } @@ -158,11 +150,11 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { if (width && height > 100000000 / width) { PNG_CLEANUP #ifdef HAS_DUMMY_CRASH - #ifdef __aarch64__ - asm volatile (".word 0xf7f0a000\n"); - #else - asm("ud2"); - #endif + #ifdef __aarch64__ + asm volatile(".word 0xf7f0a000\n"); + #else + asm("ud2"); + #endif #endif return 0; } @@ -178,9 +170,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); - png_handler.row_ptr = png_malloc( - png_handler.png_ptr, png_get_rowbytes(png_handler.png_ptr, - png_handler.info_ptr)); + png_handler.row_ptr = + png_malloc(png_handler.png_ptr, + png_get_rowbytes(png_handler.png_ptr, png_handler.info_ptr)); for (int pass = 0; pass < passes; ++pass) { for (png_uint_32 y = 0; y < height; ++y) { @@ -194,4 +186,3 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { PNG_CLEANUP return 0; } - diff --git a/fuzzers/libfuzzer_reachability/weak.c b/fuzzers/libfuzzer_reachability/weak.c index 8d898f8592..4228811157 100644 --- a/fuzzers/libfuzzer_reachability/weak.c +++ b/fuzzers/libfuzzer_reachability/weak.c @@ -1,3 +1,2 @@ -#include +#include __attribute__((weak, visibility("default"))) size_t *__libafl_target_list; - diff --git a/fuzzers/libfuzzer_stb_image/harness.c b/fuzzers/libfuzzer_stb_image/harness.c index 44b62a113b..b560ce511a 100644 --- a/fuzzers/libfuzzer_stb_image/harness.c +++ b/fuzzers/libfuzzer_stb_image/harness.c @@ -9,20 +9,21 @@ #include "stb_image.h" -int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) -{ - int x, y, channels; +int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + int x, y, channels; - if(!stbi_info_from_memory(data, size, &x, &y, &channels)) return 0; + if (!stbi_info_from_memory(data, size, &x, &y, &channels)) {return 0;} - /* exit if the image is larger than ~80MB */ - if(y && x > (80000000 / 4) / y) return 0; + /* exit if the image is larger than ~80MB */ + if (y && x > (80000000 / 4) / y){ return 0;} - unsigned char *img = stbi_load_from_memory(data, size, &x, &y, &channels, 4); + unsigned char *img = stbi_load_from_memory(data, size, &x, &y, &channels, 4); free(img); // if (x > 10000) free(img); // free crash - return 0; + // if (x > 10000) {free(img);} // free crash + + return 0; } diff --git a/fuzzers/libfuzzer_stb_image/stb_image.h b/fuzzers/libfuzzer_stb_image/stb_image.h index 94ae88d11d..c1b2fff9d3 100644 --- a/fuzzers/libfuzzer_stb_image/stb_image.h +++ b/fuzzers/libfuzzer_stb_image/stb_image.h @@ -3,7 +3,8 @@ Do this: #define STB_IMAGE_IMPLEMENTATION - before you include this file in *one* C or C++ file to create the implementation. + before you include this file in *one* C or C++ file to create the +implementation. // i.e. it should look like this: #include ... @@ -13,15 +14,16 @@ #include "stb_image.h" You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. - And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using +malloc,realloc,free QUICK NOTES: Primarily of interest to game developers and other people who can avoid problematic images and only need the trivial interface - JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) - PNG 1/2/4/8/16-bit-per-channel + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as +stock IJG lib) PNG 1/2/4/8/16-bit-per-channel TGA (not sure what subset, if a subset) BMP non-1bpp, non-RLE @@ -50,25 +52,22 @@ RECENT REVISION HISTORY: 2.26 (2020-07-13) many minor fixes 2.25 (2020-02-02) fix warnings - 2.24 (2020-02-02) fix warnings; thread-local failure_reason and flip_vertically - 2.23 (2019-08-11) fix clang static analysis warning - 2.22 (2019-03-04) gif fixes, fix warnings - 2.21 (2019-02-25) fix typo in comment - 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.24 (2020-02-02) fix warnings; thread-local failure_reason and +flip_vertically 2.23 (2019-08-11) fix clang static analysis warning 2.22 +(2019-03-04) gif fixes, fix warnings 2.21 (2019-02-25) fix typo in comment 2.20 +(2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs 2.19 (2018-02-11) fix warning 2.18 (2018-01-30) fix warnings 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings - 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes - 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC - 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs - 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes - 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes - 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 - RGB-format JPEG; remove white matting in PSD; - allocate large structures on the stack; - correct channel count for PNG & BMP - 2.10 (2016-01-22) avoid warning introduced in 2.09 - 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; +bugfixes 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE +detection on GCC 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for +Imagenet JPGs 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; +fixes 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes 2.11 +(2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 RGB-format JPEG; remove +white matting in PSD; allocate large structures on the stack; correct channel +count for PNG & BMP 2.10 (2016-01-22) avoid warning introduced in 2.09 2.09 +(2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED See end of file for full revision history. @@ -86,42 +85,41 @@ RECENT REVISION HISTORY: github:urraka (animated gif) Junggon Kim (PNM comments) Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) socks-the-fox (16-bit PNG) - Jeremy Sawicki (handle all ImageNet JPGs) - Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Jeremy Sawicki (handle all ImageNet +JPGs) Optimizations & bugfixes Mikhail Morozov (1-bit BMP) Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) Arseny Kapoulkine John-Mark Allen Carmelo J Fdez-Aguera Bug & warning fixes - Marc LeBlanc David Woo Guillaume George Martins Mozeiko - Christpher Lloyd Jerry Jansson Joseph Thomson Blazej Dariusz Roszkowski - Phil Jordan Dave Moore Roy Eltham - Hayaki Saito Nathan Reed Won Chun - Luke Graham Johan Duparc Nick Verigakis the Horde3D community - Thomas Ruf Ronny Chevalier github:rlyeh - Janez Zemva John Bartholomew Michal Cichon github:romigrou - Jonathan Blow Ken Hamada Tero Hanninen github:svdijk - Laurent Gomila Cort Stratton github:snagar - Aruelien Pocheville Sergio Gonzalez Thibault Reuille github:Zelex - Cass Everitt Ryamond Barbiero github:grim210 - Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw - Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus - Josh Tobin Matthew Gregan github:poppolopoppo - Julian Raschke Gregory Mullen Christian Floisand github:darealshinji - Baldur Karlsson Kevin Schmidt JR Smith github:Michaelangel007 - Brad Weinberger Matvey Cherevko [reserved] - Luca Sas Alexander Veselov Zack Middleton [reserved] + Marc LeBlanc David Woo Guillaume George Martins +Mozeiko Christpher Lloyd Jerry Jansson Joseph Thomson Blazej +Dariusz Roszkowski Phil Jordan Dave Moore Roy +Eltham Hayaki Saito Nathan Reed Won Chun Luke Graham Johan +Duparc Nick Verigakis the Horde3D community Thomas Ruf Ronny +Chevalier github:rlyeh Janez Zemva John +Bartholomew Michal Cichon github:romigrou Jonathan Blow Ken +Hamada Tero Hanninen github:svdijk Laurent Gomila Cort +Stratton github:snagar Aruelien Pocheville Sergio Gonzalez Thibault +Reuille github:Zelex Cass Everitt Ryamond Barbiero github:grim210 + Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw + Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus + Josh Tobin Matthew Gregan +github:poppolopoppo Julian Raschke Gregory Mullen Christian +Floisand github:darealshinji Baldur Karlsson Kevin Schmidt JR +Smith github:Michaelangel007 Brad Weinberger Matvey Cherevko +[reserved] Luca Sas Alexander Veselov Zack Middleton [reserved] Ryan C. Gordon [reserved] [reserved] DO NOT ADD YOUR NAME HERE - To add your name to the credits, pick a random blank space in the middle and fill it. - 80% of merge conflicts on stb PRs are due to people adding their name at the end - of the credits. + To add your name to the credits, pick a random blank space in the middle and +fill it. 80% of merge conflicts on stb PRs are due to people adding their name +at the end of the credits. */ #ifndef STBI_INCLUDE_STB_IMAGE_H -#define STBI_INCLUDE_STB_IMAGE_H + #define STBI_INCLUDE_STB_IMAGE_H // DOCUMENTATION // @@ -136,14 +134,15 @@ RECENT REVISION HISTORY: // // ... process data if not NULL ... // // ... x = width, y = height, n = # 8-bit components per pixel ... // // ... replace '0' with '1'..'4' to force that many components per pixel -// // ... but 'n' will always be the number that it would have been if you said 0 -// stbi_image_free(data) +// // ... but 'n' will always be the number that it would have been if you +// said 0 stbi_image_free(data) // // Standard parameters: // int *x -- outputs image width in pixels // int *y -- outputs image height in pixels // int *channels_in_file -- outputs # of image components in image file -// int desired_channels -- if non-zero, # of image components requested in result +// int desired_channels -- if non-zero, # of image components requested in +// result // // The return value from an image loader is an 'unsigned char *' which points // to the pixel data, or NULL on an allocation failure or if the image is @@ -171,8 +170,8 @@ RECENT REVISION HISTORY: // and *x, *y, *channels_in_file will be unchanged. The function // stbi_failure_reason() can be queried for an extremely brief, end-user // unfriendly explanation of why the load failed. Define STBI_NO_FAILURE_STRINGS -// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly -// more user-friendly ones. +// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get +// slightly more user-friendly ones. // // Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. // @@ -196,11 +195,12 @@ RECENT REVISION HISTORY: // 2. easy to maintain // 3. good performance // -// Sometimes I let "good performance" creep up in priority over "easy to maintain", -// and for best performance I may provide less-easy-to-use APIs that give higher -// performance, in addition to the easy-to-use ones. Nevertheless, it's important -// to keep in mind that from the standpoint of you, a client of this library, -// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// Sometimes I let "good performance" creep up in priority over "easy to +// maintain", and for best performance I may provide less-easy-to-use APIs that +// give higher performance, in addition to the easy-to-use ones. Nevertheless, +// it's important to keep in mind that from the standpoint of you, a client of +// this library, all you care about is #1 and #3, and stb libraries DO NOT +// emphasize #3 above all. // // Some secondary priorities arise directly from the first two, some of which // provide more explicit reasons why performance can't be emphasized. @@ -219,7 +219,8 @@ RECENT REVISION HISTORY: // overhead. // // The three functions you must define are "read" (reads some bytes of data), -// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the +// end). // // =========================================================================== // @@ -247,10 +248,11 @@ RECENT REVISION HISTORY: // HDR image support (disable by defining STBI_NO_HDR) // // stb_image supports loading HDR images in general, and currently the Radiance -// .HDR file format specifically. You can still load any file through the existing -// interface; if you attempt to load an HDR file, it will be automatically remapped -// to LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; -// both of these constants can be reconfigured through this interface: +// .HDR file format specifically. You can still load any file through the +// existing interface; if you attempt to load an HDR file, it will be +// automatically remapped to LDR, assuming gamma 2.2 and an arbitrary scale +// factor defaulting to 1; both of these constants can be reconfigured through +// this interface: // // stbi_hdr_to_ldr_gamma(2.2f); // stbi_hdr_to_ldr_scale(1.0f); @@ -336,37 +338,36 @@ RECENT REVISION HISTORY: // default this is set to (1 << 24), which is 16777216, but that's still // very big. -#ifndef STBI_NO_STDIO -#include -#endif // STBI_NO_STDIO + #ifndef STBI_NO_STDIO + #include + #endif // STBI_NO_STDIO -#define STBI_VERSION 1 + #define STBI_VERSION 1 -enum -{ - STBI_default = 0, // only used for desired_channels +enum { + STBI_default = 0, // only used for desired_channels - STBI_grey = 1, - STBI_grey_alpha = 2, - STBI_rgb = 3, - STBI_rgb_alpha = 4 + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 }; -#include -typedef unsigned char stbi_uc; + #include +typedef unsigned char stbi_uc; typedef unsigned short stbi_us; -#ifdef __cplusplus + #ifdef __cplusplus extern "C" { -#endif + #endif -#ifndef STBIDEF -#ifdef STB_IMAGE_STATIC -#define STBIDEF static -#else -#define STBIDEF extern -#endif -#endif + #ifndef STBIDEF + #ifdef STB_IMAGE_STATIC + #define STBIDEF static + #else + #define STBIDEF extern + #endif + #endif ////////////////////////////////////////////////////////////////////////////// // @@ -377,11 +378,13 @@ extern "C" { // load image by filename, open file, or memory buffer // -typedef struct -{ - int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read - void (*skip) (void *user,int n); // skip the next 'n' bytes, or 'unget' the last -n bytes if negative - int (*eof) (void *user); // returns nonzero if we are at end of file/data +typedef struct { + int (*read)(void *user, char *data, + int size); // fill 'data' with 'size' bytes. return number of + // bytes actually read + void (*skip)(void *user, int n); // skip the next 'n' bytes, or 'unget' the + // last -n bytes if negative + int (*eof)(void *user); // returns nonzero if we are at end of file/data } stbi_io_callbacks; //////////////////////////////////// @@ -389,376 +392,420 @@ typedef struct // 8-bits-per-channel interface // -STBIDEF stbi_uc *stbi_load_from_memory (stbi_uc const *buffer, int len , int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk , void *user, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *channels_in_file, + int desired_channels); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); -#ifndef STBI_NO_STDIO -STBIDEF stbi_uc *stbi_load (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); -// for stbi_load_from_file, file pointer is left pointing immediately after image -#endif + #ifndef STBI_NO_STDIO +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + // for stbi_load_from_file, file pointer is left pointing immediately after + // image + #endif -#ifndef STBI_NO_GIF -STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp); -#endif + #ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, + int **delays, int *x, int *y, int *z, + int *comp, int req_comp); + #endif -#ifdef STBI_WINDOWS_UTF8 -STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input); -#endif + #ifdef STBI_WINDOWS_UTF8 +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, + const wchar_t *input); + #endif //////////////////////////////////// // // 16-bits-per-channel interface // -STBIDEF stbi_us *stbi_load_16_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, + int *x, int *y, int *channels_in_file, + int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); -#ifndef STBI_NO_STDIO -STBIDEF stbi_us *stbi_load_16 (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); -#endif + #ifndef STBI_NO_STDIO +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + #endif -//////////////////////////////////// -// -// float-per-channel interface -// -#ifndef STBI_NO_LINEAR - STBIDEF float *stbi_loadf_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); - STBIDEF float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + //////////////////////////////////// + // + // float-per-channel interface + // + #ifndef STBI_NO_LINEAR +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *channels_in_file, + int desired_channels); +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); - #ifndef STBI_NO_STDIO - STBIDEF float *stbi_loadf (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); - STBIDEF float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); - #endif -#endif + #ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + #endif + #endif -#ifndef STBI_NO_HDR - STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); - STBIDEF void stbi_hdr_to_ldr_scale(float scale); -#endif // STBI_NO_HDR + #ifndef STBI_NO_HDR +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); +STBIDEF void stbi_hdr_to_ldr_scale(float scale); + #endif // STBI_NO_HDR -#ifndef STBI_NO_LINEAR - STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); - STBIDEF void stbi_ldr_to_hdr_scale(float scale); -#endif // STBI_NO_LINEAR + #ifndef STBI_NO_LINEAR +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); +STBIDEF void stbi_ldr_to_hdr_scale(float scale); + #endif // STBI_NO_LINEAR // stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR -STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); -STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); -#ifndef STBI_NO_STDIO -STBIDEF int stbi_is_hdr (char const *filename); -STBIDEF int stbi_is_hdr_from_file(FILE *f); -#endif // STBI_NO_STDIO - +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, + void *user); +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); + #ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr(char const *filename); +STBIDEF int stbi_is_hdr_from_file(FILE *f); + #endif // STBI_NO_STDIO // get a VERY brief reason for failure // on most compilers (and ALL modern mainstream compilers) this is threadsafe -STBIDEF const char *stbi_failure_reason (void); +STBIDEF const char *stbi_failure_reason(void); // free the loaded image -- this is just free() -STBIDEF void stbi_image_free (void *retval_from_stbi_load); +STBIDEF void stbi_image_free(void *retval_from_stbi_load); // get image dimensions & components without fully decoding -STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); -STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); -STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); -STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, void *user); - -#ifndef STBI_NO_STDIO -STBIDEF int stbi_info (char const *filename, int *x, int *y, int *comp); -STBIDEF int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); -STBIDEF int stbi_is_16_bit (char const *filename); -STBIDEF int stbi_is_16_bit_from_file(FILE *f); -#endif - +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, + int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, + void *user); + #ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit(char const *filename); +STBIDEF int stbi_is_16_bit_from_file(FILE *f); + #endif // for image formats that explicitly notate that they have premultiplied alpha, // we just return the colors as stored in the file. set this flag to force // unpremultiplication. results are undefined if the unpremultiply overflow. -STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); +STBIDEF void stbi_set_unpremultiply_on_load( + int flag_true_if_should_unpremultiply); // indicate whether we should process iphone images back to canonical format, // or just pass them through "as-is" STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); -// flip the image vertically, so the first pixel in the output array is the bottom left +// flip the image vertically, so the first pixel in the output array is the +// bottom left STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); -// as above, but only applies to images loaded on the thread that calls the function -// this function is only available if your compiler supports thread-local variables; -// calling it will fail to link if your compiler doesn't -STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip); +// as above, but only applies to images loaded on the thread that calls the +// function this function is only available if your compiler supports +// thread-local variables; calling it will fail to link if your compiler doesn't +STBIDEF void stbi_set_flip_vertically_on_load_thread( + int flag_true_if_should_flip); // ZLIB client - used by PNG, available for other purposes -STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); -STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, + int initial_size, int *outlen); +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, + int len, + int initial_size, + int *outlen, + int parse_header); STBIDEF char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); -STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen); -STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); -STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); +STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, + int *outlen); +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen); - -#ifdef __cplusplus + #ifdef __cplusplus } -#endif + #endif // // //// end header file ///////////////////////////////////////////////////// -#endif // STBI_INCLUDE_STB_IMAGE_H +#endif // STBI_INCLUDE_STB_IMAGE_H #ifdef STB_IMAGE_IMPLEMENTATION -#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ - || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ - || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ - || defined(STBI_ONLY_ZLIB) - #ifndef STBI_ONLY_JPEG - #define STBI_NO_JPEG - #endif - #ifndef STBI_ONLY_PNG - #define STBI_NO_PNG - #endif - #ifndef STBI_ONLY_BMP - #define STBI_NO_BMP - #endif - #ifndef STBI_ONLY_PSD - #define STBI_NO_PSD - #endif - #ifndef STBI_ONLY_TGA - #define STBI_NO_TGA - #endif - #ifndef STBI_ONLY_GIF - #define STBI_NO_GIF - #endif - #ifndef STBI_ONLY_HDR - #define STBI_NO_HDR - #endif - #ifndef STBI_ONLY_PIC - #define STBI_NO_PIC - #endif - #ifndef STBI_ONLY_PNM - #define STBI_NO_PNM - #endif -#endif + #if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || \ + defined(STBI_ONLY_BMP) || defined(STBI_ONLY_TGA) || \ + defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) || \ + defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || \ + defined(STBI_ONLY_PNM) || defined(STBI_ONLY_ZLIB) + #ifndef STBI_ONLY_JPEG + #define STBI_NO_JPEG + #endif + #ifndef STBI_ONLY_PNG + #define STBI_NO_PNG + #endif + #ifndef STBI_ONLY_BMP + #define STBI_NO_BMP + #endif + #ifndef STBI_ONLY_PSD + #define STBI_NO_PSD + #endif + #ifndef STBI_ONLY_TGA + #define STBI_NO_TGA + #endif + #ifndef STBI_ONLY_GIF + #define STBI_NO_GIF + #endif + #ifndef STBI_ONLY_HDR + #define STBI_NO_HDR + #endif + #ifndef STBI_ONLY_PIC + #define STBI_NO_PIC + #endif + #ifndef STBI_ONLY_PNM + #define STBI_NO_PNM + #endif + #endif -#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) -#define STBI_NO_ZLIB -#endif + #if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && \ + !defined(STBI_NO_ZLIB) + #define STBI_NO_ZLIB + #endif + #include + #include // ptrdiff_t on osx + #include + #include + #include -#include -#include // ptrdiff_t on osx -#include -#include -#include + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) + #include // ldexp, pow + #endif -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -#include // ldexp, pow -#endif + #ifndef STBI_NO_STDIO + #include + #endif -#ifndef STBI_NO_STDIO -#include -#endif + #ifndef STBI_ASSERT + #include + #define STBI_ASSERT(x) assert(x) + #endif -#ifndef STBI_ASSERT -#include -#define STBI_ASSERT(x) assert(x) -#endif + #ifdef __cplusplus + #define STBI_EXTERN extern "C" + #else + #define STBI_EXTERN extern + #endif -#ifdef __cplusplus -#define STBI_EXTERN extern "C" -#else -#define STBI_EXTERN extern -#endif + #ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif + #else + #define stbi_inline __forceinline + #endif + #ifndef STBI_NO_THREAD_LOCALS + #if defined(__cplusplus) && __cplusplus >= 201103L + #define STBI_THREAD_LOCAL thread_local + #elif defined(__GNUC__) && __GNUC__ < 5 + #define STBI_THREAD_LOCAL __thread + #elif defined(_MSC_VER) + #define STBI_THREAD_LOCAL __declspec(thread) + #elif defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && \ + !defined(__STDC_NO_THREADS__) + #define STBI_THREAD_LOCAL _Thread_local + #endif -#ifndef _MSC_VER - #ifdef __cplusplus - #define stbi_inline inline - #else - #define stbi_inline - #endif -#else - #define stbi_inline __forceinline -#endif - -#ifndef STBI_NO_THREAD_LOCALS - #if defined(__cplusplus) && __cplusplus >= 201103L - #define STBI_THREAD_LOCAL thread_local - #elif defined(__GNUC__) && __GNUC__ < 5 - #define STBI_THREAD_LOCAL __thread - #elif defined(_MSC_VER) - #define STBI_THREAD_LOCAL __declspec(thread) - #elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && !defined(__STDC_NO_THREADS__) - #define STBI_THREAD_LOCAL _Thread_local - #endif - - #ifndef STBI_THREAD_LOCAL + #ifndef STBI_THREAD_LOCAL #if defined(__GNUC__) - #define STBI_THREAD_LOCAL __thread + #define STBI_THREAD_LOCAL __thread #endif - #endif -#endif + #endif + #endif -#ifdef _MSC_VER + #ifdef _MSC_VER typedef unsigned short stbi__uint16; -typedef signed short stbi__int16; +typedef signed short stbi__int16; typedef unsigned int stbi__uint32; -typedef signed int stbi__int32; -#else -#include +typedef signed int stbi__int32; + #else + #include typedef uint16_t stbi__uint16; typedef int16_t stbi__int16; typedef uint32_t stbi__uint32; typedef int32_t stbi__int32; -#endif + #endif // should produce compiler error if size is wrong -typedef unsigned char validate_uint32[sizeof(stbi__uint32)==4 ? 1 : -1]; +typedef unsigned char validate_uint32[sizeof(stbi__uint32) == 4 ? 1 : -1]; -#ifdef _MSC_VER -#define STBI_NOTUSED(v) (void)(v) -#else -#define STBI_NOTUSED(v) (void)sizeof(v) -#endif + #ifdef _MSC_VER + #define STBI_NOTUSED(v) (void)(v) + #else + #define STBI_NOTUSED(v) (void)sizeof(v) + #endif -#ifdef _MSC_VER -#define STBI_HAS_LROTL -#endif + #ifdef _MSC_VER + #define STBI_HAS_LROTL + #endif -#ifdef STBI_HAS_LROTL - #define stbi_lrot(x,y) _lrotl(x,y) -#else - #define stbi_lrot(x,y) (((x) << (y)) | ((x) >> (32 - (y)))) -#endif + #ifdef STBI_HAS_LROTL + #define stbi_lrot(x, y) _lrotl(x, y) + #else + #define stbi_lrot(x, y) (((x) << (y)) | ((x) >> (32 - (y)))) + #endif -#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) -// ok -#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) -// ok -#else -#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." -#endif + #if defined(STBI_MALLOC) && defined(STBI_FREE) && \ + (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) + // ok + #elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && \ + !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) + // ok + #else + #error \ + "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." + #endif -#ifndef STBI_MALLOC -#define STBI_MALLOC(sz) malloc(sz) -#define STBI_REALLOC(p,newsz) realloc(p,newsz) -#define STBI_FREE(p) free(p) -#endif + #ifndef STBI_MALLOC + #define STBI_MALLOC(sz) malloc(sz) + #define STBI_REALLOC(p, newsz) realloc(p, newsz) + #define STBI_FREE(p) free(p) + #endif -#ifndef STBI_REALLOC_SIZED -#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) -#endif + #ifndef STBI_REALLOC_SIZED + #define STBI_REALLOC_SIZED(p, oldsz, newsz) STBI_REALLOC(p, newsz) + #endif -// x86/x64 detection -#if defined(__x86_64__) || defined(_M_X64) -#define STBI__X64_TARGET -#elif defined(__i386) || defined(_M_IX86) -#define STBI__X86_TARGET -#endif + // x86/x64 detection + #if defined(__x86_64__) || defined(_M_X64) + #define STBI__X64_TARGET + #elif defined(__i386) || defined(_M_IX86) + #define STBI__X86_TARGET + #endif -#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) -// gcc doesn't support sse2 intrinsics unless you compile with -msse2, -// which in turn means it gets to use SSE2 everywhere. This is unfortunate, -// but previous attempts to provide the SSE2 functions with runtime -// detection caused numerous issues. The way architecture extensions are -// exposed in GCC/Clang is, sadly, not really suited for one-file libs. -// New behavior: if compiled with -msse2, we use SSE2 without any -// detection; if not, we don't use it at all. -#define STBI_NO_SIMD -#endif + #if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && \ + !defined(STBI_NO_SIMD) + // gcc doesn't support sse2 intrinsics unless you compile with -msse2, + // which in turn means it gets to use SSE2 everywhere. This is unfortunate, + // but previous attempts to provide the SSE2 functions with runtime + // detection caused numerous issues. The way architecture extensions are + // exposed in GCC/Clang is, sadly, not really suited for one-file libs. + // New behavior: if compiled with -msse2, we use SSE2 without any + // detection; if not, we don't use it at all. + #define STBI_NO_SIMD + #endif -#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) -// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET -// -// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the -// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. -// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not -// simultaneously enabling "-mstackrealign". -// -// See https://github.com/nothings/stb/issues/81 for more information. -// -// So default to no SSE2 on 32-bit MinGW. If you've read this far and added -// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. -#define STBI_NO_SIMD -#endif + #if defined(__MINGW32__) && defined(STBI__X86_TARGET) && \ + !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) + // Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid + // STBI__X64_TARGET + // + // 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the + // Windows ABI and VC++ as well as Windows DLLs don't maintain that + // invariant. As a result, enabling SSE2 on 32-bit MinGW is dangerous when + // not simultaneously enabling "-mstackrealign". + // + // See https://github.com/nothings/stb/issues/81 for more information. + // + // So default to no SSE2 on 32-bit MinGW. If you've read this far and added + // -mstackrealign to your build settings, feel free to #define + // STBI_MINGW_ENABLE_SSE2. + #define STBI_NO_SIMD + #endif -#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) -#define STBI_SSE2 -#include + #if !defined(STBI_NO_SIMD) && \ + (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) + #define STBI_SSE2 + #include -#ifdef _MSC_VER + #ifdef _MSC_VER -#if _MSC_VER >= 1400 // not VC6 -#include // __cpuid -static int stbi__cpuid3(void) -{ - int info[4]; - __cpuid(info,1); - return info[3]; + #if _MSC_VER >= 1400 // not VC6 + #include // __cpuid +static int stbi__cpuid3(void) { + int info[4]; + __cpuid(info, 1); + return info[3]; } -#else -static int stbi__cpuid3(void) -{ - int res; - __asm { + #else +static int stbi__cpuid3(void) { + int res; + __asm { mov eax,1 cpuid mov res,edx - } - return res; + } + return res; } -#endif + #endif -#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + #define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name -#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) -static int stbi__sse2_available(void) -{ - int info3 = stbi__cpuid3(); - return ((info3 >> 26) & 1) != 0; + #if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) { + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; } -#endif + #endif -#else // assume GCC-style if not VC++ -#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + #else // assume GCC-style if not VC++ + #define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) -#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) -static int stbi__sse2_available(void) -{ - // If we're even attempting to compile this on GCC/Clang, that means - // -msse2 is on, which means the compiler is allowed to use SSE2 - // instructions at will, and so are we. - return 1; + #if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) { + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; } -#endif + #endif -#endif -#endif + #endif + #endif -// ARM NEON -#if defined(STBI_NO_SIMD) && defined(STBI_NEON) -#undef STBI_NEON -#endif + // ARM NEON + #if defined(STBI_NO_SIMD) && defined(STBI_NEON) + #undef STBI_NEON + #endif -#ifdef STBI_NEON -#include -// assume GCC or Clang on ARM targets -#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) -#endif + #ifdef STBI_NEON + #include + // assume GCC or Clang on ARM targets + #define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + #endif -#ifndef STBI_SIMD_ALIGN -#define STBI_SIMD_ALIGN(type, name) type name -#endif + #ifndef STBI_SIMD_ALIGN + #define STBI_SIMD_ALIGN(type, name) type name + #endif -#ifndef STBI_MAX_DIMENSIONS -#define STBI_MAX_DIMENSIONS (1 << 24) -#endif + #ifndef STBI_MAX_DIMENSIONS + #define STBI_MAX_DIMENSIONS (1 << 24) + #endif /////////////////////////////////////////////// // @@ -766,188 +813,182 @@ static int stbi__sse2_available(void) // stbi__context structure is our basic context used by all images, so it // contains all the IO context, plus some basic image information -typedef struct -{ - stbi__uint32 img_x, img_y; - int img_n, img_out_n; +typedef struct { + stbi__uint32 img_x, img_y; + int img_n, img_out_n; - stbi_io_callbacks io; - void *io_user_data; + stbi_io_callbacks io; + void *io_user_data; - int read_from_callbacks; - int buflen; - stbi_uc buffer_start[128]; - int callback_already_read; + int read_from_callbacks; + int buflen; + stbi_uc buffer_start[128]; + int callback_already_read; - stbi_uc *img_buffer, *img_buffer_end; - stbi_uc *img_buffer_original, *img_buffer_original_end; + stbi_uc *img_buffer, *img_buffer_end; + stbi_uc *img_buffer_original, *img_buffer_original_end; } stbi__context; - static void stbi__refill_buffer(stbi__context *s); // initialize a memory-decode context -static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) -{ - s->io.read = NULL; - s->read_from_callbacks = 0; - s->callback_already_read = 0; - s->img_buffer = s->img_buffer_original = (stbi_uc *) buffer; - s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *) buffer+len; +static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) { + s->io.read = NULL; + s->read_from_callbacks = 0; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = (stbi_uc *)buffer; + s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *)buffer + len; } // initialize a callback-based context -static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, void *user) -{ - s->io = *c; - s->io_user_data = user; - s->buflen = sizeof(s->buffer_start); - s->read_from_callbacks = 1; - s->callback_already_read = 0; - s->img_buffer = s->img_buffer_original = s->buffer_start; - stbi__refill_buffer(s); - s->img_buffer_original_end = s->img_buffer_end; +static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, + void *user) { + s->io = *c; + s->io_user_data = user; + s->buflen = sizeof(s->buffer_start); + s->read_from_callbacks = 1; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = s->buffer_start; + stbi__refill_buffer(s); + s->img_buffer_original_end = s->img_buffer_end; } -#ifndef STBI_NO_STDIO + #ifndef STBI_NO_STDIO -static int stbi__stdio_read(void *user, char *data, int size) -{ - return (int) fread(data,1,size,(FILE*) user); +static int stbi__stdio_read(void *user, char *data, int size) { + return (int)fread(data, 1, size, (FILE *)user); } -static void stbi__stdio_skip(void *user, int n) -{ - int ch; - fseek((FILE*) user, n, SEEK_CUR); - ch = fgetc((FILE*) user); /* have to read a byte to reset feof()'s flag */ - if (ch != EOF) { - ungetc(ch, (FILE *) user); /* push byte back onto stream if valid. */ - } +static void stbi__stdio_skip(void *user, int n) { + int ch; + fseek((FILE *)user, n, SEEK_CUR); + ch = fgetc((FILE *)user); /* have to read a byte to reset feof()'s flag */ + if (ch != EOF) { + ungetc(ch, (FILE *)user); /* push byte back onto stream if valid. */ + } } -static int stbi__stdio_eof(void *user) -{ - return feof((FILE*) user) || ferror((FILE *) user); +static int stbi__stdio_eof(void *user) { + return feof((FILE *)user) || ferror((FILE *)user); } -static stbi_io_callbacks stbi__stdio_callbacks = -{ - stbi__stdio_read, - stbi__stdio_skip, - stbi__stdio_eof, +static stbi_io_callbacks stbi__stdio_callbacks = { + stbi__stdio_read, + stbi__stdio_skip, + stbi__stdio_eof, }; -static void stbi__start_file(stbi__context *s, FILE *f) -{ - stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *) f); +static void stbi__start_file(stbi__context *s, FILE *f) { + stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *)f); } -//static void stop_file(stbi__context *s) { } + // static void stop_file(stbi__context *s) { } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -static void stbi__rewind(stbi__context *s) -{ - // conceptually rewind SHOULD rewind to the beginning of the stream, - // but we just rewind to the beginning of the initial buffer, because - // we only use it after doing 'test', which only ever looks at at most 92 bytes - s->img_buffer = s->img_buffer_original; - s->img_buffer_end = s->img_buffer_original_end; +static void stbi__rewind(stbi__context *s) { + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 + // bytes + s->img_buffer = s->img_buffer_original; + s->img_buffer_end = s->img_buffer_original_end; } -enum -{ - STBI_ORDER_RGB, - STBI_ORDER_BGR -}; +enum { STBI_ORDER_RGB, STBI_ORDER_BGR }; -typedef struct -{ - int bits_per_channel; - int num_channels; - int channel_order; +typedef struct { + int bits_per_channel; + int num_channels; + int channel_order; } stbi__result_info; -#ifndef STBI_NO_JPEG -static int stbi__jpeg_test(stbi__context *s); -static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context *s); +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PNG -static int stbi__png_test(stbi__context *s); -static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); -static int stbi__png_is16(stbi__context *s); -#endif + #ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context *s); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__png_is16(stbi__context *s); + #endif -#ifndef STBI_NO_BMP -static int stbi__bmp_test(stbi__context *s); -static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context *s); +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_TGA -static int stbi__tga_test(stbi__context *s); -static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s); +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PSD -static int stbi__psd_test(stbi__context *s); -static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc); -static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); -static int stbi__psd_is16(stbi__context *s); -#endif + #ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc); +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__psd_is16(stbi__context *s); + #endif -#ifndef STBI_NO_HDR -static int stbi__hdr_test(stbi__context *s); -static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_HDR +static int stbi__hdr_test(stbi__context *s); +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PIC -static int stbi__pic_test(stbi__context *s); -static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context *s); +static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_GIF -static int stbi__gif_test(stbi__context *s); -static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp); -static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context *s); +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, + int *z, int *comp, int req_comp); +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PNM -static int stbi__pnm_test(stbi__context *s); -static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context *s); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); + #endif static -#ifdef STBI_THREAD_LOCAL -STBI_THREAD_LOCAL -#endif -const char *stbi__g_failure_reason; + #ifdef STBI_THREAD_LOCAL + STBI_THREAD_LOCAL + #endif + const char *stbi__g_failure_reason; -STBIDEF const char *stbi_failure_reason(void) -{ - return stbi__g_failure_reason; +STBIDEF const char *stbi_failure_reason(void) { + return stbi__g_failure_reason; } -#ifndef STBI_NO_FAILURE_STRINGS -static int stbi__err(const char *str) -{ - stbi__g_failure_reason = str; - return 0; + #ifndef STBI_NO_FAILURE_STRINGS +static int stbi__err(const char *str) { + stbi__g_failure_reason = str; + return 0; } -#endif + #endif -static void *stbi__malloc(size_t size) -{ - return STBI_MALLOC(size); +static void *stbi__malloc(size_t size) { + return STBI_MALLOC(size); } // stb_image uses ints pervasively, including for offset calculations. @@ -962,317 +1003,336 @@ static void *stbi__malloc(size_t size) // return 1 if the sum is valid, 0 on overflow. // negative terms are considered invalid. -static int stbi__addsizes_valid(int a, int b) -{ - if (b < 0) return 0; - // now 0 <= b <= INT_MAX, hence also - // 0 <= INT_MAX - b <= INTMAX. - // And "a + b <= INT_MAX" (which might overflow) is the - // same as a <= INT_MAX - b (no overflow) - return a <= INT_MAX - b; +static int stbi__addsizes_valid(int a, int b) { + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; } // returns 1 if the product is valid, 0 on overflow. // negative factors are considered invalid. -static int stbi__mul2sizes_valid(int a, int b) -{ - if (a < 0 || b < 0) return 0; - if (b == 0) return 1; // mul-by-0 is always safe - // portable way to check for no overflows in a*b - return a <= INT_MAX/b; +static int stbi__mul2sizes_valid(int a, int b) { + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX / b; } -#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) + #if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || \ + !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) // returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow -static int stbi__mad2sizes_valid(int a, int b, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a*b, add); +static int stbi__mad2sizes_valid(int a, int b, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a * b, add); } -#endif + #endif // returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow -static int stbi__mad3sizes_valid(int a, int b, int c, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && - stbi__addsizes_valid(a*b*c, add); +static int stbi__mad3sizes_valid(int a, int b, int c, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__addsizes_valid(a * b * c, add); } -// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && - stbi__mul2sizes_valid(a*b*c, d) && stbi__addsizes_valid(a*b*c*d, add); + // returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't + // overflow + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__mul2sizes_valid(a * b * c, d) && + stbi__addsizes_valid(a * b * c * d, add); } -#endif + #endif -#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) + #if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || \ + !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) // mallocs with size overflow checking -static void *stbi__malloc_mad2(int a, int b, int add) -{ - if (!stbi__mad2sizes_valid(a, b, add)) return NULL; - return stbi__malloc(a*b + add); +static void *stbi__malloc_mad2(int a, int b, int add) { + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a * b + add); } -#endif + #endif -static void *stbi__malloc_mad3(int a, int b, int c, int add) -{ - if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; - return stbi__malloc(a*b*c + add); +static void *stbi__malloc_mad3(int a, int b, int c, int add) { + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a * b * c + add); } -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) -{ - if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; - return stbi__malloc(a*b*c*d + add); + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) { + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a * b * c * d + add); } -#endif + #endif // stbi__err - error // stbi__errpf - error returning pointer to float // stbi__errpuc - error returning pointer to unsigned char -#ifdef STBI_NO_FAILURE_STRINGS - #define stbi__err(x,y) 0 -#elif defined(STBI_FAILURE_USERMSG) - #define stbi__err(x,y) stbi__err(y) -#else - #define stbi__err(x,y) stbi__err(x) -#endif + #ifdef STBI_NO_FAILURE_STRINGS + #define stbi__err(x, y) 0 + #elif defined(STBI_FAILURE_USERMSG) + #define stbi__err(x, y) stbi__err(y) + #else + #define stbi__err(x, y) stbi__err(x) + #endif -#define stbi__errpf(x,y) ((float *)(size_t) (stbi__err(x,y)?NULL:NULL)) -#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + #define stbi__errpf(x, y) ((float *)(size_t)(stbi__err(x, y) ? NULL : NULL)) + #define stbi__errpuc(x, y) \ + ((unsigned char *)(size_t)(stbi__err(x, y) ? NULL : NULL)) -STBIDEF void stbi_image_free(void *retval_from_stbi_load) -{ - STBI_FREE(retval_from_stbi_load); +STBIDEF void stbi_image_free(void *retval_from_stbi_load) { + STBI_FREE(retval_from_stbi_load); } -#ifndef STBI_NO_LINEAR -static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); -#endif + #ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); + #endif -#ifndef STBI_NO_HDR -static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); -#endif + #ifndef STBI_NO_HDR +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); + #endif static int stbi__vertically_flip_on_load_global = 0; -STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) -{ - stbi__vertically_flip_on_load_global = flag_true_if_should_flip; +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) { + stbi__vertically_flip_on_load_global = flag_true_if_should_flip; } -#ifndef STBI_THREAD_LOCAL -#define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global -#else -static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, stbi__vertically_flip_on_load_set; + #ifndef STBI_THREAD_LOCAL + #define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global + #else +static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, + stbi__vertically_flip_on_load_set; -STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip) -{ - stbi__vertically_flip_on_load_local = flag_true_if_should_flip; - stbi__vertically_flip_on_load_set = 1; +STBIDEF void stbi_set_flip_vertically_on_load_thread( + int flag_true_if_should_flip) { + stbi__vertically_flip_on_load_local = flag_true_if_should_flip; + stbi__vertically_flip_on_load_set = 1; } -#define stbi__vertically_flip_on_load (stbi__vertically_flip_on_load_set \ - ? stbi__vertically_flip_on_load_local \ - : stbi__vertically_flip_on_load_global) -#endif // STBI_THREAD_LOCAL + #define stbi__vertically_flip_on_load \ + (stbi__vertically_flip_on_load_set \ + ? stbi__vertically_flip_on_load_local \ + : stbi__vertically_flip_on_load_global) + #endif // STBI_THREAD_LOCAL -static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) -{ - memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields - ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed - ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order - ri->num_channels = 0; +static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc) { + memset(ri, 0, + sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = + 8; // default is 8 so most paths don't have to be changed + ri->channel_order = + STBI_ORDER_RGB; // all current input & output are this, but this is here + // so we can add BGR order + ri->num_channels = 0; - #ifndef STBI_NO_JPEG - if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PNG - if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_BMP - if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_GIF - if (stbi__gif_test(s)) return stbi__gif_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PSD - if (stbi__psd_test(s)) return stbi__psd_load(s,x,y,comp,req_comp, ri, bpc); - #else - STBI_NOTUSED(bpc); - #endif - #ifndef STBI_NO_PIC - if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PNM - if (stbi__pnm_test(s)) return stbi__pnm_load(s,x,y,comp,req_comp, ri); - #endif + #ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PSD + if (stbi__psd_test(s)) + return stbi__psd_load(s, x, y, comp, req_comp, ri, bpc); + #else + STBI_NOTUSED(bpc); + #endif + #ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s, x, y, comp, req_comp, ri); + #endif - #ifndef STBI_NO_HDR - if (stbi__hdr_test(s)) { - float *hdr = stbi__hdr_load(s, x,y,comp,req_comp, ri); - return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); - } - #endif + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float *hdr = stbi__hdr_load(s, x, y, comp, req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif - #ifndef STBI_NO_TGA - // test tga last because it's a crappy test! - if (stbi__tga_test(s)) - return stbi__tga_load(s,x,y,comp,req_comp, ri); - #endif + #ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) return stbi__tga_load(s, x, y, comp, req_comp, ri); + #endif - return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); + return stbi__errpuc("unknown image type", + "Image not of any known type, or corrupt"); } -static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, int channels) -{ - int i; - int img_len = w * h * channels; - stbi_uc *reduced; +static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, + int channels) { + int i; + int img_len = w * h * channels; + stbi_uc *reduced; - reduced = (stbi_uc *) stbi__malloc(img_len); - if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + reduced = (stbi_uc *)stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); - for (i = 0; i < img_len; ++i) - reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + for (i = 0; i < img_len; ++i) { + reduced[i] = (stbi_uc)((orig[i] >> 8) & + 0xFF); // top half of each byte is sufficient approx + // of 16->8 bit scaling + } - STBI_FREE(orig); - return reduced; + STBI_FREE(orig); + return reduced; } -static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, int channels) -{ - int i; - int img_len = w * h * channels; - stbi__uint16 *enlarged; +static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, + int channels) { + int i; + int img_len = w * h * channels; + stbi__uint16 *enlarged; - enlarged = (stbi__uint16 *) stbi__malloc(img_len*2); - if (enlarged == NULL) return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + enlarged = (stbi__uint16 *)stbi__malloc(img_len * 2); + if (enlarged == NULL) + return (stbi__uint16 *)stbi__errpuc("outofmem", "Out of memory"); - for (i = 0; i < img_len; ++i) - enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + for (i = 0; i < img_len; ++i) { + enlarged[i] = (stbi__uint16)((orig[i] << 8) + + orig[i]); // replicate to high and low byte, + // maps 0->0, 255->0xffff + } - STBI_FREE(orig); - return enlarged; + STBI_FREE(orig); + return enlarged; } -static void stbi__vertical_flip(void *image, int w, int h, int bytes_per_pixel) -{ - int row; - size_t bytes_per_row = (size_t)w * bytes_per_pixel; - stbi_uc temp[2048]; - stbi_uc *bytes = (stbi_uc *)image; +static void stbi__vertical_flip(void *image, int w, int h, + int bytes_per_pixel) { + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc *bytes = (stbi_uc *)image; - for (row = 0; row < (h>>1); row++) { - stbi_uc *row0 = bytes + row*bytes_per_row; - stbi_uc *row1 = bytes + (h - row - 1)*bytes_per_row; - // swap row0 with row1 - size_t bytes_left = bytes_per_row; - while (bytes_left) { - size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); - memcpy(temp, row0, bytes_copy); - memcpy(row0, row1, bytes_copy); - memcpy(row1, temp, bytes_copy); - row0 += bytes_copy; - row1 += bytes_copy; - bytes_left -= bytes_copy; - } - } + for (row = 0; row < (h >> 1); row++) { + stbi_uc *row0 = bytes + row * bytes_per_row; + stbi_uc *row1 = bytes + (h - row - 1) * bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = + (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } } -#ifndef STBI_NO_GIF -static void stbi__vertical_flip_slices(void *image, int w, int h, int z, int bytes_per_pixel) -{ - int slice; - int slice_size = w * h * bytes_per_pixel; + #ifndef STBI_NO_GIF +static void stbi__vertical_flip_slices(void *image, int w, int h, int z, + int bytes_per_pixel) { + int slice; + int slice_size = w * h * bytes_per_pixel; - stbi_uc *bytes = (stbi_uc *)image; - for (slice = 0; slice < z; ++slice) { - stbi__vertical_flip(bytes, w, h, bytes_per_pixel); - bytes += slice_size; - } + stbi_uc *bytes = (stbi_uc *)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } } -#endif + #endif -static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - stbi__result_info ri; - void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); +static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, + int *y, int *comp, + int req_comp) { + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); - if (result == NULL) - return NULL; + if (result == NULL) return NULL; - // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. - STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + // it is the responsibility of the loaders to make sure we get either 8 or 16 + // bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); - if (ri.bits_per_channel != 8) { - result = stbi__convert_16_to_8((stbi__uint16 *) result, *x, *y, req_comp == 0 ? *comp : req_comp); - ri.bits_per_channel = 8; - } + if (ri.bits_per_channel != 8) { + result = stbi__convert_16_to_8((stbi__uint16 *)result, *x, *y, + req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } - // @TODO: move stbi__convert_format to here + // @TODO: move stbi__convert_format to here - if (stbi__vertically_flip_on_load) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); - } + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } - return (unsigned char *) result; + return (unsigned char *)result; } -static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - stbi__result_info ri; - void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); +static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, + int *y, int *comp, + int req_comp) { + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); - if (result == NULL) - return NULL; + if (result == NULL) return NULL; - // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. - STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + // it is the responsibility of the loaders to make sure we get either 8 or 16 + // bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); - if (ri.bits_per_channel != 16) { - result = stbi__convert_8_to_16((stbi_uc *) result, *x, *y, req_comp == 0 ? *comp : req_comp); - ri.bits_per_channel = 16; - } + if (ri.bits_per_channel != 16) { + result = stbi__convert_8_to_16((stbi_uc *)result, *x, *y, + req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } - // @TODO: move stbi__convert_format16 to here - // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to + // keep more precision - if (stbi__vertically_flip_on_load) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); - } + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } - return (stbi__uint16 *) result; + return (stbi__uint16 *)result; } -#if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) -static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, int req_comp) -{ - if (stbi__vertically_flip_on_load && result != NULL) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); - } + #if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, + int req_comp) { + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } } -#endif + #endif -#ifndef STBI_NO_STDIO + #ifndef STBI_NO_STDIO -#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) -STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide); -STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default); -#endif + #if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) +STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar( + unsigned int cp, unsigned long flags, const char *str, int cbmb, + wchar_t *widestr, int cchwide); +STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte( + unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, + char *str, int cbmb, const char *defchar, int *used_default); + #endif #if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input) { return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL); } -#endif + #endif static FILE *stbi__fopen(char const *filename, char const *mode) { @@ -1293,384 +1353,386 @@ static FILE *stbi__fopen(char const *filename, char const *mode) f = _wfopen(wFilename, wMode); #endif -#elif defined(_MSC_VER) && _MSC_VER >= 1400 - if (0 != fopen_s(&f, filename, mode)) - f=0; -#else - f = fopen(filename, mode); -#endif - return f; + #elif defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) f = 0; + #else + f = fopen(filename, mode); + #endif + return f; } - -STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - unsigned char *result; - if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); - result = stbi_load_from_file(f,x,y,comp,req_comp); - fclose(f); - return result; +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, + int req_comp) { + FILE *f = stbi__fopen(filename, "rb"); + unsigned char *result; + if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f, x, y, comp, req_comp); + fclose(f); + return result; } -STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - unsigned char *result; - stbi__context s; - stbi__start_file(&s,f); - result = stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); - if (result) { - // need to 'unget' all the characters in the IO buffer - fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); - } - return result; +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, + int req_comp) { + unsigned char *result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, -(int)(s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; } -STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - stbi__uint16 *result; - stbi__context s; - stbi__start_file(&s,f); - result = stbi__load_and_postprocess_16bit(&s,x,y,comp,req_comp); - if (result) { - // need to 'unget' all the characters in the IO buffer - fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); - } - return result; +STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, + int req_comp) { + stbi__uint16 *result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_16bit(&s, x, y, comp, req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, -(int)(s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; } -STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - stbi__uint16 *result; - if (!f) return (stbi_us *) stbi__errpuc("can't fopen", "Unable to open file"); - result = stbi_load_from_file_16(f,x,y,comp,req_comp); - fclose(f); - return result; +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, + int req_comp) { + FILE *f = stbi__fopen(filename, "rb"); + stbi__uint16 *result; + if (!f) return (stbi_us *)stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f, x, y, comp, req_comp); + fclose(f); + return result; } + #endif //! STBI_NO_STDIO -#endif //!STBI_NO_STDIO - -STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, + int *x, int *y, int *channels_in_file, + int desired_channels) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, + desired_channels); } -STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); - return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, + desired_channels); } -STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp, int req_comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); } -STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); } -#ifndef STBI_NO_GIF -STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp) -{ - unsigned char *result; - stbi__context s; - stbi__start_mem(&s,buffer,len); + #ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, + int **delays, int *x, int *y, int *z, + int *comp, int req_comp) { + unsigned char *result; + stbi__context s; + stbi__start_mem(&s, buffer, len); - result = (unsigned char*) stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); - if (stbi__vertically_flip_on_load) { - stbi__vertical_flip_slices( result, *x, *y, *z, *comp ); - } + result = + (unsigned char *)stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices(result, *x, *y, *z, *comp); + } - return result; + return result; } -#endif + #endif -#ifndef STBI_NO_LINEAR -static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - unsigned char *data; - #ifndef STBI_NO_HDR - if (stbi__hdr_test(s)) { - stbi__result_info ri; - float *hdr_data = stbi__hdr_load(s,x,y,comp,req_comp, &ri); - if (hdr_data) - stbi__float_postprocess(hdr_data,x,y,comp,req_comp); - return hdr_data; - } - #endif - data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); - if (data) - return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); - return stbi__errpf("unknown image type", "Image not of any known type, or corrupt"); + #ifndef STBI_NO_LINEAR +static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, + int req_comp) { + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + stbi__result_info ri; + float *hdr_data = stbi__hdr_load(s, x, y, comp, req_comp, &ri); + if (hdr_data) stbi__float_postprocess(hdr_data, x, y, comp, req_comp); + return hdr_data; + } + #endif + data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); + if (data) return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return stbi__errpf("unknown image type", + "Image not of any known type, or corrupt"); } -STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp, int req_comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -#ifndef STBI_NO_STDIO -STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - float *result; - FILE *f = stbi__fopen(filename, "rb"); - if (!f) return stbi__errpf("can't fopen", "Unable to open file"); - result = stbi_loadf_from_file(f,x,y,comp,req_comp); - fclose(f); - return result; + #ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, + int req_comp) { + float *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) return stbi__errpf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f, x, y, comp, req_comp); + fclose(f); + return result; } -STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_file(&s,f); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_file(&s, f); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -#endif // !STBI_NO_LINEAR + #endif // !STBI_NO_LINEAR // these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is // defined, for API simplicity; if STBI_NO_LINEAR is defined, it always // reports false! -STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) -{ - #ifndef STBI_NO_HDR - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__hdr_test(&s); - #else - STBI_NOTUSED(buffer); - STBI_NOTUSED(len); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) { + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif } -#ifndef STBI_NO_STDIO -STBIDEF int stbi_is_hdr (char const *filename) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result=0; - if (f) { - result = stbi_is_hdr_from_file(f); - fclose(f); - } - return result; + #ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr(char const *filename) { + FILE *f = stbi__fopen(filename, "rb"); + int result = 0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; } -STBIDEF int stbi_is_hdr_from_file(FILE *f) -{ - #ifndef STBI_NO_HDR - long pos = ftell(f); - int res; - stbi__context s; - stbi__start_file(&s,f); - res = stbi__hdr_test(&s); - fseek(f, pos, SEEK_SET); - return res; - #else - STBI_NOTUSED(f); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_file(FILE *f) { + #ifndef STBI_NO_HDR + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s, f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; + #else + STBI_NOTUSED(f); + return 0; + #endif } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) -{ - #ifndef STBI_NO_HDR - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__hdr_test(&s); - #else - STBI_NOTUSED(clbk); - STBI_NOTUSED(user); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, + void *user) { + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; + #endif } -#ifndef STBI_NO_LINEAR -static float stbi__l2h_gamma=2.2f, stbi__l2h_scale=1.0f; + #ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma = 2.2f, stbi__l2h_scale = 1.0f; -STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } -STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } -#endif +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { + stbi__l2h_gamma = gamma; +} +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { + stbi__l2h_scale = scale; +} + #endif -static float stbi__h2l_gamma_i=1.0f/2.2f, stbi__h2l_scale_i=1.0f; - -STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1/gamma; } -STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1/scale; } +static float stbi__h2l_gamma_i = 1.0f / 2.2f, stbi__h2l_scale_i = 1.0f; +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { + stbi__h2l_gamma_i = 1 / gamma; +} +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { + stbi__h2l_scale_i = 1 / scale; +} ////////////////////////////////////////////////////////////////////////////// // // Common code used by all image loaders // -enum -{ - STBI__SCAN_load=0, - STBI__SCAN_type, - STBI__SCAN_header -}; +enum { STBI__SCAN_load = 0, STBI__SCAN_type, STBI__SCAN_header }; -static void stbi__refill_buffer(stbi__context *s) -{ - int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); - s->callback_already_read += (int) (s->img_buffer - s->img_buffer_original); - if (n == 0) { - // at end of file, treat same as if from memory, but need to handle case - // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file - s->read_from_callbacks = 0; - s->img_buffer = s->buffer_start; - s->img_buffer_end = s->buffer_start+1; - *s->img_buffer = 0; - } else { - s->img_buffer = s->buffer_start; - s->img_buffer_end = s->buffer_start + n; - } +static void stbi__refill_buffer(stbi__context *s) { + int n = (s->io.read)(s->io_user_data, (char *)s->buffer_start, s->buflen); + s->callback_already_read += (int)(s->img_buffer - s->img_buffer_original); + if (n == 0) { + // at end of file, treat same as if from memory, but need to handle case + // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file + s->read_from_callbacks = 0; + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + 1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } } -stbi_inline static stbi_uc stbi__get8(stbi__context *s) -{ - if (s->img_buffer < s->img_buffer_end) - return *s->img_buffer++; - if (s->read_from_callbacks) { - stbi__refill_buffer(s); - return *s->img_buffer++; - } - return 0; +stbi_inline static stbi_uc stbi__get8(stbi__context *s) { + if (s->img_buffer < s->img_buffer_end) return *s->img_buffer++; + if (s->read_from_callbacks) { + stbi__refill_buffer(s); + return *s->img_buffer++; + } + return 0; } -#if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else -stbi_inline static int stbi__at_eof(stbi__context *s) -{ - if (s->io.read) { - if (!(s->io.eof)(s->io_user_data)) return 0; - // if feof() is true, check if buffer = end - // special case: we've only got the special 0 character at the end - if (s->read_from_callbacks == 0) return 1; - } + #if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && \ + defined(STBI_NO_PNM) + // nothing + #else +stbi_inline static int stbi__at_eof(stbi__context *s) { + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } - return s->img_buffer >= s->img_buffer_end; + return s->img_buffer >= s->img_buffer_end; } -#endif + #endif -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) -// nothing -#else -static void stbi__skip(stbi__context *s, int n) -{ - if (n == 0) return; // already there! - if (n < 0) { + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && \ + defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && \ + defined(STBI_NO_PIC) + // nothing + #else +static void stbi__skip(stbi__context *s, int n) { + if (n == 0) return; // already there! + if (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); return; - } - if (s->io.read) { - int blen = (int) (s->img_buffer_end - s->img_buffer); - if (blen < n) { - s->img_buffer = s->img_buffer_end; - (s->io.skip)(s->io_user_data, n - blen); - return; - } - } - s->img_buffer += n; + } + } + s->img_buffer += n; } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && defined(STBI_NO_PNM) -// nothing -#else -static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) -{ - if (s->io.read) { - int blen = (int) (s->img_buffer_end - s->img_buffer); - if (blen < n) { - int res, count; + #if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && \ + defined(STBI_NO_PNM) + // nothing + #else +static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) { + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { + int res, count; - memcpy(buffer, s->img_buffer, blen); + memcpy(buffer, s->img_buffer, blen); - count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen); - res = (count == (n-blen)); - s->img_buffer = s->img_buffer_end; - return res; - } - } + count = (s->io.read)(s->io_user_data, (char *)buffer + blen, n - blen); + res = (count == (n - blen)); + s->img_buffer = s->img_buffer_end; + return res; + } + } - if (s->img_buffer+n <= s->img_buffer_end) { - memcpy(buffer, s->img_buffer, n); - s->img_buffer += n; - return 1; - } else - return 0; + if (s->img_buffer + n <= s->img_buffer_end) { + memcpy(buffer, s->img_buffer, n); + s->img_buffer += n; + return 1; + } else + return 0; } -#endif + #endif -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) -// nothing -#else -static int stbi__get16be(stbi__context *s) -{ - int z = stbi__get8(s); - return (z << 8) + stbi__get8(s); + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && \ + defined(STBI_NO_PIC) + // nothing + #else +static int stbi__get16be(stbi__context *s) { + int z = stbi__get8(s); + return (z << 8) + stbi__get8(s); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) -// nothing -#else -static stbi__uint32 stbi__get32be(stbi__context *s) -{ - stbi__uint32 z = stbi__get16be(s); - return (z << 16) + stbi__get16be(s); + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) + // nothing + #else +static stbi__uint32 stbi__get32be(stbi__context *s) { + stbi__uint32 z = stbi__get16be(s); + return (z << 16) + stbi__get16be(s); } -#endif + #endif -#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) -// nothing -#else -static int stbi__get16le(stbi__context *s) -{ - int z = stbi__get8(s); - return z + (stbi__get8(s) << 8); + #if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) + // nothing + #else +static int stbi__get16le(stbi__context *s) { + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); } -#endif + #endif -#ifndef STBI_NO_BMP -static stbi__uint32 stbi__get32le(stbi__context *s) -{ - stbi__uint32 z = stbi__get16le(s); - return z + (stbi__get16le(s) << 16); + #ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context *s) { + stbi__uint32 z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); } -#endif + #endif -#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + #define STBI__BYTECAST(x) \ + ((stbi_uc)((x)&255)) // truncate int to byte without warnings -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && \ + defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && \ + defined(STBI_NO_PIC) && defined(STBI_NO_PNM) + // nothing + #else ////////////////////////////////////////////////////////////////////////////// // // generic converter from built-in img_n to req_comp @@ -1682,171 +1744,295 @@ static stbi__uint32 stbi__get32le(stbi__context *s) // assume data buffer is malloced, so malloc a new one and free that one // only failure mode is malloc failing -static stbi_uc stbi__compute_y(int r, int g, int b) -{ - return (stbi_uc) (((r*77) + (g*150) + (29*b)) >> 8); +static stbi_uc stbi__compute_y(int r, int g, int b) { + return (stbi_uc)(((r * 77) + (g * 150) + (29 * b)) >> 8); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else -static unsigned char *stbi__convert_format(unsigned char *data, int img_n, int req_comp, unsigned int x, unsigned int y) -{ - int i,j; - unsigned char *good; + #if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && \ + defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && \ + defined(STBI_NO_PNM) + // nothing + #else +static unsigned char *stbi__convert_format(unsigned char *data, int img_n, + int req_comp, unsigned int x, + unsigned int y) { + int i, j; + unsigned char *good; - if (req_comp == img_n) return data; - STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); - good = (unsigned char *) stbi__malloc_mad3(req_comp, x, y, 0); - if (good == NULL) { - STBI_FREE(data); - return stbi__errpuc("outofmem", "Out of memory"); - } + good = (unsigned char *)stbi__malloc_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } - for (j=0; j < (int) y; ++j) { - unsigned char *src = data + j * x * img_n ; - unsigned char *dest = good + j * x * req_comp; + for (j = 0; j < (int)y; ++j) { + unsigned char *src = data + j * x * img_n; + unsigned char *dest = good + j * x * req_comp; - #define STBI__COMBO(a,b) ((a)*8+(b)) - #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) - // convert source image with img_n components to one with req_comp components; - // avoid switch per pixel, so use switch per scanline and massive macros - switch (STBI__COMBO(img_n, req_comp)) { - STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=255; } break; - STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=255; } break; - STBI__CASE(2,1) { dest[0]=src[0]; } break; - STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; - STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=255; } break; - STBI__CASE(3,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; - STBI__CASE(3,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = 255; } break; - STBI__CASE(4,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; - STBI__CASE(4,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = src[3]; } break; - STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; - default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return stbi__errpuc("unsupported", "Unsupported format conversion"); + #define STBI__COMBO(a, b) ((a)*8 + (b)) + #define STBI__CASE(a, b) \ + case STBI__COMBO(a, b): \ + for (i = x - 1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp + // components; avoid switch per pixel, so use switch per scanline and + // massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { + dest[0] = src[0]; + dest[1] = 255; } - #undef STBI__CASE - } + break; + STBI__CASE(1, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(1, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = 255; + } + break; + STBI__CASE(2, 1) { + dest[0] = src[0]; + } + break; + STBI__CASE(2, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(2, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = src[1]; + } + break; + STBI__CASE(3, 4) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + dest[3] = 255; + } + break; + STBI__CASE(3, 1) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + } + break; + STBI__CASE(3, 2) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + dest[1] = 255; + } + break; + STBI__CASE(4, 1) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + } + break; + STBI__CASE(4, 2) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + dest[1] = src[3]; + } + break; + STBI__CASE(4, 3) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + } + break; + default: + STBI_ASSERT(0); + STBI_FREE(data); + STBI_FREE(good); + return stbi__errpuc("unsupported", "Unsupported format conversion"); + } + #undef STBI__CASE + } - STBI_FREE(data); - return good; + STBI_FREE(data); + return good; } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) -// nothing -#else -static stbi__uint16 stbi__compute_y_16(int r, int g, int b) -{ - return (stbi__uint16) (((r*77) + (g*150) + (29*b)) >> 8); + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) + // nothing + #else +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) { + return (stbi__uint16)(((r * 77) + (g * 150) + (29 * b)) >> 8); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) -// nothing -#else -static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, int req_comp, unsigned int x, unsigned int y) -{ - int i,j; - stbi__uint16 *good; + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) + // nothing + #else +static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, + int req_comp, unsigned int x, + unsigned int y) { + int i, j; + stbi__uint16 *good; - if (req_comp == img_n) return data; - STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); - good = (stbi__uint16 *) stbi__malloc(req_comp * x * y * 2); - if (good == NULL) { - STBI_FREE(data); - return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); - } + good = (stbi__uint16 *)stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16 *)stbi__errpuc("outofmem", "Out of memory"); + } - for (j=0; j < (int) y; ++j) { - stbi__uint16 *src = data + j * x * img_n ; - stbi__uint16 *dest = good + j * x * req_comp; + for (j = 0; j < (int)y; ++j) { + stbi__uint16 *src = data + j * x * img_n; + stbi__uint16 *dest = good + j * x * req_comp; - #define STBI__COMBO(a,b) ((a)*8+(b)) - #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) - // convert source image with img_n components to one with req_comp components; - // avoid switch per pixel, so use switch per scanline and massive macros - switch (STBI__COMBO(img_n, req_comp)) { - STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=0xffff; } break; - STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=0xffff; } break; - STBI__CASE(2,1) { dest[0]=src[0]; } break; - STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; - STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=0xffff; } break; - STBI__CASE(3,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; - STBI__CASE(3,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = 0xffff; } break; - STBI__CASE(4,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; - STBI__CASE(4,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = src[3]; } break; - STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; - default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return (stbi__uint16*) stbi__errpuc("unsupported", "Unsupported format conversion"); + #define STBI__COMBO(a, b) ((a)*8 + (b)) + #define STBI__CASE(a, b) \ + case STBI__COMBO(a, b): \ + for (i = x - 1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp + // components; avoid switch per pixel, so use switch per scanline and + // massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { + dest[0] = src[0]; + dest[1] = 0xffff; } - #undef STBI__CASE - } + break; + STBI__CASE(1, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(1, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = 0xffff; + } + break; + STBI__CASE(2, 1) { + dest[0] = src[0]; + } + break; + STBI__CASE(2, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(2, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = src[1]; + } + break; + STBI__CASE(3, 4) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + dest[3] = 0xffff; + } + break; + STBI__CASE(3, 1) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + } + break; + STBI__CASE(3, 2) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + dest[1] = 0xffff; + } + break; + STBI__CASE(4, 1) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + } + break; + STBI__CASE(4, 2) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + dest[1] = src[3]; + } + break; + STBI__CASE(4, 3) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + } + break; + default: + STBI_ASSERT(0); + STBI_FREE(data); + STBI_FREE(good); + return (stbi__uint16 *)stbi__errpuc("unsupported", + "Unsupported format conversion"); + } + #undef STBI__CASE + } - STBI_FREE(data); - return good; + STBI_FREE(data); + return good; } -#endif + #endif -#ifndef STBI_NO_LINEAR -static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) -{ - int i,k,n; - float *output; - if (!data) return NULL; - output = (float *) stbi__malloc_mad4(x, y, comp, sizeof(float), 0); - if (output == NULL) { STBI_FREE(data); return stbi__errpf("outofmem", "Out of memory"); } - // compute number of non-alpha components - if (comp & 1) n = comp; else n = comp-1; - for (i=0; i < x*y; ++i) { - for (k=0; k < n; ++k) { - output[i*comp + k] = (float) (pow(data[i*comp+k]/255.0f, stbi__l2h_gamma) * stbi__l2h_scale); - } - } - if (n < comp) { - for (i=0; i < x*y; ++i) { - output[i*comp + n] = data[i*comp + n]/255.0f; - } - } - STBI_FREE(data); - return output; + #ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) { + int i, k, n; + float *output; + if (!data) return NULL; + output = (float *)stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { + STBI_FREE(data); + return stbi__errpf("outofmem", "Out of memory"); + } + // compute number of non-alpha components + if (comp & 1) + n = comp; + else + n = comp - 1; + for (i = 0; i < x * y; ++i) { + for (k = 0; k < n; ++k) { + output[i * comp + k] = + (float)(pow(data[i * comp + k] / 255.0f, stbi__l2h_gamma) * + stbi__l2h_scale); + } + } + if (n < comp) { + for (i = 0; i < x * y; ++i) { + output[i * comp + n] = data[i * comp + n] / 255.0f; + } + } + STBI_FREE(data); + return output; } -#endif + #endif -#ifndef STBI_NO_HDR -#define stbi__float2int(x) ((int) (x)) -static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) -{ - int i,k,n; - stbi_uc *output; - if (!data) return NULL; - output = (stbi_uc *) stbi__malloc_mad3(x, y, comp, 0); - if (output == NULL) { STBI_FREE(data); return stbi__errpuc("outofmem", "Out of memory"); } - // compute number of non-alpha components - if (comp & 1) n = comp; else n = comp-1; - for (i=0; i < x*y; ++i) { - for (k=0; k < n; ++k) { - float z = (float) pow(data[i*comp+k]*stbi__h2l_scale_i, stbi__h2l_gamma_i) * 255 + 0.5f; - if (z < 0) z = 0; - if (z > 255) z = 255; - output[i*comp + k] = (stbi_uc) stbi__float2int(z); - } - if (k < comp) { - float z = data[i*comp+k] * 255 + 0.5f; - if (z < 0) z = 0; - if (z > 255) z = 255; - output[i*comp + k] = (stbi_uc) stbi__float2int(z); - } - } - STBI_FREE(data); - return output; + #ifndef STBI_NO_HDR + #define stbi__float2int(x) ((int)(x)) +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) { + int i, k, n; + stbi_uc *output; + if (!data) return NULL; + output = (stbi_uc *)stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } + // compute number of non-alpha components + if (comp & 1) + n = comp; + else + n = comp - 1; + for (i = 0; i < x * y; ++i) { + for (k = 0; k < n; ++k) { + float z = (float)pow(data[i * comp + k] * stbi__h2l_scale_i, + stbi__h2l_gamma_i) * + 255 + + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i * comp + k] = (stbi_uc)stbi__float2int(z); + } + if (k < comp) { + float z = data[i * comp + k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i * comp + k] = (stbi_uc)stbi__float2int(z); + } + } + STBI_FREE(data); + return output; } -#endif + #endif ////////////////////////////////////////////////////////////////////////////// // @@ -1869,2103 +2055,2259 @@ static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) // - some SIMD kernels for common paths on targets with SSE2/NEON // - uses a lot of intermediate memory, could cache poorly -#ifndef STBI_NO_JPEG + #ifndef STBI_NO_JPEG -// huffman decoding acceleration -#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + // huffman decoding acceleration + #define FAST_BITS 9 // larger handles more cases; smaller stomps less cache -typedef struct -{ - stbi_uc fast[1 << FAST_BITS]; - // weirdly, repacking this into AoS is a 10% speed loss, instead of a win - stbi__uint16 code[256]; - stbi_uc values[256]; - stbi_uc size[257]; - unsigned int maxcode[18]; - int delta[17]; // old 'firstsymbol' - old 'firstcode' +typedef struct { + stbi_uc fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + stbi__uint16 code[256]; + stbi_uc values[256]; + stbi_uc size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' } stbi__huffman; -typedef struct -{ - stbi__context *s; - stbi__huffman huff_dc[4]; - stbi__huffman huff_ac[4]; - stbi__uint16 dequant[4][64]; - stbi__int16 fast_ac[4][1 << FAST_BITS]; +typedef struct { + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; -// sizes for components, interleaved MCUs - int img_h_max, img_v_max; - int img_mcu_x, img_mcu_y; - int img_mcu_w, img_mcu_h; + // sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; -// definition of jpeg image component - struct - { - int id; - int h,v; - int tq; - int hd,ha; - int dc_pred; + // definition of jpeg image component + struct { + int id; + int h, v; + int tq; + int hd, ha; + int dc_pred; - int x,y,w2,h2; - stbi_uc *data; - void *raw_data, *raw_coeff; - stbi_uc *linebuf; - short *coeff; // progressive only - int coeff_w, coeff_h; // number of 8x8 coefficient blocks - } img_comp[4]; + int x, y, w2, h2; + stbi_uc *data; + void *raw_data, *raw_coeff; + stbi_uc *linebuf; + short *coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } img_comp[4]; - stbi__uint32 code_buffer; // jpeg entropy-coded buffer - int code_bits; // number of valid bits - unsigned char marker; // marker seen while filling entropy buffer - int nomore; // flag if we saw a marker so must stop + stbi__uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop - int progressive; - int spec_start; - int spec_end; - int succ_high; - int succ_low; - int eob_run; - int jfif; - int app14_color_transform; // Adobe APP14 tag - int rgb; + int progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; - int scan_n, order[4]; - int restart_interval, todo; + int scan_n, order[4]; + int restart_interval, todo; -// kernels - void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); - void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step); - stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs); + // kernels + void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, + const stbi_uc *pcb, const stbi_uc *pcr, int count, + int step); + stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs); } stbi__jpeg; -static int stbi__build_huffman(stbi__huffman *h, int *count) -{ - int i,j,k=0; - unsigned int code; - // build size list for each symbol (from JPEG spec) - for (i=0; i < 16; ++i) - for (j=0; j < count[i]; ++j) - h->size[k++] = (stbi_uc) (i+1); - h->size[k] = 0; +static int stbi__build_huffman(stbi__huffman *h, int *count) { + int i, j, k = 0; + unsigned int code; + // build size list for each symbol (from JPEG spec) + for (i = 0; i < 16; ++i) + for (j = 0; j < count[i]; ++j) + h->size[k++] = (stbi_uc)(i + 1); + h->size[k] = 0; - // compute actual symbols (from jpeg spec) - code = 0; - k = 0; - for(j=1; j <= 16; ++j) { - // compute delta to add to code to compute symbol id - h->delta[j] = k - code; - if (h->size[k] == j) { - while (h->size[k] == j) - h->code[k++] = (stbi__uint16) (code++); - if (code-1 >= (1u << j)) return stbi__err("bad code lengths","Corrupt JPEG"); - } - // compute largest code + 1 for this size, preshifted as needed later - h->maxcode[j] = code << (16-j); - code <<= 1; - } - h->maxcode[j] = 0xffffffff; + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for (j = 1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (stbi__uint16)(code++); + if (code - 1 >= (1u << j)) + return stbi__err("bad code lengths", "Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16 - j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; - // build non-spec acceleration table; 255 is flag for not-accelerated - memset(h->fast, 255, 1 << FAST_BITS); - for (i=0; i < k; ++i) { - int s = h->size[i]; - if (s <= FAST_BITS) { - int c = h->code[i] << (FAST_BITS-s); - int m = 1 << (FAST_BITS-s); - for (j=0; j < m; ++j) { - h->fast[c+j] = (stbi_uc) i; - } + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i = 0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS - s); + int m = 1 << (FAST_BITS - s); + for (j = 0; j < m; ++j) { + h->fast[c + j] = (stbi_uc)i; } - } - return 1; + } + } + return 1; } // build a table that decodes both magnitude and value of small ACs in // one go. -static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) -{ - int i; - for (i=0; i < (1 << FAST_BITS); ++i) { - stbi_uc fast = h->fast[i]; - fast_ac[i] = 0; - if (fast < 255) { - int rs = h->values[fast]; - int run = (rs >> 4) & 15; - int magbits = rs & 15; - int len = h->size[fast]; +static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) { + int i; + for (i = 0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; - if (magbits && len + magbits <= FAST_BITS) { - // magnitude code followed by receive_extend code - int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); - int m = 1 << (magbits - 1); - if (k < m) k += (~0U << magbits) + 1; - // if the result is small enough, we can fit it in fast_ac table - if (k >= -128 && k <= 127) - fast_ac[i] = (stbi__int16) ((k * 256) + (run * 16) + (len + magbits)); - } + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16)((k * 256) + (run * 16) + (len + magbits)); } - } + } + } } -static void stbi__grow_buffer_unsafe(stbi__jpeg *j) -{ - do { - unsigned int b = j->nomore ? 0 : stbi__get8(j->s); - if (b == 0xff) { - int c = stbi__get8(j->s); - while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes - if (c != 0) { - j->marker = (unsigned char) c; - j->nomore = 1; - return; - } +static void stbi__grow_buffer_unsafe(stbi__jpeg *j) { + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) + c = stbi__get8(j->s); // consume fill bytes + if (c != 0) { + j->marker = (unsigned char)c; + j->nomore = 1; + return; } - j->code_buffer |= b << (24 - j->code_bits); - j->code_bits += 8; - } while (j->code_bits <= 24); + } + j->code_buffer |= b << (24 - j->code_bits); + j->code_bits += 8; + } while (j->code_bits <= 24); } // (1 << n) - 1 -static const stbi__uint32 stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; +static const stbi__uint32 stbi__bmask[17] = { + 0, 1, 3, 7, 15, 31, 63, 127, 255, + 511, 1023, 2047, 4095, 8191, 16383, 32767, 65535}; // decode a jpeg huffman value from the bitstream -stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) -{ - unsigned int temp; - int c,k; +stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) { + unsigned int temp; + int c, k; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - // look at the top FAST_BITS and determine what symbol ID it is, - // if the code is <= FAST_BITS - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - k = h->fast[c]; - if (k < 255) { - int s = h->size[k]; - if (s > j->code_bits) - return -1; - j->code_buffer <<= s; - j->code_bits -= s; - return h->values[k]; - } + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } - // naive test is to shift the code_buffer down so k bits are - // valid, then test against maxcode. To speed this up, we've - // preshifted maxcode left so that it has (16-k) 0s at the - // end; in other words, regardless of the number of bits, it - // wants to be compared against something shifted to have 16; - // that way we don't need to shift inside the loop. - temp = j->code_buffer >> 16; - for (k=FAST_BITS+1 ; ; ++k) - if (temp < h->maxcode[k]) - break; - if (k == 17) { - // error! code not found - j->code_bits -= 16; - return -1; - } + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k = FAST_BITS + 1;; ++k) + if (temp < h->maxcode[k]) break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } - if (k > j->code_bits) - return -1; + if (k > j->code_bits) return -1; - // convert the huffman code to the symbol id - c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; - STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & + stbi__bmask[h->size[c]]) == h->code[c]); - // convert the id to a symbol - j->code_bits -= k; - j->code_buffer <<= k; - return h->values[c]; + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; } // bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); +stbi_inline static int stbi__extend_receive(stbi__jpeg *j, int n) { + unsigned int k; + int sgn; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); - sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB - k = stbi_lrot(j->code_buffer, n); - if (n < 0 || n >= (int) (sizeof(stbi__bmask)/sizeof(*stbi__bmask))) return 0; - j->code_buffer = k & ~stbi__bmask[n]; - k &= stbi__bmask[n]; - j->code_bits -= n; - return k + (stbi__jbias[n] & ~sgn); + sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB + k = stbi_lrot(j->code_buffer, n); + if (n < 0 || n >= (int)(sizeof(stbi__bmask) / sizeof(*stbi__bmask))) return 0; + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & ~sgn); } // get some unsigned bits -stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) -{ - unsigned int k; - if (j->code_bits < n) stbi__grow_buffer_unsafe(j); - k = stbi_lrot(j->code_buffer, n); - j->code_buffer = k & ~stbi__bmask[n]; - k &= stbi__bmask[n]; - j->code_bits -= n; - return k; +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) { + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; } -stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) -{ - unsigned int k; - if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); - k = j->code_buffer; - j->code_buffer <<= 1; - --j->code_bits; - return k & 0x80000000; +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) { + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; } // given a value that's at position X in the zigzag stream, // where does it appear in the 8x8 matrix coded as row-major? -static const stbi_uc stbi__jpeg_dezigzag[64+15] = -{ - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 32, 25, 18, 11, 4, 5, - 12, 19, 26, 33, 40, 48, 41, 34, - 27, 20, 13, 6, 7, 14, 21, 28, - 35, 42, 49, 56, 57, 50, 43, 36, - 29, 22, 15, 23, 30, 37, 44, 51, - 58, 59, 52, 45, 38, 31, 39, 46, - 53, 60, 61, 54, 47, 55, 62, 63, - // let corrupt input sample past end - 63, 63, 63, 63, 63, 63, 63, 63, - 63, 63, 63, 63, 63, 63, 63 -}; +static const stbi_uc stbi__jpeg_dezigzag[64 + 15] = { + 0, 1, 8, 16, 9, 2, 3, 10, 17, 24, 32, 25, 18, 11, 4, 5, 12, 19, 26, 33, 40, + 48, 41, 34, 27, 20, 13, 6, 7, 14, 21, 28, 35, 42, 49, 56, 57, 50, 43, 36, + 29, 22, 15, 23, 30, 37, 44, 51, 58, 59, 52, 45, 38, 31, 39, 46, 53, 60, 61, + 54, 47, 55, 62, 63, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63}; // decode one 64-entry block-- -static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, stbi__int16 *fac, int b, stbi__uint16 *dequant) -{ - int diff,dc,k; - int t; +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], + stbi__huffman *hdc, stbi__huffman *hac, + stbi__int16 *fac, int b, + stbi__uint16 *dequant) { + int diff, dc, k; + int t; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - t = stbi__jpeg_huff_decode(j, hdc); - if (t < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); - // 0 all the ac values now so we can do it 32-bits at a time - memset(data,0,64*sizeof(data[0])); + // 0 all the ac values now so we can do it 32-bits at a time + memset(data, 0, 64 * sizeof(data[0])); - diff = t ? stbi__extend_receive(j, t) : 0; - dc = j->img_comp[b].dc_pred + diff; - j->img_comp[b].dc_pred = dc; - data[0] = (short) (dc * dequant[0]); + diff = t ? stbi__extend_receive(j, t) : 0; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc * dequant[0]); - // decode AC components, see JPEG spec - k = 1; - do { - unsigned int zig; - int c,r,s; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - r = fac[c]; - if (r) { // fast-AC path - k += (r >> 4) & 15; // run - s = r & 15; // combined length - j->code_buffer <<= s; - j->code_bits -= s; - // decode into unzigzag'd location - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) ((r >> 8) * dequant[zig]); + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) * dequant[zig]); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; } else { - int rs = stbi__jpeg_huff_decode(j, hac); - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (rs != 0xf0) break; // end block - k += 16; - } else { - k += r; - // decode into unzigzag'd location - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) (stbi__extend_receive(j,s) * dequant[zig]); - } + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) * dequant[zig]); } - } while (k < 64); - return 1; + } + } while (k < 64); + return 1; } -static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], stbi__huffman *hdc, int b) -{ - int diff,dc; - int t; - if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], + stbi__huffman *hdc, int b) { + int diff, dc; + int t; + if (j->spec_end != 0) + return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - if (j->succ_high == 0) { - // first scan for DC coefficient, must be first - memset(data,0,64*sizeof(data[0])); // 0 all the ac values now - t = stbi__jpeg_huff_decode(j, hdc); - if (t == -1) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - diff = t ? stbi__extend_receive(j, t) : 0; + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data, 0, 64 * sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + if (t == -1) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + diff = t ? stbi__extend_receive(j, t) : 0; - dc = j->img_comp[b].dc_pred + diff; - j->img_comp[b].dc_pred = dc; - data[0] = (short) (dc << j->succ_low); - } else { - // refinement scan for DC coefficient - if (stbi__jpeg_get_bit(j)) - data[0] += (short) (1 << j->succ_low); - } - return 1; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc << j->succ_low); + } else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) data[0] += (short)(1 << j->succ_low); + } + return 1; } // @OPTIMIZE: store non-zigzagged during the decode passes, // and only de-zigzag when dequantizing -static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], stbi__huffman *hac, stbi__int16 *fac) -{ - int k; - if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], + stbi__huffman *hac, + stbi__int16 *fac) { + int k; + if (j->spec_start == 0) + return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - if (j->succ_high == 0) { - int shift = j->succ_low; + if (j->succ_high == 0) { + int shift = j->succ_low; - if (j->eob_run) { - --j->eob_run; - return 1; + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) << shift); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) << shift); + } } + } while (k <= j->spec_end); + } else { + // refinement scan for these AC coefficients + short bit = (short)(1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } + } else { k = j->spec_start; do { - unsigned int zig; - int c,r,s; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - r = fac[c]; - if (r) { // fast-AC path - k += (r >> 4) & 15; // run - s = r & 15; // combined length - j->code_buffer <<= s; - j->code_bits -= s; - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) ((r >> 8) << shift); - } else { - int rs = stbi__jpeg_huff_decode(j, hac); - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (r < 15) { - j->eob_run = (1 << r); - if (r) - j->eob_run += stbi__jpeg_get_bits(j, r); - --j->eob_run; - break; - } - k += 16; - } else { - k += r; - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) (stbi__extend_receive(j,s) << shift); + int r, s; + int rs = stbi__jpeg_huff_decode( + j, hac); // @OPTIMIZE see if we can use the fast path here, + // advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short *p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } else { + if (r == 0) { + *p = (short)s; + break; } - } + --r; + } + } } while (k <= j->spec_end); - } else { - // refinement scan for these AC coefficients - - short bit = (short) (1 << j->succ_low); - - if (j->eob_run) { - --j->eob_run; - for (k = j->spec_start; k <= j->spec_end; ++k) { - short *p = &data[stbi__jpeg_dezigzag[k]]; - if (*p != 0) - if (stbi__jpeg_get_bit(j)) - if ((*p & bit)==0) { - if (*p > 0) - *p += bit; - else - *p -= bit; - } - } - } else { - k = j->spec_start; - do { - int r,s; - int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (r < 15) { - j->eob_run = (1 << r) - 1; - if (r) - j->eob_run += stbi__jpeg_get_bits(j, r); - r = 64; // force end of block - } else { - // r=15 s=0 should write 16 0s, so we just do - // a run of 15 0s and then write s (which is 0), - // so we don't have to do anything special here - } - } else { - if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); - // sign bit - if (stbi__jpeg_get_bit(j)) - s = bit; - else - s = -bit; - } - - // advance by r - while (k <= j->spec_end) { - short *p = &data[stbi__jpeg_dezigzag[k++]]; - if (*p != 0) { - if (stbi__jpeg_get_bit(j)) - if ((*p & bit)==0) { - if (*p > 0) - *p += bit; - else - *p -= bit; - } - } else { - if (r == 0) { - *p = (short) s; - break; - } - --r; - } - } - } while (k <= j->spec_end); - } - } - return 1; + } + } + return 1; } // take a -128..127 value and stbi__clamp it and convert to 0..255 -stbi_inline static stbi_uc stbi__clamp(int x) -{ - // trick to use a single test to catch both cases - if ((unsigned int) x > 255) { - if (x < 0) return 0; - if (x > 255) return 255; - } - return (stbi_uc) x; +stbi_inline static stbi_uc stbi__clamp(int x) { + // trick to use a single test to catch both cases + if ((unsigned int)x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (stbi_uc)x; } -#define stbi__f2f(x) ((int) (((x) * 4096 + 0.5))) -#define stbi__fsh(x) ((x) * 4096) + #define stbi__f2f(x) ((int)(((x)*4096 + 0.5))) + #define stbi__fsh(x) ((x)*4096) -// derived from jidctint -- DCT_ISLOW -#define STBI__IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7) \ - int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \ - p2 = s2; \ - p3 = s6; \ - p1 = (p2+p3) * stbi__f2f(0.5411961f); \ - t2 = p1 + p3*stbi__f2f(-1.847759065f); \ - t3 = p1 + p2*stbi__f2f( 0.765366865f); \ - p2 = s0; \ - p3 = s4; \ - t0 = stbi__fsh(p2+p3); \ - t1 = stbi__fsh(p2-p3); \ - x0 = t0+t3; \ - x3 = t0-t3; \ - x1 = t1+t2; \ - x2 = t1-t2; \ - t0 = s7; \ - t1 = s5; \ - t2 = s3; \ - t3 = s1; \ - p3 = t0+t2; \ - p4 = t1+t3; \ - p1 = t0+t3; \ - p2 = t1+t2; \ - p5 = (p3+p4)*stbi__f2f( 1.175875602f); \ - t0 = t0*stbi__f2f( 0.298631336f); \ - t1 = t1*stbi__f2f( 2.053119869f); \ - t2 = t2*stbi__f2f( 3.072711026f); \ - t3 = t3*stbi__f2f( 1.501321110f); \ - p1 = p5 + p1*stbi__f2f(-0.899976223f); \ - p2 = p5 + p2*stbi__f2f(-2.562915447f); \ - p3 = p3*stbi__f2f(-1.961570560f); \ - p4 = p4*stbi__f2f(-0.390180644f); \ - t3 += p1+p4; \ - t2 += p2+p3; \ - t1 += p2+p4; \ - t0 += p1+p3; + // derived from jidctint -- DCT_ISLOW + #define STBI__IDCT_1D(s0, s1, s2, s3, s4, s5, s6, s7) \ + int t0, t1, t2, t3, p1, p2, p3, p4, p5, x0, x1, x2, x3; \ + p2 = s2; \ + p3 = s6; \ + p1 = (p2 + p3) * stbi__f2f(0.5411961f); \ + t2 = p1 + p3 * stbi__f2f(-1.847759065f); \ + t3 = p1 + p2 * stbi__f2f(0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = stbi__fsh(p2 + p3); \ + t1 = stbi__fsh(p2 - p3); \ + x0 = t0 + t3; \ + x3 = t0 - t3; \ + x1 = t1 + t2; \ + x2 = t1 - t2; \ + t0 = s7; \ + t1 = s5; \ + t2 = s3; \ + t3 = s1; \ + p3 = t0 + t2; \ + p4 = t1 + t3; \ + p1 = t0 + t3; \ + p2 = t1 + t2; \ + p5 = (p3 + p4) * stbi__f2f(1.175875602f); \ + t0 = t0 * stbi__f2f(0.298631336f); \ + t1 = t1 * stbi__f2f(2.053119869f); \ + t2 = t2 * stbi__f2f(3.072711026f); \ + t3 = t3 * stbi__f2f(1.501321110f); \ + p1 = p5 + p1 * stbi__f2f(-0.899976223f); \ + p2 = p5 + p2 * stbi__f2f(-2.562915447f); \ + p3 = p3 * stbi__f2f(-1.961570560f); \ + p4 = p4 * stbi__f2f(-0.390180644f); \ + t3 += p1 + p4; \ + t2 += p2 + p3; \ + t1 += p2 + p4; \ + t0 += p1 + p3; -static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) -{ - int i,val[64],*v=val; - stbi_uc *o; - short *d = data; +static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) { + int i, val[64], *v = val; + stbi_uc *o; + short *d = data; - // columns - for (i=0; i < 8; ++i,++d, ++v) { - // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing - if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 - && d[40]==0 && d[48]==0 && d[56]==0) { - // no shortcut 0 seconds - // (1|2|3|4|5|6|7)==0 0 seconds - // all separate -0.047 seconds - // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds - int dcterm = d[0]*4; - v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; - } else { - STBI__IDCT_1D(d[ 0],d[ 8],d[16],d[24],d[32],d[40],d[48],d[56]) - // constants scaled things up by 1<<12; let's bring them back - // down, but keep 2 extra bits of precision - x0 += 512; x1 += 512; x2 += 512; x3 += 512; - v[ 0] = (x0+t3) >> 10; - v[56] = (x0-t3) >> 10; - v[ 8] = (x1+t2) >> 10; - v[48] = (x1-t2) >> 10; - v[16] = (x2+t1) >> 10; - v[40] = (x2-t1) >> 10; - v[24] = (x3+t0) >> 10; - v[32] = (x3-t0) >> 10; - } - } + // columns + for (i = 0; i < 8; ++i, ++d, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[8] == 0 && d[16] == 0 && d[24] == 0 && d[32] == 0 && d[40] == 0 && + d[48] == 0 && d[56] == 0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0] * 4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[0], d[8], d[16], d[24], d[32], d[40], d[48], d[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; + x1 += 512; + x2 += 512; + x3 += 512; + v[0] = (x0 + t3) >> 10; + v[56] = (x0 - t3) >> 10; + v[8] = (x1 + t2) >> 10; + v[48] = (x1 - t2) >> 10; + v[16] = (x2 + t1) >> 10; + v[40] = (x2 - t1) >> 10; + v[24] = (x3 + t0) >> 10; + v[32] = (x3 - t0) >> 10; + } + } - for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { - // no fast case since the first 1D IDCT spread components out - STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) - // constants scaled things up by 1<<12, plus we had 1<<2 from first - // loop, plus horizontal and vertical each scale by sqrt(8) so together - // we've got an extra 1<<3, so 1<<17 total we need to remove. - // so we want to round that, which means adding 0.5 * 1<<17, - // aka 65536. Also, we'll end up with -128 to 127 that we want - // to encode as 0..255 by adding 128, so we'll add that before the shift - x0 += 65536 + (128<<17); - x1 += 65536 + (128<<17); - x2 += 65536 + (128<<17); - x3 += 65536 + (128<<17); - // tried computing the shifts into temps, or'ing the temps to see - // if any were out of range, but that was slower - o[0] = stbi__clamp((x0+t3) >> 17); - o[7] = stbi__clamp((x0-t3) >> 17); - o[1] = stbi__clamp((x1+t2) >> 17); - o[6] = stbi__clamp((x1-t2) >> 17); - o[2] = stbi__clamp((x2+t1) >> 17); - o[5] = stbi__clamp((x2-t1) >> 17); - o[3] = stbi__clamp((x3+t0) >> 17); - o[4] = stbi__clamp((x3-t0) >> 17); - } + for (i = 0, v = val, o = out; i < 8; ++i, v += 8, o += out_stride) { + // no fast case since the first 1D IDCT spread components out + STBI__IDCT_1D(v[0], v[1], v[2], v[3], v[4], v[5], v[6], v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128 << 17); + x1 += 65536 + (128 << 17); + x2 += 65536 + (128 << 17); + x3 += 65536 + (128 << 17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = stbi__clamp((x0 + t3) >> 17); + o[7] = stbi__clamp((x0 - t3) >> 17); + o[1] = stbi__clamp((x1 + t2) >> 17); + o[6] = stbi__clamp((x1 - t2) >> 17); + o[2] = stbi__clamp((x2 + t1) >> 17); + o[5] = stbi__clamp((x2 - t1) >> 17); + o[3] = stbi__clamp((x3 + t0) >> 17); + o[4] = stbi__clamp((x3 - t0) >> 17); + } } -#ifdef STBI_SSE2 + #ifdef STBI_SSE2 // sse2 integer IDCT. not the fastest possible implementation but it // produces bit-identical results to the generic C version so it's // fully "transparent". -static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) -{ - // This is constructed to match our regular (generic) integer IDCT exactly. - __m128i row0, row1, row2, row3, row4, row5, row6, row7; - __m128i tmp; +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) { + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; - // dot product constant: even elems=x, odd elems=y - #define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + // dot product constant: even elems=x, odd elems=y + #define dct_const(x, y) \ + _mm_setr_epi16((x), (y), (x), (y), (x), (y), (x), (y)) - // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) - // out(1) = c1[even]*x + c1[odd]*y - #define dct_rot(out0,out1, x,y,c0,c1) \ - __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ - __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ - __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ - __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ - __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ - __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) + // out(1) = c1[even]*x + c1[odd]*y + #define dct_rot(out0, out1, x, y, c0, c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x), (y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x), (y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) - // out = in << 12 (in 16-bit, out 32-bit) - #define dct_widen(out, in) \ - __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ - __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + // out = in << 12 (in 16-bit, out 32-bit) + #define dct_widen(out, in) \ + __m128i out##_l = \ + _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = \ + _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) - // wide add - #define dct_wadd(out, a, b) \ - __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ - __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + // wide add + #define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) - // wide sub - #define dct_wsub(out, a, b) \ - __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ - __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + // wide sub + #define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) - // butterfly a/b, add bias, then shift by "s" and pack - #define dct_bfly32o(out0, out1, a,b,bias,s) \ - { \ - __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ - __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ - dct_wadd(sum, abiased, b); \ - dct_wsub(dif, abiased, b); \ - out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ - out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ - } + // butterfly a/b, add bias, then shift by "s" and pack + #define dct_bfly32o(out0, out1, a, b, bias, s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), \ + _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), \ + _mm_srai_epi32(dif_h, s)); \ + } - // 8-bit interleave step (for transposes) - #define dct_interleave8(a, b) \ - tmp = a; \ - a = _mm_unpacklo_epi8(a, b); \ - b = _mm_unpackhi_epi8(tmp, b) + // 8-bit interleave step (for transposes) + #define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) - // 16-bit interleave step (for transposes) - #define dct_interleave16(a, b) \ - tmp = a; \ - a = _mm_unpacklo_epi16(a, b); \ - b = _mm_unpackhi_epi16(tmp, b) + // 16-bit interleave step (for transposes) + #define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) - #define dct_pass(bias,shift) \ - { \ - /* even part */ \ - dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ - __m128i sum04 = _mm_add_epi16(row0, row4); \ - __m128i dif04 = _mm_sub_epi16(row0, row4); \ - dct_widen(t0e, sum04); \ - dct_widen(t1e, dif04); \ - dct_wadd(x0, t0e, t3e); \ - dct_wsub(x3, t0e, t3e); \ - dct_wadd(x1, t1e, t2e); \ - dct_wsub(x2, t1e, t2e); \ - /* odd part */ \ - dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ - dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ - __m128i sum17 = _mm_add_epi16(row1, row7); \ - __m128i sum35 = _mm_add_epi16(row3, row5); \ - dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ - dct_wadd(x4, y0o, y4o); \ - dct_wadd(x5, y1o, y5o); \ - dct_wadd(x6, y2o, y5o); \ - dct_wadd(x7, y3o, y4o); \ - dct_bfly32o(row0,row7, x0,x7,bias,shift); \ - dct_bfly32o(row1,row6, x1,x6,bias,shift); \ - dct_bfly32o(row2,row5, x2,x5,bias,shift); \ - dct_bfly32o(row3,row4, x3,x4,bias,shift); \ - } + #define dct_pass(bias, shift) \ + { \ + /* even part */ \ + dct_rot(t2e, t3e, row2, row6, rot0_0, rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o, y2o, row7, row3, rot2_0, rot2_1); \ + dct_rot(y1o, y3o, row5, row1, rot3_0, rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o, y5o, sum17, sum35, rot1_0, rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0, row7, x0, x7, bias, shift); \ + dct_bfly32o(row1, row6, x1, x6, bias, shift); \ + dct_bfly32o(row2, row5, x2, x5, bias, shift); \ + dct_bfly32o(row3, row4, x3, x4, bias, shift); \ + } - __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); - __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f( 0.765366865f), stbi__f2f(0.5411961f)); - __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); - __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); - __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f( 0.298631336f), stbi__f2f(-1.961570560f)); - __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f( 3.072711026f)); - __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f( 2.053119869f), stbi__f2f(-0.390180644f)); - __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f( 1.501321110f)); + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), + stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f(0.765366865f), + stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), + stbi__f2f(1.175875602f)); + __m128i rot1_1 = + dct_const(stbi__f2f(1.175875602f), + stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f(0.298631336f), + stbi__f2f(-1.961570560f)); + __m128i rot2_1 = + dct_const(stbi__f2f(-1.961570560f), + stbi__f2f(-1.961570560f) + stbi__f2f(3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f(2.053119869f), + stbi__f2f(-0.390180644f)); + __m128i rot3_1 = + dct_const(stbi__f2f(-0.390180644f), + stbi__f2f(-0.390180644f) + stbi__f2f(1.501321110f)); - // rounding biases in column/row passes, see stbi__idct_block for explanation. - __m128i bias_0 = _mm_set1_epi32(512); - __m128i bias_1 = _mm_set1_epi32(65536 + (128<<17)); + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128 << 17)); - // load - row0 = _mm_load_si128((const __m128i *) (data + 0*8)); - row1 = _mm_load_si128((const __m128i *) (data + 1*8)); - row2 = _mm_load_si128((const __m128i *) (data + 2*8)); - row3 = _mm_load_si128((const __m128i *) (data + 3*8)); - row4 = _mm_load_si128((const __m128i *) (data + 4*8)); - row5 = _mm_load_si128((const __m128i *) (data + 5*8)); - row6 = _mm_load_si128((const __m128i *) (data + 6*8)); - row7 = _mm_load_si128((const __m128i *) (data + 7*8)); + // load + row0 = _mm_load_si128((const __m128i *)(data + 0 * 8)); + row1 = _mm_load_si128((const __m128i *)(data + 1 * 8)); + row2 = _mm_load_si128((const __m128i *)(data + 2 * 8)); + row3 = _mm_load_si128((const __m128i *)(data + 3 * 8)); + row4 = _mm_load_si128((const __m128i *)(data + 4 * 8)); + row5 = _mm_load_si128((const __m128i *)(data + 5 * 8)); + row6 = _mm_load_si128((const __m128i *)(data + 6 * 8)); + row7 = _mm_load_si128((const __m128i *)(data + 7 * 8)); - // column pass - dct_pass(bias_0, 10); + // column pass + dct_pass(bias_0, 10); - { - // 16bit 8x8 transpose pass 1 - dct_interleave16(row0, row4); - dct_interleave16(row1, row5); - dct_interleave16(row2, row6); - dct_interleave16(row3, row7); + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); - // transpose pass 2 - dct_interleave16(row0, row2); - dct_interleave16(row1, row3); - dct_interleave16(row4, row6); - dct_interleave16(row5, row7); + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); - // transpose pass 3 - dct_interleave16(row0, row1); - dct_interleave16(row2, row3); - dct_interleave16(row4, row5); - dct_interleave16(row6, row7); - } + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } - // row pass - dct_pass(bias_1, 17); + // row pass + dct_pass(bias_1, 17); - { - // pack - __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 - __m128i p1 = _mm_packus_epi16(row2, row3); - __m128i p2 = _mm_packus_epi16(row4, row5); - __m128i p3 = _mm_packus_epi16(row6, row7); + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); - // 8bit 8x8 transpose pass 1 - dct_interleave8(p0, p2); // a0e0a1e1... - dct_interleave8(p1, p3); // c0g0c1g1... + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... - // transpose pass 2 - dct_interleave8(p0, p1); // a0c0e0g0... - dct_interleave8(p2, p3); // b0d0f0h0... + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... - // transpose pass 3 - dct_interleave8(p0, p2); // a0b0c0d0... - dct_interleave8(p1, p3); // a4b4c4d4... + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... - // store - _mm_storel_epi64((__m128i *) out, p0); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p2); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p1); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p3); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p3, 0x4e)); - } + // store + _mm_storel_epi64((__m128i *)out, p0); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p0, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p2); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p2, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p1); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p1, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p3); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p3, 0x4e)); + } -#undef dct_const -#undef dct_rot -#undef dct_widen -#undef dct_wadd -#undef dct_wsub -#undef dct_bfly32o -#undef dct_interleave8 -#undef dct_interleave16 -#undef dct_pass + #undef dct_const + #undef dct_rot + #undef dct_widen + #undef dct_wadd + #undef dct_wsub + #undef dct_bfly32o + #undef dct_interleave8 + #undef dct_interleave16 + #undef dct_pass } -#endif // STBI_SSE2 + #endif // STBI_SSE2 -#ifdef STBI_NEON + #ifdef STBI_NEON // NEON integer IDCT. should produce bit-identical // results to the generic C version. -static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) -{ - int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) { + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; - int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); - int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); - int16x4_t rot0_2 = vdup_n_s16(stbi__f2f( 0.765366865f)); - int16x4_t rot1_0 = vdup_n_s16(stbi__f2f( 1.175875602f)); - int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); - int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); - int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); - int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); - int16x4_t rot3_0 = vdup_n_s16(stbi__f2f( 0.298631336f)); - int16x4_t rot3_1 = vdup_n_s16(stbi__f2f( 2.053119869f)); - int16x4_t rot3_2 = vdup_n_s16(stbi__f2f( 3.072711026f)); - int16x4_t rot3_3 = vdup_n_s16(stbi__f2f( 1.501321110f)); + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f(0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f(1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f(0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f(2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f(3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f(1.501321110f)); -#define dct_long_mul(out, inq, coeff) \ - int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ - int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + #define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) -#define dct_long_mac(out, acc, inq, coeff) \ - int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ - int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + #define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) -#define dct_widen(out, inq) \ - int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ - int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + #define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) -// wide add -#define dct_wadd(out, a, b) \ - int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ - int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + // wide add + #define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) -// wide sub -#define dct_wsub(out, a, b) \ - int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ - int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + // wide sub + #define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) -// butterfly a/b, then shift using "shiftop" by "s" and pack -#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ - { \ - dct_wadd(sum, a, b); \ - dct_wsub(dif, a, b); \ - out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ - out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ - } + // butterfly a/b, then shift using "shiftop" by "s" and pack + #define dct_bfly32o(out0, out1, a, b, shiftop, s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } -#define dct_pass(shiftop, shift) \ - { \ - /* even part */ \ - int16x8_t sum26 = vaddq_s16(row2, row6); \ - dct_long_mul(p1e, sum26, rot0_0); \ - dct_long_mac(t2e, p1e, row6, rot0_1); \ - dct_long_mac(t3e, p1e, row2, rot0_2); \ - int16x8_t sum04 = vaddq_s16(row0, row4); \ - int16x8_t dif04 = vsubq_s16(row0, row4); \ - dct_widen(t0e, sum04); \ - dct_widen(t1e, dif04); \ - dct_wadd(x0, t0e, t3e); \ - dct_wsub(x3, t0e, t3e); \ - dct_wadd(x1, t1e, t2e); \ - dct_wsub(x2, t1e, t2e); \ - /* odd part */ \ - int16x8_t sum15 = vaddq_s16(row1, row5); \ - int16x8_t sum17 = vaddq_s16(row1, row7); \ - int16x8_t sum35 = vaddq_s16(row3, row5); \ - int16x8_t sum37 = vaddq_s16(row3, row7); \ - int16x8_t sumodd = vaddq_s16(sum17, sum35); \ - dct_long_mul(p5o, sumodd, rot1_0); \ - dct_long_mac(p1o, p5o, sum17, rot1_1); \ - dct_long_mac(p2o, p5o, sum35, rot1_2); \ - dct_long_mul(p3o, sum37, rot2_0); \ - dct_long_mul(p4o, sum15, rot2_1); \ - dct_wadd(sump13o, p1o, p3o); \ - dct_wadd(sump24o, p2o, p4o); \ - dct_wadd(sump23o, p2o, p3o); \ - dct_wadd(sump14o, p1o, p4o); \ - dct_long_mac(x4, sump13o, row7, rot3_0); \ - dct_long_mac(x5, sump24o, row5, rot3_1); \ - dct_long_mac(x6, sump23o, row3, rot3_2); \ - dct_long_mac(x7, sump14o, row1, rot3_3); \ - dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ - dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ - dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ - dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ - } + #define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0, row7, x0, x7, shiftop, shift); \ + dct_bfly32o(row1, row6, x1, x6, shiftop, shift); \ + dct_bfly32o(row2, row5, x2, x5, shiftop, shift); \ + dct_bfly32o(row3, row4, x3, x4, shiftop, shift); \ + } - // load - row0 = vld1q_s16(data + 0*8); - row1 = vld1q_s16(data + 1*8); - row2 = vld1q_s16(data + 2*8); - row3 = vld1q_s16(data + 3*8); - row4 = vld1q_s16(data + 4*8); - row5 = vld1q_s16(data + 5*8); - row6 = vld1q_s16(data + 6*8); - row7 = vld1q_s16(data + 7*8); + // load + row0 = vld1q_s16(data + 0 * 8); + row1 = vld1q_s16(data + 1 * 8); + row2 = vld1q_s16(data + 2 * 8); + row3 = vld1q_s16(data + 3 * 8); + row4 = vld1q_s16(data + 4 * 8); + row5 = vld1q_s16(data + 5 * 8); + row6 = vld1q_s16(data + 6 * 8); + row7 = vld1q_s16(data + 7 * 8); - // add DC bias - row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); - // column pass - dct_pass(vrshrn_n_s32, 10); + // column pass + dct_pass(vrshrn_n_s32, 10); - // 16bit 8x8 transpose - { -// these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. -// whether compilers actually get this is another story, sadly. -#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } -#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } -#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + // 16bit 8x8 transpose + { + // these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. + // whether compilers actually get this is another story, sadly. + #define dct_trn16(x, y) \ + { \ + int16x8x2_t t = vtrnq_s16(x, y); \ + x = t.val[0]; \ + y = t.val[1]; \ + } + #define dct_trn32(x, y) \ + { \ + int32x4x2_t t = \ + vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); \ + x = vreinterpretq_s16_s32(t.val[0]); \ + y = vreinterpretq_s16_s32(t.val[1]); \ + } + #define dct_trn64(x, y) \ + { \ + int16x8_t x0 = x; \ + int16x8_t y0 = y; \ + x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); \ + y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); \ + } - // pass 1 - dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 - dct_trn16(row2, row3); - dct_trn16(row4, row5); - dct_trn16(row6, row7); + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); - // pass 2 - dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 - dct_trn32(row1, row3); - dct_trn32(row4, row6); - dct_trn32(row5, row7); + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); - // pass 3 - dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 - dct_trn64(row1, row5); - dct_trn64(row2, row6); - dct_trn64(row3, row7); + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); -#undef dct_trn16 -#undef dct_trn32 -#undef dct_trn64 - } + #undef dct_trn16 + #undef dct_trn32 + #undef dct_trn64 + } - // row pass - // vrshrn_n_s32 only supports shifts up to 16, we need - // 17. so do a non-rounding shift of 16 first then follow - // up with a rounding shift by 1. - dct_pass(vshrn_n_s32, 16); + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); - { - // pack and round - uint8x8_t p0 = vqrshrun_n_s16(row0, 1); - uint8x8_t p1 = vqrshrun_n_s16(row1, 1); - uint8x8_t p2 = vqrshrun_n_s16(row2, 1); - uint8x8_t p3 = vqrshrun_n_s16(row3, 1); - uint8x8_t p4 = vqrshrun_n_s16(row4, 1); - uint8x8_t p5 = vqrshrun_n_s16(row5, 1); - uint8x8_t p6 = vqrshrun_n_s16(row6, 1); - uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); - // again, these can translate into one instruction, but often don't. -#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } -#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } -#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + // again, these can translate into one instruction, but often don't. + #define dct_trn8_8(x, y) \ + { \ + uint8x8x2_t t = vtrn_u8(x, y); \ + x = t.val[0]; \ + y = t.val[1]; \ + } + #define dct_trn8_16(x, y) \ + { \ + uint16x4x2_t t = \ + vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); \ + x = vreinterpret_u8_u16(t.val[0]); \ + y = vreinterpret_u8_u16(t.val[1]); \ + } + #define dct_trn8_32(x, y) \ + { \ + uint32x2x2_t t = \ + vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); \ + x = vreinterpret_u8_u32(t.val[0]); \ + y = vreinterpret_u8_u32(t.val[1]); \ + } - // sadly can't use interleaved stores here since we only write - // 8 bytes to each scan line! + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! - // 8x8 8-bit transpose pass 1 - dct_trn8_8(p0, p1); - dct_trn8_8(p2, p3); - dct_trn8_8(p4, p5); - dct_trn8_8(p6, p7); + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); - // pass 2 - dct_trn8_16(p0, p2); - dct_trn8_16(p1, p3); - dct_trn8_16(p4, p6); - dct_trn8_16(p5, p7); + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); - // pass 3 - dct_trn8_32(p0, p4); - dct_trn8_32(p1, p5); - dct_trn8_32(p2, p6); - dct_trn8_32(p3, p7); + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); - // store - vst1_u8(out, p0); out += out_stride; - vst1_u8(out, p1); out += out_stride; - vst1_u8(out, p2); out += out_stride; - vst1_u8(out, p3); out += out_stride; - vst1_u8(out, p4); out += out_stride; - vst1_u8(out, p5); out += out_stride; - vst1_u8(out, p6); out += out_stride; - vst1_u8(out, p7); + // store + vst1_u8(out, p0); + out += out_stride; + vst1_u8(out, p1); + out += out_stride; + vst1_u8(out, p2); + out += out_stride; + vst1_u8(out, p3); + out += out_stride; + vst1_u8(out, p4); + out += out_stride; + vst1_u8(out, p5); + out += out_stride; + vst1_u8(out, p6); + out += out_stride; + vst1_u8(out, p7); -#undef dct_trn8_8 -#undef dct_trn8_16 -#undef dct_trn8_32 - } + #undef dct_trn8_8 + #undef dct_trn8_16 + #undef dct_trn8_32 + } -#undef dct_long_mul -#undef dct_long_mac -#undef dct_widen -#undef dct_wadd -#undef dct_wsub -#undef dct_bfly32o -#undef dct_pass + #undef dct_long_mul + #undef dct_long_mac + #undef dct_widen + #undef dct_wadd + #undef dct_wsub + #undef dct_bfly32o + #undef dct_pass } -#endif // STBI_NEON + #endif // STBI_NEON -#define STBI__MARKER_none 0xff + #define STBI__MARKER_none 0xff // if there's a pending marker from the entropy stream, return that // otherwise, fetch from the stream and get a marker. if there's no // marker, return 0xff, which is never a valid marker value -static stbi_uc stbi__get_marker(stbi__jpeg *j) -{ - stbi_uc x; - if (j->marker != STBI__MARKER_none) { x = j->marker; j->marker = STBI__MARKER_none; return x; } - x = stbi__get8(j->s); - if (x != 0xff) return STBI__MARKER_none; - while (x == 0xff) - x = stbi__get8(j->s); // consume repeated 0xff fill bytes - return x; +static stbi_uc stbi__get_marker(stbi__jpeg *j) { + stbi_uc x; + if (j->marker != STBI__MARKER_none) { + x = j->marker; + j->marker = STBI__MARKER_none; + return x; + } + x = stbi__get8(j->s); + if (x != 0xff) return STBI__MARKER_none; + while (x == 0xff) + x = stbi__get8(j->s); // consume repeated 0xff fill bytes + return x; } -// in each scan, we'll have scan_n components, and the order -// of the components is specified by order[] -#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + // in each scan, we'll have scan_n components, and the order + // of the components is specified by order[] + #define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) // after a restart interval, stbi__jpeg_reset the entropy decoder and // the dc prediction -static void stbi__jpeg_reset(stbi__jpeg *j) -{ - j->code_bits = 0; - j->code_buffer = 0; - j->nomore = 0; - j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = j->img_comp[3].dc_pred = 0; - j->marker = STBI__MARKER_none; - j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; - j->eob_run = 0; - // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, - // since we don't even allow 1<<30 pixels +static void stbi__jpeg_reset(stbi__jpeg *j) { + j->code_bits = 0; + j->code_buffer = 0; + j->nomore = 0; + j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = + j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels } -static int stbi__parse_entropy_coded_data(stbi__jpeg *z) -{ - stbi__jpeg_reset(z); - if (!z->progressive) { - if (z->scan_n == 1) { - int i,j; - STBI_SIMD_ALIGN(short, data[64]); - int n = z->order[0]; - // non-interleaved data, we just need to process one block at a time, - // in trivial scanline order - // number of blocks to do just depends on how many actual "pixels" this - // component has, independent of interleaved MCU blocking and such - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); - // every data block is an MCU, so countdown the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - // if it's NOT a restart, then just bail, so we get corrupt data - // rather than no data - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } else { // interleaved - int i,j,k,x,y; - STBI_SIMD_ALIGN(short, data[64]); - for (j=0; j < z->img_mcu_y; ++j) { - for (i=0; i < z->img_mcu_x; ++i) { - // scan an interleaved mcu... process scan_n components in order - for (k=0; k < z->scan_n; ++k) { - int n = z->order[k]; - // scan out an mcu's worth of this component; that's just determined - // by the basic H and V specified for the component - for (y=0; y < z->img_comp[n].v; ++y) { - for (x=0; x < z->img_comp[n].h; ++x) { - int x2 = (i*z->img_comp[n].h + x)*8; - int y2 = (j*z->img_comp[n].v + y)*8; - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data); - } - } - } - // after all interleaved components, that's an interleaved MCU, - // so now count down the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) { + stbi__jpeg_reset(z); + if (!z->progressive) { + if (z->scan_n == 1) { + int i, j; + STBI_SIMD_ALIGN(short, data[64]); + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc + z->img_comp[n].hd, + z->huff_ac + ha, z->fast_ac[ha], n, + z->dequant[z->img_comp[n].tq])) + return 0; + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, + z->img_comp[n].w2, data); + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } } - } else { - if (z->scan_n == 1) { - int i,j; - int n = z->order[0]; - // non-interleaved data, we just need to process one block at a time, - // in trivial scanline order - // number of blocks to do just depends on how many actual "pixels" this - // component has, independent of interleaved MCU blocking and such - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); - if (z->spec_start == 0) { - if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) - return 0; - } else { - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) - return 0; - } - // every data block is an MCU, so countdown the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } else { // interleaved - int i,j,k,x,y; - for (j=0; j < z->img_mcu_y; ++j) { - for (i=0; i < z->img_mcu_x; ++i) { - // scan an interleaved mcu... process scan_n components in order - for (k=0; k < z->scan_n; ++k) { - int n = z->order[k]; - // scan out an mcu's worth of this component; that's just determined - // by the basic H and V specified for the component - for (y=0; y < z->img_comp[n].v; ++y) { - for (x=0; x < z->img_comp[n].h; ++x) { - int x2 = (i*z->img_comp[n].h + x); - int y2 = (j*z->img_comp[n].v + y); - short *data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); - if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) - return 0; - } - } - } - // after all interleaved components, that's an interleaved MCU, - // so now count down the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } - } -} - -static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) -{ - int i; - for (i=0; i < 64; ++i) - data[i] *= dequant[i]; -} - -static void stbi__jpeg_finish(stbi__jpeg *z) -{ - if (z->progressive) { - // dequantize and idct the data - int i,j,n; - for (n=0; n < z->s->img_n; ++n) { - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); - stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); - } - } - } - } -} - -static int stbi__process_marker(stbi__jpeg *z, int m) -{ - int L; - switch (m) { - case STBI__MARKER_none: // no marker found - return stbi__err("expected marker","Corrupt JPEG"); - - case 0xDD: // DRI - specify restart interval - if (stbi__get16be(z->s) != 4) return stbi__err("bad DRI len","Corrupt JPEG"); - z->restart_interval = stbi__get16be(z->s); - return 1; - - case 0xDB: // DQT - define quantization table - L = stbi__get16be(z->s)-2; - while (L > 0) { - int q = stbi__get8(z->s); - int p = q >> 4, sixteen = (p != 0); - int t = q & 15,i; - if (p != 0 && p != 1) return stbi__err("bad DQT type","Corrupt JPEG"); - if (t > 3) return stbi__err("bad DQT table","Corrupt JPEG"); - - for (i=0; i < 64; ++i) - z->dequant[t][stbi__jpeg_dezigzag[i]] = (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); - L -= (sixteen ? 129 : 65); - } - return L==0; - - case 0xC4: // DHT - define huffman table - L = stbi__get16be(z->s)-2; - while (L > 0) { - stbi_uc *v; - int sizes[16],i,n=0; - int q = stbi__get8(z->s); - int tc = q >> 4; - int th = q & 15; - if (tc > 1 || th > 3) return stbi__err("bad DHT header","Corrupt JPEG"); - for (i=0; i < 16; ++i) { - sizes[i] = stbi__get8(z->s); - n += sizes[i]; - } - L -= 17; - if (tc == 0) { - if (!stbi__build_huffman(z->huff_dc+th, sizes)) return 0; - v = z->huff_dc[th].values; - } else { - if (!stbi__build_huffman(z->huff_ac+th, sizes)) return 0; - v = z->huff_ac[th].values; - } - for (i=0; i < n; ++i) - v[i] = stbi__get8(z->s); - if (tc != 0) - stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); - L -= n; - } - return L==0; - } - - // check for comment block or APP blocks - if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { - L = stbi__get16be(z->s); - if (L < 2) { - if (m == 0xFE) - return stbi__err("bad COM len","Corrupt JPEG"); - else - return stbi__err("bad APP len","Corrupt JPEG"); - } - L -= 2; - - if (m == 0xE0 && L >= 5) { // JFIF APP0 segment - static const unsigned char tag[5] = {'J','F','I','F','\0'}; - int ok = 1; - int i; - for (i=0; i < 5; ++i) - if (stbi__get8(z->s) != tag[i]) - ok = 0; - L -= 5; - if (ok) - z->jfif = 1; - } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment - static const unsigned char tag[6] = {'A','d','o','b','e','\0'}; - int ok = 1; - int i; - for (i=0; i < 6; ++i) - if (stbi__get8(z->s) != tag[i]) - ok = 0; - L -= 6; - if (ok) { - stbi__get8(z->s); // version - stbi__get16be(z->s); // flags0 - stbi__get16be(z->s); // flags1 - z->app14_color_transform = stbi__get8(z->s); // color transform - L -= 6; - } - } - - stbi__skip(z->s, L); return 1; - } + } else { // interleaved + int i, j, k, x, y; + STBI_SIMD_ALIGN(short, data[64]); + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k = 0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y = 0; y < z->img_comp[n].v; ++y) { + for (x = 0; x < z->img_comp[n].h; ++x) { + int x2 = (i * z->img_comp[n].h + x) * 8; + int y2 = (j * z->img_comp[n].v + y) * 8; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, + z->huff_dc + z->img_comp[n].hd, + z->huff_ac + ha, z->fast_ac[ha], n, + z->dequant[z->img_comp[n].tq])) + return 0; + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * y2 + x2, + z->img_comp[n].w2, data); + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } else { + if (z->scan_n == 1) { + int i, j; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short *data = + z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc( + z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], + z->fast_ac[ha])) + return 0; + } + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i, j, k, x, y; + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k = 0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y = 0; y < z->img_comp[n].v; ++y) { + for (x = 0; x < z->img_comp[n].h; ++x) { + int x2 = (i * z->img_comp[n].h + x); + int y2 = (j * z->img_comp[n].v + y); + short *data = z->img_comp[n].coeff + + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc( + z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} - return stbi__err("unknown marker","Corrupt JPEG"); +static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) { + int i; + for (i = 0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg *z) { + if (z->progressive) { + // dequantize and idct the data + int i, j, n; + for (n = 0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short *data = + z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, + z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg *z, int m) { + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker", "Corrupt JPEG"); + + case 0xDD: // DRI - specify restart interval + if (stbi__get16be(z->s) != 4) + return stbi__err("bad DRI len", "Corrupt JPEG"); + z->restart_interval = stbi__get16be(z->s); + return 1; + + case 0xDB: // DQT - define quantization table + L = stbi__get16be(z->s) - 2; + while (L > 0) { + int q = stbi__get8(z->s); + int p = q >> 4, sixteen = (p != 0); + int t = q & 15, i; + if (p != 0 && p != 1) return stbi__err("bad DQT type", "Corrupt JPEG"); + if (t > 3) return stbi__err("bad DQT table", "Corrupt JPEG"); + + for (i = 0; i < 64; ++i) + z->dequant[t][stbi__jpeg_dezigzag[i]] = + (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 65); + } + return L == 0; + + case 0xC4: // DHT - define huffman table + L = stbi__get16be(z->s) - 2; + while (L > 0) { + stbi_uc *v; + int sizes[16], i, n = 0; + int q = stbi__get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) + return stbi__err("bad DHT header", "Corrupt JPEG"); + for (i = 0; i < 16; ++i) { + sizes[i] = stbi__get8(z->s); + n += sizes[i]; + } + L -= 17; + if (tc == 0) { + if (!stbi__build_huffman(z->huff_dc + th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!stbi__build_huffman(z->huff_ac + th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i = 0; i < n; ++i) + v[i] = stbi__get8(z->s); + if (tc != 0) stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L == 0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len", "Corrupt JPEG"); + else + return stbi__err("bad APP len", "Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = {'J', 'F', 'I', 'F', '\0'}; + int ok = 1; + int i; + for (i = 0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) ok = 0; + L -= 5; + if (ok) z->jfif = 1; + } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = {'A', 'd', 'o', 'b', 'e', '\0'}; + int ok = 1; + int i; + for (i = 0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker", "Corrupt JPEG"); } // after we see SOS -static int stbi__process_scan_header(stbi__jpeg *z) -{ - int i; - int Ls = stbi__get16be(z->s); - z->scan_n = stbi__get8(z->s); - if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return stbi__err("bad SOS component count","Corrupt JPEG"); - if (Ls != 6+2*z->scan_n) return stbi__err("bad SOS len","Corrupt JPEG"); - for (i=0; i < z->scan_n; ++i) { - int id = stbi__get8(z->s), which; - int q = stbi__get8(z->s); - for (which = 0; which < z->s->img_n; ++which) - if (z->img_comp[which].id == id) - break; - if (which == z->s->img_n) return 0; // no match - z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff","Corrupt JPEG"); - z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff","Corrupt JPEG"); - z->order[i] = which; - } +static int stbi__process_scan_header(stbi__jpeg *z) { + int i; + int Ls = stbi__get16be(z->s); + z->scan_n = stbi__get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int)z->s->img_n) + return stbi__err("bad SOS component count", "Corrupt JPEG"); + if (Ls != 6 + 2 * z->scan_n) return stbi__err("bad SOS len", "Corrupt JPEG"); + for (i = 0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; + if (z->img_comp[which].hd > 3) + return stbi__err("bad DC huff", "Corrupt JPEG"); + z->img_comp[which].ha = q & 15; + if (z->img_comp[which].ha > 3) + return stbi__err("bad AC huff", "Corrupt JPEG"); + z->order[i] = which; + } - { - int aa; - z->spec_start = stbi__get8(z->s); - z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 - aa = stbi__get8(z->s); - z->succ_high = (aa >> 4); - z->succ_low = (aa & 15); - if (z->progressive) { - if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) - return stbi__err("bad SOS", "Corrupt JPEG"); - } else { - if (z->spec_start != 0) return stbi__err("bad SOS","Corrupt JPEG"); - if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS","Corrupt JPEG"); - z->spec_end = 63; - } - } + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || + z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } else { + if (z->spec_start != 0) return stbi__err("bad SOS", "Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) + return stbi__err("bad SOS", "Corrupt JPEG"); + z->spec_end = 63; + } + } - return 1; + return 1; } -static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) -{ - int i; - for (i=0; i < ncomp; ++i) { - if (z->img_comp[i].raw_data) { - STBI_FREE(z->img_comp[i].raw_data); - z->img_comp[i].raw_data = NULL; - z->img_comp[i].data = NULL; - } - if (z->img_comp[i].raw_coeff) { - STBI_FREE(z->img_comp[i].raw_coeff); - z->img_comp[i].raw_coeff = 0; - z->img_comp[i].coeff = 0; - } - if (z->img_comp[i].linebuf) { - STBI_FREE(z->img_comp[i].linebuf); - z->img_comp[i].linebuf = NULL; - } - } - return why; -} - -static int stbi__process_frame_header(stbi__jpeg *z, int scan) -{ - stbi__context *s = z->s; - int Lf,p,i,q, h_max=1,v_max=1,c; - Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len","Corrupt JPEG"); // JPEG - p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline - s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG - s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); // JPEG requires - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - c = stbi__get8(s); - if (c != 3 && c != 1 && c != 4) return stbi__err("bad component count","Corrupt JPEG"); - s->img_n = c; - for (i=0; i < c; ++i) { +static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) { + int i; + for (i = 0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; z->img_comp[i].data = NULL; - z->img_comp[i].linebuf = NULL; - } - - if (Lf != 8+3*s->img_n) return stbi__err("bad SOF len","Corrupt JPEG"); - - z->rgb = 0; - for (i=0; i < s->img_n; ++i) { - static const unsigned char rgb[3] = { 'R', 'G', 'B' }; - z->img_comp[i].id = stbi__get8(s); - if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) - ++z->rgb; - q = stbi__get8(s); - z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H","Corrupt JPEG"); - z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V","Corrupt JPEG"); - z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ","Corrupt JPEG"); - } - - if (scan != STBI__SCAN_load) return 1; - - if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) return stbi__err("too large", "Image too large to decode"); - - for (i=0; i < s->img_n; ++i) { - if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; - if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; - } - - // compute interleaved mcu info - z->img_h_max = h_max; - z->img_v_max = v_max; - z->img_mcu_w = h_max * 8; - z->img_mcu_h = v_max * 8; - // these sizes can't be more than 17 bits - z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; - z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; - - for (i=0; i < s->img_n; ++i) { - // number of effective pixels (e.g. for non-interleaved MCU) - z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; - z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; - // to simplify generation, we'll allocate enough memory to decode - // the bogus oversized data from using interleaved MCUs and their - // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't - // discard the extra data until colorspace conversion - // - // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) - // so these muls can't overflow with 32-bit ints (which we require) - z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; - z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; - z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); z->img_comp[i].linebuf = NULL; - z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); - if (z->img_comp[i].raw_data == NULL) - return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); - // align blocks for idct using mmx/sse - z->img_comp[i].data = (stbi_uc*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); - if (z->progressive) { - // w2, h2 are multiples of 8 (see above) - z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; - z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; - z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); - if (z->img_comp[i].raw_coeff == NULL) - return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); - z->img_comp[i].coeff = (short*) (((size_t) z->img_comp[i].raw_coeff + 15) & ~15); - } - } - - return 1; + } + } + return why; } -// use comparisons since in some cases we handle more than one case (e.g. SOF) -#define stbi__DNL(x) ((x) == 0xdc) -#define stbi__SOI(x) ((x) == 0xd8) -#define stbi__EOI(x) ((x) == 0xd9) -#define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) -#define stbi__SOS(x) ((x) == 0xda) +static int stbi__process_frame_header(stbi__jpeg *z, int scan) { + stbi__context *s = z->s; + int Lf, p, i, q, h_max = 1, v_max = 1, c; + Lf = stbi__get16be(s); + if (Lf < 11) return stbi__err("bad SOF len", "Corrupt JPEG"); // JPEG + p = stbi__get8(s); + if (p != 8) + return stbi__err("only 8-bit", + "JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = stbi__get16be(s); + if (s->img_y == 0) + return stbi__err( + "no header height", + "JPEG format not supported: delayed height"); // Legal, but we don't + // handle it--but neither + // does IJG + s->img_x = stbi__get16be(s); + if (s->img_x == 0) + return stbi__err("0 width", "Corrupt JPEG"); // JPEG requires + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + c = stbi__get8(s); + if (c != 3 && c != 1 && c != 4) + return stbi__err("bad component count", "Corrupt JPEG"); + s->img_n = c; + for (i = 0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } -#define stbi__SOF_progressive(x) ((x) == 0xc2) + if (Lf != 8 + 3 * s->img_n) return stbi__err("bad SOF len", "Corrupt JPEG"); -static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) -{ - int m; - z->jfif = 0; - z->app14_color_transform = -1; // valid values are 0,1,2 - z->marker = STBI__MARKER_none; // initialize cached marker to empty - m = stbi__get_marker(z); - if (!stbi__SOI(m)) return stbi__err("no SOI","Corrupt JPEG"); - if (scan == STBI__SCAN_type) return 1; - m = stbi__get_marker(z); - while (!stbi__SOF(m)) { - if (!stbi__process_marker(z,m)) return 0; + z->rgb = 0; + for (i = 0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = {'R', 'G', 'B'}; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); + if (!z->img_comp[i].h || z->img_comp[i].h > 4) + return stbi__err("bad H", "Corrupt JPEG"); + z->img_comp[i].v = q & 15; + if (!z->img_comp[i].v || z->img_comp[i].v > 4) + return stbi__err("bad V", "Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); + if (z->img_comp[i].tq > 3) return stbi__err("bad TQ", "Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) + return stbi__err("too large", "Image too large to decode"); + + for (i = 0; i < s->img_n; ++i) { + if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; + if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + // these sizes can't be more than 17 bits + z->img_mcu_x = (s->img_x + z->img_mcu_w - 1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h - 1) / z->img_mcu_h; + + for (i = 0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max - 1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max - 1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked + // earlier) so these muls can't overflow with 32-bit ints (which we require) + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = + stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i + 1, + stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = + (stbi_uc *)(((size_t)z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3( + z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components( + z, i + 1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = + (short *)(((size_t)z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + + // use comparisons since in some cases we handle more than one case (e.g. + // SOF) + #define stbi__DNL(x) ((x) == 0xdc) + #define stbi__SOI(x) ((x) == 0xd8) + #define stbi__EOI(x) ((x) == 0xd9) + #define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) + #define stbi__SOS(x) ((x) == 0xda) + + #define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) { + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI", "Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z, m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); m = stbi__get_marker(z); - while (m == STBI__MARKER_none) { - // some files have extra padding after their blocks, so ok, we'll scan - if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); - m = stbi__get_marker(z); - } - } - z->progressive = stbi__SOF_progressive(m); - if (!stbi__process_frame_header(z, scan)) return 0; - return 1; + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; } // decode image to YCbCr format -static int stbi__decode_jpeg_image(stbi__jpeg *j) -{ - int m; - for (m = 0; m < 4; m++) { - j->img_comp[m].raw_data = NULL; - j->img_comp[m].raw_coeff = NULL; - } - j->restart_interval = 0; - if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; - m = stbi__get_marker(j); - while (!stbi__EOI(m)) { - if (stbi__SOS(m)) { - if (!stbi__process_scan_header(j)) return 0; - if (!stbi__parse_entropy_coded_data(j)) return 0; - if (j->marker == STBI__MARKER_none ) { - // handle 0s at the end of image data from IP Kamera 9060 - while (!stbi__at_eof(j->s)) { - int x = stbi__get8(j->s); - if (x == 255) { - j->marker = stbi__get8(j->s); - break; - } - } - // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 - } - } else if (stbi__DNL(m)) { - int Ld = stbi__get16be(j->s); - stbi__uint32 NL = stbi__get16be(j->s); - if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); - if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); - } else { - if (!stbi__process_marker(j, m)) return 0; +static int stbi__decode_jpeg_image(stbi__jpeg *j) { + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none) { + // handle 0s at the end of image data from IP Kamera 9060 + while (!stbi__at_eof(j->s)) { + int x = stbi__get8(j->s); + if (x == 255) { + j->marker = stbi__get8(j->s); + break; + } + } + // if we reach eof without hitting a marker, stbi__get_marker() below + // will fail and we'll eventually return 0 } - m = stbi__get_marker(j); - } - if (j->progressive) - stbi__jpeg_finish(j); - return 1; + } else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + } else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + if (j->progressive) stbi__jpeg_finish(j); + return 1; } // static jfif-centered resampling (across block boundaries) typedef stbi_uc *(*resample_row_func)(stbi_uc *out, stbi_uc *in0, stbi_uc *in1, - int w, int hs); + int w, int hs); -#define stbi__div4(x) ((stbi_uc) ((x) >> 2)) + #define stbi__div4(x) ((stbi_uc)((x) >> 2)) -static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - STBI_NOTUSED(out); - STBI_NOTUSED(in_far); - STBI_NOTUSED(w); - STBI_NOTUSED(hs); - return in_near; +static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, + int w, int hs) { + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; } -static stbi_uc* stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate two samples vertically for every one in input - int i; - STBI_NOTUSED(hs); - for (i=0; i < w; ++i) - out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); - return out; +static stbi_uc *stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i = 0; i < w; ++i) + out[i] = stbi__div4(3 * in_near[i] + in_far[i] + 2); + return out; } -static stbi_uc* stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate two samples horizontally for every one in input - int i; - stbi_uc *input = in_near; +static stbi_uc *stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate two samples horizontally for every one in input + int i; + stbi_uc *input = in_near; - if (w == 1) { - // if only one sample, can't do any interpolation - out[0] = out[1] = input[0]; - return out; - } + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } - out[0] = input[0]; - out[1] = stbi__div4(input[0]*3 + input[1] + 2); - for (i=1; i < w-1; ++i) { - int n = 3*input[i]+2; - out[i*2+0] = stbi__div4(n+input[i-1]); - out[i*2+1] = stbi__div4(n+input[i+1]); - } - out[i*2+0] = stbi__div4(input[w-2]*3 + input[w-1] + 2); - out[i*2+1] = input[w-1]; + out[0] = input[0]; + out[1] = stbi__div4(input[0] * 3 + input[1] + 2); + for (i = 1; i < w - 1; ++i) { + int n = 3 * input[i] + 2; + out[i * 2 + 0] = stbi__div4(n + input[i - 1]); + out[i * 2 + 1] = stbi__div4(n + input[i + 1]); + } + out[i * 2 + 0] = stbi__div4(input[w - 2] * 3 + input[w - 1] + 2); + out[i * 2 + 1] = input[w - 1]; - STBI_NOTUSED(in_far); - STBI_NOTUSED(hs); + STBI_NOTUSED(in_far); + STBI_NOTUSED(hs); - return out; + return out; } -#define stbi__div16(x) ((stbi_uc) ((x) >> 4)) + #define stbi__div16(x) ((stbi_uc)((x) >> 4)) -static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate 2x2 samples for every one in input - int i,t0,t1; - if (w == 1) { - out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); - return out; - } +static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate 2x2 samples for every one in input + int i, t0, t1; + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } - t1 = 3*in_near[0] + in_far[0]; - out[0] = stbi__div4(t1+2); - for (i=1; i < w; ++i) { - t0 = t1; - t1 = 3*in_near[i]+in_far[i]; - out[i*2-1] = stbi__div16(3*t0 + t1 + 8); - out[i*2 ] = stbi__div16(3*t1 + t0 + 8); - } - out[w*2-1] = stbi__div4(t1+2); + t1 = 3 * in_near[0] + in_far[0]; + out[0] = stbi__div4(t1 + 2); + for (i = 1; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); - STBI_NOTUSED(hs); + STBI_NOTUSED(hs); - return out; + return out; } -#if defined(STBI_SSE2) || defined(STBI_NEON) -static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate 2x2 samples for every one in input - int i=0,t0,t1; + #if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate 2x2 samples for every one in input + int i = 0, t0, t1; - if (w == 1) { - out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); - return out; - } + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } - t1 = 3*in_near[0] + in_far[0]; - // process groups of 8 pixels for as long as we can. - // note we can't handle the last pixel in a row in this loop - // because we need to handle the filter boundary conditions. - for (; i < ((w-1) & ~7); i += 8) { -#if defined(STBI_SSE2) - // load and perform the vertical filtering pass - // this uses 3*x + y = 4*x + (y - x) - __m128i zero = _mm_setzero_si128(); - __m128i farb = _mm_loadl_epi64((__m128i *) (in_far + i)); - __m128i nearb = _mm_loadl_epi64((__m128i *) (in_near + i)); - __m128i farw = _mm_unpacklo_epi8(farb, zero); - __m128i nearw = _mm_unpacklo_epi8(nearb, zero); - __m128i diff = _mm_sub_epi16(farw, nearw); - __m128i nears = _mm_slli_epi16(nearw, 2); - __m128i curr = _mm_add_epi16(nears, diff); // current row + t1 = 3 * in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w - 1) & ~7); i += 8) { + #if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i *)(in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i *)(in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row - // horizontal filter works the same based on shifted vers of current - // row. "prev" is current row shifted right by 1 pixel; we need to - // insert the previous pixel value (from t1). - // "next" is current row shifted left by 1 pixel, with first pixel - // of next block of 8 pixels added in. - __m128i prv0 = _mm_slli_si128(curr, 2); - __m128i nxt0 = _mm_srli_si128(curr, 2); - __m128i prev = _mm_insert_epi16(prv0, t1, 0); - __m128i next = _mm_insert_epi16(nxt0, 3*in_near[i+8] + in_far[i+8], 7); + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = + _mm_insert_epi16(nxt0, 3 * in_near[i + 8] + in_far[i + 8], 7); - // horizontal filter, polyphase implementation since it's convenient: - // even pixels = 3*cur + prev = cur*4 + (prev - cur) - // odd pixels = 3*cur + next = cur*4 + (next - cur) - // note the shared term. - __m128i bias = _mm_set1_epi16(8); - __m128i curs = _mm_slli_epi16(curr, 2); - __m128i prvd = _mm_sub_epi16(prev, curr); - __m128i nxtd = _mm_sub_epi16(next, curr); - __m128i curb = _mm_add_epi16(curs, bias); - __m128i even = _mm_add_epi16(prvd, curb); - __m128i odd = _mm_add_epi16(nxtd, curb); + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); - // interleave even and odd pixels, then undo scaling. - __m128i int0 = _mm_unpacklo_epi16(even, odd); - __m128i int1 = _mm_unpackhi_epi16(even, odd); - __m128i de0 = _mm_srli_epi16(int0, 4); - __m128i de1 = _mm_srli_epi16(int1, 4); + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); - // pack and write output - __m128i outv = _mm_packus_epi16(de0, de1); - _mm_storeu_si128((__m128i *) (out + i*2), outv); -#elif defined(STBI_NEON) - // load and perform the vertical filtering pass - // this uses 3*x + y = 4*x + (y - x) - uint8x8_t farb = vld1_u8(in_far + i); - uint8x8_t nearb = vld1_u8(in_near + i); - int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); - int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); - int16x8_t curr = vaddq_s16(nears, diff); // current row + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i *)(out + i * 2), outv); + #elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row - // horizontal filter works the same based on shifted vers of current - // row. "prev" is current row shifted right by 1 pixel; we need to - // insert the previous pixel value (from t1). - // "next" is current row shifted left by 1 pixel, with first pixel - // of next block of 8 pixels added in. - int16x8_t prv0 = vextq_s16(curr, curr, 7); - int16x8_t nxt0 = vextq_s16(curr, curr, 1); - int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); - int16x8_t next = vsetq_lane_s16(3*in_near[i+8] + in_far[i+8], nxt0, 7); + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = + vsetq_lane_s16(3 * in_near[i + 8] + in_far[i + 8], nxt0, 7); - // horizontal filter, polyphase implementation since it's convenient: - // even pixels = 3*cur + prev = cur*4 + (prev - cur) - // odd pixels = 3*cur + next = cur*4 + (next - cur) - // note the shared term. - int16x8_t curs = vshlq_n_s16(curr, 2); - int16x8_t prvd = vsubq_s16(prev, curr); - int16x8_t nxtd = vsubq_s16(next, curr); - int16x8_t even = vaddq_s16(curs, prvd); - int16x8_t odd = vaddq_s16(curs, nxtd); + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); - // undo scaling and round, then store with even/odd phases interleaved - uint8x8x2_t o; - o.val[0] = vqrshrun_n_s16(even, 4); - o.val[1] = vqrshrun_n_s16(odd, 4); - vst2_u8(out + i*2, o); -#endif + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i * 2, o); + #endif - // "previous" value for next iter - t1 = 3*in_near[i+7] + in_far[i+7]; - } + // "previous" value for next iter + t1 = 3 * in_near[i + 7] + in_far[i + 7]; + } - t0 = t1; - t1 = 3*in_near[i] + in_far[i]; - out[i*2] = stbi__div16(3*t1 + t0 + 8); + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); - for (++i; i < w; ++i) { - t0 = t1; - t1 = 3*in_near[i]+in_far[i]; - out[i*2-1] = stbi__div16(3*t0 + t1 + 8); - out[i*2 ] = stbi__div16(3*t1 + t0 + 8); - } - out[w*2-1] = stbi__div4(t1+2); + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); - STBI_NOTUSED(hs); + STBI_NOTUSED(hs); - return out; + return out; } -#endif + #endif -static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // resample with nearest-neighbor - int i,j; - STBI_NOTUSED(in_far); - for (i=0; i < w; ++i) - for (j=0; j < hs; ++j) - out[i*hs+j] = in_near[i]; - return out; +static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // resample with nearest-neighbor + int i, j; + STBI_NOTUSED(in_far); + for (i = 0; i < w; ++i) + for (j = 0; j < hs; ++j) + out[i * hs + j] = in_near[i]; + return out; } -// this is a reduced-precision calculation of YCbCr-to-RGB introduced -// to make sure the code produces the same results in both SIMD and scalar -#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) -static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step) -{ - int i; - for (i=0; i < count; ++i) { - int y_fixed = (y[i] << 20) + (1<<19); // rounding - int r,g,b; - int cr = pcr[i] - 128; - int cb = pcb[i] - 128; - r = y_fixed + cr* stbi__float2fixed(1.40200f); - g = y_fixed + (cr*-stbi__float2fixed(0.71414f)) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); - b = y_fixed + cb* stbi__float2fixed(1.77200f); - r >>= 20; - g >>= 20; - b >>= 20; - if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } - if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } - if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } - out[0] = (stbi_uc)r; - out[1] = (stbi_uc)g; - out[2] = (stbi_uc)b; - out[3] = 255; - out += step; - } + // this is a reduced-precision calculation of YCbCr-to-RGB introduced + // to make sure the code produces the same results in both SIMD and scalar + #define stbi__float2fixed(x) (((int)((x)*4096.0f + 0.5f)) << 8) +static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, + const stbi_uc *pcb, const stbi_uc *pcr, + int count, int step) { + int i; + for (i = 0; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + (cr * -stbi__float2fixed(0.71414f)) + + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { + if (r < 0) + r = 0; + else + r = 255; + } + if ((unsigned)g > 255) { + if (g < 0) + g = 0; + else + g = 255; + } + if ((unsigned)b > 255) { + if (b < 0) + b = 0; + else + b = 255; + } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } } -#if defined(STBI_SSE2) || defined(STBI_NEON) -static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, stbi_uc const *pcb, stbi_uc const *pcr, int count, int step) -{ - int i = 0; + #if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, + stbi_uc const *pcb, stbi_uc const *pcr, + int count, int step) { + int i = 0; -#ifdef STBI_SSE2 - // step == 3 is pretty ugly on the final interleave, and i'm not convinced - // it's useful in practice (you wouldn't use it for textures, for example). - // so just accelerate step == 4 case. - if (step == 4) { - // this is a fairly straightforward implementation and not super-optimized. - __m128i signflip = _mm_set1_epi8(-0x80); - __m128i cr_const0 = _mm_set1_epi16( (short) ( 1.40200f*4096.0f+0.5f)); - __m128i cr_const1 = _mm_set1_epi16( - (short) ( 0.71414f*4096.0f+0.5f)); - __m128i cb_const0 = _mm_set1_epi16( - (short) ( 0.34414f*4096.0f+0.5f)); - __m128i cb_const1 = _mm_set1_epi16( (short) ( 1.77200f*4096.0f+0.5f)); - __m128i y_bias = _mm_set1_epi8((char) (unsigned char) 128); - __m128i xw = _mm_set1_epi16(255); // alpha channel + #ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16((short)(1.40200f * 4096.0f + 0.5f)); + __m128i cr_const1 = _mm_set1_epi16(-(short)(0.71414f * 4096.0f + 0.5f)); + __m128i cb_const0 = _mm_set1_epi16(-(short)(0.34414f * 4096.0f + 0.5f)); + __m128i cb_const1 = _mm_set1_epi16((short)(1.77200f * 4096.0f + 0.5f)); + __m128i y_bias = _mm_set1_epi8((char)(unsigned char)128); + __m128i xw = _mm_set1_epi16(255); // alpha channel - for (; i+7 < count; i += 8) { - // load - __m128i y_bytes = _mm_loadl_epi64((__m128i *) (y+i)); - __m128i cr_bytes = _mm_loadl_epi64((__m128i *) (pcr+i)); - __m128i cb_bytes = _mm_loadl_epi64((__m128i *) (pcb+i)); - __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 - __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + for (; i + 7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i *)(y + i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i *)(pcr + i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i *)(pcb + i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 - // unpack to short (and left-shift cr, cb by 8) - __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); - __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); - __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); - // color transform - __m128i yws = _mm_srli_epi16(yw, 4); - __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); - __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); - __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); - __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); - __m128i rws = _mm_add_epi16(cr0, yws); - __m128i gwt = _mm_add_epi16(cb0, yws); - __m128i bws = _mm_add_epi16(yws, cb1); - __m128i gws = _mm_add_epi16(gwt, cr1); + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); - // descale - __m128i rw = _mm_srai_epi16(rws, 4); - __m128i bw = _mm_srai_epi16(bws, 4); - __m128i gw = _mm_srai_epi16(gws, 4); + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); - // back to byte, set up for transpose - __m128i brb = _mm_packus_epi16(rw, bw); - __m128i gxb = _mm_packus_epi16(gw, xw); + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); - // transpose to interleave channels - __m128i t0 = _mm_unpacklo_epi8(brb, gxb); - __m128i t1 = _mm_unpackhi_epi8(brb, gxb); - __m128i o0 = _mm_unpacklo_epi16(t0, t1); - __m128i o1 = _mm_unpackhi_epi16(t0, t1); + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); - // store - _mm_storeu_si128((__m128i *) (out + 0), o0); - _mm_storeu_si128((__m128i *) (out + 16), o1); - out += 32; - } - } -#endif + // store + _mm_storeu_si128((__m128i *)(out + 0), o0); + _mm_storeu_si128((__m128i *)(out + 16), o1); + out += 32; + } + } + #endif -#ifdef STBI_NEON - // in this version, step=3 support would be easy to add. but is there demand? - if (step == 4) { - // this is a fairly straightforward implementation and not super-optimized. - uint8x8_t signflip = vdup_n_u8(0x80); - int16x8_t cr_const0 = vdupq_n_s16( (short) ( 1.40200f*4096.0f+0.5f)); - int16x8_t cr_const1 = vdupq_n_s16( - (short) ( 0.71414f*4096.0f+0.5f)); - int16x8_t cb_const0 = vdupq_n_s16( - (short) ( 0.34414f*4096.0f+0.5f)); - int16x8_t cb_const1 = vdupq_n_s16( (short) ( 1.77200f*4096.0f+0.5f)); + #ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16((short)(1.40200f * 4096.0f + 0.5f)); + int16x8_t cr_const1 = vdupq_n_s16(-(short)(0.71414f * 4096.0f + 0.5f)); + int16x8_t cb_const0 = vdupq_n_s16(-(short)(0.34414f * 4096.0f + 0.5f)); + int16x8_t cb_const1 = vdupq_n_s16((short)(1.77200f * 4096.0f + 0.5f)); - for (; i+7 < count; i += 8) { - // load - uint8x8_t y_bytes = vld1_u8(y + i); - uint8x8_t cr_bytes = vld1_u8(pcr + i); - uint8x8_t cb_bytes = vld1_u8(pcb + i); - int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); - int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + for (; i + 7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); - // expand to s16 - int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); - int16x8_t crw = vshll_n_s8(cr_biased, 7); - int16x8_t cbw = vshll_n_s8(cb_biased, 7); + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); - // color transform - int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); - int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); - int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); - int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); - int16x8_t rws = vaddq_s16(yws, cr0); - int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); - int16x8_t bws = vaddq_s16(yws, cb1); + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); - // undo scaling, round, convert to byte - uint8x8x4_t o; - o.val[0] = vqrshrun_n_s16(rws, 4); - o.val[1] = vqrshrun_n_s16(gws, 4); - o.val[2] = vqrshrun_n_s16(bws, 4); - o.val[3] = vdup_n_u8(255); + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); - // store, interleaving r/g/b/a - vst4_u8(out, o); - out += 8*4; - } - } -#endif + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8 * 4; + } + } + #endif - for (; i < count; ++i) { - int y_fixed = (y[i] << 20) + (1<<19); // rounding - int r,g,b; - int cr = pcr[i] - 128; - int cb = pcb[i] - 128; - r = y_fixed + cr* stbi__float2fixed(1.40200f); - g = y_fixed + cr*-stbi__float2fixed(0.71414f) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); - b = y_fixed + cb* stbi__float2fixed(1.77200f); - r >>= 20; - g >>= 20; - b >>= 20; - if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } - if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } - if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } - out[0] = (stbi_uc)r; - out[1] = (stbi_uc)g; - out[2] = (stbi_uc)b; - out[3] = 255; - out += step; - } + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + cr * -stbi__float2fixed(0.71414f) + + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { + if (r < 0) + r = 0; + else + r = 255; + } + if ((unsigned)g > 255) { + if (g < 0) + g = 0; + else + g = 255; + } + if ((unsigned)b > 255) { + if (b < 0) + b = 0; + else + b = 255; + } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } } -#endif + #endif // set up the kernels -static void stbi__setup_jpeg(stbi__jpeg *j) -{ - j->idct_block_kernel = stbi__idct_block; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; +static void stbi__setup_jpeg(stbi__jpeg *j) { + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; -#ifdef STBI_SSE2 - if (stbi__sse2_available()) { - j->idct_block_kernel = stbi__idct_simd; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; - } -#endif + #ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } + #endif -#ifdef STBI_NEON - j->idct_block_kernel = stbi__idct_simd; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; -#endif + #ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + #endif } // clean up the temporary component buffers -static void stbi__cleanup_jpeg(stbi__jpeg *j) -{ - stbi__free_jpeg_components(j, j->s->img_n, 0); +static void stbi__cleanup_jpeg(stbi__jpeg *j) { + stbi__free_jpeg_components(j, j->s->img_n, 0); } -typedef struct -{ - resample_row_func resample; - stbi_uc *line0,*line1; - int hs,vs; // expansion factor in each axis - int w_lores; // horizontal pixels pre-expansion - int ystep; // how far through vertical expansion we are - int ypos; // which pre-expansion row we're on +typedef struct { + resample_row_func resample; + stbi_uc *line0, *line1; + int hs, vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on } stbi__resample; // fast 0..255 * 0..255 => 0..255 rounded multiplication -static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) -{ - unsigned int t = x*y + 128; - return (stbi_uc) ((t + (t >>8)) >> 8); +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) { + unsigned int t = x * y + 128; + return (stbi_uc)((t + (t >> 8)) >> 8); } -static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) -{ - int n, decode_n, is_rgb; - z->s->img_n = 0; // make stbi__cleanup_jpeg safe +static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, + int *comp, int req_comp) { + int n, decode_n, is_rgb; + z->s->img_n = 0; // make stbi__cleanup_jpeg safe - // validate req_comp - if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + // validate req_comp + if (req_comp < 0 || req_comp > 4) + return stbi__errpuc("bad req_comp", "Internal error"); - // load a jpeg image from whichever source, but leave in YCbCr format - if (!stbi__decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + // load a jpeg image from whichever source, but leave in YCbCr format + if (!stbi__decode_jpeg_image(z)) { + stbi__cleanup_jpeg(z); + return NULL; + } - // determine actual number of components to generate - n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; - is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + is_rgb = z->s->img_n == 3 && + (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); - if (z->s->img_n == 3 && n < 3 && !is_rgb) - decode_n = 1; - else - decode_n = z->s->img_n; + if (z->s->img_n == 3 && n < 3 && !is_rgb) + decode_n = 1; + else + decode_n = z->s->img_n; - // resample and color-convert - { - int k; - unsigned int i,j; - stbi_uc *output; - stbi_uc *coutput[4] = { NULL, NULL, NULL, NULL }; + // resample and color-convert + { + int k; + unsigned int i, j; + stbi_uc *output; + stbi_uc *coutput[4] = {NULL, NULL, NULL, NULL}; - stbi__resample res_comp[4]; + stbi__resample res_comp[4]; - for (k=0; k < decode_n; ++k) { - stbi__resample *r = &res_comp[k]; + for (k = 0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; - // allocate line buffer big enough for upsampling off the edges - // with upsample factor of 4 - z->img_comp[k].linebuf = (stbi_uc *) stbi__malloc(z->s->img_x + 3); - if (!z->img_comp[k].linebuf) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } - - r->hs = z->img_h_max / z->img_comp[k].h; - r->vs = z->img_v_max / z->img_comp[k].v; - r->ystep = r->vs >> 1; - r->w_lores = (z->s->img_x + r->hs-1) / r->hs; - r->ypos = 0; - r->line0 = r->line1 = z->img_comp[k].data; - - if (r->hs == 1 && r->vs == 1) r->resample = resample_row_1; - else if (r->hs == 1 && r->vs == 2) r->resample = stbi__resample_row_v_2; - else if (r->hs == 2 && r->vs == 1) r->resample = stbi__resample_row_h_2; - else if (r->hs == 2 && r->vs == 2) r->resample = z->resample_row_hv_2_kernel; - else r->resample = stbi__resample_row_generic; + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (stbi_uc *)stbi__malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { + stbi__cleanup_jpeg(z); + return stbi__errpuc("outofmem", "Out of memory"); } - // can't error after this so, this is safe - output = (stbi_uc *) stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); - if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + r->hs = z->img_h_max / z->img_comp[k].h; + r->vs = z->img_v_max / z->img_comp[k].v; + r->ystep = r->vs >> 1; + r->w_lores = (z->s->img_x + r->hs - 1) / r->hs; + r->ypos = 0; + r->line0 = r->line1 = z->img_comp[k].data; - // now go ahead and resample - for (j=0; j < z->s->img_y; ++j) { - stbi_uc *out = output + n * z->s->img_x * j; - for (k=0; k < decode_n; ++k) { - stbi__resample *r = &res_comp[k]; - int y_bot = r->ystep >= (r->vs >> 1); - coutput[k] = r->resample(z->img_comp[k].linebuf, - y_bot ? r->line1 : r->line0, - y_bot ? r->line0 : r->line1, - r->w_lores, r->hs); - if (++r->ystep >= r->vs) { - r->ystep = 0; - r->line0 = r->line1; - if (++r->ypos < z->img_comp[k].y) - r->line1 += z->img_comp[k].w2; - } - } - if (n >= 3) { - stbi_uc *y = coutput[0]; - if (z->s->img_n == 3) { - if (is_rgb) { - for (i=0; i < z->s->img_x; ++i) { - out[0] = y[i]; - out[1] = coutput[1][i]; - out[2] = coutput[2][i]; - out[3] = 255; - out += n; - } - } else { - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - } - } else if (z->s->img_n == 4) { - if (z->app14_color_transform == 0) { // CMYK - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - out[0] = stbi__blinn_8x8(coutput[0][i], m); - out[1] = stbi__blinn_8x8(coutput[1][i], m); - out[2] = stbi__blinn_8x8(coutput[2][i], m); - out[3] = 255; - out += n; - } - } else if (z->app14_color_transform == 2) { // YCCK - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - out[0] = stbi__blinn_8x8(255 - out[0], m); - out[1] = stbi__blinn_8x8(255 - out[1], m); - out[2] = stbi__blinn_8x8(255 - out[2], m); - out += n; - } - } else { // YCbCr + alpha? Ignore the fourth channel for now - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - } - } else - for (i=0; i < z->s->img_x; ++i) { - out[0] = out[1] = out[2] = y[i]; - out[3] = 255; // not used if n==3 - out += n; - } - } else { - if (is_rgb) { - if (n == 1) - for (i=0; i < z->s->img_x; ++i) - *out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); - else { - for (i=0; i < z->s->img_x; ++i, out += 2) { - out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); - out[1] = 255; - } - } - } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); - stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); - stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); - out[0] = stbi__compute_y(r, g, b); - out[1] = 255; - out += n; - } - } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { - for (i=0; i < z->s->img_x; ++i) { - out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); - out[1] = 255; - out += n; - } - } else { - stbi_uc *y = coutput[0]; - if (n == 1) - for (i=0; i < z->s->img_x; ++i) out[i] = y[i]; - else - for (i=0; i < z->s->img_x; ++i) { *out++ = y[i]; *out++ = 255; } - } - } - } + if (r->hs == 1 && r->vs == 1) + r->resample = resample_row_1; + else if (r->hs == 1 && r->vs == 2) + r->resample = stbi__resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) + r->resample = stbi__resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) + r->resample = z->resample_row_hv_2_kernel; + else + r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc *)stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); + if (!output) { stbi__cleanup_jpeg(z); - *out_x = z->s->img_x; - *out_y = z->s->img_y; - if (comp) *comp = z->s->img_n >= 3 ? 3 : 1; // report original components, not output - return output; - } + return stbi__errpuc("outofmem", "Out of memory"); + } + + // now go ahead and resample + for (j = 0; j < z->s->img_y; ++j) { + stbi_uc *out = output + n * z->s->img_x * j; + for (k = 0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + int y_bot = r->ystep >= (r->vs >> 1); + coutput[k] = + r->resample(z->img_comp[k].linebuf, y_bot ? r->line1 : r->line0, + y_bot ? r->line0 : r->line1, r->w_lores, r->hs); + if (++r->ystep >= r->vs) { + r->ystep = 0; + r->line0 = r->line1; + if (++r->ypos < z->img_comp[k].y) r->line1 += z->img_comp[k].w2; + } + } + if (n >= 3) { + stbi_uc *y = coutput[0]; + if (z->s->img_n == 3) { + if (is_rgb) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + } + } else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + } + } else + for (i = 0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + if (is_rgb) { + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + *out++ = + stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i = 0; i < z->s->img_x; ++i, out += 2) { + out[0] = + stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + out += n; + } + } else { + stbi_uc *y = coutput[0]; + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + out[i] = y[i]; + else + for (i = 0; i < z->s->img_x; ++i) { + *out++ = y[i]; + *out++ = 255; + } + } + } + } + stbi__cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) + *comp = + z->s->img_n >= 3 ? 3 : 1; // report original components, not output + return output; + } } -static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - unsigned char* result; - stbi__jpeg* j = (stbi__jpeg*) stbi__malloc(sizeof(stbi__jpeg)); - STBI_NOTUSED(ri); - j->s = s; - stbi__setup_jpeg(j); - result = load_jpeg_image(j, x,y,comp,req_comp); - STBI_FREE(j); - return result; +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + unsigned char *result; + stbi__jpeg *j = (stbi__jpeg *)stbi__malloc(sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x, y, comp, req_comp); + STBI_FREE(j); + return result; } -static int stbi__jpeg_test(stbi__context *s) -{ - int r; - stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); - j->s = s; - stbi__setup_jpeg(j); - r = stbi__decode_jpeg_header(j, STBI__SCAN_type); - stbi__rewind(s); - STBI_FREE(j); - return r; +static int stbi__jpeg_test(stbi__context *s) { + int r; + stbi__jpeg *j = (stbi__jpeg *)stbi__malloc(sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; } -static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) -{ - if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { - stbi__rewind( j->s ); - return 0; - } - if (x) *x = j->s->img_x; - if (y) *y = j->s->img_y; - if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; - return 1; +static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) { + if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { + stbi__rewind(j->s); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; + return 1; } -static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) -{ - int result; - stbi__jpeg* j = (stbi__jpeg*) (stbi__malloc(sizeof(stbi__jpeg))); - j->s = s; - result = stbi__jpeg_info_raw(j, x, y, comp); - STBI_FREE(j); - return result; +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) { + int result; + stbi__jpeg *j = (stbi__jpeg *)(stbi__malloc(sizeof(stbi__jpeg))); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; } -#endif + #endif // public domain zlib decode v0.2 Sean Barrett 2006-11-18 // simple implementation @@ -3974,86 +4316,83 @@ static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) // performance // - fast huffman -#ifndef STBI_NO_ZLIB + #ifndef STBI_NO_ZLIB -// fast-way is faster to check than jpeg huffman, but slow way is slower -#define STBI__ZFAST_BITS 9 // accelerate all cases in default tables -#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) + // fast-way is faster to check than jpeg huffman, but slow way is slower + #define STBI__ZFAST_BITS 9 // accelerate all cases in default tables + #define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) // zlib-style huffman encoding // (jpegs packs from left, zlib from right, so can't share code) -typedef struct -{ - stbi__uint16 fast[1 << STBI__ZFAST_BITS]; - stbi__uint16 firstcode[16]; - int maxcode[17]; - stbi__uint16 firstsymbol[16]; - stbi_uc size[288]; - stbi__uint16 value[288]; +typedef struct { + stbi__uint16 fast[1 << STBI__ZFAST_BITS]; + stbi__uint16 firstcode[16]; + int maxcode[17]; + stbi__uint16 firstsymbol[16]; + stbi_uc size[288]; + stbi__uint16 value[288]; } stbi__zhuffman; -stbi_inline static int stbi__bitreverse16(int n) -{ - n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); - n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); - n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); - n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); +stbi_inline static int stbi__bitreverse16(int n) { + n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); + n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); + n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); + n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); return n; } -stbi_inline static int stbi__bit_reverse(int v, int bits) -{ - STBI_ASSERT(bits <= 16); - // to bit reverse n bits, reverse 16 and shift - // e.g. 11 bits, bit reverse and shift away 5 - return stbi__bitreverse16(v) >> (16-bits); +stbi_inline static int stbi__bit_reverse(int v, int bits) { + STBI_ASSERT(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return stbi__bitreverse16(v) >> (16 - bits); } -static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int num) -{ - int i,k=0; - int code, next_code[16], sizes[17]; +static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, + int num) { + int i, k = 0; + int code, next_code[16], sizes[17]; - // DEFLATE spec for generating codes - memset(sizes, 0, sizeof(sizes)); - memset(z->fast, 0, sizeof(z->fast)); - for (i=0; i < num; ++i) - ++sizes[sizelist[i]]; - sizes[0] = 0; - for (i=1; i < 16; ++i) - if (sizes[i] > (1 << i)) - return stbi__err("bad sizes", "Corrupt PNG"); - code = 0; - for (i=1; i < 16; ++i) { - next_code[i] = code; - z->firstcode[i] = (stbi__uint16) code; - z->firstsymbol[i] = (stbi__uint16) k; - code = (code + sizes[i]); - if (sizes[i]) - if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt PNG"); - z->maxcode[i] = code << (16-i); // preshift for inner loop - code <<= 1; - k += sizes[i]; - } - z->maxcode[16] = 0x10000; // sentinel - for (i=0; i < num; ++i) { - int s = sizelist[i]; - if (s) { - int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; - stbi__uint16 fastv = (stbi__uint16) ((s << 9) | i); - z->size [c] = (stbi_uc ) s; - z->value[c] = (stbi__uint16) i; - if (s <= STBI__ZFAST_BITS) { - int j = stbi__bit_reverse(next_code[s],s); - while (j < (1 << STBI__ZFAST_BITS)) { - z->fast[j] = fastv; - j += (1 << s); - } - } - ++next_code[s]; + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 0, sizeof(z->fast)); + for (i = 0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i = 1; i < 16; ++i) + if (sizes[i] > (1 << i)) return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i = 1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (stbi__uint16)code; + z->firstsymbol[i] = (stbi__uint16)k; + code = (code + sizes[i]); + if (sizes[i]) + if (code - 1 >= (1 << i)) + return stbi__err("bad codelengths", "Corrupt PNG"); + z->maxcode[i] = code << (16 - i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i = 0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + stbi__uint16 fastv = (stbi__uint16)((s << 9) | i); + z->size[c] = (stbi_uc)s; + z->value[c] = (stbi__uint16)i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s], s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (1 << s); + } } - } - return 1; + ++next_code[s]; + } + } + return 1; } // zlib-from-memory implementation for PNG reading @@ -4062,277 +4401,291 @@ static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int // we require PNG read all the IDATs and combine them into a single // memory buffer -typedef struct -{ - stbi_uc *zbuffer, *zbuffer_end; - int num_bits; - stbi__uint32 code_buffer; +typedef struct { + stbi_uc *zbuffer, *zbuffer_end; + int num_bits; + stbi__uint32 code_buffer; - char *zout; - char *zout_start; - char *zout_end; - int z_expandable; + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; - stbi__zhuffman z_length, z_distance; + stbi__zhuffman z_length, z_distance; } stbi__zbuf; -stbi_inline static int stbi__zeof(stbi__zbuf *z) -{ - return (z->zbuffer >= z->zbuffer_end); +stbi_inline static int stbi__zeof(stbi__zbuf *z) { + return (z->zbuffer >= z->zbuffer_end); } -stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) -{ - return stbi__zeof(z) ? 0 : *z->zbuffer++; +stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) { + return stbi__zeof(z) ? 0 : *z->zbuffer++; } -static void stbi__fill_bits(stbi__zbuf *z) -{ - do { - if (z->code_buffer >= (1U << z->num_bits)) { - z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ - return; - } - z->code_buffer |= (unsigned int) stbi__zget8(z) << z->num_bits; - z->num_bits += 8; - } while (z->num_bits <= 24); +static void stbi__fill_bits(stbi__zbuf *z) { + do { + if (z->code_buffer >= (1U << z->num_bits)) { + z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ + return; + } + z->code_buffer |= (unsigned int)stbi__zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); } -stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) -{ - unsigned int k; - if (z->num_bits < n) stbi__fill_bits(z); - k = z->code_buffer & ((1 << n) - 1); - z->code_buffer >>= n; - z->num_bits -= n; - return k; +stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) { + unsigned int k; + if (z->num_bits < n) stbi__fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; } -static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) -{ - int b,s,k; - // not resolved by fast table, so compute it the slow way - // use jpeg approach, which requires MSbits at top - k = stbi__bit_reverse(a->code_buffer, 16); - for (s=STBI__ZFAST_BITS+1; ; ++s) - if (k < z->maxcode[s]) - break; - if (s >= 16) return -1; // invalid code! - // code size is s, so: - b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; - if (b >= sizeof (z->size)) return -1; // some data was corrupt somewhere! - if (z->size[b] != s) return -1; // was originally an assert, but report failure instead. - a->code_buffer >>= s; - a->num_bits -= s; - return z->value[b]; +static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) { + int b, s, k; + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = stbi__bit_reverse(a->code_buffer, 16); + for (s = STBI__ZFAST_BITS + 1;; ++s) + if (k < z->maxcode[s]) break; + if (s >= 16) return -1; // invalid code! + // code size is s, so: + b = (k >> (16 - s)) - z->firstcode[s] + z->firstsymbol[s]; + if (b >= sizeof(z->size)) return -1; // some data was corrupt somewhere! + if (z->size[b] != s) + return -1; // was originally an assert, but report failure instead. + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; } -stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) -{ - int b,s; - if (a->num_bits < 16) { - if (stbi__zeof(a)) { - return -1; /* report error for unexpected end of data. */ - } - stbi__fill_bits(a); - } - b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; - if (b) { - s = b >> 9; - a->code_buffer >>= s; - a->num_bits -= s; - return b & 511; - } - return stbi__zhuffman_decode_slowpath(a, z); +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) { + int b, s; + if (a->num_bits < 16) { + if (stbi__zeof(a)) { + return -1; /* report error for unexpected end of data. */ + } + stbi__fill_bits(a); + } + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); } -static int stbi__zexpand(stbi__zbuf *z, char *zout, int n) // need to make room for n bytes +static int stbi__zexpand(stbi__zbuf *z, char *zout, + int n) // need to make room for n bytes { - char *q; - unsigned int cur, limit, old_limit; - z->zout = zout; - if (!z->z_expandable) return stbi__err("output buffer limit","Corrupt PNG"); - cur = (unsigned int) (z->zout - z->zout_start); - limit = old_limit = (unsigned) (z->zout_end - z->zout_start); - if (UINT_MAX - cur < (unsigned) n) return stbi__err("outofmem", "Out of memory"); - while (cur + n > limit) { - if(limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); - limit *= 2; - } - q = (char *) STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); - STBI_NOTUSED(old_limit); - if (q == NULL) return stbi__err("outofmem", "Out of memory"); - z->zout_start = q; - z->zout = q + cur; - z->zout_end = q + limit; - return 1; + char *q; + unsigned int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit", "Corrupt PNG"); + cur = (unsigned int)(z->zout - z->zout_start); + limit = old_limit = (unsigned)(z->zout_end - z->zout_start); + if (UINT_MAX - cur < (unsigned)n) + return stbi__err("outofmem", "Out of memory"); + while (cur + n > limit) { + if (limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); + limit *= 2; + } + q = (char *)STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_limit); + if (q == NULL) return stbi__err("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; } static const int stbi__zlength_base[31] = { - 3,4,5,6,7,8,9,10,11,13, - 15,17,19,23,27,31,35,43,51,59, - 67,83,99,115,131,163,195,227,258,0,0 }; + 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, + 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0}; -static const int stbi__zlength_extra[31]= -{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 }; +static const int stbi__zlength_extra[31] = {0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, + 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, + 4, 4, 5, 5, 5, 5, 0, 0, 0}; -static const int stbi__zdist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193, -257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0}; +static const int stbi__zdist_base[32] = { + 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, + 49, 65, 97, 129, 193, 257, 385, 513, 769, 1025, 1537, + 2049, 3073, 4097, 6145, 8193, 12289, 16385, 24577, 0, 0}; -static const int stbi__zdist_extra[32] = -{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; +static const int stbi__zdist_extra[32] = {0, 0, 0, 0, 1, 1, 2, 2, 3, 3, + 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, + 9, 9, 10, 10, 11, 11, 12, 12, 13, 13}; -static int stbi__parse_huffman_block(stbi__zbuf *a) -{ - char *zout = a->zout; - for(;;) { - int z = stbi__zhuffman_decode(a, &a->z_length); - if (z < 256) { - if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); // error in huffman codes - if (zout >= a->zout_end) { - if (!stbi__zexpand(a, zout, 1)) return 0; - zout = a->zout; - } - *zout++ = (char) z; +static int stbi__parse_huffman_block(stbi__zbuf *a) { + char *zout = a->zout; + for (;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) + return stbi__err("bad huffman code", + "Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char)z; + } else { + stbi_uc *p; + int len, dist; + if (z == 256) { + a->zout = zout; + return 1; + } + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) + len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0) return stbi__err("bad huffman code", "Corrupt PNG"); + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) + return stbi__err("bad dist", "Corrupt PNG"); + if (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc *)(zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { + do + *zout++ = v; + while (--len); + } } else { - stbi_uc *p; - int len,dist; - if (z == 256) { - a->zout = zout; - return 1; - } - z -= 257; - len = stbi__zlength_base[z]; - if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); - z = stbi__zhuffman_decode(a, &a->z_distance); - if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); - dist = stbi__zdist_base[z]; - if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); - if (zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); - if (zout + len > a->zout_end) { - if (!stbi__zexpand(a, zout, len)) return 0; - zout = a->zout; - } - p = (stbi_uc *) (zout - dist); - if (dist == 1) { // run of one byte; common in images. - stbi_uc v = *p; - if (len) { do *zout++ = v; while (--len); } - } else { - if (len) { do *zout++ = *p++; while (--len); } - } + if (len) { + do + *zout++ = *p++; + while (--len); + } } - } + } + } } -static int stbi__compute_huffman_codes(stbi__zbuf *a) -{ - static const stbi_uc length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; - stbi__zhuffman z_codelength; - stbi_uc lencodes[286+32+137];//padding for maximum single op - stbi_uc codelength_sizes[19]; - int i,n; +static int stbi__compute_huffman_codes(stbi__zbuf *a) { + static const stbi_uc length_dezigzag[19] = { + 16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; + stbi__zhuffman z_codelength; + stbi_uc lencodes[286 + 32 + 137]; // padding for maximum single op + stbi_uc codelength_sizes[19]; + int i, n; - int hlit = stbi__zreceive(a,5) + 257; - int hdist = stbi__zreceive(a,5) + 1; - int hclen = stbi__zreceive(a,4) + 4; - int ntot = hlit + hdist; + int hlit = stbi__zreceive(a, 5) + 257; + int hdist = stbi__zreceive(a, 5) + 1; + int hclen = stbi__zreceive(a, 4) + 4; + int ntot = hlit + hdist; - memset(codelength_sizes, 0, sizeof(codelength_sizes)); - for (i=0; i < hclen; ++i) { - int s = stbi__zreceive(a,3); - codelength_sizes[length_dezigzag[i]] = (stbi_uc) s; - } - if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i = 0; i < hclen; ++i) { + int s = stbi__zreceive(a, 3); + codelength_sizes[length_dezigzag[i]] = (stbi_uc)s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; - n = 0; - while (n < ntot) { - int c = stbi__zhuffman_decode(a, &z_codelength); - if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); - if (c < 16) - lencodes[n++] = (stbi_uc) c; - else { - stbi_uc fill = 0; - if (c == 16) { - c = stbi__zreceive(a,2)+3; - if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); - fill = lencodes[n-1]; - } else if (c == 17) { - c = stbi__zreceive(a,3)+3; - } else if (c == 18) { - c = stbi__zreceive(a,7)+11; - } else { - return stbi__err("bad codelengths", "Corrupt PNG"); - } - if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); - memset(lencodes+n, fill, c); - n += c; + n = 0; + while (n < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc)c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a, 2) + 3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n - 1]; + } else if (c == 17) { + c = stbi__zreceive(a, 3) + 3; + } else if (c == 18) { + c = stbi__zreceive(a, 7) + 11; + } else { + return stbi__err("bad codelengths", "Corrupt PNG"); } - } - if (n != ntot) return stbi__err("bad codelengths","Corrupt PNG"); - if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; - if (!stbi__zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; - return 1; + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes + n, fill, c); + n += c; + } + } + if (n != ntot) return stbi__err("bad codelengths", "Corrupt PNG"); + if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, lencodes + hlit, hdist)) return 0; + return 1; } -static int stbi__parse_uncompressed_block(stbi__zbuf *a) -{ - stbi_uc header[4]; - int len,nlen,k; - if (a->num_bits & 7) - stbi__zreceive(a, a->num_bits & 7); // discard - // drain the bit-packed data into header - k = 0; - while (a->num_bits > 0) { - header[k++] = (stbi_uc) (a->code_buffer & 255); // suppress MSVC run-time check - a->code_buffer >>= 8; - a->num_bits -= 8; - } - if (a->num_bits < 0) return stbi__err("zlib corrupt","Corrupt PNG"); - // now fill header the normal way - while (k < 4) - header[k++] = stbi__zget8(a); - len = header[1] * 256 + header[0]; - nlen = header[3] * 256 + header[2]; - if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt","Corrupt PNG"); - if (a->zbuffer + len > a->zbuffer_end) return stbi__err("read past buffer","Corrupt PNG"); - if (a->zout + len > a->zout_end) - if (!stbi__zexpand(a, a->zout, len)) return 0; - memcpy(a->zout, a->zbuffer, len); - a->zbuffer += len; - a->zout += len; - return 1; +static int stbi__parse_uncompressed_block(stbi__zbuf *a) { + stbi_uc header[4]; + int len, nlen, k; + if (a->num_bits & 7) stbi__zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = + (stbi_uc)(a->code_buffer & 255); // suppress MSVC run-time check + a->code_buffer >>= 8; + a->num_bits -= 8; + } + if (a->num_bits < 0) return stbi__err("zlib corrupt", "Corrupt PNG"); + // now fill header the normal way + while (k < 4) + header[k++] = stbi__zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt", "Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) + return stbi__err("read past buffer", "Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!stbi__zexpand(a, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; } -static int stbi__parse_zlib_header(stbi__zbuf *a) -{ - int cmf = stbi__zget8(a); - int cm = cmf & 15; - /* int cinfo = cmf >> 4; */ - int flg = stbi__zget8(a); - if (stbi__zeof(a)) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec - if ((cmf*256+flg) % 31 != 0) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec - if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png - if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); // DEFLATE required for png - // window = 1 << (8 + cinfo)... but who cares, we fully buffer output - return 1; +static int stbi__parse_zlib_header(stbi__zbuf *a) { + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if (stbi__zeof(a)) + return stbi__err("bad zlib header", "Corrupt PNG"); // zlib spec + if ((cmf * 256 + flg) % 31 != 0) + return stbi__err("bad zlib header", "Corrupt PNG"); // zlib spec + if (flg & 32) + return stbi__err("no preset dict", + "Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) + return stbi__err("bad compression", + "Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; } -static const stbi_uc stbi__zdefault_length[288] = -{ - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 -}; -static const stbi_uc stbi__zdefault_distance[32] = -{ - 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 -}; +static const stbi_uc stbi__zdefault_length[288] = { + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 8}; +static const stbi_uc stbi__zdefault_distance[32] = { + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5}; /* Init algorithm: { @@ -4346,119 +4699,123 @@ Init algorithm: } */ -static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) -{ - int final, type; - if (parse_header) - if (!stbi__parse_zlib_header(a)) return 0; - a->num_bits = 0; - a->code_buffer = 0; - do { - final = stbi__zreceive(a,1); - type = stbi__zreceive(a,2); - if (type == 0) { - if (!stbi__parse_uncompressed_block(a)) return 0; - } else if (type == 3) { - return 0; +static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) { + int final, type; + if (parse_header) + if (!stbi__parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + do { + final = stbi__zreceive(a, 1); + type = stbi__zreceive(a, 2); + if (type == 0) { + if (!stbi__parse_uncompressed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + // use fixed code lengths + if (!stbi__zbuild_huffman(&a->z_length, stbi__zdefault_length, 288)) + return 0; + if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) + return 0; } else { - if (type == 1) { - // use fixed code lengths - if (!stbi__zbuild_huffman(&a->z_length , stbi__zdefault_length , 288)) return 0; - if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) return 0; - } else { - if (!stbi__compute_huffman_codes(a)) return 0; - } - if (!stbi__parse_huffman_block(a)) return 0; + if (!stbi__compute_huffman_codes(a)) return 0; } - } while (!final); - return 1; + if (!stbi__parse_huffman_block(a)) return 0; + } + } while (!final); + return 1; } -static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, int parse_header) -{ - a->zout_start = obuf; - a->zout = obuf; - a->zout_end = obuf + olen; - a->z_expandable = exp; +static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, + int parse_header) { + a->zout_start = obuf; + a->zout = obuf; + a->zout_end = obuf + olen; + a->z_expandable = exp; - return stbi__parse_zlib(a, parse_header); + return stbi__parse_zlib(a, parse_header); } -STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(initial_size); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer + len; - if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, + int initial_size, int *outlen) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) -{ - return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, + int *outlen) { + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); } -STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(initial_size); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer + len; - if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, + int len, + int initial_size, + int *outlen, + int parse_header) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) -{ - stbi__zbuf a; - a.zbuffer = (stbi_uc *) ibuffer; - a.zbuffer_end = (stbi_uc *) ibuffer + ilen; - if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) - return (int) (a.zout - a.zout_start); - else - return -1; +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, + char const *ibuffer, int ilen) { + stbi__zbuf a; + a.zbuffer = (stbi_uc *)ibuffer; + a.zbuffer_end = (stbi_uc *)ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) + return (int)(a.zout - a.zout_start); + else + return -1; } -STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(16384); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer+len; - if (stbi__do_zlib(&a, p, 16384, 1, 0)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, + int *outlen) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) -{ - stbi__zbuf a; - a.zbuffer = (stbi_uc *) ibuffer; - a.zbuffer_end = (stbi_uc *) ibuffer + ilen; - if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) - return (int) (a.zout - a.zout_start); - else - return -1; +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen) { + stbi__zbuf a; + a.zbuffer = (stbi_uc *)ibuffer; + a.zbuffer_end = (stbi_uc *)ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) + return (int)(a.zout - a.zout_start); + else + return -1; } -#endif + #endif // public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 // simple implementation @@ -4470,713 +4827,827 @@ STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char // performance // - uses stb_zlib, a PD zlib implementation with fast huffman decoding -#ifndef STBI_NO_PNG -typedef struct -{ - stbi__uint32 length; - stbi__uint32 type; + #ifndef STBI_NO_PNG +typedef struct { + stbi__uint32 length; + stbi__uint32 type; } stbi__pngchunk; -static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) -{ - stbi__pngchunk c; - c.length = stbi__get32be(s); - c.type = stbi__get32be(s); - return c; +static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) { + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; } -static int stbi__check_png_header(stbi__context *s) -{ - static const stbi_uc png_sig[8] = { 137,80,78,71,13,10,26,10 }; - int i; - for (i=0; i < 8; ++i) - if (stbi__get8(s) != png_sig[i]) return stbi__err("bad png sig","Not a PNG"); - return 1; +static int stbi__check_png_header(stbi__context *s) { + static const stbi_uc png_sig[8] = {137, 80, 78, 71, 13, 10, 26, 10}; + int i; + for (i = 0; i < 8; ++i) + if (stbi__get8(s) != png_sig[i]) + return stbi__err("bad png sig", "Not a PNG"); + return 1; } -typedef struct -{ - stbi__context *s; - stbi_uc *idata, *expanded, *out; - int depth; +typedef struct { + stbi__context *s; + stbi_uc *idata, *expanded, *out; + int depth; } stbi__png; - enum { - STBI__F_none=0, - STBI__F_sub=1, - STBI__F_up=2, - STBI__F_avg=3, - STBI__F_paeth=4, - // synthetic filters used for first scanline to avoid needing a dummy row of 0s - STBI__F_avg_first, - STBI__F_paeth_first + STBI__F_none = 0, + STBI__F_sub = 1, + STBI__F_up = 2, + STBI__F_avg = 3, + STBI__F_paeth = 4, + // synthetic filters used for first scanline to avoid needing a dummy row of + // 0s + STBI__F_avg_first, + STBI__F_paeth_first }; -static stbi_uc first_row_filter[5] = -{ - STBI__F_none, - STBI__F_sub, - STBI__F_none, - STBI__F_avg_first, - STBI__F_paeth_first -}; +static stbi_uc first_row_filter[5] = {STBI__F_none, STBI__F_sub, STBI__F_none, + STBI__F_avg_first, STBI__F_paeth_first}; -static int stbi__paeth(int a, int b, int c) -{ - int p = a + b - c; - int pa = abs(p-a); - int pb = abs(p-b); - int pc = abs(p-c); - if (pa <= pb && pa <= pc) return a; - if (pb <= pc) return b; - return c; +static int stbi__paeth(int a, int b, int c) { + int p = a + b - c; + int pa = abs(p - a); + int pb = abs(p - b); + int pc = abs(p - c); + if (pa <= pb && pa <= pc) return a; + if (pb <= pc) return b; + return c; } -static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; +static const stbi_uc stbi__depth_scale_table[9] = {0, 0xff, 0x55, 0, 0x11, + 0, 0, 0, 0x01}; // create the png data from post-deflated data -static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y, int depth, int color) -{ - int bytes = (depth == 16? 2 : 1); - stbi__context *s = a->s; - stbi__uint32 i,j,stride = x*out_n*bytes; - stbi__uint32 img_len, img_width_bytes; - int k; - int img_n = s->img_n; // copy it into a local for later +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, + stbi__uint32 raw_len, int out_n, + stbi__uint32 x, stbi__uint32 y, int depth, + int color) { + int bytes = (depth == 16 ? 2 : 1); + stbi__context *s = a->s; + stbi__uint32 i, j, stride = x * out_n * bytes; + stbi__uint32 img_len, img_width_bytes; + int k; + int img_n = s->img_n; // copy it into a local for later - int output_bytes = out_n*bytes; - int filter_bytes = img_n*bytes; - int width = x; + int output_bytes = out_n * bytes; + int filter_bytes = img_n * bytes; + int width = x; - STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); - a->out = (stbi_uc *) stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into - if (!a->out) return stbi__err("outofmem", "Out of memory"); + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n + 1); + a->out = (stbi_uc *)stbi__malloc_mad3( + x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); - if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); - img_width_bytes = (((img_n * x * depth) + 7) >> 3); - img_len = (img_width_bytes + 1) * y; + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) + return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; - // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, - // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), - // so just check for raw_len < img_len always. - if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + // we used to check for exact match between raw_len and img_len on + // non-interlaced PNGs, but issue #276 reported a PNG in the wild that had + // extra data at the end (all zeros), so just check for raw_len < img_len + // always. + if (raw_len < img_len) return stbi__err("not enough pixels", "Corrupt PNG"); - for (j=0; j < y; ++j) { - stbi_uc *cur = a->out + stride*j; - stbi_uc *prior; - int filter = *raw++; + for (j = 0; j < y; ++j) { + stbi_uc *cur = a->out + stride * j; + stbi_uc *prior; + int filter = *raw++; - if (filter > 4) - return stbi__err("invalid filter","Corrupt PNG"); + if (filter > 4) return stbi__err("invalid filter", "Corrupt PNG"); - if (depth < 8) { - if (img_width_bytes > x) return stbi__err("invalid width","Corrupt PNG"); - cur += x*out_n - img_width_bytes; // store output to the rightmost img_len bytes, so we can decode in place - filter_bytes = 1; - width = img_width_bytes; + if (depth < 8) { + if (img_width_bytes > x) return stbi__err("invalid width", "Corrupt PNG"); + cur += + x * out_n - img_width_bytes; // store output to the rightmost img_len + // bytes, so we can decode in place + filter_bytes = 1; + width = img_width_bytes; + } + prior = cur - stride; // bugfix: need to compute this after 'cur +=' + // computation above + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // handle first byte explicitly + for (k = 0; k < filter_bytes; ++k) { + switch (filter) { + case STBI__F_none: + cur[k] = raw[k]; + break; + case STBI__F_sub: + cur[k] = raw[k]; + break; + case STBI__F_up: + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + break; + case STBI__F_avg: + cur[k] = STBI__BYTECAST(raw[k] + (prior[k] >> 1)); + break; + case STBI__F_paeth: + cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0, prior[k], 0)); + break; + case STBI__F_avg_first: + cur[k] = raw[k]; + break; + case STBI__F_paeth_first: + cur[k] = raw[k]; + break; } - prior = cur - stride; // bugfix: need to compute this after 'cur +=' computation above + } - // if first row, use special filter that doesn't sample previous row - if (j == 0) filter = first_row_filter[filter]; - - // handle first byte explicitly - for (k=0; k < filter_bytes; ++k) { - switch (filter) { - case STBI__F_none : cur[k] = raw[k]; break; - case STBI__F_sub : cur[k] = raw[k]; break; - case STBI__F_up : cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; - case STBI__F_avg : cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); break; - case STBI__F_paeth : cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0,prior[k],0)); break; - case STBI__F_avg_first : cur[k] = raw[k]; break; - case STBI__F_paeth_first: cur[k] = raw[k]; break; - } + if (depth == 8) { + if (img_n != out_n) cur[img_n] = 255; // first pixel + raw += img_n; + cur += out_n; + prior += out_n; + } else if (depth == 16) { + if (img_n != out_n) { + cur[filter_bytes] = 255; // first pixel top byte + cur[filter_bytes + 1] = 255; // first pixel bottom byte } + raw += filter_bytes; + cur += output_bytes; + prior += output_bytes; + } else { + raw += 1; + cur += 1; + prior += 1; + } - if (depth == 8) { - if (img_n != out_n) - cur[img_n] = 255; // first pixel - raw += img_n; - cur += out_n; - prior += out_n; - } else if (depth == 16) { - if (img_n != out_n) { - cur[filter_bytes] = 255; // first pixel top byte - cur[filter_bytes+1] = 255; // first pixel bottom byte - } - raw += filter_bytes; - cur += output_bytes; - prior += output_bytes; - } else { - raw += 1; - cur += 1; - prior += 1; + // this is a little gross, so that we don't switch per-pixel or + // per-component + if (depth < 8 || img_n == out_n) { + int nk = (width - 1) * filter_bytes; + #define STBI__CASE(f) \ + case f: \ + for (k = 0; k < nk; ++k) + switch (filter) { + // "none" filter turns into a memcpy here; make that explicit. + case STBI__F_none: + memcpy(cur, raw, nk); + break; + STBI__CASE(STBI__F_sub) { + cur[k] = STBI__BYTECAST(raw[k] + cur[k - filter_bytes]); + } + break; + STBI__CASE(STBI__F_up) { + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + } + break; + STBI__CASE(STBI__F_avg) { + cur[k] = STBI__BYTECAST(raw[k] + + ((prior[k] + cur[k - filter_bytes]) >> 1)); + } + break; + STBI__CASE(STBI__F_paeth) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - filter_bytes], prior[k], + prior[k - filter_bytes])); + } + break; + STBI__CASE(STBI__F_avg_first) { + cur[k] = STBI__BYTECAST(raw[k] + (cur[k - filter_bytes] >> 1)); + } + break; + STBI__CASE(STBI__F_paeth_first) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - filter_bytes], 0, 0)); + } + break; } - - // this is a little gross, so that we don't switch per-pixel or per-component - if (depth < 8 || img_n == out_n) { - int nk = (width - 1)*filter_bytes; - #define STBI__CASE(f) \ - case f: \ - for (k=0; k < nk; ++k) - switch (filter) { - // "none" filter turns into a memcpy here; make that explicit. - case STBI__F_none: memcpy(cur, raw, nk); break; - STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); } break; - STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; - STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); } break; - STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],prior[k],prior[k-filter_bytes])); } break; - STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); } break; - STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],0,0)); } break; - } - #undef STBI__CASE - raw += nk; - } else { - STBI_ASSERT(img_n+1 == out_n); - #define STBI__CASE(f) \ - case f: \ - for (i=x-1; i >= 1; --i, cur[filter_bytes]=255,raw+=filter_bytes,cur+=output_bytes,prior+=output_bytes) \ - for (k=0; k < filter_bytes; ++k) - switch (filter) { - STBI__CASE(STBI__F_none) { cur[k] = raw[k]; } break; - STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k- output_bytes]); } break; - STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; - STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k- output_bytes])>>1)); } break; - STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],prior[k],prior[k- output_bytes])); } break; - STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k- output_bytes] >> 1)); } break; - STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],0,0)); } break; - } - #undef STBI__CASE - - // the loop above sets the high byte of the pixels' alpha, but for - // 16 bit png files we also need the low byte set. we'll do that here. - if (depth == 16) { - cur = a->out + stride*j; // start at the beginning of the row again - for (i=0; i < x; ++i,cur+=output_bytes) { - cur[filter_bytes+1] = 255; - } - } + #undef STBI__CASE + raw += nk; + } else { + STBI_ASSERT(img_n + 1 == out_n); + #define STBI__CASE(f) \ + case f: \ + for (i = x - 1; i >= 1; --i, cur[filter_bytes] = 255, \ + raw += filter_bytes, cur += output_bytes, prior += output_bytes) \ + for (k = 0; k < filter_bytes; ++k) + switch (filter) { + STBI__CASE(STBI__F_none) { + cur[k] = raw[k]; + } + break; + STBI__CASE(STBI__F_sub) { + cur[k] = STBI__BYTECAST(raw[k] + cur[k - output_bytes]); + } + break; + STBI__CASE(STBI__F_up) { + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + } + break; + STBI__CASE(STBI__F_avg) { + cur[k] = STBI__BYTECAST(raw[k] + + ((prior[k] + cur[k - output_bytes]) >> 1)); + } + break; + STBI__CASE(STBI__F_paeth) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - output_bytes], prior[k], + prior[k - output_bytes])); + } + break; + STBI__CASE(STBI__F_avg_first) { + cur[k] = STBI__BYTECAST(raw[k] + (cur[k - output_bytes] >> 1)); + } + break; + STBI__CASE(STBI__F_paeth_first) { + cur[k] = + STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - output_bytes], 0, 0)); + } + break; } - } + #undef STBI__CASE - // we make a separate pass to expand bits to pixels; for performance, - // this could run two scanlines behind the above code, so it won't - // intefere with filtering but will still be in the cache. - if (depth < 8) { - for (j=0; j < y; ++j) { - stbi_uc *cur = a->out + stride*j; - stbi_uc *in = a->out + stride*j + x*out_n - img_width_bytes; - // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common 8-bit path optimal at minimal cost for 1/2/4-bit - // png guarante byte alignment, if width is not multiple of 8/4/2 we'll decode dummy trailing data that will be skipped in the later loop - stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range - - // note that the final byte might overshoot and write more data than desired. - // we can allocate enough data that this never writes out of memory, but it - // could also overwrite the next scanline. can it overwrite non-empty data - // on the next scanline? yes, consider 1-pixel-wide scanlines with 1-bit-per-pixel. - // so we need to explicitly clamp the final ones - - if (depth == 4) { - for (k=x*img_n; k >= 2; k-=2, ++in) { - *cur++ = scale * ((*in >> 4) ); - *cur++ = scale * ((*in ) & 0x0f); - } - if (k > 0) *cur++ = scale * ((*in >> 4) ); - } else if (depth == 2) { - for (k=x*img_n; k >= 4; k-=4, ++in) { - *cur++ = scale * ((*in >> 6) ); - *cur++ = scale * ((*in >> 4) & 0x03); - *cur++ = scale * ((*in >> 2) & 0x03); - *cur++ = scale * ((*in ) & 0x03); - } - if (k > 0) *cur++ = scale * ((*in >> 6) ); - if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); - if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); - } else if (depth == 1) { - for (k=x*img_n; k >= 8; k-=8, ++in) { - *cur++ = scale * ((*in >> 7) ); - *cur++ = scale * ((*in >> 6) & 0x01); - *cur++ = scale * ((*in >> 5) & 0x01); - *cur++ = scale * ((*in >> 4) & 0x01); - *cur++ = scale * ((*in >> 3) & 0x01); - *cur++ = scale * ((*in >> 2) & 0x01); - *cur++ = scale * ((*in >> 1) & 0x01); - *cur++ = scale * ((*in ) & 0x01); - } - if (k > 0) *cur++ = scale * ((*in >> 7) ); - if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); - if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); - if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); - if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); - if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); - if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); - } - if (img_n != out_n) { - int q; - // insert alpha = 255 - cur = a->out + stride*j; - if (img_n == 1) { - for (q=x-1; q >= 0; --q) { - cur[q*2+1] = 255; - cur[q*2+0] = cur[q]; - } - } else { - STBI_ASSERT(img_n == 3); - for (q=x-1; q >= 0; --q) { - cur[q*4+3] = 255; - cur[q*4+2] = cur[q*3+2]; - cur[q*4+1] = cur[q*3+1]; - cur[q*4+0] = cur[q*3+0]; - } - } - } + // the loop above sets the high byte of the pixels' alpha, but for + // 16 bit png files we also need the low byte set. we'll do that here. + if (depth == 16) { + cur = a->out + stride * j; // start at the beginning of the row again + for (i = 0; i < x; ++i, cur += output_bytes) { + cur[filter_bytes + 1] = 255; + } } - } else if (depth == 16) { - // force the image data from big-endian to platform-native. - // this is done in a separate pass due to the decoding relying - // on the data being untouched, but could probably be done - // per-line during decode if care is taken. - stbi_uc *cur = a->out; - stbi__uint16 *cur16 = (stbi__uint16*)cur; + } + } - for(i=0; i < x*y*out_n; ++i,cur16++,cur+=2) { - *cur16 = (cur[0] << 8) | cur[1]; + // we make a separate pass to expand bits to pixels; for performance, + // this could run two scanlines behind the above code, so it won't + // intefere with filtering but will still be in the cache. + if (depth < 8) { + for (j = 0; j < y; ++j) { + stbi_uc *cur = a->out + stride * j; + stbi_uc *in = a->out + stride * j + x * out_n - img_width_bytes; + // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common + // 8-bit path optimal at minimal cost for 1/2/4-bit png guarante byte + // alignment, if width is not multiple of 8/4/2 we'll decode dummy + // trailing data that will be skipped in the later loop + stbi_uc scale = (color == 0) + ? stbi__depth_scale_table[depth] + : 1; // scale grayscale values to 0..255 range + + // note that the final byte might overshoot and write more data than + // desired. we can allocate enough data that this never writes out of + // memory, but it could also overwrite the next scanline. can it overwrite + // non-empty data on the next scanline? yes, consider 1-pixel-wide + // scanlines with 1-bit-per-pixel. so we need to explicitly clamp the + // final ones + + if (depth == 4) { + for (k = x * img_n; k >= 2; k -= 2, ++in) { + *cur++ = scale * ((*in >> 4)); + *cur++ = scale * ((*in) & 0x0f); + } + if (k > 0) *cur++ = scale * ((*in >> 4)); + } else if (depth == 2) { + for (k = x * img_n; k >= 4; k -= 4, ++in) { + *cur++ = scale * ((*in >> 6)); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in) & 0x03); + } + if (k > 0) *cur++ = scale * ((*in >> 6)); + if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); + } else if (depth == 1) { + for (k = x * img_n; k >= 8; k -= 8, ++in) { + *cur++ = scale * ((*in >> 7)); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in) & 0x01); + } + if (k > 0) *cur++ = scale * ((*in >> 7)); + if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); } - } + if (img_n != out_n) { + int q; + // insert alpha = 255 + cur = a->out + stride * j; + if (img_n == 1) { + for (q = x - 1; q >= 0; --q) { + cur[q * 2 + 1] = 255; + cur[q * 2 + 0] = cur[q]; + } + } else { + STBI_ASSERT(img_n == 3); + for (q = x - 1; q >= 0; --q) { + cur[q * 4 + 3] = 255; + cur[q * 4 + 2] = cur[q * 3 + 2]; + cur[q * 4 + 1] = cur[q * 3 + 1]; + cur[q * 4 + 0] = cur[q * 3 + 0]; + } + } + } + } + } else if (depth == 16) { + // force the image data from big-endian to platform-native. + // this is done in a separate pass due to the decoding relying + // on the data being untouched, but could probably be done + // per-line during decode if care is taken. + stbi_uc *cur = a->out; + stbi__uint16 *cur16 = (stbi__uint16 *)cur; - return 1; + for (i = 0; i < x * y * out_n; ++i, cur16++, cur += 2) { + *cur16 = (cur[0] << 8) | cur[1]; + } + } + + return 1; } -static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) -{ - int bytes = (depth == 16 ? 2 : 1); - int out_bytes = out_n * bytes; - stbi_uc *final; - int p; - if (!interlaced) - return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); +static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, + stbi__uint32 image_data_len, int out_n, + int depth, int color, int interlaced) { + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, + a->s->img_x, a->s->img_y, depth, color); - // de-interlacing - final = (stbi_uc *) stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); - for (p=0; p < 7; ++p) { - int xorig[] = { 0,4,0,2,0,1,0 }; - int yorig[] = { 0,0,4,0,2,0,1 }; - int xspc[] = { 8,8,4,4,2,2,1 }; - int yspc[] = { 8,8,8,4,4,2,2 }; - int i,j,x,y; - // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 - x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; - y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; - if (x && y) { - stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; - if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { - STBI_FREE(final); - return 0; - } - for (j=0; j < y; ++j) { - for (i=0; i < x; ++i) { - int out_y = j*yspc[p]+yorig[p]; - int out_x = i*xspc[p]+xorig[p]; - memcpy(final + out_y*a->s->img_x*out_bytes + out_x*out_bytes, - a->out + (j*x+i)*out_bytes, out_bytes); - } - } - STBI_FREE(a->out); - image_data += img_len; - image_data_len -= img_len; + // de-interlacing + final = (stbi_uc *)stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + for (p = 0; p < 7; ++p) { + int xorig[] = {0, 4, 0, 2, 0, 1, 0}; + int yorig[] = {0, 0, 4, 0, 2, 0, 1}; + int xspc[] = {8, 8, 4, 4, 2, 2, 1}; + int yspc[] = {8, 8, 8, 4, 4, 2, 2}; + int i, j, x, y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p] - 1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p] - 1) / yspc[p]; + if (x && y) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, + y, depth, color)) { + STBI_FREE(final); + return 0; } - } - a->out = final; + for (j = 0; j < y; ++j) { + for (i = 0; i < x; ++i) { + int out_y = j * yspc[p] + yorig[p]; + int out_x = i * xspc[p] + xorig[p]; + memcpy(final + out_y * a->s->img_x * out_bytes + out_x * out_bytes, + a->out + (j * x + i) * out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; - return 1; + return 1; } -static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi_uc *p = z->out; +static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; - // compute color-based transparency, assuming we've - // already got 255 as the alpha value in the output - STBI_ASSERT(out_n == 2 || out_n == 4); + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); - if (out_n == 2) { - for (i=0; i < pixel_count; ++i) { - p[1] = (p[0] == tc[0] ? 0 : 255); - p += 2; - } - } else { - for (i=0; i < pixel_count; ++i) { - if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) - p[3] = 0; - p += 4; - } - } - return 1; + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 255); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) p[3] = 0; + p += 4; + } + } + return 1; } -static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], int out_n) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi__uint16 *p = (stbi__uint16*) z->out; +static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], + int out_n) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16 *p = (stbi__uint16 *)z->out; - // compute color-based transparency, assuming we've - // already got 65535 as the alpha value in the output - STBI_ASSERT(out_n == 2 || out_n == 4); + // compute color-based transparency, assuming we've + // already got 65535 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); - if (out_n == 2) { - for (i = 0; i < pixel_count; ++i) { - p[1] = (p[0] == tc[0] ? 0 : 65535); - p += 2; - } - } else { - for (i = 0; i < pixel_count; ++i) { - if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) - p[3] = 0; - p += 4; - } - } - return 1; + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 65535); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) p[3] = 0; + p += 4; + } + } + return 1; } -static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, int pal_img_n) -{ - stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; - stbi_uc *p, *temp_out, *orig = a->out; +static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, + int pal_img_n) { + stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; + stbi_uc *p, *temp_out, *orig = a->out; - p = (stbi_uc *) stbi__malloc_mad2(pixel_count, pal_img_n, 0); - if (p == NULL) return stbi__err("outofmem", "Out of memory"); + p = (stbi_uc *)stbi__malloc_mad2(pixel_count, pal_img_n, 0); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); - // between here and free(out) below, exitting would leak - temp_out = p; + // between here and free(out) below, exitting would leak + temp_out = p; - if (pal_img_n == 3) { - for (i=0; i < pixel_count; ++i) { - int n = orig[i]*4; - p[0] = palette[n ]; - p[1] = palette[n+1]; - p[2] = palette[n+2]; - p += 3; - } - } else { - for (i=0; i < pixel_count; ++i) { - int n = orig[i]*4; - p[0] = palette[n ]; - p[1] = palette[n+1]; - p[2] = palette[n+2]; - p[3] = palette[n+3]; - p += 4; - } - } - STBI_FREE(a->out); - a->out = temp_out; + if (pal_img_n == 3) { + for (i = 0; i < pixel_count; ++i) { + int n = orig[i] * 4; + p[0] = palette[n]; + p[1] = palette[n + 1]; + p[2] = palette[n + 2]; + p += 3; + } + } else { + for (i = 0; i < pixel_count; ++i) { + int n = orig[i] * 4; + p[0] = palette[n]; + p[1] = palette[n + 1]; + p[2] = palette[n + 2]; + p[3] = palette[n + 3]; + p += 4; + } + } + STBI_FREE(a->out); + a->out = temp_out; - STBI_NOTUSED(len); + STBI_NOTUSED(len); - return 1; + return 1; } static int stbi__unpremultiply_on_load = 0; static int stbi__de_iphone_flag = 0; -STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) -{ - stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; +STBIDEF void stbi_set_unpremultiply_on_load( + int flag_true_if_should_unpremultiply) { + stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; } -STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) -{ - stbi__de_iphone_flag = flag_true_if_should_convert; +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) { + stbi__de_iphone_flag = flag_true_if_should_convert; } -static void stbi__de_iphone(stbi__png *z) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi_uc *p = z->out; +static void stbi__de_iphone(stbi__png *z) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; - if (s->img_out_n == 3) { // convert bgr to rgb - for (i=0; i < pixel_count; ++i) { - stbi_uc t = p[0]; - p[0] = p[2]; - p[2] = t; - p += 3; + if (s->img_out_n == 3) { // convert bgr to rgb + for (i = 0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } else { + STBI_ASSERT(s->img_out_n == 4); + if (stbi__unpremultiply_on_load) { + // convert bgr to rgb and unpremultiply + for (i = 0; i < pixel_count; ++i) { + stbi_uc a = p[3]; + stbi_uc t = p[0]; + if (a) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = (t * 255 + half) / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; } - } else { - STBI_ASSERT(s->img_out_n == 4); - if (stbi__unpremultiply_on_load) { - // convert bgr to rgb and unpremultiply - for (i=0; i < pixel_count; ++i) { - stbi_uc a = p[3]; - stbi_uc t = p[0]; - if (a) { - stbi_uc half = a / 2; - p[0] = (p[2] * 255 + half) / a; - p[1] = (p[1] * 255 + half) / a; - p[2] = ( t * 255 + half) / a; - } else { - p[0] = p[2]; - p[2] = t; - } - p += 4; - } - } else { - // convert bgr to rgb - for (i=0; i < pixel_count; ++i) { - stbi_uc t = p[0]; - p[0] = p[2]; - p[2] = t; - p += 4; - } + } else { + // convert bgr to rgb + for (i = 0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; } - } + } + } } -#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + #define STBI__PNG_TYPE(a, b, c, d) \ + (((unsigned)(a) << 24) + ((unsigned)(b) << 16) + ((unsigned)(c) << 8) + \ + (unsigned)(d)) -static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) -{ - stbi_uc palette[1024], pal_img_n=0; - stbi_uc has_trans=0, tc[3]={0}; - stbi__uint16 tc16[3]; - stbi__uint32 ioff=0, idata_limit=0, i, pal_len=0; - int first=1,k,interlace=0, color=0, is_iphone=0; - stbi__context *s = z->s; +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) { + stbi_uc palette[1024], pal_img_n = 0; + stbi_uc has_trans = 0, tc[3] = {0}; + stbi__uint16 tc16[3]; + stbi__uint32 ioff = 0, idata_limit = 0, i, pal_len = 0; + int first = 1, k, interlace = 0, color = 0, is_iphone = 0; + stbi__context *s = z->s; - z->expanded = NULL; - z->idata = NULL; - z->out = NULL; + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; - if (!stbi__check_png_header(s)) return 0; + if (!stbi__check_png_header(s)) return 0; - if (scan == STBI__SCAN_type) return 1; + if (scan == STBI__SCAN_type) return 1; - for (;;) { - stbi__pngchunk c = stbi__get_chunk_header(s); - switch (c.type) { - case STBI__PNG_TYPE('C','g','B','I'): - is_iphone = 1; - stbi__skip(s, c.length); - break; - case STBI__PNG_TYPE('I','H','D','R'): { - int comp,filter; - if (!first) return stbi__err("multiple IHDR","Corrupt PNG"); - first = 0; - if (c.length != 13) return stbi__err("bad IHDR len","Corrupt PNG"); - s->img_x = stbi__get32be(s); - s->img_y = stbi__get32be(s); - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only","PNG not supported: 1/2/4/8/16-bit only"); - color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype","Corrupt PNG"); - if (color == 3 && z->depth == 16) return stbi__err("bad ctype","Corrupt PNG"); - if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype","Corrupt PNG"); - comp = stbi__get8(s); if (comp) return stbi__err("bad comp method","Corrupt PNG"); - filter= stbi__get8(s); if (filter) return stbi__err("bad filter method","Corrupt PNG"); - interlace = stbi__get8(s); if (interlace>1) return stbi__err("bad interlace method","Corrupt PNG"); - if (!s->img_x || !s->img_y) return stbi__err("0-pixel image","Corrupt PNG"); - if (!pal_img_n) { - s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); - if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); - if (scan == STBI__SCAN_header) return 1; - } else { - // if paletted, then pal_n is our final components, and - // img_n is # components to decompress/filter. - s->img_n = 1; - if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); - // if SCAN_header, have to scan to see if we have a tRNS - } - break; - } + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C', 'g', 'B', 'I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I', 'H', 'D', 'R'): { + int comp, filter; + if (!first) return stbi__err("multiple IHDR", "Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len", "Corrupt PNG"); + s->img_x = stbi__get32be(s); + s->img_y = stbi__get32be(s); + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + z->depth = stbi__get8(s); + if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && + z->depth != 16) + return stbi__err("1/2/4/8/16-bit only", + "PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); + if (color > 6) return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3 && z->depth == 16) + return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3) + pal_img_n = 3; + else if (color & 1) + return stbi__err("bad ctype", "Corrupt PNG"); + comp = stbi__get8(s); + if (comp) return stbi__err("bad comp method", "Corrupt PNG"); + filter = stbi__get8(s); + if (filter) return stbi__err("bad filter method", "Corrupt PNG"); + interlace = stbi__get8(s); + if (interlace > 1) + return stbi__err("bad interlace method", "Corrupt PNG"); + if (!s->img_x || !s->img_y) + return stbi__err("0-pixel image", "Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) + return stbi__err("too large", "Image too large to decode"); + if (scan == STBI__SCAN_header) return 1; + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) + return stbi__err("too large", "Corrupt PNG"); + // if SCAN_header, have to scan to see if we have a tRNS + } + break; + } - case STBI__PNG_TYPE('P','L','T','E'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (c.length > 256*3) return stbi__err("invalid PLTE","Corrupt PNG"); - pal_len = c.length / 3; - if (pal_len * 3 != c.length) return stbi__err("invalid PLTE","Corrupt PNG"); - for (i=0; i < pal_len; ++i) { - palette[i*4+0] = stbi__get8(s); - palette[i*4+1] = stbi__get8(s); - palette[i*4+2] = stbi__get8(s); - palette[i*4+3] = 255; - } - break; - } + case STBI__PNG_TYPE('P', 'L', 'T', 'E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256 * 3) return stbi__err("invalid PLTE", "Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) + return stbi__err("invalid PLTE", "Corrupt PNG"); + for (i = 0; i < pal_len; ++i) { + palette[i * 4 + 0] = stbi__get8(s); + palette[i * 4 + 1] = stbi__get8(s); + palette[i * 4 + 2] = stbi__get8(s); + palette[i * 4 + 3] = 255; + } + break; + } - case STBI__PNG_TYPE('t','R','N','S'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (z->idata) return stbi__err("tRNS after IDAT","Corrupt PNG"); - if (pal_img_n) { - if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } - if (pal_len == 0) return stbi__err("tRNS before PLTE","Corrupt PNG"); - if (c.length > pal_len) return stbi__err("bad tRNS len","Corrupt PNG"); - pal_img_n = 4; - for (i=0; i < c.length; ++i) - palette[i*4+3] = stbi__get8(s); - } else { - if (!(s->img_n & 1)) return stbi__err("tRNS with alpha","Corrupt PNG"); - if (c.length != (stbi__uint32) s->img_n*2) return stbi__err("bad tRNS len","Corrupt PNG"); - has_trans = 1; - if (z->depth == 16) { - for (k = 0; k < s->img_n; ++k) tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is - } else { - for (k = 0; k < s->img_n; ++k) tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger - } - } - break; - } - - case STBI__PNG_TYPE('I','D','A','T'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); - if (scan == STBI__SCAN_header) { s->img_n = pal_img_n; return 1; } - if ((int)(ioff + c.length) < (int)ioff) return 0; - if (ioff + c.length > idata_limit) { - stbi__uint32 idata_limit_old = idata_limit; - stbi_uc *p; - if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; - while (ioff + c.length > idata_limit) - idata_limit *= 2; - STBI_NOTUSED(idata_limit_old); - p = (stbi_uc *) STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); - z->idata = p; - } - if (!stbi__getn(s, z->idata+ioff,c.length)) return stbi__err("outofdata","Corrupt PNG"); - ioff += c.length; - break; - } - - case STBI__PNG_TYPE('I','E','N','D'): { - stbi__uint32 raw_len, bpl; - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (scan != STBI__SCAN_load) return 1; - if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); - // initial guess for decoded data size to avoid unnecessary reallocs - bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component - raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; - z->expanded = (stbi_uc *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, raw_len, (int *) &raw_len, !is_iphone); - if (z->expanded == NULL) return 0; // zlib should set error - STBI_FREE(z->idata); z->idata = NULL; - if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) - s->img_out_n = s->img_n+1; - else - s->img_out_n = s->img_n; - if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, z->depth, color, interlace)) return 0; - if (has_trans) { - if (z->depth == 16) { - if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; - } else { - if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; - } - } - if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) - stbi__de_iphone(z); - if (pal_img_n) { - // pal_img_n == 3 or 4 - s->img_n = pal_img_n; // record the actual colors we had - s->img_out_n = pal_img_n; - if (req_comp >= 3) s->img_out_n = req_comp; - if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) - return 0; - } else if (has_trans) { - // non-paletted image with tRNS -> source image has (constant) alpha - ++s->img_n; - } - STBI_FREE(z->expanded); z->expanded = NULL; - // end of PNG chunk, read and skip CRC - stbi__get32be(s); + case STBI__PNG_TYPE('t', 'R', 'N', 'S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT", "Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { + s->img_n = 4; return 1; - } - - default: - // if critical, fail - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if ((c.type & (1 << 29)) == 0) { - #ifndef STBI_NO_FAILURE_STRINGS - // not threadsafe - static char invalid_chunk[] = "XXXX PNG chunk not known"; - invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); - invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); - invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); - invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); - #endif - return stbi__err(invalid_chunk, "PNG not supported: unknown PNG chunk type"); - } - stbi__skip(s, c.length); - break; + } + if (pal_len == 0) return stbi__err("tRNS before PLTE", "Corrupt PNG"); + if (c.length > pal_len) + return stbi__err("bad tRNS len", "Corrupt PNG"); + pal_img_n = 4; + for (i = 0; i < c.length; ++i) + palette[i * 4 + 3] = stbi__get8(s); + } else { + if (!(s->img_n & 1)) + return stbi__err("tRNS with alpha", "Corrupt PNG"); + if (c.length != (stbi__uint32)s->img_n * 2) + return stbi__err("bad tRNS len", "Corrupt PNG"); + has_trans = 1; + if (z->depth == 16) { + for (k = 0; k < s->img_n; ++k) + tc16[k] = + (stbi__uint16)stbi__get16be(s); // copy the values as-is + } else { + for (k = 0; k < s->img_n; ++k) + tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * + stbi__depth_scale_table[z->depth]; // non 8-bit images + // will be larger + } + } + break; } - // end of PNG chunk, read and skip CRC - stbi__get32be(s); - } + + case STBI__PNG_TYPE('I', 'D', 'A', 'T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE", "Corrupt PNG"); + if (scan == STBI__SCAN_header) { + s->img_n = pal_img_n; + return 1; + } + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = idata_limit; + stbi_uc *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc *)STBI_REALLOC_SIZED(z->idata, idata_limit_old, + idata_limit); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata + ioff, c.length)) + return stbi__err("outofdata", "Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I', 'E', 'N', 'D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT", "Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc *)stbi_zlib_decode_malloc_guesssize_headerflag( + (char *)z->idata, ioff, raw_len, (int *)&raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_FREE(z->idata); + z->idata = NULL; + if ((req_comp == s->img_n + 1 && req_comp != 3 && !pal_img_n) || + has_trans) + s->img_out_n = s->img_n + 1; + else + s->img_out_n = s->img_n; + if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, + z->depth, color, interlace)) + return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } else { + if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; + } + } + if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) + stbi__de_iphone(z); + if (pal_img_n) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_FREE(z->expanded); + z->expanded = NULL; + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + return 1; + } + + default: + // if critical, fail + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + static char invalid_chunk[] = "XXXX PNG chunk not known"; + invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); + invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); + invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); + invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); + #endif + return stbi__err(invalid_chunk, + "PNG not supported: unknown PNG chunk type"); + } + stbi__skip(s, c.length); + break; + } + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + } } -static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, stbi__result_info *ri) -{ - void *result=NULL; - if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); - if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { - if (p->depth <= 8) - ri->bits_per_channel = 8; - else if (p->depth == 16) - ri->bits_per_channel = 16; +static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, + stbi__result_info *ri) { + void *result = NULL; + if (req_comp < 0 || req_comp > 4) + return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth <= 8) + ri->bits_per_channel = 8; + else if (p->depth == 16) + ri->bits_per_channel = 16; + else + return stbi__errpuc("bad bits_per_channel", + "PNG not supported: unsupported color depth"); + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char *)result, p->s->img_out_n, + req_comp, p->s->img_x, p->s->img_y); else - return stbi__errpuc("bad bits_per_channel", "PNG not supported: unsupported color depth"); - result = p->out; - p->out = NULL; - if (req_comp && req_comp != p->s->img_out_n) { - if (ri->bits_per_channel == 8) - result = stbi__convert_format((unsigned char *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); - else - result = stbi__convert_format16((stbi__uint16 *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); - p->s->img_out_n = req_comp; - if (result == NULL) return result; - } - *x = p->s->img_x; - *y = p->s->img_y; - if (n) *n = p->s->img_n; - } - STBI_FREE(p->out); p->out = NULL; - STBI_FREE(p->expanded); p->expanded = NULL; - STBI_FREE(p->idata); p->idata = NULL; + result = stbi__convert_format16((stbi__uint16 *)result, p->s->img_out_n, + req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n) *n = p->s->img_n; + } + STBI_FREE(p->out); + p->out = NULL; + STBI_FREE(p->expanded); + p->expanded = NULL; + STBI_FREE(p->idata); + p->idata = NULL; - return result; + return result; } -static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi__png p; - p.s = s; - return stbi__do_png(&p, x,y,comp,req_comp, ri); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi__png p; + p.s = s; + return stbi__do_png(&p, x, y, comp, req_comp, ri); } -static int stbi__png_test(stbi__context *s) -{ - int r; - r = stbi__check_png_header(s); - stbi__rewind(s); - return r; +static int stbi__png_test(stbi__context *s) { + int r; + r = stbi__check_png_header(s); + stbi__rewind(s); + return r; } -static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) -{ - if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { - stbi__rewind( p->s ); - return 0; - } - if (x) *x = p->s->img_x; - if (y) *y = p->s->img_y; - if (comp) *comp = p->s->img_n; - return 1; +static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) { + if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { + stbi__rewind(p->s); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; } -static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) -{ - stbi__png p; - p.s = s; - return stbi__png_info_raw(&p, x, y, comp); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) { + stbi__png p; + p.s = s; + return stbi__png_info_raw(&p, x, y, comp); } static int stbi__png_is16(stbi__context *s) @@ -5191,958 +5662,1015 @@ static int stbi__png_is16(stbi__context *s) } return 1; } -#endif + #endif // Microsoft/Windows BMP image -#ifndef STBI_NO_BMP -static int stbi__bmp_test_raw(stbi__context *s) -{ - int r; - int sz; - if (stbi__get8(s) != 'B') return 0; - if (stbi__get8(s) != 'M') return 0; - stbi__get32le(s); // discard filesize - stbi__get16le(s); // discard reserved - stbi__get16le(s); // discard reserved - stbi__get32le(s); // discard data offset - sz = stbi__get32le(s); - r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); - return r; + #ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context *s) { + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + stbi__get32le(s); // discard data offset + sz = stbi__get32le(s); + r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); + return r; } -static int stbi__bmp_test(stbi__context *s) -{ - int r = stbi__bmp_test_raw(s); - stbi__rewind(s); - return r; +static int stbi__bmp_test(stbi__context *s) { + int r = stbi__bmp_test_raw(s); + stbi__rewind(s); + return r; } - // returns 0..31 for the highest set bit -static int stbi__high_bit(unsigned int z) -{ - int n=0; - if (z == 0) return -1; - if (z >= 0x10000) { n += 16; z >>= 16; } - if (z >= 0x00100) { n += 8; z >>= 8; } - if (z >= 0x00010) { n += 4; z >>= 4; } - if (z >= 0x00004) { n += 2; z >>= 2; } - if (z >= 0x00002) { n += 1;/* >>= 1;*/ } - return n; +static int stbi__high_bit(unsigned int z) { + int n = 0; + if (z == 0) return -1; + if (z >= 0x10000) { + n += 16; + z >>= 16; + } + if (z >= 0x00100) { + n += 8; + z >>= 8; + } + if (z >= 0x00010) { + n += 4; + z >>= 4; + } + if (z >= 0x00004) { + n += 2; + z >>= 2; + } + if (z >= 0x00002) { n += 1; /* >>= 1;*/ } + return n; } -static int stbi__bitcount(unsigned int a) -{ - a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 - a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 - a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits - a = (a + (a >> 8)); // max 16 per 8 bits - a = (a + (a >> 16)); // max 32 per 8 bits - return a & 0xff; +static int stbi__bitcount(unsigned int a) { + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; } // extract an arbitrarily-aligned N-bit value (N=bits) // from v, and then make it 8-bits long and fractionally // extend it to full full range. -static int stbi__shiftsigned(unsigned int v, int shift, int bits) -{ - static unsigned int mul_table[9] = { +static int stbi__shiftsigned(unsigned int v, int shift, int bits) { + static unsigned int mul_table[9] = { 0, - 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, - 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, - }; - static unsigned int shift_table[9] = { - 0, 0,0,1,0,2,4,6,0, - }; - if (shift < 0) - v <<= -shift; - else - v >>= shift; - STBI_ASSERT(v < 256); - v >>= (8-bits); - STBI_ASSERT(bits >= 0 && bits <= 8); - return (int) ((unsigned) v * mul_table[bits]) >> shift_table[bits]; + 0xff /*0b11111111*/, + 0x55 /*0b01010101*/, + 0x49 /*0b01001001*/, + 0x11 /*0b00010001*/, + 0x21 /*0b00100001*/, + 0x41 /*0b01000001*/, + 0x81 /*0b10000001*/, + 0x01 /*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0, 0, 1, 0, 2, 4, 6, 0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v < 256); + v >>= (8 - bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int)((unsigned)v * mul_table[bits]) >> shift_table[bits]; } -typedef struct -{ - int bpp, offset, hsz; - unsigned int mr,mg,mb,ma, all_a; - int extra_read; +typedef struct { + int bpp, offset, hsz; + unsigned int mr, mg, mb, ma, all_a; + int extra_read; } stbi__bmp_data; -static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) -{ - int hsz; - if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') return stbi__errpuc("not BMP", "Corrupt BMP"); - stbi__get32le(s); // discard filesize - stbi__get16le(s); // discard reserved - stbi__get16le(s); // discard reserved - info->offset = stbi__get32le(s); - info->hsz = hsz = stbi__get32le(s); - info->mr = info->mg = info->mb = info->ma = 0; - info->extra_read = 14; +static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) { + int hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') + return stbi__errpuc("not BMP", "Corrupt BMP"); + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + info->extra_read = 14; - if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); + if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); - if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); - if (hsz == 12) { - s->img_x = stbi__get16le(s); - s->img_y = stbi__get16le(s); - } else { - s->img_x = stbi__get32le(s); - s->img_y = stbi__get32le(s); - } - if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); - info->bpp = stbi__get16le(s); - if (hsz != 12) { - int compress = stbi__get32le(s); - if (compress == 1 || compress == 2) return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); - stbi__get32le(s); // discard sizeof - stbi__get32le(s); // discard hres - stbi__get32le(s); // discard vres - stbi__get32le(s); // discard colorsused - stbi__get32le(s); // discard max important - if (hsz == 40 || hsz == 56) { - if (hsz == 56) { - stbi__get32le(s); - stbi__get32le(s); - stbi__get32le(s); - stbi__get32le(s); - } - if (info->bpp == 16 || info->bpp == 32) { - if (compress == 0) { - if (info->bpp == 32) { - info->mr = 0xffu << 16; - info->mg = 0xffu << 8; - info->mb = 0xffu << 0; - info->ma = 0xffu << 24; - info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 - } else { - info->mr = 31u << 10; - info->mg = 31u << 5; - info->mb = 31u << 0; - } - } else if (compress == 3) { - info->mr = stbi__get32le(s); - info->mg = stbi__get32le(s); - info->mb = stbi__get32le(s); - info->extra_read += 12; - // not documented, but generated by photoshop and handled by mspaint - if (info->mr == info->mg && info->mg == info->mb) { - // ?!?!? - return stbi__errpuc("bad BMP", "bad BMP"); - } - } else - return stbi__errpuc("bad BMP", "bad BMP"); - } - } else { - int i; - if (hsz != 108 && hsz != 124) + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) + return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = stbi__get16le(s); + s->img_y = stbi__get16le(s); + } else { + s->img_x = stbi__get32le(s); + s->img_y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int compress = stbi__get32le(s); + if (compress == 1 || compress == 2) + return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); + stbi__get32le(s); // discard sizeof + stbi__get32le(s); // discard hres + stbi__get32le(s); // discard vres + stbi__get32le(s); // discard colorsused + stbi__get32le(s); // discard max important + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + } + if (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha + // channel but it was all 0 + } else { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } + } else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->extra_read += 12; + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? return stbi__errpuc("bad BMP", "bad BMP"); - info->mr = stbi__get32le(s); - info->mg = stbi__get32le(s); - info->mb = stbi__get32le(s); - info->ma = stbi__get32le(s); - stbi__get32le(s); // discard color space - for (i=0; i < 12; ++i) - stbi__get32le(s); // discard color space parameters - if (hsz == 124) { - stbi__get32le(s); // discard rendering intent - stbi__get32le(s); // discard offset of profile data - stbi__get32le(s); // discard size of profile data - stbi__get32le(s); // discard reserved - } + } + } else + return stbi__errpuc("bad BMP", "bad BMP"); } - } - return (void *) 1; -} - - -static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *out; - unsigned int mr=0,mg=0,mb=0,ma=0, all_a; - stbi_uc pal[256][4]; - int psize=0,i,j,width; - int flip_vertically, pad, target; - stbi__bmp_data info; - STBI_NOTUSED(ri); - - info.all_a = 255; - if (stbi__bmp_parse_header(s, &info) == NULL) - return NULL; // error code already set - - flip_vertically = ((int) s->img_y) > 0; - s->img_y = abs((int) s->img_y); - - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - - mr = info.mr; - mg = info.mg; - mb = info.mb; - ma = info.ma; - all_a = info.all_a; - - if (info.hsz == 12) { - if (info.bpp < 24) - psize = (info.offset - info.extra_read - 24) / 3; - } else { - if (info.bpp < 16) - psize = (info.offset - info.extra_read - info.hsz) >> 2; - } - if (psize == 0) { - STBI_ASSERT(info.offset == s->callback_already_read + (int) (s->img_buffer - s->img_buffer_original)); - if (info.offset != s->callback_already_read + (s->img_buffer - s->buffer_start)) { - return stbi__errpuc("bad offset", "Corrupt BMP"); - } - } - - if (info.bpp == 24 && ma == 0xff000000) - s->img_n = 3; - else - s->img_n = ma ? 4 : 3; - if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 - target = req_comp; - else - target = s->img_n; // if they want monochrome, we'll post-convert - - // sanity-check size - if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) - return stbi__errpuc("too large", "Corrupt BMP"); - - out = (stbi_uc *) stbi__malloc_mad3(target, s->img_x, s->img_y, 0); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - if (info.bpp < 16) { - int z=0; - if (psize == 0 || psize > 256) { STBI_FREE(out); return stbi__errpuc("invalid", "Corrupt BMP"); } - for (i=0; i < psize; ++i) { - pal[i][2] = stbi__get8(s); - pal[i][1] = stbi__get8(s); - pal[i][0] = stbi__get8(s); - if (info.hsz != 12) stbi__get8(s); - pal[i][3] = 255; - } - stbi__skip(s, info.offset - info.extra_read - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); - if (info.bpp == 1) width = (s->img_x + 7) >> 3; - else if (info.bpp == 4) width = (s->img_x + 1) >> 1; - else if (info.bpp == 8) width = s->img_x; - else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } - pad = (-width)&3; - if (info.bpp == 1) { - for (j=0; j < (int) s->img_y; ++j) { - int bit_offset = 7, v = stbi__get8(s); - for (i=0; i < (int) s->img_x; ++i) { - int color = (v>>bit_offset)&0x1; - out[z++] = pal[color][0]; - out[z++] = pal[color][1]; - out[z++] = pal[color][2]; - if (target == 4) out[z++] = 255; - if (i+1 == (int) s->img_x) break; - if((--bit_offset) < 0) { - bit_offset = 7; - v = stbi__get8(s); - } - } - stbi__skip(s, pad); - } - } else { - for (j=0; j < (int) s->img_y; ++j) { - for (i=0; i < (int) s->img_x; i += 2) { - int v=stbi__get8(s),v2=0; - if (info.bpp == 4) { - v2 = v & 15; - v >>= 4; - } - out[z++] = pal[v][0]; - out[z++] = pal[v][1]; - out[z++] = pal[v][2]; - if (target == 4) out[z++] = 255; - if (i+1 == (int) s->img_x) break; - v = (info.bpp == 8) ? stbi__get8(s) : v2; - out[z++] = pal[v][0]; - out[z++] = pal[v][1]; - out[z++] = pal[v][2]; - if (target == 4) out[z++] = 255; - } - stbi__skip(s, pad); - } - } - } else { - int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0; - int z = 0; - int easy=0; - stbi__skip(s, info.offset - info.extra_read - info.hsz); - if (info.bpp == 24) width = 3 * s->img_x; - else if (info.bpp == 16) width = 2*s->img_x; - else /* bpp = 32 and pad = 0 */ width=0; - pad = (-width) & 3; - if (info.bpp == 24) { - easy = 1; - } else if (info.bpp == 32) { - if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) - easy = 2; - } - if (!easy) { - if (!mr || !mg || !mb) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } - // right shift amt to put high bit in position #7 - rshift = stbi__high_bit(mr)-7; rcount = stbi__bitcount(mr); - gshift = stbi__high_bit(mg)-7; gcount = stbi__bitcount(mg); - bshift = stbi__high_bit(mb)-7; bcount = stbi__bitcount(mb); - ashift = stbi__high_bit(ma)-7; acount = stbi__bitcount(ma); - if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } - } - for (j=0; j < (int) s->img_y; ++j) { - if (easy) { - for (i=0; i < (int) s->img_x; ++i) { - unsigned char a; - out[z+2] = stbi__get8(s); - out[z+1] = stbi__get8(s); - out[z+0] = stbi__get8(s); - z += 3; - a = (easy == 2 ? stbi__get8(s) : 255); - all_a |= a; - if (target == 4) out[z++] = a; - } - } else { - int bpp = info.bpp; - for (i=0; i < (int) s->img_x; ++i) { - stbi__uint32 v = (bpp == 16 ? (stbi__uint32) stbi__get16le(s) : stbi__get32le(s)); - unsigned int a; - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); - a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); - all_a |= a; - if (target == 4) out[z++] = STBI__BYTECAST(a); - } - } - stbi__skip(s, pad); - } - } - - // if alpha channel is all 0s, replace with all 255s - if (target == 4 && all_a == 0) - for (i=4*s->img_x*s->img_y-1; i >= 0; i -= 4) - out[i] = 255; - - if (flip_vertically) { - stbi_uc t; - for (j=0; j < (int) s->img_y>>1; ++j) { - stbi_uc *p1 = out + j *s->img_x*target; - stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; - for (i=0; i < (int) s->img_x*target; ++i) { - t = p1[i]; p1[i] = p2[i]; p2[i] = t; - } - } - } - - if (req_comp && req_comp != target) { - out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } - - *x = s->img_x; - *y = s->img_y; - if (comp) *comp = s->img_n; - return out; -} -#endif - -// Targa Truevision - TGA -// by Jonathan Dummer -#ifndef STBI_NO_TGA -// returns STBI_rgb or whatever, 0 on error -static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) -{ - // only RGB or RGBA (incl. 16bit) or grey allowed - if (is_rgb16) *is_rgb16 = 0; - switch(bits_per_pixel) { - case 8: return STBI_grey; - case 16: if(is_grey) return STBI_grey_alpha; - // fallthrough - case 15: if(is_rgb16) *is_rgb16 = 1; - return STBI_rgb; - case 24: // fallthrough - case 32: return bits_per_pixel/8; - default: return 0; - } -} - -static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) -{ - int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; - int sz, tga_colormap_type; - stbi__get8(s); // discard Offset - tga_colormap_type = stbi__get8(s); // colormap type - if( tga_colormap_type > 1 ) { - stbi__rewind(s); - return 0; // only RGB or indexed allowed - } - tga_image_type = stbi__get8(s); // image type - if ( tga_colormap_type == 1 ) { // colormapped (paletted) image - if (tga_image_type != 1 && tga_image_type != 9) { - stbi__rewind(s); - return 0; - } - stbi__skip(s,4); // skip index of first colormap entry and number of entries - sz = stbi__get8(s); // check bits per palette color entry - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) { - stbi__rewind(s); - return 0; - } - stbi__skip(s,4); // skip image x and y origin - tga_colormap_bpp = sz; - } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE - if ( (tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11) ) { - stbi__rewind(s); - return 0; // only RGB or grey allowed, +/- RLE - } - stbi__skip(s,9); // skip colormap specification and image x/y origin - tga_colormap_bpp = 0; - } - tga_w = stbi__get16le(s); - if( tga_w < 1 ) { - stbi__rewind(s); - return 0; // test width - } - tga_h = stbi__get16le(s); - if( tga_h < 1 ) { - stbi__rewind(s); - return 0; // test height - } - tga_bits_per_pixel = stbi__get8(s); // bits per pixel - stbi__get8(s); // ignore alpha bits - if (tga_colormap_bpp != 0) { - if((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { - // when using a colormap, tga_bits_per_pixel is the size of the indexes - // I don't think anything but 8 or 16bit indexes makes sense - stbi__rewind(s); - return 0; - } - tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); } else { - tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + int i; + if (hsz != 108 && hsz != 124) return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->ma = stbi__get32le(s); + stbi__get32le(s); // discard color space + for (i = 0; i < 12; ++i) + stbi__get32le(s); // discard color space parameters + if (hsz == 124) { + stbi__get32le(s); // discard rendering intent + stbi__get32le(s); // discard offset of profile data + stbi__get32le(s); // discard size of profile data + stbi__get32le(s); // discard reserved + } } - if(!tga_comp) { + } + return (void *)1; +} + +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *out; + unsigned int mr = 0, mg = 0, mb = 0, ma = 0, all_a; + stbi_uc pal[256][4]; + int psize = 0, i, j, width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int)s->img_y) > 0; + s->img_y = abs((int)s->img_y); + + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) psize = (info.offset - info.extra_read - 24) / 3; + } else { + if (info.bpp < 16) psize = (info.offset - info.extra_read - info.hsz) >> 2; + } + if (psize == 0) { + STBI_ASSERT(info.offset == + s->callback_already_read + + (int)(s->img_buffer - s->img_buffer_original)); + if (info.offset != + s->callback_already_read + (s->img_buffer - s->buffer_start)) { + return stbi__errpuc("bad offset", "Corrupt BMP"); + } + } + + if (info.bpp == 24 && ma == 0xff000000) + s->img_n = 3; + else + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc *)stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z = 0; + if (psize == 0 || psize > 256) { + STBI_FREE(out); + return stbi__errpuc("invalid", "Corrupt BMP"); + } + for (i = 0; i < psize; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + if (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - info.extra_read - info.hsz - + psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) + width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) + width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) + width = s->img_x; + else { + STBI_FREE(out); + return stbi__errpuc("bad bpp", "Corrupt BMP"); + } + pad = (-width) & 3; + if (info.bpp == 1) { + for (j = 0; j < (int)s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i = 0; i < (int)s->img_x; ++i) { + int color = (v >> bit_offset) & 0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if (target == 4) out[z++] = 255; + if (i + 1 == (int)s->img_x) break; + if ((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } else { + for (j = 0; j < (int)s->img_y; ++j) { + for (i = 0; i < (int)s->img_x; i += 2) { + int v = stbi__get8(s), v2 = 0; + if (info.bpp == 4) { + v2 = v & 15; + v >>= 4; + } + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + if (i + 1 == (int)s->img_x) break; + v = (info.bpp == 8) ? stbi__get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + stbi__skip(s, pad); + } + } + } else { + int rshift = 0, gshift = 0, bshift = 0, ashift = 0, rcount = 0, gcount = 0, + bcount = 0, acount = 0; + int z = 0; + int easy = 0; + stbi__skip(s, info.offset - info.extra_read - info.hsz); + if (info.bpp == 24) + width = 3 * s->img_x; + else if (info.bpp == 16) + width = 2 * s->img_x; + else /* bpp = 32 and pad = 0 */ + width = 0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { + STBI_FREE(out); + return stbi__errpuc("bad masks", "Corrupt BMP"); + } + // right shift amt to put high bit in position #7 + rshift = stbi__high_bit(mr) - 7; + rcount = stbi__bitcount(mr); + gshift = stbi__high_bit(mg) - 7; + gcount = stbi__bitcount(mg); + bshift = stbi__high_bit(mb) - 7; + bcount = stbi__bitcount(mb); + ashift = stbi__high_bit(ma) - 7; + acount = stbi__bitcount(ma); + if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { + STBI_FREE(out); + return stbi__errpuc("bad masks", "Corrupt BMP"); + } + } + for (j = 0; j < (int)s->img_y; ++j) { + if (easy) { + for (i = 0; i < (int)s->img_x; ++i) { + unsigned char a; + out[z + 2] = stbi__get8(s); + out[z + 1] = stbi__get8(s); + out[z + 0] = stbi__get8(s); + z += 3; + a = (easy == 2 ? stbi__get8(s) : 255); + all_a |= a; + if (target == 4) out[z++] = a; + } + } else { + int bpp = info.bpp; + for (i = 0; i < (int)s->img_x; ++i) { + stbi__uint32 v = + (bpp == 16 ? (stbi__uint32)stbi__get16le(s) : stbi__get32le(s)); + unsigned int a; + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); + a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i = 4 * s->img_x * s->img_y - 1; i >= 0; i -= 4) + out[i] = 255; + + if (flip_vertically) { + stbi_uc t; + for (j = 0; j < (int)s->img_y >> 1; ++j) { + stbi_uc *p1 = out + j * s->img_x * target; + stbi_uc *p2 = out + (s->img_y - 1 - j) * s->img_x * target; + for (i = 0; i < (int)s->img_x * target; ++i) { + t = p1[i]; + p1[i] = p2[i]; + p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} + #endif + + // Targa Truevision - TGA + // by Jonathan Dummer + #ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int *is_rgb16) { + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16) *is_rgb16 = 0; + switch (bits_per_pixel) { + case 8: + return STBI_grey; + case 16: + if (is_grey) return STBI_grey_alpha; + // fallthrough + case 15: + if (is_rgb16) *is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: + return bits_per_pixel / 8; + default: + return 0; + } +} + +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) { + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, + tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if (tga_colormap_type > 1) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if (tga_colormap_type == 1) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { stbi__rewind(s); return 0; } - if (x) *x = tga_w; - if (y) *y = tga_h; - if (comp) *comp = tga_comp; - return 1; // seems to have passed everything + stbi__skip(s, + 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 4); // skip image x and y origin + tga_colormap_bpp = sz; + } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ((tga_image_type != 2) && (tga_image_type != 3) && + (tga_image_type != 10) && (tga_image_type != 11)) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s, 9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + tga_w = stbi__get16le(s); + if (tga_w < 1) { + stbi__rewind(s); + return 0; // test width + } + tga_h = stbi__get16le(s); + if (tga_h < 1) { + stbi__rewind(s); + return 0; // test height + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } else { + tga_comp = stbi__tga_get_comp( + tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), + NULL); + } + if (!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp; + return 1; // seems to have passed everything } -static int stbi__tga_test(stbi__context *s) -{ - int res = 0; - int sz, tga_color_type; - stbi__get8(s); // discard Offset - tga_color_type = stbi__get8(s); // color type - if ( tga_color_type > 1 ) goto errorEnd; // only RGB or indexed allowed - sz = stbi__get8(s); // image type - if ( tga_color_type == 1 ) { // colormapped (paletted) image - if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 - stbi__skip(s,4); // skip index of first colormap entry and number of entries - sz = stbi__get8(s); // check bits per palette color entry - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; - stbi__skip(s,4); // skip image x and y origin - } else { // "normal" image w/o colormap - if ( (sz != 2) && (sz != 3) && (sz != 10) && (sz != 11) ) goto errorEnd; // only RGB or grey allowed, +/- RLE - stbi__skip(s,9); // skip colormap specification and image x/y origin - } - if ( stbi__get16le(s) < 1 ) goto errorEnd; // test width - if ( stbi__get16le(s) < 1 ) goto errorEnd; // test height - sz = stbi__get8(s); // bits per pixel - if ( (tga_color_type == 1) && (sz != 8) && (sz != 16) ) goto errorEnd; // for colormapped images, bpp is size of an index - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; +static int stbi__tga_test(stbi__context *s) { + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if (tga_color_type > 1) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if (tga_color_type == 1) { // colormapped (paletted) image + if (sz != 1 && sz != 9) + goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s, + 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) + goto errorEnd; + stbi__skip(s, 4); // skip image x and y origin + } else { // "normal" image w/o colormap + if ((sz != 2) && (sz != 3) && (sz != 10) && (sz != 11)) + goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s, 9); // skip colormap specification and image x/y origin + } + if (stbi__get16le(s) < 1) goto errorEnd; // test width + if (stbi__get16le(s) < 1) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ((tga_color_type == 1) && (sz != 8) && (sz != 16)) + goto errorEnd; // for colormapped images, bpp is size of an index + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) + goto errorEnd; - res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + res = 1; // if we got this far, everything's good and we can return 1 instead + // of 0 errorEnd: - stbi__rewind(s); - return res; + stbi__rewind(s); + return res; } // read 16bit value and convert to 24bit RGB -static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc* out) -{ - stbi__uint16 px = (stbi__uint16)stbi__get16le(s); - stbi__uint16 fiveBitMask = 31; - // we have 3 channels with 5bits each - int r = (px >> 10) & fiveBitMask; - int g = (px >> 5) & fiveBitMask; - int b = px & fiveBitMask; - // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later - out[0] = (stbi_uc)((r * 255)/31); - out[1] = (stbi_uc)((g * 255)/31); - out[2] = (stbi_uc)((b * 255)/31); +static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc *out) { + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be + // swapped later + out[0] = (stbi_uc)((r * 255) / 31); + out[1] = (stbi_uc)((g * 255) / 31); + out[2] = (stbi_uc)((b * 255) / 31); - // some people claim that the most significant bit might be used for alpha - // (possibly if an alpha-bit is set in the "image descriptor byte") - // but that only made 16bit test images completely translucent.. - // so let's treat all 15 and 16bit TGAs as RGB with no alpha. + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. } -static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - // read in the TGA header stuff - int tga_offset = stbi__get8(s); - int tga_indexed = stbi__get8(s); - int tga_image_type = stbi__get8(s); - int tga_is_RLE = 0; - int tga_palette_start = stbi__get16le(s); - int tga_palette_len = stbi__get16le(s); - int tga_palette_bits = stbi__get8(s); - int tga_x_origin = stbi__get16le(s); - int tga_y_origin = stbi__get16le(s); - int tga_width = stbi__get16le(s); - int tga_height = stbi__get16le(s); - int tga_bits_per_pixel = stbi__get8(s); - int tga_comp, tga_rgb16=0; - int tga_inverted = stbi__get8(s); - // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) - // image data - unsigned char *tga_data; - unsigned char *tga_palette = NULL; - int i, j; - unsigned char raw_data[4] = {0}; - int RLE_count = 0; - int RLE_repeating = 0; - int read_next_pixel = 1; - STBI_NOTUSED(ri); - STBI_NOTUSED(tga_x_origin); // @TODO - STBI_NOTUSED(tga_y_origin); // @TODO +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + // read in the TGA header stuff + int tga_offset = stbi__get8(s); + int tga_indexed = stbi__get8(s); + int tga_image_type = stbi__get8(s); + int tga_is_RLE = 0; + int tga_palette_start = stbi__get16le(s); + int tga_palette_len = stbi__get16le(s); + int tga_palette_bits = stbi__get8(s); + int tga_x_origin = stbi__get16le(s); + int tga_y_origin = stbi__get16le(s); + int tga_width = stbi__get16le(s); + int tga_height = stbi__get16le(s); + int tga_bits_per_pixel = stbi__get8(s); + int tga_comp, tga_rgb16 = 0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused + // (useless?) + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = {0}; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + STBI_NOTUSED(tga_x_origin); // @TODO + STBI_NOTUSED(tga_y_origin); // @TODO - if (tga_height > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (tga_width > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (tga_height > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (tga_width > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - // do a tiny bit of precessing - if ( tga_image_type >= 8 ) - { - tga_image_type -= 8; - tga_is_RLE = 1; - } - tga_inverted = 1 - ((tga_inverted >> 5) & 1); + // do a tiny bit of precessing + if (tga_image_type >= 8) { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); - // If I'm paletted, then I'll use the number of bits from the palette - if ( tga_indexed ) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); - else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + // If I'm paletted, then I'll use the number of bits from the palette + if (tga_indexed) + tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), + &tga_rgb16); - if(!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency - return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + if (!tga_comp) // shouldn't really happen, stbi__tga_test() should have + // ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); - // tga info - *x = tga_width; - *y = tga_height; - if (comp) *comp = tga_comp; + // tga info + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; - if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) - return stbi__errpuc("too large", "Corrupt TGA"); + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); - tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); - if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + tga_data = + (unsigned char *)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); - // skip to the data's starting position (offset usually = 0) - stbi__skip(s, tga_offset ); + // skip to the data's starting position (offset usually = 0) + stbi__skip(s, tga_offset); - if ( !tga_indexed && !tga_is_RLE && !tga_rgb16 ) { - for (i=0; i < tga_height; ++i) { - int row = tga_inverted ? tga_height -i - 1 : i; - stbi_uc *tga_row = tga_data + row*tga_width*tga_comp; - stbi__getn(s, tga_row, tga_width * tga_comp); + if (!tga_indexed && !tga_is_RLE && !tga_rgb16) { + for (i = 0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height - i - 1 : i; + stbi_uc *tga_row = tga_data + row * tga_width * tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } else { + // do I need to load a palette? + if (tga_indexed) { + if (tga_palette_len == 0) { /* you have to have at least one entry! */ + STBI_FREE(tga_data); + return stbi__errpuc("bad palette", "Corrupt TGA"); } - } else { - // do I need to load a palette? - if ( tga_indexed) - { - if (tga_palette_len == 0) { /* you have to have at least one entry! */ - STBI_FREE(tga_data); - return stbi__errpuc("bad palette", "Corrupt TGA"); - } - // any data to skip? (offset usually = 0) - stbi__skip(s, tga_palette_start ); - // load the palette - tga_palette = (unsigned char*)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); - if (!tga_palette) { - STBI_FREE(tga_data); - return stbi__errpuc("outofmem", "Out of memory"); - } - if (tga_rgb16) { - stbi_uc *pal_entry = tga_palette; - STBI_ASSERT(tga_comp == STBI_rgb); - for (i=0; i < tga_palette_len; ++i) { - stbi__tga_read_rgb16(s, pal_entry); - pal_entry += tga_comp; - } - } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { - STBI_FREE(tga_data); - STBI_FREE(tga_palette); - return stbi__errpuc("bad palette", "Corrupt TGA"); - } + // any data to skip? (offset usually = 0) + stbi__skip(s, tga_palette_start); + // load the palette + tga_palette = + (unsigned char *)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); } - // load the data - for (i=0; i < tga_width * tga_height; ++i) - { - // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? - if ( tga_is_RLE ) - { - if ( RLE_count == 0 ) - { - // yep, get the next byte as a RLE command - int RLE_cmd = stbi__get8(s); - RLE_count = 1 + (RLE_cmd & 127); - RLE_repeating = RLE_cmd >> 7; - read_next_pixel = 1; - } else if ( !RLE_repeating ) - { - read_next_pixel = 1; - } - } else - { - read_next_pixel = 1; - } - // OK, if I need to read a pixel, do it now - if ( read_next_pixel ) - { - // load however much data we did have - if ( tga_indexed ) - { - // read in index, then perform the lookup - int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); - if ( pal_idx >= tga_palette_len ) { - // invalid index - pal_idx = 0; - } - pal_idx *= tga_comp; - for (j = 0; j < tga_comp; ++j) { - raw_data[j] = tga_palette[pal_idx+j]; - } - } else if(tga_rgb16) { - STBI_ASSERT(tga_comp == STBI_rgb); - stbi__tga_read_rgb16(s, raw_data); - } else { - // read in the data raw - for (j = 0; j < tga_comp; ++j) { - raw_data[j] = stbi__get8(s); - } - } - // clear the reading flag for the next pixel - read_next_pixel = 0; - } // end of reading a pixel - - // copy data - for (j = 0; j < tga_comp; ++j) - tga_data[i*tga_comp+j] = raw_data[j]; - - // in case we're in RLE mode, keep counting down - --RLE_count; + if (tga_rgb16) { + stbi_uc *pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i = 0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_FREE(tga_palette); + return stbi__errpuc("bad palette", "Corrupt TGA"); } - // do I need to invert the image? - if ( tga_inverted ) - { - for (j = 0; j*2 < tga_height; ++j) - { - int index1 = j * tga_width * tga_comp; - int index2 = (tga_height - 1 - j) * tga_width * tga_comp; - for (i = tga_width * tga_comp; i > 0; --i) - { - unsigned char temp = tga_data[index1]; - tga_data[index1] = tga_data[index2]; - tga_data[index2] = temp; - ++index1; - ++index2; - } - } + } + // load the data + for (i = 0; i < tga_width * tga_height; ++i) { + // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? + if (tga_is_RLE) { + if (RLE_count == 0) { + // yep, get the next byte as a RLE command + int RLE_cmd = stbi__get8(s); + RLE_count = 1 + (RLE_cmd & 127); + RLE_repeating = RLE_cmd >> 7; + read_next_pixel = 1; + } else if (!RLE_repeating) { + read_next_pixel = 1; + } + } else { + read_next_pixel = 1; } - // clear my palette, if I had one - if ( tga_palette != NULL ) - { - STBI_FREE( tga_palette ); + // OK, if I need to read a pixel, do it now + if (read_next_pixel) { + // load however much data we did have + if (tga_indexed) { + // read in index, then perform the lookup + int pal_idx = + (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if (pal_idx >= tga_palette_len) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx + j]; + } + } else if (tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } else { + // read in the data raw + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = stbi__get8(s); + } + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + + // copy data + for (j = 0; j < tga_comp; ++j) + tga_data[i * tga_comp + j] = raw_data[j]; + + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if (tga_inverted) { + for (j = 0; j * 2 < tga_height; ++j) { + int index1 = j * tga_width * tga_comp; + int index2 = (tga_height - 1 - j) * tga_width * tga_comp; + for (i = tga_width * tga_comp; i > 0; --i) { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } } - } + } + // clear my palette, if I had one + if (tga_palette != NULL) { STBI_FREE(tga_palette); } + } - // swap RGB - if the source data was RGB16, it already is in the right order - if (tga_comp >= 3 && !tga_rgb16) - { - unsigned char* tga_pixel = tga_data; - for (i=0; i < tga_width * tga_height; ++i) - { - unsigned char temp = tga_pixel[0]; - tga_pixel[0] = tga_pixel[2]; - tga_pixel[2] = temp; - tga_pixel += tga_comp; - } - } + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) { + unsigned char *tga_pixel = tga_data; + for (i = 0; i < tga_width * tga_height; ++i) { + unsigned char temp = tga_pixel[0]; + tga_pixel[0] = tga_pixel[2]; + tga_pixel[2] = temp; + tga_pixel += tga_comp; + } + } - // convert to target component count - if (req_comp && req_comp != tga_comp) - tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + // convert to target component count + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, + tga_height); - // the things I do to get rid of an error message, and yet keep - // Microsoft's C compilers happy... [8^( - tga_palette_start = tga_palette_len = tga_palette_bits = - tga_x_origin = tga_y_origin = 0; - STBI_NOTUSED(tga_palette_start); - // OK, done - return tga_data; + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = tga_x_origin = + tga_y_origin = 0; + STBI_NOTUSED(tga_palette_start); + // OK, done + return tga_data; } -#endif + #endif // ************************************************************************************************* -// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, +// tweaked by STB -#ifndef STBI_NO_PSD -static int stbi__psd_test(stbi__context *s) -{ - int r = (stbi__get32be(s) == 0x38425053); - stbi__rewind(s); - return r; + #ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s) { + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; } -static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) -{ - int count, nleft, len; +static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) { + int count, nleft, len; - count = 0; - while ((nleft = pixelCount - count) > 0) { - len = stbi__get8(s); - if (len == 128) { - // No-op. - } else if (len < 128) { - // Copy next len+1 bytes literally. - len++; - if (len > nleft) return 0; // corrupt data - count += len; - while (len) { - *p = stbi__get8(s); - p += 4; - len--; - } - } else if (len > 128) { - stbi_uc val; - // Next -len+1 bytes in the dest are replicated from next source byte. - // (Interpret len as a negative 8-bit int.) - len = 257 - len; - if (len > nleft) return 0; // corrupt data - val = stbi__get8(s); - count += len; - while (len) { - *p = val; - p += 4; - len--; - } + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + count += len; + while (len) { + *p = stbi__get8(s); + p += 4; + len--; } - } + } else if (len > 128) { + stbi_uc val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } - return 1; + return 1; } -static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) -{ - int pixelCount; - int channelCount, compression; - int channel, i; - int bitdepth; - int w,h; - stbi_uc *out; - STBI_NOTUSED(ri); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc) { + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w, h; + stbi_uc *out; + STBI_NOTUSED(ri); - // Check identifier - if (stbi__get32be(s) != 0x38425053) // "8BPS" - return stbi__errpuc("not PSD", "Corrupt PSD image"); + // Check identifier + if (stbi__get32be(s) != 0x38425053) // "8BPS" + return stbi__errpuc("not PSD", "Corrupt PSD image"); - // Check file type version. - if (stbi__get16be(s) != 1) - return stbi__errpuc("wrong version", "Unsupported version of PSD image"); + // Check file type version. + if (stbi__get16be(s) != 1) + return stbi__errpuc("wrong version", "Unsupported version of PSD image"); - // Skip 6 reserved bytes. - stbi__skip(s, 6 ); + // Skip 6 reserved bytes. + stbi__skip(s, 6); - // Read the number of channels (R, G, B, A, etc). - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) - return stbi__errpuc("wrong channel count", "Unsupported number of channels in PSD image"); + // Read the number of channels (R, G, B, A, etc). + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) + return stbi__errpuc("wrong channel count", + "Unsupported number of channels in PSD image"); - // Read the rows and columns of the image. - h = stbi__get32be(s); - w = stbi__get32be(s); + // Read the rows and columns of the image. + h = stbi__get32be(s); + w = stbi__get32be(s); - if (h > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (w > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (h > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (w > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - // Make sure the depth is 8 bits. - bitdepth = stbi__get16be(s); - if (bitdepth != 8 && bitdepth != 16) - return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 bit"); + // Make sure the depth is 8 bits. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", + "PSD bit depth is not 8 or 16 bit"); - // Make sure the color mode is RGB. - // Valid options are: - // 0: Bitmap - // 1: Grayscale - // 2: Indexed color - // 3: RGB color - // 4: CMYK color - // 7: Multichannel - // 8: Duotone - // 9: Lab color - if (stbi__get16be(s) != 3) - return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (stbi__get16be(s) != 3) + return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); - // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) - stbi__skip(s,stbi__get32be(s) ); + // Skip the Mode Data. (It's the palette for indexed color; other info for + // other modes.) + stbi__skip(s, stbi__get32be(s)); - // Skip the image resources. (resolution, pen tool paths, etc) - stbi__skip(s, stbi__get32be(s) ); + // Skip the image resources. (resolution, pen tool paths, etc) + stbi__skip(s, stbi__get32be(s)); - // Skip the reserved data. - stbi__skip(s, stbi__get32be(s) ); + // Skip the reserved data. + stbi__skip(s, stbi__get32be(s)); - // Find out if the data is compressed. - // Known values: - // 0: no compression - // 1: RLE compressed - compression = stbi__get16be(s); - if (compression > 1) - return stbi__errpuc("bad compression", "PSD has an unknown compression format"); + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = stbi__get16be(s); + if (compression > 1) + return stbi__errpuc("bad compression", + "PSD has an unknown compression format"); - // Check size - if (!stbi__mad3sizes_valid(4, w, h, 0)) - return stbi__errpuc("too large", "Corrupt PSD"); + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); - // Create the destination image. + // Create the destination image. - if (!compression && bitdepth == 16 && bpc == 16) { - out = (stbi_uc *) stbi__malloc_mad3(8, w, h, 0); - ri->bits_per_channel = 16; - } else - out = (stbi_uc *) stbi__malloc(4 * w*h); + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc *)stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } else + out = (stbi_uc *)stbi__malloc(4 * w * h); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - pixelCount = w*h; + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + pixelCount = w * h; - // Initialize the data to zero. - //memset( out, 0, pixelCount * 4 ); + // Initialize the data to zero. + // memset( out, 0, pixelCount * 4 ); - // Finally, the image data. - if (compression) { - // RLE as used by .PSD and .TIFF - // Loop until you get the number of unpacked bytes you are expecting: - // Read the next source byte into n. - // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. - // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. - // Else if n is 128, noop. - // Endloop + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes + // literally. Else if n is between -127 and -1 inclusive, copy the next + // byte -n+1 times. Else if n is 128, noop. + // Endloop - // The RLE-compressed data is preceded by a 2-byte data count for each row in the data, - // which we're going to just skip. - stbi__skip(s, h * channelCount * 2 ); + // The RLE-compressed data is preceded by a 2-byte data count for each row + // in the data, which we're going to just skip. + stbi__skip(s, h * channelCount * 2); - // Read the RLE data by channel. - for (channel = 0; channel < 4; channel++) { - stbi_uc *p; + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; - p = out+channel; - if (channel >= channelCount) { - // Fill this channel with default data. - for (i = 0; i < pixelCount; i++, p += 4) - *p = (channel == 3 ? 255 : 0); - } else { - // Read the RLE data. - if (!stbi__psd_decode_rle(s, p, pixelCount)) { - STBI_FREE(out); - return stbi__errpuc("corrupt", "bad RLE data"); - } - } - } - - } else { - // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) - // where each channel consists of an 8-bit (or 16-bit) value for each pixel in the image. - - // Read the data by channel. - for (channel = 0; channel < 4; channel++) { - if (channel >= channelCount) { - // Fill this channel with default data. - if (bitdepth == 16 && bpc == 16) { - stbi__uint16 *q = ((stbi__uint16 *) out) + channel; - stbi__uint16 val = channel == 3 ? 65535 : 0; - for (i = 0; i < pixelCount; i++, q += 4) - *q = val; - } else { - stbi_uc *p = out+channel; - stbi_uc val = channel == 3 ? 255 : 0; - for (i = 0; i < pixelCount; i++, p += 4) - *p = val; - } - } else { - if (ri->bits_per_channel == 16) { // output bpc - stbi__uint16 *q = ((stbi__uint16 *) out) + channel; - for (i = 0; i < pixelCount; i++, q += 4) - *q = (stbi__uint16) stbi__get16be(s); - } else { - stbi_uc *p = out+channel; - if (bitdepth == 16) { // input bpc - for (i = 0; i < pixelCount; i++, p += 4) - *p = (stbi_uc) (stbi__get16be(s) >> 8); - } else { - for (i = 0; i < pixelCount; i++, p += 4) - *p = stbi__get8(s); - } - } - } - } - } - - // remove weird white matte from PSD - if (channelCount >= 4) { - if (ri->bits_per_channel == 16) { - for (i=0; i < w*h; ++i) { - stbi__uint16 *pixel = (stbi__uint16 *) out + 4*i; - if (pixel[3] != 0 && pixel[3] != 65535) { - float a = pixel[3] / 65535.0f; - float ra = 1.0f / a; - float inv_a = 65535.0f * (1 - ra); - pixel[0] = (stbi__uint16) (pixel[0]*ra + inv_a); - pixel[1] = (stbi__uint16) (pixel[1]*ra + inv_a); - pixel[2] = (stbi__uint16) (pixel[2]*ra + inv_a); - } - } + p = out + channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++, p += 4) + *p = (channel == 3 ? 255 : 0); } else { - for (i=0; i < w*h; ++i) { - unsigned char *pixel = out + 4*i; - if (pixel[3] != 0 && pixel[3] != 255) { - float a = pixel[3] / 255.0f; - float ra = 1.0f / a; - float inv_a = 255.0f * (1 - ra); - pixel[0] = (unsigned char) (pixel[0]*ra + inv_a); - pixel[1] = (unsigned char) (pixel[1]*ra + inv_a); - pixel[2] = (unsigned char) (pixel[2]*ra + inv_a); - } - } + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } } - } + } - // convert to desired output format - if (req_comp && req_comp != 4) { - if (ri->bits_per_channel == 16) - out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, 4, req_comp, w, h); - else - out = stbi__convert_format(out, 4, req_comp, w, h); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } + } else { + // We're at the raw image data. It's each channel in order (Red, Green, + // Blue, Alpha, ...) where each channel consists of an 8-bit (or 16-bit) + // value for each pixel in the image. - if (comp) *comp = 4; - *y = h; - *x = w; + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16 *q = ((stbi__uint16 *)out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + *q = val; + } else { + stbi_uc *p = out + channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + *p = val; + } + } else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16 *q = ((stbi__uint16 *)out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + *q = (stbi__uint16)stbi__get16be(s); + } else { + stbi_uc *p = out + channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + *p = (stbi_uc)(stbi__get16be(s) >> 8); + } else { + for (i = 0; i < pixelCount; i++, p += 4) + *p = stbi__get8(s); + } + } + } + } + } - return out; + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i = 0; i < w * h; ++i) { + stbi__uint16 *pixel = (stbi__uint16 *)out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16)(pixel[0] * ra + inv_a); + pixel[1] = (stbi__uint16)(pixel[1] * ra + inv_a); + pixel[2] = (stbi__uint16)(pixel[2] * ra + inv_a); + } + } + } else { + for (i = 0; i < w * h; ++i) { + unsigned char *pixel = out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char)(pixel[0] * ra + inv_a); + pixel[1] = (unsigned char)(pixel[1] * ra + inv_a); + pixel[2] = (unsigned char)(pixel[2] * ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc *)stbi__convert_format16((stbi__uint16 *)out, 4, req_comp, + w, h); + else + out = stbi__convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + if (comp) *comp = 4; + *y = h; + *x = w; + + return out; } -#endif + #endif // ************************************************************************************************* // Softimage PIC loader @@ -6151,1130 +6679,1139 @@ static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req // See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format // See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ -#ifndef STBI_NO_PIC -static int stbi__pic_is4(stbi__context *s,const char *str) -{ - int i; - for (i=0; i<4; ++i) - if (stbi__get8(s) != (stbi_uc)str[i]) - return 0; + #ifndef STBI_NO_PIC +static int stbi__pic_is4(stbi__context *s, const char *str) { + int i; + for (i = 0; i < 4; ++i) + if (stbi__get8(s) != (stbi_uc)str[i]) return 0; - return 1; + return 1; } -static int stbi__pic_test_core(stbi__context *s) -{ - int i; +static int stbi__pic_test_core(stbi__context *s) { + int i; - if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) - return 0; + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) return 0; - for(i=0;i<84;++i) - stbi__get8(s); + for (i = 0; i < 84; ++i) + stbi__get8(s); - if (!stbi__pic_is4(s,"PICT")) - return 0; + if (!stbi__pic_is4(s, "PICT")) return 0; - return 1; + return 1; } -typedef struct -{ - stbi_uc size,type,channel; +typedef struct { + stbi_uc size, type, channel; } stbi__pic_packet; -static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) -{ - int mask=0x80, i; +static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) { + int mask = 0x80, i; - for (i=0; i<4; ++i, mask>>=1) { - if (channel & mask) { - if (stbi__at_eof(s)) return stbi__errpuc("bad file","PIC file too short"); - dest[i]=stbi__get8(s); - } - } + for (i = 0; i < 4; ++i, mask >>= 1) { + if (channel & mask) { + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "PIC file too short"); + dest[i] = stbi__get8(s); + } + } - return dest; + return dest; } -static void stbi__copyval(int channel,stbi_uc *dest,const stbi_uc *src) -{ - int mask=0x80,i; +static void stbi__copyval(int channel, stbi_uc *dest, const stbi_uc *src) { + int mask = 0x80, i; - for (i=0;i<4; ++i, mask>>=1) - if (channel&mask) - dest[i]=src[i]; + for (i = 0; i < 4; ++i, mask >>= 1) + if (channel & mask) dest[i] = src[i]; } -static stbi_uc *stbi__pic_load_core(stbi__context *s,int width,int height,int *comp, stbi_uc *result) -{ - int act_comp=0,num_packets=0,y,chained; - stbi__pic_packet packets[10]; +static stbi_uc *stbi__pic_load_core(stbi__context *s, int width, int height, + int *comp, stbi_uc *result) { + int act_comp = 0, num_packets = 0, y, chained; + stbi__pic_packet packets[10]; - // this will (should...) cater for even some bizarre stuff like having data - // for the same channel in multiple packets. - do { - stbi__pic_packet *packet; + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + stbi__pic_packet *packet; - if (num_packets==sizeof(packets)/sizeof(packets[0])) - return stbi__errpuc("bad format","too many packets"); + if (num_packets == sizeof(packets) / sizeof(packets[0])) + return stbi__errpuc("bad format", "too many packets"); - packet = &packets[num_packets++]; + packet = &packets[num_packets++]; - chained = stbi__get8(s); - packet->size = stbi__get8(s); - packet->type = stbi__get8(s); - packet->channel = stbi__get8(s); + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); - act_comp |= packet->channel; + act_comp |= packet->channel; - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (reading packets)"); - if (packet->size != 8) return stbi__errpuc("bad format","packet isn't 8bpp"); - } while (chained); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "file too short (reading packets)"); + if (packet->size != 8) + return stbi__errpuc("bad format", "packet isn't 8bpp"); + } while (chained); - *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? - for(y=0; ytype) { - default: - return stbi__errpuc("bad format","packet has bad compression type"); + switch (packet->type) { + default: + return stbi__errpuc("bad format", "packet has bad compression type"); - case 0: {//uncompressed - int x; + case 0: { // uncompressed + int x; - for(x=0;xchannel,dest)) - return 0; - break; + for (x = 0; x < width; ++x, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) return 0; + break; + } + + case 1: // Pure RLE + { + int left = width, i; + + while (left > 0) { + stbi_uc count, value[4]; + + count = stbi__get8(s); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", + "file too short (pure read count)"); + + if (count > left) count = (stbi_uc)left; + + if (!stbi__readval(s, packet->channel, value)) return 0; + + for (i = 0; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + left -= count; + } + } break; + + case 2: { // Mixed RLE + int left = width; + while (left > 0) { + int count = stbi__get8(s), i; + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", + "file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + + if (count == 128) + count = stbi__get16be(s); + else + count -= 127; + if (count > left) + return stbi__errpuc("bad file", "scanline overrun"); + + if (!stbi__readval(s, packet->channel, value)) return 0; + + for (i = 0; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + } else { // Raw + ++count; + if (count > left) + return stbi__errpuc("bad file", "scanline overrun"); + + for (i = 0; i < count; ++i, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) return 0; } - - case 1://Pure RLE - { - int left=width, i; - - while (left>0) { - stbi_uc count,value[4]; - - count=stbi__get8(s); - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pure read count)"); - - if (count > left) - count = (stbi_uc) left; - - if (!stbi__readval(s,packet->channel,value)) return 0; - - for(i=0; ichannel,dest,value); - left -= count; - } - } - break; - - case 2: {//Mixed RLE - int left=width; - while (left>0) { - int count = stbi__get8(s), i; - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (mixed read count)"); - - if (count >= 128) { // Repeated - stbi_uc value[4]; - - if (count==128) - count = stbi__get16be(s); - else - count -= 127; - if (count > left) - return stbi__errpuc("bad file","scanline overrun"); - - if (!stbi__readval(s,packet->channel,value)) - return 0; - - for(i=0;ichannel,dest,value); - } else { // Raw - ++count; - if (count>left) return stbi__errpuc("bad file","scanline overrun"); - - for(i=0;ichannel,dest)) - return 0; - } - left-=count; - } - break; - } - } + left -= count; + } + break; + } } - } + } + } - return result; + return result; } -static void *stbi__pic_load(stbi__context *s,int *px,int *py,int *comp,int req_comp, stbi__result_info *ri) -{ - stbi_uc *result; - int i, x,y, internal_comp; - STBI_NOTUSED(ri); +static void *stbi__pic_load(stbi__context *s, int *px, int *py, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *result; + int i, x, y, internal_comp; + STBI_NOTUSED(ri); - if (!comp) comp = &internal_comp; + if (!comp) comp = &internal_comp; - for (i=0; i<92; ++i) - stbi__get8(s); + for (i = 0; i < 92; ++i) + stbi__get8(s); - x = stbi__get16be(s); - y = stbi__get16be(s); + x = stbi__get16be(s); + y = stbi__get16be(s); - if (y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pic header)"); - if (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC image too large to decode"); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "file too short (pic header)"); + if (!stbi__mad3sizes_valid(x, y, 4, 0)) + return stbi__errpuc("too large", "PIC image too large to decode"); - stbi__get32be(s); //skip `ratio' - stbi__get16be(s); //skip `fields' - stbi__get16be(s); //skip `pad' + stbi__get32be(s); // skip `ratio' + stbi__get16be(s); // skip `fields' + stbi__get16be(s); // skip `pad' - // intermediate buffer is RGBA - result = (stbi_uc *) stbi__malloc_mad3(x, y, 4, 0); - memset(result, 0xff, x*y*4); + // intermediate buffer is RGBA + result = (stbi_uc *)stbi__malloc_mad3(x, y, 4, 0); + memset(result, 0xff, x * y * 4); - if (!stbi__pic_load_core(s,x,y,comp, result)) { - STBI_FREE(result); - result=0; - } - *px = x; - *py = y; - if (req_comp == 0) req_comp = *comp; - result=stbi__convert_format(result,4,req_comp,x,y); + if (!stbi__pic_load_core(s, x, y, comp, result)) { + STBI_FREE(result); + result = 0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result = stbi__convert_format(result, 4, req_comp, x, y); - return result; + return result; } -static int stbi__pic_test(stbi__context *s) -{ - int r = stbi__pic_test_core(s); - stbi__rewind(s); - return r; +static int stbi__pic_test(stbi__context *s) { + int r = stbi__pic_test_core(s); + stbi__rewind(s); + return r; } -#endif + #endif // ************************************************************************************************* // GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb -#ifndef STBI_NO_GIF -typedef struct -{ - stbi__int16 prefix; - stbi_uc first; - stbi_uc suffix; + #ifndef STBI_NO_GIF +typedef struct { + stbi__int16 prefix; + stbi_uc first; + stbi_uc suffix; } stbi__gif_lzw; -typedef struct -{ - int w,h; - stbi_uc *out; // output buffer (always 4 components) - stbi_uc *background; // The current "background" as far as a gif is concerned - stbi_uc *history; - int flags, bgindex, ratio, transparent, eflags; - stbi_uc pal[256][4]; - stbi_uc lpal[256][4]; - stbi__gif_lzw codes[8192]; - stbi_uc *color_table; - int parse, step; - int lflags; - int start_x, start_y; - int max_x, max_y; - int cur_x, cur_y; - int line_size; - int delay; +typedef struct { + int w, h; + stbi_uc *out; // output buffer (always 4 components) + stbi_uc *background; // The current "background" as far as a gif is concerned + stbi_uc *history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + stbi_uc *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; + int delay; } stbi__gif; -static int stbi__gif_test_raw(stbi__context *s) -{ - int sz; - if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') return 0; - sz = stbi__get8(s); - if (sz != '9' && sz != '7') return 0; - if (stbi__get8(s) != 'a') return 0; - return 1; +static int stbi__gif_test_raw(stbi__context *s) { + int sz; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || + stbi__get8(s) != '8') + return 0; + sz = stbi__get8(s); + if (sz != '9' && sz != '7') return 0; + if (stbi__get8(s) != 'a') return 0; + return 1; } -static int stbi__gif_test(stbi__context *s) -{ - int r = stbi__gif_test_raw(s); - stbi__rewind(s); - return r; +static int stbi__gif_test(stbi__context *s) { + int r = stbi__gif_test_raw(s); + stbi__rewind(s); + return r; } -static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], int num_entries, int transp) -{ - int i; - for (i=0; i < num_entries; ++i) { - pal[i][2] = stbi__get8(s); - pal[i][1] = stbi__get8(s); - pal[i][0] = stbi__get8(s); - pal[i][3] = transp == i ? 0 : 255; - } +static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], + int num_entries, int transp) { + int i; + for (i = 0; i < num_entries; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + pal[i][3] = transp == i ? 0 : 255; + } } -static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, int is_info) -{ - stbi_uc version; - if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') - return stbi__err("not GIF", "Corrupt GIF"); +static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, + int is_info) { + stbi_uc version; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || + stbi__get8(s) != '8') + return stbi__err("not GIF", "Corrupt GIF"); - version = stbi__get8(s); - if (version != '7' && version != '9') return stbi__err("not GIF", "Corrupt GIF"); - if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); + version = stbi__get8(s); + if (version != '7' && version != '9') + return stbi__err("not GIF", "Corrupt GIF"); + if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); - stbi__g_failure_reason = ""; - g->w = stbi__get16le(s); - g->h = stbi__get16le(s); - g->flags = stbi__get8(s); - g->bgindex = stbi__get8(s); - g->ratio = stbi__get8(s); - g->transparent = -1; + stbi__g_failure_reason = ""; + g->w = stbi__get16le(s); + g->h = stbi__get16le(s); + g->flags = stbi__get8(s); + g->bgindex = stbi__get8(s); + g->ratio = stbi__get8(s); + g->transparent = -1; - if (g->w > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (g->h > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (g->w > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (g->h > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); - if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + if (comp != 0) + *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the + // comments - if (is_info) return 1; + if (is_info) return 1; - if (g->flags & 0x80) - stbi__gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + if (g->flags & 0x80) + stbi__gif_parse_colortable(s, g->pal, 2 << (g->flags & 7), -1); - return 1; + return 1; } -static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) -{ - stbi__gif* g = (stbi__gif*) stbi__malloc(sizeof(stbi__gif)); - if (!stbi__gif_header(s, g, comp, 1)) { - STBI_FREE(g); - stbi__rewind( s ); - return 0; - } - if (x) *x = g->w; - if (y) *y = g->h; - STBI_FREE(g); - return 1; +static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) { + stbi__gif *g = (stbi__gif *)stbi__malloc(sizeof(stbi__gif)); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind(s); + return 0; + } + if (x) *x = g->w; + if (y) *y = g->h; + STBI_FREE(g); + return 1; } -static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) -{ - stbi_uc *p, *c; - int idx; +static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) { + stbi_uc *p, *c; + int idx; - // recurse to decode the prefixes, since the linked-list is backwards, - // and working backwards through an interleaved image would be nasty - if (g->codes[code].prefix >= 0) - stbi__out_gif_code(g, g->codes[code].prefix); + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) stbi__out_gif_code(g, g->codes[code].prefix); - if (g->cur_y >= g->max_y) return; + if (g->cur_y >= g->max_y) return; - idx = g->cur_x + g->cur_y; - p = &g->out[idx]; - g->history[idx / 4] = 1; + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; - c = &g->color_table[g->codes[code].suffix * 4]; - if (c[3] > 128) { // don't render transparent pixels; - p[0] = c[2]; - p[1] = c[1]; - p[2] = c[0]; - p[3] = c[3]; - } - g->cur_x += 4; + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; - if (g->cur_x >= g->max_x) { - g->cur_x = g->start_x; - g->cur_y += g->step; + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; - while (g->cur_y >= g->max_y && g->parse > 0) { - g->step = (1 << g->parse) * g->line_size; - g->cur_y = g->start_y + (g->step >> 1); - --g->parse; + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) { + stbi_uc lzw_cs; + stbi__int32 len, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw *p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc)init_code; + g->codes[init_code].suffix = (stbi_uc)init_code; + } + + // support no starting clear code + avail = clear + 2; + oldcode = -1; + + len = 0; + for (;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = stbi__get8(s); // start new block + if (len == 0) return g->out; } - } + --len; + bits |= (stbi__int32)stbi__get8(s) << valid_bits; + valid_bits += 8; + } else { + stbi__int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + stbi__skip(s, len); + while ((len = stbi__get8(s)) > 0) + stbi__skip(s, len); + return g->out; + } else if (code <= avail) { + if (first) { return stbi__errpuc("no clear code", "Corrupt GIF"); } + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 8192) { + return stbi__errpuc("too many codes", "Corrupt GIF"); + } + + p->prefix = (stbi__int16)oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + + stbi__out_gif_code(g, (stbi__uint16)code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + } + } + } } -static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) -{ - stbi_uc lzw_cs; - stbi__int32 len, init_code; - stbi__uint32 first; - stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; - stbi__gif_lzw *p; +// this function is designed to support animated gifs, although stb_image +// doesn't support it two back is the image from two frames ago, used for a very +// specific disposal format +static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, + int req_comp, stbi_uc *two_back) { + int dispose; + int first_frame; + int pi; + int pcount; + STBI_NOTUSED(req_comp); - lzw_cs = stbi__get8(s); - if (lzw_cs > 12) return NULL; - clear = 1 << lzw_cs; - first = 1; - codesize = lzw_cs + 1; - codemask = (1 << codesize) - 1; - bits = 0; - valid_bits = 0; - for (init_code = 0; init_code < clear; init_code++) { - g->codes[init_code].prefix = -1; - g->codes[init_code].first = (stbi_uc) init_code; - g->codes[init_code].suffix = (stbi_uc) init_code; - } + // on first frame, any non-written pixels get the background colour + // (non-transparent) + first_frame = 0; + if (g->out == 0) { + if (!stbi__gif_header(s, g, comp, 0)) + return 0; // stbi__g_failure_reason set by stbi__gif_header + if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) + return stbi__errpuc("too large", "GIF image is too large"); + pcount = g->w * g->h; + g->out = (stbi_uc *)stbi__malloc(4 * pcount); + g->background = (stbi_uc *)stbi__malloc(4 * pcount); + g->history = (stbi_uc *)stbi__malloc(pcount); + if (!g->out || !g->background || !g->history) + return stbi__errpuc("outofmem", "Out of memory"); - // support no starting clear code - avail = clear+2; - oldcode = -1; + // image is treated as "transparent" at the start - ie, nothing overwrites + // the current background; background colour is only used for pixels that + // are not rendered first frame, after that "background" color refers to the + // color that was there the previous frame. + memset(g->out, 0x00, 4 * pcount); + memset(g->background, 0x00, + 4 * pcount); // state of the background (starts transparent) + memset(g->history, 0x00, + pcount); // pixels that were affected previous frame + first_frame = 1; + } else { + // second frame - how do we dispose of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; - len = 0; - for(;;) { - if (valid_bits < codesize) { - if (len == 0) { - len = stbi__get8(s); // start new block - if (len == 0) - return g->out; - } - --len; - bits |= (stbi__int32) stbi__get8(s) << valid_bits; - valid_bits += 8; - } else { - stbi__int32 code = bits & codemask; - bits >>= codesize; - valid_bits -= codesize; - // @OPTIMIZE: is there some way we can accelerate the non-clear path? - if (code == clear) { // clear code - codesize = lzw_cs + 1; - codemask = (1 << codesize) - 1; - avail = clear + 2; - oldcode = -1; - first = 0; - } else if (code == clear + 1) { // end of stream code + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to + // the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { memcpy(&g->out[pi * 4], &two_back[pi * 4], 4); } + } + } else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy(&g->out[pi * 4], &g->background[pi * 4], 4); + } + } + } else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy(g->background, g->out, 4 * g->w * g->h); + } + + // clear my history; + memset(g->history, 0x00, + g->w * g->h); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + case 0x2C: /* Image Descriptor */ + { + stbi__int32 x, y, w, h; + stbi_uc *o; + + x = stbi__get16le(s); + y = stbi__get16le(s); + w = stbi__get16le(s); + h = stbi__get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + // if the width of the specified rectangle is 0, that means + // we may not see *any* pixels or the image is malformed; + // to make sure this is caught, move the current y down to + // max_y (which is what out_gif_code checks). + if (w == 0) g->cur_y = g->max_y; + + g->lflags = stbi__get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi__gif_parse_colortable(s, g->lpal, 2 << (g->lflags & 7), + g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (stbi_uc *)g->lpal; + } else if (g->flags & 0x80) { + g->color_table = (stbi_uc *)g->pal; + } else + return stbi__errpuc("missing color table", "Corrupt GIF"); + + o = stbi__process_gif_raster(s, g); + if (!o) return NULL; + + // if this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = + 255; // just in case it was made transparent, undo that; It + // will be reset next frame if need be; + memcpy(&g->out[pi * 4], &g->pal[g->bgindex], 4); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = + 10 * + stbi__get16le( + s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { g->pal[g->transparent][3] = 255; } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { g->pal[g->transparent][3] = 0; } + } else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } else { stbi__skip(s, len); - while ((len = stbi__get8(s)) > 0) - stbi__skip(s,len); - return g->out; - } else if (code <= avail) { - if (first) { - return stbi__errpuc("no clear code", "Corrupt GIF"); - } - - if (oldcode >= 0) { - p = &g->codes[avail++]; - if (avail > 8192) { - return stbi__errpuc("too many codes", "Corrupt GIF"); - } - - p->prefix = (stbi__int16) oldcode; - p->first = g->codes[oldcode].first; - p->suffix = (code == avail) ? p->first : g->codes[code].first; - } else if (code == avail) - return stbi__errpuc("illegal code in raster", "Corrupt GIF"); - - stbi__out_gif_code(g, (stbi__uint16) code); - - if ((avail & codemask) == 0 && avail <= 0x0FFF) { - codesize++; - codemask = (1 << codesize) - 1; - } - - oldcode = code; - } else { - return stbi__errpuc("illegal code in raster", "Corrupt GIF"); - } - } - } -} - -// this function is designed to support animated gifs, although stb_image doesn't support it -// two back is the image from two frames ago, used for a very specific disposal format -static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, int req_comp, stbi_uc *two_back) -{ - int dispose; - int first_frame; - int pi; - int pcount; - STBI_NOTUSED(req_comp); - - // on first frame, any non-written pixels get the background colour (non-transparent) - first_frame = 0; - if (g->out == 0) { - if (!stbi__gif_header(s, g, comp,0)) return 0; // stbi__g_failure_reason set by stbi__gif_header - if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) - return stbi__errpuc("too large", "GIF image is too large"); - pcount = g->w * g->h; - g->out = (stbi_uc *) stbi__malloc(4 * pcount); - g->background = (stbi_uc *) stbi__malloc(4 * pcount); - g->history = (stbi_uc *) stbi__malloc(pcount); - if (!g->out || !g->background || !g->history) - return stbi__errpuc("outofmem", "Out of memory"); - - // image is treated as "transparent" at the start - ie, nothing overwrites the current background; - // background colour is only used for pixels that are not rendered first frame, after that "background" - // color refers to the color that was there the previous frame. - memset(g->out, 0x00, 4 * pcount); - memset(g->background, 0x00, 4 * pcount); // state of the background (starts transparent) - memset(g->history, 0x00, pcount); // pixels that were affected previous frame - first_frame = 1; - } else { - // second frame - how do we dispose of the previous one? - dispose = (g->eflags & 0x1C) >> 2; - pcount = g->w * g->h; - - if ((dispose == 3) && (two_back == 0)) { - dispose = 2; // if I don't have an image to revert back to, default to the old background - } - - if (dispose == 3) { // use previous graphic - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi]) { - memcpy( &g->out[pi * 4], &two_back[pi * 4], 4 ); - } - } - } else if (dispose == 2) { - // restore what was changed last frame to background before that frame; - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi]) { - memcpy( &g->out[pi * 4], &g->background[pi * 4], 4 ); - } - } - } else { - // This is a non-disposal case eithe way, so just - // leave the pixels as is, and they will become the new background - // 1: do not dispose - // 0: not specified. - } - - // background is what out is after the undoing of the previou frame; - memcpy( g->background, g->out, 4 * g->w * g->h ); - } - - // clear my history; - memset( g->history, 0x00, g->w * g->h ); // pixels that were affected previous frame - - for (;;) { - int tag = stbi__get8(s); - switch (tag) { - case 0x2C: /* Image Descriptor */ - { - stbi__int32 x, y, w, h; - stbi_uc *o; - - x = stbi__get16le(s); - y = stbi__get16le(s); - w = stbi__get16le(s); - h = stbi__get16le(s); - if (((x + w) > (g->w)) || ((y + h) > (g->h))) - return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); - - g->line_size = g->w * 4; - g->start_x = x * 4; - g->start_y = y * g->line_size; - g->max_x = g->start_x + w * 4; - g->max_y = g->start_y + h * g->line_size; - g->cur_x = g->start_x; - g->cur_y = g->start_y; - - // if the width of the specified rectangle is 0, that means - // we may not see *any* pixels or the image is malformed; - // to make sure this is caught, move the current y down to - // max_y (which is what out_gif_code checks). - if (w == 0) - g->cur_y = g->max_y; - - g->lflags = stbi__get8(s); - - if (g->lflags & 0x40) { - g->step = 8 * g->line_size; // first interlaced spacing - g->parse = 3; - } else { - g->step = g->line_size; - g->parse = 0; - } - - if (g->lflags & 0x80) { - stbi__gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); - g->color_table = (stbi_uc *) g->lpal; - } else if (g->flags & 0x80) { - g->color_table = (stbi_uc *) g->pal; - } else - return stbi__errpuc("missing color table", "Corrupt GIF"); - - o = stbi__process_gif_raster(s, g); - if (!o) return NULL; - - // if this was the first frame, - pcount = g->w * g->h; - if (first_frame && (g->bgindex > 0)) { - // if first frame, any pixel not drawn to gets the background color - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi] == 0) { - g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; - memcpy( &g->out[pi * 4], &g->pal[g->bgindex], 4 ); - } - } - } - - return o; - } - - case 0x21: // Comment Extension. - { - int len; - int ext = stbi__get8(s); - if (ext == 0xF9) { // Graphic Control Extension. - len = stbi__get8(s); - if (len == 4) { - g->eflags = stbi__get8(s); - g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. - - // unset old transparent - if (g->transparent >= 0) { - g->pal[g->transparent][3] = 255; - } - if (g->eflags & 0x01) { - g->transparent = stbi__get8(s); - if (g->transparent >= 0) { - g->pal[g->transparent][3] = 0; - } - } else { - // don't need transparent - stbi__skip(s, 1); - g->transparent = -1; - } - } else { - stbi__skip(s, len); - break; - } - } - while ((len = stbi__get8(s)) != 0) { - stbi__skip(s, len); - } break; - } - - case 0x3B: // gif stream termination code - return (stbi_uc *) s; // using '1' causes warning on some compilers - - default: - return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } + while ((len = stbi__get8(s)) != 0) { + stbi__skip(s, len); + } + break; } - } + + case 0x3B: // gif stream termination code + return (stbi_uc *)s; // using '1' causes warning on some compilers + + default: + return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } } -static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp) -{ - if (stbi__gif_test(s)) { - int layers = 0; - stbi_uc *u = 0; - stbi_uc *out = 0; - stbi_uc *two_back = 0; - stbi__gif g; - int stride; - int out_size = 0; - int delays_size = 0; - memset(&g, 0, sizeof(g)); - if (delays) { - *delays = 0; +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, + int *z, int *comp, int req_comp) { + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc *u = 0; + stbi_uc *out = 0; + stbi_uc *two_back = 0; + stbi__gif g; + int stride; + int out_size = 0; + int delays_size = 0; + memset(&g, 0, sizeof(g)); + if (delays) { *delays = 0; } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc *)s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + void *tmp = + (stbi_uc *)STBI_REALLOC_SIZED(out, out_size, layers * stride); + if (NULL == tmp) { + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + return stbi__errpuc("outofmem", "Out of memory"); + } else { + out = (stbi_uc *)tmp; + out_size = layers * stride; + } + + if (delays) { + *delays = (int *)STBI_REALLOC_SIZED(*delays, delays_size, + sizeof(int) * layers); + delays_size = layers * sizeof(int); + } + } else { + out = (stbi_uc *)stbi__malloc(layers * stride); + out_size = layers * stride; + if (delays) { + *delays = (int *)stbi__malloc(layers * sizeof(int)); + delays_size = layers * sizeof(int); + } + } + memcpy(out + ((layers - 1) * stride), u, stride); + if (layers >= 2) { two_back = out - 2 * stride; } + + if (delays) { (*delays)[layers - 1U] = g.delay; } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + STBI_NOTUSED(ri); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc *)s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } else if (g.out) { + // if there was an error and we allocated an image buffer, free it! + STBI_FREE(g.out); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + return u; +} + +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) { + return stbi__gif_info_raw(s, x, y, comp); +} + #endif + + // ************************************************************************************************* + // Radiance RGBE HDR loader + // originally by Nicolas Schulz + #ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context *s, const char *signature) { + int i; + for (i = 0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context *s) { + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if (!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + stbi__rewind(s); + } + return r; +} + + #define STBI__HDR_BUFLEN 1024 +static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) { + int len = 0; + char c = '\0'; + + c = (char)stbi__get8(z); + + while (!stbi__at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == STBI__HDR_BUFLEN - 1) { + // flush to end of line + while (!stbi__at_eof(z) && stbi__get8(z) != '\n') + ; + break; + } + c = (char)stbi__get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) { + if (input[3] != 0) { + float f1; + // Exponent + f1 = (float)ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: + output[3] = 1; /* fallthrough */ + case 3: + output[0] = output[1] = output[2] = 0; + break; + case 2: + output[1] = 1; /* fallthrough */ + case 1: + output[0] = 0; + break; + } + } +} + +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1, c2, z; + const char *headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s, buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && + strcmp(headerToken, "#?RGBE") != 0) + return stbi__errpf("not HDR", "Corrupt HDR image"); + + // Parse header + for (;;) { + token = stbi__hdr_gettoken(s, buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) + return stbi__errpf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = stbi__hdr_gettoken(s, buffer); + if (strncmp(token, "-Y ", 3)) + return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = (int)strtol(token, &token, 10); + while (*token == ' ') + ++token; + if (strncmp(token, "+X ", 3)) + return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = (int)strtol(token, NULL, 10); + + if (height > STBI_MAX_DIMENSIONS) + return stbi__errpf("too large", "Very large image (corrupt?)"); + if (width > STBI_MAX_DIMENSIONS) + return stbi__errpf("too large", "Very large image (corrupt?)"); + + *x = width; + *y = height; + + if (comp) *comp = 3; + if (req_comp == 0) req_comp = 3; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = + (float *)stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) return stbi__errpf("outofmem", "Out of memory"); + + // Load image data + // image data is stored as some number of sca + if (width < 8 || width >= 32768) { + // Read flat data + for (j = 0; j < height; ++j) { + for (i = 0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + stbi__getn(s, rgbe, 4); + stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, + req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = stbi__get8(s); + c2 = stbi__get8(s); + len = stbi__get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a + // decoded pixel (note this can't be a valid pixel--one of RGB must be + // >= 128) + stbi_uc rgbe[4]; + rgbe[0] = (stbi_uc)c1; + rgbe[1] = (stbi_uc)c2; + rgbe[2] = (stbi_uc)len; + rgbe[3] = (stbi_uc)stbi__get8(s); + stbi__hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); + } + if (scanline == NULL) { + scanline = (stbi_uc *)stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } } - do { - u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); - if (u == (stbi_uc *) s) u = 0; // end of animated gif marker - - if (u) { - *x = g.w; - *y = g.h; - ++layers; - stride = g.w * g.h * 4; - - if (out) { - void *tmp = (stbi_uc*) STBI_REALLOC_SIZED( out, out_size, layers * stride ); - if (NULL == tmp) { - STBI_FREE(g.out); - STBI_FREE(g.history); - STBI_FREE(g.background); - return stbi__errpuc("outofmem", "Out of memory"); - } - else { - out = (stbi_uc*) tmp; - out_size = layers * stride; - } - - if (delays) { - *delays = (int*) STBI_REALLOC_SIZED( *delays, delays_size, sizeof(int) * layers ); - delays_size = layers * sizeof(int); - } - } else { - out = (stbi_uc*)stbi__malloc( layers * stride ); - out_size = layers * stride; - if (delays) { - *delays = (int*) stbi__malloc( layers * sizeof(int) ); - delays_size = layers * sizeof(int); - } + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if (count > nleft) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("corrupt", "bad RLE data in HDR"); } - memcpy( out + ((layers - 1) * stride), u, stride ); - if (layers >= 2) { - two_back = out - 2 * stride; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + if (count > nleft) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("corrupt", "bad RLE data in HDR"); } - - if (delays) { - (*delays)[layers - 1U] = g.delay; - } - } - } while (u != 0); - - // free temp buffer; - STBI_FREE(g.out); - STBI_FREE(g.history); - STBI_FREE(g.background); - - // do the final conversion after loading everything; - if (req_comp && req_comp != 4) - out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); - - *z = layers; - return out; - } else { - return stbi__errpuc("not GIF", "Image was not as a gif type."); - } -} - -static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *u = 0; - stbi__gif g; - memset(&g, 0, sizeof(g)); - STBI_NOTUSED(ri); - - u = stbi__gif_load_next(s, &g, comp, req_comp, 0); - if (u == (stbi_uc *) s) u = 0; // end of animated gif marker - if (u) { - *x = g.w; - *y = g.h; - - // moved conversion to after successful load so that the same - // can be done for multiple frames. - if (req_comp && req_comp != 4) - u = stbi__convert_format(u, 4, req_comp, g.w, g.h); - } else if (g.out) { - // if there was an error and we allocated an image buffer, free it! - STBI_FREE(g.out); - } - - // free buffers needed for multiple frame loading; - STBI_FREE(g.history); - STBI_FREE(g.background); - - return u; -} - -static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) -{ - return stbi__gif_info_raw(s,x,y,comp); -} -#endif - -// ************************************************************************************************* -// Radiance RGBE HDR loader -// originally by Nicolas Schulz -#ifndef STBI_NO_HDR -static int stbi__hdr_test_core(stbi__context *s, const char *signature) -{ - int i; - for (i=0; signature[i]; ++i) - if (stbi__get8(s) != signature[i]) - return 0; - stbi__rewind(s); - return 1; -} - -static int stbi__hdr_test(stbi__context* s) -{ - int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); - stbi__rewind(s); - if(!r) { - r = stbi__hdr_test_core(s, "#?RGBE\n"); - stbi__rewind(s); - } - return r; -} - -#define STBI__HDR_BUFLEN 1024 -static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) -{ - int len=0; - char c = '\0'; - - c = (char) stbi__get8(z); - - while (!stbi__at_eof(z) && c != '\n') { - buffer[len++] = c; - if (len == STBI__HDR_BUFLEN-1) { - // flush to end of line - while (!stbi__at_eof(z) && stbi__get8(z) != '\n') - ; - break; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = stbi__get8(s); + } + } } - c = (char) stbi__get8(z); - } + for (i = 0; i < width; ++i) + stbi__hdr_convert(hdr_data + (j * width + i) * req_comp, + scanline + i * 4, req_comp); + } + if (scanline) STBI_FREE(scanline); + } - buffer[len] = 0; - return buffer; + return hdr_data; } -static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) -{ - if ( input[3] != 0 ) { - float f1; - // Exponent - f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); - if (req_comp <= 2) - output[0] = (input[0] + input[1] + input[2]) * f1 / 3; - else { - output[0] = input[0] * f1; - output[1] = input[1] * f1; - output[2] = input[2] * f1; - } - if (req_comp == 2) output[1] = 1; - if (req_comp == 4) output[3] = 1; - } else { - switch (req_comp) { - case 4: output[3] = 1; /* fallthrough */ - case 3: output[0] = output[1] = output[2] = 0; - break; - case 2: output[1] = 1; /* fallthrough */ - case 1: output[0] = 0; - break; - } - } +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) { + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 0) { + stbi__rewind(s); + return 0; + } + + for (;;) { + token = stbi__hdr_gettoken(s, buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi__rewind(s); + return 0; + } + token = stbi__hdr_gettoken(s, buffer); + if (strncmp(token, "-Y ", 3)) { + stbi__rewind(s); + return 0; + } + token += 3; + *y = (int)strtol(token, &token, 10); + while (*token == ' ') + ++token; + if (strncmp(token, "+X ", 3)) { + stbi__rewind(s); + return 0; + } + token += 3; + *x = (int)strtol(token, NULL, 10); + *comp = 3; + return 1; +} + #endif // STBI_NO_HDR + + #ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) { + void *p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + stbi__rewind(s); + if (p == NULL) return 0; + if (x) *x = s->img_x; + if (y) *y = s->img_y; + if (comp) { + if (info.bpp == 24 && info.ma == 0xff000000) + *comp = 3; + else + *comp = info.ma ? 4 : 3; + } + return 1; +} + #endif + + #ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) { + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind(s); + return 0; + } + *y = stbi__get32be(s); + *x = stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind(s); + return 0; + } + *comp = 4; + return 1; } -static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - char buffer[STBI__HDR_BUFLEN]; - char *token; - int valid = 0; - int width, height; - stbi_uc *scanline; - float *hdr_data; - int len; - unsigned char count, value; - int i, j, k, c1,c2, z; - const char *headerToken; - STBI_NOTUSED(ri); - - // Check identifier - headerToken = stbi__hdr_gettoken(s,buffer); - if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 0) - return stbi__errpf("not HDR", "Corrupt HDR image"); - - // Parse header - for(;;) { - token = stbi__hdr_gettoken(s,buffer); - if (token[0] == 0) break; - if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; - } - - if (!valid) return stbi__errpf("unsupported format", "Unsupported HDR format"); - - // Parse width and height - // can't use sscanf() if we're not using stdio! - token = stbi__hdr_gettoken(s,buffer); - if (strncmp(token, "-Y ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); - token += 3; - height = (int) strtol(token, &token, 10); - while (*token == ' ') ++token; - if (strncmp(token, "+X ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); - token += 3; - width = (int) strtol(token, NULL, 10); - - if (height > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); - if (width > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); - - *x = width; - *y = height; - - if (comp) *comp = 3; - if (req_comp == 0) req_comp = 3; - - if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) - return stbi__errpf("too large", "HDR image is too large"); - - // Read data - hdr_data = (float *) stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); - if (!hdr_data) - return stbi__errpf("outofmem", "Out of memory"); - - // Load image data - // image data is stored as some number of sca - if ( width < 8 || width >= 32768) { - // Read flat data - for (j=0; j < height; ++j) { - for (i=0; i < width; ++i) { - stbi_uc rgbe[4]; - main_decode_loop: - stbi__getn(s, rgbe, 4); - stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); - } - } - } else { - // Read RLE-encoded data - scanline = NULL; - - for (j = 0; j < height; ++j) { - c1 = stbi__get8(s); - c2 = stbi__get8(s); - len = stbi__get8(s); - if (c1 != 2 || c2 != 2 || (len & 0x80)) { - // not run-length encoded, so we have to actually use THIS data as a decoded - // pixel (note this can't be a valid pixel--one of RGB must be >= 128) - stbi_uc rgbe[4]; - rgbe[0] = (stbi_uc) c1; - rgbe[1] = (stbi_uc) c2; - rgbe[2] = (stbi_uc) len; - rgbe[3] = (stbi_uc) stbi__get8(s); - stbi__hdr_convert(hdr_data, rgbe, req_comp); - i = 1; - j = 0; - STBI_FREE(scanline); - goto main_decode_loop; // yes, this makes no sense - } - len <<= 8; - len |= stbi__get8(s); - if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } - if (scanline == NULL) { - scanline = (stbi_uc *) stbi__malloc_mad2(width, 4, 0); - if (!scanline) { - STBI_FREE(hdr_data); - return stbi__errpf("outofmem", "Out of memory"); - } - } - - for (k = 0; k < 4; ++k) { - int nleft; - i = 0; - while ((nleft = width - i) > 0) { - count = stbi__get8(s); - if (count > 128) { - // Run - value = stbi__get8(s); - count -= 128; - if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } - for (z = 0; z < count; ++z) - scanline[i++ * 4 + k] = value; - } else { - // Dump - if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } - for (z = 0; z < count; ++z) - scanline[i++ * 4 + k] = stbi__get8(s); - } - } - } - for (i=0; i < width; ++i) - stbi__hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); - } - if (scanline) - STBI_FREE(scanline); - } - - return hdr_data; +static int stbi__psd_is16(stbi__context *s) { + int channelCount, depth; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind(s); + return 0; + } + (void)stbi__get32be(s); + (void)stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind(s); + return 0; + } + return 1; } + #endif -static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) -{ - char buffer[STBI__HDR_BUFLEN]; - char *token; - int valid = 0; - int dummy; + #ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) { + int act_comp = 0, num_packets = 0, chained, dummy; + stbi__pic_packet packets[10]; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; - if (stbi__hdr_test(s) == 0) { - stbi__rewind( s ); - return 0; - } + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } - for(;;) { - token = stbi__hdr_gettoken(s,buffer); - if (token[0] == 0) break; - if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; - } + stbi__skip(s, 88); - if (!valid) { - stbi__rewind( s ); - return 0; - } - token = stbi__hdr_gettoken(s,buffer); - if (strncmp(token, "-Y ", 3)) { - stbi__rewind( s ); - return 0; - } - token += 3; - *y = (int) strtol(token, &token, 10); - while (*token == ' ') ++token; - if (strncmp(token, "+X ", 3)) { - stbi__rewind( s ); - return 0; - } - token += 3; - *x = (int) strtol(token, NULL, 10); - *comp = 3; - return 1; -} -#endif // STBI_NO_HDR + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind(s); + return 0; + } + if ((*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi__rewind(s); + return 0; + } -#ifndef STBI_NO_BMP -static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) -{ - void *p; - stbi__bmp_data info; + stbi__skip(s, 8); - info.all_a = 255; - p = stbi__bmp_parse_header(s, &info); - stbi__rewind( s ); - if (p == NULL) - return 0; - if (x) *x = s->img_x; - if (y) *y = s->img_y; - if (comp) { - if (info.bpp == 24 && info.ma == 0xff000000) - *comp = 3; - else - *comp = info.ma ? 4 : 3; - } - return 1; -} -#endif + do { + stbi__pic_packet *packet; -#ifndef STBI_NO_PSD -static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) -{ - int channelCount, dummy, depth; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; - if (stbi__get32be(s) != 0x38425053) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 1) { - stbi__rewind( s ); - return 0; - } - stbi__skip(s, 6); - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) { - stbi__rewind( s ); - return 0; - } - *y = stbi__get32be(s); - *x = stbi__get32be(s); - depth = stbi__get16be(s); - if (depth != 8 && depth != 16) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 3) { - stbi__rewind( s ); - return 0; - } - *comp = 4; - return 1; -} + if (num_packets == sizeof(packets) / sizeof(packets[0])) return 0; -static int stbi__psd_is16(stbi__context *s) -{ - int channelCount, depth; - if (stbi__get32be(s) != 0x38425053) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 1) { - stbi__rewind( s ); - return 0; - } - stbi__skip(s, 6); - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) { - stbi__rewind( s ); - return 0; - } - (void) stbi__get32be(s); - (void) stbi__get32be(s); - depth = stbi__get16be(s); - if (depth != 16) { - stbi__rewind( s ); - return 0; - } - return 1; -} -#endif + packet = &packets[num_packets++]; + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + act_comp |= packet->channel; -#ifndef STBI_NO_PIC -static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) -{ - int act_comp=0,num_packets=0,chained,dummy; - stbi__pic_packet packets[10]; - - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; - - if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) { + if (stbi__at_eof(s)) { stbi__rewind(s); return 0; - } - - stbi__skip(s, 88); - - *x = stbi__get16be(s); - *y = stbi__get16be(s); - if (stbi__at_eof(s)) { - stbi__rewind( s); + } + if (packet->size != 8) { + stbi__rewind(s); return 0; - } - if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { - stbi__rewind( s ); - return 0; - } + } + } while (chained); - stbi__skip(s, 8); + *comp = (act_comp & 0x10 ? 4 : 3); - do { - stbi__pic_packet *packet; - - if (num_packets==sizeof(packets)/sizeof(packets[0])) - return 0; - - packet = &packets[num_packets++]; - chained = stbi__get8(s); - packet->size = stbi__get8(s); - packet->type = stbi__get8(s); - packet->channel = stbi__get8(s); - act_comp |= packet->channel; - - if (stbi__at_eof(s)) { - stbi__rewind( s ); - return 0; - } - if (packet->size != 8) { - stbi__rewind( s ); - return 0; - } - } while (chained); - - *comp = (act_comp & 0x10 ? 4 : 3); - - return 1; + return 1; } -#endif + #endif // ************************************************************************************************* // Portable Gray Map and Portable Pixel Map loader @@ -7288,259 +7825,248 @@ static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) // Does not support ASCII image data (formats P2 and P3) // Does not support 16-bit-per-channel -#ifndef STBI_NO_PNM + #ifndef STBI_NO_PNM -static int stbi__pnm_test(stbi__context *s) -{ - char p, t; - p = (char) stbi__get8(s); - t = (char) stbi__get8(s); - if (p != 'P' || (t != '5' && t != '6')) { - stbi__rewind( s ); - return 0; - } - return 1; +static int stbi__pnm_test(stbi__context *s) { + char p, t; + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + return 1; } -static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *out; - STBI_NOTUSED(ri); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *out; + STBI_NOTUSED(ri); - if (!stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n)) - return 0; + if (!stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n)) + return 0; - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - *x = s->img_x; - *y = s->img_y; - if (comp) *comp = s->img_n; + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; - if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) - return stbi__errpuc("too large", "PNM too large"); + if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "PNM too large"); - out = (stbi_uc *) stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - stbi__getn(s, out, s->img_n * s->img_x * s->img_y); + out = (stbi_uc *)stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + stbi__getn(s, out, s->img_n * s->img_x * s->img_y); - if (req_comp && req_comp != s->img_n) { - out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } - return out; + if (req_comp && req_comp != s->img_n) { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; } -static int stbi__pnm_isspace(char c) -{ - return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +static int stbi__pnm_isspace(char c) { + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || + c == '\r'; } -static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) -{ - for (;;) { - while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) - *c = (char) stbi__get8(s); +static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) { + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + *c = (char)stbi__get8(s); - if (stbi__at_eof(s) || *c != '#') - break; + if (stbi__at_eof(s) || *c != '#') break; - while (!stbi__at_eof(s) && *c != '\n' && *c != '\r' ) - *c = (char) stbi__get8(s); - } + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r') + *c = (char)stbi__get8(s); + } } -static int stbi__pnm_isdigit(char c) -{ - return c >= '0' && c <= '9'; +static int stbi__pnm_isdigit(char c) { + return c >= '0' && c <= '9'; } -static int stbi__pnm_getinteger(stbi__context *s, char *c) -{ - int value = 0; +static int stbi__pnm_getinteger(stbi__context *s, char *c) { + int value = 0; - while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { - value = value*10 + (*c - '0'); - *c = (char) stbi__get8(s); - } + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value * 10 + (*c - '0'); + *c = (char)stbi__get8(s); + } - return value; + return value; } -static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) -{ - int maxv, dummy; - char c, p, t; +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) { + int maxv, dummy; + char c, p, t; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; - stbi__rewind(s); + stbi__rewind(s); - // Get identifier - p = (char) stbi__get8(s); - t = (char) stbi__get8(s); - if (p != 'P' || (t != '5' && t != '6')) { - stbi__rewind(s); - return 0; - } + // Get identifier + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } - *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + *comp = + (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm - c = (char) stbi__get8(s); - stbi__pnm_skip_whitespace(s, &c); + c = (char)stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); - *x = stbi__pnm_getinteger(s, &c); // read width - stbi__pnm_skip_whitespace(s, &c); + *x = stbi__pnm_getinteger(s, &c); // read width + stbi__pnm_skip_whitespace(s, &c); - *y = stbi__pnm_getinteger(s, &c); // read height - stbi__pnm_skip_whitespace(s, &c); + *y = stbi__pnm_getinteger(s, &c); // read height + stbi__pnm_skip_whitespace(s, &c); - maxv = stbi__pnm_getinteger(s, &c); // read max value + maxv = stbi__pnm_getinteger(s, &c); // read max value - if (maxv > 255) - return stbi__err("max value > 255", "PPM image not 8-bit"); - else - return 1; + if (maxv > 255) + return stbi__err("max value > 255", "PPM image not 8-bit"); + else + return 1; } -#endif + #endif -static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) -{ - #ifndef STBI_NO_JPEG - if (stbi__jpeg_info(s, x, y, comp)) return 1; - #endif +static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) { + #ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PNG - if (stbi__png_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_GIF - if (stbi__gif_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_BMP - if (stbi__bmp_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PSD - if (stbi__psd_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PIC - if (stbi__pic_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PNM - if (stbi__pnm_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_HDR - if (stbi__hdr_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; + #endif - // test tga last because it's a crappy test! - #ifndef STBI_NO_TGA - if (stbi__tga_info(s, x, y, comp)) - return 1; - #endif - return stbi__err("unknown image type", "Image not of any known type, or corrupt"); + // test tga last because it's a crappy test! + #ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) return 1; + #endif + return stbi__err("unknown image type", + "Image not of any known type, or corrupt"); } -static int stbi__is_16_main(stbi__context *s) -{ - #ifndef STBI_NO_PNG - if (stbi__png_is16(s)) return 1; - #endif +static int stbi__is_16_main(stbi__context *s) { + #ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; + #endif - #ifndef STBI_NO_PSD - if (stbi__psd_is16(s)) return 1; - #endif + #ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; + #endif - return 0; + return 0; } -#ifndef STBI_NO_STDIO -STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result; - if (!f) return stbi__err("can't fopen", "Unable to open file"); - result = stbi_info_from_file(f, x, y, comp); - fclose(f); - return result; + #ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) { + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; } -STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) -{ - int r; - stbi__context s; - long pos = ftell(f); - stbi__start_file(&s, f); - r = stbi__info_main(&s,x,y,comp); - fseek(f,pos,SEEK_SET); - return r; +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) { + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__info_main(&s, x, y, comp); + fseek(f, pos, SEEK_SET); + return r; } -STBIDEF int stbi_is_16_bit(char const *filename) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result; - if (!f) return stbi__err("can't fopen", "Unable to open file"); - result = stbi_is_16_bit_from_file(f); - fclose(f); - return result; +STBIDEF int stbi_is_16_bit(char const *filename) { + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; } -STBIDEF int stbi_is_16_bit_from_file(FILE *f) -{ - int r; - stbi__context s; - long pos = ftell(f); - stbi__start_file(&s, f); - r = stbi__is_16_main(&s); - fseek(f,pos,SEEK_SET); - return r; +STBIDEF int stbi_is_16_bit_from_file(FILE *f) { + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + fseek(f, pos, SEEK_SET); + return r; } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__info_main(&s,x,y,comp); +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__info_main(&s, x, y, comp); } -STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); - return stbi__info_main(&s,x,y,comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, + int *x, int *y, int *comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)c, user); + return stbi__info_main(&s, x, y, comp); } -STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__is_16_main(&s); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__is_16_main(&s); } -STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); - return stbi__is_16_main(&s); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, + void *user) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)c, user); + return stbi__is_16_main(&s); } -#endif // STB_IMAGE_IMPLEMENTATION +#endif // STB_IMAGE_IMPLEMENTATION /* revision history: - 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs - 2.19 (2018-02-11) fix warning - 2.18 (2018-01-30) fix warnings - 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and + platform ifdefs 2.19 (2018-02-11) fix warning 2.18 (2018-01-30) fix + warnings 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug 1-bit BMP *_is_16_bit api avoid warnings @@ -7555,13 +8081,11 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user warning fixes; disable run-time SSE detection on gcc; uniform handling of optional "return" values; thread-safe initialization of zlib tables - 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs - 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now - 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes - 2.11 (2016-04-02) allocate large structures on the stack - remove white matting for transparent PSD - fix reported channel count for PNG & BMP - re-enable SSE2 in non-gcc 64-bit + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet + JPGs 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now 2.12 + (2016-04-02) fix typo in 2.11 PSD fix that caused crashes 2.11 (2016-04-02) + allocate large structures on the stack remove white matting for transparent + PSD fix reported channel count for PNG & BMP re-enable SSE2 in non-gcc 64-bit support RGB-formatted JPEG read 16-bit PNGs (only as 8-bit) 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED @@ -7569,11 +8093,9 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 16-bit-per-pixel TGA (not bit-per-component) info() for TGA could break due to .hdr handling info() for BMP to shares code instead of sloppy parse - can use STBI_REALLOC_SIZED if allocator doesn't support realloc - code cleanup - 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA - 2.07 (2015-09-13) fix compiler warnings - partial animated GIF support + can use STBI_REALLOC_SIZED if allocator doesn't support + realloc code cleanup 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD + as RGBA 2.07 (2015-09-13) fix compiler warnings partial animated GIF support limited 16-bpc PSD support #ifdef unused functions bug with < 92 byte PIC,PNM,HDR,TGA @@ -7584,23 +8106,18 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user stbi_set_flip_vertically_on_load (nguillemot) fix NEON support; fix mingw support 2.02 (2015-01-19) fix incorrect assert, fix warning - 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 - 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG - 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) - progressive JPEG (stb) - PGM/PPM support (Ken Miller) - STBI_MALLOC,STBI_REALLOC,STBI_FREE + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit + without -msse2 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG 2.00 + (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) progressive + JPEG (stb) PGM/PPM support (Ken Miller) STBI_MALLOC,STBI_REALLOC,STBI_FREE GIF bugfix -- seemingly never worked STBI_NO_*, STBI_ONLY_* 1.48 (2014-12-14) fix incorrectly-named assert() - 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) - optimize PNG (ryg) - fix bug in interlaced PNG with user-specified channel count (stb) - 1.46 (2014-08-26) - fix broken tRNS chunk (colorkey-style transparency) in non-paletted PNG - 1.45 (2014-08-16) - fix MSVC-ARM internal compiler error by wrapping malloc - 1.44 (2014-08-07) + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar + Cornut & stb) optimize PNG (ryg) fix bug in interlaced PNG with + user-specified channel count (stb) 1.46 (2014-08-26) fix broken tRNS chunk + (colorkey-style transparency) in non-paletted PNG 1.45 (2014-08-16) fix + MSVC-ARM internal compiler error by wrapping malloc 1.44 (2014-08-07) various warning fixes from Ronny Chevalier 1.43 (2014-07-15) fix MSVC-only compiler problem in code changed in 1.42 @@ -7609,73 +8126,48 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user fixes to stbi__cleanup_jpeg path added STBI_ASSERT to avoid requiring assert.h 1.41 (2014-06-25) - fix search&replace from 1.36 that messed up comments/error messages - 1.40 (2014-06-22) - fix gcc struct-initialization warning - 1.39 (2014-06-15) - fix to TGA optimization when req_comp != number of components in TGA; - fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my test suite) - add support for BMP version 5 (more ignored fields) - 1.38 (2014-06-06) - suppress MSVC warnings on integer casts truncating values - fix accidental rename of 'skip' field of I/O - 1.37 (2014-06-04) - remove duplicate typedef - 1.36 (2014-06-03) - convert to header file single-file library - if de-iphone isn't set, load iphone images color-swapped instead of returning NULL - 1.35 (2014-05-27) - various warnings - fix broken STBI_SIMD path - fix bug where stbi_load_from_file no longer left file pointer in correct place - fix broken non-easy path for 32-bit BMP (possibly never used) - TGA optimization by Arseny Kapoulkine - 1.34 (unknown) - use STBI_NOTUSED in stbi__resample_row_generic(), fix one more leak in tga failure case - 1.33 (2011-07-14) - make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements - 1.32 (2011-07-13) - support for "info" function for all supported filetypes (SpartanJ) - 1.31 (2011-06-20) - a few more leak fixes, bug in PNG handling (SpartanJ) - 1.30 (2011-06-11) - added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + fix search&replace from 1.36 that messed up comments/error + messages 1.40 (2014-06-22) fix gcc struct-initialization warning 1.39 + (2014-06-15) fix to TGA optimization when req_comp != number of components in + TGA; fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my + test suite) add support for BMP version 5 (more ignored fields) 1.38 + (2014-06-06) suppress MSVC warnings on integer casts truncating values fix + accidental rename of 'skip' field of I/O 1.37 (2014-06-04) remove duplicate + typedef 1.36 (2014-06-03) convert to header file single-file library if + de-iphone isn't set, load iphone images color-swapped instead of returning + NULL 1.35 (2014-05-27) various warnings fix broken STBI_SIMD path fix bug + where stbi_load_from_file no longer left file pointer in correct place fix + broken non-easy path for 32-bit BMP (possibly never used) TGA optimization by + Arseny Kapoulkine 1.34 (unknown) use STBI_NOTUSED in + stbi__resample_row_generic(), fix one more leak in tga failure case 1.33 + (2011-07-14) make stbi_is_hdr work in STBI_NO_HDR (as specified), minor + compiler-friendly improvements 1.32 (2011-07-13) support for "info" function + for all supported filetypes (SpartanJ) 1.31 (2011-06-20) a few more leak + fixes, bug in PNG handling (SpartanJ) 1.30 (2011-06-11) added ability to + load files via callbacks to accomidate custom input streams (Ben Wenger) removed deprecated format-specific test/load functions - removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway - error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) - fix inefficiency in decoding 32-bit BMP (David Woo) - 1.29 (2010-08-16) - various warning fixes from Aurelien Pocheville - 1.28 (2010-08-01) - fix bug in GIF palette transparency (SpartanJ) - 1.27 (2010-08-01) - cast-to-stbi_uc to fix warnings - 1.26 (2010-07-24) - fix bug in file buffering for PNG reported by SpartanJ - 1.25 (2010-07-17) - refix trans_data warning (Won Chun) - 1.24 (2010-07-12) - perf improvements reading from files on platforms with lock-heavy fgetc() - minor perf improvements for jpeg - deprecated type-specific functions so we'll get feedback if they're needed - attempt to fix trans_data warning (Won Chun) - 1.23 fixed bug in iPhone support - 1.22 (2010-07-10) - removed image *writing* support - stbi_info support from Jetro Lauha - GIF support from Jean-Marc Lienher + removed support for installable file formats (stbi_loader) -- + would have been broken for IO callbacks anyway error cases in bmp and tga + give messages and don't leak (Raymond Barbiero, grisha) fix inefficiency in + decoding 32-bit BMP (David Woo) 1.29 (2010-08-16) various warning fixes from + Aurelien Pocheville 1.28 (2010-08-01) fix bug in GIF palette transparency + (SpartanJ) 1.27 (2010-08-01) cast-to-stbi_uc to fix warnings 1.26 + (2010-07-24) fix bug in file buffering for PNG reported by SpartanJ 1.25 + (2010-07-17) refix trans_data warning (Won Chun) 1.24 (2010-07-12) perf + improvements reading from files on platforms with lock-heavy fgetc() minor + perf improvements for jpeg deprecated type-specific functions so we'll get + feedback if they're needed attempt to fix trans_data warning (Won Chun) 1.23 + fixed bug in iPhone support 1.22 (2010-07-10) removed image *writing* + support stbi_info support from Jetro Lauha GIF support from Jean-Marc Lienher iPhone PNG-extensions from James Brown - warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. Janez (U+017D)emva) - 1.21 fix use of 'stbi_uc' in header (reported by jon blow) - 1.20 added support for Softimage PIC, by Tom Seddon - 1.19 bug in interlaced PNG corruption check (found by ryg) - 1.18 (2008-08-02) - fix a threading bug (local mutable static) - 1.17 support interlaced PNG - 1.16 major bugfix - stbi__convert_format converted one too many pixels - 1.15 initialize some fields for thread safety - 1.14 fix threadsafe conversion bug - header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. + Janez (U+017D)emva) 1.21 fix use of 'stbi_uc' in header (reported by jon + blow) 1.20 added support for Softimage PIC, by Tom Seddon 1.19 bug in + interlaced PNG corruption check (found by ryg) 1.18 (2008-08-02) fix a + threading bug (local mutable static) 1.17 support interlaced PNG 1.16 + major bugfix - stbi__convert_format converted one too many pixels 1.15 + initialize some fields for thread safety 1.14 fix threadsafe conversion + bug header-file-only version (#define STBI_HEADER_FILE_ONLY before including) 1.13 threadsafe 1.12 const qualifiers in the API 1.11 Support installable IDCT, colorspace conversion routines @@ -7685,15 +8177,14 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz 1.07 attempt to fix C++ warning/errors again 1.06 attempt to fix C++ warning/errors again - 1.05 fix TGA loading to return correct *comp and use good luminance calc - 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free - 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR - 1.02 support for (subset of) HDR files, float interface for preferred access to them - 1.01 fix bug: possible bug in handling right-side up bmps... not sure - fix bug: the stbi__bmp_load() and stbi__tga_load() functions didn't work at all - 1.00 interface to zlib that skips zlib header - 0.99 correct handling of alpha in palette - 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 1.05 fix TGA loading to return correct *comp and use good luminance + calc 1.04 default float alpha is 1, not 255; use 'void *' for + stbi_image_free 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR 1.02 support + for (subset of) HDR files, float interface for preferred access to them 1.01 + fix bug: possible bug in handling right-side up bmps... not sure fix bug: the + stbi__bmp_load() and stbi__tga_load() functions didn't work at all 1.00 + interface to zlib that skips zlib header 0.99 correct handling of alpha in + palette 0.98 TGA loader by lonesock; dynamically add loaders (untested) 0.97 jpeg errors on too large a file; also catch another malloc failure 0.96 fix detection of invalid v value - particleman@mollyrocket forum 0.95 during header scan, seek to markers in case of padding @@ -7706,8 +8197,8 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 0.60 fix compiling as c++ 0.59 fix warnings: merge Dave Moore's -Wall fixes 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian - 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available - 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but + less than 16 available 0.56 fix bug: zlib uncompressed mode len vs. nlen 0.55 fix bug: restart_interval not initialized to 0 0.54 allow NULL for 'int *comp' 0.53 fix bug in png 3->4; speedup png decoding @@ -7718,7 +8209,6 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user first released version */ - /* ------------------------------------------------------------------------------ This software is available under 2 licenses -- choose whichever you prefer. diff --git a/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness.c b/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness.c index 44b62a113b..dbf367e47e 100644 --- a/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness.c +++ b/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness.c @@ -9,20 +9,21 @@ #include "stb_image.h" -int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) -{ - int x, y, channels; +int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + int x, y, channels; - if(!stbi_info_from_memory(data, size, &x, &y, &channels)) return 0; + if (!stbi_info_from_memory(data, size, &x, &y, &channels)) { return 0; } - /* exit if the image is larger than ~80MB */ - if(y && x > (80000000 / 4) / y) return 0; + /* exit if the image is larger than ~80MB */ + if (y && x > (80000000 / 4) / y) { return 0; } - unsigned char *img = stbi_load_from_memory(data, size, &x, &y, &channels, 4); + unsigned char *img = stbi_load_from_memory(data, size, &x, &y, &channels, 4); - free(img); - - // if (x > 10000) free(img); // free crash + free(img); - return 0; + // if (x > 10000) free(img); // free crash + + // if (x > 10000) {free(img);} // free crash + + return 0; } diff --git a/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness_symcc.c b/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness_symcc.c index f84ad0f2a1..eb1c5e2b3c 100644 --- a/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness_symcc.c +++ b/fuzzers/libfuzzer_stb_image_concolic/fuzzer/harness_symcc.c @@ -14,26 +14,22 @@ #include "stb_image.h" int main(int argc, char **argv) { - if (argc < 2) { - return -1; - } + if (argc < 2) { return -1; } char *file_path = argv[1]; int x, y, channels; - if (!stbi_info(file_path, &x, &y, &channels)) - return 0; + if (!stbi_info(file_path, &x, &y, &channels)) { return 0; } /* exit if the image is larger than ~80MB */ - if (y && x > (80000000 / 4) / y) - return 0; + if (y && x > (80000000 / 4) / y) { return 0; } unsigned char *img = stbi_load(file_path, &x, &y, &channels, 4); free(img); - // if (x > 10000) free(img); // free crash + // if (x > 10000) {free(img);} // free crash return 0; } diff --git a/fuzzers/libfuzzer_stb_image_concolic/fuzzer/stb_image.h b/fuzzers/libfuzzer_stb_image_concolic/fuzzer/stb_image.h index 94ae88d11d..eb0f8a3f71 100644 --- a/fuzzers/libfuzzer_stb_image_concolic/fuzzer/stb_image.h +++ b/fuzzers/libfuzzer_stb_image_concolic/fuzzer/stb_image.h @@ -3,7 +3,8 @@ Do this: #define STB_IMAGE_IMPLEMENTATION - before you include this file in *one* C or C++ file to create the implementation. + before you include this file in *one* C or C++ file to create the +implementation. // i.e. it should look like this: #include ... @@ -13,15 +14,16 @@ #include "stb_image.h" You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. - And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using +malloc,realloc,free QUICK NOTES: Primarily of interest to game developers and other people who can avoid problematic images and only need the trivial interface - JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) - PNG 1/2/4/8/16-bit-per-channel + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as +stock IJG lib) PNG 1/2/4/8/16-bit-per-channel TGA (not sure what subset, if a subset) BMP non-1bpp, non-RLE @@ -50,25 +52,22 @@ RECENT REVISION HISTORY: 2.26 (2020-07-13) many minor fixes 2.25 (2020-02-02) fix warnings - 2.24 (2020-02-02) fix warnings; thread-local failure_reason and flip_vertically - 2.23 (2019-08-11) fix clang static analysis warning - 2.22 (2019-03-04) gif fixes, fix warnings - 2.21 (2019-02-25) fix typo in comment - 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.24 (2020-02-02) fix warnings; thread-local failure_reason and +flip_vertically 2.23 (2019-08-11) fix clang static analysis warning 2.22 +(2019-03-04) gif fixes, fix warnings 2.21 (2019-02-25) fix typo in comment 2.20 +(2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs 2.19 (2018-02-11) fix warning 2.18 (2018-01-30) fix warnings 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings - 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes - 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC - 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs - 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes - 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes - 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 - RGB-format JPEG; remove white matting in PSD; - allocate large structures on the stack; - correct channel count for PNG & BMP - 2.10 (2016-01-22) avoid warning introduced in 2.09 - 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; +bugfixes 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE +detection on GCC 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for +Imagenet JPGs 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; +fixes 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes 2.11 +(2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 RGB-format JPEG; remove +white matting in PSD; allocate large structures on the stack; correct channel +count for PNG & BMP 2.10 (2016-01-22) avoid warning introduced in 2.09 2.09 +(2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED See end of file for full revision history. @@ -86,42 +85,41 @@ RECENT REVISION HISTORY: github:urraka (animated gif) Junggon Kim (PNM comments) Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) socks-the-fox (16-bit PNG) - Jeremy Sawicki (handle all ImageNet JPGs) - Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Jeremy Sawicki (handle all ImageNet +JPGs) Optimizations & bugfixes Mikhail Morozov (1-bit BMP) Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) Arseny Kapoulkine John-Mark Allen Carmelo J Fdez-Aguera Bug & warning fixes - Marc LeBlanc David Woo Guillaume George Martins Mozeiko - Christpher Lloyd Jerry Jansson Joseph Thomson Blazej Dariusz Roszkowski - Phil Jordan Dave Moore Roy Eltham - Hayaki Saito Nathan Reed Won Chun - Luke Graham Johan Duparc Nick Verigakis the Horde3D community - Thomas Ruf Ronny Chevalier github:rlyeh - Janez Zemva John Bartholomew Michal Cichon github:romigrou - Jonathan Blow Ken Hamada Tero Hanninen github:svdijk - Laurent Gomila Cort Stratton github:snagar - Aruelien Pocheville Sergio Gonzalez Thibault Reuille github:Zelex - Cass Everitt Ryamond Barbiero github:grim210 - Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw - Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus - Josh Tobin Matthew Gregan github:poppolopoppo - Julian Raschke Gregory Mullen Christian Floisand github:darealshinji - Baldur Karlsson Kevin Schmidt JR Smith github:Michaelangel007 - Brad Weinberger Matvey Cherevko [reserved] - Luca Sas Alexander Veselov Zack Middleton [reserved] + Marc LeBlanc David Woo Guillaume George Martins +Mozeiko Christpher Lloyd Jerry Jansson Joseph Thomson Blazej +Dariusz Roszkowski Phil Jordan Dave Moore Roy +Eltham Hayaki Saito Nathan Reed Won Chun Luke Graham Johan +Duparc Nick Verigakis the Horde3D community Thomas Ruf Ronny +Chevalier github:rlyeh Janez Zemva John +Bartholomew Michal Cichon github:romigrou Jonathan Blow Ken +Hamada Tero Hanninen github:svdijk Laurent Gomila Cort +Stratton github:snagar Aruelien Pocheville Sergio Gonzalez Thibault +Reuille github:Zelex Cass Everitt Ryamond Barbiero github:grim210 + Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw + Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus + Josh Tobin Matthew Gregan +github:poppolopoppo Julian Raschke Gregory Mullen Christian +Floisand github:darealshinji Baldur Karlsson Kevin Schmidt JR +Smith github:Michaelangel007 Brad Weinberger Matvey Cherevko +[reserved] Luca Sas Alexander Veselov Zack Middleton [reserved] Ryan C. Gordon [reserved] [reserved] DO NOT ADD YOUR NAME HERE - To add your name to the credits, pick a random blank space in the middle and fill it. - 80% of merge conflicts on stb PRs are due to people adding their name at the end - of the credits. + To add your name to the credits, pick a random blank space in the middle and +fill it. 80% of merge conflicts on stb PRs are due to people adding their name +at the end of the credits. */ #ifndef STBI_INCLUDE_STB_IMAGE_H -#define STBI_INCLUDE_STB_IMAGE_H + #define STBI_INCLUDE_STB_IMAGE_H // DOCUMENTATION // @@ -136,14 +134,15 @@ RECENT REVISION HISTORY: // // ... process data if not NULL ... // // ... x = width, y = height, n = # 8-bit components per pixel ... // // ... replace '0' with '1'..'4' to force that many components per pixel -// // ... but 'n' will always be the number that it would have been if you said 0 -// stbi_image_free(data) +// // ... but 'n' will always be the number that it would have been if you +// said 0 stbi_image_free(data) // // Standard parameters: // int *x -- outputs image width in pixels // int *y -- outputs image height in pixels // int *channels_in_file -- outputs # of image components in image file -// int desired_channels -- if non-zero, # of image components requested in result +// int desired_channels -- if non-zero, # of image components requested in +// result // // The return value from an image loader is an 'unsigned char *' which points // to the pixel data, or NULL on an allocation failure or if the image is @@ -171,8 +170,8 @@ RECENT REVISION HISTORY: // and *x, *y, *channels_in_file will be unchanged. The function // stbi_failure_reason() can be queried for an extremely brief, end-user // unfriendly explanation of why the load failed. Define STBI_NO_FAILURE_STRINGS -// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly -// more user-friendly ones. +// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get +// slightly more user-friendly ones. // // Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. // @@ -196,11 +195,12 @@ RECENT REVISION HISTORY: // 2. easy to maintain // 3. good performance // -// Sometimes I let "good performance" creep up in priority over "easy to maintain", -// and for best performance I may provide less-easy-to-use APIs that give higher -// performance, in addition to the easy-to-use ones. Nevertheless, it's important -// to keep in mind that from the standpoint of you, a client of this library, -// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// Sometimes I let "good performance" creep up in priority over "easy to +// maintain", and for best performance I may provide less-easy-to-use APIs that +// give higher performance, in addition to the easy-to-use ones. Nevertheless, +// it's important to keep in mind that from the standpoint of you, a client of +// this library, all you care about is #1 and #3, and stb libraries DO NOT +// emphasize #3 above all. // // Some secondary priorities arise directly from the first two, some of which // provide more explicit reasons why performance can't be emphasized. @@ -219,7 +219,8 @@ RECENT REVISION HISTORY: // overhead. // // The three functions you must define are "read" (reads some bytes of data), -// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the +// end). // // =========================================================================== // @@ -247,10 +248,11 @@ RECENT REVISION HISTORY: // HDR image support (disable by defining STBI_NO_HDR) // // stb_image supports loading HDR images in general, and currently the Radiance -// .HDR file format specifically. You can still load any file through the existing -// interface; if you attempt to load an HDR file, it will be automatically remapped -// to LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; -// both of these constants can be reconfigured through this interface: +// .HDR file format specifically. You can still load any file through the +// existing interface; if you attempt to load an HDR file, it will be +// automatically remapped to LDR, assuming gamma 2.2 and an arbitrary scale +// factor defaulting to 1; both of these constants can be reconfigured through +// this interface: // // stbi_hdr_to_ldr_gamma(2.2f); // stbi_hdr_to_ldr_scale(1.0f); @@ -336,37 +338,36 @@ RECENT REVISION HISTORY: // default this is set to (1 << 24), which is 16777216, but that's still // very big. -#ifndef STBI_NO_STDIO -#include -#endif // STBI_NO_STDIO + #ifndef STBI_NO_STDIO + #include + #endif // STBI_NO_STDIO -#define STBI_VERSION 1 + #define STBI_VERSION 1 -enum -{ - STBI_default = 0, // only used for desired_channels +enum { + STBI_default = 0, // only used for desired_channels - STBI_grey = 1, - STBI_grey_alpha = 2, - STBI_rgb = 3, - STBI_rgb_alpha = 4 + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 }; -#include -typedef unsigned char stbi_uc; + #include +typedef unsigned char stbi_uc; typedef unsigned short stbi_us; -#ifdef __cplusplus + #ifdef __cplusplus extern "C" { -#endif + #endif -#ifndef STBIDEF -#ifdef STB_IMAGE_STATIC -#define STBIDEF static -#else -#define STBIDEF extern -#endif -#endif + #ifndef STBIDEF + #ifdef STB_IMAGE_STATIC + #define STBIDEF static + #else + #define STBIDEF extern + #endif + #endif ////////////////////////////////////////////////////////////////////////////// // @@ -377,11 +378,13 @@ extern "C" { // load image by filename, open file, or memory buffer // -typedef struct -{ - int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read - void (*skip) (void *user,int n); // skip the next 'n' bytes, or 'unget' the last -n bytes if negative - int (*eof) (void *user); // returns nonzero if we are at end of file/data +typedef struct { + int (*read)(void *user, char *data, + int size); // fill 'data' with 'size' bytes. return number of + // bytes actually read + void (*skip)(void *user, int n); // skip the next 'n' bytes, or 'unget' the + // last -n bytes if negative + int (*eof)(void *user); // returns nonzero if we are at end of file/data } stbi_io_callbacks; //////////////////////////////////// @@ -389,376 +392,420 @@ typedef struct // 8-bits-per-channel interface // -STBIDEF stbi_uc *stbi_load_from_memory (stbi_uc const *buffer, int len , int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk , void *user, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *channels_in_file, + int desired_channels); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); -#ifndef STBI_NO_STDIO -STBIDEF stbi_uc *stbi_load (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); -// for stbi_load_from_file, file pointer is left pointing immediately after image -#endif + #ifndef STBI_NO_STDIO +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + // for stbi_load_from_file, file pointer is left pointing immediately after + // image + #endif -#ifndef STBI_NO_GIF -STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp); -#endif + #ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, + int **delays, int *x, int *y, int *z, + int *comp, int req_comp); + #endif -#ifdef STBI_WINDOWS_UTF8 -STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input); -#endif + #ifdef STBI_WINDOWS_UTF8 +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, + const wchar_t *input); + #endif //////////////////////////////////// // // 16-bits-per-channel interface // -STBIDEF stbi_us *stbi_load_16_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, + int *x, int *y, int *channels_in_file, + int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); -#ifndef STBI_NO_STDIO -STBIDEF stbi_us *stbi_load_16 (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); -#endif + #ifndef STBI_NO_STDIO +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + #endif -//////////////////////////////////// -// -// float-per-channel interface -// -#ifndef STBI_NO_LINEAR - STBIDEF float *stbi_loadf_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); - STBIDEF float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + //////////////////////////////////// + // + // float-per-channel interface + // + #ifndef STBI_NO_LINEAR +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *channels_in_file, + int desired_channels); +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); - #ifndef STBI_NO_STDIO - STBIDEF float *stbi_loadf (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); - STBIDEF float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); - #endif -#endif + #ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + #endif + #endif -#ifndef STBI_NO_HDR - STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); - STBIDEF void stbi_hdr_to_ldr_scale(float scale); -#endif // STBI_NO_HDR + #ifndef STBI_NO_HDR +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); +STBIDEF void stbi_hdr_to_ldr_scale(float scale); + #endif // STBI_NO_HDR -#ifndef STBI_NO_LINEAR - STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); - STBIDEF void stbi_ldr_to_hdr_scale(float scale); -#endif // STBI_NO_LINEAR + #ifndef STBI_NO_LINEAR +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); +STBIDEF void stbi_ldr_to_hdr_scale(float scale); + #endif // STBI_NO_LINEAR // stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR -STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); -STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); -#ifndef STBI_NO_STDIO -STBIDEF int stbi_is_hdr (char const *filename); -STBIDEF int stbi_is_hdr_from_file(FILE *f); -#endif // STBI_NO_STDIO - +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, + void *user); +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); + #ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr(char const *filename); +STBIDEF int stbi_is_hdr_from_file(FILE *f); + #endif // STBI_NO_STDIO // get a VERY brief reason for failure // on most compilers (and ALL modern mainstream compilers) this is threadsafe -STBIDEF const char *stbi_failure_reason (void); +STBIDEF const char *stbi_failure_reason(void); // free the loaded image -- this is just free() -STBIDEF void stbi_image_free (void *retval_from_stbi_load); +STBIDEF void stbi_image_free(void *retval_from_stbi_load); // get image dimensions & components without fully decoding -STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); -STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); -STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); -STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, void *user); - -#ifndef STBI_NO_STDIO -STBIDEF int stbi_info (char const *filename, int *x, int *y, int *comp); -STBIDEF int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); -STBIDEF int stbi_is_16_bit (char const *filename); -STBIDEF int stbi_is_16_bit_from_file(FILE *f); -#endif - +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, + int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, + void *user); + #ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit(char const *filename); +STBIDEF int stbi_is_16_bit_from_file(FILE *f); + #endif // for image formats that explicitly notate that they have premultiplied alpha, // we just return the colors as stored in the file. set this flag to force // unpremultiplication. results are undefined if the unpremultiply overflow. -STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); +STBIDEF void stbi_set_unpremultiply_on_load( + int flag_true_if_should_unpremultiply); // indicate whether we should process iphone images back to canonical format, // or just pass them through "as-is" STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); -// flip the image vertically, so the first pixel in the output array is the bottom left +// flip the image vertically, so the first pixel in the output array is the +// bottom left STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); -// as above, but only applies to images loaded on the thread that calls the function -// this function is only available if your compiler supports thread-local variables; -// calling it will fail to link if your compiler doesn't -STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip); +// as above, but only applies to images loaded on the thread that calls the +// function this function is only available if your compiler supports +// thread-local variables; calling it will fail to link if your compiler doesn't +STBIDEF void stbi_set_flip_vertically_on_load_thread( + int flag_true_if_should_flip); // ZLIB client - used by PNG, available for other purposes -STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); -STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, + int initial_size, int *outlen); +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, + int len, + int initial_size, + int *outlen, + int parse_header); STBIDEF char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); -STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen); -STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); -STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); +STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, + int *outlen); +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen); - -#ifdef __cplusplus + #ifdef __cplusplus } -#endif + #endif // // //// end header file ///////////////////////////////////////////////////// -#endif // STBI_INCLUDE_STB_IMAGE_H +#endif // STBI_INCLUDE_STB_IMAGE_H #ifdef STB_IMAGE_IMPLEMENTATION -#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ - || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ - || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ - || defined(STBI_ONLY_ZLIB) - #ifndef STBI_ONLY_JPEG - #define STBI_NO_JPEG - #endif - #ifndef STBI_ONLY_PNG - #define STBI_NO_PNG - #endif - #ifndef STBI_ONLY_BMP - #define STBI_NO_BMP - #endif - #ifndef STBI_ONLY_PSD - #define STBI_NO_PSD - #endif - #ifndef STBI_ONLY_TGA - #define STBI_NO_TGA - #endif - #ifndef STBI_ONLY_GIF - #define STBI_NO_GIF - #endif - #ifndef STBI_ONLY_HDR - #define STBI_NO_HDR - #endif - #ifndef STBI_ONLY_PIC - #define STBI_NO_PIC - #endif - #ifndef STBI_ONLY_PNM - #define STBI_NO_PNM - #endif -#endif + #if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || \ + defined(STBI_ONLY_BMP) || defined(STBI_ONLY_TGA) || \ + defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) || \ + defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || \ + defined(STBI_ONLY_PNM) || defined(STBI_ONLY_ZLIB) + #ifndef STBI_ONLY_JPEG + #define STBI_NO_JPEG + #endif + #ifndef STBI_ONLY_PNG + #define STBI_NO_PNG + #endif + #ifndef STBI_ONLY_BMP + #define STBI_NO_BMP + #endif + #ifndef STBI_ONLY_PSD + #define STBI_NO_PSD + #endif + #ifndef STBI_ONLY_TGA + #define STBI_NO_TGA + #endif + #ifndef STBI_ONLY_GIF + #define STBI_NO_GIF + #endif + #ifndef STBI_ONLY_HDR + #define STBI_NO_HDR + #endif + #ifndef STBI_ONLY_PIC + #define STBI_NO_PIC + #endif + #ifndef STBI_ONLY_PNM + #define STBI_NO_PNM + #endif + #endif -#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) -#define STBI_NO_ZLIB -#endif + #if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && \ + !defined(STBI_NO_ZLIB) + #define STBI_NO_ZLIB + #endif + #include + #include // ptrdiff_t on osx + #include + #include + #include -#include -#include // ptrdiff_t on osx -#include -#include -#include + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) + #include // ldexp, pow + #endif -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -#include // ldexp, pow -#endif + #ifndef STBI_NO_STDIO + #include + #endif -#ifndef STBI_NO_STDIO -#include -#endif + #ifndef STBI_ASSERT + #include + #define STBI_ASSERT(x) assert(x) + #endif -#ifndef STBI_ASSERT -#include -#define STBI_ASSERT(x) assert(x) -#endif + #ifdef __cplusplus + #define STBI_EXTERN extern "C" + #else + #define STBI_EXTERN extern + #endif -#ifdef __cplusplus -#define STBI_EXTERN extern "C" -#else -#define STBI_EXTERN extern -#endif + #ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif + #else + #define stbi_inline __forceinline + #endif + #ifndef STBI_NO_THREAD_LOCALS + #if defined(__cplusplus) && __cplusplus >= 201103L + #define STBI_THREAD_LOCAL thread_local + #elif defined(__GNUC__) && __GNUC__ < 5 + #define STBI_THREAD_LOCAL __thread + #elif defined(_MSC_VER) + #define STBI_THREAD_LOCAL __declspec(thread) + #elif defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && \ + !defined(__STDC_NO_THREADS__) + #define STBI_THREAD_LOCAL _Thread_local + #endif -#ifndef _MSC_VER - #ifdef __cplusplus - #define stbi_inline inline - #else - #define stbi_inline - #endif -#else - #define stbi_inline __forceinline -#endif - -#ifndef STBI_NO_THREAD_LOCALS - #if defined(__cplusplus) && __cplusplus >= 201103L - #define STBI_THREAD_LOCAL thread_local - #elif defined(__GNUC__) && __GNUC__ < 5 - #define STBI_THREAD_LOCAL __thread - #elif defined(_MSC_VER) - #define STBI_THREAD_LOCAL __declspec(thread) - #elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && !defined(__STDC_NO_THREADS__) - #define STBI_THREAD_LOCAL _Thread_local - #endif - - #ifndef STBI_THREAD_LOCAL + #ifndef STBI_THREAD_LOCAL #if defined(__GNUC__) - #define STBI_THREAD_LOCAL __thread + #define STBI_THREAD_LOCAL __thread #endif - #endif -#endif + #endif + #endif -#ifdef _MSC_VER + #ifdef _MSC_VER typedef unsigned short stbi__uint16; -typedef signed short stbi__int16; +typedef signed short stbi__int16; typedef unsigned int stbi__uint32; -typedef signed int stbi__int32; -#else -#include +typedef signed int stbi__int32; + #else + #include typedef uint16_t stbi__uint16; typedef int16_t stbi__int16; typedef uint32_t stbi__uint32; typedef int32_t stbi__int32; -#endif + #endif // should produce compiler error if size is wrong -typedef unsigned char validate_uint32[sizeof(stbi__uint32)==4 ? 1 : -1]; +typedef unsigned char validate_uint32[sizeof(stbi__uint32) == 4 ? 1 : -1]; -#ifdef _MSC_VER -#define STBI_NOTUSED(v) (void)(v) -#else -#define STBI_NOTUSED(v) (void)sizeof(v) -#endif + #ifdef _MSC_VER + #define STBI_NOTUSED(v) (void)(v) + #else + #define STBI_NOTUSED(v) (void)sizeof(v) + #endif -#ifdef _MSC_VER -#define STBI_HAS_LROTL -#endif + #ifdef _MSC_VER + #define STBI_HAS_LROTL + #endif -#ifdef STBI_HAS_LROTL - #define stbi_lrot(x,y) _lrotl(x,y) -#else - #define stbi_lrot(x,y) (((x) << (y)) | ((x) >> (32 - (y)))) -#endif + #ifdef STBI_HAS_LROTL + #define stbi_lrot(x, y) _lrotl(x, y) + #else + #define stbi_lrot(x, y) (((x) << (y)) | ((x) >> (32 - (y)))) + #endif -#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) -// ok -#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) -// ok -#else -#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." -#endif + #if defined(STBI_MALLOC) && defined(STBI_FREE) && \ + (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) + // ok + #elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && \ + !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) + // ok + #else + #error \ + "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." + #endif -#ifndef STBI_MALLOC -#define STBI_MALLOC(sz) malloc(sz) -#define STBI_REALLOC(p,newsz) realloc(p,newsz) -#define STBI_FREE(p) free(p) -#endif + #ifndef STBI_MALLOC + #define STBI_MALLOC(sz) malloc(sz) + #define STBI_REALLOC(p, newsz) realloc(p, newsz) + #define STBI_FREE(p) free(p) + #endif -#ifndef STBI_REALLOC_SIZED -#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) -#endif + #ifndef STBI_REALLOC_SIZED + #define STBI_REALLOC_SIZED(p, oldsz, newsz) STBI_REALLOC(p, newsz) + #endif -// x86/x64 detection -#if defined(__x86_64__) || defined(_M_X64) -#define STBI__X64_TARGET -#elif defined(__i386) || defined(_M_IX86) -#define STBI__X86_TARGET -#endif + // x86/x64 detection + #if defined(__x86_64__) || defined(_M_X64) + #define STBI__X64_TARGET + #elif defined(__i386) || defined(_M_IX86) + #define STBI__X86_TARGET + #endif -#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) -// gcc doesn't support sse2 intrinsics unless you compile with -msse2, -// which in turn means it gets to use SSE2 everywhere. This is unfortunate, -// but previous attempts to provide the SSE2 functions with runtime -// detection caused numerous issues. The way architecture extensions are -// exposed in GCC/Clang is, sadly, not really suited for one-file libs. -// New behavior: if compiled with -msse2, we use SSE2 without any -// detection; if not, we don't use it at all. -#define STBI_NO_SIMD -#endif + #if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && \ + !defined(STBI_NO_SIMD) + // gcc doesn't support sse2 intrinsics unless you compile with -msse2, + // which in turn means it gets to use SSE2 everywhere. This is unfortunate, + // but previous attempts to provide the SSE2 functions with runtime + // detection caused numerous issues. The way architecture extensions are + // exposed in GCC/Clang is, sadly, not really suited for one-file libs. + // New behavior: if compiled with -msse2, we use SSE2 without any + // detection; if not, we don't use it at all. + #define STBI_NO_SIMD + #endif -#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) -// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET -// -// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the -// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. -// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not -// simultaneously enabling "-mstackrealign". -// -// See https://github.com/nothings/stb/issues/81 for more information. -// -// So default to no SSE2 on 32-bit MinGW. If you've read this far and added -// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. -#define STBI_NO_SIMD -#endif + #if defined(__MINGW32__) && defined(STBI__X86_TARGET) && \ + !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) + // Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid + // STBI__X64_TARGET + // + // 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the + // Windows ABI and VC++ as well as Windows DLLs don't maintain that + // invariant. As a result, enabling SSE2 on 32-bit MinGW is dangerous when + // not simultaneously enabling "-mstackrealign". + // + // See https://github.com/nothings/stb/issues/81 for more information. + // + // So default to no SSE2 on 32-bit MinGW. If you've read this far and added + // -mstackrealign to your build settings, feel free to #define + // STBI_MINGW_ENABLE_SSE2. + #define STBI_NO_SIMD + #endif -#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) -#define STBI_SSE2 -#include + #if !defined(STBI_NO_SIMD) && \ + (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) + #define STBI_SSE2 + #include -#ifdef _MSC_VER + #ifdef _MSC_VER -#if _MSC_VER >= 1400 // not VC6 -#include // __cpuid -static int stbi__cpuid3(void) -{ - int info[4]; - __cpuid(info,1); - return info[3]; + #if _MSC_VER >= 1400 // not VC6 + #include // __cpuid +static int stbi__cpuid3(void) { + int info[4]; + __cpuid(info, 1); + return info[3]; } -#else -static int stbi__cpuid3(void) -{ - int res; - __asm { + #else +static int stbi__cpuid3(void) { + int res; + __asm { mov eax,1 cpuid mov res,edx - } - return res; + } + return res; } -#endif + #endif -#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + #define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name -#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) -static int stbi__sse2_available(void) -{ - int info3 = stbi__cpuid3(); - return ((info3 >> 26) & 1) != 0; + #if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) { + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; } -#endif + #endif -#else // assume GCC-style if not VC++ -#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + #else // assume GCC-style if not VC++ + #define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) -#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) -static int stbi__sse2_available(void) -{ - // If we're even attempting to compile this on GCC/Clang, that means - // -msse2 is on, which means the compiler is allowed to use SSE2 - // instructions at will, and so are we. - return 1; + #if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) { + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; } -#endif + #endif -#endif -#endif + #endif + #endif -// ARM NEON -#if defined(STBI_NO_SIMD) && defined(STBI_NEON) -#undef STBI_NEON -#endif + // ARM NEON + #if defined(STBI_NO_SIMD) && defined(STBI_NEON) + #undef STBI_NEON + #endif -#ifdef STBI_NEON -#include -// assume GCC or Clang on ARM targets -#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) -#endif + #ifdef STBI_NEON + #include + // assume GCC or Clang on ARM targets + #define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + #endif -#ifndef STBI_SIMD_ALIGN -#define STBI_SIMD_ALIGN(type, name) type name -#endif + #ifndef STBI_SIMD_ALIGN + #define STBI_SIMD_ALIGN(type, name) type name + #endif -#ifndef STBI_MAX_DIMENSIONS -#define STBI_MAX_DIMENSIONS (1 << 24) -#endif + #ifndef STBI_MAX_DIMENSIONS + #define STBI_MAX_DIMENSIONS (1 << 24) + #endif /////////////////////////////////////////////// // @@ -766,188 +813,182 @@ static int stbi__sse2_available(void) // stbi__context structure is our basic context used by all images, so it // contains all the IO context, plus some basic image information -typedef struct -{ - stbi__uint32 img_x, img_y; - int img_n, img_out_n; +typedef struct { + stbi__uint32 img_x, img_y; + int img_n, img_out_n; - stbi_io_callbacks io; - void *io_user_data; + stbi_io_callbacks io; + void *io_user_data; - int read_from_callbacks; - int buflen; - stbi_uc buffer_start[128]; - int callback_already_read; + int read_from_callbacks; + int buflen; + stbi_uc buffer_start[128]; + int callback_already_read; - stbi_uc *img_buffer, *img_buffer_end; - stbi_uc *img_buffer_original, *img_buffer_original_end; + stbi_uc *img_buffer, *img_buffer_end; + stbi_uc *img_buffer_original, *img_buffer_original_end; } stbi__context; - static void stbi__refill_buffer(stbi__context *s); // initialize a memory-decode context -static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) -{ - s->io.read = NULL; - s->read_from_callbacks = 0; - s->callback_already_read = 0; - s->img_buffer = s->img_buffer_original = (stbi_uc *) buffer; - s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *) buffer+len; +static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) { + s->io.read = NULL; + s->read_from_callbacks = 0; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = (stbi_uc *)buffer; + s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *)buffer + len; } // initialize a callback-based context -static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, void *user) -{ - s->io = *c; - s->io_user_data = user; - s->buflen = sizeof(s->buffer_start); - s->read_from_callbacks = 1; - s->callback_already_read = 0; - s->img_buffer = s->img_buffer_original = s->buffer_start; - stbi__refill_buffer(s); - s->img_buffer_original_end = s->img_buffer_end; +static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, + void *user) { + s->io = *c; + s->io_user_data = user; + s->buflen = sizeof(s->buffer_start); + s->read_from_callbacks = 1; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = s->buffer_start; + stbi__refill_buffer(s); + s->img_buffer_original_end = s->img_buffer_end; } -#ifndef STBI_NO_STDIO + #ifndef STBI_NO_STDIO -static int stbi__stdio_read(void *user, char *data, int size) -{ - return (int) fread(data,1,size,(FILE*) user); +static int stbi__stdio_read(void *user, char *data, int size) { + return (int)fread(data, 1, size, (FILE *)user); } -static void stbi__stdio_skip(void *user, int n) -{ - int ch; - fseek((FILE*) user, n, SEEK_CUR); - ch = fgetc((FILE*) user); /* have to read a byte to reset feof()'s flag */ - if (ch != EOF) { - ungetc(ch, (FILE *) user); /* push byte back onto stream if valid. */ - } +static void stbi__stdio_skip(void *user, int n) { + int ch; + fseek((FILE *)user, n, SEEK_CUR); + ch = fgetc((FILE *)user); /* have to read a byte to reset feof()'s flag */ + if (ch != EOF) { + ungetc(ch, (FILE *)user); /* push byte back onto stream if valid. */ + } } -static int stbi__stdio_eof(void *user) -{ - return feof((FILE*) user) || ferror((FILE *) user); +static int stbi__stdio_eof(void *user) { + return feof((FILE *)user) || ferror((FILE *)user); } -static stbi_io_callbacks stbi__stdio_callbacks = -{ - stbi__stdio_read, - stbi__stdio_skip, - stbi__stdio_eof, +static stbi_io_callbacks stbi__stdio_callbacks = { + stbi__stdio_read, + stbi__stdio_skip, + stbi__stdio_eof, }; -static void stbi__start_file(stbi__context *s, FILE *f) -{ - stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *) f); +static void stbi__start_file(stbi__context *s, FILE *f) { + stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *)f); } -//static void stop_file(stbi__context *s) { } + // static void stop_file(stbi__context *s) { } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -static void stbi__rewind(stbi__context *s) -{ - // conceptually rewind SHOULD rewind to the beginning of the stream, - // but we just rewind to the beginning of the initial buffer, because - // we only use it after doing 'test', which only ever looks at at most 92 bytes - s->img_buffer = s->img_buffer_original; - s->img_buffer_end = s->img_buffer_original_end; +static void stbi__rewind(stbi__context *s) { + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 + // bytes + s->img_buffer = s->img_buffer_original; + s->img_buffer_end = s->img_buffer_original_end; } -enum -{ - STBI_ORDER_RGB, - STBI_ORDER_BGR -}; +enum { STBI_ORDER_RGB, STBI_ORDER_BGR }; -typedef struct -{ - int bits_per_channel; - int num_channels; - int channel_order; +typedef struct { + int bits_per_channel; + int num_channels; + int channel_order; } stbi__result_info; -#ifndef STBI_NO_JPEG -static int stbi__jpeg_test(stbi__context *s); -static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context *s); +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PNG -static int stbi__png_test(stbi__context *s); -static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); -static int stbi__png_is16(stbi__context *s); -#endif + #ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context *s); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__png_is16(stbi__context *s); + #endif -#ifndef STBI_NO_BMP -static int stbi__bmp_test(stbi__context *s); -static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context *s); +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_TGA -static int stbi__tga_test(stbi__context *s); -static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s); +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PSD -static int stbi__psd_test(stbi__context *s); -static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc); -static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); -static int stbi__psd_is16(stbi__context *s); -#endif + #ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc); +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__psd_is16(stbi__context *s); + #endif -#ifndef STBI_NO_HDR -static int stbi__hdr_test(stbi__context *s); -static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_HDR +static int stbi__hdr_test(stbi__context *s); +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PIC -static int stbi__pic_test(stbi__context *s); -static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context *s); +static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_GIF -static int stbi__gif_test(stbi__context *s); -static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp); -static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context *s); +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, + int *z, int *comp, int req_comp); +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PNM -static int stbi__pnm_test(stbi__context *s); -static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context *s); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); + #endif static -#ifdef STBI_THREAD_LOCAL -STBI_THREAD_LOCAL -#endif -const char *stbi__g_failure_reason; + #ifdef STBI_THREAD_LOCAL + STBI_THREAD_LOCAL + #endif + const char *stbi__g_failure_reason; -STBIDEF const char *stbi_failure_reason(void) -{ - return stbi__g_failure_reason; +STBIDEF const char *stbi_failure_reason(void) { + return stbi__g_failure_reason; } -#ifndef STBI_NO_FAILURE_STRINGS -static int stbi__err(const char *str) -{ - stbi__g_failure_reason = str; - return 0; + #ifndef STBI_NO_FAILURE_STRINGS +static int stbi__err(const char *str) { + stbi__g_failure_reason = str; + return 0; } -#endif + #endif -static void *stbi__malloc(size_t size) -{ - return STBI_MALLOC(size); +static void *stbi__malloc(size_t size) { + return STBI_MALLOC(size); } // stb_image uses ints pervasively, including for offset calculations. @@ -962,317 +1003,334 @@ static void *stbi__malloc(size_t size) // return 1 if the sum is valid, 0 on overflow. // negative terms are considered invalid. -static int stbi__addsizes_valid(int a, int b) -{ - if (b < 0) return 0; - // now 0 <= b <= INT_MAX, hence also - // 0 <= INT_MAX - b <= INTMAX. - // And "a + b <= INT_MAX" (which might overflow) is the - // same as a <= INT_MAX - b (no overflow) - return a <= INT_MAX - b; +static int stbi__addsizes_valid(int a, int b) { + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; } // returns 1 if the product is valid, 0 on overflow. // negative factors are considered invalid. -static int stbi__mul2sizes_valid(int a, int b) -{ - if (a < 0 || b < 0) return 0; - if (b == 0) return 1; // mul-by-0 is always safe - // portable way to check for no overflows in a*b - return a <= INT_MAX/b; +static int stbi__mul2sizes_valid(int a, int b) { + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX / b; } -#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) + #if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || \ + !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) // returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow -static int stbi__mad2sizes_valid(int a, int b, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a*b, add); +static int stbi__mad2sizes_valid(int a, int b, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a * b, add); } -#endif + #endif // returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow -static int stbi__mad3sizes_valid(int a, int b, int c, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && - stbi__addsizes_valid(a*b*c, add); +static int stbi__mad3sizes_valid(int a, int b, int c, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__addsizes_valid(a * b * c, add); } -// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && - stbi__mul2sizes_valid(a*b*c, d) && stbi__addsizes_valid(a*b*c*d, add); + // returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't + // overflow + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__mul2sizes_valid(a * b * c, d) && + stbi__addsizes_valid(a * b * c * d, add); } -#endif + #endif -#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) + #if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || \ + !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) // mallocs with size overflow checking -static void *stbi__malloc_mad2(int a, int b, int add) -{ - if (!stbi__mad2sizes_valid(a, b, add)) return NULL; - return stbi__malloc(a*b + add); +static void *stbi__malloc_mad2(int a, int b, int add) { + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a * b + add); } -#endif + #endif -static void *stbi__malloc_mad3(int a, int b, int c, int add) -{ - if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; - return stbi__malloc(a*b*c + add); +static void *stbi__malloc_mad3(int a, int b, int c, int add) { + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a * b * c + add); } -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) -{ - if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; - return stbi__malloc(a*b*c*d + add); + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) { + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a * b * c * d + add); } -#endif + #endif // stbi__err - error // stbi__errpf - error returning pointer to float // stbi__errpuc - error returning pointer to unsigned char -#ifdef STBI_NO_FAILURE_STRINGS - #define stbi__err(x,y) 0 -#elif defined(STBI_FAILURE_USERMSG) - #define stbi__err(x,y) stbi__err(y) -#else - #define stbi__err(x,y) stbi__err(x) -#endif + #ifdef STBI_NO_FAILURE_STRINGS + #define stbi__err(x, y) 0 + #elif defined(STBI_FAILURE_USERMSG) + #define stbi__err(x, y) stbi__err(y) + #else + #define stbi__err(x, y) stbi__err(x) + #endif -#define stbi__errpf(x,y) ((float *)(size_t) (stbi__err(x,y)?NULL:NULL)) -#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + #define stbi__errpf(x, y) ((float *)(size_t)(stbi__err(x, y) ? NULL : NULL)) + #define stbi__errpuc(x, y) \ + ((unsigned char *)(size_t)(stbi__err(x, y) ? NULL : NULL)) -STBIDEF void stbi_image_free(void *retval_from_stbi_load) -{ - STBI_FREE(retval_from_stbi_load); +STBIDEF void stbi_image_free(void *retval_from_stbi_load) { + STBI_FREE(retval_from_stbi_load); } -#ifndef STBI_NO_LINEAR -static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); -#endif + #ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); + #endif -#ifndef STBI_NO_HDR -static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); -#endif + #ifndef STBI_NO_HDR +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); + #endif static int stbi__vertically_flip_on_load_global = 0; -STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) -{ - stbi__vertically_flip_on_load_global = flag_true_if_should_flip; +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) { + stbi__vertically_flip_on_load_global = flag_true_if_should_flip; } -#ifndef STBI_THREAD_LOCAL -#define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global -#else -static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, stbi__vertically_flip_on_load_set; + #ifndef STBI_THREAD_LOCAL + #define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global + #else +static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, + stbi__vertically_flip_on_load_set; -STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip) -{ - stbi__vertically_flip_on_load_local = flag_true_if_should_flip; - stbi__vertically_flip_on_load_set = 1; +STBIDEF void stbi_set_flip_vertically_on_load_thread( + int flag_true_if_should_flip) { + stbi__vertically_flip_on_load_local = flag_true_if_should_flip; + stbi__vertically_flip_on_load_set = 1; } -#define stbi__vertically_flip_on_load (stbi__vertically_flip_on_load_set \ - ? stbi__vertically_flip_on_load_local \ - : stbi__vertically_flip_on_load_global) -#endif // STBI_THREAD_LOCAL + #define stbi__vertically_flip_on_load \ + (stbi__vertically_flip_on_load_set \ + ? stbi__vertically_flip_on_load_local \ + : stbi__vertically_flip_on_load_global) + #endif // STBI_THREAD_LOCAL -static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) -{ - memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields - ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed - ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order - ri->num_channels = 0; +static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc) { + memset(ri, 0, + sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = + 8; // default is 8 so most paths don't have to be changed + ri->channel_order = + STBI_ORDER_RGB; // all current input & output are this, but this is here + // so we can add BGR order + ri->num_channels = 0; - #ifndef STBI_NO_JPEG - if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PNG - if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_BMP - if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_GIF - if (stbi__gif_test(s)) return stbi__gif_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PSD - if (stbi__psd_test(s)) return stbi__psd_load(s,x,y,comp,req_comp, ri, bpc); - #else - STBI_NOTUSED(bpc); - #endif - #ifndef STBI_NO_PIC - if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PNM - if (stbi__pnm_test(s)) return stbi__pnm_load(s,x,y,comp,req_comp, ri); - #endif + #ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PSD + if (stbi__psd_test(s)) + return stbi__psd_load(s, x, y, comp, req_comp, ri, bpc); + #else + STBI_NOTUSED(bpc); + #endif + #ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s, x, y, comp, req_comp, ri); + #endif - #ifndef STBI_NO_HDR - if (stbi__hdr_test(s)) { - float *hdr = stbi__hdr_load(s, x,y,comp,req_comp, ri); - return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); - } - #endif + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float *hdr = stbi__hdr_load(s, x, y, comp, req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif - #ifndef STBI_NO_TGA - // test tga last because it's a crappy test! - if (stbi__tga_test(s)) - return stbi__tga_load(s,x,y,comp,req_comp, ri); - #endif + #ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) return stbi__tga_load(s, x, y, comp, req_comp, ri); + #endif - return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); + return stbi__errpuc("unknown image type", + "Image not of any known type, or corrupt"); } -static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, int channels) -{ - int i; - int img_len = w * h * channels; - stbi_uc *reduced; +static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, + int channels) { + int i; + int img_len = w * h * channels; + stbi_uc *reduced; - reduced = (stbi_uc *) stbi__malloc(img_len); - if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + reduced = (stbi_uc *)stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); - for (i = 0; i < img_len; ++i) - reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + for (i = 0; i < img_len; ++i) + reduced[i] = (stbi_uc)((orig[i] >> 8) & + 0xFF); // top half of each byte is sufficient approx + // of 16->8 bit scaling - STBI_FREE(orig); - return reduced; + STBI_FREE(orig); + return reduced; } -static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, int channels) -{ - int i; - int img_len = w * h * channels; - stbi__uint16 *enlarged; +static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, + int channels) { + int i; + int img_len = w * h * channels; + stbi__uint16 *enlarged; - enlarged = (stbi__uint16 *) stbi__malloc(img_len*2); - if (enlarged == NULL) return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + enlarged = (stbi__uint16 *)stbi__malloc(img_len * 2); + if (enlarged == NULL) + return (stbi__uint16 *)stbi__errpuc("outofmem", "Out of memory"); - for (i = 0; i < img_len; ++i) - enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + for (i = 0; i < img_len; ++i) + enlarged[i] = (stbi__uint16)((orig[i] << 8) + + orig[i]); // replicate to high and low byte, + // maps 0->0, 255->0xffff - STBI_FREE(orig); - return enlarged; + STBI_FREE(orig); + return enlarged; } -static void stbi__vertical_flip(void *image, int w, int h, int bytes_per_pixel) -{ - int row; - size_t bytes_per_row = (size_t)w * bytes_per_pixel; - stbi_uc temp[2048]; - stbi_uc *bytes = (stbi_uc *)image; +static void stbi__vertical_flip(void *image, int w, int h, + int bytes_per_pixel) { + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc *bytes = (stbi_uc *)image; - for (row = 0; row < (h>>1); row++) { - stbi_uc *row0 = bytes + row*bytes_per_row; - stbi_uc *row1 = bytes + (h - row - 1)*bytes_per_row; - // swap row0 with row1 - size_t bytes_left = bytes_per_row; - while (bytes_left) { - size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); - memcpy(temp, row0, bytes_copy); - memcpy(row0, row1, bytes_copy); - memcpy(row1, temp, bytes_copy); - row0 += bytes_copy; - row1 += bytes_copy; - bytes_left -= bytes_copy; - } - } + for (row = 0; row < (h >> 1); row++) { + stbi_uc *row0 = bytes + row * bytes_per_row; + stbi_uc *row1 = bytes + (h - row - 1) * bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = + (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } } -#ifndef STBI_NO_GIF -static void stbi__vertical_flip_slices(void *image, int w, int h, int z, int bytes_per_pixel) -{ - int slice; - int slice_size = w * h * bytes_per_pixel; + #ifndef STBI_NO_GIF +static void stbi__vertical_flip_slices(void *image, int w, int h, int z, + int bytes_per_pixel) { + int slice; + int slice_size = w * h * bytes_per_pixel; - stbi_uc *bytes = (stbi_uc *)image; - for (slice = 0; slice < z; ++slice) { - stbi__vertical_flip(bytes, w, h, bytes_per_pixel); - bytes += slice_size; - } + stbi_uc *bytes = (stbi_uc *)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } } -#endif + #endif -static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - stbi__result_info ri; - void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); +static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, + int *y, int *comp, + int req_comp) { + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); - if (result == NULL) - return NULL; + if (result == NULL) return NULL; - // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. - STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + // it is the responsibility of the loaders to make sure we get either 8 or 16 + // bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); - if (ri.bits_per_channel != 8) { - result = stbi__convert_16_to_8((stbi__uint16 *) result, *x, *y, req_comp == 0 ? *comp : req_comp); - ri.bits_per_channel = 8; - } + if (ri.bits_per_channel != 8) { + result = stbi__convert_16_to_8((stbi__uint16 *)result, *x, *y, + req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } - // @TODO: move stbi__convert_format to here + // @TODO: move stbi__convert_format to here - if (stbi__vertically_flip_on_load) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); - } + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } - return (unsigned char *) result; + return (unsigned char *)result; } -static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - stbi__result_info ri; - void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); +static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, + int *y, int *comp, + int req_comp) { + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); - if (result == NULL) - return NULL; + if (result == NULL) return NULL; - // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. - STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + // it is the responsibility of the loaders to make sure we get either 8 or 16 + // bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); - if (ri.bits_per_channel != 16) { - result = stbi__convert_8_to_16((stbi_uc *) result, *x, *y, req_comp == 0 ? *comp : req_comp); - ri.bits_per_channel = 16; - } + if (ri.bits_per_channel != 16) { + result = stbi__convert_8_to_16((stbi_uc *)result, *x, *y, + req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } - // @TODO: move stbi__convert_format16 to here - // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to + // keep more precision - if (stbi__vertically_flip_on_load) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); - } + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } - return (stbi__uint16 *) result; + return (stbi__uint16 *)result; } -#if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) -static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, int req_comp) -{ - if (stbi__vertically_flip_on_load && result != NULL) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); - } + #if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, + int req_comp) { + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } } -#endif + #endif -#ifndef STBI_NO_STDIO + #ifndef STBI_NO_STDIO -#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) -STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide); -STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default); -#endif + #if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) +STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar( + unsigned int cp, unsigned long flags, const char *str, int cbmb, + wchar_t *widestr, int cchwide); +STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte( + unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, + char *str, int cbmb, const char *defchar, int *used_default); + #endif #if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input) { return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL); } -#endif + #endif static FILE *stbi__fopen(char const *filename, char const *mode) { @@ -1293,384 +1351,386 @@ static FILE *stbi__fopen(char const *filename, char const *mode) f = _wfopen(wFilename, wMode); #endif -#elif defined(_MSC_VER) && _MSC_VER >= 1400 - if (0 != fopen_s(&f, filename, mode)) - f=0; -#else - f = fopen(filename, mode); -#endif - return f; + #elif defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) f = 0; + #else + f = fopen(filename, mode); + #endif + return f; } - -STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - unsigned char *result; - if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); - result = stbi_load_from_file(f,x,y,comp,req_comp); - fclose(f); - return result; +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, + int req_comp) { + FILE *f = stbi__fopen(filename, "rb"); + unsigned char *result; + if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f, x, y, comp, req_comp); + fclose(f); + return result; } -STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - unsigned char *result; - stbi__context s; - stbi__start_file(&s,f); - result = stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); - if (result) { - // need to 'unget' all the characters in the IO buffer - fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); - } - return result; +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, + int req_comp) { + unsigned char *result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, -(int)(s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; } -STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - stbi__uint16 *result; - stbi__context s; - stbi__start_file(&s,f); - result = stbi__load_and_postprocess_16bit(&s,x,y,comp,req_comp); - if (result) { - // need to 'unget' all the characters in the IO buffer - fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); - } - return result; +STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, + int req_comp) { + stbi__uint16 *result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_16bit(&s, x, y, comp, req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, -(int)(s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; } -STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - stbi__uint16 *result; - if (!f) return (stbi_us *) stbi__errpuc("can't fopen", "Unable to open file"); - result = stbi_load_from_file_16(f,x,y,comp,req_comp); - fclose(f); - return result; +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, + int req_comp) { + FILE *f = stbi__fopen(filename, "rb"); + stbi__uint16 *result; + if (!f) return (stbi_us *)stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f, x, y, comp, req_comp); + fclose(f); + return result; } + #endif //! STBI_NO_STDIO -#endif //!STBI_NO_STDIO - -STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, + int *x, int *y, int *channels_in_file, + int desired_channels) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, + desired_channels); } -STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); - return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, + desired_channels); } -STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp, int req_comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); } -STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); } -#ifndef STBI_NO_GIF -STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp) -{ - unsigned char *result; - stbi__context s; - stbi__start_mem(&s,buffer,len); + #ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, + int **delays, int *x, int *y, int *z, + int *comp, int req_comp) { + unsigned char *result; + stbi__context s; + stbi__start_mem(&s, buffer, len); - result = (unsigned char*) stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); - if (stbi__vertically_flip_on_load) { - stbi__vertical_flip_slices( result, *x, *y, *z, *comp ); - } + result = + (unsigned char *)stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices(result, *x, *y, *z, *comp); + } - return result; + return result; } -#endif + #endif -#ifndef STBI_NO_LINEAR -static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - unsigned char *data; - #ifndef STBI_NO_HDR - if (stbi__hdr_test(s)) { - stbi__result_info ri; - float *hdr_data = stbi__hdr_load(s,x,y,comp,req_comp, &ri); - if (hdr_data) - stbi__float_postprocess(hdr_data,x,y,comp,req_comp); - return hdr_data; - } - #endif - data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); - if (data) - return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); - return stbi__errpf("unknown image type", "Image not of any known type, or corrupt"); + #ifndef STBI_NO_LINEAR +static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, + int req_comp) { + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + stbi__result_info ri; + float *hdr_data = stbi__hdr_load(s, x, y, comp, req_comp, &ri); + if (hdr_data) stbi__float_postprocess(hdr_data, x, y, comp, req_comp); + return hdr_data; + } + #endif + data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); + if (data) return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return stbi__errpf("unknown image type", + "Image not of any known type, or corrupt"); } -STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp, int req_comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -#ifndef STBI_NO_STDIO -STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - float *result; - FILE *f = stbi__fopen(filename, "rb"); - if (!f) return stbi__errpf("can't fopen", "Unable to open file"); - result = stbi_loadf_from_file(f,x,y,comp,req_comp); - fclose(f); - return result; + #ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, + int req_comp) { + float *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) return stbi__errpf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f, x, y, comp, req_comp); + fclose(f); + return result; } -STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_file(&s,f); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_file(&s, f); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -#endif // !STBI_NO_LINEAR + #endif // !STBI_NO_LINEAR // these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is // defined, for API simplicity; if STBI_NO_LINEAR is defined, it always // reports false! -STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) -{ - #ifndef STBI_NO_HDR - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__hdr_test(&s); - #else - STBI_NOTUSED(buffer); - STBI_NOTUSED(len); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) { + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif } -#ifndef STBI_NO_STDIO -STBIDEF int stbi_is_hdr (char const *filename) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result=0; - if (f) { - result = stbi_is_hdr_from_file(f); - fclose(f); - } - return result; + #ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr(char const *filename) { + FILE *f = stbi__fopen(filename, "rb"); + int result = 0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; } -STBIDEF int stbi_is_hdr_from_file(FILE *f) -{ - #ifndef STBI_NO_HDR - long pos = ftell(f); - int res; - stbi__context s; - stbi__start_file(&s,f); - res = stbi__hdr_test(&s); - fseek(f, pos, SEEK_SET); - return res; - #else - STBI_NOTUSED(f); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_file(FILE *f) { + #ifndef STBI_NO_HDR + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s, f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; + #else + STBI_NOTUSED(f); + return 0; + #endif } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) -{ - #ifndef STBI_NO_HDR - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__hdr_test(&s); - #else - STBI_NOTUSED(clbk); - STBI_NOTUSED(user); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, + void *user) { + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; + #endif } -#ifndef STBI_NO_LINEAR -static float stbi__l2h_gamma=2.2f, stbi__l2h_scale=1.0f; + #ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma = 2.2f, stbi__l2h_scale = 1.0f; -STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } -STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } -#endif +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { + stbi__l2h_gamma = gamma; +} +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { + stbi__l2h_scale = scale; +} + #endif -static float stbi__h2l_gamma_i=1.0f/2.2f, stbi__h2l_scale_i=1.0f; - -STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1/gamma; } -STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1/scale; } +static float stbi__h2l_gamma_i = 1.0f / 2.2f, stbi__h2l_scale_i = 1.0f; +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { + stbi__h2l_gamma_i = 1 / gamma; +} +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { + stbi__h2l_scale_i = 1 / scale; +} ////////////////////////////////////////////////////////////////////////////// // // Common code used by all image loaders // -enum -{ - STBI__SCAN_load=0, - STBI__SCAN_type, - STBI__SCAN_header -}; +enum { STBI__SCAN_load = 0, STBI__SCAN_type, STBI__SCAN_header }; -static void stbi__refill_buffer(stbi__context *s) -{ - int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); - s->callback_already_read += (int) (s->img_buffer - s->img_buffer_original); - if (n == 0) { - // at end of file, treat same as if from memory, but need to handle case - // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file - s->read_from_callbacks = 0; - s->img_buffer = s->buffer_start; - s->img_buffer_end = s->buffer_start+1; - *s->img_buffer = 0; - } else { - s->img_buffer = s->buffer_start; - s->img_buffer_end = s->buffer_start + n; - } +static void stbi__refill_buffer(stbi__context *s) { + int n = (s->io.read)(s->io_user_data, (char *)s->buffer_start, s->buflen); + s->callback_already_read += (int)(s->img_buffer - s->img_buffer_original); + if (n == 0) { + // at end of file, treat same as if from memory, but need to handle case + // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file + s->read_from_callbacks = 0; + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + 1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } } -stbi_inline static stbi_uc stbi__get8(stbi__context *s) -{ - if (s->img_buffer < s->img_buffer_end) - return *s->img_buffer++; - if (s->read_from_callbacks) { - stbi__refill_buffer(s); - return *s->img_buffer++; - } - return 0; +stbi_inline static stbi_uc stbi__get8(stbi__context *s) { + if (s->img_buffer < s->img_buffer_end) return *s->img_buffer++; + if (s->read_from_callbacks) { + stbi__refill_buffer(s); + return *s->img_buffer++; + } + return 0; } -#if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else -stbi_inline static int stbi__at_eof(stbi__context *s) -{ - if (s->io.read) { - if (!(s->io.eof)(s->io_user_data)) return 0; - // if feof() is true, check if buffer = end - // special case: we've only got the special 0 character at the end - if (s->read_from_callbacks == 0) return 1; - } + #if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && \ + defined(STBI_NO_PNM) + // nothing + #else +stbi_inline static int stbi__at_eof(stbi__context *s) { + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } - return s->img_buffer >= s->img_buffer_end; + return s->img_buffer >= s->img_buffer_end; } -#endif + #endif -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) -// nothing -#else -static void stbi__skip(stbi__context *s, int n) -{ - if (n == 0) return; // already there! - if (n < 0) { + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && \ + defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && \ + defined(STBI_NO_PIC) + // nothing + #else +static void stbi__skip(stbi__context *s, int n) { + if (n == 0) return; // already there! + if (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); return; - } - if (s->io.read) { - int blen = (int) (s->img_buffer_end - s->img_buffer); - if (blen < n) { - s->img_buffer = s->img_buffer_end; - (s->io.skip)(s->io_user_data, n - blen); - return; - } - } - s->img_buffer += n; + } + } + s->img_buffer += n; } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && defined(STBI_NO_PNM) -// nothing -#else -static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) -{ - if (s->io.read) { - int blen = (int) (s->img_buffer_end - s->img_buffer); - if (blen < n) { - int res, count; + #if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && \ + defined(STBI_NO_PNM) + // nothing + #else +static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) { + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { + int res, count; - memcpy(buffer, s->img_buffer, blen); + memcpy(buffer, s->img_buffer, blen); - count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen); - res = (count == (n-blen)); - s->img_buffer = s->img_buffer_end; - return res; - } - } + count = (s->io.read)(s->io_user_data, (char *)buffer + blen, n - blen); + res = (count == (n - blen)); + s->img_buffer = s->img_buffer_end; + return res; + } + } - if (s->img_buffer+n <= s->img_buffer_end) { - memcpy(buffer, s->img_buffer, n); - s->img_buffer += n; - return 1; - } else - return 0; + if (s->img_buffer + n <= s->img_buffer_end) { + memcpy(buffer, s->img_buffer, n); + s->img_buffer += n; + return 1; + } else + return 0; } -#endif + #endif -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) -// nothing -#else -static int stbi__get16be(stbi__context *s) -{ - int z = stbi__get8(s); - return (z << 8) + stbi__get8(s); + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && \ + defined(STBI_NO_PIC) + // nothing + #else +static int stbi__get16be(stbi__context *s) { + int z = stbi__get8(s); + return (z << 8) + stbi__get8(s); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) -// nothing -#else -static stbi__uint32 stbi__get32be(stbi__context *s) -{ - stbi__uint32 z = stbi__get16be(s); - return (z << 16) + stbi__get16be(s); + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) + // nothing + #else +static stbi__uint32 stbi__get32be(stbi__context *s) { + stbi__uint32 z = stbi__get16be(s); + return (z << 16) + stbi__get16be(s); } -#endif + #endif -#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) -// nothing -#else -static int stbi__get16le(stbi__context *s) -{ - int z = stbi__get8(s); - return z + (stbi__get8(s) << 8); + #if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) + // nothing + #else +static int stbi__get16le(stbi__context *s) { + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); } -#endif + #endif -#ifndef STBI_NO_BMP -static stbi__uint32 stbi__get32le(stbi__context *s) -{ - stbi__uint32 z = stbi__get16le(s); - return z + (stbi__get16le(s) << 16); + #ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context *s) { + stbi__uint32 z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); } -#endif + #endif -#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + #define STBI__BYTECAST(x) \ + ((stbi_uc)((x)&255)) // truncate int to byte without warnings -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && \ + defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && \ + defined(STBI_NO_PIC) && defined(STBI_NO_PNM) + // nothing + #else ////////////////////////////////////////////////////////////////////////////// // // generic converter from built-in img_n to req_comp @@ -1682,171 +1742,295 @@ static stbi__uint32 stbi__get32le(stbi__context *s) // assume data buffer is malloced, so malloc a new one and free that one // only failure mode is malloc failing -static stbi_uc stbi__compute_y(int r, int g, int b) -{ - return (stbi_uc) (((r*77) + (g*150) + (29*b)) >> 8); +static stbi_uc stbi__compute_y(int r, int g, int b) { + return (stbi_uc)(((r * 77) + (g * 150) + (29 * b)) >> 8); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else -static unsigned char *stbi__convert_format(unsigned char *data, int img_n, int req_comp, unsigned int x, unsigned int y) -{ - int i,j; - unsigned char *good; + #if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && \ + defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && \ + defined(STBI_NO_PNM) + // nothing + #else +static unsigned char *stbi__convert_format(unsigned char *data, int img_n, + int req_comp, unsigned int x, + unsigned int y) { + int i, j; + unsigned char *good; - if (req_comp == img_n) return data; - STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); - good = (unsigned char *) stbi__malloc_mad3(req_comp, x, y, 0); - if (good == NULL) { - STBI_FREE(data); - return stbi__errpuc("outofmem", "Out of memory"); - } + good = (unsigned char *)stbi__malloc_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } - for (j=0; j < (int) y; ++j) { - unsigned char *src = data + j * x * img_n ; - unsigned char *dest = good + j * x * req_comp; + for (j = 0; j < (int)y; ++j) { + unsigned char *src = data + j * x * img_n; + unsigned char *dest = good + j * x * req_comp; - #define STBI__COMBO(a,b) ((a)*8+(b)) - #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) - // convert source image with img_n components to one with req_comp components; - // avoid switch per pixel, so use switch per scanline and massive macros - switch (STBI__COMBO(img_n, req_comp)) { - STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=255; } break; - STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=255; } break; - STBI__CASE(2,1) { dest[0]=src[0]; } break; - STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; - STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=255; } break; - STBI__CASE(3,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; - STBI__CASE(3,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = 255; } break; - STBI__CASE(4,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; - STBI__CASE(4,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = src[3]; } break; - STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; - default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return stbi__errpuc("unsupported", "Unsupported format conversion"); + #define STBI__COMBO(a, b) ((a)*8 + (b)) + #define STBI__CASE(a, b) \ + case STBI__COMBO(a, b): \ + for (i = x - 1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp + // components; avoid switch per pixel, so use switch per scanline and + // massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { + dest[0] = src[0]; + dest[1] = 255; } - #undef STBI__CASE - } + break; + STBI__CASE(1, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(1, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = 255; + } + break; + STBI__CASE(2, 1) { + dest[0] = src[0]; + } + break; + STBI__CASE(2, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(2, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = src[1]; + } + break; + STBI__CASE(3, 4) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + dest[3] = 255; + } + break; + STBI__CASE(3, 1) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + } + break; + STBI__CASE(3, 2) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + dest[1] = 255; + } + break; + STBI__CASE(4, 1) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + } + break; + STBI__CASE(4, 2) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + dest[1] = src[3]; + } + break; + STBI__CASE(4, 3) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + } + break; + default: + STBI_ASSERT(0); + STBI_FREE(data); + STBI_FREE(good); + return stbi__errpuc("unsupported", "Unsupported format conversion"); + } + #undef STBI__CASE + } - STBI_FREE(data); - return good; + STBI_FREE(data); + return good; } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) -// nothing -#else -static stbi__uint16 stbi__compute_y_16(int r, int g, int b) -{ - return (stbi__uint16) (((r*77) + (g*150) + (29*b)) >> 8); + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) + // nothing + #else +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) { + return (stbi__uint16)(((r * 77) + (g * 150) + (29 * b)) >> 8); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) -// nothing -#else -static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, int req_comp, unsigned int x, unsigned int y) -{ - int i,j; - stbi__uint16 *good; + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) + // nothing + #else +static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, + int req_comp, unsigned int x, + unsigned int y) { + int i, j; + stbi__uint16 *good; - if (req_comp == img_n) return data; - STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); - good = (stbi__uint16 *) stbi__malloc(req_comp * x * y * 2); - if (good == NULL) { - STBI_FREE(data); - return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); - } + good = (stbi__uint16 *)stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16 *)stbi__errpuc("outofmem", "Out of memory"); + } - for (j=0; j < (int) y; ++j) { - stbi__uint16 *src = data + j * x * img_n ; - stbi__uint16 *dest = good + j * x * req_comp; + for (j = 0; j < (int)y; ++j) { + stbi__uint16 *src = data + j * x * img_n; + stbi__uint16 *dest = good + j * x * req_comp; - #define STBI__COMBO(a,b) ((a)*8+(b)) - #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) - // convert source image with img_n components to one with req_comp components; - // avoid switch per pixel, so use switch per scanline and massive macros - switch (STBI__COMBO(img_n, req_comp)) { - STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=0xffff; } break; - STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=0xffff; } break; - STBI__CASE(2,1) { dest[0]=src[0]; } break; - STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; - STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=0xffff; } break; - STBI__CASE(3,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; - STBI__CASE(3,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = 0xffff; } break; - STBI__CASE(4,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; - STBI__CASE(4,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = src[3]; } break; - STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; - default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return (stbi__uint16*) stbi__errpuc("unsupported", "Unsupported format conversion"); + #define STBI__COMBO(a, b) ((a)*8 + (b)) + #define STBI__CASE(a, b) \ + case STBI__COMBO(a, b): \ + for (i = x - 1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp + // components; avoid switch per pixel, so use switch per scanline and + // massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { + dest[0] = src[0]; + dest[1] = 0xffff; } - #undef STBI__CASE - } + break; + STBI__CASE(1, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(1, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = 0xffff; + } + break; + STBI__CASE(2, 1) { + dest[0] = src[0]; + } + break; + STBI__CASE(2, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(2, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = src[1]; + } + break; + STBI__CASE(3, 4) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + dest[3] = 0xffff; + } + break; + STBI__CASE(3, 1) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + } + break; + STBI__CASE(3, 2) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + dest[1] = 0xffff; + } + break; + STBI__CASE(4, 1) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + } + break; + STBI__CASE(4, 2) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + dest[1] = src[3]; + } + break; + STBI__CASE(4, 3) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + } + break; + default: + STBI_ASSERT(0); + STBI_FREE(data); + STBI_FREE(good); + return (stbi__uint16 *)stbi__errpuc("unsupported", + "Unsupported format conversion"); + } + #undef STBI__CASE + } - STBI_FREE(data); - return good; + STBI_FREE(data); + return good; } -#endif + #endif -#ifndef STBI_NO_LINEAR -static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) -{ - int i,k,n; - float *output; - if (!data) return NULL; - output = (float *) stbi__malloc_mad4(x, y, comp, sizeof(float), 0); - if (output == NULL) { STBI_FREE(data); return stbi__errpf("outofmem", "Out of memory"); } - // compute number of non-alpha components - if (comp & 1) n = comp; else n = comp-1; - for (i=0; i < x*y; ++i) { - for (k=0; k < n; ++k) { - output[i*comp + k] = (float) (pow(data[i*comp+k]/255.0f, stbi__l2h_gamma) * stbi__l2h_scale); - } - } - if (n < comp) { - for (i=0; i < x*y; ++i) { - output[i*comp + n] = data[i*comp + n]/255.0f; - } - } - STBI_FREE(data); - return output; + #ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) { + int i, k, n; + float *output; + if (!data) return NULL; + output = (float *)stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { + STBI_FREE(data); + return stbi__errpf("outofmem", "Out of memory"); + } + // compute number of non-alpha components + if (comp & 1) + n = comp; + else + n = comp - 1; + for (i = 0; i < x * y; ++i) { + for (k = 0; k < n; ++k) { + output[i * comp + k] = + (float)(pow(data[i * comp + k] / 255.0f, stbi__l2h_gamma) * + stbi__l2h_scale); + } + } + if (n < comp) { + for (i = 0; i < x * y; ++i) { + output[i * comp + n] = data[i * comp + n] / 255.0f; + } + } + STBI_FREE(data); + return output; } -#endif + #endif -#ifndef STBI_NO_HDR -#define stbi__float2int(x) ((int) (x)) -static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) -{ - int i,k,n; - stbi_uc *output; - if (!data) return NULL; - output = (stbi_uc *) stbi__malloc_mad3(x, y, comp, 0); - if (output == NULL) { STBI_FREE(data); return stbi__errpuc("outofmem", "Out of memory"); } - // compute number of non-alpha components - if (comp & 1) n = comp; else n = comp-1; - for (i=0; i < x*y; ++i) { - for (k=0; k < n; ++k) { - float z = (float) pow(data[i*comp+k]*stbi__h2l_scale_i, stbi__h2l_gamma_i) * 255 + 0.5f; - if (z < 0) z = 0; - if (z > 255) z = 255; - output[i*comp + k] = (stbi_uc) stbi__float2int(z); - } - if (k < comp) { - float z = data[i*comp+k] * 255 + 0.5f; - if (z < 0) z = 0; - if (z > 255) z = 255; - output[i*comp + k] = (stbi_uc) stbi__float2int(z); - } - } - STBI_FREE(data); - return output; + #ifndef STBI_NO_HDR + #define stbi__float2int(x) ((int)(x)) +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) { + int i, k, n; + stbi_uc *output; + if (!data) return NULL; + output = (stbi_uc *)stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } + // compute number of non-alpha components + if (comp & 1) + n = comp; + else + n = comp - 1; + for (i = 0; i < x * y; ++i) { + for (k = 0; k < n; ++k) { + float z = (float)pow(data[i * comp + k] * stbi__h2l_scale_i, + stbi__h2l_gamma_i) * + 255 + + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i * comp + k] = (stbi_uc)stbi__float2int(z); + } + if (k < comp) { + float z = data[i * comp + k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i * comp + k] = (stbi_uc)stbi__float2int(z); + } + } + STBI_FREE(data); + return output; } -#endif + #endif ////////////////////////////////////////////////////////////////////////////// // @@ -1869,2103 +2053,2259 @@ static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) // - some SIMD kernels for common paths on targets with SSE2/NEON // - uses a lot of intermediate memory, could cache poorly -#ifndef STBI_NO_JPEG + #ifndef STBI_NO_JPEG -// huffman decoding acceleration -#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + // huffman decoding acceleration + #define FAST_BITS 9 // larger handles more cases; smaller stomps less cache -typedef struct -{ - stbi_uc fast[1 << FAST_BITS]; - // weirdly, repacking this into AoS is a 10% speed loss, instead of a win - stbi__uint16 code[256]; - stbi_uc values[256]; - stbi_uc size[257]; - unsigned int maxcode[18]; - int delta[17]; // old 'firstsymbol' - old 'firstcode' +typedef struct { + stbi_uc fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + stbi__uint16 code[256]; + stbi_uc values[256]; + stbi_uc size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' } stbi__huffman; -typedef struct -{ - stbi__context *s; - stbi__huffman huff_dc[4]; - stbi__huffman huff_ac[4]; - stbi__uint16 dequant[4][64]; - stbi__int16 fast_ac[4][1 << FAST_BITS]; +typedef struct { + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; -// sizes for components, interleaved MCUs - int img_h_max, img_v_max; - int img_mcu_x, img_mcu_y; - int img_mcu_w, img_mcu_h; + // sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; -// definition of jpeg image component - struct - { - int id; - int h,v; - int tq; - int hd,ha; - int dc_pred; + // definition of jpeg image component + struct { + int id; + int h, v; + int tq; + int hd, ha; + int dc_pred; - int x,y,w2,h2; - stbi_uc *data; - void *raw_data, *raw_coeff; - stbi_uc *linebuf; - short *coeff; // progressive only - int coeff_w, coeff_h; // number of 8x8 coefficient blocks - } img_comp[4]; + int x, y, w2, h2; + stbi_uc *data; + void *raw_data, *raw_coeff; + stbi_uc *linebuf; + short *coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } img_comp[4]; - stbi__uint32 code_buffer; // jpeg entropy-coded buffer - int code_bits; // number of valid bits - unsigned char marker; // marker seen while filling entropy buffer - int nomore; // flag if we saw a marker so must stop + stbi__uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop - int progressive; - int spec_start; - int spec_end; - int succ_high; - int succ_low; - int eob_run; - int jfif; - int app14_color_transform; // Adobe APP14 tag - int rgb; + int progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; - int scan_n, order[4]; - int restart_interval, todo; + int scan_n, order[4]; + int restart_interval, todo; -// kernels - void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); - void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step); - stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs); + // kernels + void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, + const stbi_uc *pcb, const stbi_uc *pcr, int count, + int step); + stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs); } stbi__jpeg; -static int stbi__build_huffman(stbi__huffman *h, int *count) -{ - int i,j,k=0; - unsigned int code; - // build size list for each symbol (from JPEG spec) - for (i=0; i < 16; ++i) - for (j=0; j < count[i]; ++j) - h->size[k++] = (stbi_uc) (i+1); - h->size[k] = 0; +static int stbi__build_huffman(stbi__huffman *h, int *count) { + int i, j, k = 0; + unsigned int code; + // build size list for each symbol (from JPEG spec) + for (i = 0; i < 16; ++i) + for (j = 0; j < count[i]; ++j) + h->size[k++] = (stbi_uc)(i + 1); + h->size[k] = 0; - // compute actual symbols (from jpeg spec) - code = 0; - k = 0; - for(j=1; j <= 16; ++j) { - // compute delta to add to code to compute symbol id - h->delta[j] = k - code; - if (h->size[k] == j) { - while (h->size[k] == j) - h->code[k++] = (stbi__uint16) (code++); - if (code-1 >= (1u << j)) return stbi__err("bad code lengths","Corrupt JPEG"); - } - // compute largest code + 1 for this size, preshifted as needed later - h->maxcode[j] = code << (16-j); - code <<= 1; - } - h->maxcode[j] = 0xffffffff; + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for (j = 1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (stbi__uint16)(code++); + if (code - 1 >= (1u << j)) + return stbi__err("bad code lengths", "Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16 - j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; - // build non-spec acceleration table; 255 is flag for not-accelerated - memset(h->fast, 255, 1 << FAST_BITS); - for (i=0; i < k; ++i) { - int s = h->size[i]; - if (s <= FAST_BITS) { - int c = h->code[i] << (FAST_BITS-s); - int m = 1 << (FAST_BITS-s); - for (j=0; j < m; ++j) { - h->fast[c+j] = (stbi_uc) i; - } + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i = 0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS - s); + int m = 1 << (FAST_BITS - s); + for (j = 0; j < m; ++j) { + h->fast[c + j] = (stbi_uc)i; } - } - return 1; + } + } + return 1; } // build a table that decodes both magnitude and value of small ACs in // one go. -static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) -{ - int i; - for (i=0; i < (1 << FAST_BITS); ++i) { - stbi_uc fast = h->fast[i]; - fast_ac[i] = 0; - if (fast < 255) { - int rs = h->values[fast]; - int run = (rs >> 4) & 15; - int magbits = rs & 15; - int len = h->size[fast]; +static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) { + int i; + for (i = 0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; - if (magbits && len + magbits <= FAST_BITS) { - // magnitude code followed by receive_extend code - int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); - int m = 1 << (magbits - 1); - if (k < m) k += (~0U << magbits) + 1; - // if the result is small enough, we can fit it in fast_ac table - if (k >= -128 && k <= 127) - fast_ac[i] = (stbi__int16) ((k * 256) + (run * 16) + (len + magbits)); - } + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16)((k * 256) + (run * 16) + (len + magbits)); } - } + } + } } -static void stbi__grow_buffer_unsafe(stbi__jpeg *j) -{ - do { - unsigned int b = j->nomore ? 0 : stbi__get8(j->s); - if (b == 0xff) { - int c = stbi__get8(j->s); - while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes - if (c != 0) { - j->marker = (unsigned char) c; - j->nomore = 1; - return; - } +static void stbi__grow_buffer_unsafe(stbi__jpeg *j) { + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) + c = stbi__get8(j->s); // consume fill bytes + if (c != 0) { + j->marker = (unsigned char)c; + j->nomore = 1; + return; } - j->code_buffer |= b << (24 - j->code_bits); - j->code_bits += 8; - } while (j->code_bits <= 24); + } + j->code_buffer |= b << (24 - j->code_bits); + j->code_bits += 8; + } while (j->code_bits <= 24); } // (1 << n) - 1 -static const stbi__uint32 stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; +static const stbi__uint32 stbi__bmask[17] = { + 0, 1, 3, 7, 15, 31, 63, 127, 255, + 511, 1023, 2047, 4095, 8191, 16383, 32767, 65535}; // decode a jpeg huffman value from the bitstream -stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) -{ - unsigned int temp; - int c,k; +stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) { + unsigned int temp; + int c, k; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - // look at the top FAST_BITS and determine what symbol ID it is, - // if the code is <= FAST_BITS - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - k = h->fast[c]; - if (k < 255) { - int s = h->size[k]; - if (s > j->code_bits) - return -1; - j->code_buffer <<= s; - j->code_bits -= s; - return h->values[k]; - } + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } - // naive test is to shift the code_buffer down so k bits are - // valid, then test against maxcode. To speed this up, we've - // preshifted maxcode left so that it has (16-k) 0s at the - // end; in other words, regardless of the number of bits, it - // wants to be compared against something shifted to have 16; - // that way we don't need to shift inside the loop. - temp = j->code_buffer >> 16; - for (k=FAST_BITS+1 ; ; ++k) - if (temp < h->maxcode[k]) - break; - if (k == 17) { - // error! code not found - j->code_bits -= 16; - return -1; - } + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k = FAST_BITS + 1;; ++k) + if (temp < h->maxcode[k]) break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } - if (k > j->code_bits) - return -1; + if (k > j->code_bits) return -1; - // convert the huffman code to the symbol id - c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; - STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & + stbi__bmask[h->size[c]]) == h->code[c]); - // convert the id to a symbol - j->code_bits -= k; - j->code_buffer <<= k; - return h->values[c]; + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; } // bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); +stbi_inline static int stbi__extend_receive(stbi__jpeg *j, int n) { + unsigned int k; + int sgn; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); - sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB - k = stbi_lrot(j->code_buffer, n); - if (n < 0 || n >= (int) (sizeof(stbi__bmask)/sizeof(*stbi__bmask))) return 0; - j->code_buffer = k & ~stbi__bmask[n]; - k &= stbi__bmask[n]; - j->code_bits -= n; - return k + (stbi__jbias[n] & ~sgn); + sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB + k = stbi_lrot(j->code_buffer, n); + if (n < 0 || n >= (int)(sizeof(stbi__bmask) / sizeof(*stbi__bmask))) return 0; + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & ~sgn); } // get some unsigned bits -stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) -{ - unsigned int k; - if (j->code_bits < n) stbi__grow_buffer_unsafe(j); - k = stbi_lrot(j->code_buffer, n); - j->code_buffer = k & ~stbi__bmask[n]; - k &= stbi__bmask[n]; - j->code_bits -= n; - return k; +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) { + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; } -stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) -{ - unsigned int k; - if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); - k = j->code_buffer; - j->code_buffer <<= 1; - --j->code_bits; - return k & 0x80000000; +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) { + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; } // given a value that's at position X in the zigzag stream, // where does it appear in the 8x8 matrix coded as row-major? -static const stbi_uc stbi__jpeg_dezigzag[64+15] = -{ - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 32, 25, 18, 11, 4, 5, - 12, 19, 26, 33, 40, 48, 41, 34, - 27, 20, 13, 6, 7, 14, 21, 28, - 35, 42, 49, 56, 57, 50, 43, 36, - 29, 22, 15, 23, 30, 37, 44, 51, - 58, 59, 52, 45, 38, 31, 39, 46, - 53, 60, 61, 54, 47, 55, 62, 63, - // let corrupt input sample past end - 63, 63, 63, 63, 63, 63, 63, 63, - 63, 63, 63, 63, 63, 63, 63 -}; +static const stbi_uc stbi__jpeg_dezigzag[64 + 15] = { + 0, 1, 8, 16, 9, 2, 3, 10, 17, 24, 32, 25, 18, 11, 4, 5, 12, 19, 26, 33, 40, + 48, 41, 34, 27, 20, 13, 6, 7, 14, 21, 28, 35, 42, 49, 56, 57, 50, 43, 36, + 29, 22, 15, 23, 30, 37, 44, 51, 58, 59, 52, 45, 38, 31, 39, 46, 53, 60, 61, + 54, 47, 55, 62, 63, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63}; // decode one 64-entry block-- -static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, stbi__int16 *fac, int b, stbi__uint16 *dequant) -{ - int diff,dc,k; - int t; +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], + stbi__huffman *hdc, stbi__huffman *hac, + stbi__int16 *fac, int b, + stbi__uint16 *dequant) { + int diff, dc, k; + int t; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - t = stbi__jpeg_huff_decode(j, hdc); - if (t < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); - // 0 all the ac values now so we can do it 32-bits at a time - memset(data,0,64*sizeof(data[0])); + // 0 all the ac values now so we can do it 32-bits at a time + memset(data, 0, 64 * sizeof(data[0])); - diff = t ? stbi__extend_receive(j, t) : 0; - dc = j->img_comp[b].dc_pred + diff; - j->img_comp[b].dc_pred = dc; - data[0] = (short) (dc * dequant[0]); + diff = t ? stbi__extend_receive(j, t) : 0; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc * dequant[0]); - // decode AC components, see JPEG spec - k = 1; - do { - unsigned int zig; - int c,r,s; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - r = fac[c]; - if (r) { // fast-AC path - k += (r >> 4) & 15; // run - s = r & 15; // combined length - j->code_buffer <<= s; - j->code_bits -= s; - // decode into unzigzag'd location - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) ((r >> 8) * dequant[zig]); + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) * dequant[zig]); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; } else { - int rs = stbi__jpeg_huff_decode(j, hac); - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (rs != 0xf0) break; // end block - k += 16; - } else { - k += r; - // decode into unzigzag'd location - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) (stbi__extend_receive(j,s) * dequant[zig]); - } + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) * dequant[zig]); } - } while (k < 64); - return 1; + } + } while (k < 64); + return 1; } -static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], stbi__huffman *hdc, int b) -{ - int diff,dc; - int t; - if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], + stbi__huffman *hdc, int b) { + int diff, dc; + int t; + if (j->spec_end != 0) + return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - if (j->succ_high == 0) { - // first scan for DC coefficient, must be first - memset(data,0,64*sizeof(data[0])); // 0 all the ac values now - t = stbi__jpeg_huff_decode(j, hdc); - if (t == -1) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - diff = t ? stbi__extend_receive(j, t) : 0; + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data, 0, 64 * sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + if (t == -1) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + diff = t ? stbi__extend_receive(j, t) : 0; - dc = j->img_comp[b].dc_pred + diff; - j->img_comp[b].dc_pred = dc; - data[0] = (short) (dc << j->succ_low); - } else { - // refinement scan for DC coefficient - if (stbi__jpeg_get_bit(j)) - data[0] += (short) (1 << j->succ_low); - } - return 1; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc << j->succ_low); + } else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) data[0] += (short)(1 << j->succ_low); + } + return 1; } // @OPTIMIZE: store non-zigzagged during the decode passes, // and only de-zigzag when dequantizing -static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], stbi__huffman *hac, stbi__int16 *fac) -{ - int k; - if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], + stbi__huffman *hac, + stbi__int16 *fac) { + int k; + if (j->spec_start == 0) + return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - if (j->succ_high == 0) { - int shift = j->succ_low; + if (j->succ_high == 0) { + int shift = j->succ_low; - if (j->eob_run) { - --j->eob_run; - return 1; + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) << shift); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) << shift); + } } + } while (k <= j->spec_end); + } else { + // refinement scan for these AC coefficients + short bit = (short)(1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } + } else { k = j->spec_start; do { - unsigned int zig; - int c,r,s; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - r = fac[c]; - if (r) { // fast-AC path - k += (r >> 4) & 15; // run - s = r & 15; // combined length - j->code_buffer <<= s; - j->code_bits -= s; - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) ((r >> 8) << shift); - } else { - int rs = stbi__jpeg_huff_decode(j, hac); - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (r < 15) { - j->eob_run = (1 << r); - if (r) - j->eob_run += stbi__jpeg_get_bits(j, r); - --j->eob_run; - break; - } - k += 16; - } else { - k += r; - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) (stbi__extend_receive(j,s) << shift); + int r, s; + int rs = stbi__jpeg_huff_decode( + j, hac); // @OPTIMIZE see if we can use the fast path here, + // advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short *p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } else { + if (r == 0) { + *p = (short)s; + break; } - } + --r; + } + } } while (k <= j->spec_end); - } else { - // refinement scan for these AC coefficients - - short bit = (short) (1 << j->succ_low); - - if (j->eob_run) { - --j->eob_run; - for (k = j->spec_start; k <= j->spec_end; ++k) { - short *p = &data[stbi__jpeg_dezigzag[k]]; - if (*p != 0) - if (stbi__jpeg_get_bit(j)) - if ((*p & bit)==0) { - if (*p > 0) - *p += bit; - else - *p -= bit; - } - } - } else { - k = j->spec_start; - do { - int r,s; - int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (r < 15) { - j->eob_run = (1 << r) - 1; - if (r) - j->eob_run += stbi__jpeg_get_bits(j, r); - r = 64; // force end of block - } else { - // r=15 s=0 should write 16 0s, so we just do - // a run of 15 0s and then write s (which is 0), - // so we don't have to do anything special here - } - } else { - if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); - // sign bit - if (stbi__jpeg_get_bit(j)) - s = bit; - else - s = -bit; - } - - // advance by r - while (k <= j->spec_end) { - short *p = &data[stbi__jpeg_dezigzag[k++]]; - if (*p != 0) { - if (stbi__jpeg_get_bit(j)) - if ((*p & bit)==0) { - if (*p > 0) - *p += bit; - else - *p -= bit; - } - } else { - if (r == 0) { - *p = (short) s; - break; - } - --r; - } - } - } while (k <= j->spec_end); - } - } - return 1; + } + } + return 1; } // take a -128..127 value and stbi__clamp it and convert to 0..255 -stbi_inline static stbi_uc stbi__clamp(int x) -{ - // trick to use a single test to catch both cases - if ((unsigned int) x > 255) { - if (x < 0) return 0; - if (x > 255) return 255; - } - return (stbi_uc) x; +stbi_inline static stbi_uc stbi__clamp(int x) { + // trick to use a single test to catch both cases + if ((unsigned int)x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (stbi_uc)x; } -#define stbi__f2f(x) ((int) (((x) * 4096 + 0.5))) -#define stbi__fsh(x) ((x) * 4096) + #define stbi__f2f(x) ((int)(((x)*4096 + 0.5))) + #define stbi__fsh(x) ((x)*4096) -// derived from jidctint -- DCT_ISLOW -#define STBI__IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7) \ - int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \ - p2 = s2; \ - p3 = s6; \ - p1 = (p2+p3) * stbi__f2f(0.5411961f); \ - t2 = p1 + p3*stbi__f2f(-1.847759065f); \ - t3 = p1 + p2*stbi__f2f( 0.765366865f); \ - p2 = s0; \ - p3 = s4; \ - t0 = stbi__fsh(p2+p3); \ - t1 = stbi__fsh(p2-p3); \ - x0 = t0+t3; \ - x3 = t0-t3; \ - x1 = t1+t2; \ - x2 = t1-t2; \ - t0 = s7; \ - t1 = s5; \ - t2 = s3; \ - t3 = s1; \ - p3 = t0+t2; \ - p4 = t1+t3; \ - p1 = t0+t3; \ - p2 = t1+t2; \ - p5 = (p3+p4)*stbi__f2f( 1.175875602f); \ - t0 = t0*stbi__f2f( 0.298631336f); \ - t1 = t1*stbi__f2f( 2.053119869f); \ - t2 = t2*stbi__f2f( 3.072711026f); \ - t3 = t3*stbi__f2f( 1.501321110f); \ - p1 = p5 + p1*stbi__f2f(-0.899976223f); \ - p2 = p5 + p2*stbi__f2f(-2.562915447f); \ - p3 = p3*stbi__f2f(-1.961570560f); \ - p4 = p4*stbi__f2f(-0.390180644f); \ - t3 += p1+p4; \ - t2 += p2+p3; \ - t1 += p2+p4; \ - t0 += p1+p3; + // derived from jidctint -- DCT_ISLOW + #define STBI__IDCT_1D(s0, s1, s2, s3, s4, s5, s6, s7) \ + int t0, t1, t2, t3, p1, p2, p3, p4, p5, x0, x1, x2, x3; \ + p2 = s2; \ + p3 = s6; \ + p1 = (p2 + p3) * stbi__f2f(0.5411961f); \ + t2 = p1 + p3 * stbi__f2f(-1.847759065f); \ + t3 = p1 + p2 * stbi__f2f(0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = stbi__fsh(p2 + p3); \ + t1 = stbi__fsh(p2 - p3); \ + x0 = t0 + t3; \ + x3 = t0 - t3; \ + x1 = t1 + t2; \ + x2 = t1 - t2; \ + t0 = s7; \ + t1 = s5; \ + t2 = s3; \ + t3 = s1; \ + p3 = t0 + t2; \ + p4 = t1 + t3; \ + p1 = t0 + t3; \ + p2 = t1 + t2; \ + p5 = (p3 + p4) * stbi__f2f(1.175875602f); \ + t0 = t0 * stbi__f2f(0.298631336f); \ + t1 = t1 * stbi__f2f(2.053119869f); \ + t2 = t2 * stbi__f2f(3.072711026f); \ + t3 = t3 * stbi__f2f(1.501321110f); \ + p1 = p5 + p1 * stbi__f2f(-0.899976223f); \ + p2 = p5 + p2 * stbi__f2f(-2.562915447f); \ + p3 = p3 * stbi__f2f(-1.961570560f); \ + p4 = p4 * stbi__f2f(-0.390180644f); \ + t3 += p1 + p4; \ + t2 += p2 + p3; \ + t1 += p2 + p4; \ + t0 += p1 + p3; -static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) -{ - int i,val[64],*v=val; - stbi_uc *o; - short *d = data; +static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) { + int i, val[64], *v = val; + stbi_uc *o; + short *d = data; - // columns - for (i=0; i < 8; ++i,++d, ++v) { - // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing - if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 - && d[40]==0 && d[48]==0 && d[56]==0) { - // no shortcut 0 seconds - // (1|2|3|4|5|6|7)==0 0 seconds - // all separate -0.047 seconds - // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds - int dcterm = d[0]*4; - v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; - } else { - STBI__IDCT_1D(d[ 0],d[ 8],d[16],d[24],d[32],d[40],d[48],d[56]) - // constants scaled things up by 1<<12; let's bring them back - // down, but keep 2 extra bits of precision - x0 += 512; x1 += 512; x2 += 512; x3 += 512; - v[ 0] = (x0+t3) >> 10; - v[56] = (x0-t3) >> 10; - v[ 8] = (x1+t2) >> 10; - v[48] = (x1-t2) >> 10; - v[16] = (x2+t1) >> 10; - v[40] = (x2-t1) >> 10; - v[24] = (x3+t0) >> 10; - v[32] = (x3-t0) >> 10; - } - } + // columns + for (i = 0; i < 8; ++i, ++d, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[8] == 0 && d[16] == 0 && d[24] == 0 && d[32] == 0 && d[40] == 0 && + d[48] == 0 && d[56] == 0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0] * 4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[0], d[8], d[16], d[24], d[32], d[40], d[48], d[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; + x1 += 512; + x2 += 512; + x3 += 512; + v[0] = (x0 + t3) >> 10; + v[56] = (x0 - t3) >> 10; + v[8] = (x1 + t2) >> 10; + v[48] = (x1 - t2) >> 10; + v[16] = (x2 + t1) >> 10; + v[40] = (x2 - t1) >> 10; + v[24] = (x3 + t0) >> 10; + v[32] = (x3 - t0) >> 10; + } + } - for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { - // no fast case since the first 1D IDCT spread components out - STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) - // constants scaled things up by 1<<12, plus we had 1<<2 from first - // loop, plus horizontal and vertical each scale by sqrt(8) so together - // we've got an extra 1<<3, so 1<<17 total we need to remove. - // so we want to round that, which means adding 0.5 * 1<<17, - // aka 65536. Also, we'll end up with -128 to 127 that we want - // to encode as 0..255 by adding 128, so we'll add that before the shift - x0 += 65536 + (128<<17); - x1 += 65536 + (128<<17); - x2 += 65536 + (128<<17); - x3 += 65536 + (128<<17); - // tried computing the shifts into temps, or'ing the temps to see - // if any were out of range, but that was slower - o[0] = stbi__clamp((x0+t3) >> 17); - o[7] = stbi__clamp((x0-t3) >> 17); - o[1] = stbi__clamp((x1+t2) >> 17); - o[6] = stbi__clamp((x1-t2) >> 17); - o[2] = stbi__clamp((x2+t1) >> 17); - o[5] = stbi__clamp((x2-t1) >> 17); - o[3] = stbi__clamp((x3+t0) >> 17); - o[4] = stbi__clamp((x3-t0) >> 17); - } + for (i = 0, v = val, o = out; i < 8; ++i, v += 8, o += out_stride) { + // no fast case since the first 1D IDCT spread components out + STBI__IDCT_1D(v[0], v[1], v[2], v[3], v[4], v[5], v[6], v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128 << 17); + x1 += 65536 + (128 << 17); + x2 += 65536 + (128 << 17); + x3 += 65536 + (128 << 17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = stbi__clamp((x0 + t3) >> 17); + o[7] = stbi__clamp((x0 - t3) >> 17); + o[1] = stbi__clamp((x1 + t2) >> 17); + o[6] = stbi__clamp((x1 - t2) >> 17); + o[2] = stbi__clamp((x2 + t1) >> 17); + o[5] = stbi__clamp((x2 - t1) >> 17); + o[3] = stbi__clamp((x3 + t0) >> 17); + o[4] = stbi__clamp((x3 - t0) >> 17); + } } -#ifdef STBI_SSE2 + #ifdef STBI_SSE2 // sse2 integer IDCT. not the fastest possible implementation but it // produces bit-identical results to the generic C version so it's // fully "transparent". -static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) -{ - // This is constructed to match our regular (generic) integer IDCT exactly. - __m128i row0, row1, row2, row3, row4, row5, row6, row7; - __m128i tmp; +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) { + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; - // dot product constant: even elems=x, odd elems=y - #define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + // dot product constant: even elems=x, odd elems=y + #define dct_const(x, y) \ + _mm_setr_epi16((x), (y), (x), (y), (x), (y), (x), (y)) - // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) - // out(1) = c1[even]*x + c1[odd]*y - #define dct_rot(out0,out1, x,y,c0,c1) \ - __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ - __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ - __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ - __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ - __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ - __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) + // out(1) = c1[even]*x + c1[odd]*y + #define dct_rot(out0, out1, x, y, c0, c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x), (y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x), (y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) - // out = in << 12 (in 16-bit, out 32-bit) - #define dct_widen(out, in) \ - __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ - __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + // out = in << 12 (in 16-bit, out 32-bit) + #define dct_widen(out, in) \ + __m128i out##_l = \ + _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = \ + _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) - // wide add - #define dct_wadd(out, a, b) \ - __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ - __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + // wide add + #define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) - // wide sub - #define dct_wsub(out, a, b) \ - __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ - __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + // wide sub + #define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) - // butterfly a/b, add bias, then shift by "s" and pack - #define dct_bfly32o(out0, out1, a,b,bias,s) \ - { \ - __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ - __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ - dct_wadd(sum, abiased, b); \ - dct_wsub(dif, abiased, b); \ - out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ - out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ - } + // butterfly a/b, add bias, then shift by "s" and pack + #define dct_bfly32o(out0, out1, a, b, bias, s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), \ + _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), \ + _mm_srai_epi32(dif_h, s)); \ + } - // 8-bit interleave step (for transposes) - #define dct_interleave8(a, b) \ - tmp = a; \ - a = _mm_unpacklo_epi8(a, b); \ - b = _mm_unpackhi_epi8(tmp, b) + // 8-bit interleave step (for transposes) + #define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) - // 16-bit interleave step (for transposes) - #define dct_interleave16(a, b) \ - tmp = a; \ - a = _mm_unpacklo_epi16(a, b); \ - b = _mm_unpackhi_epi16(tmp, b) + // 16-bit interleave step (for transposes) + #define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) - #define dct_pass(bias,shift) \ - { \ - /* even part */ \ - dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ - __m128i sum04 = _mm_add_epi16(row0, row4); \ - __m128i dif04 = _mm_sub_epi16(row0, row4); \ - dct_widen(t0e, sum04); \ - dct_widen(t1e, dif04); \ - dct_wadd(x0, t0e, t3e); \ - dct_wsub(x3, t0e, t3e); \ - dct_wadd(x1, t1e, t2e); \ - dct_wsub(x2, t1e, t2e); \ - /* odd part */ \ - dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ - dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ - __m128i sum17 = _mm_add_epi16(row1, row7); \ - __m128i sum35 = _mm_add_epi16(row3, row5); \ - dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ - dct_wadd(x4, y0o, y4o); \ - dct_wadd(x5, y1o, y5o); \ - dct_wadd(x6, y2o, y5o); \ - dct_wadd(x7, y3o, y4o); \ - dct_bfly32o(row0,row7, x0,x7,bias,shift); \ - dct_bfly32o(row1,row6, x1,x6,bias,shift); \ - dct_bfly32o(row2,row5, x2,x5,bias,shift); \ - dct_bfly32o(row3,row4, x3,x4,bias,shift); \ - } + #define dct_pass(bias, shift) \ + { \ + /* even part */ \ + dct_rot(t2e, t3e, row2, row6, rot0_0, rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o, y2o, row7, row3, rot2_0, rot2_1); \ + dct_rot(y1o, y3o, row5, row1, rot3_0, rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o, y5o, sum17, sum35, rot1_0, rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0, row7, x0, x7, bias, shift); \ + dct_bfly32o(row1, row6, x1, x6, bias, shift); \ + dct_bfly32o(row2, row5, x2, x5, bias, shift); \ + dct_bfly32o(row3, row4, x3, x4, bias, shift); \ + } - __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); - __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f( 0.765366865f), stbi__f2f(0.5411961f)); - __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); - __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); - __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f( 0.298631336f), stbi__f2f(-1.961570560f)); - __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f( 3.072711026f)); - __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f( 2.053119869f), stbi__f2f(-0.390180644f)); - __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f( 1.501321110f)); + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), + stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f(0.765366865f), + stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), + stbi__f2f(1.175875602f)); + __m128i rot1_1 = + dct_const(stbi__f2f(1.175875602f), + stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f(0.298631336f), + stbi__f2f(-1.961570560f)); + __m128i rot2_1 = + dct_const(stbi__f2f(-1.961570560f), + stbi__f2f(-1.961570560f) + stbi__f2f(3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f(2.053119869f), + stbi__f2f(-0.390180644f)); + __m128i rot3_1 = + dct_const(stbi__f2f(-0.390180644f), + stbi__f2f(-0.390180644f) + stbi__f2f(1.501321110f)); - // rounding biases in column/row passes, see stbi__idct_block for explanation. - __m128i bias_0 = _mm_set1_epi32(512); - __m128i bias_1 = _mm_set1_epi32(65536 + (128<<17)); + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128 << 17)); - // load - row0 = _mm_load_si128((const __m128i *) (data + 0*8)); - row1 = _mm_load_si128((const __m128i *) (data + 1*8)); - row2 = _mm_load_si128((const __m128i *) (data + 2*8)); - row3 = _mm_load_si128((const __m128i *) (data + 3*8)); - row4 = _mm_load_si128((const __m128i *) (data + 4*8)); - row5 = _mm_load_si128((const __m128i *) (data + 5*8)); - row6 = _mm_load_si128((const __m128i *) (data + 6*8)); - row7 = _mm_load_si128((const __m128i *) (data + 7*8)); + // load + row0 = _mm_load_si128((const __m128i *)(data + 0 * 8)); + row1 = _mm_load_si128((const __m128i *)(data + 1 * 8)); + row2 = _mm_load_si128((const __m128i *)(data + 2 * 8)); + row3 = _mm_load_si128((const __m128i *)(data + 3 * 8)); + row4 = _mm_load_si128((const __m128i *)(data + 4 * 8)); + row5 = _mm_load_si128((const __m128i *)(data + 5 * 8)); + row6 = _mm_load_si128((const __m128i *)(data + 6 * 8)); + row7 = _mm_load_si128((const __m128i *)(data + 7 * 8)); - // column pass - dct_pass(bias_0, 10); + // column pass + dct_pass(bias_0, 10); - { - // 16bit 8x8 transpose pass 1 - dct_interleave16(row0, row4); - dct_interleave16(row1, row5); - dct_interleave16(row2, row6); - dct_interleave16(row3, row7); + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); - // transpose pass 2 - dct_interleave16(row0, row2); - dct_interleave16(row1, row3); - dct_interleave16(row4, row6); - dct_interleave16(row5, row7); + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); - // transpose pass 3 - dct_interleave16(row0, row1); - dct_interleave16(row2, row3); - dct_interleave16(row4, row5); - dct_interleave16(row6, row7); - } + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } - // row pass - dct_pass(bias_1, 17); + // row pass + dct_pass(bias_1, 17); - { - // pack - __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 - __m128i p1 = _mm_packus_epi16(row2, row3); - __m128i p2 = _mm_packus_epi16(row4, row5); - __m128i p3 = _mm_packus_epi16(row6, row7); + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); - // 8bit 8x8 transpose pass 1 - dct_interleave8(p0, p2); // a0e0a1e1... - dct_interleave8(p1, p3); // c0g0c1g1... + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... - // transpose pass 2 - dct_interleave8(p0, p1); // a0c0e0g0... - dct_interleave8(p2, p3); // b0d0f0h0... + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... - // transpose pass 3 - dct_interleave8(p0, p2); // a0b0c0d0... - dct_interleave8(p1, p3); // a4b4c4d4... + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... - // store - _mm_storel_epi64((__m128i *) out, p0); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p2); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p1); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p3); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p3, 0x4e)); - } + // store + _mm_storel_epi64((__m128i *)out, p0); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p0, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p2); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p2, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p1); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p1, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p3); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p3, 0x4e)); + } -#undef dct_const -#undef dct_rot -#undef dct_widen -#undef dct_wadd -#undef dct_wsub -#undef dct_bfly32o -#undef dct_interleave8 -#undef dct_interleave16 -#undef dct_pass + #undef dct_const + #undef dct_rot + #undef dct_widen + #undef dct_wadd + #undef dct_wsub + #undef dct_bfly32o + #undef dct_interleave8 + #undef dct_interleave16 + #undef dct_pass } -#endif // STBI_SSE2 + #endif // STBI_SSE2 -#ifdef STBI_NEON + #ifdef STBI_NEON // NEON integer IDCT. should produce bit-identical // results to the generic C version. -static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) -{ - int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) { + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; - int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); - int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); - int16x4_t rot0_2 = vdup_n_s16(stbi__f2f( 0.765366865f)); - int16x4_t rot1_0 = vdup_n_s16(stbi__f2f( 1.175875602f)); - int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); - int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); - int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); - int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); - int16x4_t rot3_0 = vdup_n_s16(stbi__f2f( 0.298631336f)); - int16x4_t rot3_1 = vdup_n_s16(stbi__f2f( 2.053119869f)); - int16x4_t rot3_2 = vdup_n_s16(stbi__f2f( 3.072711026f)); - int16x4_t rot3_3 = vdup_n_s16(stbi__f2f( 1.501321110f)); + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f(0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f(1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f(0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f(2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f(3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f(1.501321110f)); -#define dct_long_mul(out, inq, coeff) \ - int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ - int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + #define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) -#define dct_long_mac(out, acc, inq, coeff) \ - int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ - int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + #define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) -#define dct_widen(out, inq) \ - int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ - int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + #define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) -// wide add -#define dct_wadd(out, a, b) \ - int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ - int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + // wide add + #define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) -// wide sub -#define dct_wsub(out, a, b) \ - int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ - int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + // wide sub + #define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) -// butterfly a/b, then shift using "shiftop" by "s" and pack -#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ - { \ - dct_wadd(sum, a, b); \ - dct_wsub(dif, a, b); \ - out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ - out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ - } + // butterfly a/b, then shift using "shiftop" by "s" and pack + #define dct_bfly32o(out0, out1, a, b, shiftop, s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } -#define dct_pass(shiftop, shift) \ - { \ - /* even part */ \ - int16x8_t sum26 = vaddq_s16(row2, row6); \ - dct_long_mul(p1e, sum26, rot0_0); \ - dct_long_mac(t2e, p1e, row6, rot0_1); \ - dct_long_mac(t3e, p1e, row2, rot0_2); \ - int16x8_t sum04 = vaddq_s16(row0, row4); \ - int16x8_t dif04 = vsubq_s16(row0, row4); \ - dct_widen(t0e, sum04); \ - dct_widen(t1e, dif04); \ - dct_wadd(x0, t0e, t3e); \ - dct_wsub(x3, t0e, t3e); \ - dct_wadd(x1, t1e, t2e); \ - dct_wsub(x2, t1e, t2e); \ - /* odd part */ \ - int16x8_t sum15 = vaddq_s16(row1, row5); \ - int16x8_t sum17 = vaddq_s16(row1, row7); \ - int16x8_t sum35 = vaddq_s16(row3, row5); \ - int16x8_t sum37 = vaddq_s16(row3, row7); \ - int16x8_t sumodd = vaddq_s16(sum17, sum35); \ - dct_long_mul(p5o, sumodd, rot1_0); \ - dct_long_mac(p1o, p5o, sum17, rot1_1); \ - dct_long_mac(p2o, p5o, sum35, rot1_2); \ - dct_long_mul(p3o, sum37, rot2_0); \ - dct_long_mul(p4o, sum15, rot2_1); \ - dct_wadd(sump13o, p1o, p3o); \ - dct_wadd(sump24o, p2o, p4o); \ - dct_wadd(sump23o, p2o, p3o); \ - dct_wadd(sump14o, p1o, p4o); \ - dct_long_mac(x4, sump13o, row7, rot3_0); \ - dct_long_mac(x5, sump24o, row5, rot3_1); \ - dct_long_mac(x6, sump23o, row3, rot3_2); \ - dct_long_mac(x7, sump14o, row1, rot3_3); \ - dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ - dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ - dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ - dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ - } + #define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0, row7, x0, x7, shiftop, shift); \ + dct_bfly32o(row1, row6, x1, x6, shiftop, shift); \ + dct_bfly32o(row2, row5, x2, x5, shiftop, shift); \ + dct_bfly32o(row3, row4, x3, x4, shiftop, shift); \ + } - // load - row0 = vld1q_s16(data + 0*8); - row1 = vld1q_s16(data + 1*8); - row2 = vld1q_s16(data + 2*8); - row3 = vld1q_s16(data + 3*8); - row4 = vld1q_s16(data + 4*8); - row5 = vld1q_s16(data + 5*8); - row6 = vld1q_s16(data + 6*8); - row7 = vld1q_s16(data + 7*8); + // load + row0 = vld1q_s16(data + 0 * 8); + row1 = vld1q_s16(data + 1 * 8); + row2 = vld1q_s16(data + 2 * 8); + row3 = vld1q_s16(data + 3 * 8); + row4 = vld1q_s16(data + 4 * 8); + row5 = vld1q_s16(data + 5 * 8); + row6 = vld1q_s16(data + 6 * 8); + row7 = vld1q_s16(data + 7 * 8); - // add DC bias - row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); - // column pass - dct_pass(vrshrn_n_s32, 10); + // column pass + dct_pass(vrshrn_n_s32, 10); - // 16bit 8x8 transpose - { -// these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. -// whether compilers actually get this is another story, sadly. -#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } -#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } -#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + // 16bit 8x8 transpose + { + // these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. + // whether compilers actually get this is another story, sadly. + #define dct_trn16(x, y) \ + { \ + int16x8x2_t t = vtrnq_s16(x, y); \ + x = t.val[0]; \ + y = t.val[1]; \ + } + #define dct_trn32(x, y) \ + { \ + int32x4x2_t t = \ + vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); \ + x = vreinterpretq_s16_s32(t.val[0]); \ + y = vreinterpretq_s16_s32(t.val[1]); \ + } + #define dct_trn64(x, y) \ + { \ + int16x8_t x0 = x; \ + int16x8_t y0 = y; \ + x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); \ + y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); \ + } - // pass 1 - dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 - dct_trn16(row2, row3); - dct_trn16(row4, row5); - dct_trn16(row6, row7); + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); - // pass 2 - dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 - dct_trn32(row1, row3); - dct_trn32(row4, row6); - dct_trn32(row5, row7); + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); - // pass 3 - dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 - dct_trn64(row1, row5); - dct_trn64(row2, row6); - dct_trn64(row3, row7); + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); -#undef dct_trn16 -#undef dct_trn32 -#undef dct_trn64 - } + #undef dct_trn16 + #undef dct_trn32 + #undef dct_trn64 + } - // row pass - // vrshrn_n_s32 only supports shifts up to 16, we need - // 17. so do a non-rounding shift of 16 first then follow - // up with a rounding shift by 1. - dct_pass(vshrn_n_s32, 16); + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); - { - // pack and round - uint8x8_t p0 = vqrshrun_n_s16(row0, 1); - uint8x8_t p1 = vqrshrun_n_s16(row1, 1); - uint8x8_t p2 = vqrshrun_n_s16(row2, 1); - uint8x8_t p3 = vqrshrun_n_s16(row3, 1); - uint8x8_t p4 = vqrshrun_n_s16(row4, 1); - uint8x8_t p5 = vqrshrun_n_s16(row5, 1); - uint8x8_t p6 = vqrshrun_n_s16(row6, 1); - uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); - // again, these can translate into one instruction, but often don't. -#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } -#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } -#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + // again, these can translate into one instruction, but often don't. + #define dct_trn8_8(x, y) \ + { \ + uint8x8x2_t t = vtrn_u8(x, y); \ + x = t.val[0]; \ + y = t.val[1]; \ + } + #define dct_trn8_16(x, y) \ + { \ + uint16x4x2_t t = \ + vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); \ + x = vreinterpret_u8_u16(t.val[0]); \ + y = vreinterpret_u8_u16(t.val[1]); \ + } + #define dct_trn8_32(x, y) \ + { \ + uint32x2x2_t t = \ + vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); \ + x = vreinterpret_u8_u32(t.val[0]); \ + y = vreinterpret_u8_u32(t.val[1]); \ + } - // sadly can't use interleaved stores here since we only write - // 8 bytes to each scan line! + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! - // 8x8 8-bit transpose pass 1 - dct_trn8_8(p0, p1); - dct_trn8_8(p2, p3); - dct_trn8_8(p4, p5); - dct_trn8_8(p6, p7); + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); - // pass 2 - dct_trn8_16(p0, p2); - dct_trn8_16(p1, p3); - dct_trn8_16(p4, p6); - dct_trn8_16(p5, p7); + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); - // pass 3 - dct_trn8_32(p0, p4); - dct_trn8_32(p1, p5); - dct_trn8_32(p2, p6); - dct_trn8_32(p3, p7); + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); - // store - vst1_u8(out, p0); out += out_stride; - vst1_u8(out, p1); out += out_stride; - vst1_u8(out, p2); out += out_stride; - vst1_u8(out, p3); out += out_stride; - vst1_u8(out, p4); out += out_stride; - vst1_u8(out, p5); out += out_stride; - vst1_u8(out, p6); out += out_stride; - vst1_u8(out, p7); + // store + vst1_u8(out, p0); + out += out_stride; + vst1_u8(out, p1); + out += out_stride; + vst1_u8(out, p2); + out += out_stride; + vst1_u8(out, p3); + out += out_stride; + vst1_u8(out, p4); + out += out_stride; + vst1_u8(out, p5); + out += out_stride; + vst1_u8(out, p6); + out += out_stride; + vst1_u8(out, p7); -#undef dct_trn8_8 -#undef dct_trn8_16 -#undef dct_trn8_32 - } + #undef dct_trn8_8 + #undef dct_trn8_16 + #undef dct_trn8_32 + } -#undef dct_long_mul -#undef dct_long_mac -#undef dct_widen -#undef dct_wadd -#undef dct_wsub -#undef dct_bfly32o -#undef dct_pass + #undef dct_long_mul + #undef dct_long_mac + #undef dct_widen + #undef dct_wadd + #undef dct_wsub + #undef dct_bfly32o + #undef dct_pass } -#endif // STBI_NEON + #endif // STBI_NEON -#define STBI__MARKER_none 0xff + #define STBI__MARKER_none 0xff // if there's a pending marker from the entropy stream, return that // otherwise, fetch from the stream and get a marker. if there's no // marker, return 0xff, which is never a valid marker value -static stbi_uc stbi__get_marker(stbi__jpeg *j) -{ - stbi_uc x; - if (j->marker != STBI__MARKER_none) { x = j->marker; j->marker = STBI__MARKER_none; return x; } - x = stbi__get8(j->s); - if (x != 0xff) return STBI__MARKER_none; - while (x == 0xff) - x = stbi__get8(j->s); // consume repeated 0xff fill bytes - return x; +static stbi_uc stbi__get_marker(stbi__jpeg *j) { + stbi_uc x; + if (j->marker != STBI__MARKER_none) { + x = j->marker; + j->marker = STBI__MARKER_none; + return x; + } + x = stbi__get8(j->s); + if (x != 0xff) return STBI__MARKER_none; + while (x == 0xff) + x = stbi__get8(j->s); // consume repeated 0xff fill bytes + return x; } -// in each scan, we'll have scan_n components, and the order -// of the components is specified by order[] -#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + // in each scan, we'll have scan_n components, and the order + // of the components is specified by order[] + #define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) // after a restart interval, stbi__jpeg_reset the entropy decoder and // the dc prediction -static void stbi__jpeg_reset(stbi__jpeg *j) -{ - j->code_bits = 0; - j->code_buffer = 0; - j->nomore = 0; - j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = j->img_comp[3].dc_pred = 0; - j->marker = STBI__MARKER_none; - j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; - j->eob_run = 0; - // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, - // since we don't even allow 1<<30 pixels +static void stbi__jpeg_reset(stbi__jpeg *j) { + j->code_bits = 0; + j->code_buffer = 0; + j->nomore = 0; + j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = + j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels } -static int stbi__parse_entropy_coded_data(stbi__jpeg *z) -{ - stbi__jpeg_reset(z); - if (!z->progressive) { - if (z->scan_n == 1) { - int i,j; - STBI_SIMD_ALIGN(short, data[64]); - int n = z->order[0]; - // non-interleaved data, we just need to process one block at a time, - // in trivial scanline order - // number of blocks to do just depends on how many actual "pixels" this - // component has, independent of interleaved MCU blocking and such - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); - // every data block is an MCU, so countdown the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - // if it's NOT a restart, then just bail, so we get corrupt data - // rather than no data - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } else { // interleaved - int i,j,k,x,y; - STBI_SIMD_ALIGN(short, data[64]); - for (j=0; j < z->img_mcu_y; ++j) { - for (i=0; i < z->img_mcu_x; ++i) { - // scan an interleaved mcu... process scan_n components in order - for (k=0; k < z->scan_n; ++k) { - int n = z->order[k]; - // scan out an mcu's worth of this component; that's just determined - // by the basic H and V specified for the component - for (y=0; y < z->img_comp[n].v; ++y) { - for (x=0; x < z->img_comp[n].h; ++x) { - int x2 = (i*z->img_comp[n].h + x)*8; - int y2 = (j*z->img_comp[n].v + y)*8; - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data); - } - } - } - // after all interleaved components, that's an interleaved MCU, - // so now count down the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) { + stbi__jpeg_reset(z); + if (!z->progressive) { + if (z->scan_n == 1) { + int i, j; + STBI_SIMD_ALIGN(short, data[64]); + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc + z->img_comp[n].hd, + z->huff_ac + ha, z->fast_ac[ha], n, + z->dequant[z->img_comp[n].tq])) + return 0; + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, + z->img_comp[n].w2, data); + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } } - } else { - if (z->scan_n == 1) { - int i,j; - int n = z->order[0]; - // non-interleaved data, we just need to process one block at a time, - // in trivial scanline order - // number of blocks to do just depends on how many actual "pixels" this - // component has, independent of interleaved MCU blocking and such - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); - if (z->spec_start == 0) { - if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) - return 0; - } else { - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) - return 0; - } - // every data block is an MCU, so countdown the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } else { // interleaved - int i,j,k,x,y; - for (j=0; j < z->img_mcu_y; ++j) { - for (i=0; i < z->img_mcu_x; ++i) { - // scan an interleaved mcu... process scan_n components in order - for (k=0; k < z->scan_n; ++k) { - int n = z->order[k]; - // scan out an mcu's worth of this component; that's just determined - // by the basic H and V specified for the component - for (y=0; y < z->img_comp[n].v; ++y) { - for (x=0; x < z->img_comp[n].h; ++x) { - int x2 = (i*z->img_comp[n].h + x); - int y2 = (j*z->img_comp[n].v + y); - short *data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); - if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) - return 0; - } - } - } - // after all interleaved components, that's an interleaved MCU, - // so now count down the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } - } -} - -static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) -{ - int i; - for (i=0; i < 64; ++i) - data[i] *= dequant[i]; -} - -static void stbi__jpeg_finish(stbi__jpeg *z) -{ - if (z->progressive) { - // dequantize and idct the data - int i,j,n; - for (n=0; n < z->s->img_n; ++n) { - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); - stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); - } - } - } - } -} - -static int stbi__process_marker(stbi__jpeg *z, int m) -{ - int L; - switch (m) { - case STBI__MARKER_none: // no marker found - return stbi__err("expected marker","Corrupt JPEG"); - - case 0xDD: // DRI - specify restart interval - if (stbi__get16be(z->s) != 4) return stbi__err("bad DRI len","Corrupt JPEG"); - z->restart_interval = stbi__get16be(z->s); - return 1; - - case 0xDB: // DQT - define quantization table - L = stbi__get16be(z->s)-2; - while (L > 0) { - int q = stbi__get8(z->s); - int p = q >> 4, sixteen = (p != 0); - int t = q & 15,i; - if (p != 0 && p != 1) return stbi__err("bad DQT type","Corrupt JPEG"); - if (t > 3) return stbi__err("bad DQT table","Corrupt JPEG"); - - for (i=0; i < 64; ++i) - z->dequant[t][stbi__jpeg_dezigzag[i]] = (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); - L -= (sixteen ? 129 : 65); - } - return L==0; - - case 0xC4: // DHT - define huffman table - L = stbi__get16be(z->s)-2; - while (L > 0) { - stbi_uc *v; - int sizes[16],i,n=0; - int q = stbi__get8(z->s); - int tc = q >> 4; - int th = q & 15; - if (tc > 1 || th > 3) return stbi__err("bad DHT header","Corrupt JPEG"); - for (i=0; i < 16; ++i) { - sizes[i] = stbi__get8(z->s); - n += sizes[i]; - } - L -= 17; - if (tc == 0) { - if (!stbi__build_huffman(z->huff_dc+th, sizes)) return 0; - v = z->huff_dc[th].values; - } else { - if (!stbi__build_huffman(z->huff_ac+th, sizes)) return 0; - v = z->huff_ac[th].values; - } - for (i=0; i < n; ++i) - v[i] = stbi__get8(z->s); - if (tc != 0) - stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); - L -= n; - } - return L==0; - } - - // check for comment block or APP blocks - if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { - L = stbi__get16be(z->s); - if (L < 2) { - if (m == 0xFE) - return stbi__err("bad COM len","Corrupt JPEG"); - else - return stbi__err("bad APP len","Corrupt JPEG"); - } - L -= 2; - - if (m == 0xE0 && L >= 5) { // JFIF APP0 segment - static const unsigned char tag[5] = {'J','F','I','F','\0'}; - int ok = 1; - int i; - for (i=0; i < 5; ++i) - if (stbi__get8(z->s) != tag[i]) - ok = 0; - L -= 5; - if (ok) - z->jfif = 1; - } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment - static const unsigned char tag[6] = {'A','d','o','b','e','\0'}; - int ok = 1; - int i; - for (i=0; i < 6; ++i) - if (stbi__get8(z->s) != tag[i]) - ok = 0; - L -= 6; - if (ok) { - stbi__get8(z->s); // version - stbi__get16be(z->s); // flags0 - stbi__get16be(z->s); // flags1 - z->app14_color_transform = stbi__get8(z->s); // color transform - L -= 6; - } - } - - stbi__skip(z->s, L); return 1; - } + } else { // interleaved + int i, j, k, x, y; + STBI_SIMD_ALIGN(short, data[64]); + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k = 0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y = 0; y < z->img_comp[n].v; ++y) { + for (x = 0; x < z->img_comp[n].h; ++x) { + int x2 = (i * z->img_comp[n].h + x) * 8; + int y2 = (j * z->img_comp[n].v + y) * 8; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, + z->huff_dc + z->img_comp[n].hd, + z->huff_ac + ha, z->fast_ac[ha], n, + z->dequant[z->img_comp[n].tq])) + return 0; + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * y2 + x2, + z->img_comp[n].w2, data); + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } else { + if (z->scan_n == 1) { + int i, j; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short *data = + z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc( + z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], + z->fast_ac[ha])) + return 0; + } + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i, j, k, x, y; + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k = 0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y = 0; y < z->img_comp[n].v; ++y) { + for (x = 0; x < z->img_comp[n].h; ++x) { + int x2 = (i * z->img_comp[n].h + x); + int y2 = (j * z->img_comp[n].v + y); + short *data = z->img_comp[n].coeff + + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc( + z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} - return stbi__err("unknown marker","Corrupt JPEG"); +static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) { + int i; + for (i = 0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg *z) { + if (z->progressive) { + // dequantize and idct the data + int i, j, n; + for (n = 0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short *data = + z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, + z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg *z, int m) { + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker", "Corrupt JPEG"); + + case 0xDD: // DRI - specify restart interval + if (stbi__get16be(z->s) != 4) + return stbi__err("bad DRI len", "Corrupt JPEG"); + z->restart_interval = stbi__get16be(z->s); + return 1; + + case 0xDB: // DQT - define quantization table + L = stbi__get16be(z->s) - 2; + while (L > 0) { + int q = stbi__get8(z->s); + int p = q >> 4, sixteen = (p != 0); + int t = q & 15, i; + if (p != 0 && p != 1) return stbi__err("bad DQT type", "Corrupt JPEG"); + if (t > 3) return stbi__err("bad DQT table", "Corrupt JPEG"); + + for (i = 0; i < 64; ++i) + z->dequant[t][stbi__jpeg_dezigzag[i]] = + (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 65); + } + return L == 0; + + case 0xC4: // DHT - define huffman table + L = stbi__get16be(z->s) - 2; + while (L > 0) { + stbi_uc *v; + int sizes[16], i, n = 0; + int q = stbi__get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) + return stbi__err("bad DHT header", "Corrupt JPEG"); + for (i = 0; i < 16; ++i) { + sizes[i] = stbi__get8(z->s); + n += sizes[i]; + } + L -= 17; + if (tc == 0) { + if (!stbi__build_huffman(z->huff_dc + th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!stbi__build_huffman(z->huff_ac + th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i = 0; i < n; ++i) + v[i] = stbi__get8(z->s); + if (tc != 0) stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L == 0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len", "Corrupt JPEG"); + else + return stbi__err("bad APP len", "Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = {'J', 'F', 'I', 'F', '\0'}; + int ok = 1; + int i; + for (i = 0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) ok = 0; + L -= 5; + if (ok) z->jfif = 1; + } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = {'A', 'd', 'o', 'b', 'e', '\0'}; + int ok = 1; + int i; + for (i = 0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker", "Corrupt JPEG"); } // after we see SOS -static int stbi__process_scan_header(stbi__jpeg *z) -{ - int i; - int Ls = stbi__get16be(z->s); - z->scan_n = stbi__get8(z->s); - if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return stbi__err("bad SOS component count","Corrupt JPEG"); - if (Ls != 6+2*z->scan_n) return stbi__err("bad SOS len","Corrupt JPEG"); - for (i=0; i < z->scan_n; ++i) { - int id = stbi__get8(z->s), which; - int q = stbi__get8(z->s); - for (which = 0; which < z->s->img_n; ++which) - if (z->img_comp[which].id == id) - break; - if (which == z->s->img_n) return 0; // no match - z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff","Corrupt JPEG"); - z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff","Corrupt JPEG"); - z->order[i] = which; - } +static int stbi__process_scan_header(stbi__jpeg *z) { + int i; + int Ls = stbi__get16be(z->s); + z->scan_n = stbi__get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int)z->s->img_n) + return stbi__err("bad SOS component count", "Corrupt JPEG"); + if (Ls != 6 + 2 * z->scan_n) return stbi__err("bad SOS len", "Corrupt JPEG"); + for (i = 0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; + if (z->img_comp[which].hd > 3) + return stbi__err("bad DC huff", "Corrupt JPEG"); + z->img_comp[which].ha = q & 15; + if (z->img_comp[which].ha > 3) + return stbi__err("bad AC huff", "Corrupt JPEG"); + z->order[i] = which; + } - { - int aa; - z->spec_start = stbi__get8(z->s); - z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 - aa = stbi__get8(z->s); - z->succ_high = (aa >> 4); - z->succ_low = (aa & 15); - if (z->progressive) { - if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) - return stbi__err("bad SOS", "Corrupt JPEG"); - } else { - if (z->spec_start != 0) return stbi__err("bad SOS","Corrupt JPEG"); - if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS","Corrupt JPEG"); - z->spec_end = 63; - } - } + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || + z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } else { + if (z->spec_start != 0) return stbi__err("bad SOS", "Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) + return stbi__err("bad SOS", "Corrupt JPEG"); + z->spec_end = 63; + } + } - return 1; + return 1; } -static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) -{ - int i; - for (i=0; i < ncomp; ++i) { - if (z->img_comp[i].raw_data) { - STBI_FREE(z->img_comp[i].raw_data); - z->img_comp[i].raw_data = NULL; - z->img_comp[i].data = NULL; - } - if (z->img_comp[i].raw_coeff) { - STBI_FREE(z->img_comp[i].raw_coeff); - z->img_comp[i].raw_coeff = 0; - z->img_comp[i].coeff = 0; - } - if (z->img_comp[i].linebuf) { - STBI_FREE(z->img_comp[i].linebuf); - z->img_comp[i].linebuf = NULL; - } - } - return why; -} - -static int stbi__process_frame_header(stbi__jpeg *z, int scan) -{ - stbi__context *s = z->s; - int Lf,p,i,q, h_max=1,v_max=1,c; - Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len","Corrupt JPEG"); // JPEG - p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline - s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG - s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); // JPEG requires - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - c = stbi__get8(s); - if (c != 3 && c != 1 && c != 4) return stbi__err("bad component count","Corrupt JPEG"); - s->img_n = c; - for (i=0; i < c; ++i) { +static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) { + int i; + for (i = 0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; z->img_comp[i].data = NULL; - z->img_comp[i].linebuf = NULL; - } - - if (Lf != 8+3*s->img_n) return stbi__err("bad SOF len","Corrupt JPEG"); - - z->rgb = 0; - for (i=0; i < s->img_n; ++i) { - static const unsigned char rgb[3] = { 'R', 'G', 'B' }; - z->img_comp[i].id = stbi__get8(s); - if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) - ++z->rgb; - q = stbi__get8(s); - z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H","Corrupt JPEG"); - z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V","Corrupt JPEG"); - z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ","Corrupt JPEG"); - } - - if (scan != STBI__SCAN_load) return 1; - - if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) return stbi__err("too large", "Image too large to decode"); - - for (i=0; i < s->img_n; ++i) { - if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; - if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; - } - - // compute interleaved mcu info - z->img_h_max = h_max; - z->img_v_max = v_max; - z->img_mcu_w = h_max * 8; - z->img_mcu_h = v_max * 8; - // these sizes can't be more than 17 bits - z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; - z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; - - for (i=0; i < s->img_n; ++i) { - // number of effective pixels (e.g. for non-interleaved MCU) - z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; - z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; - // to simplify generation, we'll allocate enough memory to decode - // the bogus oversized data from using interleaved MCUs and their - // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't - // discard the extra data until colorspace conversion - // - // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) - // so these muls can't overflow with 32-bit ints (which we require) - z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; - z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; - z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); z->img_comp[i].linebuf = NULL; - z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); - if (z->img_comp[i].raw_data == NULL) - return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); - // align blocks for idct using mmx/sse - z->img_comp[i].data = (stbi_uc*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); - if (z->progressive) { - // w2, h2 are multiples of 8 (see above) - z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; - z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; - z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); - if (z->img_comp[i].raw_coeff == NULL) - return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); - z->img_comp[i].coeff = (short*) (((size_t) z->img_comp[i].raw_coeff + 15) & ~15); - } - } - - return 1; + } + } + return why; } -// use comparisons since in some cases we handle more than one case (e.g. SOF) -#define stbi__DNL(x) ((x) == 0xdc) -#define stbi__SOI(x) ((x) == 0xd8) -#define stbi__EOI(x) ((x) == 0xd9) -#define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) -#define stbi__SOS(x) ((x) == 0xda) +static int stbi__process_frame_header(stbi__jpeg *z, int scan) { + stbi__context *s = z->s; + int Lf, p, i, q, h_max = 1, v_max = 1, c; + Lf = stbi__get16be(s); + if (Lf < 11) return stbi__err("bad SOF len", "Corrupt JPEG"); // JPEG + p = stbi__get8(s); + if (p != 8) + return stbi__err("only 8-bit", + "JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = stbi__get16be(s); + if (s->img_y == 0) + return stbi__err( + "no header height", + "JPEG format not supported: delayed height"); // Legal, but we don't + // handle it--but neither + // does IJG + s->img_x = stbi__get16be(s); + if (s->img_x == 0) + return stbi__err("0 width", "Corrupt JPEG"); // JPEG requires + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + c = stbi__get8(s); + if (c != 3 && c != 1 && c != 4) + return stbi__err("bad component count", "Corrupt JPEG"); + s->img_n = c; + for (i = 0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } -#define stbi__SOF_progressive(x) ((x) == 0xc2) + if (Lf != 8 + 3 * s->img_n) return stbi__err("bad SOF len", "Corrupt JPEG"); -static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) -{ - int m; - z->jfif = 0; - z->app14_color_transform = -1; // valid values are 0,1,2 - z->marker = STBI__MARKER_none; // initialize cached marker to empty - m = stbi__get_marker(z); - if (!stbi__SOI(m)) return stbi__err("no SOI","Corrupt JPEG"); - if (scan == STBI__SCAN_type) return 1; - m = stbi__get_marker(z); - while (!stbi__SOF(m)) { - if (!stbi__process_marker(z,m)) return 0; + z->rgb = 0; + for (i = 0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = {'R', 'G', 'B'}; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); + if (!z->img_comp[i].h || z->img_comp[i].h > 4) + return stbi__err("bad H", "Corrupt JPEG"); + z->img_comp[i].v = q & 15; + if (!z->img_comp[i].v || z->img_comp[i].v > 4) + return stbi__err("bad V", "Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); + if (z->img_comp[i].tq > 3) return stbi__err("bad TQ", "Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) + return stbi__err("too large", "Image too large to decode"); + + for (i = 0; i < s->img_n; ++i) { + if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; + if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + // these sizes can't be more than 17 bits + z->img_mcu_x = (s->img_x + z->img_mcu_w - 1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h - 1) / z->img_mcu_h; + + for (i = 0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max - 1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max - 1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked + // earlier) so these muls can't overflow with 32-bit ints (which we require) + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = + stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i + 1, + stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = + (stbi_uc *)(((size_t)z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3( + z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components( + z, i + 1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = + (short *)(((size_t)z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + + // use comparisons since in some cases we handle more than one case (e.g. + // SOF) + #define stbi__DNL(x) ((x) == 0xdc) + #define stbi__SOI(x) ((x) == 0xd8) + #define stbi__EOI(x) ((x) == 0xd9) + #define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) + #define stbi__SOS(x) ((x) == 0xda) + + #define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) { + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI", "Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z, m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); m = stbi__get_marker(z); - while (m == STBI__MARKER_none) { - // some files have extra padding after their blocks, so ok, we'll scan - if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); - m = stbi__get_marker(z); - } - } - z->progressive = stbi__SOF_progressive(m); - if (!stbi__process_frame_header(z, scan)) return 0; - return 1; + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; } // decode image to YCbCr format -static int stbi__decode_jpeg_image(stbi__jpeg *j) -{ - int m; - for (m = 0; m < 4; m++) { - j->img_comp[m].raw_data = NULL; - j->img_comp[m].raw_coeff = NULL; - } - j->restart_interval = 0; - if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; - m = stbi__get_marker(j); - while (!stbi__EOI(m)) { - if (stbi__SOS(m)) { - if (!stbi__process_scan_header(j)) return 0; - if (!stbi__parse_entropy_coded_data(j)) return 0; - if (j->marker == STBI__MARKER_none ) { - // handle 0s at the end of image data from IP Kamera 9060 - while (!stbi__at_eof(j->s)) { - int x = stbi__get8(j->s); - if (x == 255) { - j->marker = stbi__get8(j->s); - break; - } - } - // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 - } - } else if (stbi__DNL(m)) { - int Ld = stbi__get16be(j->s); - stbi__uint32 NL = stbi__get16be(j->s); - if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); - if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); - } else { - if (!stbi__process_marker(j, m)) return 0; +static int stbi__decode_jpeg_image(stbi__jpeg *j) { + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none) { + // handle 0s at the end of image data from IP Kamera 9060 + while (!stbi__at_eof(j->s)) { + int x = stbi__get8(j->s); + if (x == 255) { + j->marker = stbi__get8(j->s); + break; + } + } + // if we reach eof without hitting a marker, stbi__get_marker() below + // will fail and we'll eventually return 0 } - m = stbi__get_marker(j); - } - if (j->progressive) - stbi__jpeg_finish(j); - return 1; + } else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + } else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + if (j->progressive) stbi__jpeg_finish(j); + return 1; } // static jfif-centered resampling (across block boundaries) typedef stbi_uc *(*resample_row_func)(stbi_uc *out, stbi_uc *in0, stbi_uc *in1, - int w, int hs); + int w, int hs); -#define stbi__div4(x) ((stbi_uc) ((x) >> 2)) + #define stbi__div4(x) ((stbi_uc)((x) >> 2)) -static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - STBI_NOTUSED(out); - STBI_NOTUSED(in_far); - STBI_NOTUSED(w); - STBI_NOTUSED(hs); - return in_near; +static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, + int w, int hs) { + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; } -static stbi_uc* stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate two samples vertically for every one in input - int i; - STBI_NOTUSED(hs); - for (i=0; i < w; ++i) - out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); - return out; +static stbi_uc *stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i = 0; i < w; ++i) + out[i] = stbi__div4(3 * in_near[i] + in_far[i] + 2); + return out; } -static stbi_uc* stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate two samples horizontally for every one in input - int i; - stbi_uc *input = in_near; +static stbi_uc *stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate two samples horizontally for every one in input + int i; + stbi_uc *input = in_near; - if (w == 1) { - // if only one sample, can't do any interpolation - out[0] = out[1] = input[0]; - return out; - } + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } - out[0] = input[0]; - out[1] = stbi__div4(input[0]*3 + input[1] + 2); - for (i=1; i < w-1; ++i) { - int n = 3*input[i]+2; - out[i*2+0] = stbi__div4(n+input[i-1]); - out[i*2+1] = stbi__div4(n+input[i+1]); - } - out[i*2+0] = stbi__div4(input[w-2]*3 + input[w-1] + 2); - out[i*2+1] = input[w-1]; + out[0] = input[0]; + out[1] = stbi__div4(input[0] * 3 + input[1] + 2); + for (i = 1; i < w - 1; ++i) { + int n = 3 * input[i] + 2; + out[i * 2 + 0] = stbi__div4(n + input[i - 1]); + out[i * 2 + 1] = stbi__div4(n + input[i + 1]); + } + out[i * 2 + 0] = stbi__div4(input[w - 2] * 3 + input[w - 1] + 2); + out[i * 2 + 1] = input[w - 1]; - STBI_NOTUSED(in_far); - STBI_NOTUSED(hs); + STBI_NOTUSED(in_far); + STBI_NOTUSED(hs); - return out; + return out; } -#define stbi__div16(x) ((stbi_uc) ((x) >> 4)) + #define stbi__div16(x) ((stbi_uc)((x) >> 4)) -static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate 2x2 samples for every one in input - int i,t0,t1; - if (w == 1) { - out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); - return out; - } +static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate 2x2 samples for every one in input + int i, t0, t1; + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } - t1 = 3*in_near[0] + in_far[0]; - out[0] = stbi__div4(t1+2); - for (i=1; i < w; ++i) { - t0 = t1; - t1 = 3*in_near[i]+in_far[i]; - out[i*2-1] = stbi__div16(3*t0 + t1 + 8); - out[i*2 ] = stbi__div16(3*t1 + t0 + 8); - } - out[w*2-1] = stbi__div4(t1+2); + t1 = 3 * in_near[0] + in_far[0]; + out[0] = stbi__div4(t1 + 2); + for (i = 1; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); - STBI_NOTUSED(hs); + STBI_NOTUSED(hs); - return out; + return out; } -#if defined(STBI_SSE2) || defined(STBI_NEON) -static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate 2x2 samples for every one in input - int i=0,t0,t1; + #if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate 2x2 samples for every one in input + int i = 0, t0, t1; - if (w == 1) { - out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); - return out; - } + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } - t1 = 3*in_near[0] + in_far[0]; - // process groups of 8 pixels for as long as we can. - // note we can't handle the last pixel in a row in this loop - // because we need to handle the filter boundary conditions. - for (; i < ((w-1) & ~7); i += 8) { -#if defined(STBI_SSE2) - // load and perform the vertical filtering pass - // this uses 3*x + y = 4*x + (y - x) - __m128i zero = _mm_setzero_si128(); - __m128i farb = _mm_loadl_epi64((__m128i *) (in_far + i)); - __m128i nearb = _mm_loadl_epi64((__m128i *) (in_near + i)); - __m128i farw = _mm_unpacklo_epi8(farb, zero); - __m128i nearw = _mm_unpacklo_epi8(nearb, zero); - __m128i diff = _mm_sub_epi16(farw, nearw); - __m128i nears = _mm_slli_epi16(nearw, 2); - __m128i curr = _mm_add_epi16(nears, diff); // current row + t1 = 3 * in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w - 1) & ~7); i += 8) { + #if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i *)(in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i *)(in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row - // horizontal filter works the same based on shifted vers of current - // row. "prev" is current row shifted right by 1 pixel; we need to - // insert the previous pixel value (from t1). - // "next" is current row shifted left by 1 pixel, with first pixel - // of next block of 8 pixels added in. - __m128i prv0 = _mm_slli_si128(curr, 2); - __m128i nxt0 = _mm_srli_si128(curr, 2); - __m128i prev = _mm_insert_epi16(prv0, t1, 0); - __m128i next = _mm_insert_epi16(nxt0, 3*in_near[i+8] + in_far[i+8], 7); + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = + _mm_insert_epi16(nxt0, 3 * in_near[i + 8] + in_far[i + 8], 7); - // horizontal filter, polyphase implementation since it's convenient: - // even pixels = 3*cur + prev = cur*4 + (prev - cur) - // odd pixels = 3*cur + next = cur*4 + (next - cur) - // note the shared term. - __m128i bias = _mm_set1_epi16(8); - __m128i curs = _mm_slli_epi16(curr, 2); - __m128i prvd = _mm_sub_epi16(prev, curr); - __m128i nxtd = _mm_sub_epi16(next, curr); - __m128i curb = _mm_add_epi16(curs, bias); - __m128i even = _mm_add_epi16(prvd, curb); - __m128i odd = _mm_add_epi16(nxtd, curb); + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); - // interleave even and odd pixels, then undo scaling. - __m128i int0 = _mm_unpacklo_epi16(even, odd); - __m128i int1 = _mm_unpackhi_epi16(even, odd); - __m128i de0 = _mm_srli_epi16(int0, 4); - __m128i de1 = _mm_srli_epi16(int1, 4); + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); - // pack and write output - __m128i outv = _mm_packus_epi16(de0, de1); - _mm_storeu_si128((__m128i *) (out + i*2), outv); -#elif defined(STBI_NEON) - // load and perform the vertical filtering pass - // this uses 3*x + y = 4*x + (y - x) - uint8x8_t farb = vld1_u8(in_far + i); - uint8x8_t nearb = vld1_u8(in_near + i); - int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); - int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); - int16x8_t curr = vaddq_s16(nears, diff); // current row + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i *)(out + i * 2), outv); + #elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row - // horizontal filter works the same based on shifted vers of current - // row. "prev" is current row shifted right by 1 pixel; we need to - // insert the previous pixel value (from t1). - // "next" is current row shifted left by 1 pixel, with first pixel - // of next block of 8 pixels added in. - int16x8_t prv0 = vextq_s16(curr, curr, 7); - int16x8_t nxt0 = vextq_s16(curr, curr, 1); - int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); - int16x8_t next = vsetq_lane_s16(3*in_near[i+8] + in_far[i+8], nxt0, 7); + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = + vsetq_lane_s16(3 * in_near[i + 8] + in_far[i + 8], nxt0, 7); - // horizontal filter, polyphase implementation since it's convenient: - // even pixels = 3*cur + prev = cur*4 + (prev - cur) - // odd pixels = 3*cur + next = cur*4 + (next - cur) - // note the shared term. - int16x8_t curs = vshlq_n_s16(curr, 2); - int16x8_t prvd = vsubq_s16(prev, curr); - int16x8_t nxtd = vsubq_s16(next, curr); - int16x8_t even = vaddq_s16(curs, prvd); - int16x8_t odd = vaddq_s16(curs, nxtd); + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); - // undo scaling and round, then store with even/odd phases interleaved - uint8x8x2_t o; - o.val[0] = vqrshrun_n_s16(even, 4); - o.val[1] = vqrshrun_n_s16(odd, 4); - vst2_u8(out + i*2, o); -#endif + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i * 2, o); + #endif - // "previous" value for next iter - t1 = 3*in_near[i+7] + in_far[i+7]; - } + // "previous" value for next iter + t1 = 3 * in_near[i + 7] + in_far[i + 7]; + } - t0 = t1; - t1 = 3*in_near[i] + in_far[i]; - out[i*2] = stbi__div16(3*t1 + t0 + 8); + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); - for (++i; i < w; ++i) { - t0 = t1; - t1 = 3*in_near[i]+in_far[i]; - out[i*2-1] = stbi__div16(3*t0 + t1 + 8); - out[i*2 ] = stbi__div16(3*t1 + t0 + 8); - } - out[w*2-1] = stbi__div4(t1+2); + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); - STBI_NOTUSED(hs); + STBI_NOTUSED(hs); - return out; + return out; } -#endif + #endif -static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // resample with nearest-neighbor - int i,j; - STBI_NOTUSED(in_far); - for (i=0; i < w; ++i) - for (j=0; j < hs; ++j) - out[i*hs+j] = in_near[i]; - return out; +static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // resample with nearest-neighbor + int i, j; + STBI_NOTUSED(in_far); + for (i = 0; i < w; ++i) + for (j = 0; j < hs; ++j) + out[i * hs + j] = in_near[i]; + return out; } -// this is a reduced-precision calculation of YCbCr-to-RGB introduced -// to make sure the code produces the same results in both SIMD and scalar -#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) -static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step) -{ - int i; - for (i=0; i < count; ++i) { - int y_fixed = (y[i] << 20) + (1<<19); // rounding - int r,g,b; - int cr = pcr[i] - 128; - int cb = pcb[i] - 128; - r = y_fixed + cr* stbi__float2fixed(1.40200f); - g = y_fixed + (cr*-stbi__float2fixed(0.71414f)) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); - b = y_fixed + cb* stbi__float2fixed(1.77200f); - r >>= 20; - g >>= 20; - b >>= 20; - if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } - if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } - if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } - out[0] = (stbi_uc)r; - out[1] = (stbi_uc)g; - out[2] = (stbi_uc)b; - out[3] = 255; - out += step; - } + // this is a reduced-precision calculation of YCbCr-to-RGB introduced + // to make sure the code produces the same results in both SIMD and scalar + #define stbi__float2fixed(x) (((int)((x)*4096.0f + 0.5f)) << 8) +static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, + const stbi_uc *pcb, const stbi_uc *pcr, + int count, int step) { + int i; + for (i = 0; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + (cr * -stbi__float2fixed(0.71414f)) + + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { + if (r < 0) + r = 0; + else + r = 255; + } + if ((unsigned)g > 255) { + if (g < 0) + g = 0; + else + g = 255; + } + if ((unsigned)b > 255) { + if (b < 0) + b = 0; + else + b = 255; + } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } } -#if defined(STBI_SSE2) || defined(STBI_NEON) -static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, stbi_uc const *pcb, stbi_uc const *pcr, int count, int step) -{ - int i = 0; + #if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, + stbi_uc const *pcb, stbi_uc const *pcr, + int count, int step) { + int i = 0; -#ifdef STBI_SSE2 - // step == 3 is pretty ugly on the final interleave, and i'm not convinced - // it's useful in practice (you wouldn't use it for textures, for example). - // so just accelerate step == 4 case. - if (step == 4) { - // this is a fairly straightforward implementation and not super-optimized. - __m128i signflip = _mm_set1_epi8(-0x80); - __m128i cr_const0 = _mm_set1_epi16( (short) ( 1.40200f*4096.0f+0.5f)); - __m128i cr_const1 = _mm_set1_epi16( - (short) ( 0.71414f*4096.0f+0.5f)); - __m128i cb_const0 = _mm_set1_epi16( - (short) ( 0.34414f*4096.0f+0.5f)); - __m128i cb_const1 = _mm_set1_epi16( (short) ( 1.77200f*4096.0f+0.5f)); - __m128i y_bias = _mm_set1_epi8((char) (unsigned char) 128); - __m128i xw = _mm_set1_epi16(255); // alpha channel + #ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16((short)(1.40200f * 4096.0f + 0.5f)); + __m128i cr_const1 = _mm_set1_epi16(-(short)(0.71414f * 4096.0f + 0.5f)); + __m128i cb_const0 = _mm_set1_epi16(-(short)(0.34414f * 4096.0f + 0.5f)); + __m128i cb_const1 = _mm_set1_epi16((short)(1.77200f * 4096.0f + 0.5f)); + __m128i y_bias = _mm_set1_epi8((char)(unsigned char)128); + __m128i xw = _mm_set1_epi16(255); // alpha channel - for (; i+7 < count; i += 8) { - // load - __m128i y_bytes = _mm_loadl_epi64((__m128i *) (y+i)); - __m128i cr_bytes = _mm_loadl_epi64((__m128i *) (pcr+i)); - __m128i cb_bytes = _mm_loadl_epi64((__m128i *) (pcb+i)); - __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 - __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + for (; i + 7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i *)(y + i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i *)(pcr + i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i *)(pcb + i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 - // unpack to short (and left-shift cr, cb by 8) - __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); - __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); - __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); - // color transform - __m128i yws = _mm_srli_epi16(yw, 4); - __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); - __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); - __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); - __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); - __m128i rws = _mm_add_epi16(cr0, yws); - __m128i gwt = _mm_add_epi16(cb0, yws); - __m128i bws = _mm_add_epi16(yws, cb1); - __m128i gws = _mm_add_epi16(gwt, cr1); + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); - // descale - __m128i rw = _mm_srai_epi16(rws, 4); - __m128i bw = _mm_srai_epi16(bws, 4); - __m128i gw = _mm_srai_epi16(gws, 4); + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); - // back to byte, set up for transpose - __m128i brb = _mm_packus_epi16(rw, bw); - __m128i gxb = _mm_packus_epi16(gw, xw); + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); - // transpose to interleave channels - __m128i t0 = _mm_unpacklo_epi8(brb, gxb); - __m128i t1 = _mm_unpackhi_epi8(brb, gxb); - __m128i o0 = _mm_unpacklo_epi16(t0, t1); - __m128i o1 = _mm_unpackhi_epi16(t0, t1); + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); - // store - _mm_storeu_si128((__m128i *) (out + 0), o0); - _mm_storeu_si128((__m128i *) (out + 16), o1); - out += 32; - } - } -#endif + // store + _mm_storeu_si128((__m128i *)(out + 0), o0); + _mm_storeu_si128((__m128i *)(out + 16), o1); + out += 32; + } + } + #endif -#ifdef STBI_NEON - // in this version, step=3 support would be easy to add. but is there demand? - if (step == 4) { - // this is a fairly straightforward implementation and not super-optimized. - uint8x8_t signflip = vdup_n_u8(0x80); - int16x8_t cr_const0 = vdupq_n_s16( (short) ( 1.40200f*4096.0f+0.5f)); - int16x8_t cr_const1 = vdupq_n_s16( - (short) ( 0.71414f*4096.0f+0.5f)); - int16x8_t cb_const0 = vdupq_n_s16( - (short) ( 0.34414f*4096.0f+0.5f)); - int16x8_t cb_const1 = vdupq_n_s16( (short) ( 1.77200f*4096.0f+0.5f)); + #ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16((short)(1.40200f * 4096.0f + 0.5f)); + int16x8_t cr_const1 = vdupq_n_s16(-(short)(0.71414f * 4096.0f + 0.5f)); + int16x8_t cb_const0 = vdupq_n_s16(-(short)(0.34414f * 4096.0f + 0.5f)); + int16x8_t cb_const1 = vdupq_n_s16((short)(1.77200f * 4096.0f + 0.5f)); - for (; i+7 < count; i += 8) { - // load - uint8x8_t y_bytes = vld1_u8(y + i); - uint8x8_t cr_bytes = vld1_u8(pcr + i); - uint8x8_t cb_bytes = vld1_u8(pcb + i); - int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); - int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + for (; i + 7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); - // expand to s16 - int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); - int16x8_t crw = vshll_n_s8(cr_biased, 7); - int16x8_t cbw = vshll_n_s8(cb_biased, 7); + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); - // color transform - int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); - int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); - int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); - int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); - int16x8_t rws = vaddq_s16(yws, cr0); - int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); - int16x8_t bws = vaddq_s16(yws, cb1); + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); - // undo scaling, round, convert to byte - uint8x8x4_t o; - o.val[0] = vqrshrun_n_s16(rws, 4); - o.val[1] = vqrshrun_n_s16(gws, 4); - o.val[2] = vqrshrun_n_s16(bws, 4); - o.val[3] = vdup_n_u8(255); + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); - // store, interleaving r/g/b/a - vst4_u8(out, o); - out += 8*4; - } - } -#endif + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8 * 4; + } + } + #endif - for (; i < count; ++i) { - int y_fixed = (y[i] << 20) + (1<<19); // rounding - int r,g,b; - int cr = pcr[i] - 128; - int cb = pcb[i] - 128; - r = y_fixed + cr* stbi__float2fixed(1.40200f); - g = y_fixed + cr*-stbi__float2fixed(0.71414f) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); - b = y_fixed + cb* stbi__float2fixed(1.77200f); - r >>= 20; - g >>= 20; - b >>= 20; - if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } - if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } - if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } - out[0] = (stbi_uc)r; - out[1] = (stbi_uc)g; - out[2] = (stbi_uc)b; - out[3] = 255; - out += step; - } + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + cr * -stbi__float2fixed(0.71414f) + + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { + if (r < 0) + r = 0; + else + r = 255; + } + if ((unsigned)g > 255) { + if (g < 0) + g = 0; + else + g = 255; + } + if ((unsigned)b > 255) { + if (b < 0) + b = 0; + else + b = 255; + } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } } -#endif + #endif // set up the kernels -static void stbi__setup_jpeg(stbi__jpeg *j) -{ - j->idct_block_kernel = stbi__idct_block; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; +static void stbi__setup_jpeg(stbi__jpeg *j) { + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; -#ifdef STBI_SSE2 - if (stbi__sse2_available()) { - j->idct_block_kernel = stbi__idct_simd; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; - } -#endif + #ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } + #endif -#ifdef STBI_NEON - j->idct_block_kernel = stbi__idct_simd; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; -#endif + #ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + #endif } // clean up the temporary component buffers -static void stbi__cleanup_jpeg(stbi__jpeg *j) -{ - stbi__free_jpeg_components(j, j->s->img_n, 0); +static void stbi__cleanup_jpeg(stbi__jpeg *j) { + stbi__free_jpeg_components(j, j->s->img_n, 0); } -typedef struct -{ - resample_row_func resample; - stbi_uc *line0,*line1; - int hs,vs; // expansion factor in each axis - int w_lores; // horizontal pixels pre-expansion - int ystep; // how far through vertical expansion we are - int ypos; // which pre-expansion row we're on +typedef struct { + resample_row_func resample; + stbi_uc *line0, *line1; + int hs, vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on } stbi__resample; // fast 0..255 * 0..255 => 0..255 rounded multiplication -static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) -{ - unsigned int t = x*y + 128; - return (stbi_uc) ((t + (t >>8)) >> 8); +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) { + unsigned int t = x * y + 128; + return (stbi_uc)((t + (t >> 8)) >> 8); } -static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) -{ - int n, decode_n, is_rgb; - z->s->img_n = 0; // make stbi__cleanup_jpeg safe +static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, + int *comp, int req_comp) { + int n, decode_n, is_rgb; + z->s->img_n = 0; // make stbi__cleanup_jpeg safe - // validate req_comp - if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + // validate req_comp + if (req_comp < 0 || req_comp > 4) + return stbi__errpuc("bad req_comp", "Internal error"); - // load a jpeg image from whichever source, but leave in YCbCr format - if (!stbi__decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + // load a jpeg image from whichever source, but leave in YCbCr format + if (!stbi__decode_jpeg_image(z)) { + stbi__cleanup_jpeg(z); + return NULL; + } - // determine actual number of components to generate - n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; - is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + is_rgb = z->s->img_n == 3 && + (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); - if (z->s->img_n == 3 && n < 3 && !is_rgb) - decode_n = 1; - else - decode_n = z->s->img_n; + if (z->s->img_n == 3 && n < 3 && !is_rgb) + decode_n = 1; + else + decode_n = z->s->img_n; - // resample and color-convert - { - int k; - unsigned int i,j; - stbi_uc *output; - stbi_uc *coutput[4] = { NULL, NULL, NULL, NULL }; + // resample and color-convert + { + int k; + unsigned int i, j; + stbi_uc *output; + stbi_uc *coutput[4] = {NULL, NULL, NULL, NULL}; - stbi__resample res_comp[4]; + stbi__resample res_comp[4]; - for (k=0; k < decode_n; ++k) { - stbi__resample *r = &res_comp[k]; + for (k = 0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; - // allocate line buffer big enough for upsampling off the edges - // with upsample factor of 4 - z->img_comp[k].linebuf = (stbi_uc *) stbi__malloc(z->s->img_x + 3); - if (!z->img_comp[k].linebuf) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } - - r->hs = z->img_h_max / z->img_comp[k].h; - r->vs = z->img_v_max / z->img_comp[k].v; - r->ystep = r->vs >> 1; - r->w_lores = (z->s->img_x + r->hs-1) / r->hs; - r->ypos = 0; - r->line0 = r->line1 = z->img_comp[k].data; - - if (r->hs == 1 && r->vs == 1) r->resample = resample_row_1; - else if (r->hs == 1 && r->vs == 2) r->resample = stbi__resample_row_v_2; - else if (r->hs == 2 && r->vs == 1) r->resample = stbi__resample_row_h_2; - else if (r->hs == 2 && r->vs == 2) r->resample = z->resample_row_hv_2_kernel; - else r->resample = stbi__resample_row_generic; + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (stbi_uc *)stbi__malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { + stbi__cleanup_jpeg(z); + return stbi__errpuc("outofmem", "Out of memory"); } - // can't error after this so, this is safe - output = (stbi_uc *) stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); - if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + r->hs = z->img_h_max / z->img_comp[k].h; + r->vs = z->img_v_max / z->img_comp[k].v; + r->ystep = r->vs >> 1; + r->w_lores = (z->s->img_x + r->hs - 1) / r->hs; + r->ypos = 0; + r->line0 = r->line1 = z->img_comp[k].data; - // now go ahead and resample - for (j=0; j < z->s->img_y; ++j) { - stbi_uc *out = output + n * z->s->img_x * j; - for (k=0; k < decode_n; ++k) { - stbi__resample *r = &res_comp[k]; - int y_bot = r->ystep >= (r->vs >> 1); - coutput[k] = r->resample(z->img_comp[k].linebuf, - y_bot ? r->line1 : r->line0, - y_bot ? r->line0 : r->line1, - r->w_lores, r->hs); - if (++r->ystep >= r->vs) { - r->ystep = 0; - r->line0 = r->line1; - if (++r->ypos < z->img_comp[k].y) - r->line1 += z->img_comp[k].w2; - } - } - if (n >= 3) { - stbi_uc *y = coutput[0]; - if (z->s->img_n == 3) { - if (is_rgb) { - for (i=0; i < z->s->img_x; ++i) { - out[0] = y[i]; - out[1] = coutput[1][i]; - out[2] = coutput[2][i]; - out[3] = 255; - out += n; - } - } else { - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - } - } else if (z->s->img_n == 4) { - if (z->app14_color_transform == 0) { // CMYK - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - out[0] = stbi__blinn_8x8(coutput[0][i], m); - out[1] = stbi__blinn_8x8(coutput[1][i], m); - out[2] = stbi__blinn_8x8(coutput[2][i], m); - out[3] = 255; - out += n; - } - } else if (z->app14_color_transform == 2) { // YCCK - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - out[0] = stbi__blinn_8x8(255 - out[0], m); - out[1] = stbi__blinn_8x8(255 - out[1], m); - out[2] = stbi__blinn_8x8(255 - out[2], m); - out += n; - } - } else { // YCbCr + alpha? Ignore the fourth channel for now - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - } - } else - for (i=0; i < z->s->img_x; ++i) { - out[0] = out[1] = out[2] = y[i]; - out[3] = 255; // not used if n==3 - out += n; - } - } else { - if (is_rgb) { - if (n == 1) - for (i=0; i < z->s->img_x; ++i) - *out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); - else { - for (i=0; i < z->s->img_x; ++i, out += 2) { - out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); - out[1] = 255; - } - } - } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); - stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); - stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); - out[0] = stbi__compute_y(r, g, b); - out[1] = 255; - out += n; - } - } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { - for (i=0; i < z->s->img_x; ++i) { - out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); - out[1] = 255; - out += n; - } - } else { - stbi_uc *y = coutput[0]; - if (n == 1) - for (i=0; i < z->s->img_x; ++i) out[i] = y[i]; - else - for (i=0; i < z->s->img_x; ++i) { *out++ = y[i]; *out++ = 255; } - } - } - } + if (r->hs == 1 && r->vs == 1) + r->resample = resample_row_1; + else if (r->hs == 1 && r->vs == 2) + r->resample = stbi__resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) + r->resample = stbi__resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) + r->resample = z->resample_row_hv_2_kernel; + else + r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc *)stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); + if (!output) { stbi__cleanup_jpeg(z); - *out_x = z->s->img_x; - *out_y = z->s->img_y; - if (comp) *comp = z->s->img_n >= 3 ? 3 : 1; // report original components, not output - return output; - } + return stbi__errpuc("outofmem", "Out of memory"); + } + + // now go ahead and resample + for (j = 0; j < z->s->img_y; ++j) { + stbi_uc *out = output + n * z->s->img_x * j; + for (k = 0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + int y_bot = r->ystep >= (r->vs >> 1); + coutput[k] = + r->resample(z->img_comp[k].linebuf, y_bot ? r->line1 : r->line0, + y_bot ? r->line0 : r->line1, r->w_lores, r->hs); + if (++r->ystep >= r->vs) { + r->ystep = 0; + r->line0 = r->line1; + if (++r->ypos < z->img_comp[k].y) r->line1 += z->img_comp[k].w2; + } + } + if (n >= 3) { + stbi_uc *y = coutput[0]; + if (z->s->img_n == 3) { + if (is_rgb) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + } + } else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + } + } else + for (i = 0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + if (is_rgb) { + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + *out++ = + stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i = 0; i < z->s->img_x; ++i, out += 2) { + out[0] = + stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + out += n; + } + } else { + stbi_uc *y = coutput[0]; + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + out[i] = y[i]; + else + for (i = 0; i < z->s->img_x; ++i) { + *out++ = y[i]; + *out++ = 255; + } + } + } + } + stbi__cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) + *comp = + z->s->img_n >= 3 ? 3 : 1; // report original components, not output + return output; + } } -static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - unsigned char* result; - stbi__jpeg* j = (stbi__jpeg*) stbi__malloc(sizeof(stbi__jpeg)); - STBI_NOTUSED(ri); - j->s = s; - stbi__setup_jpeg(j); - result = load_jpeg_image(j, x,y,comp,req_comp); - STBI_FREE(j); - return result; +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + unsigned char *result; + stbi__jpeg *j = (stbi__jpeg *)stbi__malloc(sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x, y, comp, req_comp); + STBI_FREE(j); + return result; } -static int stbi__jpeg_test(stbi__context *s) -{ - int r; - stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); - j->s = s; - stbi__setup_jpeg(j); - r = stbi__decode_jpeg_header(j, STBI__SCAN_type); - stbi__rewind(s); - STBI_FREE(j); - return r; +static int stbi__jpeg_test(stbi__context *s) { + int r; + stbi__jpeg *j = (stbi__jpeg *)stbi__malloc(sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; } -static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) -{ - if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { - stbi__rewind( j->s ); - return 0; - } - if (x) *x = j->s->img_x; - if (y) *y = j->s->img_y; - if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; - return 1; +static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) { + if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { + stbi__rewind(j->s); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; + return 1; } -static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) -{ - int result; - stbi__jpeg* j = (stbi__jpeg*) (stbi__malloc(sizeof(stbi__jpeg))); - j->s = s; - result = stbi__jpeg_info_raw(j, x, y, comp); - STBI_FREE(j); - return result; +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) { + int result; + stbi__jpeg *j = (stbi__jpeg *)(stbi__malloc(sizeof(stbi__jpeg))); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; } -#endif + #endif // public domain zlib decode v0.2 Sean Barrett 2006-11-18 // simple implementation @@ -3974,86 +4314,83 @@ static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) // performance // - fast huffman -#ifndef STBI_NO_ZLIB + #ifndef STBI_NO_ZLIB -// fast-way is faster to check than jpeg huffman, but slow way is slower -#define STBI__ZFAST_BITS 9 // accelerate all cases in default tables -#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) + // fast-way is faster to check than jpeg huffman, but slow way is slower + #define STBI__ZFAST_BITS 9 // accelerate all cases in default tables + #define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) // zlib-style huffman encoding // (jpegs packs from left, zlib from right, so can't share code) -typedef struct -{ - stbi__uint16 fast[1 << STBI__ZFAST_BITS]; - stbi__uint16 firstcode[16]; - int maxcode[17]; - stbi__uint16 firstsymbol[16]; - stbi_uc size[288]; - stbi__uint16 value[288]; +typedef struct { + stbi__uint16 fast[1 << STBI__ZFAST_BITS]; + stbi__uint16 firstcode[16]; + int maxcode[17]; + stbi__uint16 firstsymbol[16]; + stbi_uc size[288]; + stbi__uint16 value[288]; } stbi__zhuffman; -stbi_inline static int stbi__bitreverse16(int n) -{ - n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); - n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); - n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); - n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); +stbi_inline static int stbi__bitreverse16(int n) { + n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); + n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); + n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); + n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); return n; } -stbi_inline static int stbi__bit_reverse(int v, int bits) -{ - STBI_ASSERT(bits <= 16); - // to bit reverse n bits, reverse 16 and shift - // e.g. 11 bits, bit reverse and shift away 5 - return stbi__bitreverse16(v) >> (16-bits); +stbi_inline static int stbi__bit_reverse(int v, int bits) { + STBI_ASSERT(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return stbi__bitreverse16(v) >> (16 - bits); } -static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int num) -{ - int i,k=0; - int code, next_code[16], sizes[17]; +static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, + int num) { + int i, k = 0; + int code, next_code[16], sizes[17]; - // DEFLATE spec for generating codes - memset(sizes, 0, sizeof(sizes)); - memset(z->fast, 0, sizeof(z->fast)); - for (i=0; i < num; ++i) - ++sizes[sizelist[i]]; - sizes[0] = 0; - for (i=1; i < 16; ++i) - if (sizes[i] > (1 << i)) - return stbi__err("bad sizes", "Corrupt PNG"); - code = 0; - for (i=1; i < 16; ++i) { - next_code[i] = code; - z->firstcode[i] = (stbi__uint16) code; - z->firstsymbol[i] = (stbi__uint16) k; - code = (code + sizes[i]); - if (sizes[i]) - if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt PNG"); - z->maxcode[i] = code << (16-i); // preshift for inner loop - code <<= 1; - k += sizes[i]; - } - z->maxcode[16] = 0x10000; // sentinel - for (i=0; i < num; ++i) { - int s = sizelist[i]; - if (s) { - int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; - stbi__uint16 fastv = (stbi__uint16) ((s << 9) | i); - z->size [c] = (stbi_uc ) s; - z->value[c] = (stbi__uint16) i; - if (s <= STBI__ZFAST_BITS) { - int j = stbi__bit_reverse(next_code[s],s); - while (j < (1 << STBI__ZFAST_BITS)) { - z->fast[j] = fastv; - j += (1 << s); - } - } - ++next_code[s]; + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 0, sizeof(z->fast)); + for (i = 0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i = 1; i < 16; ++i) + if (sizes[i] > (1 << i)) return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i = 1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (stbi__uint16)code; + z->firstsymbol[i] = (stbi__uint16)k; + code = (code + sizes[i]); + if (sizes[i]) + if (code - 1 >= (1 << i)) + return stbi__err("bad codelengths", "Corrupt PNG"); + z->maxcode[i] = code << (16 - i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i = 0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + stbi__uint16 fastv = (stbi__uint16)((s << 9) | i); + z->size[c] = (stbi_uc)s; + z->value[c] = (stbi__uint16)i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s], s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (1 << s); + } } - } - return 1; + ++next_code[s]; + } + } + return 1; } // zlib-from-memory implementation for PNG reading @@ -4062,277 +4399,291 @@ static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int // we require PNG read all the IDATs and combine them into a single // memory buffer -typedef struct -{ - stbi_uc *zbuffer, *zbuffer_end; - int num_bits; - stbi__uint32 code_buffer; +typedef struct { + stbi_uc *zbuffer, *zbuffer_end; + int num_bits; + stbi__uint32 code_buffer; - char *zout; - char *zout_start; - char *zout_end; - int z_expandable; + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; - stbi__zhuffman z_length, z_distance; + stbi__zhuffman z_length, z_distance; } stbi__zbuf; -stbi_inline static int stbi__zeof(stbi__zbuf *z) -{ - return (z->zbuffer >= z->zbuffer_end); +stbi_inline static int stbi__zeof(stbi__zbuf *z) { + return (z->zbuffer >= z->zbuffer_end); } -stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) -{ - return stbi__zeof(z) ? 0 : *z->zbuffer++; +stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) { + return stbi__zeof(z) ? 0 : *z->zbuffer++; } -static void stbi__fill_bits(stbi__zbuf *z) -{ - do { - if (z->code_buffer >= (1U << z->num_bits)) { - z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ - return; - } - z->code_buffer |= (unsigned int) stbi__zget8(z) << z->num_bits; - z->num_bits += 8; - } while (z->num_bits <= 24); +static void stbi__fill_bits(stbi__zbuf *z) { + do { + if (z->code_buffer >= (1U << z->num_bits)) { + z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ + return; + } + z->code_buffer |= (unsigned int)stbi__zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); } -stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) -{ - unsigned int k; - if (z->num_bits < n) stbi__fill_bits(z); - k = z->code_buffer & ((1 << n) - 1); - z->code_buffer >>= n; - z->num_bits -= n; - return k; +stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) { + unsigned int k; + if (z->num_bits < n) stbi__fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; } -static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) -{ - int b,s,k; - // not resolved by fast table, so compute it the slow way - // use jpeg approach, which requires MSbits at top - k = stbi__bit_reverse(a->code_buffer, 16); - for (s=STBI__ZFAST_BITS+1; ; ++s) - if (k < z->maxcode[s]) - break; - if (s >= 16) return -1; // invalid code! - // code size is s, so: - b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; - if (b >= sizeof (z->size)) return -1; // some data was corrupt somewhere! - if (z->size[b] != s) return -1; // was originally an assert, but report failure instead. - a->code_buffer >>= s; - a->num_bits -= s; - return z->value[b]; +static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) { + int b, s, k; + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = stbi__bit_reverse(a->code_buffer, 16); + for (s = STBI__ZFAST_BITS + 1;; ++s) + if (k < z->maxcode[s]) break; + if (s >= 16) return -1; // invalid code! + // code size is s, so: + b = (k >> (16 - s)) - z->firstcode[s] + z->firstsymbol[s]; + if (b >= sizeof(z->size)) return -1; // some data was corrupt somewhere! + if (z->size[b] != s) + return -1; // was originally an assert, but report failure instead. + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; } -stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) -{ - int b,s; - if (a->num_bits < 16) { - if (stbi__zeof(a)) { - return -1; /* report error for unexpected end of data. */ - } - stbi__fill_bits(a); - } - b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; - if (b) { - s = b >> 9; - a->code_buffer >>= s; - a->num_bits -= s; - return b & 511; - } - return stbi__zhuffman_decode_slowpath(a, z); +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) { + int b, s; + if (a->num_bits < 16) { + if (stbi__zeof(a)) { + return -1; /* report error for unexpected end of data. */ + } + stbi__fill_bits(a); + } + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); } -static int stbi__zexpand(stbi__zbuf *z, char *zout, int n) // need to make room for n bytes +static int stbi__zexpand(stbi__zbuf *z, char *zout, + int n) // need to make room for n bytes { - char *q; - unsigned int cur, limit, old_limit; - z->zout = zout; - if (!z->z_expandable) return stbi__err("output buffer limit","Corrupt PNG"); - cur = (unsigned int) (z->zout - z->zout_start); - limit = old_limit = (unsigned) (z->zout_end - z->zout_start); - if (UINT_MAX - cur < (unsigned) n) return stbi__err("outofmem", "Out of memory"); - while (cur + n > limit) { - if(limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); - limit *= 2; - } - q = (char *) STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); - STBI_NOTUSED(old_limit); - if (q == NULL) return stbi__err("outofmem", "Out of memory"); - z->zout_start = q; - z->zout = q + cur; - z->zout_end = q + limit; - return 1; + char *q; + unsigned int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit", "Corrupt PNG"); + cur = (unsigned int)(z->zout - z->zout_start); + limit = old_limit = (unsigned)(z->zout_end - z->zout_start); + if (UINT_MAX - cur < (unsigned)n) + return stbi__err("outofmem", "Out of memory"); + while (cur + n > limit) { + if (limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); + limit *= 2; + } + q = (char *)STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_limit); + if (q == NULL) return stbi__err("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; } static const int stbi__zlength_base[31] = { - 3,4,5,6,7,8,9,10,11,13, - 15,17,19,23,27,31,35,43,51,59, - 67,83,99,115,131,163,195,227,258,0,0 }; + 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, + 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0}; -static const int stbi__zlength_extra[31]= -{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 }; +static const int stbi__zlength_extra[31] = {0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, + 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, + 4, 4, 5, 5, 5, 5, 0, 0, 0}; -static const int stbi__zdist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193, -257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0}; +static const int stbi__zdist_base[32] = { + 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, + 49, 65, 97, 129, 193, 257, 385, 513, 769, 1025, 1537, + 2049, 3073, 4097, 6145, 8193, 12289, 16385, 24577, 0, 0}; -static const int stbi__zdist_extra[32] = -{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; +static const int stbi__zdist_extra[32] = {0, 0, 0, 0, 1, 1, 2, 2, 3, 3, + 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, + 9, 9, 10, 10, 11, 11, 12, 12, 13, 13}; -static int stbi__parse_huffman_block(stbi__zbuf *a) -{ - char *zout = a->zout; - for(;;) { - int z = stbi__zhuffman_decode(a, &a->z_length); - if (z < 256) { - if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); // error in huffman codes - if (zout >= a->zout_end) { - if (!stbi__zexpand(a, zout, 1)) return 0; - zout = a->zout; - } - *zout++ = (char) z; +static int stbi__parse_huffman_block(stbi__zbuf *a) { + char *zout = a->zout; + for (;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) + return stbi__err("bad huffman code", + "Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char)z; + } else { + stbi_uc *p; + int len, dist; + if (z == 256) { + a->zout = zout; + return 1; + } + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) + len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0) return stbi__err("bad huffman code", "Corrupt PNG"); + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) + return stbi__err("bad dist", "Corrupt PNG"); + if (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc *)(zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { + do + *zout++ = v; + while (--len); + } } else { - stbi_uc *p; - int len,dist; - if (z == 256) { - a->zout = zout; - return 1; - } - z -= 257; - len = stbi__zlength_base[z]; - if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); - z = stbi__zhuffman_decode(a, &a->z_distance); - if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); - dist = stbi__zdist_base[z]; - if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); - if (zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); - if (zout + len > a->zout_end) { - if (!stbi__zexpand(a, zout, len)) return 0; - zout = a->zout; - } - p = (stbi_uc *) (zout - dist); - if (dist == 1) { // run of one byte; common in images. - stbi_uc v = *p; - if (len) { do *zout++ = v; while (--len); } - } else { - if (len) { do *zout++ = *p++; while (--len); } - } + if (len) { + do + *zout++ = *p++; + while (--len); + } } - } + } + } } -static int stbi__compute_huffman_codes(stbi__zbuf *a) -{ - static const stbi_uc length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; - stbi__zhuffman z_codelength; - stbi_uc lencodes[286+32+137];//padding for maximum single op - stbi_uc codelength_sizes[19]; - int i,n; +static int stbi__compute_huffman_codes(stbi__zbuf *a) { + static const stbi_uc length_dezigzag[19] = { + 16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; + stbi__zhuffman z_codelength; + stbi_uc lencodes[286 + 32 + 137]; // padding for maximum single op + stbi_uc codelength_sizes[19]; + int i, n; - int hlit = stbi__zreceive(a,5) + 257; - int hdist = stbi__zreceive(a,5) + 1; - int hclen = stbi__zreceive(a,4) + 4; - int ntot = hlit + hdist; + int hlit = stbi__zreceive(a, 5) + 257; + int hdist = stbi__zreceive(a, 5) + 1; + int hclen = stbi__zreceive(a, 4) + 4; + int ntot = hlit + hdist; - memset(codelength_sizes, 0, sizeof(codelength_sizes)); - for (i=0; i < hclen; ++i) { - int s = stbi__zreceive(a,3); - codelength_sizes[length_dezigzag[i]] = (stbi_uc) s; - } - if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i = 0; i < hclen; ++i) { + int s = stbi__zreceive(a, 3); + codelength_sizes[length_dezigzag[i]] = (stbi_uc)s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; - n = 0; - while (n < ntot) { - int c = stbi__zhuffman_decode(a, &z_codelength); - if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); - if (c < 16) - lencodes[n++] = (stbi_uc) c; - else { - stbi_uc fill = 0; - if (c == 16) { - c = stbi__zreceive(a,2)+3; - if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); - fill = lencodes[n-1]; - } else if (c == 17) { - c = stbi__zreceive(a,3)+3; - } else if (c == 18) { - c = stbi__zreceive(a,7)+11; - } else { - return stbi__err("bad codelengths", "Corrupt PNG"); - } - if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); - memset(lencodes+n, fill, c); - n += c; + n = 0; + while (n < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc)c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a, 2) + 3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n - 1]; + } else if (c == 17) { + c = stbi__zreceive(a, 3) + 3; + } else if (c == 18) { + c = stbi__zreceive(a, 7) + 11; + } else { + return stbi__err("bad codelengths", "Corrupt PNG"); } - } - if (n != ntot) return stbi__err("bad codelengths","Corrupt PNG"); - if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; - if (!stbi__zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; - return 1; + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes + n, fill, c); + n += c; + } + } + if (n != ntot) return stbi__err("bad codelengths", "Corrupt PNG"); + if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, lencodes + hlit, hdist)) return 0; + return 1; } -static int stbi__parse_uncompressed_block(stbi__zbuf *a) -{ - stbi_uc header[4]; - int len,nlen,k; - if (a->num_bits & 7) - stbi__zreceive(a, a->num_bits & 7); // discard - // drain the bit-packed data into header - k = 0; - while (a->num_bits > 0) { - header[k++] = (stbi_uc) (a->code_buffer & 255); // suppress MSVC run-time check - a->code_buffer >>= 8; - a->num_bits -= 8; - } - if (a->num_bits < 0) return stbi__err("zlib corrupt","Corrupt PNG"); - // now fill header the normal way - while (k < 4) - header[k++] = stbi__zget8(a); - len = header[1] * 256 + header[0]; - nlen = header[3] * 256 + header[2]; - if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt","Corrupt PNG"); - if (a->zbuffer + len > a->zbuffer_end) return stbi__err("read past buffer","Corrupt PNG"); - if (a->zout + len > a->zout_end) - if (!stbi__zexpand(a, a->zout, len)) return 0; - memcpy(a->zout, a->zbuffer, len); - a->zbuffer += len; - a->zout += len; - return 1; +static int stbi__parse_uncompressed_block(stbi__zbuf *a) { + stbi_uc header[4]; + int len, nlen, k; + if (a->num_bits & 7) stbi__zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = + (stbi_uc)(a->code_buffer & 255); // suppress MSVC run-time check + a->code_buffer >>= 8; + a->num_bits -= 8; + } + if (a->num_bits < 0) return stbi__err("zlib corrupt", "Corrupt PNG"); + // now fill header the normal way + while (k < 4) + header[k++] = stbi__zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt", "Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) + return stbi__err("read past buffer", "Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!stbi__zexpand(a, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; } -static int stbi__parse_zlib_header(stbi__zbuf *a) -{ - int cmf = stbi__zget8(a); - int cm = cmf & 15; - /* int cinfo = cmf >> 4; */ - int flg = stbi__zget8(a); - if (stbi__zeof(a)) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec - if ((cmf*256+flg) % 31 != 0) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec - if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png - if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); // DEFLATE required for png - // window = 1 << (8 + cinfo)... but who cares, we fully buffer output - return 1; +static int stbi__parse_zlib_header(stbi__zbuf *a) { + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if (stbi__zeof(a)) + return stbi__err("bad zlib header", "Corrupt PNG"); // zlib spec + if ((cmf * 256 + flg) % 31 != 0) + return stbi__err("bad zlib header", "Corrupt PNG"); // zlib spec + if (flg & 32) + return stbi__err("no preset dict", + "Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) + return stbi__err("bad compression", + "Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; } -static const stbi_uc stbi__zdefault_length[288] = -{ - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 -}; -static const stbi_uc stbi__zdefault_distance[32] = -{ - 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 -}; +static const stbi_uc stbi__zdefault_length[288] = { + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 8}; +static const stbi_uc stbi__zdefault_distance[32] = { + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5}; /* Init algorithm: { @@ -4346,119 +4697,123 @@ Init algorithm: } */ -static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) -{ - int final, type; - if (parse_header) - if (!stbi__parse_zlib_header(a)) return 0; - a->num_bits = 0; - a->code_buffer = 0; - do { - final = stbi__zreceive(a,1); - type = stbi__zreceive(a,2); - if (type == 0) { - if (!stbi__parse_uncompressed_block(a)) return 0; - } else if (type == 3) { - return 0; +static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) { + int final, type; + if (parse_header) + if (!stbi__parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + do { + final = stbi__zreceive(a, 1); + type = stbi__zreceive(a, 2); + if (type == 0) { + if (!stbi__parse_uncompressed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + // use fixed code lengths + if (!stbi__zbuild_huffman(&a->z_length, stbi__zdefault_length, 288)) + return 0; + if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) + return 0; } else { - if (type == 1) { - // use fixed code lengths - if (!stbi__zbuild_huffman(&a->z_length , stbi__zdefault_length , 288)) return 0; - if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) return 0; - } else { - if (!stbi__compute_huffman_codes(a)) return 0; - } - if (!stbi__parse_huffman_block(a)) return 0; + if (!stbi__compute_huffman_codes(a)) return 0; } - } while (!final); - return 1; + if (!stbi__parse_huffman_block(a)) return 0; + } + } while (!final); + return 1; } -static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, int parse_header) -{ - a->zout_start = obuf; - a->zout = obuf; - a->zout_end = obuf + olen; - a->z_expandable = exp; +static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, + int parse_header) { + a->zout_start = obuf; + a->zout = obuf; + a->zout_end = obuf + olen; + a->z_expandable = exp; - return stbi__parse_zlib(a, parse_header); + return stbi__parse_zlib(a, parse_header); } -STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(initial_size); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer + len; - if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, + int initial_size, int *outlen) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) -{ - return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, + int *outlen) { + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); } -STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(initial_size); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer + len; - if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, + int len, + int initial_size, + int *outlen, + int parse_header) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) -{ - stbi__zbuf a; - a.zbuffer = (stbi_uc *) ibuffer; - a.zbuffer_end = (stbi_uc *) ibuffer + ilen; - if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) - return (int) (a.zout - a.zout_start); - else - return -1; +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, + char const *ibuffer, int ilen) { + stbi__zbuf a; + a.zbuffer = (stbi_uc *)ibuffer; + a.zbuffer_end = (stbi_uc *)ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) + return (int)(a.zout - a.zout_start); + else + return -1; } -STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(16384); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer+len; - if (stbi__do_zlib(&a, p, 16384, 1, 0)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, + int *outlen) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) -{ - stbi__zbuf a; - a.zbuffer = (stbi_uc *) ibuffer; - a.zbuffer_end = (stbi_uc *) ibuffer + ilen; - if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) - return (int) (a.zout - a.zout_start); - else - return -1; +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen) { + stbi__zbuf a; + a.zbuffer = (stbi_uc *)ibuffer; + a.zbuffer_end = (stbi_uc *)ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) + return (int)(a.zout - a.zout_start); + else + return -1; } -#endif + #endif // public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 // simple implementation @@ -4470,713 +4825,827 @@ STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char // performance // - uses stb_zlib, a PD zlib implementation with fast huffman decoding -#ifndef STBI_NO_PNG -typedef struct -{ - stbi__uint32 length; - stbi__uint32 type; + #ifndef STBI_NO_PNG +typedef struct { + stbi__uint32 length; + stbi__uint32 type; } stbi__pngchunk; -static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) -{ - stbi__pngchunk c; - c.length = stbi__get32be(s); - c.type = stbi__get32be(s); - return c; +static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) { + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; } -static int stbi__check_png_header(stbi__context *s) -{ - static const stbi_uc png_sig[8] = { 137,80,78,71,13,10,26,10 }; - int i; - for (i=0; i < 8; ++i) - if (stbi__get8(s) != png_sig[i]) return stbi__err("bad png sig","Not a PNG"); - return 1; +static int stbi__check_png_header(stbi__context *s) { + static const stbi_uc png_sig[8] = {137, 80, 78, 71, 13, 10, 26, 10}; + int i; + for (i = 0; i < 8; ++i) + if (stbi__get8(s) != png_sig[i]) + return stbi__err("bad png sig", "Not a PNG"); + return 1; } -typedef struct -{ - stbi__context *s; - stbi_uc *idata, *expanded, *out; - int depth; +typedef struct { + stbi__context *s; + stbi_uc *idata, *expanded, *out; + int depth; } stbi__png; - enum { - STBI__F_none=0, - STBI__F_sub=1, - STBI__F_up=2, - STBI__F_avg=3, - STBI__F_paeth=4, - // synthetic filters used for first scanline to avoid needing a dummy row of 0s - STBI__F_avg_first, - STBI__F_paeth_first + STBI__F_none = 0, + STBI__F_sub = 1, + STBI__F_up = 2, + STBI__F_avg = 3, + STBI__F_paeth = 4, + // synthetic filters used for first scanline to avoid needing a dummy row of + // 0s + STBI__F_avg_first, + STBI__F_paeth_first }; -static stbi_uc first_row_filter[5] = -{ - STBI__F_none, - STBI__F_sub, - STBI__F_none, - STBI__F_avg_first, - STBI__F_paeth_first -}; +static stbi_uc first_row_filter[5] = {STBI__F_none, STBI__F_sub, STBI__F_none, + STBI__F_avg_first, STBI__F_paeth_first}; -static int stbi__paeth(int a, int b, int c) -{ - int p = a + b - c; - int pa = abs(p-a); - int pb = abs(p-b); - int pc = abs(p-c); - if (pa <= pb && pa <= pc) return a; - if (pb <= pc) return b; - return c; +static int stbi__paeth(int a, int b, int c) { + int p = a + b - c; + int pa = abs(p - a); + int pb = abs(p - b); + int pc = abs(p - c); + if (pa <= pb && pa <= pc) return a; + if (pb <= pc) return b; + return c; } -static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; +static const stbi_uc stbi__depth_scale_table[9] = {0, 0xff, 0x55, 0, 0x11, + 0, 0, 0, 0x01}; // create the png data from post-deflated data -static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y, int depth, int color) -{ - int bytes = (depth == 16? 2 : 1); - stbi__context *s = a->s; - stbi__uint32 i,j,stride = x*out_n*bytes; - stbi__uint32 img_len, img_width_bytes; - int k; - int img_n = s->img_n; // copy it into a local for later +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, + stbi__uint32 raw_len, int out_n, + stbi__uint32 x, stbi__uint32 y, int depth, + int color) { + int bytes = (depth == 16 ? 2 : 1); + stbi__context *s = a->s; + stbi__uint32 i, j, stride = x * out_n * bytes; + stbi__uint32 img_len, img_width_bytes; + int k; + int img_n = s->img_n; // copy it into a local for later - int output_bytes = out_n*bytes; - int filter_bytes = img_n*bytes; - int width = x; + int output_bytes = out_n * bytes; + int filter_bytes = img_n * bytes; + int width = x; - STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); - a->out = (stbi_uc *) stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into - if (!a->out) return stbi__err("outofmem", "Out of memory"); + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n + 1); + a->out = (stbi_uc *)stbi__malloc_mad3( + x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); - if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); - img_width_bytes = (((img_n * x * depth) + 7) >> 3); - img_len = (img_width_bytes + 1) * y; + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) + return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; - // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, - // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), - // so just check for raw_len < img_len always. - if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + // we used to check for exact match between raw_len and img_len on + // non-interlaced PNGs, but issue #276 reported a PNG in the wild that had + // extra data at the end (all zeros), so just check for raw_len < img_len + // always. + if (raw_len < img_len) return stbi__err("not enough pixels", "Corrupt PNG"); - for (j=0; j < y; ++j) { - stbi_uc *cur = a->out + stride*j; - stbi_uc *prior; - int filter = *raw++; + for (j = 0; j < y; ++j) { + stbi_uc *cur = a->out + stride * j; + stbi_uc *prior; + int filter = *raw++; - if (filter > 4) - return stbi__err("invalid filter","Corrupt PNG"); + if (filter > 4) return stbi__err("invalid filter", "Corrupt PNG"); - if (depth < 8) { - if (img_width_bytes > x) return stbi__err("invalid width","Corrupt PNG"); - cur += x*out_n - img_width_bytes; // store output to the rightmost img_len bytes, so we can decode in place - filter_bytes = 1; - width = img_width_bytes; + if (depth < 8) { + if (img_width_bytes > x) return stbi__err("invalid width", "Corrupt PNG"); + cur += + x * out_n - img_width_bytes; // store output to the rightmost img_len + // bytes, so we can decode in place + filter_bytes = 1; + width = img_width_bytes; + } + prior = cur - stride; // bugfix: need to compute this after 'cur +=' + // computation above + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // handle first byte explicitly + for (k = 0; k < filter_bytes; ++k) { + switch (filter) { + case STBI__F_none: + cur[k] = raw[k]; + break; + case STBI__F_sub: + cur[k] = raw[k]; + break; + case STBI__F_up: + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + break; + case STBI__F_avg: + cur[k] = STBI__BYTECAST(raw[k] + (prior[k] >> 1)); + break; + case STBI__F_paeth: + cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0, prior[k], 0)); + break; + case STBI__F_avg_first: + cur[k] = raw[k]; + break; + case STBI__F_paeth_first: + cur[k] = raw[k]; + break; } - prior = cur - stride; // bugfix: need to compute this after 'cur +=' computation above + } - // if first row, use special filter that doesn't sample previous row - if (j == 0) filter = first_row_filter[filter]; - - // handle first byte explicitly - for (k=0; k < filter_bytes; ++k) { - switch (filter) { - case STBI__F_none : cur[k] = raw[k]; break; - case STBI__F_sub : cur[k] = raw[k]; break; - case STBI__F_up : cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; - case STBI__F_avg : cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); break; - case STBI__F_paeth : cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0,prior[k],0)); break; - case STBI__F_avg_first : cur[k] = raw[k]; break; - case STBI__F_paeth_first: cur[k] = raw[k]; break; - } + if (depth == 8) { + if (img_n != out_n) cur[img_n] = 255; // first pixel + raw += img_n; + cur += out_n; + prior += out_n; + } else if (depth == 16) { + if (img_n != out_n) { + cur[filter_bytes] = 255; // first pixel top byte + cur[filter_bytes + 1] = 255; // first pixel bottom byte } + raw += filter_bytes; + cur += output_bytes; + prior += output_bytes; + } else { + raw += 1; + cur += 1; + prior += 1; + } - if (depth == 8) { - if (img_n != out_n) - cur[img_n] = 255; // first pixel - raw += img_n; - cur += out_n; - prior += out_n; - } else if (depth == 16) { - if (img_n != out_n) { - cur[filter_bytes] = 255; // first pixel top byte - cur[filter_bytes+1] = 255; // first pixel bottom byte - } - raw += filter_bytes; - cur += output_bytes; - prior += output_bytes; - } else { - raw += 1; - cur += 1; - prior += 1; + // this is a little gross, so that we don't switch per-pixel or + // per-component + if (depth < 8 || img_n == out_n) { + int nk = (width - 1) * filter_bytes; + #define STBI__CASE(f) \ + case f: \ + for (k = 0; k < nk; ++k) + switch (filter) { + // "none" filter turns into a memcpy here; make that explicit. + case STBI__F_none: + memcpy(cur, raw, nk); + break; + STBI__CASE(STBI__F_sub) { + cur[k] = STBI__BYTECAST(raw[k] + cur[k - filter_bytes]); + } + break; + STBI__CASE(STBI__F_up) { + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + } + break; + STBI__CASE(STBI__F_avg) { + cur[k] = STBI__BYTECAST(raw[k] + + ((prior[k] + cur[k - filter_bytes]) >> 1)); + } + break; + STBI__CASE(STBI__F_paeth) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - filter_bytes], prior[k], + prior[k - filter_bytes])); + } + break; + STBI__CASE(STBI__F_avg_first) { + cur[k] = STBI__BYTECAST(raw[k] + (cur[k - filter_bytes] >> 1)); + } + break; + STBI__CASE(STBI__F_paeth_first) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - filter_bytes], 0, 0)); + } + break; } - - // this is a little gross, so that we don't switch per-pixel or per-component - if (depth < 8 || img_n == out_n) { - int nk = (width - 1)*filter_bytes; - #define STBI__CASE(f) \ - case f: \ - for (k=0; k < nk; ++k) - switch (filter) { - // "none" filter turns into a memcpy here; make that explicit. - case STBI__F_none: memcpy(cur, raw, nk); break; - STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); } break; - STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; - STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); } break; - STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],prior[k],prior[k-filter_bytes])); } break; - STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); } break; - STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],0,0)); } break; - } - #undef STBI__CASE - raw += nk; - } else { - STBI_ASSERT(img_n+1 == out_n); - #define STBI__CASE(f) \ - case f: \ - for (i=x-1; i >= 1; --i, cur[filter_bytes]=255,raw+=filter_bytes,cur+=output_bytes,prior+=output_bytes) \ - for (k=0; k < filter_bytes; ++k) - switch (filter) { - STBI__CASE(STBI__F_none) { cur[k] = raw[k]; } break; - STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k- output_bytes]); } break; - STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; - STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k- output_bytes])>>1)); } break; - STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],prior[k],prior[k- output_bytes])); } break; - STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k- output_bytes] >> 1)); } break; - STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],0,0)); } break; - } - #undef STBI__CASE - - // the loop above sets the high byte of the pixels' alpha, but for - // 16 bit png files we also need the low byte set. we'll do that here. - if (depth == 16) { - cur = a->out + stride*j; // start at the beginning of the row again - for (i=0; i < x; ++i,cur+=output_bytes) { - cur[filter_bytes+1] = 255; - } - } + #undef STBI__CASE + raw += nk; + } else { + STBI_ASSERT(img_n + 1 == out_n); + #define STBI__CASE(f) \ + case f: \ + for (i = x - 1; i >= 1; --i, cur[filter_bytes] = 255, \ + raw += filter_bytes, cur += output_bytes, prior += output_bytes) \ + for (k = 0; k < filter_bytes; ++k) + switch (filter) { + STBI__CASE(STBI__F_none) { + cur[k] = raw[k]; + } + break; + STBI__CASE(STBI__F_sub) { + cur[k] = STBI__BYTECAST(raw[k] + cur[k - output_bytes]); + } + break; + STBI__CASE(STBI__F_up) { + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + } + break; + STBI__CASE(STBI__F_avg) { + cur[k] = STBI__BYTECAST(raw[k] + + ((prior[k] + cur[k - output_bytes]) >> 1)); + } + break; + STBI__CASE(STBI__F_paeth) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - output_bytes], prior[k], + prior[k - output_bytes])); + } + break; + STBI__CASE(STBI__F_avg_first) { + cur[k] = STBI__BYTECAST(raw[k] + (cur[k - output_bytes] >> 1)); + } + break; + STBI__CASE(STBI__F_paeth_first) { + cur[k] = + STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - output_bytes], 0, 0)); + } + break; } - } + #undef STBI__CASE - // we make a separate pass to expand bits to pixels; for performance, - // this could run two scanlines behind the above code, so it won't - // intefere with filtering but will still be in the cache. - if (depth < 8) { - for (j=0; j < y; ++j) { - stbi_uc *cur = a->out + stride*j; - stbi_uc *in = a->out + stride*j + x*out_n - img_width_bytes; - // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common 8-bit path optimal at minimal cost for 1/2/4-bit - // png guarante byte alignment, if width is not multiple of 8/4/2 we'll decode dummy trailing data that will be skipped in the later loop - stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range - - // note that the final byte might overshoot and write more data than desired. - // we can allocate enough data that this never writes out of memory, but it - // could also overwrite the next scanline. can it overwrite non-empty data - // on the next scanline? yes, consider 1-pixel-wide scanlines with 1-bit-per-pixel. - // so we need to explicitly clamp the final ones - - if (depth == 4) { - for (k=x*img_n; k >= 2; k-=2, ++in) { - *cur++ = scale * ((*in >> 4) ); - *cur++ = scale * ((*in ) & 0x0f); - } - if (k > 0) *cur++ = scale * ((*in >> 4) ); - } else if (depth == 2) { - for (k=x*img_n; k >= 4; k-=4, ++in) { - *cur++ = scale * ((*in >> 6) ); - *cur++ = scale * ((*in >> 4) & 0x03); - *cur++ = scale * ((*in >> 2) & 0x03); - *cur++ = scale * ((*in ) & 0x03); - } - if (k > 0) *cur++ = scale * ((*in >> 6) ); - if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); - if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); - } else if (depth == 1) { - for (k=x*img_n; k >= 8; k-=8, ++in) { - *cur++ = scale * ((*in >> 7) ); - *cur++ = scale * ((*in >> 6) & 0x01); - *cur++ = scale * ((*in >> 5) & 0x01); - *cur++ = scale * ((*in >> 4) & 0x01); - *cur++ = scale * ((*in >> 3) & 0x01); - *cur++ = scale * ((*in >> 2) & 0x01); - *cur++ = scale * ((*in >> 1) & 0x01); - *cur++ = scale * ((*in ) & 0x01); - } - if (k > 0) *cur++ = scale * ((*in >> 7) ); - if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); - if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); - if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); - if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); - if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); - if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); - } - if (img_n != out_n) { - int q; - // insert alpha = 255 - cur = a->out + stride*j; - if (img_n == 1) { - for (q=x-1; q >= 0; --q) { - cur[q*2+1] = 255; - cur[q*2+0] = cur[q]; - } - } else { - STBI_ASSERT(img_n == 3); - for (q=x-1; q >= 0; --q) { - cur[q*4+3] = 255; - cur[q*4+2] = cur[q*3+2]; - cur[q*4+1] = cur[q*3+1]; - cur[q*4+0] = cur[q*3+0]; - } - } - } + // the loop above sets the high byte of the pixels' alpha, but for + // 16 bit png files we also need the low byte set. we'll do that here. + if (depth == 16) { + cur = a->out + stride * j; // start at the beginning of the row again + for (i = 0; i < x; ++i, cur += output_bytes) { + cur[filter_bytes + 1] = 255; + } } - } else if (depth == 16) { - // force the image data from big-endian to platform-native. - // this is done in a separate pass due to the decoding relying - // on the data being untouched, but could probably be done - // per-line during decode if care is taken. - stbi_uc *cur = a->out; - stbi__uint16 *cur16 = (stbi__uint16*)cur; + } + } - for(i=0; i < x*y*out_n; ++i,cur16++,cur+=2) { - *cur16 = (cur[0] << 8) | cur[1]; + // we make a separate pass to expand bits to pixels; for performance, + // this could run two scanlines behind the above code, so it won't + // intefere with filtering but will still be in the cache. + if (depth < 8) { + for (j = 0; j < y; ++j) { + stbi_uc *cur = a->out + stride * j; + stbi_uc *in = a->out + stride * j + x * out_n - img_width_bytes; + // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common + // 8-bit path optimal at minimal cost for 1/2/4-bit png guarante byte + // alignment, if width is not multiple of 8/4/2 we'll decode dummy + // trailing data that will be skipped in the later loop + stbi_uc scale = (color == 0) + ? stbi__depth_scale_table[depth] + : 1; // scale grayscale values to 0..255 range + + // note that the final byte might overshoot and write more data than + // desired. we can allocate enough data that this never writes out of + // memory, but it could also overwrite the next scanline. can it overwrite + // non-empty data on the next scanline? yes, consider 1-pixel-wide + // scanlines with 1-bit-per-pixel. so we need to explicitly clamp the + // final ones + + if (depth == 4) { + for (k = x * img_n; k >= 2; k -= 2, ++in) { + *cur++ = scale * ((*in >> 4)); + *cur++ = scale * ((*in) & 0x0f); + } + if (k > 0) *cur++ = scale * ((*in >> 4)); + } else if (depth == 2) { + for (k = x * img_n; k >= 4; k -= 4, ++in) { + *cur++ = scale * ((*in >> 6)); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in) & 0x03); + } + if (k > 0) *cur++ = scale * ((*in >> 6)); + if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); + } else if (depth == 1) { + for (k = x * img_n; k >= 8; k -= 8, ++in) { + *cur++ = scale * ((*in >> 7)); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in) & 0x01); + } + if (k > 0) *cur++ = scale * ((*in >> 7)); + if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); } - } + if (img_n != out_n) { + int q; + // insert alpha = 255 + cur = a->out + stride * j; + if (img_n == 1) { + for (q = x - 1; q >= 0; --q) { + cur[q * 2 + 1] = 255; + cur[q * 2 + 0] = cur[q]; + } + } else { + STBI_ASSERT(img_n == 3); + for (q = x - 1; q >= 0; --q) { + cur[q * 4 + 3] = 255; + cur[q * 4 + 2] = cur[q * 3 + 2]; + cur[q * 4 + 1] = cur[q * 3 + 1]; + cur[q * 4 + 0] = cur[q * 3 + 0]; + } + } + } + } + } else if (depth == 16) { + // force the image data from big-endian to platform-native. + // this is done in a separate pass due to the decoding relying + // on the data being untouched, but could probably be done + // per-line during decode if care is taken. + stbi_uc *cur = a->out; + stbi__uint16 *cur16 = (stbi__uint16 *)cur; - return 1; + for (i = 0; i < x * y * out_n; ++i, cur16++, cur += 2) { + *cur16 = (cur[0] << 8) | cur[1]; + } + } + + return 1; } -static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) -{ - int bytes = (depth == 16 ? 2 : 1); - int out_bytes = out_n * bytes; - stbi_uc *final; - int p; - if (!interlaced) - return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); +static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, + stbi__uint32 image_data_len, int out_n, + int depth, int color, int interlaced) { + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, + a->s->img_x, a->s->img_y, depth, color); - // de-interlacing - final = (stbi_uc *) stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); - for (p=0; p < 7; ++p) { - int xorig[] = { 0,4,0,2,0,1,0 }; - int yorig[] = { 0,0,4,0,2,0,1 }; - int xspc[] = { 8,8,4,4,2,2,1 }; - int yspc[] = { 8,8,8,4,4,2,2 }; - int i,j,x,y; - // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 - x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; - y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; - if (x && y) { - stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; - if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { - STBI_FREE(final); - return 0; - } - for (j=0; j < y; ++j) { - for (i=0; i < x; ++i) { - int out_y = j*yspc[p]+yorig[p]; - int out_x = i*xspc[p]+xorig[p]; - memcpy(final + out_y*a->s->img_x*out_bytes + out_x*out_bytes, - a->out + (j*x+i)*out_bytes, out_bytes); - } - } - STBI_FREE(a->out); - image_data += img_len; - image_data_len -= img_len; + // de-interlacing + final = (stbi_uc *)stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + for (p = 0; p < 7; ++p) { + int xorig[] = {0, 4, 0, 2, 0, 1, 0}; + int yorig[] = {0, 0, 4, 0, 2, 0, 1}; + int xspc[] = {8, 8, 4, 4, 2, 2, 1}; + int yspc[] = {8, 8, 8, 4, 4, 2, 2}; + int i, j, x, y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p] - 1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p] - 1) / yspc[p]; + if (x && y) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, + y, depth, color)) { + STBI_FREE(final); + return 0; } - } - a->out = final; + for (j = 0; j < y; ++j) { + for (i = 0; i < x; ++i) { + int out_y = j * yspc[p] + yorig[p]; + int out_x = i * xspc[p] + xorig[p]; + memcpy(final + out_y * a->s->img_x * out_bytes + out_x * out_bytes, + a->out + (j * x + i) * out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; - return 1; + return 1; } -static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi_uc *p = z->out; +static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; - // compute color-based transparency, assuming we've - // already got 255 as the alpha value in the output - STBI_ASSERT(out_n == 2 || out_n == 4); + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); - if (out_n == 2) { - for (i=0; i < pixel_count; ++i) { - p[1] = (p[0] == tc[0] ? 0 : 255); - p += 2; - } - } else { - for (i=0; i < pixel_count; ++i) { - if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) - p[3] = 0; - p += 4; - } - } - return 1; + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 255); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) p[3] = 0; + p += 4; + } + } + return 1; } -static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], int out_n) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi__uint16 *p = (stbi__uint16*) z->out; +static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], + int out_n) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16 *p = (stbi__uint16 *)z->out; - // compute color-based transparency, assuming we've - // already got 65535 as the alpha value in the output - STBI_ASSERT(out_n == 2 || out_n == 4); + // compute color-based transparency, assuming we've + // already got 65535 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); - if (out_n == 2) { - for (i = 0; i < pixel_count; ++i) { - p[1] = (p[0] == tc[0] ? 0 : 65535); - p += 2; - } - } else { - for (i = 0; i < pixel_count; ++i) { - if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) - p[3] = 0; - p += 4; - } - } - return 1; + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 65535); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) p[3] = 0; + p += 4; + } + } + return 1; } -static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, int pal_img_n) -{ - stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; - stbi_uc *p, *temp_out, *orig = a->out; +static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, + int pal_img_n) { + stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; + stbi_uc *p, *temp_out, *orig = a->out; - p = (stbi_uc *) stbi__malloc_mad2(pixel_count, pal_img_n, 0); - if (p == NULL) return stbi__err("outofmem", "Out of memory"); + p = (stbi_uc *)stbi__malloc_mad2(pixel_count, pal_img_n, 0); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); - // between here and free(out) below, exitting would leak - temp_out = p; + // between here and free(out) below, exitting would leak + temp_out = p; - if (pal_img_n == 3) { - for (i=0; i < pixel_count; ++i) { - int n = orig[i]*4; - p[0] = palette[n ]; - p[1] = palette[n+1]; - p[2] = palette[n+2]; - p += 3; - } - } else { - for (i=0; i < pixel_count; ++i) { - int n = orig[i]*4; - p[0] = palette[n ]; - p[1] = palette[n+1]; - p[2] = palette[n+2]; - p[3] = palette[n+3]; - p += 4; - } - } - STBI_FREE(a->out); - a->out = temp_out; + if (pal_img_n == 3) { + for (i = 0; i < pixel_count; ++i) { + int n = orig[i] * 4; + p[0] = palette[n]; + p[1] = palette[n + 1]; + p[2] = palette[n + 2]; + p += 3; + } + } else { + for (i = 0; i < pixel_count; ++i) { + int n = orig[i] * 4; + p[0] = palette[n]; + p[1] = palette[n + 1]; + p[2] = palette[n + 2]; + p[3] = palette[n + 3]; + p += 4; + } + } + STBI_FREE(a->out); + a->out = temp_out; - STBI_NOTUSED(len); + STBI_NOTUSED(len); - return 1; + return 1; } static int stbi__unpremultiply_on_load = 0; static int stbi__de_iphone_flag = 0; -STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) -{ - stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; +STBIDEF void stbi_set_unpremultiply_on_load( + int flag_true_if_should_unpremultiply) { + stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; } -STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) -{ - stbi__de_iphone_flag = flag_true_if_should_convert; +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) { + stbi__de_iphone_flag = flag_true_if_should_convert; } -static void stbi__de_iphone(stbi__png *z) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi_uc *p = z->out; +static void stbi__de_iphone(stbi__png *z) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; - if (s->img_out_n == 3) { // convert bgr to rgb - for (i=0; i < pixel_count; ++i) { - stbi_uc t = p[0]; - p[0] = p[2]; - p[2] = t; - p += 3; + if (s->img_out_n == 3) { // convert bgr to rgb + for (i = 0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } else { + STBI_ASSERT(s->img_out_n == 4); + if (stbi__unpremultiply_on_load) { + // convert bgr to rgb and unpremultiply + for (i = 0; i < pixel_count; ++i) { + stbi_uc a = p[3]; + stbi_uc t = p[0]; + if (a) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = (t * 255 + half) / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; } - } else { - STBI_ASSERT(s->img_out_n == 4); - if (stbi__unpremultiply_on_load) { - // convert bgr to rgb and unpremultiply - for (i=0; i < pixel_count; ++i) { - stbi_uc a = p[3]; - stbi_uc t = p[0]; - if (a) { - stbi_uc half = a / 2; - p[0] = (p[2] * 255 + half) / a; - p[1] = (p[1] * 255 + half) / a; - p[2] = ( t * 255 + half) / a; - } else { - p[0] = p[2]; - p[2] = t; - } - p += 4; - } - } else { - // convert bgr to rgb - for (i=0; i < pixel_count; ++i) { - stbi_uc t = p[0]; - p[0] = p[2]; - p[2] = t; - p += 4; - } + } else { + // convert bgr to rgb + for (i = 0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; } - } + } + } } -#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + #define STBI__PNG_TYPE(a, b, c, d) \ + (((unsigned)(a) << 24) + ((unsigned)(b) << 16) + ((unsigned)(c) << 8) + \ + (unsigned)(d)) -static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) -{ - stbi_uc palette[1024], pal_img_n=0; - stbi_uc has_trans=0, tc[3]={0}; - stbi__uint16 tc16[3]; - stbi__uint32 ioff=0, idata_limit=0, i, pal_len=0; - int first=1,k,interlace=0, color=0, is_iphone=0; - stbi__context *s = z->s; +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) { + stbi_uc palette[1024], pal_img_n = 0; + stbi_uc has_trans = 0, tc[3] = {0}; + stbi__uint16 tc16[3]; + stbi__uint32 ioff = 0, idata_limit = 0, i, pal_len = 0; + int first = 1, k, interlace = 0, color = 0, is_iphone = 0; + stbi__context *s = z->s; - z->expanded = NULL; - z->idata = NULL; - z->out = NULL; + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; - if (!stbi__check_png_header(s)) return 0; + if (!stbi__check_png_header(s)) return 0; - if (scan == STBI__SCAN_type) return 1; + if (scan == STBI__SCAN_type) return 1; - for (;;) { - stbi__pngchunk c = stbi__get_chunk_header(s); - switch (c.type) { - case STBI__PNG_TYPE('C','g','B','I'): - is_iphone = 1; - stbi__skip(s, c.length); - break; - case STBI__PNG_TYPE('I','H','D','R'): { - int comp,filter; - if (!first) return stbi__err("multiple IHDR","Corrupt PNG"); - first = 0; - if (c.length != 13) return stbi__err("bad IHDR len","Corrupt PNG"); - s->img_x = stbi__get32be(s); - s->img_y = stbi__get32be(s); - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only","PNG not supported: 1/2/4/8/16-bit only"); - color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype","Corrupt PNG"); - if (color == 3 && z->depth == 16) return stbi__err("bad ctype","Corrupt PNG"); - if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype","Corrupt PNG"); - comp = stbi__get8(s); if (comp) return stbi__err("bad comp method","Corrupt PNG"); - filter= stbi__get8(s); if (filter) return stbi__err("bad filter method","Corrupt PNG"); - interlace = stbi__get8(s); if (interlace>1) return stbi__err("bad interlace method","Corrupt PNG"); - if (!s->img_x || !s->img_y) return stbi__err("0-pixel image","Corrupt PNG"); - if (!pal_img_n) { - s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); - if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); - if (scan == STBI__SCAN_header) return 1; - } else { - // if paletted, then pal_n is our final components, and - // img_n is # components to decompress/filter. - s->img_n = 1; - if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); - // if SCAN_header, have to scan to see if we have a tRNS - } - break; - } + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C', 'g', 'B', 'I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I', 'H', 'D', 'R'): { + int comp, filter; + if (!first) return stbi__err("multiple IHDR", "Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len", "Corrupt PNG"); + s->img_x = stbi__get32be(s); + s->img_y = stbi__get32be(s); + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + z->depth = stbi__get8(s); + if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && + z->depth != 16) + return stbi__err("1/2/4/8/16-bit only", + "PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); + if (color > 6) return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3 && z->depth == 16) + return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3) + pal_img_n = 3; + else if (color & 1) + return stbi__err("bad ctype", "Corrupt PNG"); + comp = stbi__get8(s); + if (comp) return stbi__err("bad comp method", "Corrupt PNG"); + filter = stbi__get8(s); + if (filter) return stbi__err("bad filter method", "Corrupt PNG"); + interlace = stbi__get8(s); + if (interlace > 1) + return stbi__err("bad interlace method", "Corrupt PNG"); + if (!s->img_x || !s->img_y) + return stbi__err("0-pixel image", "Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) + return stbi__err("too large", "Image too large to decode"); + if (scan == STBI__SCAN_header) return 1; + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) + return stbi__err("too large", "Corrupt PNG"); + // if SCAN_header, have to scan to see if we have a tRNS + } + break; + } - case STBI__PNG_TYPE('P','L','T','E'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (c.length > 256*3) return stbi__err("invalid PLTE","Corrupt PNG"); - pal_len = c.length / 3; - if (pal_len * 3 != c.length) return stbi__err("invalid PLTE","Corrupt PNG"); - for (i=0; i < pal_len; ++i) { - palette[i*4+0] = stbi__get8(s); - palette[i*4+1] = stbi__get8(s); - palette[i*4+2] = stbi__get8(s); - palette[i*4+3] = 255; - } - break; - } + case STBI__PNG_TYPE('P', 'L', 'T', 'E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256 * 3) return stbi__err("invalid PLTE", "Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) + return stbi__err("invalid PLTE", "Corrupt PNG"); + for (i = 0; i < pal_len; ++i) { + palette[i * 4 + 0] = stbi__get8(s); + palette[i * 4 + 1] = stbi__get8(s); + palette[i * 4 + 2] = stbi__get8(s); + palette[i * 4 + 3] = 255; + } + break; + } - case STBI__PNG_TYPE('t','R','N','S'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (z->idata) return stbi__err("tRNS after IDAT","Corrupt PNG"); - if (pal_img_n) { - if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } - if (pal_len == 0) return stbi__err("tRNS before PLTE","Corrupt PNG"); - if (c.length > pal_len) return stbi__err("bad tRNS len","Corrupt PNG"); - pal_img_n = 4; - for (i=0; i < c.length; ++i) - palette[i*4+3] = stbi__get8(s); - } else { - if (!(s->img_n & 1)) return stbi__err("tRNS with alpha","Corrupt PNG"); - if (c.length != (stbi__uint32) s->img_n*2) return stbi__err("bad tRNS len","Corrupt PNG"); - has_trans = 1; - if (z->depth == 16) { - for (k = 0; k < s->img_n; ++k) tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is - } else { - for (k = 0; k < s->img_n; ++k) tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger - } - } - break; - } - - case STBI__PNG_TYPE('I','D','A','T'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); - if (scan == STBI__SCAN_header) { s->img_n = pal_img_n; return 1; } - if ((int)(ioff + c.length) < (int)ioff) return 0; - if (ioff + c.length > idata_limit) { - stbi__uint32 idata_limit_old = idata_limit; - stbi_uc *p; - if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; - while (ioff + c.length > idata_limit) - idata_limit *= 2; - STBI_NOTUSED(idata_limit_old); - p = (stbi_uc *) STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); - z->idata = p; - } - if (!stbi__getn(s, z->idata+ioff,c.length)) return stbi__err("outofdata","Corrupt PNG"); - ioff += c.length; - break; - } - - case STBI__PNG_TYPE('I','E','N','D'): { - stbi__uint32 raw_len, bpl; - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (scan != STBI__SCAN_load) return 1; - if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); - // initial guess for decoded data size to avoid unnecessary reallocs - bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component - raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; - z->expanded = (stbi_uc *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, raw_len, (int *) &raw_len, !is_iphone); - if (z->expanded == NULL) return 0; // zlib should set error - STBI_FREE(z->idata); z->idata = NULL; - if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) - s->img_out_n = s->img_n+1; - else - s->img_out_n = s->img_n; - if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, z->depth, color, interlace)) return 0; - if (has_trans) { - if (z->depth == 16) { - if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; - } else { - if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; - } - } - if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) - stbi__de_iphone(z); - if (pal_img_n) { - // pal_img_n == 3 or 4 - s->img_n = pal_img_n; // record the actual colors we had - s->img_out_n = pal_img_n; - if (req_comp >= 3) s->img_out_n = req_comp; - if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) - return 0; - } else if (has_trans) { - // non-paletted image with tRNS -> source image has (constant) alpha - ++s->img_n; - } - STBI_FREE(z->expanded); z->expanded = NULL; - // end of PNG chunk, read and skip CRC - stbi__get32be(s); + case STBI__PNG_TYPE('t', 'R', 'N', 'S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT", "Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { + s->img_n = 4; return 1; - } - - default: - // if critical, fail - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if ((c.type & (1 << 29)) == 0) { - #ifndef STBI_NO_FAILURE_STRINGS - // not threadsafe - static char invalid_chunk[] = "XXXX PNG chunk not known"; - invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); - invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); - invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); - invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); - #endif - return stbi__err(invalid_chunk, "PNG not supported: unknown PNG chunk type"); - } - stbi__skip(s, c.length); - break; + } + if (pal_len == 0) return stbi__err("tRNS before PLTE", "Corrupt PNG"); + if (c.length > pal_len) + return stbi__err("bad tRNS len", "Corrupt PNG"); + pal_img_n = 4; + for (i = 0; i < c.length; ++i) + palette[i * 4 + 3] = stbi__get8(s); + } else { + if (!(s->img_n & 1)) + return stbi__err("tRNS with alpha", "Corrupt PNG"); + if (c.length != (stbi__uint32)s->img_n * 2) + return stbi__err("bad tRNS len", "Corrupt PNG"); + has_trans = 1; + if (z->depth == 16) { + for (k = 0; k < s->img_n; ++k) + tc16[k] = + (stbi__uint16)stbi__get16be(s); // copy the values as-is + } else { + for (k = 0; k < s->img_n; ++k) + tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * + stbi__depth_scale_table[z->depth]; // non 8-bit images + // will be larger + } + } + break; } - // end of PNG chunk, read and skip CRC - stbi__get32be(s); - } + + case STBI__PNG_TYPE('I', 'D', 'A', 'T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE", "Corrupt PNG"); + if (scan == STBI__SCAN_header) { + s->img_n = pal_img_n; + return 1; + } + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = idata_limit; + stbi_uc *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc *)STBI_REALLOC_SIZED(z->idata, idata_limit_old, + idata_limit); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata + ioff, c.length)) + return stbi__err("outofdata", "Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I', 'E', 'N', 'D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT", "Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc *)stbi_zlib_decode_malloc_guesssize_headerflag( + (char *)z->idata, ioff, raw_len, (int *)&raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_FREE(z->idata); + z->idata = NULL; + if ((req_comp == s->img_n + 1 && req_comp != 3 && !pal_img_n) || + has_trans) + s->img_out_n = s->img_n + 1; + else + s->img_out_n = s->img_n; + if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, + z->depth, color, interlace)) + return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } else { + if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; + } + } + if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) + stbi__de_iphone(z); + if (pal_img_n) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_FREE(z->expanded); + z->expanded = NULL; + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + return 1; + } + + default: + // if critical, fail + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + static char invalid_chunk[] = "XXXX PNG chunk not known"; + invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); + invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); + invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); + invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); + #endif + return stbi__err(invalid_chunk, + "PNG not supported: unknown PNG chunk type"); + } + stbi__skip(s, c.length); + break; + } + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + } } -static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, stbi__result_info *ri) -{ - void *result=NULL; - if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); - if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { - if (p->depth <= 8) - ri->bits_per_channel = 8; - else if (p->depth == 16) - ri->bits_per_channel = 16; +static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, + stbi__result_info *ri) { + void *result = NULL; + if (req_comp < 0 || req_comp > 4) + return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth <= 8) + ri->bits_per_channel = 8; + else if (p->depth == 16) + ri->bits_per_channel = 16; + else + return stbi__errpuc("bad bits_per_channel", + "PNG not supported: unsupported color depth"); + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char *)result, p->s->img_out_n, + req_comp, p->s->img_x, p->s->img_y); else - return stbi__errpuc("bad bits_per_channel", "PNG not supported: unsupported color depth"); - result = p->out; - p->out = NULL; - if (req_comp && req_comp != p->s->img_out_n) { - if (ri->bits_per_channel == 8) - result = stbi__convert_format((unsigned char *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); - else - result = stbi__convert_format16((stbi__uint16 *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); - p->s->img_out_n = req_comp; - if (result == NULL) return result; - } - *x = p->s->img_x; - *y = p->s->img_y; - if (n) *n = p->s->img_n; - } - STBI_FREE(p->out); p->out = NULL; - STBI_FREE(p->expanded); p->expanded = NULL; - STBI_FREE(p->idata); p->idata = NULL; + result = stbi__convert_format16((stbi__uint16 *)result, p->s->img_out_n, + req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n) *n = p->s->img_n; + } + STBI_FREE(p->out); + p->out = NULL; + STBI_FREE(p->expanded); + p->expanded = NULL; + STBI_FREE(p->idata); + p->idata = NULL; - return result; + return result; } -static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi__png p; - p.s = s; - return stbi__do_png(&p, x,y,comp,req_comp, ri); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi__png p; + p.s = s; + return stbi__do_png(&p, x, y, comp, req_comp, ri); } -static int stbi__png_test(stbi__context *s) -{ - int r; - r = stbi__check_png_header(s); - stbi__rewind(s); - return r; +static int stbi__png_test(stbi__context *s) { + int r; + r = stbi__check_png_header(s); + stbi__rewind(s); + return r; } -static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) -{ - if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { - stbi__rewind( p->s ); - return 0; - } - if (x) *x = p->s->img_x; - if (y) *y = p->s->img_y; - if (comp) *comp = p->s->img_n; - return 1; +static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) { + if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { + stbi__rewind(p->s); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; } -static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) -{ - stbi__png p; - p.s = s; - return stbi__png_info_raw(&p, x, y, comp); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) { + stbi__png p; + p.s = s; + return stbi__png_info_raw(&p, x, y, comp); } static int stbi__png_is16(stbi__context *s) @@ -5191,958 +5660,1015 @@ static int stbi__png_is16(stbi__context *s) } return 1; } -#endif + #endif // Microsoft/Windows BMP image -#ifndef STBI_NO_BMP -static int stbi__bmp_test_raw(stbi__context *s) -{ - int r; - int sz; - if (stbi__get8(s) != 'B') return 0; - if (stbi__get8(s) != 'M') return 0; - stbi__get32le(s); // discard filesize - stbi__get16le(s); // discard reserved - stbi__get16le(s); // discard reserved - stbi__get32le(s); // discard data offset - sz = stbi__get32le(s); - r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); - return r; + #ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context *s) { + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + stbi__get32le(s); // discard data offset + sz = stbi__get32le(s); + r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); + return r; } -static int stbi__bmp_test(stbi__context *s) -{ - int r = stbi__bmp_test_raw(s); - stbi__rewind(s); - return r; +static int stbi__bmp_test(stbi__context *s) { + int r = stbi__bmp_test_raw(s); + stbi__rewind(s); + return r; } - // returns 0..31 for the highest set bit -static int stbi__high_bit(unsigned int z) -{ - int n=0; - if (z == 0) return -1; - if (z >= 0x10000) { n += 16; z >>= 16; } - if (z >= 0x00100) { n += 8; z >>= 8; } - if (z >= 0x00010) { n += 4; z >>= 4; } - if (z >= 0x00004) { n += 2; z >>= 2; } - if (z >= 0x00002) { n += 1;/* >>= 1;*/ } - return n; +static int stbi__high_bit(unsigned int z) { + int n = 0; + if (z == 0) return -1; + if (z >= 0x10000) { + n += 16; + z >>= 16; + } + if (z >= 0x00100) { + n += 8; + z >>= 8; + } + if (z >= 0x00010) { + n += 4; + z >>= 4; + } + if (z >= 0x00004) { + n += 2; + z >>= 2; + } + if (z >= 0x00002) { n += 1; /* >>= 1;*/ } + return n; } -static int stbi__bitcount(unsigned int a) -{ - a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 - a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 - a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits - a = (a + (a >> 8)); // max 16 per 8 bits - a = (a + (a >> 16)); // max 32 per 8 bits - return a & 0xff; +static int stbi__bitcount(unsigned int a) { + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; } // extract an arbitrarily-aligned N-bit value (N=bits) // from v, and then make it 8-bits long and fractionally // extend it to full full range. -static int stbi__shiftsigned(unsigned int v, int shift, int bits) -{ - static unsigned int mul_table[9] = { +static int stbi__shiftsigned(unsigned int v, int shift, int bits) { + static unsigned int mul_table[9] = { 0, - 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, - 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, - }; - static unsigned int shift_table[9] = { - 0, 0,0,1,0,2,4,6,0, - }; - if (shift < 0) - v <<= -shift; - else - v >>= shift; - STBI_ASSERT(v < 256); - v >>= (8-bits); - STBI_ASSERT(bits >= 0 && bits <= 8); - return (int) ((unsigned) v * mul_table[bits]) >> shift_table[bits]; + 0xff /*0b11111111*/, + 0x55 /*0b01010101*/, + 0x49 /*0b01001001*/, + 0x11 /*0b00010001*/, + 0x21 /*0b00100001*/, + 0x41 /*0b01000001*/, + 0x81 /*0b10000001*/, + 0x01 /*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0, 0, 1, 0, 2, 4, 6, 0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v < 256); + v >>= (8 - bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int)((unsigned)v * mul_table[bits]) >> shift_table[bits]; } -typedef struct -{ - int bpp, offset, hsz; - unsigned int mr,mg,mb,ma, all_a; - int extra_read; +typedef struct { + int bpp, offset, hsz; + unsigned int mr, mg, mb, ma, all_a; + int extra_read; } stbi__bmp_data; -static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) -{ - int hsz; - if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') return stbi__errpuc("not BMP", "Corrupt BMP"); - stbi__get32le(s); // discard filesize - stbi__get16le(s); // discard reserved - stbi__get16le(s); // discard reserved - info->offset = stbi__get32le(s); - info->hsz = hsz = stbi__get32le(s); - info->mr = info->mg = info->mb = info->ma = 0; - info->extra_read = 14; +static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) { + int hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') + return stbi__errpuc("not BMP", "Corrupt BMP"); + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + info->extra_read = 14; - if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); + if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); - if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); - if (hsz == 12) { - s->img_x = stbi__get16le(s); - s->img_y = stbi__get16le(s); - } else { - s->img_x = stbi__get32le(s); - s->img_y = stbi__get32le(s); - } - if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); - info->bpp = stbi__get16le(s); - if (hsz != 12) { - int compress = stbi__get32le(s); - if (compress == 1 || compress == 2) return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); - stbi__get32le(s); // discard sizeof - stbi__get32le(s); // discard hres - stbi__get32le(s); // discard vres - stbi__get32le(s); // discard colorsused - stbi__get32le(s); // discard max important - if (hsz == 40 || hsz == 56) { - if (hsz == 56) { - stbi__get32le(s); - stbi__get32le(s); - stbi__get32le(s); - stbi__get32le(s); - } - if (info->bpp == 16 || info->bpp == 32) { - if (compress == 0) { - if (info->bpp == 32) { - info->mr = 0xffu << 16; - info->mg = 0xffu << 8; - info->mb = 0xffu << 0; - info->ma = 0xffu << 24; - info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 - } else { - info->mr = 31u << 10; - info->mg = 31u << 5; - info->mb = 31u << 0; - } - } else if (compress == 3) { - info->mr = stbi__get32le(s); - info->mg = stbi__get32le(s); - info->mb = stbi__get32le(s); - info->extra_read += 12; - // not documented, but generated by photoshop and handled by mspaint - if (info->mr == info->mg && info->mg == info->mb) { - // ?!?!? - return stbi__errpuc("bad BMP", "bad BMP"); - } - } else - return stbi__errpuc("bad BMP", "bad BMP"); - } - } else { - int i; - if (hsz != 108 && hsz != 124) + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) + return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = stbi__get16le(s); + s->img_y = stbi__get16le(s); + } else { + s->img_x = stbi__get32le(s); + s->img_y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int compress = stbi__get32le(s); + if (compress == 1 || compress == 2) + return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); + stbi__get32le(s); // discard sizeof + stbi__get32le(s); // discard hres + stbi__get32le(s); // discard vres + stbi__get32le(s); // discard colorsused + stbi__get32le(s); // discard max important + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + } + if (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha + // channel but it was all 0 + } else { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } + } else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->extra_read += 12; + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? return stbi__errpuc("bad BMP", "bad BMP"); - info->mr = stbi__get32le(s); - info->mg = stbi__get32le(s); - info->mb = stbi__get32le(s); - info->ma = stbi__get32le(s); - stbi__get32le(s); // discard color space - for (i=0; i < 12; ++i) - stbi__get32le(s); // discard color space parameters - if (hsz == 124) { - stbi__get32le(s); // discard rendering intent - stbi__get32le(s); // discard offset of profile data - stbi__get32le(s); // discard size of profile data - stbi__get32le(s); // discard reserved - } + } + } else + return stbi__errpuc("bad BMP", "bad BMP"); } - } - return (void *) 1; -} - - -static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *out; - unsigned int mr=0,mg=0,mb=0,ma=0, all_a; - stbi_uc pal[256][4]; - int psize=0,i,j,width; - int flip_vertically, pad, target; - stbi__bmp_data info; - STBI_NOTUSED(ri); - - info.all_a = 255; - if (stbi__bmp_parse_header(s, &info) == NULL) - return NULL; // error code already set - - flip_vertically = ((int) s->img_y) > 0; - s->img_y = abs((int) s->img_y); - - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - - mr = info.mr; - mg = info.mg; - mb = info.mb; - ma = info.ma; - all_a = info.all_a; - - if (info.hsz == 12) { - if (info.bpp < 24) - psize = (info.offset - info.extra_read - 24) / 3; - } else { - if (info.bpp < 16) - psize = (info.offset - info.extra_read - info.hsz) >> 2; - } - if (psize == 0) { - STBI_ASSERT(info.offset == s->callback_already_read + (int) (s->img_buffer - s->img_buffer_original)); - if (info.offset != s->callback_already_read + (s->img_buffer - s->buffer_start)) { - return stbi__errpuc("bad offset", "Corrupt BMP"); - } - } - - if (info.bpp == 24 && ma == 0xff000000) - s->img_n = 3; - else - s->img_n = ma ? 4 : 3; - if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 - target = req_comp; - else - target = s->img_n; // if they want monochrome, we'll post-convert - - // sanity-check size - if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) - return stbi__errpuc("too large", "Corrupt BMP"); - - out = (stbi_uc *) stbi__malloc_mad3(target, s->img_x, s->img_y, 0); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - if (info.bpp < 16) { - int z=0; - if (psize == 0 || psize > 256) { STBI_FREE(out); return stbi__errpuc("invalid", "Corrupt BMP"); } - for (i=0; i < psize; ++i) { - pal[i][2] = stbi__get8(s); - pal[i][1] = stbi__get8(s); - pal[i][0] = stbi__get8(s); - if (info.hsz != 12) stbi__get8(s); - pal[i][3] = 255; - } - stbi__skip(s, info.offset - info.extra_read - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); - if (info.bpp == 1) width = (s->img_x + 7) >> 3; - else if (info.bpp == 4) width = (s->img_x + 1) >> 1; - else if (info.bpp == 8) width = s->img_x; - else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } - pad = (-width)&3; - if (info.bpp == 1) { - for (j=0; j < (int) s->img_y; ++j) { - int bit_offset = 7, v = stbi__get8(s); - for (i=0; i < (int) s->img_x; ++i) { - int color = (v>>bit_offset)&0x1; - out[z++] = pal[color][0]; - out[z++] = pal[color][1]; - out[z++] = pal[color][2]; - if (target == 4) out[z++] = 255; - if (i+1 == (int) s->img_x) break; - if((--bit_offset) < 0) { - bit_offset = 7; - v = stbi__get8(s); - } - } - stbi__skip(s, pad); - } - } else { - for (j=0; j < (int) s->img_y; ++j) { - for (i=0; i < (int) s->img_x; i += 2) { - int v=stbi__get8(s),v2=0; - if (info.bpp == 4) { - v2 = v & 15; - v >>= 4; - } - out[z++] = pal[v][0]; - out[z++] = pal[v][1]; - out[z++] = pal[v][2]; - if (target == 4) out[z++] = 255; - if (i+1 == (int) s->img_x) break; - v = (info.bpp == 8) ? stbi__get8(s) : v2; - out[z++] = pal[v][0]; - out[z++] = pal[v][1]; - out[z++] = pal[v][2]; - if (target == 4) out[z++] = 255; - } - stbi__skip(s, pad); - } - } - } else { - int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0; - int z = 0; - int easy=0; - stbi__skip(s, info.offset - info.extra_read - info.hsz); - if (info.bpp == 24) width = 3 * s->img_x; - else if (info.bpp == 16) width = 2*s->img_x; - else /* bpp = 32 and pad = 0 */ width=0; - pad = (-width) & 3; - if (info.bpp == 24) { - easy = 1; - } else if (info.bpp == 32) { - if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) - easy = 2; - } - if (!easy) { - if (!mr || !mg || !mb) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } - // right shift amt to put high bit in position #7 - rshift = stbi__high_bit(mr)-7; rcount = stbi__bitcount(mr); - gshift = stbi__high_bit(mg)-7; gcount = stbi__bitcount(mg); - bshift = stbi__high_bit(mb)-7; bcount = stbi__bitcount(mb); - ashift = stbi__high_bit(ma)-7; acount = stbi__bitcount(ma); - if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } - } - for (j=0; j < (int) s->img_y; ++j) { - if (easy) { - for (i=0; i < (int) s->img_x; ++i) { - unsigned char a; - out[z+2] = stbi__get8(s); - out[z+1] = stbi__get8(s); - out[z+0] = stbi__get8(s); - z += 3; - a = (easy == 2 ? stbi__get8(s) : 255); - all_a |= a; - if (target == 4) out[z++] = a; - } - } else { - int bpp = info.bpp; - for (i=0; i < (int) s->img_x; ++i) { - stbi__uint32 v = (bpp == 16 ? (stbi__uint32) stbi__get16le(s) : stbi__get32le(s)); - unsigned int a; - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); - a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); - all_a |= a; - if (target == 4) out[z++] = STBI__BYTECAST(a); - } - } - stbi__skip(s, pad); - } - } - - // if alpha channel is all 0s, replace with all 255s - if (target == 4 && all_a == 0) - for (i=4*s->img_x*s->img_y-1; i >= 0; i -= 4) - out[i] = 255; - - if (flip_vertically) { - stbi_uc t; - for (j=0; j < (int) s->img_y>>1; ++j) { - stbi_uc *p1 = out + j *s->img_x*target; - stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; - for (i=0; i < (int) s->img_x*target; ++i) { - t = p1[i]; p1[i] = p2[i]; p2[i] = t; - } - } - } - - if (req_comp && req_comp != target) { - out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } - - *x = s->img_x; - *y = s->img_y; - if (comp) *comp = s->img_n; - return out; -} -#endif - -// Targa Truevision - TGA -// by Jonathan Dummer -#ifndef STBI_NO_TGA -// returns STBI_rgb or whatever, 0 on error -static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) -{ - // only RGB or RGBA (incl. 16bit) or grey allowed - if (is_rgb16) *is_rgb16 = 0; - switch(bits_per_pixel) { - case 8: return STBI_grey; - case 16: if(is_grey) return STBI_grey_alpha; - // fallthrough - case 15: if(is_rgb16) *is_rgb16 = 1; - return STBI_rgb; - case 24: // fallthrough - case 32: return bits_per_pixel/8; - default: return 0; - } -} - -static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) -{ - int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; - int sz, tga_colormap_type; - stbi__get8(s); // discard Offset - tga_colormap_type = stbi__get8(s); // colormap type - if( tga_colormap_type > 1 ) { - stbi__rewind(s); - return 0; // only RGB or indexed allowed - } - tga_image_type = stbi__get8(s); // image type - if ( tga_colormap_type == 1 ) { // colormapped (paletted) image - if (tga_image_type != 1 && tga_image_type != 9) { - stbi__rewind(s); - return 0; - } - stbi__skip(s,4); // skip index of first colormap entry and number of entries - sz = stbi__get8(s); // check bits per palette color entry - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) { - stbi__rewind(s); - return 0; - } - stbi__skip(s,4); // skip image x and y origin - tga_colormap_bpp = sz; - } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE - if ( (tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11) ) { - stbi__rewind(s); - return 0; // only RGB or grey allowed, +/- RLE - } - stbi__skip(s,9); // skip colormap specification and image x/y origin - tga_colormap_bpp = 0; - } - tga_w = stbi__get16le(s); - if( tga_w < 1 ) { - stbi__rewind(s); - return 0; // test width - } - tga_h = stbi__get16le(s); - if( tga_h < 1 ) { - stbi__rewind(s); - return 0; // test height - } - tga_bits_per_pixel = stbi__get8(s); // bits per pixel - stbi__get8(s); // ignore alpha bits - if (tga_colormap_bpp != 0) { - if((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { - // when using a colormap, tga_bits_per_pixel is the size of the indexes - // I don't think anything but 8 or 16bit indexes makes sense - stbi__rewind(s); - return 0; - } - tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); } else { - tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + int i; + if (hsz != 108 && hsz != 124) return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->ma = stbi__get32le(s); + stbi__get32le(s); // discard color space + for (i = 0; i < 12; ++i) + stbi__get32le(s); // discard color space parameters + if (hsz == 124) { + stbi__get32le(s); // discard rendering intent + stbi__get32le(s); // discard offset of profile data + stbi__get32le(s); // discard size of profile data + stbi__get32le(s); // discard reserved + } } - if(!tga_comp) { + } + return (void *)1; +} + +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *out; + unsigned int mr = 0, mg = 0, mb = 0, ma = 0, all_a; + stbi_uc pal[256][4]; + int psize = 0, i, j, width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int)s->img_y) > 0; + s->img_y = abs((int)s->img_y); + + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) psize = (info.offset - info.extra_read - 24) / 3; + } else { + if (info.bpp < 16) psize = (info.offset - info.extra_read - info.hsz) >> 2; + } + if (psize == 0) { + STBI_ASSERT(info.offset == + s->callback_already_read + + (int)(s->img_buffer - s->img_buffer_original)); + if (info.offset != + s->callback_already_read + (s->img_buffer - s->buffer_start)) { + return stbi__errpuc("bad offset", "Corrupt BMP"); + } + } + + if (info.bpp == 24 && ma == 0xff000000) + s->img_n = 3; + else + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc *)stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z = 0; + if (psize == 0 || psize > 256) { + STBI_FREE(out); + return stbi__errpuc("invalid", "Corrupt BMP"); + } + for (i = 0; i < psize; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + if (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - info.extra_read - info.hsz - + psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) + width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) + width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) + width = s->img_x; + else { + STBI_FREE(out); + return stbi__errpuc("bad bpp", "Corrupt BMP"); + } + pad = (-width) & 3; + if (info.bpp == 1) { + for (j = 0; j < (int)s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i = 0; i < (int)s->img_x; ++i) { + int color = (v >> bit_offset) & 0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if (target == 4) out[z++] = 255; + if (i + 1 == (int)s->img_x) break; + if ((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } else { + for (j = 0; j < (int)s->img_y; ++j) { + for (i = 0; i < (int)s->img_x; i += 2) { + int v = stbi__get8(s), v2 = 0; + if (info.bpp == 4) { + v2 = v & 15; + v >>= 4; + } + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + if (i + 1 == (int)s->img_x) break; + v = (info.bpp == 8) ? stbi__get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + stbi__skip(s, pad); + } + } + } else { + int rshift = 0, gshift = 0, bshift = 0, ashift = 0, rcount = 0, gcount = 0, + bcount = 0, acount = 0; + int z = 0; + int easy = 0; + stbi__skip(s, info.offset - info.extra_read - info.hsz); + if (info.bpp == 24) + width = 3 * s->img_x; + else if (info.bpp == 16) + width = 2 * s->img_x; + else /* bpp = 32 and pad = 0 */ + width = 0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { + STBI_FREE(out); + return stbi__errpuc("bad masks", "Corrupt BMP"); + } + // right shift amt to put high bit in position #7 + rshift = stbi__high_bit(mr) - 7; + rcount = stbi__bitcount(mr); + gshift = stbi__high_bit(mg) - 7; + gcount = stbi__bitcount(mg); + bshift = stbi__high_bit(mb) - 7; + bcount = stbi__bitcount(mb); + ashift = stbi__high_bit(ma) - 7; + acount = stbi__bitcount(ma); + if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { + STBI_FREE(out); + return stbi__errpuc("bad masks", "Corrupt BMP"); + } + } + for (j = 0; j < (int)s->img_y; ++j) { + if (easy) { + for (i = 0; i < (int)s->img_x; ++i) { + unsigned char a; + out[z + 2] = stbi__get8(s); + out[z + 1] = stbi__get8(s); + out[z + 0] = stbi__get8(s); + z += 3; + a = (easy == 2 ? stbi__get8(s) : 255); + all_a |= a; + if (target == 4) out[z++] = a; + } + } else { + int bpp = info.bpp; + for (i = 0; i < (int)s->img_x; ++i) { + stbi__uint32 v = + (bpp == 16 ? (stbi__uint32)stbi__get16le(s) : stbi__get32le(s)); + unsigned int a; + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); + a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i = 4 * s->img_x * s->img_y - 1; i >= 0; i -= 4) + out[i] = 255; + + if (flip_vertically) { + stbi_uc t; + for (j = 0; j < (int)s->img_y >> 1; ++j) { + stbi_uc *p1 = out + j * s->img_x * target; + stbi_uc *p2 = out + (s->img_y - 1 - j) * s->img_x * target; + for (i = 0; i < (int)s->img_x * target; ++i) { + t = p1[i]; + p1[i] = p2[i]; + p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} + #endif + + // Targa Truevision - TGA + // by Jonathan Dummer + #ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int *is_rgb16) { + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16) *is_rgb16 = 0; + switch (bits_per_pixel) { + case 8: + return STBI_grey; + case 16: + if (is_grey) return STBI_grey_alpha; + // fallthrough + case 15: + if (is_rgb16) *is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: + return bits_per_pixel / 8; + default: + return 0; + } +} + +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) { + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, + tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if (tga_colormap_type > 1) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if (tga_colormap_type == 1) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { stbi__rewind(s); return 0; } - if (x) *x = tga_w; - if (y) *y = tga_h; - if (comp) *comp = tga_comp; - return 1; // seems to have passed everything + stbi__skip(s, + 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 4); // skip image x and y origin + tga_colormap_bpp = sz; + } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ((tga_image_type != 2) && (tga_image_type != 3) && + (tga_image_type != 10) && (tga_image_type != 11)) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s, 9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + tga_w = stbi__get16le(s); + if (tga_w < 1) { + stbi__rewind(s); + return 0; // test width + } + tga_h = stbi__get16le(s); + if (tga_h < 1) { + stbi__rewind(s); + return 0; // test height + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } else { + tga_comp = stbi__tga_get_comp( + tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), + NULL); + } + if (!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp; + return 1; // seems to have passed everything } -static int stbi__tga_test(stbi__context *s) -{ - int res = 0; - int sz, tga_color_type; - stbi__get8(s); // discard Offset - tga_color_type = stbi__get8(s); // color type - if ( tga_color_type > 1 ) goto errorEnd; // only RGB or indexed allowed - sz = stbi__get8(s); // image type - if ( tga_color_type == 1 ) { // colormapped (paletted) image - if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 - stbi__skip(s,4); // skip index of first colormap entry and number of entries - sz = stbi__get8(s); // check bits per palette color entry - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; - stbi__skip(s,4); // skip image x and y origin - } else { // "normal" image w/o colormap - if ( (sz != 2) && (sz != 3) && (sz != 10) && (sz != 11) ) goto errorEnd; // only RGB or grey allowed, +/- RLE - stbi__skip(s,9); // skip colormap specification and image x/y origin - } - if ( stbi__get16le(s) < 1 ) goto errorEnd; // test width - if ( stbi__get16le(s) < 1 ) goto errorEnd; // test height - sz = stbi__get8(s); // bits per pixel - if ( (tga_color_type == 1) && (sz != 8) && (sz != 16) ) goto errorEnd; // for colormapped images, bpp is size of an index - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; +static int stbi__tga_test(stbi__context *s) { + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if (tga_color_type > 1) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if (tga_color_type == 1) { // colormapped (paletted) image + if (sz != 1 && sz != 9) + goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s, + 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) + goto errorEnd; + stbi__skip(s, 4); // skip image x and y origin + } else { // "normal" image w/o colormap + if ((sz != 2) && (sz != 3) && (sz != 10) && (sz != 11)) + goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s, 9); // skip colormap specification and image x/y origin + } + if (stbi__get16le(s) < 1) goto errorEnd; // test width + if (stbi__get16le(s) < 1) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ((tga_color_type == 1) && (sz != 8) && (sz != 16)) + goto errorEnd; // for colormapped images, bpp is size of an index + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) + goto errorEnd; - res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + res = 1; // if we got this far, everything's good and we can return 1 instead + // of 0 errorEnd: - stbi__rewind(s); - return res; + stbi__rewind(s); + return res; } // read 16bit value and convert to 24bit RGB -static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc* out) -{ - stbi__uint16 px = (stbi__uint16)stbi__get16le(s); - stbi__uint16 fiveBitMask = 31; - // we have 3 channels with 5bits each - int r = (px >> 10) & fiveBitMask; - int g = (px >> 5) & fiveBitMask; - int b = px & fiveBitMask; - // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later - out[0] = (stbi_uc)((r * 255)/31); - out[1] = (stbi_uc)((g * 255)/31); - out[2] = (stbi_uc)((b * 255)/31); +static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc *out) { + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be + // swapped later + out[0] = (stbi_uc)((r * 255) / 31); + out[1] = (stbi_uc)((g * 255) / 31); + out[2] = (stbi_uc)((b * 255) / 31); - // some people claim that the most significant bit might be used for alpha - // (possibly if an alpha-bit is set in the "image descriptor byte") - // but that only made 16bit test images completely translucent.. - // so let's treat all 15 and 16bit TGAs as RGB with no alpha. + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. } -static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - // read in the TGA header stuff - int tga_offset = stbi__get8(s); - int tga_indexed = stbi__get8(s); - int tga_image_type = stbi__get8(s); - int tga_is_RLE = 0; - int tga_palette_start = stbi__get16le(s); - int tga_palette_len = stbi__get16le(s); - int tga_palette_bits = stbi__get8(s); - int tga_x_origin = stbi__get16le(s); - int tga_y_origin = stbi__get16le(s); - int tga_width = stbi__get16le(s); - int tga_height = stbi__get16le(s); - int tga_bits_per_pixel = stbi__get8(s); - int tga_comp, tga_rgb16=0; - int tga_inverted = stbi__get8(s); - // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) - // image data - unsigned char *tga_data; - unsigned char *tga_palette = NULL; - int i, j; - unsigned char raw_data[4] = {0}; - int RLE_count = 0; - int RLE_repeating = 0; - int read_next_pixel = 1; - STBI_NOTUSED(ri); - STBI_NOTUSED(tga_x_origin); // @TODO - STBI_NOTUSED(tga_y_origin); // @TODO +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + // read in the TGA header stuff + int tga_offset = stbi__get8(s); + int tga_indexed = stbi__get8(s); + int tga_image_type = stbi__get8(s); + int tga_is_RLE = 0; + int tga_palette_start = stbi__get16le(s); + int tga_palette_len = stbi__get16le(s); + int tga_palette_bits = stbi__get8(s); + int tga_x_origin = stbi__get16le(s); + int tga_y_origin = stbi__get16le(s); + int tga_width = stbi__get16le(s); + int tga_height = stbi__get16le(s); + int tga_bits_per_pixel = stbi__get8(s); + int tga_comp, tga_rgb16 = 0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused + // (useless?) + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = {0}; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + STBI_NOTUSED(tga_x_origin); // @TODO + STBI_NOTUSED(tga_y_origin); // @TODO - if (tga_height > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (tga_width > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (tga_height > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (tga_width > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - // do a tiny bit of precessing - if ( tga_image_type >= 8 ) - { - tga_image_type -= 8; - tga_is_RLE = 1; - } - tga_inverted = 1 - ((tga_inverted >> 5) & 1); + // do a tiny bit of precessing + if (tga_image_type >= 8) { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); - // If I'm paletted, then I'll use the number of bits from the palette - if ( tga_indexed ) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); - else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + // If I'm paletted, then I'll use the number of bits from the palette + if (tga_indexed) + tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), + &tga_rgb16); - if(!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency - return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + if (!tga_comp) // shouldn't really happen, stbi__tga_test() should have + // ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); - // tga info - *x = tga_width; - *y = tga_height; - if (comp) *comp = tga_comp; + // tga info + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; - if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) - return stbi__errpuc("too large", "Corrupt TGA"); + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); - tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); - if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + tga_data = + (unsigned char *)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); - // skip to the data's starting position (offset usually = 0) - stbi__skip(s, tga_offset ); + // skip to the data's starting position (offset usually = 0) + stbi__skip(s, tga_offset); - if ( !tga_indexed && !tga_is_RLE && !tga_rgb16 ) { - for (i=0; i < tga_height; ++i) { - int row = tga_inverted ? tga_height -i - 1 : i; - stbi_uc *tga_row = tga_data + row*tga_width*tga_comp; - stbi__getn(s, tga_row, tga_width * tga_comp); + if (!tga_indexed && !tga_is_RLE && !tga_rgb16) { + for (i = 0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height - i - 1 : i; + stbi_uc *tga_row = tga_data + row * tga_width * tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } else { + // do I need to load a palette? + if (tga_indexed) { + if (tga_palette_len == 0) { /* you have to have at least one entry! */ + STBI_FREE(tga_data); + return stbi__errpuc("bad palette", "Corrupt TGA"); } - } else { - // do I need to load a palette? - if ( tga_indexed) - { - if (tga_palette_len == 0) { /* you have to have at least one entry! */ - STBI_FREE(tga_data); - return stbi__errpuc("bad palette", "Corrupt TGA"); - } - // any data to skip? (offset usually = 0) - stbi__skip(s, tga_palette_start ); - // load the palette - tga_palette = (unsigned char*)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); - if (!tga_palette) { - STBI_FREE(tga_data); - return stbi__errpuc("outofmem", "Out of memory"); - } - if (tga_rgb16) { - stbi_uc *pal_entry = tga_palette; - STBI_ASSERT(tga_comp == STBI_rgb); - for (i=0; i < tga_palette_len; ++i) { - stbi__tga_read_rgb16(s, pal_entry); - pal_entry += tga_comp; - } - } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { - STBI_FREE(tga_data); - STBI_FREE(tga_palette); - return stbi__errpuc("bad palette", "Corrupt TGA"); - } + // any data to skip? (offset usually = 0) + stbi__skip(s, tga_palette_start); + // load the palette + tga_palette = + (unsigned char *)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); } - // load the data - for (i=0; i < tga_width * tga_height; ++i) - { - // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? - if ( tga_is_RLE ) - { - if ( RLE_count == 0 ) - { - // yep, get the next byte as a RLE command - int RLE_cmd = stbi__get8(s); - RLE_count = 1 + (RLE_cmd & 127); - RLE_repeating = RLE_cmd >> 7; - read_next_pixel = 1; - } else if ( !RLE_repeating ) - { - read_next_pixel = 1; - } - } else - { - read_next_pixel = 1; - } - // OK, if I need to read a pixel, do it now - if ( read_next_pixel ) - { - // load however much data we did have - if ( tga_indexed ) - { - // read in index, then perform the lookup - int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); - if ( pal_idx >= tga_palette_len ) { - // invalid index - pal_idx = 0; - } - pal_idx *= tga_comp; - for (j = 0; j < tga_comp; ++j) { - raw_data[j] = tga_palette[pal_idx+j]; - } - } else if(tga_rgb16) { - STBI_ASSERT(tga_comp == STBI_rgb); - stbi__tga_read_rgb16(s, raw_data); - } else { - // read in the data raw - for (j = 0; j < tga_comp; ++j) { - raw_data[j] = stbi__get8(s); - } - } - // clear the reading flag for the next pixel - read_next_pixel = 0; - } // end of reading a pixel - - // copy data - for (j = 0; j < tga_comp; ++j) - tga_data[i*tga_comp+j] = raw_data[j]; - - // in case we're in RLE mode, keep counting down - --RLE_count; + if (tga_rgb16) { + stbi_uc *pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i = 0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_FREE(tga_palette); + return stbi__errpuc("bad palette", "Corrupt TGA"); } - // do I need to invert the image? - if ( tga_inverted ) - { - for (j = 0; j*2 < tga_height; ++j) - { - int index1 = j * tga_width * tga_comp; - int index2 = (tga_height - 1 - j) * tga_width * tga_comp; - for (i = tga_width * tga_comp; i > 0; --i) - { - unsigned char temp = tga_data[index1]; - tga_data[index1] = tga_data[index2]; - tga_data[index2] = temp; - ++index1; - ++index2; - } - } + } + // load the data + for (i = 0; i < tga_width * tga_height; ++i) { + // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? + if (tga_is_RLE) { + if (RLE_count == 0) { + // yep, get the next byte as a RLE command + int RLE_cmd = stbi__get8(s); + RLE_count = 1 + (RLE_cmd & 127); + RLE_repeating = RLE_cmd >> 7; + read_next_pixel = 1; + } else if (!RLE_repeating) { + read_next_pixel = 1; + } + } else { + read_next_pixel = 1; } - // clear my palette, if I had one - if ( tga_palette != NULL ) - { - STBI_FREE( tga_palette ); + // OK, if I need to read a pixel, do it now + if (read_next_pixel) { + // load however much data we did have + if (tga_indexed) { + // read in index, then perform the lookup + int pal_idx = + (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if (pal_idx >= tga_palette_len) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx + j]; + } + } else if (tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } else { + // read in the data raw + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = stbi__get8(s); + } + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + + // copy data + for (j = 0; j < tga_comp; ++j) + tga_data[i * tga_comp + j] = raw_data[j]; + + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if (tga_inverted) { + for (j = 0; j * 2 < tga_height; ++j) { + int index1 = j * tga_width * tga_comp; + int index2 = (tga_height - 1 - j) * tga_width * tga_comp; + for (i = tga_width * tga_comp; i > 0; --i) { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } } - } + } + // clear my palette, if I had one + if (tga_palette != NULL) { STBI_FREE(tga_palette); } + } - // swap RGB - if the source data was RGB16, it already is in the right order - if (tga_comp >= 3 && !tga_rgb16) - { - unsigned char* tga_pixel = tga_data; - for (i=0; i < tga_width * tga_height; ++i) - { - unsigned char temp = tga_pixel[0]; - tga_pixel[0] = tga_pixel[2]; - tga_pixel[2] = temp; - tga_pixel += tga_comp; - } - } + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) { + unsigned char *tga_pixel = tga_data; + for (i = 0; i < tga_width * tga_height; ++i) { + unsigned char temp = tga_pixel[0]; + tga_pixel[0] = tga_pixel[2]; + tga_pixel[2] = temp; + tga_pixel += tga_comp; + } + } - // convert to target component count - if (req_comp && req_comp != tga_comp) - tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + // convert to target component count + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, + tga_height); - // the things I do to get rid of an error message, and yet keep - // Microsoft's C compilers happy... [8^( - tga_palette_start = tga_palette_len = tga_palette_bits = - tga_x_origin = tga_y_origin = 0; - STBI_NOTUSED(tga_palette_start); - // OK, done - return tga_data; + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = tga_x_origin = + tga_y_origin = 0; + STBI_NOTUSED(tga_palette_start); + // OK, done + return tga_data; } -#endif + #endif // ************************************************************************************************* -// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, +// tweaked by STB -#ifndef STBI_NO_PSD -static int stbi__psd_test(stbi__context *s) -{ - int r = (stbi__get32be(s) == 0x38425053); - stbi__rewind(s); - return r; + #ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s) { + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; } -static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) -{ - int count, nleft, len; +static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) { + int count, nleft, len; - count = 0; - while ((nleft = pixelCount - count) > 0) { - len = stbi__get8(s); - if (len == 128) { - // No-op. - } else if (len < 128) { - // Copy next len+1 bytes literally. - len++; - if (len > nleft) return 0; // corrupt data - count += len; - while (len) { - *p = stbi__get8(s); - p += 4; - len--; - } - } else if (len > 128) { - stbi_uc val; - // Next -len+1 bytes in the dest are replicated from next source byte. - // (Interpret len as a negative 8-bit int.) - len = 257 - len; - if (len > nleft) return 0; // corrupt data - val = stbi__get8(s); - count += len; - while (len) { - *p = val; - p += 4; - len--; - } + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + count += len; + while (len) { + *p = stbi__get8(s); + p += 4; + len--; } - } + } else if (len > 128) { + stbi_uc val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } - return 1; + return 1; } -static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) -{ - int pixelCount; - int channelCount, compression; - int channel, i; - int bitdepth; - int w,h; - stbi_uc *out; - STBI_NOTUSED(ri); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc) { + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w, h; + stbi_uc *out; + STBI_NOTUSED(ri); - // Check identifier - if (stbi__get32be(s) != 0x38425053) // "8BPS" - return stbi__errpuc("not PSD", "Corrupt PSD image"); + // Check identifier + if (stbi__get32be(s) != 0x38425053) // "8BPS" + return stbi__errpuc("not PSD", "Corrupt PSD image"); - // Check file type version. - if (stbi__get16be(s) != 1) - return stbi__errpuc("wrong version", "Unsupported version of PSD image"); + // Check file type version. + if (stbi__get16be(s) != 1) + return stbi__errpuc("wrong version", "Unsupported version of PSD image"); - // Skip 6 reserved bytes. - stbi__skip(s, 6 ); + // Skip 6 reserved bytes. + stbi__skip(s, 6); - // Read the number of channels (R, G, B, A, etc). - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) - return stbi__errpuc("wrong channel count", "Unsupported number of channels in PSD image"); + // Read the number of channels (R, G, B, A, etc). + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) + return stbi__errpuc("wrong channel count", + "Unsupported number of channels in PSD image"); - // Read the rows and columns of the image. - h = stbi__get32be(s); - w = stbi__get32be(s); + // Read the rows and columns of the image. + h = stbi__get32be(s); + w = stbi__get32be(s); - if (h > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (w > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (h > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (w > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - // Make sure the depth is 8 bits. - bitdepth = stbi__get16be(s); - if (bitdepth != 8 && bitdepth != 16) - return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 bit"); + // Make sure the depth is 8 bits. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", + "PSD bit depth is not 8 or 16 bit"); - // Make sure the color mode is RGB. - // Valid options are: - // 0: Bitmap - // 1: Grayscale - // 2: Indexed color - // 3: RGB color - // 4: CMYK color - // 7: Multichannel - // 8: Duotone - // 9: Lab color - if (stbi__get16be(s) != 3) - return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (stbi__get16be(s) != 3) + return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); - // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) - stbi__skip(s,stbi__get32be(s) ); + // Skip the Mode Data. (It's the palette for indexed color; other info for + // other modes.) + stbi__skip(s, stbi__get32be(s)); - // Skip the image resources. (resolution, pen tool paths, etc) - stbi__skip(s, stbi__get32be(s) ); + // Skip the image resources. (resolution, pen tool paths, etc) + stbi__skip(s, stbi__get32be(s)); - // Skip the reserved data. - stbi__skip(s, stbi__get32be(s) ); + // Skip the reserved data. + stbi__skip(s, stbi__get32be(s)); - // Find out if the data is compressed. - // Known values: - // 0: no compression - // 1: RLE compressed - compression = stbi__get16be(s); - if (compression > 1) - return stbi__errpuc("bad compression", "PSD has an unknown compression format"); + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = stbi__get16be(s); + if (compression > 1) + return stbi__errpuc("bad compression", + "PSD has an unknown compression format"); - // Check size - if (!stbi__mad3sizes_valid(4, w, h, 0)) - return stbi__errpuc("too large", "Corrupt PSD"); + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); - // Create the destination image. + // Create the destination image. - if (!compression && bitdepth == 16 && bpc == 16) { - out = (stbi_uc *) stbi__malloc_mad3(8, w, h, 0); - ri->bits_per_channel = 16; - } else - out = (stbi_uc *) stbi__malloc(4 * w*h); + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc *)stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } else + out = (stbi_uc *)stbi__malloc(4 * w * h); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - pixelCount = w*h; + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + pixelCount = w * h; - // Initialize the data to zero. - //memset( out, 0, pixelCount * 4 ); + // Initialize the data to zero. + // memset( out, 0, pixelCount * 4 ); - // Finally, the image data. - if (compression) { - // RLE as used by .PSD and .TIFF - // Loop until you get the number of unpacked bytes you are expecting: - // Read the next source byte into n. - // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. - // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. - // Else if n is 128, noop. - // Endloop + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes + // literally. Else if n is between -127 and -1 inclusive, copy the next + // byte -n+1 times. Else if n is 128, noop. + // Endloop - // The RLE-compressed data is preceded by a 2-byte data count for each row in the data, - // which we're going to just skip. - stbi__skip(s, h * channelCount * 2 ); + // The RLE-compressed data is preceded by a 2-byte data count for each row + // in the data, which we're going to just skip. + stbi__skip(s, h * channelCount * 2); - // Read the RLE data by channel. - for (channel = 0; channel < 4; channel++) { - stbi_uc *p; + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; - p = out+channel; - if (channel >= channelCount) { - // Fill this channel with default data. - for (i = 0; i < pixelCount; i++, p += 4) - *p = (channel == 3 ? 255 : 0); - } else { - // Read the RLE data. - if (!stbi__psd_decode_rle(s, p, pixelCount)) { - STBI_FREE(out); - return stbi__errpuc("corrupt", "bad RLE data"); - } - } - } - - } else { - // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) - // where each channel consists of an 8-bit (or 16-bit) value for each pixel in the image. - - // Read the data by channel. - for (channel = 0; channel < 4; channel++) { - if (channel >= channelCount) { - // Fill this channel with default data. - if (bitdepth == 16 && bpc == 16) { - stbi__uint16 *q = ((stbi__uint16 *) out) + channel; - stbi__uint16 val = channel == 3 ? 65535 : 0; - for (i = 0; i < pixelCount; i++, q += 4) - *q = val; - } else { - stbi_uc *p = out+channel; - stbi_uc val = channel == 3 ? 255 : 0; - for (i = 0; i < pixelCount; i++, p += 4) - *p = val; - } - } else { - if (ri->bits_per_channel == 16) { // output bpc - stbi__uint16 *q = ((stbi__uint16 *) out) + channel; - for (i = 0; i < pixelCount; i++, q += 4) - *q = (stbi__uint16) stbi__get16be(s); - } else { - stbi_uc *p = out+channel; - if (bitdepth == 16) { // input bpc - for (i = 0; i < pixelCount; i++, p += 4) - *p = (stbi_uc) (stbi__get16be(s) >> 8); - } else { - for (i = 0; i < pixelCount; i++, p += 4) - *p = stbi__get8(s); - } - } - } - } - } - - // remove weird white matte from PSD - if (channelCount >= 4) { - if (ri->bits_per_channel == 16) { - for (i=0; i < w*h; ++i) { - stbi__uint16 *pixel = (stbi__uint16 *) out + 4*i; - if (pixel[3] != 0 && pixel[3] != 65535) { - float a = pixel[3] / 65535.0f; - float ra = 1.0f / a; - float inv_a = 65535.0f * (1 - ra); - pixel[0] = (stbi__uint16) (pixel[0]*ra + inv_a); - pixel[1] = (stbi__uint16) (pixel[1]*ra + inv_a); - pixel[2] = (stbi__uint16) (pixel[2]*ra + inv_a); - } - } + p = out + channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++, p += 4) + *p = (channel == 3 ? 255 : 0); } else { - for (i=0; i < w*h; ++i) { - unsigned char *pixel = out + 4*i; - if (pixel[3] != 0 && pixel[3] != 255) { - float a = pixel[3] / 255.0f; - float ra = 1.0f / a; - float inv_a = 255.0f * (1 - ra); - pixel[0] = (unsigned char) (pixel[0]*ra + inv_a); - pixel[1] = (unsigned char) (pixel[1]*ra + inv_a); - pixel[2] = (unsigned char) (pixel[2]*ra + inv_a); - } - } + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } } - } + } - // convert to desired output format - if (req_comp && req_comp != 4) { - if (ri->bits_per_channel == 16) - out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, 4, req_comp, w, h); - else - out = stbi__convert_format(out, 4, req_comp, w, h); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } + } else { + // We're at the raw image data. It's each channel in order (Red, Green, + // Blue, Alpha, ...) where each channel consists of an 8-bit (or 16-bit) + // value for each pixel in the image. - if (comp) *comp = 4; - *y = h; - *x = w; + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16 *q = ((stbi__uint16 *)out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + *q = val; + } else { + stbi_uc *p = out + channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + *p = val; + } + } else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16 *q = ((stbi__uint16 *)out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + *q = (stbi__uint16)stbi__get16be(s); + } else { + stbi_uc *p = out + channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + *p = (stbi_uc)(stbi__get16be(s) >> 8); + } else { + for (i = 0; i < pixelCount; i++, p += 4) + *p = stbi__get8(s); + } + } + } + } + } - return out; + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i = 0; i < w * h; ++i) { + stbi__uint16 *pixel = (stbi__uint16 *)out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16)(pixel[0] * ra + inv_a); + pixel[1] = (stbi__uint16)(pixel[1] * ra + inv_a); + pixel[2] = (stbi__uint16)(pixel[2] * ra + inv_a); + } + } + } else { + for (i = 0; i < w * h; ++i) { + unsigned char *pixel = out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char)(pixel[0] * ra + inv_a); + pixel[1] = (unsigned char)(pixel[1] * ra + inv_a); + pixel[2] = (unsigned char)(pixel[2] * ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc *)stbi__convert_format16((stbi__uint16 *)out, 4, req_comp, + w, h); + else + out = stbi__convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + if (comp) *comp = 4; + *y = h; + *x = w; + + return out; } -#endif + #endif // ************************************************************************************************* // Softimage PIC loader @@ -6151,1130 +6677,1139 @@ static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req // See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format // See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ -#ifndef STBI_NO_PIC -static int stbi__pic_is4(stbi__context *s,const char *str) -{ - int i; - for (i=0; i<4; ++i) - if (stbi__get8(s) != (stbi_uc)str[i]) - return 0; + #ifndef STBI_NO_PIC +static int stbi__pic_is4(stbi__context *s, const char *str) { + int i; + for (i = 0; i < 4; ++i) + if (stbi__get8(s) != (stbi_uc)str[i]) return 0; - return 1; + return 1; } -static int stbi__pic_test_core(stbi__context *s) -{ - int i; +static int stbi__pic_test_core(stbi__context *s) { + int i; - if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) - return 0; + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) return 0; - for(i=0;i<84;++i) - stbi__get8(s); + for (i = 0; i < 84; ++i) + stbi__get8(s); - if (!stbi__pic_is4(s,"PICT")) - return 0; + if (!stbi__pic_is4(s, "PICT")) return 0; - return 1; + return 1; } -typedef struct -{ - stbi_uc size,type,channel; +typedef struct { + stbi_uc size, type, channel; } stbi__pic_packet; -static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) -{ - int mask=0x80, i; +static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) { + int mask = 0x80, i; - for (i=0; i<4; ++i, mask>>=1) { - if (channel & mask) { - if (stbi__at_eof(s)) return stbi__errpuc("bad file","PIC file too short"); - dest[i]=stbi__get8(s); - } - } + for (i = 0; i < 4; ++i, mask >>= 1) { + if (channel & mask) { + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "PIC file too short"); + dest[i] = stbi__get8(s); + } + } - return dest; + return dest; } -static void stbi__copyval(int channel,stbi_uc *dest,const stbi_uc *src) -{ - int mask=0x80,i; +static void stbi__copyval(int channel, stbi_uc *dest, const stbi_uc *src) { + int mask = 0x80, i; - for (i=0;i<4; ++i, mask>>=1) - if (channel&mask) - dest[i]=src[i]; + for (i = 0; i < 4; ++i, mask >>= 1) + if (channel & mask) dest[i] = src[i]; } -static stbi_uc *stbi__pic_load_core(stbi__context *s,int width,int height,int *comp, stbi_uc *result) -{ - int act_comp=0,num_packets=0,y,chained; - stbi__pic_packet packets[10]; +static stbi_uc *stbi__pic_load_core(stbi__context *s, int width, int height, + int *comp, stbi_uc *result) { + int act_comp = 0, num_packets = 0, y, chained; + stbi__pic_packet packets[10]; - // this will (should...) cater for even some bizarre stuff like having data - // for the same channel in multiple packets. - do { - stbi__pic_packet *packet; + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + stbi__pic_packet *packet; - if (num_packets==sizeof(packets)/sizeof(packets[0])) - return stbi__errpuc("bad format","too many packets"); + if (num_packets == sizeof(packets) / sizeof(packets[0])) + return stbi__errpuc("bad format", "too many packets"); - packet = &packets[num_packets++]; + packet = &packets[num_packets++]; - chained = stbi__get8(s); - packet->size = stbi__get8(s); - packet->type = stbi__get8(s); - packet->channel = stbi__get8(s); + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); - act_comp |= packet->channel; + act_comp |= packet->channel; - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (reading packets)"); - if (packet->size != 8) return stbi__errpuc("bad format","packet isn't 8bpp"); - } while (chained); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "file too short (reading packets)"); + if (packet->size != 8) + return stbi__errpuc("bad format", "packet isn't 8bpp"); + } while (chained); - *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? - for(y=0; ytype) { - default: - return stbi__errpuc("bad format","packet has bad compression type"); + switch (packet->type) { + default: + return stbi__errpuc("bad format", "packet has bad compression type"); - case 0: {//uncompressed - int x; + case 0: { // uncompressed + int x; - for(x=0;xchannel,dest)) - return 0; - break; + for (x = 0; x < width; ++x, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) return 0; + break; + } + + case 1: // Pure RLE + { + int left = width, i; + + while (left > 0) { + stbi_uc count, value[4]; + + count = stbi__get8(s); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", + "file too short (pure read count)"); + + if (count > left) count = (stbi_uc)left; + + if (!stbi__readval(s, packet->channel, value)) return 0; + + for (i = 0; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + left -= count; + } + } break; + + case 2: { // Mixed RLE + int left = width; + while (left > 0) { + int count = stbi__get8(s), i; + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", + "file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + + if (count == 128) + count = stbi__get16be(s); + else + count -= 127; + if (count > left) + return stbi__errpuc("bad file", "scanline overrun"); + + if (!stbi__readval(s, packet->channel, value)) return 0; + + for (i = 0; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + } else { // Raw + ++count; + if (count > left) + return stbi__errpuc("bad file", "scanline overrun"); + + for (i = 0; i < count; ++i, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) return 0; } - - case 1://Pure RLE - { - int left=width, i; - - while (left>0) { - stbi_uc count,value[4]; - - count=stbi__get8(s); - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pure read count)"); - - if (count > left) - count = (stbi_uc) left; - - if (!stbi__readval(s,packet->channel,value)) return 0; - - for(i=0; ichannel,dest,value); - left -= count; - } - } - break; - - case 2: {//Mixed RLE - int left=width; - while (left>0) { - int count = stbi__get8(s), i; - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (mixed read count)"); - - if (count >= 128) { // Repeated - stbi_uc value[4]; - - if (count==128) - count = stbi__get16be(s); - else - count -= 127; - if (count > left) - return stbi__errpuc("bad file","scanline overrun"); - - if (!stbi__readval(s,packet->channel,value)) - return 0; - - for(i=0;ichannel,dest,value); - } else { // Raw - ++count; - if (count>left) return stbi__errpuc("bad file","scanline overrun"); - - for(i=0;ichannel,dest)) - return 0; - } - left-=count; - } - break; - } - } + left -= count; + } + break; + } } - } + } + } - return result; + return result; } -static void *stbi__pic_load(stbi__context *s,int *px,int *py,int *comp,int req_comp, stbi__result_info *ri) -{ - stbi_uc *result; - int i, x,y, internal_comp; - STBI_NOTUSED(ri); +static void *stbi__pic_load(stbi__context *s, int *px, int *py, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *result; + int i, x, y, internal_comp; + STBI_NOTUSED(ri); - if (!comp) comp = &internal_comp; + if (!comp) comp = &internal_comp; - for (i=0; i<92; ++i) - stbi__get8(s); + for (i = 0; i < 92; ++i) + stbi__get8(s); - x = stbi__get16be(s); - y = stbi__get16be(s); + x = stbi__get16be(s); + y = stbi__get16be(s); - if (y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pic header)"); - if (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC image too large to decode"); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "file too short (pic header)"); + if (!stbi__mad3sizes_valid(x, y, 4, 0)) + return stbi__errpuc("too large", "PIC image too large to decode"); - stbi__get32be(s); //skip `ratio' - stbi__get16be(s); //skip `fields' - stbi__get16be(s); //skip `pad' + stbi__get32be(s); // skip `ratio' + stbi__get16be(s); // skip `fields' + stbi__get16be(s); // skip `pad' - // intermediate buffer is RGBA - result = (stbi_uc *) stbi__malloc_mad3(x, y, 4, 0); - memset(result, 0xff, x*y*4); + // intermediate buffer is RGBA + result = (stbi_uc *)stbi__malloc_mad3(x, y, 4, 0); + memset(result, 0xff, x * y * 4); - if (!stbi__pic_load_core(s,x,y,comp, result)) { - STBI_FREE(result); - result=0; - } - *px = x; - *py = y; - if (req_comp == 0) req_comp = *comp; - result=stbi__convert_format(result,4,req_comp,x,y); + if (!stbi__pic_load_core(s, x, y, comp, result)) { + STBI_FREE(result); + result = 0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result = stbi__convert_format(result, 4, req_comp, x, y); - return result; + return result; } -static int stbi__pic_test(stbi__context *s) -{ - int r = stbi__pic_test_core(s); - stbi__rewind(s); - return r; +static int stbi__pic_test(stbi__context *s) { + int r = stbi__pic_test_core(s); + stbi__rewind(s); + return r; } -#endif + #endif // ************************************************************************************************* // GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb -#ifndef STBI_NO_GIF -typedef struct -{ - stbi__int16 prefix; - stbi_uc first; - stbi_uc suffix; + #ifndef STBI_NO_GIF +typedef struct { + stbi__int16 prefix; + stbi_uc first; + stbi_uc suffix; } stbi__gif_lzw; -typedef struct -{ - int w,h; - stbi_uc *out; // output buffer (always 4 components) - stbi_uc *background; // The current "background" as far as a gif is concerned - stbi_uc *history; - int flags, bgindex, ratio, transparent, eflags; - stbi_uc pal[256][4]; - stbi_uc lpal[256][4]; - stbi__gif_lzw codes[8192]; - stbi_uc *color_table; - int parse, step; - int lflags; - int start_x, start_y; - int max_x, max_y; - int cur_x, cur_y; - int line_size; - int delay; +typedef struct { + int w, h; + stbi_uc *out; // output buffer (always 4 components) + stbi_uc *background; // The current "background" as far as a gif is concerned + stbi_uc *history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + stbi_uc *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; + int delay; } stbi__gif; -static int stbi__gif_test_raw(stbi__context *s) -{ - int sz; - if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') return 0; - sz = stbi__get8(s); - if (sz != '9' && sz != '7') return 0; - if (stbi__get8(s) != 'a') return 0; - return 1; +static int stbi__gif_test_raw(stbi__context *s) { + int sz; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || + stbi__get8(s) != '8') + return 0; + sz = stbi__get8(s); + if (sz != '9' && sz != '7') return 0; + if (stbi__get8(s) != 'a') return 0; + return 1; } -static int stbi__gif_test(stbi__context *s) -{ - int r = stbi__gif_test_raw(s); - stbi__rewind(s); - return r; +static int stbi__gif_test(stbi__context *s) { + int r = stbi__gif_test_raw(s); + stbi__rewind(s); + return r; } -static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], int num_entries, int transp) -{ - int i; - for (i=0; i < num_entries; ++i) { - pal[i][2] = stbi__get8(s); - pal[i][1] = stbi__get8(s); - pal[i][0] = stbi__get8(s); - pal[i][3] = transp == i ? 0 : 255; - } +static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], + int num_entries, int transp) { + int i; + for (i = 0; i < num_entries; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + pal[i][3] = transp == i ? 0 : 255; + } } -static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, int is_info) -{ - stbi_uc version; - if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') - return stbi__err("not GIF", "Corrupt GIF"); +static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, + int is_info) { + stbi_uc version; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || + stbi__get8(s) != '8') + return stbi__err("not GIF", "Corrupt GIF"); - version = stbi__get8(s); - if (version != '7' && version != '9') return stbi__err("not GIF", "Corrupt GIF"); - if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); + version = stbi__get8(s); + if (version != '7' && version != '9') + return stbi__err("not GIF", "Corrupt GIF"); + if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); - stbi__g_failure_reason = ""; - g->w = stbi__get16le(s); - g->h = stbi__get16le(s); - g->flags = stbi__get8(s); - g->bgindex = stbi__get8(s); - g->ratio = stbi__get8(s); - g->transparent = -1; + stbi__g_failure_reason = ""; + g->w = stbi__get16le(s); + g->h = stbi__get16le(s); + g->flags = stbi__get8(s); + g->bgindex = stbi__get8(s); + g->ratio = stbi__get8(s); + g->transparent = -1; - if (g->w > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (g->h > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (g->w > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (g->h > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); - if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + if (comp != 0) + *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the + // comments - if (is_info) return 1; + if (is_info) return 1; - if (g->flags & 0x80) - stbi__gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + if (g->flags & 0x80) + stbi__gif_parse_colortable(s, g->pal, 2 << (g->flags & 7), -1); - return 1; + return 1; } -static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) -{ - stbi__gif* g = (stbi__gif*) stbi__malloc(sizeof(stbi__gif)); - if (!stbi__gif_header(s, g, comp, 1)) { - STBI_FREE(g); - stbi__rewind( s ); - return 0; - } - if (x) *x = g->w; - if (y) *y = g->h; - STBI_FREE(g); - return 1; +static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) { + stbi__gif *g = (stbi__gif *)stbi__malloc(sizeof(stbi__gif)); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind(s); + return 0; + } + if (x) *x = g->w; + if (y) *y = g->h; + STBI_FREE(g); + return 1; } -static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) -{ - stbi_uc *p, *c; - int idx; +static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) { + stbi_uc *p, *c; + int idx; - // recurse to decode the prefixes, since the linked-list is backwards, - // and working backwards through an interleaved image would be nasty - if (g->codes[code].prefix >= 0) - stbi__out_gif_code(g, g->codes[code].prefix); + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) stbi__out_gif_code(g, g->codes[code].prefix); - if (g->cur_y >= g->max_y) return; + if (g->cur_y >= g->max_y) return; - idx = g->cur_x + g->cur_y; - p = &g->out[idx]; - g->history[idx / 4] = 1; + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; - c = &g->color_table[g->codes[code].suffix * 4]; - if (c[3] > 128) { // don't render transparent pixels; - p[0] = c[2]; - p[1] = c[1]; - p[2] = c[0]; - p[3] = c[3]; - } - g->cur_x += 4; + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; - if (g->cur_x >= g->max_x) { - g->cur_x = g->start_x; - g->cur_y += g->step; + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; - while (g->cur_y >= g->max_y && g->parse > 0) { - g->step = (1 << g->parse) * g->line_size; - g->cur_y = g->start_y + (g->step >> 1); - --g->parse; + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) { + stbi_uc lzw_cs; + stbi__int32 len, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw *p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc)init_code; + g->codes[init_code].suffix = (stbi_uc)init_code; + } + + // support no starting clear code + avail = clear + 2; + oldcode = -1; + + len = 0; + for (;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = stbi__get8(s); // start new block + if (len == 0) return g->out; } - } + --len; + bits |= (stbi__int32)stbi__get8(s) << valid_bits; + valid_bits += 8; + } else { + stbi__int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + stbi__skip(s, len); + while ((len = stbi__get8(s)) > 0) + stbi__skip(s, len); + return g->out; + } else if (code <= avail) { + if (first) { return stbi__errpuc("no clear code", "Corrupt GIF"); } + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 8192) { + return stbi__errpuc("too many codes", "Corrupt GIF"); + } + + p->prefix = (stbi__int16)oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + + stbi__out_gif_code(g, (stbi__uint16)code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + } + } + } } -static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) -{ - stbi_uc lzw_cs; - stbi__int32 len, init_code; - stbi__uint32 first; - stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; - stbi__gif_lzw *p; +// this function is designed to support animated gifs, although stb_image +// doesn't support it two back is the image from two frames ago, used for a very +// specific disposal format +static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, + int req_comp, stbi_uc *two_back) { + int dispose; + int first_frame; + int pi; + int pcount; + STBI_NOTUSED(req_comp); - lzw_cs = stbi__get8(s); - if (lzw_cs > 12) return NULL; - clear = 1 << lzw_cs; - first = 1; - codesize = lzw_cs + 1; - codemask = (1 << codesize) - 1; - bits = 0; - valid_bits = 0; - for (init_code = 0; init_code < clear; init_code++) { - g->codes[init_code].prefix = -1; - g->codes[init_code].first = (stbi_uc) init_code; - g->codes[init_code].suffix = (stbi_uc) init_code; - } + // on first frame, any non-written pixels get the background colour + // (non-transparent) + first_frame = 0; + if (g->out == 0) { + if (!stbi__gif_header(s, g, comp, 0)) + return 0; // stbi__g_failure_reason set by stbi__gif_header + if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) + return stbi__errpuc("too large", "GIF image is too large"); + pcount = g->w * g->h; + g->out = (stbi_uc *)stbi__malloc(4 * pcount); + g->background = (stbi_uc *)stbi__malloc(4 * pcount); + g->history = (stbi_uc *)stbi__malloc(pcount); + if (!g->out || !g->background || !g->history) + return stbi__errpuc("outofmem", "Out of memory"); - // support no starting clear code - avail = clear+2; - oldcode = -1; + // image is treated as "transparent" at the start - ie, nothing overwrites + // the current background; background colour is only used for pixels that + // are not rendered first frame, after that "background" color refers to the + // color that was there the previous frame. + memset(g->out, 0x00, 4 * pcount); + memset(g->background, 0x00, + 4 * pcount); // state of the background (starts transparent) + memset(g->history, 0x00, + pcount); // pixels that were affected previous frame + first_frame = 1; + } else { + // second frame - how do we dispose of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; - len = 0; - for(;;) { - if (valid_bits < codesize) { - if (len == 0) { - len = stbi__get8(s); // start new block - if (len == 0) - return g->out; - } - --len; - bits |= (stbi__int32) stbi__get8(s) << valid_bits; - valid_bits += 8; - } else { - stbi__int32 code = bits & codemask; - bits >>= codesize; - valid_bits -= codesize; - // @OPTIMIZE: is there some way we can accelerate the non-clear path? - if (code == clear) { // clear code - codesize = lzw_cs + 1; - codemask = (1 << codesize) - 1; - avail = clear + 2; - oldcode = -1; - first = 0; - } else if (code == clear + 1) { // end of stream code + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to + // the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { memcpy(&g->out[pi * 4], &two_back[pi * 4], 4); } + } + } else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy(&g->out[pi * 4], &g->background[pi * 4], 4); + } + } + } else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy(g->background, g->out, 4 * g->w * g->h); + } + + // clear my history; + memset(g->history, 0x00, + g->w * g->h); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + case 0x2C: /* Image Descriptor */ + { + stbi__int32 x, y, w, h; + stbi_uc *o; + + x = stbi__get16le(s); + y = stbi__get16le(s); + w = stbi__get16le(s); + h = stbi__get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + // if the width of the specified rectangle is 0, that means + // we may not see *any* pixels or the image is malformed; + // to make sure this is caught, move the current y down to + // max_y (which is what out_gif_code checks). + if (w == 0) g->cur_y = g->max_y; + + g->lflags = stbi__get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi__gif_parse_colortable(s, g->lpal, 2 << (g->lflags & 7), + g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (stbi_uc *)g->lpal; + } else if (g->flags & 0x80) { + g->color_table = (stbi_uc *)g->pal; + } else + return stbi__errpuc("missing color table", "Corrupt GIF"); + + o = stbi__process_gif_raster(s, g); + if (!o) return NULL; + + // if this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = + 255; // just in case it was made transparent, undo that; It + // will be reset next frame if need be; + memcpy(&g->out[pi * 4], &g->pal[g->bgindex], 4); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = + 10 * + stbi__get16le( + s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { g->pal[g->transparent][3] = 255; } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { g->pal[g->transparent][3] = 0; } + } else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } else { stbi__skip(s, len); - while ((len = stbi__get8(s)) > 0) - stbi__skip(s,len); - return g->out; - } else if (code <= avail) { - if (first) { - return stbi__errpuc("no clear code", "Corrupt GIF"); - } - - if (oldcode >= 0) { - p = &g->codes[avail++]; - if (avail > 8192) { - return stbi__errpuc("too many codes", "Corrupt GIF"); - } - - p->prefix = (stbi__int16) oldcode; - p->first = g->codes[oldcode].first; - p->suffix = (code == avail) ? p->first : g->codes[code].first; - } else if (code == avail) - return stbi__errpuc("illegal code in raster", "Corrupt GIF"); - - stbi__out_gif_code(g, (stbi__uint16) code); - - if ((avail & codemask) == 0 && avail <= 0x0FFF) { - codesize++; - codemask = (1 << codesize) - 1; - } - - oldcode = code; - } else { - return stbi__errpuc("illegal code in raster", "Corrupt GIF"); - } - } - } -} - -// this function is designed to support animated gifs, although stb_image doesn't support it -// two back is the image from two frames ago, used for a very specific disposal format -static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, int req_comp, stbi_uc *two_back) -{ - int dispose; - int first_frame; - int pi; - int pcount; - STBI_NOTUSED(req_comp); - - // on first frame, any non-written pixels get the background colour (non-transparent) - first_frame = 0; - if (g->out == 0) { - if (!stbi__gif_header(s, g, comp,0)) return 0; // stbi__g_failure_reason set by stbi__gif_header - if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) - return stbi__errpuc("too large", "GIF image is too large"); - pcount = g->w * g->h; - g->out = (stbi_uc *) stbi__malloc(4 * pcount); - g->background = (stbi_uc *) stbi__malloc(4 * pcount); - g->history = (stbi_uc *) stbi__malloc(pcount); - if (!g->out || !g->background || !g->history) - return stbi__errpuc("outofmem", "Out of memory"); - - // image is treated as "transparent" at the start - ie, nothing overwrites the current background; - // background colour is only used for pixels that are not rendered first frame, after that "background" - // color refers to the color that was there the previous frame. - memset(g->out, 0x00, 4 * pcount); - memset(g->background, 0x00, 4 * pcount); // state of the background (starts transparent) - memset(g->history, 0x00, pcount); // pixels that were affected previous frame - first_frame = 1; - } else { - // second frame - how do we dispose of the previous one? - dispose = (g->eflags & 0x1C) >> 2; - pcount = g->w * g->h; - - if ((dispose == 3) && (two_back == 0)) { - dispose = 2; // if I don't have an image to revert back to, default to the old background - } - - if (dispose == 3) { // use previous graphic - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi]) { - memcpy( &g->out[pi * 4], &two_back[pi * 4], 4 ); - } - } - } else if (dispose == 2) { - // restore what was changed last frame to background before that frame; - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi]) { - memcpy( &g->out[pi * 4], &g->background[pi * 4], 4 ); - } - } - } else { - // This is a non-disposal case eithe way, so just - // leave the pixels as is, and they will become the new background - // 1: do not dispose - // 0: not specified. - } - - // background is what out is after the undoing of the previou frame; - memcpy( g->background, g->out, 4 * g->w * g->h ); - } - - // clear my history; - memset( g->history, 0x00, g->w * g->h ); // pixels that were affected previous frame - - for (;;) { - int tag = stbi__get8(s); - switch (tag) { - case 0x2C: /* Image Descriptor */ - { - stbi__int32 x, y, w, h; - stbi_uc *o; - - x = stbi__get16le(s); - y = stbi__get16le(s); - w = stbi__get16le(s); - h = stbi__get16le(s); - if (((x + w) > (g->w)) || ((y + h) > (g->h))) - return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); - - g->line_size = g->w * 4; - g->start_x = x * 4; - g->start_y = y * g->line_size; - g->max_x = g->start_x + w * 4; - g->max_y = g->start_y + h * g->line_size; - g->cur_x = g->start_x; - g->cur_y = g->start_y; - - // if the width of the specified rectangle is 0, that means - // we may not see *any* pixels or the image is malformed; - // to make sure this is caught, move the current y down to - // max_y (which is what out_gif_code checks). - if (w == 0) - g->cur_y = g->max_y; - - g->lflags = stbi__get8(s); - - if (g->lflags & 0x40) { - g->step = 8 * g->line_size; // first interlaced spacing - g->parse = 3; - } else { - g->step = g->line_size; - g->parse = 0; - } - - if (g->lflags & 0x80) { - stbi__gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); - g->color_table = (stbi_uc *) g->lpal; - } else if (g->flags & 0x80) { - g->color_table = (stbi_uc *) g->pal; - } else - return stbi__errpuc("missing color table", "Corrupt GIF"); - - o = stbi__process_gif_raster(s, g); - if (!o) return NULL; - - // if this was the first frame, - pcount = g->w * g->h; - if (first_frame && (g->bgindex > 0)) { - // if first frame, any pixel not drawn to gets the background color - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi] == 0) { - g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; - memcpy( &g->out[pi * 4], &g->pal[g->bgindex], 4 ); - } - } - } - - return o; - } - - case 0x21: // Comment Extension. - { - int len; - int ext = stbi__get8(s); - if (ext == 0xF9) { // Graphic Control Extension. - len = stbi__get8(s); - if (len == 4) { - g->eflags = stbi__get8(s); - g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. - - // unset old transparent - if (g->transparent >= 0) { - g->pal[g->transparent][3] = 255; - } - if (g->eflags & 0x01) { - g->transparent = stbi__get8(s); - if (g->transparent >= 0) { - g->pal[g->transparent][3] = 0; - } - } else { - // don't need transparent - stbi__skip(s, 1); - g->transparent = -1; - } - } else { - stbi__skip(s, len); - break; - } - } - while ((len = stbi__get8(s)) != 0) { - stbi__skip(s, len); - } break; - } - - case 0x3B: // gif stream termination code - return (stbi_uc *) s; // using '1' causes warning on some compilers - - default: - return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } + while ((len = stbi__get8(s)) != 0) { + stbi__skip(s, len); + } + break; } - } + + case 0x3B: // gif stream termination code + return (stbi_uc *)s; // using '1' causes warning on some compilers + + default: + return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } } -static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp) -{ - if (stbi__gif_test(s)) { - int layers = 0; - stbi_uc *u = 0; - stbi_uc *out = 0; - stbi_uc *two_back = 0; - stbi__gif g; - int stride; - int out_size = 0; - int delays_size = 0; - memset(&g, 0, sizeof(g)); - if (delays) { - *delays = 0; +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, + int *z, int *comp, int req_comp) { + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc *u = 0; + stbi_uc *out = 0; + stbi_uc *two_back = 0; + stbi__gif g; + int stride; + int out_size = 0; + int delays_size = 0; + memset(&g, 0, sizeof(g)); + if (delays) { *delays = 0; } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc *)s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + void *tmp = + (stbi_uc *)STBI_REALLOC_SIZED(out, out_size, layers * stride); + if (NULL == tmp) { + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + return stbi__errpuc("outofmem", "Out of memory"); + } else { + out = (stbi_uc *)tmp; + out_size = layers * stride; + } + + if (delays) { + *delays = (int *)STBI_REALLOC_SIZED(*delays, delays_size, + sizeof(int) * layers); + delays_size = layers * sizeof(int); + } + } else { + out = (stbi_uc *)stbi__malloc(layers * stride); + out_size = layers * stride; + if (delays) { + *delays = (int *)stbi__malloc(layers * sizeof(int)); + delays_size = layers * sizeof(int); + } + } + memcpy(out + ((layers - 1) * stride), u, stride); + if (layers >= 2) { two_back = out - 2 * stride; } + + if (delays) { (*delays)[layers - 1U] = g.delay; } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + STBI_NOTUSED(ri); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc *)s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } else if (g.out) { + // if there was an error and we allocated an image buffer, free it! + STBI_FREE(g.out); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + return u; +} + +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) { + return stbi__gif_info_raw(s, x, y, comp); +} + #endif + + // ************************************************************************************************* + // Radiance RGBE HDR loader + // originally by Nicolas Schulz + #ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context *s, const char *signature) { + int i; + for (i = 0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context *s) { + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if (!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + stbi__rewind(s); + } + return r; +} + + #define STBI__HDR_BUFLEN 1024 +static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) { + int len = 0; + char c = '\0'; + + c = (char)stbi__get8(z); + + while (!stbi__at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == STBI__HDR_BUFLEN - 1) { + // flush to end of line + while (!stbi__at_eof(z) && stbi__get8(z) != '\n') + ; + break; + } + c = (char)stbi__get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) { + if (input[3] != 0) { + float f1; + // Exponent + f1 = (float)ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: + output[3] = 1; /* fallthrough */ + case 3: + output[0] = output[1] = output[2] = 0; + break; + case 2: + output[1] = 1; /* fallthrough */ + case 1: + output[0] = 0; + break; + } + } +} + +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1, c2, z; + const char *headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s, buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && + strcmp(headerToken, "#?RGBE") != 0) + return stbi__errpf("not HDR", "Corrupt HDR image"); + + // Parse header + for (;;) { + token = stbi__hdr_gettoken(s, buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) + return stbi__errpf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = stbi__hdr_gettoken(s, buffer); + if (strncmp(token, "-Y ", 3)) + return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = (int)strtol(token, &token, 10); + while (*token == ' ') + ++token; + if (strncmp(token, "+X ", 3)) + return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = (int)strtol(token, NULL, 10); + + if (height > STBI_MAX_DIMENSIONS) + return stbi__errpf("too large", "Very large image (corrupt?)"); + if (width > STBI_MAX_DIMENSIONS) + return stbi__errpf("too large", "Very large image (corrupt?)"); + + *x = width; + *y = height; + + if (comp) *comp = 3; + if (req_comp == 0) req_comp = 3; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = + (float *)stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) return stbi__errpf("outofmem", "Out of memory"); + + // Load image data + // image data is stored as some number of sca + if (width < 8 || width >= 32768) { + // Read flat data + for (j = 0; j < height; ++j) { + for (i = 0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + stbi__getn(s, rgbe, 4); + stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, + req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = stbi__get8(s); + c2 = stbi__get8(s); + len = stbi__get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a + // decoded pixel (note this can't be a valid pixel--one of RGB must be + // >= 128) + stbi_uc rgbe[4]; + rgbe[0] = (stbi_uc)c1; + rgbe[1] = (stbi_uc)c2; + rgbe[2] = (stbi_uc)len; + rgbe[3] = (stbi_uc)stbi__get8(s); + stbi__hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); + } + if (scanline == NULL) { + scanline = (stbi_uc *)stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } } - do { - u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); - if (u == (stbi_uc *) s) u = 0; // end of animated gif marker - - if (u) { - *x = g.w; - *y = g.h; - ++layers; - stride = g.w * g.h * 4; - - if (out) { - void *tmp = (stbi_uc*) STBI_REALLOC_SIZED( out, out_size, layers * stride ); - if (NULL == tmp) { - STBI_FREE(g.out); - STBI_FREE(g.history); - STBI_FREE(g.background); - return stbi__errpuc("outofmem", "Out of memory"); - } - else { - out = (stbi_uc*) tmp; - out_size = layers * stride; - } - - if (delays) { - *delays = (int*) STBI_REALLOC_SIZED( *delays, delays_size, sizeof(int) * layers ); - delays_size = layers * sizeof(int); - } - } else { - out = (stbi_uc*)stbi__malloc( layers * stride ); - out_size = layers * stride; - if (delays) { - *delays = (int*) stbi__malloc( layers * sizeof(int) ); - delays_size = layers * sizeof(int); - } + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if (count > nleft) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("corrupt", "bad RLE data in HDR"); } - memcpy( out + ((layers - 1) * stride), u, stride ); - if (layers >= 2) { - two_back = out - 2 * stride; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + if (count > nleft) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("corrupt", "bad RLE data in HDR"); } - - if (delays) { - (*delays)[layers - 1U] = g.delay; - } - } - } while (u != 0); - - // free temp buffer; - STBI_FREE(g.out); - STBI_FREE(g.history); - STBI_FREE(g.background); - - // do the final conversion after loading everything; - if (req_comp && req_comp != 4) - out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); - - *z = layers; - return out; - } else { - return stbi__errpuc("not GIF", "Image was not as a gif type."); - } -} - -static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *u = 0; - stbi__gif g; - memset(&g, 0, sizeof(g)); - STBI_NOTUSED(ri); - - u = stbi__gif_load_next(s, &g, comp, req_comp, 0); - if (u == (stbi_uc *) s) u = 0; // end of animated gif marker - if (u) { - *x = g.w; - *y = g.h; - - // moved conversion to after successful load so that the same - // can be done for multiple frames. - if (req_comp && req_comp != 4) - u = stbi__convert_format(u, 4, req_comp, g.w, g.h); - } else if (g.out) { - // if there was an error and we allocated an image buffer, free it! - STBI_FREE(g.out); - } - - // free buffers needed for multiple frame loading; - STBI_FREE(g.history); - STBI_FREE(g.background); - - return u; -} - -static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) -{ - return stbi__gif_info_raw(s,x,y,comp); -} -#endif - -// ************************************************************************************************* -// Radiance RGBE HDR loader -// originally by Nicolas Schulz -#ifndef STBI_NO_HDR -static int stbi__hdr_test_core(stbi__context *s, const char *signature) -{ - int i; - for (i=0; signature[i]; ++i) - if (stbi__get8(s) != signature[i]) - return 0; - stbi__rewind(s); - return 1; -} - -static int stbi__hdr_test(stbi__context* s) -{ - int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); - stbi__rewind(s); - if(!r) { - r = stbi__hdr_test_core(s, "#?RGBE\n"); - stbi__rewind(s); - } - return r; -} - -#define STBI__HDR_BUFLEN 1024 -static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) -{ - int len=0; - char c = '\0'; - - c = (char) stbi__get8(z); - - while (!stbi__at_eof(z) && c != '\n') { - buffer[len++] = c; - if (len == STBI__HDR_BUFLEN-1) { - // flush to end of line - while (!stbi__at_eof(z) && stbi__get8(z) != '\n') - ; - break; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = stbi__get8(s); + } + } } - c = (char) stbi__get8(z); - } + for (i = 0; i < width; ++i) + stbi__hdr_convert(hdr_data + (j * width + i) * req_comp, + scanline + i * 4, req_comp); + } + if (scanline) STBI_FREE(scanline); + } - buffer[len] = 0; - return buffer; + return hdr_data; } -static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) -{ - if ( input[3] != 0 ) { - float f1; - // Exponent - f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); - if (req_comp <= 2) - output[0] = (input[0] + input[1] + input[2]) * f1 / 3; - else { - output[0] = input[0] * f1; - output[1] = input[1] * f1; - output[2] = input[2] * f1; - } - if (req_comp == 2) output[1] = 1; - if (req_comp == 4) output[3] = 1; - } else { - switch (req_comp) { - case 4: output[3] = 1; /* fallthrough */ - case 3: output[0] = output[1] = output[2] = 0; - break; - case 2: output[1] = 1; /* fallthrough */ - case 1: output[0] = 0; - break; - } - } +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) { + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 0) { + stbi__rewind(s); + return 0; + } + + for (;;) { + token = stbi__hdr_gettoken(s, buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi__rewind(s); + return 0; + } + token = stbi__hdr_gettoken(s, buffer); + if (strncmp(token, "-Y ", 3)) { + stbi__rewind(s); + return 0; + } + token += 3; + *y = (int)strtol(token, &token, 10); + while (*token == ' ') + ++token; + if (strncmp(token, "+X ", 3)) { + stbi__rewind(s); + return 0; + } + token += 3; + *x = (int)strtol(token, NULL, 10); + *comp = 3; + return 1; +} + #endif // STBI_NO_HDR + + #ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) { + void *p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + stbi__rewind(s); + if (p == NULL) return 0; + if (x) *x = s->img_x; + if (y) *y = s->img_y; + if (comp) { + if (info.bpp == 24 && info.ma == 0xff000000) + *comp = 3; + else + *comp = info.ma ? 4 : 3; + } + return 1; +} + #endif + + #ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) { + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind(s); + return 0; + } + *y = stbi__get32be(s); + *x = stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind(s); + return 0; + } + *comp = 4; + return 1; } -static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - char buffer[STBI__HDR_BUFLEN]; - char *token; - int valid = 0; - int width, height; - stbi_uc *scanline; - float *hdr_data; - int len; - unsigned char count, value; - int i, j, k, c1,c2, z; - const char *headerToken; - STBI_NOTUSED(ri); - - // Check identifier - headerToken = stbi__hdr_gettoken(s,buffer); - if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 0) - return stbi__errpf("not HDR", "Corrupt HDR image"); - - // Parse header - for(;;) { - token = stbi__hdr_gettoken(s,buffer); - if (token[0] == 0) break; - if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; - } - - if (!valid) return stbi__errpf("unsupported format", "Unsupported HDR format"); - - // Parse width and height - // can't use sscanf() if we're not using stdio! - token = stbi__hdr_gettoken(s,buffer); - if (strncmp(token, "-Y ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); - token += 3; - height = (int) strtol(token, &token, 10); - while (*token == ' ') ++token; - if (strncmp(token, "+X ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); - token += 3; - width = (int) strtol(token, NULL, 10); - - if (height > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); - if (width > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); - - *x = width; - *y = height; - - if (comp) *comp = 3; - if (req_comp == 0) req_comp = 3; - - if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) - return stbi__errpf("too large", "HDR image is too large"); - - // Read data - hdr_data = (float *) stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); - if (!hdr_data) - return stbi__errpf("outofmem", "Out of memory"); - - // Load image data - // image data is stored as some number of sca - if ( width < 8 || width >= 32768) { - // Read flat data - for (j=0; j < height; ++j) { - for (i=0; i < width; ++i) { - stbi_uc rgbe[4]; - main_decode_loop: - stbi__getn(s, rgbe, 4); - stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); - } - } - } else { - // Read RLE-encoded data - scanline = NULL; - - for (j = 0; j < height; ++j) { - c1 = stbi__get8(s); - c2 = stbi__get8(s); - len = stbi__get8(s); - if (c1 != 2 || c2 != 2 || (len & 0x80)) { - // not run-length encoded, so we have to actually use THIS data as a decoded - // pixel (note this can't be a valid pixel--one of RGB must be >= 128) - stbi_uc rgbe[4]; - rgbe[0] = (stbi_uc) c1; - rgbe[1] = (stbi_uc) c2; - rgbe[2] = (stbi_uc) len; - rgbe[3] = (stbi_uc) stbi__get8(s); - stbi__hdr_convert(hdr_data, rgbe, req_comp); - i = 1; - j = 0; - STBI_FREE(scanline); - goto main_decode_loop; // yes, this makes no sense - } - len <<= 8; - len |= stbi__get8(s); - if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } - if (scanline == NULL) { - scanline = (stbi_uc *) stbi__malloc_mad2(width, 4, 0); - if (!scanline) { - STBI_FREE(hdr_data); - return stbi__errpf("outofmem", "Out of memory"); - } - } - - for (k = 0; k < 4; ++k) { - int nleft; - i = 0; - while ((nleft = width - i) > 0) { - count = stbi__get8(s); - if (count > 128) { - // Run - value = stbi__get8(s); - count -= 128; - if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } - for (z = 0; z < count; ++z) - scanline[i++ * 4 + k] = value; - } else { - // Dump - if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } - for (z = 0; z < count; ++z) - scanline[i++ * 4 + k] = stbi__get8(s); - } - } - } - for (i=0; i < width; ++i) - stbi__hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); - } - if (scanline) - STBI_FREE(scanline); - } - - return hdr_data; +static int stbi__psd_is16(stbi__context *s) { + int channelCount, depth; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind(s); + return 0; + } + (void)stbi__get32be(s); + (void)stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind(s); + return 0; + } + return 1; } + #endif -static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) -{ - char buffer[STBI__HDR_BUFLEN]; - char *token; - int valid = 0; - int dummy; + #ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) { + int act_comp = 0, num_packets = 0, chained, dummy; + stbi__pic_packet packets[10]; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; - if (stbi__hdr_test(s) == 0) { - stbi__rewind( s ); - return 0; - } + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } - for(;;) { - token = stbi__hdr_gettoken(s,buffer); - if (token[0] == 0) break; - if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; - } + stbi__skip(s, 88); - if (!valid) { - stbi__rewind( s ); - return 0; - } - token = stbi__hdr_gettoken(s,buffer); - if (strncmp(token, "-Y ", 3)) { - stbi__rewind( s ); - return 0; - } - token += 3; - *y = (int) strtol(token, &token, 10); - while (*token == ' ') ++token; - if (strncmp(token, "+X ", 3)) { - stbi__rewind( s ); - return 0; - } - token += 3; - *x = (int) strtol(token, NULL, 10); - *comp = 3; - return 1; -} -#endif // STBI_NO_HDR + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind(s); + return 0; + } + if ((*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi__rewind(s); + return 0; + } -#ifndef STBI_NO_BMP -static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) -{ - void *p; - stbi__bmp_data info; + stbi__skip(s, 8); - info.all_a = 255; - p = stbi__bmp_parse_header(s, &info); - stbi__rewind( s ); - if (p == NULL) - return 0; - if (x) *x = s->img_x; - if (y) *y = s->img_y; - if (comp) { - if (info.bpp == 24 && info.ma == 0xff000000) - *comp = 3; - else - *comp = info.ma ? 4 : 3; - } - return 1; -} -#endif + do { + stbi__pic_packet *packet; -#ifndef STBI_NO_PSD -static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) -{ - int channelCount, dummy, depth; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; - if (stbi__get32be(s) != 0x38425053) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 1) { - stbi__rewind( s ); - return 0; - } - stbi__skip(s, 6); - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) { - stbi__rewind( s ); - return 0; - } - *y = stbi__get32be(s); - *x = stbi__get32be(s); - depth = stbi__get16be(s); - if (depth != 8 && depth != 16) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 3) { - stbi__rewind( s ); - return 0; - } - *comp = 4; - return 1; -} + if (num_packets == sizeof(packets) / sizeof(packets[0])) return 0; -static int stbi__psd_is16(stbi__context *s) -{ - int channelCount, depth; - if (stbi__get32be(s) != 0x38425053) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 1) { - stbi__rewind( s ); - return 0; - } - stbi__skip(s, 6); - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) { - stbi__rewind( s ); - return 0; - } - (void) stbi__get32be(s); - (void) stbi__get32be(s); - depth = stbi__get16be(s); - if (depth != 16) { - stbi__rewind( s ); - return 0; - } - return 1; -} -#endif + packet = &packets[num_packets++]; + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + act_comp |= packet->channel; -#ifndef STBI_NO_PIC -static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) -{ - int act_comp=0,num_packets=0,chained,dummy; - stbi__pic_packet packets[10]; - - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; - - if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) { + if (stbi__at_eof(s)) { stbi__rewind(s); return 0; - } - - stbi__skip(s, 88); - - *x = stbi__get16be(s); - *y = stbi__get16be(s); - if (stbi__at_eof(s)) { - stbi__rewind( s); + } + if (packet->size != 8) { + stbi__rewind(s); return 0; - } - if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { - stbi__rewind( s ); - return 0; - } + } + } while (chained); - stbi__skip(s, 8); + *comp = (act_comp & 0x10 ? 4 : 3); - do { - stbi__pic_packet *packet; - - if (num_packets==sizeof(packets)/sizeof(packets[0])) - return 0; - - packet = &packets[num_packets++]; - chained = stbi__get8(s); - packet->size = stbi__get8(s); - packet->type = stbi__get8(s); - packet->channel = stbi__get8(s); - act_comp |= packet->channel; - - if (stbi__at_eof(s)) { - stbi__rewind( s ); - return 0; - } - if (packet->size != 8) { - stbi__rewind( s ); - return 0; - } - } while (chained); - - *comp = (act_comp & 0x10 ? 4 : 3); - - return 1; + return 1; } -#endif + #endif // ************************************************************************************************* // Portable Gray Map and Portable Pixel Map loader @@ -7288,259 +7823,248 @@ static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) // Does not support ASCII image data (formats P2 and P3) // Does not support 16-bit-per-channel -#ifndef STBI_NO_PNM + #ifndef STBI_NO_PNM -static int stbi__pnm_test(stbi__context *s) -{ - char p, t; - p = (char) stbi__get8(s); - t = (char) stbi__get8(s); - if (p != 'P' || (t != '5' && t != '6')) { - stbi__rewind( s ); - return 0; - } - return 1; +static int stbi__pnm_test(stbi__context *s) { + char p, t; + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + return 1; } -static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *out; - STBI_NOTUSED(ri); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *out; + STBI_NOTUSED(ri); - if (!stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n)) - return 0; + if (!stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n)) + return 0; - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - *x = s->img_x; - *y = s->img_y; - if (comp) *comp = s->img_n; + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; - if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) - return stbi__errpuc("too large", "PNM too large"); + if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "PNM too large"); - out = (stbi_uc *) stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - stbi__getn(s, out, s->img_n * s->img_x * s->img_y); + out = (stbi_uc *)stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + stbi__getn(s, out, s->img_n * s->img_x * s->img_y); - if (req_comp && req_comp != s->img_n) { - out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } - return out; + if (req_comp && req_comp != s->img_n) { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; } -static int stbi__pnm_isspace(char c) -{ - return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +static int stbi__pnm_isspace(char c) { + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || + c == '\r'; } -static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) -{ - for (;;) { - while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) - *c = (char) stbi__get8(s); +static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) { + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + *c = (char)stbi__get8(s); - if (stbi__at_eof(s) || *c != '#') - break; + if (stbi__at_eof(s) || *c != '#') break; - while (!stbi__at_eof(s) && *c != '\n' && *c != '\r' ) - *c = (char) stbi__get8(s); - } + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r') + *c = (char)stbi__get8(s); + } } -static int stbi__pnm_isdigit(char c) -{ - return c >= '0' && c <= '9'; +static int stbi__pnm_isdigit(char c) { + return c >= '0' && c <= '9'; } -static int stbi__pnm_getinteger(stbi__context *s, char *c) -{ - int value = 0; +static int stbi__pnm_getinteger(stbi__context *s, char *c) { + int value = 0; - while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { - value = value*10 + (*c - '0'); - *c = (char) stbi__get8(s); - } + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value * 10 + (*c - '0'); + *c = (char)stbi__get8(s); + } - return value; + return value; } -static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) -{ - int maxv, dummy; - char c, p, t; +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) { + int maxv, dummy; + char c, p, t; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; - stbi__rewind(s); + stbi__rewind(s); - // Get identifier - p = (char) stbi__get8(s); - t = (char) stbi__get8(s); - if (p != 'P' || (t != '5' && t != '6')) { - stbi__rewind(s); - return 0; - } + // Get identifier + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } - *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + *comp = + (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm - c = (char) stbi__get8(s); - stbi__pnm_skip_whitespace(s, &c); + c = (char)stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); - *x = stbi__pnm_getinteger(s, &c); // read width - stbi__pnm_skip_whitespace(s, &c); + *x = stbi__pnm_getinteger(s, &c); // read width + stbi__pnm_skip_whitespace(s, &c); - *y = stbi__pnm_getinteger(s, &c); // read height - stbi__pnm_skip_whitespace(s, &c); + *y = stbi__pnm_getinteger(s, &c); // read height + stbi__pnm_skip_whitespace(s, &c); - maxv = stbi__pnm_getinteger(s, &c); // read max value + maxv = stbi__pnm_getinteger(s, &c); // read max value - if (maxv > 255) - return stbi__err("max value > 255", "PPM image not 8-bit"); - else - return 1; + if (maxv > 255) + return stbi__err("max value > 255", "PPM image not 8-bit"); + else + return 1; } -#endif + #endif -static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) -{ - #ifndef STBI_NO_JPEG - if (stbi__jpeg_info(s, x, y, comp)) return 1; - #endif +static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) { + #ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PNG - if (stbi__png_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_GIF - if (stbi__gif_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_BMP - if (stbi__bmp_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PSD - if (stbi__psd_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PIC - if (stbi__pic_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PNM - if (stbi__pnm_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_HDR - if (stbi__hdr_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; + #endif - // test tga last because it's a crappy test! - #ifndef STBI_NO_TGA - if (stbi__tga_info(s, x, y, comp)) - return 1; - #endif - return stbi__err("unknown image type", "Image not of any known type, or corrupt"); + // test tga last because it's a crappy test! + #ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) return 1; + #endif + return stbi__err("unknown image type", + "Image not of any known type, or corrupt"); } -static int stbi__is_16_main(stbi__context *s) -{ - #ifndef STBI_NO_PNG - if (stbi__png_is16(s)) return 1; - #endif +static int stbi__is_16_main(stbi__context *s) { + #ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; + #endif - #ifndef STBI_NO_PSD - if (stbi__psd_is16(s)) return 1; - #endif + #ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; + #endif - return 0; + return 0; } -#ifndef STBI_NO_STDIO -STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result; - if (!f) return stbi__err("can't fopen", "Unable to open file"); - result = stbi_info_from_file(f, x, y, comp); - fclose(f); - return result; + #ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) { + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; } -STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) -{ - int r; - stbi__context s; - long pos = ftell(f); - stbi__start_file(&s, f); - r = stbi__info_main(&s,x,y,comp); - fseek(f,pos,SEEK_SET); - return r; +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) { + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__info_main(&s, x, y, comp); + fseek(f, pos, SEEK_SET); + return r; } -STBIDEF int stbi_is_16_bit(char const *filename) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result; - if (!f) return stbi__err("can't fopen", "Unable to open file"); - result = stbi_is_16_bit_from_file(f); - fclose(f); - return result; +STBIDEF int stbi_is_16_bit(char const *filename) { + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; } -STBIDEF int stbi_is_16_bit_from_file(FILE *f) -{ - int r; - stbi__context s; - long pos = ftell(f); - stbi__start_file(&s, f); - r = stbi__is_16_main(&s); - fseek(f,pos,SEEK_SET); - return r; +STBIDEF int stbi_is_16_bit_from_file(FILE *f) { + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + fseek(f, pos, SEEK_SET); + return r; } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__info_main(&s,x,y,comp); +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__info_main(&s, x, y, comp); } -STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); - return stbi__info_main(&s,x,y,comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, + int *x, int *y, int *comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)c, user); + return stbi__info_main(&s, x, y, comp); } -STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__is_16_main(&s); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__is_16_main(&s); } -STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); - return stbi__is_16_main(&s); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, + void *user) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)c, user); + return stbi__is_16_main(&s); } -#endif // STB_IMAGE_IMPLEMENTATION +#endif // STB_IMAGE_IMPLEMENTATION /* revision history: - 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs - 2.19 (2018-02-11) fix warning - 2.18 (2018-01-30) fix warnings - 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and + platform ifdefs 2.19 (2018-02-11) fix warning 2.18 (2018-01-30) fix + warnings 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug 1-bit BMP *_is_16_bit api avoid warnings @@ -7555,13 +8079,11 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user warning fixes; disable run-time SSE detection on gcc; uniform handling of optional "return" values; thread-safe initialization of zlib tables - 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs - 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now - 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes - 2.11 (2016-04-02) allocate large structures on the stack - remove white matting for transparent PSD - fix reported channel count for PNG & BMP - re-enable SSE2 in non-gcc 64-bit + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet + JPGs 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now 2.12 + (2016-04-02) fix typo in 2.11 PSD fix that caused crashes 2.11 (2016-04-02) + allocate large structures on the stack remove white matting for transparent + PSD fix reported channel count for PNG & BMP re-enable SSE2 in non-gcc 64-bit support RGB-formatted JPEG read 16-bit PNGs (only as 8-bit) 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED @@ -7569,11 +8091,9 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 16-bit-per-pixel TGA (not bit-per-component) info() for TGA could break due to .hdr handling info() for BMP to shares code instead of sloppy parse - can use STBI_REALLOC_SIZED if allocator doesn't support realloc - code cleanup - 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA - 2.07 (2015-09-13) fix compiler warnings - partial animated GIF support + can use STBI_REALLOC_SIZED if allocator doesn't support + realloc code cleanup 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD + as RGBA 2.07 (2015-09-13) fix compiler warnings partial animated GIF support limited 16-bpc PSD support #ifdef unused functions bug with < 92 byte PIC,PNM,HDR,TGA @@ -7584,23 +8104,18 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user stbi_set_flip_vertically_on_load (nguillemot) fix NEON support; fix mingw support 2.02 (2015-01-19) fix incorrect assert, fix warning - 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 - 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG - 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) - progressive JPEG (stb) - PGM/PPM support (Ken Miller) - STBI_MALLOC,STBI_REALLOC,STBI_FREE + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit + without -msse2 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG 2.00 + (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) progressive + JPEG (stb) PGM/PPM support (Ken Miller) STBI_MALLOC,STBI_REALLOC,STBI_FREE GIF bugfix -- seemingly never worked STBI_NO_*, STBI_ONLY_* 1.48 (2014-12-14) fix incorrectly-named assert() - 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) - optimize PNG (ryg) - fix bug in interlaced PNG with user-specified channel count (stb) - 1.46 (2014-08-26) - fix broken tRNS chunk (colorkey-style transparency) in non-paletted PNG - 1.45 (2014-08-16) - fix MSVC-ARM internal compiler error by wrapping malloc - 1.44 (2014-08-07) + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar + Cornut & stb) optimize PNG (ryg) fix bug in interlaced PNG with + user-specified channel count (stb) 1.46 (2014-08-26) fix broken tRNS chunk + (colorkey-style transparency) in non-paletted PNG 1.45 (2014-08-16) fix + MSVC-ARM internal compiler error by wrapping malloc 1.44 (2014-08-07) various warning fixes from Ronny Chevalier 1.43 (2014-07-15) fix MSVC-only compiler problem in code changed in 1.42 @@ -7609,73 +8124,48 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user fixes to stbi__cleanup_jpeg path added STBI_ASSERT to avoid requiring assert.h 1.41 (2014-06-25) - fix search&replace from 1.36 that messed up comments/error messages - 1.40 (2014-06-22) - fix gcc struct-initialization warning - 1.39 (2014-06-15) - fix to TGA optimization when req_comp != number of components in TGA; - fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my test suite) - add support for BMP version 5 (more ignored fields) - 1.38 (2014-06-06) - suppress MSVC warnings on integer casts truncating values - fix accidental rename of 'skip' field of I/O - 1.37 (2014-06-04) - remove duplicate typedef - 1.36 (2014-06-03) - convert to header file single-file library - if de-iphone isn't set, load iphone images color-swapped instead of returning NULL - 1.35 (2014-05-27) - various warnings - fix broken STBI_SIMD path - fix bug where stbi_load_from_file no longer left file pointer in correct place - fix broken non-easy path for 32-bit BMP (possibly never used) - TGA optimization by Arseny Kapoulkine - 1.34 (unknown) - use STBI_NOTUSED in stbi__resample_row_generic(), fix one more leak in tga failure case - 1.33 (2011-07-14) - make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements - 1.32 (2011-07-13) - support for "info" function for all supported filetypes (SpartanJ) - 1.31 (2011-06-20) - a few more leak fixes, bug in PNG handling (SpartanJ) - 1.30 (2011-06-11) - added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + fix search&replace from 1.36 that messed up comments/error + messages 1.40 (2014-06-22) fix gcc struct-initialization warning 1.39 + (2014-06-15) fix to TGA optimization when req_comp != number of components in + TGA; fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my + test suite) add support for BMP version 5 (more ignored fields) 1.38 + (2014-06-06) suppress MSVC warnings on integer casts truncating values fix + accidental rename of 'skip' field of I/O 1.37 (2014-06-04) remove duplicate + typedef 1.36 (2014-06-03) convert to header file single-file library if + de-iphone isn't set, load iphone images color-swapped instead of returning + NULL 1.35 (2014-05-27) various warnings fix broken STBI_SIMD path fix bug + where stbi_load_from_file no longer left file pointer in correct place fix + broken non-easy path for 32-bit BMP (possibly never used) TGA optimization by + Arseny Kapoulkine 1.34 (unknown) use STBI_NOTUSED in + stbi__resample_row_generic(), fix one more leak in tga failure case 1.33 + (2011-07-14) make stbi_is_hdr work in STBI_NO_HDR (as specified), minor + compiler-friendly improvements 1.32 (2011-07-13) support for "info" function + for all supported filetypes (SpartanJ) 1.31 (2011-06-20) a few more leak + fixes, bug in PNG handling (SpartanJ) 1.30 (2011-06-11) added ability to + load files via callbacks to accomidate custom input streams (Ben Wenger) removed deprecated format-specific test/load functions - removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway - error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) - fix inefficiency in decoding 32-bit BMP (David Woo) - 1.29 (2010-08-16) - various warning fixes from Aurelien Pocheville - 1.28 (2010-08-01) - fix bug in GIF palette transparency (SpartanJ) - 1.27 (2010-08-01) - cast-to-stbi_uc to fix warnings - 1.26 (2010-07-24) - fix bug in file buffering for PNG reported by SpartanJ - 1.25 (2010-07-17) - refix trans_data warning (Won Chun) - 1.24 (2010-07-12) - perf improvements reading from files on platforms with lock-heavy fgetc() - minor perf improvements for jpeg - deprecated type-specific functions so we'll get feedback if they're needed - attempt to fix trans_data warning (Won Chun) - 1.23 fixed bug in iPhone support - 1.22 (2010-07-10) - removed image *writing* support - stbi_info support from Jetro Lauha - GIF support from Jean-Marc Lienher + removed support for installable file formats (stbi_loader) -- + would have been broken for IO callbacks anyway error cases in bmp and tga + give messages and don't leak (Raymond Barbiero, grisha) fix inefficiency in + decoding 32-bit BMP (David Woo) 1.29 (2010-08-16) various warning fixes from + Aurelien Pocheville 1.28 (2010-08-01) fix bug in GIF palette transparency + (SpartanJ) 1.27 (2010-08-01) cast-to-stbi_uc to fix warnings 1.26 + (2010-07-24) fix bug in file buffering for PNG reported by SpartanJ 1.25 + (2010-07-17) refix trans_data warning (Won Chun) 1.24 (2010-07-12) perf + improvements reading from files on platforms with lock-heavy fgetc() minor + perf improvements for jpeg deprecated type-specific functions so we'll get + feedback if they're needed attempt to fix trans_data warning (Won Chun) 1.23 + fixed bug in iPhone support 1.22 (2010-07-10) removed image *writing* + support stbi_info support from Jetro Lauha GIF support from Jean-Marc Lienher iPhone PNG-extensions from James Brown - warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. Janez (U+017D)emva) - 1.21 fix use of 'stbi_uc' in header (reported by jon blow) - 1.20 added support for Softimage PIC, by Tom Seddon - 1.19 bug in interlaced PNG corruption check (found by ryg) - 1.18 (2008-08-02) - fix a threading bug (local mutable static) - 1.17 support interlaced PNG - 1.16 major bugfix - stbi__convert_format converted one too many pixels - 1.15 initialize some fields for thread safety - 1.14 fix threadsafe conversion bug - header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. + Janez (U+017D)emva) 1.21 fix use of 'stbi_uc' in header (reported by jon + blow) 1.20 added support for Softimage PIC, by Tom Seddon 1.19 bug in + interlaced PNG corruption check (found by ryg) 1.18 (2008-08-02) fix a + threading bug (local mutable static) 1.17 support interlaced PNG 1.16 + major bugfix - stbi__convert_format converted one too many pixels 1.15 + initialize some fields for thread safety 1.14 fix threadsafe conversion + bug header-file-only version (#define STBI_HEADER_FILE_ONLY before including) 1.13 threadsafe 1.12 const qualifiers in the API 1.11 Support installable IDCT, colorspace conversion routines @@ -7685,15 +8175,14 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz 1.07 attempt to fix C++ warning/errors again 1.06 attempt to fix C++ warning/errors again - 1.05 fix TGA loading to return correct *comp and use good luminance calc - 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free - 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR - 1.02 support for (subset of) HDR files, float interface for preferred access to them - 1.01 fix bug: possible bug in handling right-side up bmps... not sure - fix bug: the stbi__bmp_load() and stbi__tga_load() functions didn't work at all - 1.00 interface to zlib that skips zlib header - 0.99 correct handling of alpha in palette - 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 1.05 fix TGA loading to return correct *comp and use good luminance + calc 1.04 default float alpha is 1, not 255; use 'void *' for + stbi_image_free 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR 1.02 support + for (subset of) HDR files, float interface for preferred access to them 1.01 + fix bug: possible bug in handling right-side up bmps... not sure fix bug: the + stbi__bmp_load() and stbi__tga_load() functions didn't work at all 1.00 + interface to zlib that skips zlib header 0.99 correct handling of alpha in + palette 0.98 TGA loader by lonesock; dynamically add loaders (untested) 0.97 jpeg errors on too large a file; also catch another malloc failure 0.96 fix detection of invalid v value - particleman@mollyrocket forum 0.95 during header scan, seek to markers in case of padding @@ -7706,8 +8195,8 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 0.60 fix compiling as c++ 0.59 fix warnings: merge Dave Moore's -Wall fixes 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian - 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available - 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but + less than 16 available 0.56 fix bug: zlib uncompressed mode len vs. nlen 0.55 fix bug: restart_interval not initialized to 0 0.54 allow NULL for 'int *comp' 0.53 fix bug in png 3->4; speedup png decoding @@ -7718,7 +8207,6 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user first released version */ - /* ------------------------------------------------------------------------------ This software is available under 2 licenses -- choose whichever you prefer. diff --git a/fuzzers/libfuzzer_stb_image_sugar/harness.c b/fuzzers/libfuzzer_stb_image_sugar/harness.c index 44b62a113b..dbf367e47e 100644 --- a/fuzzers/libfuzzer_stb_image_sugar/harness.c +++ b/fuzzers/libfuzzer_stb_image_sugar/harness.c @@ -9,20 +9,21 @@ #include "stb_image.h" -int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) -{ - int x, y, channels; +int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + int x, y, channels; - if(!stbi_info_from_memory(data, size, &x, &y, &channels)) return 0; + if (!stbi_info_from_memory(data, size, &x, &y, &channels)) { return 0; } - /* exit if the image is larger than ~80MB */ - if(y && x > (80000000 / 4) / y) return 0; + /* exit if the image is larger than ~80MB */ + if (y && x > (80000000 / 4) / y) { return 0; } - unsigned char *img = stbi_load_from_memory(data, size, &x, &y, &channels, 4); + unsigned char *img = stbi_load_from_memory(data, size, &x, &y, &channels, 4); - free(img); - - // if (x > 10000) free(img); // free crash + free(img); - return 0; + // if (x > 10000) free(img); // free crash + + // if (x > 10000) {free(img);} // free crash + + return 0; } diff --git a/fuzzers/libfuzzer_stb_image_sugar/stb_image.h b/fuzzers/libfuzzer_stb_image_sugar/stb_image.h index 94ae88d11d..eb0f8a3f71 100644 --- a/fuzzers/libfuzzer_stb_image_sugar/stb_image.h +++ b/fuzzers/libfuzzer_stb_image_sugar/stb_image.h @@ -3,7 +3,8 @@ Do this: #define STB_IMAGE_IMPLEMENTATION - before you include this file in *one* C or C++ file to create the implementation. + before you include this file in *one* C or C++ file to create the +implementation. // i.e. it should look like this: #include ... @@ -13,15 +14,16 @@ #include "stb_image.h" You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. - And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using +malloc,realloc,free QUICK NOTES: Primarily of interest to game developers and other people who can avoid problematic images and only need the trivial interface - JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) - PNG 1/2/4/8/16-bit-per-channel + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as +stock IJG lib) PNG 1/2/4/8/16-bit-per-channel TGA (not sure what subset, if a subset) BMP non-1bpp, non-RLE @@ -50,25 +52,22 @@ RECENT REVISION HISTORY: 2.26 (2020-07-13) many minor fixes 2.25 (2020-02-02) fix warnings - 2.24 (2020-02-02) fix warnings; thread-local failure_reason and flip_vertically - 2.23 (2019-08-11) fix clang static analysis warning - 2.22 (2019-03-04) gif fixes, fix warnings - 2.21 (2019-02-25) fix typo in comment - 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.24 (2020-02-02) fix warnings; thread-local failure_reason and +flip_vertically 2.23 (2019-08-11) fix clang static analysis warning 2.22 +(2019-03-04) gif fixes, fix warnings 2.21 (2019-02-25) fix typo in comment 2.20 +(2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs 2.19 (2018-02-11) fix warning 2.18 (2018-01-30) fix warnings 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings - 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes - 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC - 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs - 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes - 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes - 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 - RGB-format JPEG; remove white matting in PSD; - allocate large structures on the stack; - correct channel count for PNG & BMP - 2.10 (2016-01-22) avoid warning introduced in 2.09 - 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; +bugfixes 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE +detection on GCC 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for +Imagenet JPGs 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; +fixes 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes 2.11 +(2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 RGB-format JPEG; remove +white matting in PSD; allocate large structures on the stack; correct channel +count for PNG & BMP 2.10 (2016-01-22) avoid warning introduced in 2.09 2.09 +(2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED See end of file for full revision history. @@ -86,42 +85,41 @@ RECENT REVISION HISTORY: github:urraka (animated gif) Junggon Kim (PNM comments) Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) socks-the-fox (16-bit PNG) - Jeremy Sawicki (handle all ImageNet JPGs) - Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Jeremy Sawicki (handle all ImageNet +JPGs) Optimizations & bugfixes Mikhail Morozov (1-bit BMP) Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) Arseny Kapoulkine John-Mark Allen Carmelo J Fdez-Aguera Bug & warning fixes - Marc LeBlanc David Woo Guillaume George Martins Mozeiko - Christpher Lloyd Jerry Jansson Joseph Thomson Blazej Dariusz Roszkowski - Phil Jordan Dave Moore Roy Eltham - Hayaki Saito Nathan Reed Won Chun - Luke Graham Johan Duparc Nick Verigakis the Horde3D community - Thomas Ruf Ronny Chevalier github:rlyeh - Janez Zemva John Bartholomew Michal Cichon github:romigrou - Jonathan Blow Ken Hamada Tero Hanninen github:svdijk - Laurent Gomila Cort Stratton github:snagar - Aruelien Pocheville Sergio Gonzalez Thibault Reuille github:Zelex - Cass Everitt Ryamond Barbiero github:grim210 - Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw - Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus - Josh Tobin Matthew Gregan github:poppolopoppo - Julian Raschke Gregory Mullen Christian Floisand github:darealshinji - Baldur Karlsson Kevin Schmidt JR Smith github:Michaelangel007 - Brad Weinberger Matvey Cherevko [reserved] - Luca Sas Alexander Veselov Zack Middleton [reserved] + Marc LeBlanc David Woo Guillaume George Martins +Mozeiko Christpher Lloyd Jerry Jansson Joseph Thomson Blazej +Dariusz Roszkowski Phil Jordan Dave Moore Roy +Eltham Hayaki Saito Nathan Reed Won Chun Luke Graham Johan +Duparc Nick Verigakis the Horde3D community Thomas Ruf Ronny +Chevalier github:rlyeh Janez Zemva John +Bartholomew Michal Cichon github:romigrou Jonathan Blow Ken +Hamada Tero Hanninen github:svdijk Laurent Gomila Cort +Stratton github:snagar Aruelien Pocheville Sergio Gonzalez Thibault +Reuille github:Zelex Cass Everitt Ryamond Barbiero github:grim210 + Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw + Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus + Josh Tobin Matthew Gregan +github:poppolopoppo Julian Raschke Gregory Mullen Christian +Floisand github:darealshinji Baldur Karlsson Kevin Schmidt JR +Smith github:Michaelangel007 Brad Weinberger Matvey Cherevko +[reserved] Luca Sas Alexander Veselov Zack Middleton [reserved] Ryan C. Gordon [reserved] [reserved] DO NOT ADD YOUR NAME HERE - To add your name to the credits, pick a random blank space in the middle and fill it. - 80% of merge conflicts on stb PRs are due to people adding their name at the end - of the credits. + To add your name to the credits, pick a random blank space in the middle and +fill it. 80% of merge conflicts on stb PRs are due to people adding their name +at the end of the credits. */ #ifndef STBI_INCLUDE_STB_IMAGE_H -#define STBI_INCLUDE_STB_IMAGE_H + #define STBI_INCLUDE_STB_IMAGE_H // DOCUMENTATION // @@ -136,14 +134,15 @@ RECENT REVISION HISTORY: // // ... process data if not NULL ... // // ... x = width, y = height, n = # 8-bit components per pixel ... // // ... replace '0' with '1'..'4' to force that many components per pixel -// // ... but 'n' will always be the number that it would have been if you said 0 -// stbi_image_free(data) +// // ... but 'n' will always be the number that it would have been if you +// said 0 stbi_image_free(data) // // Standard parameters: // int *x -- outputs image width in pixels // int *y -- outputs image height in pixels // int *channels_in_file -- outputs # of image components in image file -// int desired_channels -- if non-zero, # of image components requested in result +// int desired_channels -- if non-zero, # of image components requested in +// result // // The return value from an image loader is an 'unsigned char *' which points // to the pixel data, or NULL on an allocation failure or if the image is @@ -171,8 +170,8 @@ RECENT REVISION HISTORY: // and *x, *y, *channels_in_file will be unchanged. The function // stbi_failure_reason() can be queried for an extremely brief, end-user // unfriendly explanation of why the load failed. Define STBI_NO_FAILURE_STRINGS -// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly -// more user-friendly ones. +// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get +// slightly more user-friendly ones. // // Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. // @@ -196,11 +195,12 @@ RECENT REVISION HISTORY: // 2. easy to maintain // 3. good performance // -// Sometimes I let "good performance" creep up in priority over "easy to maintain", -// and for best performance I may provide less-easy-to-use APIs that give higher -// performance, in addition to the easy-to-use ones. Nevertheless, it's important -// to keep in mind that from the standpoint of you, a client of this library, -// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// Sometimes I let "good performance" creep up in priority over "easy to +// maintain", and for best performance I may provide less-easy-to-use APIs that +// give higher performance, in addition to the easy-to-use ones. Nevertheless, +// it's important to keep in mind that from the standpoint of you, a client of +// this library, all you care about is #1 and #3, and stb libraries DO NOT +// emphasize #3 above all. // // Some secondary priorities arise directly from the first two, some of which // provide more explicit reasons why performance can't be emphasized. @@ -219,7 +219,8 @@ RECENT REVISION HISTORY: // overhead. // // The three functions you must define are "read" (reads some bytes of data), -// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the +// end). // // =========================================================================== // @@ -247,10 +248,11 @@ RECENT REVISION HISTORY: // HDR image support (disable by defining STBI_NO_HDR) // // stb_image supports loading HDR images in general, and currently the Radiance -// .HDR file format specifically. You can still load any file through the existing -// interface; if you attempt to load an HDR file, it will be automatically remapped -// to LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; -// both of these constants can be reconfigured through this interface: +// .HDR file format specifically. You can still load any file through the +// existing interface; if you attempt to load an HDR file, it will be +// automatically remapped to LDR, assuming gamma 2.2 and an arbitrary scale +// factor defaulting to 1; both of these constants can be reconfigured through +// this interface: // // stbi_hdr_to_ldr_gamma(2.2f); // stbi_hdr_to_ldr_scale(1.0f); @@ -336,37 +338,36 @@ RECENT REVISION HISTORY: // default this is set to (1 << 24), which is 16777216, but that's still // very big. -#ifndef STBI_NO_STDIO -#include -#endif // STBI_NO_STDIO + #ifndef STBI_NO_STDIO + #include + #endif // STBI_NO_STDIO -#define STBI_VERSION 1 + #define STBI_VERSION 1 -enum -{ - STBI_default = 0, // only used for desired_channels +enum { + STBI_default = 0, // only used for desired_channels - STBI_grey = 1, - STBI_grey_alpha = 2, - STBI_rgb = 3, - STBI_rgb_alpha = 4 + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 }; -#include -typedef unsigned char stbi_uc; + #include +typedef unsigned char stbi_uc; typedef unsigned short stbi_us; -#ifdef __cplusplus + #ifdef __cplusplus extern "C" { -#endif + #endif -#ifndef STBIDEF -#ifdef STB_IMAGE_STATIC -#define STBIDEF static -#else -#define STBIDEF extern -#endif -#endif + #ifndef STBIDEF + #ifdef STB_IMAGE_STATIC + #define STBIDEF static + #else + #define STBIDEF extern + #endif + #endif ////////////////////////////////////////////////////////////////////////////// // @@ -377,11 +378,13 @@ extern "C" { // load image by filename, open file, or memory buffer // -typedef struct -{ - int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read - void (*skip) (void *user,int n); // skip the next 'n' bytes, or 'unget' the last -n bytes if negative - int (*eof) (void *user); // returns nonzero if we are at end of file/data +typedef struct { + int (*read)(void *user, char *data, + int size); // fill 'data' with 'size' bytes. return number of + // bytes actually read + void (*skip)(void *user, int n); // skip the next 'n' bytes, or 'unget' the + // last -n bytes if negative + int (*eof)(void *user); // returns nonzero if we are at end of file/data } stbi_io_callbacks; //////////////////////////////////// @@ -389,376 +392,420 @@ typedef struct // 8-bits-per-channel interface // -STBIDEF stbi_uc *stbi_load_from_memory (stbi_uc const *buffer, int len , int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk , void *user, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *channels_in_file, + int desired_channels); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); -#ifndef STBI_NO_STDIO -STBIDEF stbi_uc *stbi_load (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); -// for stbi_load_from_file, file pointer is left pointing immediately after image -#endif + #ifndef STBI_NO_STDIO +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + // for stbi_load_from_file, file pointer is left pointing immediately after + // image + #endif -#ifndef STBI_NO_GIF -STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp); -#endif + #ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, + int **delays, int *x, int *y, int *z, + int *comp, int req_comp); + #endif -#ifdef STBI_WINDOWS_UTF8 -STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input); -#endif + #ifdef STBI_WINDOWS_UTF8 +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, + const wchar_t *input); + #endif //////////////////////////////////// // // 16-bits-per-channel interface // -STBIDEF stbi_us *stbi_load_16_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, + int *x, int *y, int *channels_in_file, + int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); -#ifndef STBI_NO_STDIO -STBIDEF stbi_us *stbi_load_16 (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); -STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); -#endif + #ifndef STBI_NO_STDIO +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + #endif -//////////////////////////////////// -// -// float-per-channel interface -// -#ifndef STBI_NO_LINEAR - STBIDEF float *stbi_loadf_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); - STBIDEF float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + //////////////////////////////////// + // + // float-per-channel interface + // + #ifndef STBI_NO_LINEAR +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *channels_in_file, + int desired_channels); +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels); - #ifndef STBI_NO_STDIO - STBIDEF float *stbi_loadf (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); - STBIDEF float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); - #endif -#endif + #ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, + int *channels_in_file, int desired_channels); +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, + int *channels_in_file, + int desired_channels); + #endif + #endif -#ifndef STBI_NO_HDR - STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); - STBIDEF void stbi_hdr_to_ldr_scale(float scale); -#endif // STBI_NO_HDR + #ifndef STBI_NO_HDR +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); +STBIDEF void stbi_hdr_to_ldr_scale(float scale); + #endif // STBI_NO_HDR -#ifndef STBI_NO_LINEAR - STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); - STBIDEF void stbi_ldr_to_hdr_scale(float scale); -#endif // STBI_NO_LINEAR + #ifndef STBI_NO_LINEAR +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); +STBIDEF void stbi_ldr_to_hdr_scale(float scale); + #endif // STBI_NO_LINEAR // stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR -STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); -STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); -#ifndef STBI_NO_STDIO -STBIDEF int stbi_is_hdr (char const *filename); -STBIDEF int stbi_is_hdr_from_file(FILE *f); -#endif // STBI_NO_STDIO - +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, + void *user); +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); + #ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr(char const *filename); +STBIDEF int stbi_is_hdr_from_file(FILE *f); + #endif // STBI_NO_STDIO // get a VERY brief reason for failure // on most compilers (and ALL modern mainstream compilers) this is threadsafe -STBIDEF const char *stbi_failure_reason (void); +STBIDEF const char *stbi_failure_reason(void); // free the loaded image -- this is just free() -STBIDEF void stbi_image_free (void *retval_from_stbi_load); +STBIDEF void stbi_image_free(void *retval_from_stbi_load); // get image dimensions & components without fully decoding -STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); -STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); -STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); -STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, void *user); - -#ifndef STBI_NO_STDIO -STBIDEF int stbi_info (char const *filename, int *x, int *y, int *comp); -STBIDEF int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); -STBIDEF int stbi_is_16_bit (char const *filename); -STBIDEF int stbi_is_16_bit_from_file(FILE *f); -#endif - +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, + int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, + void *user); + #ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit(char const *filename); +STBIDEF int stbi_is_16_bit_from_file(FILE *f); + #endif // for image formats that explicitly notate that they have premultiplied alpha, // we just return the colors as stored in the file. set this flag to force // unpremultiplication. results are undefined if the unpremultiply overflow. -STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); +STBIDEF void stbi_set_unpremultiply_on_load( + int flag_true_if_should_unpremultiply); // indicate whether we should process iphone images back to canonical format, // or just pass them through "as-is" STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); -// flip the image vertically, so the first pixel in the output array is the bottom left +// flip the image vertically, so the first pixel in the output array is the +// bottom left STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); -// as above, but only applies to images loaded on the thread that calls the function -// this function is only available if your compiler supports thread-local variables; -// calling it will fail to link if your compiler doesn't -STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip); +// as above, but only applies to images loaded on the thread that calls the +// function this function is only available if your compiler supports +// thread-local variables; calling it will fail to link if your compiler doesn't +STBIDEF void stbi_set_flip_vertically_on_load_thread( + int flag_true_if_should_flip); // ZLIB client - used by PNG, available for other purposes -STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); -STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, + int initial_size, int *outlen); +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, + int len, + int initial_size, + int *outlen, + int parse_header); STBIDEF char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); -STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen); -STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); -STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); +STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, + int *outlen); +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen); - -#ifdef __cplusplus + #ifdef __cplusplus } -#endif + #endif // // //// end header file ///////////////////////////////////////////////////// -#endif // STBI_INCLUDE_STB_IMAGE_H +#endif // STBI_INCLUDE_STB_IMAGE_H #ifdef STB_IMAGE_IMPLEMENTATION -#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ - || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ - || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ - || defined(STBI_ONLY_ZLIB) - #ifndef STBI_ONLY_JPEG - #define STBI_NO_JPEG - #endif - #ifndef STBI_ONLY_PNG - #define STBI_NO_PNG - #endif - #ifndef STBI_ONLY_BMP - #define STBI_NO_BMP - #endif - #ifndef STBI_ONLY_PSD - #define STBI_NO_PSD - #endif - #ifndef STBI_ONLY_TGA - #define STBI_NO_TGA - #endif - #ifndef STBI_ONLY_GIF - #define STBI_NO_GIF - #endif - #ifndef STBI_ONLY_HDR - #define STBI_NO_HDR - #endif - #ifndef STBI_ONLY_PIC - #define STBI_NO_PIC - #endif - #ifndef STBI_ONLY_PNM - #define STBI_NO_PNM - #endif -#endif + #if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || \ + defined(STBI_ONLY_BMP) || defined(STBI_ONLY_TGA) || \ + defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) || \ + defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || \ + defined(STBI_ONLY_PNM) || defined(STBI_ONLY_ZLIB) + #ifndef STBI_ONLY_JPEG + #define STBI_NO_JPEG + #endif + #ifndef STBI_ONLY_PNG + #define STBI_NO_PNG + #endif + #ifndef STBI_ONLY_BMP + #define STBI_NO_BMP + #endif + #ifndef STBI_ONLY_PSD + #define STBI_NO_PSD + #endif + #ifndef STBI_ONLY_TGA + #define STBI_NO_TGA + #endif + #ifndef STBI_ONLY_GIF + #define STBI_NO_GIF + #endif + #ifndef STBI_ONLY_HDR + #define STBI_NO_HDR + #endif + #ifndef STBI_ONLY_PIC + #define STBI_NO_PIC + #endif + #ifndef STBI_ONLY_PNM + #define STBI_NO_PNM + #endif + #endif -#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) -#define STBI_NO_ZLIB -#endif + #if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && \ + !defined(STBI_NO_ZLIB) + #define STBI_NO_ZLIB + #endif + #include + #include // ptrdiff_t on osx + #include + #include + #include -#include -#include // ptrdiff_t on osx -#include -#include -#include + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) + #include // ldexp, pow + #endif -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -#include // ldexp, pow -#endif + #ifndef STBI_NO_STDIO + #include + #endif -#ifndef STBI_NO_STDIO -#include -#endif + #ifndef STBI_ASSERT + #include + #define STBI_ASSERT(x) assert(x) + #endif -#ifndef STBI_ASSERT -#include -#define STBI_ASSERT(x) assert(x) -#endif + #ifdef __cplusplus + #define STBI_EXTERN extern "C" + #else + #define STBI_EXTERN extern + #endif -#ifdef __cplusplus -#define STBI_EXTERN extern "C" -#else -#define STBI_EXTERN extern -#endif + #ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif + #else + #define stbi_inline __forceinline + #endif + #ifndef STBI_NO_THREAD_LOCALS + #if defined(__cplusplus) && __cplusplus >= 201103L + #define STBI_THREAD_LOCAL thread_local + #elif defined(__GNUC__) && __GNUC__ < 5 + #define STBI_THREAD_LOCAL __thread + #elif defined(_MSC_VER) + #define STBI_THREAD_LOCAL __declspec(thread) + #elif defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && \ + !defined(__STDC_NO_THREADS__) + #define STBI_THREAD_LOCAL _Thread_local + #endif -#ifndef _MSC_VER - #ifdef __cplusplus - #define stbi_inline inline - #else - #define stbi_inline - #endif -#else - #define stbi_inline __forceinline -#endif - -#ifndef STBI_NO_THREAD_LOCALS - #if defined(__cplusplus) && __cplusplus >= 201103L - #define STBI_THREAD_LOCAL thread_local - #elif defined(__GNUC__) && __GNUC__ < 5 - #define STBI_THREAD_LOCAL __thread - #elif defined(_MSC_VER) - #define STBI_THREAD_LOCAL __declspec(thread) - #elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && !defined(__STDC_NO_THREADS__) - #define STBI_THREAD_LOCAL _Thread_local - #endif - - #ifndef STBI_THREAD_LOCAL + #ifndef STBI_THREAD_LOCAL #if defined(__GNUC__) - #define STBI_THREAD_LOCAL __thread + #define STBI_THREAD_LOCAL __thread #endif - #endif -#endif + #endif + #endif -#ifdef _MSC_VER + #ifdef _MSC_VER typedef unsigned short stbi__uint16; -typedef signed short stbi__int16; +typedef signed short stbi__int16; typedef unsigned int stbi__uint32; -typedef signed int stbi__int32; -#else -#include +typedef signed int stbi__int32; + #else + #include typedef uint16_t stbi__uint16; typedef int16_t stbi__int16; typedef uint32_t stbi__uint32; typedef int32_t stbi__int32; -#endif + #endif // should produce compiler error if size is wrong -typedef unsigned char validate_uint32[sizeof(stbi__uint32)==4 ? 1 : -1]; +typedef unsigned char validate_uint32[sizeof(stbi__uint32) == 4 ? 1 : -1]; -#ifdef _MSC_VER -#define STBI_NOTUSED(v) (void)(v) -#else -#define STBI_NOTUSED(v) (void)sizeof(v) -#endif + #ifdef _MSC_VER + #define STBI_NOTUSED(v) (void)(v) + #else + #define STBI_NOTUSED(v) (void)sizeof(v) + #endif -#ifdef _MSC_VER -#define STBI_HAS_LROTL -#endif + #ifdef _MSC_VER + #define STBI_HAS_LROTL + #endif -#ifdef STBI_HAS_LROTL - #define stbi_lrot(x,y) _lrotl(x,y) -#else - #define stbi_lrot(x,y) (((x) << (y)) | ((x) >> (32 - (y)))) -#endif + #ifdef STBI_HAS_LROTL + #define stbi_lrot(x, y) _lrotl(x, y) + #else + #define stbi_lrot(x, y) (((x) << (y)) | ((x) >> (32 - (y)))) + #endif -#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) -// ok -#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) -// ok -#else -#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." -#endif + #if defined(STBI_MALLOC) && defined(STBI_FREE) && \ + (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) + // ok + #elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && \ + !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) + // ok + #else + #error \ + "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." + #endif -#ifndef STBI_MALLOC -#define STBI_MALLOC(sz) malloc(sz) -#define STBI_REALLOC(p,newsz) realloc(p,newsz) -#define STBI_FREE(p) free(p) -#endif + #ifndef STBI_MALLOC + #define STBI_MALLOC(sz) malloc(sz) + #define STBI_REALLOC(p, newsz) realloc(p, newsz) + #define STBI_FREE(p) free(p) + #endif -#ifndef STBI_REALLOC_SIZED -#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) -#endif + #ifndef STBI_REALLOC_SIZED + #define STBI_REALLOC_SIZED(p, oldsz, newsz) STBI_REALLOC(p, newsz) + #endif -// x86/x64 detection -#if defined(__x86_64__) || defined(_M_X64) -#define STBI__X64_TARGET -#elif defined(__i386) || defined(_M_IX86) -#define STBI__X86_TARGET -#endif + // x86/x64 detection + #if defined(__x86_64__) || defined(_M_X64) + #define STBI__X64_TARGET + #elif defined(__i386) || defined(_M_IX86) + #define STBI__X86_TARGET + #endif -#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) -// gcc doesn't support sse2 intrinsics unless you compile with -msse2, -// which in turn means it gets to use SSE2 everywhere. This is unfortunate, -// but previous attempts to provide the SSE2 functions with runtime -// detection caused numerous issues. The way architecture extensions are -// exposed in GCC/Clang is, sadly, not really suited for one-file libs. -// New behavior: if compiled with -msse2, we use SSE2 without any -// detection; if not, we don't use it at all. -#define STBI_NO_SIMD -#endif + #if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && \ + !defined(STBI_NO_SIMD) + // gcc doesn't support sse2 intrinsics unless you compile with -msse2, + // which in turn means it gets to use SSE2 everywhere. This is unfortunate, + // but previous attempts to provide the SSE2 functions with runtime + // detection caused numerous issues. The way architecture extensions are + // exposed in GCC/Clang is, sadly, not really suited for one-file libs. + // New behavior: if compiled with -msse2, we use SSE2 without any + // detection; if not, we don't use it at all. + #define STBI_NO_SIMD + #endif -#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) -// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET -// -// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the -// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. -// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not -// simultaneously enabling "-mstackrealign". -// -// See https://github.com/nothings/stb/issues/81 for more information. -// -// So default to no SSE2 on 32-bit MinGW. If you've read this far and added -// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. -#define STBI_NO_SIMD -#endif + #if defined(__MINGW32__) && defined(STBI__X86_TARGET) && \ + !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) + // Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid + // STBI__X64_TARGET + // + // 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the + // Windows ABI and VC++ as well as Windows DLLs don't maintain that + // invariant. As a result, enabling SSE2 on 32-bit MinGW is dangerous when + // not simultaneously enabling "-mstackrealign". + // + // See https://github.com/nothings/stb/issues/81 for more information. + // + // So default to no SSE2 on 32-bit MinGW. If you've read this far and added + // -mstackrealign to your build settings, feel free to #define + // STBI_MINGW_ENABLE_SSE2. + #define STBI_NO_SIMD + #endif -#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) -#define STBI_SSE2 -#include + #if !defined(STBI_NO_SIMD) && \ + (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) + #define STBI_SSE2 + #include -#ifdef _MSC_VER + #ifdef _MSC_VER -#if _MSC_VER >= 1400 // not VC6 -#include // __cpuid -static int stbi__cpuid3(void) -{ - int info[4]; - __cpuid(info,1); - return info[3]; + #if _MSC_VER >= 1400 // not VC6 + #include // __cpuid +static int stbi__cpuid3(void) { + int info[4]; + __cpuid(info, 1); + return info[3]; } -#else -static int stbi__cpuid3(void) -{ - int res; - __asm { + #else +static int stbi__cpuid3(void) { + int res; + __asm { mov eax,1 cpuid mov res,edx - } - return res; + } + return res; } -#endif + #endif -#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + #define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name -#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) -static int stbi__sse2_available(void) -{ - int info3 = stbi__cpuid3(); - return ((info3 >> 26) & 1) != 0; + #if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) { + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; } -#endif + #endif -#else // assume GCC-style if not VC++ -#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + #else // assume GCC-style if not VC++ + #define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) -#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) -static int stbi__sse2_available(void) -{ - // If we're even attempting to compile this on GCC/Clang, that means - // -msse2 is on, which means the compiler is allowed to use SSE2 - // instructions at will, and so are we. - return 1; + #if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) { + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; } -#endif + #endif -#endif -#endif + #endif + #endif -// ARM NEON -#if defined(STBI_NO_SIMD) && defined(STBI_NEON) -#undef STBI_NEON -#endif + // ARM NEON + #if defined(STBI_NO_SIMD) && defined(STBI_NEON) + #undef STBI_NEON + #endif -#ifdef STBI_NEON -#include -// assume GCC or Clang on ARM targets -#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) -#endif + #ifdef STBI_NEON + #include + // assume GCC or Clang on ARM targets + #define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + #endif -#ifndef STBI_SIMD_ALIGN -#define STBI_SIMD_ALIGN(type, name) type name -#endif + #ifndef STBI_SIMD_ALIGN + #define STBI_SIMD_ALIGN(type, name) type name + #endif -#ifndef STBI_MAX_DIMENSIONS -#define STBI_MAX_DIMENSIONS (1 << 24) -#endif + #ifndef STBI_MAX_DIMENSIONS + #define STBI_MAX_DIMENSIONS (1 << 24) + #endif /////////////////////////////////////////////// // @@ -766,188 +813,182 @@ static int stbi__sse2_available(void) // stbi__context structure is our basic context used by all images, so it // contains all the IO context, plus some basic image information -typedef struct -{ - stbi__uint32 img_x, img_y; - int img_n, img_out_n; +typedef struct { + stbi__uint32 img_x, img_y; + int img_n, img_out_n; - stbi_io_callbacks io; - void *io_user_data; + stbi_io_callbacks io; + void *io_user_data; - int read_from_callbacks; - int buflen; - stbi_uc buffer_start[128]; - int callback_already_read; + int read_from_callbacks; + int buflen; + stbi_uc buffer_start[128]; + int callback_already_read; - stbi_uc *img_buffer, *img_buffer_end; - stbi_uc *img_buffer_original, *img_buffer_original_end; + stbi_uc *img_buffer, *img_buffer_end; + stbi_uc *img_buffer_original, *img_buffer_original_end; } stbi__context; - static void stbi__refill_buffer(stbi__context *s); // initialize a memory-decode context -static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) -{ - s->io.read = NULL; - s->read_from_callbacks = 0; - s->callback_already_read = 0; - s->img_buffer = s->img_buffer_original = (stbi_uc *) buffer; - s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *) buffer+len; +static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) { + s->io.read = NULL; + s->read_from_callbacks = 0; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = (stbi_uc *)buffer; + s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *)buffer + len; } // initialize a callback-based context -static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, void *user) -{ - s->io = *c; - s->io_user_data = user; - s->buflen = sizeof(s->buffer_start); - s->read_from_callbacks = 1; - s->callback_already_read = 0; - s->img_buffer = s->img_buffer_original = s->buffer_start; - stbi__refill_buffer(s); - s->img_buffer_original_end = s->img_buffer_end; +static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, + void *user) { + s->io = *c; + s->io_user_data = user; + s->buflen = sizeof(s->buffer_start); + s->read_from_callbacks = 1; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = s->buffer_start; + stbi__refill_buffer(s); + s->img_buffer_original_end = s->img_buffer_end; } -#ifndef STBI_NO_STDIO + #ifndef STBI_NO_STDIO -static int stbi__stdio_read(void *user, char *data, int size) -{ - return (int) fread(data,1,size,(FILE*) user); +static int stbi__stdio_read(void *user, char *data, int size) { + return (int)fread(data, 1, size, (FILE *)user); } -static void stbi__stdio_skip(void *user, int n) -{ - int ch; - fseek((FILE*) user, n, SEEK_CUR); - ch = fgetc((FILE*) user); /* have to read a byte to reset feof()'s flag */ - if (ch != EOF) { - ungetc(ch, (FILE *) user); /* push byte back onto stream if valid. */ - } +static void stbi__stdio_skip(void *user, int n) { + int ch; + fseek((FILE *)user, n, SEEK_CUR); + ch = fgetc((FILE *)user); /* have to read a byte to reset feof()'s flag */ + if (ch != EOF) { + ungetc(ch, (FILE *)user); /* push byte back onto stream if valid. */ + } } -static int stbi__stdio_eof(void *user) -{ - return feof((FILE*) user) || ferror((FILE *) user); +static int stbi__stdio_eof(void *user) { + return feof((FILE *)user) || ferror((FILE *)user); } -static stbi_io_callbacks stbi__stdio_callbacks = -{ - stbi__stdio_read, - stbi__stdio_skip, - stbi__stdio_eof, +static stbi_io_callbacks stbi__stdio_callbacks = { + stbi__stdio_read, + stbi__stdio_skip, + stbi__stdio_eof, }; -static void stbi__start_file(stbi__context *s, FILE *f) -{ - stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *) f); +static void stbi__start_file(stbi__context *s, FILE *f) { + stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *)f); } -//static void stop_file(stbi__context *s) { } + // static void stop_file(stbi__context *s) { } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -static void stbi__rewind(stbi__context *s) -{ - // conceptually rewind SHOULD rewind to the beginning of the stream, - // but we just rewind to the beginning of the initial buffer, because - // we only use it after doing 'test', which only ever looks at at most 92 bytes - s->img_buffer = s->img_buffer_original; - s->img_buffer_end = s->img_buffer_original_end; +static void stbi__rewind(stbi__context *s) { + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 + // bytes + s->img_buffer = s->img_buffer_original; + s->img_buffer_end = s->img_buffer_original_end; } -enum -{ - STBI_ORDER_RGB, - STBI_ORDER_BGR -}; +enum { STBI_ORDER_RGB, STBI_ORDER_BGR }; -typedef struct -{ - int bits_per_channel; - int num_channels; - int channel_order; +typedef struct { + int bits_per_channel; + int num_channels; + int channel_order; } stbi__result_info; -#ifndef STBI_NO_JPEG -static int stbi__jpeg_test(stbi__context *s); -static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context *s); +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PNG -static int stbi__png_test(stbi__context *s); -static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); -static int stbi__png_is16(stbi__context *s); -#endif + #ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context *s); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__png_is16(stbi__context *s); + #endif -#ifndef STBI_NO_BMP -static int stbi__bmp_test(stbi__context *s); -static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context *s); +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_TGA -static int stbi__tga_test(stbi__context *s); -static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s); +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PSD -static int stbi__psd_test(stbi__context *s); -static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc); -static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); -static int stbi__psd_is16(stbi__context *s); -#endif + #ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc); +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__psd_is16(stbi__context *s); + #endif -#ifndef STBI_NO_HDR -static int stbi__hdr_test(stbi__context *s); -static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_HDR +static int stbi__hdr_test(stbi__context *s); +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PIC -static int stbi__pic_test(stbi__context *s); -static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context *s); +static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_GIF -static int stbi__gif_test(stbi__context *s); -static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp); -static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context *s); +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, + int *z, int *comp, int req_comp); +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); + #endif -#ifndef STBI_NO_PNM -static int stbi__pnm_test(stbi__context *s); -static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); -static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); -#endif + #ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context *s); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri); +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); + #endif static -#ifdef STBI_THREAD_LOCAL -STBI_THREAD_LOCAL -#endif -const char *stbi__g_failure_reason; + #ifdef STBI_THREAD_LOCAL + STBI_THREAD_LOCAL + #endif + const char *stbi__g_failure_reason; -STBIDEF const char *stbi_failure_reason(void) -{ - return stbi__g_failure_reason; +STBIDEF const char *stbi_failure_reason(void) { + return stbi__g_failure_reason; } -#ifndef STBI_NO_FAILURE_STRINGS -static int stbi__err(const char *str) -{ - stbi__g_failure_reason = str; - return 0; + #ifndef STBI_NO_FAILURE_STRINGS +static int stbi__err(const char *str) { + stbi__g_failure_reason = str; + return 0; } -#endif + #endif -static void *stbi__malloc(size_t size) -{ - return STBI_MALLOC(size); +static void *stbi__malloc(size_t size) { + return STBI_MALLOC(size); } // stb_image uses ints pervasively, including for offset calculations. @@ -962,317 +1003,334 @@ static void *stbi__malloc(size_t size) // return 1 if the sum is valid, 0 on overflow. // negative terms are considered invalid. -static int stbi__addsizes_valid(int a, int b) -{ - if (b < 0) return 0; - // now 0 <= b <= INT_MAX, hence also - // 0 <= INT_MAX - b <= INTMAX. - // And "a + b <= INT_MAX" (which might overflow) is the - // same as a <= INT_MAX - b (no overflow) - return a <= INT_MAX - b; +static int stbi__addsizes_valid(int a, int b) { + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; } // returns 1 if the product is valid, 0 on overflow. // negative factors are considered invalid. -static int stbi__mul2sizes_valid(int a, int b) -{ - if (a < 0 || b < 0) return 0; - if (b == 0) return 1; // mul-by-0 is always safe - // portable way to check for no overflows in a*b - return a <= INT_MAX/b; +static int stbi__mul2sizes_valid(int a, int b) { + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX / b; } -#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) + #if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || \ + !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) // returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow -static int stbi__mad2sizes_valid(int a, int b, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a*b, add); +static int stbi__mad2sizes_valid(int a, int b, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a * b, add); } -#endif + #endif // returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow -static int stbi__mad3sizes_valid(int a, int b, int c, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && - stbi__addsizes_valid(a*b*c, add); +static int stbi__mad3sizes_valid(int a, int b, int c, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__addsizes_valid(a * b * c, add); } -// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) -{ - return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && - stbi__mul2sizes_valid(a*b*c, d) && stbi__addsizes_valid(a*b*c*d, add); + // returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't + // overflow + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) { + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__mul2sizes_valid(a * b * c, d) && + stbi__addsizes_valid(a * b * c * d, add); } -#endif + #endif -#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) + #if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || \ + !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) // mallocs with size overflow checking -static void *stbi__malloc_mad2(int a, int b, int add) -{ - if (!stbi__mad2sizes_valid(a, b, add)) return NULL; - return stbi__malloc(a*b + add); +static void *stbi__malloc_mad2(int a, int b, int add) { + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a * b + add); } -#endif + #endif -static void *stbi__malloc_mad3(int a, int b, int c, int add) -{ - if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; - return stbi__malloc(a*b*c + add); +static void *stbi__malloc_mad3(int a, int b, int c, int add) { + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a * b * c + add); } -#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) -static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) -{ - if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; - return stbi__malloc(a*b*c*d + add); + #if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) { + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a * b * c * d + add); } -#endif + #endif // stbi__err - error // stbi__errpf - error returning pointer to float // stbi__errpuc - error returning pointer to unsigned char -#ifdef STBI_NO_FAILURE_STRINGS - #define stbi__err(x,y) 0 -#elif defined(STBI_FAILURE_USERMSG) - #define stbi__err(x,y) stbi__err(y) -#else - #define stbi__err(x,y) stbi__err(x) -#endif + #ifdef STBI_NO_FAILURE_STRINGS + #define stbi__err(x, y) 0 + #elif defined(STBI_FAILURE_USERMSG) + #define stbi__err(x, y) stbi__err(y) + #else + #define stbi__err(x, y) stbi__err(x) + #endif -#define stbi__errpf(x,y) ((float *)(size_t) (stbi__err(x,y)?NULL:NULL)) -#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + #define stbi__errpf(x, y) ((float *)(size_t)(stbi__err(x, y) ? NULL : NULL)) + #define stbi__errpuc(x, y) \ + ((unsigned char *)(size_t)(stbi__err(x, y) ? NULL : NULL)) -STBIDEF void stbi_image_free(void *retval_from_stbi_load) -{ - STBI_FREE(retval_from_stbi_load); +STBIDEF void stbi_image_free(void *retval_from_stbi_load) { + STBI_FREE(retval_from_stbi_load); } -#ifndef STBI_NO_LINEAR -static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); -#endif + #ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); + #endif -#ifndef STBI_NO_HDR -static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); -#endif + #ifndef STBI_NO_HDR +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); + #endif static int stbi__vertically_flip_on_load_global = 0; -STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) -{ - stbi__vertically_flip_on_load_global = flag_true_if_should_flip; +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) { + stbi__vertically_flip_on_load_global = flag_true_if_should_flip; } -#ifndef STBI_THREAD_LOCAL -#define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global -#else -static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, stbi__vertically_flip_on_load_set; + #ifndef STBI_THREAD_LOCAL + #define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global + #else +static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, + stbi__vertically_flip_on_load_set; -STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip) -{ - stbi__vertically_flip_on_load_local = flag_true_if_should_flip; - stbi__vertically_flip_on_load_set = 1; +STBIDEF void stbi_set_flip_vertically_on_load_thread( + int flag_true_if_should_flip) { + stbi__vertically_flip_on_load_local = flag_true_if_should_flip; + stbi__vertically_flip_on_load_set = 1; } -#define stbi__vertically_flip_on_load (stbi__vertically_flip_on_load_set \ - ? stbi__vertically_flip_on_load_local \ - : stbi__vertically_flip_on_load_global) -#endif // STBI_THREAD_LOCAL + #define stbi__vertically_flip_on_load \ + (stbi__vertically_flip_on_load_set \ + ? stbi__vertically_flip_on_load_local \ + : stbi__vertically_flip_on_load_global) + #endif // STBI_THREAD_LOCAL -static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) -{ - memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields - ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed - ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order - ri->num_channels = 0; +static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc) { + memset(ri, 0, + sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = + 8; // default is 8 so most paths don't have to be changed + ri->channel_order = + STBI_ORDER_RGB; // all current input & output are this, but this is here + // so we can add BGR order + ri->num_channels = 0; - #ifndef STBI_NO_JPEG - if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PNG - if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_BMP - if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_GIF - if (stbi__gif_test(s)) return stbi__gif_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PSD - if (stbi__psd_test(s)) return stbi__psd_load(s,x,y,comp,req_comp, ri, bpc); - #else - STBI_NOTUSED(bpc); - #endif - #ifndef STBI_NO_PIC - if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp, ri); - #endif - #ifndef STBI_NO_PNM - if (stbi__pnm_test(s)) return stbi__pnm_load(s,x,y,comp,req_comp, ri); - #endif + #ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PSD + if (stbi__psd_test(s)) + return stbi__psd_load(s, x, y, comp, req_comp, ri, bpc); + #else + STBI_NOTUSED(bpc); + #endif + #ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s, x, y, comp, req_comp, ri); + #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s, x, y, comp, req_comp, ri); + #endif - #ifndef STBI_NO_HDR - if (stbi__hdr_test(s)) { - float *hdr = stbi__hdr_load(s, x,y,comp,req_comp, ri); - return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); - } - #endif + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float *hdr = stbi__hdr_load(s, x, y, comp, req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif - #ifndef STBI_NO_TGA - // test tga last because it's a crappy test! - if (stbi__tga_test(s)) - return stbi__tga_load(s,x,y,comp,req_comp, ri); - #endif + #ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) return stbi__tga_load(s, x, y, comp, req_comp, ri); + #endif - return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); + return stbi__errpuc("unknown image type", + "Image not of any known type, or corrupt"); } -static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, int channels) -{ - int i; - int img_len = w * h * channels; - stbi_uc *reduced; +static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, + int channels) { + int i; + int img_len = w * h * channels; + stbi_uc *reduced; - reduced = (stbi_uc *) stbi__malloc(img_len); - if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + reduced = (stbi_uc *)stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); - for (i = 0; i < img_len; ++i) - reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + for (i = 0; i < img_len; ++i) + reduced[i] = (stbi_uc)((orig[i] >> 8) & + 0xFF); // top half of each byte is sufficient approx + // of 16->8 bit scaling - STBI_FREE(orig); - return reduced; + STBI_FREE(orig); + return reduced; } -static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, int channels) -{ - int i; - int img_len = w * h * channels; - stbi__uint16 *enlarged; +static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, + int channels) { + int i; + int img_len = w * h * channels; + stbi__uint16 *enlarged; - enlarged = (stbi__uint16 *) stbi__malloc(img_len*2); - if (enlarged == NULL) return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + enlarged = (stbi__uint16 *)stbi__malloc(img_len * 2); + if (enlarged == NULL) + return (stbi__uint16 *)stbi__errpuc("outofmem", "Out of memory"); - for (i = 0; i < img_len; ++i) - enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + for (i = 0; i < img_len; ++i) + enlarged[i] = (stbi__uint16)((orig[i] << 8) + + orig[i]); // replicate to high and low byte, + // maps 0->0, 255->0xffff - STBI_FREE(orig); - return enlarged; + STBI_FREE(orig); + return enlarged; } -static void stbi__vertical_flip(void *image, int w, int h, int bytes_per_pixel) -{ - int row; - size_t bytes_per_row = (size_t)w * bytes_per_pixel; - stbi_uc temp[2048]; - stbi_uc *bytes = (stbi_uc *)image; +static void stbi__vertical_flip(void *image, int w, int h, + int bytes_per_pixel) { + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc *bytes = (stbi_uc *)image; - for (row = 0; row < (h>>1); row++) { - stbi_uc *row0 = bytes + row*bytes_per_row; - stbi_uc *row1 = bytes + (h - row - 1)*bytes_per_row; - // swap row0 with row1 - size_t bytes_left = bytes_per_row; - while (bytes_left) { - size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); - memcpy(temp, row0, bytes_copy); - memcpy(row0, row1, bytes_copy); - memcpy(row1, temp, bytes_copy); - row0 += bytes_copy; - row1 += bytes_copy; - bytes_left -= bytes_copy; - } - } + for (row = 0; row < (h >> 1); row++) { + stbi_uc *row0 = bytes + row * bytes_per_row; + stbi_uc *row1 = bytes + (h - row - 1) * bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = + (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } } -#ifndef STBI_NO_GIF -static void stbi__vertical_flip_slices(void *image, int w, int h, int z, int bytes_per_pixel) -{ - int slice; - int slice_size = w * h * bytes_per_pixel; + #ifndef STBI_NO_GIF +static void stbi__vertical_flip_slices(void *image, int w, int h, int z, + int bytes_per_pixel) { + int slice; + int slice_size = w * h * bytes_per_pixel; - stbi_uc *bytes = (stbi_uc *)image; - for (slice = 0; slice < z; ++slice) { - stbi__vertical_flip(bytes, w, h, bytes_per_pixel); - bytes += slice_size; - } + stbi_uc *bytes = (stbi_uc *)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } } -#endif + #endif -static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - stbi__result_info ri; - void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); +static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, + int *y, int *comp, + int req_comp) { + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); - if (result == NULL) - return NULL; + if (result == NULL) return NULL; - // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. - STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + // it is the responsibility of the loaders to make sure we get either 8 or 16 + // bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); - if (ri.bits_per_channel != 8) { - result = stbi__convert_16_to_8((stbi__uint16 *) result, *x, *y, req_comp == 0 ? *comp : req_comp); - ri.bits_per_channel = 8; - } + if (ri.bits_per_channel != 8) { + result = stbi__convert_16_to_8((stbi__uint16 *)result, *x, *y, + req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } - // @TODO: move stbi__convert_format to here + // @TODO: move stbi__convert_format to here - if (stbi__vertically_flip_on_load) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); - } + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } - return (unsigned char *) result; + return (unsigned char *)result; } -static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - stbi__result_info ri; - void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); +static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, + int *y, int *comp, + int req_comp) { + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); - if (result == NULL) - return NULL; + if (result == NULL) return NULL; - // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. - STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + // it is the responsibility of the loaders to make sure we get either 8 or 16 + // bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); - if (ri.bits_per_channel != 16) { - result = stbi__convert_8_to_16((stbi_uc *) result, *x, *y, req_comp == 0 ? *comp : req_comp); - ri.bits_per_channel = 16; - } + if (ri.bits_per_channel != 16) { + result = stbi__convert_8_to_16((stbi_uc *)result, *x, *y, + req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } - // @TODO: move stbi__convert_format16 to here - // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to + // keep more precision - if (stbi__vertically_flip_on_load) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); - } + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } - return (stbi__uint16 *) result; + return (stbi__uint16 *)result; } -#if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) -static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, int req_comp) -{ - if (stbi__vertically_flip_on_load && result != NULL) { - int channels = req_comp ? req_comp : *comp; - stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); - } + #if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, + int req_comp) { + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } } -#endif + #endif -#ifndef STBI_NO_STDIO + #ifndef STBI_NO_STDIO -#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) -STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide); -STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default); -#endif + #if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) +STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar( + unsigned int cp, unsigned long flags, const char *str, int cbmb, + wchar_t *widestr, int cchwide); +STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte( + unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, + char *str, int cbmb, const char *defchar, int *used_default); + #endif #if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input) { return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL); } -#endif + #endif static FILE *stbi__fopen(char const *filename, char const *mode) { @@ -1293,384 +1351,386 @@ static FILE *stbi__fopen(char const *filename, char const *mode) f = _wfopen(wFilename, wMode); #endif -#elif defined(_MSC_VER) && _MSC_VER >= 1400 - if (0 != fopen_s(&f, filename, mode)) - f=0; -#else - f = fopen(filename, mode); -#endif - return f; + #elif defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) f = 0; + #else + f = fopen(filename, mode); + #endif + return f; } - -STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - unsigned char *result; - if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); - result = stbi_load_from_file(f,x,y,comp,req_comp); - fclose(f); - return result; +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, + int req_comp) { + FILE *f = stbi__fopen(filename, "rb"); + unsigned char *result; + if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f, x, y, comp, req_comp); + fclose(f); + return result; } -STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - unsigned char *result; - stbi__context s; - stbi__start_file(&s,f); - result = stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); - if (result) { - // need to 'unget' all the characters in the IO buffer - fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); - } - return result; +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, + int req_comp) { + unsigned char *result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, -(int)(s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; } -STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - stbi__uint16 *result; - stbi__context s; - stbi__start_file(&s,f); - result = stbi__load_and_postprocess_16bit(&s,x,y,comp,req_comp); - if (result) { - // need to 'unget' all the characters in the IO buffer - fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); - } - return result; +STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, + int req_comp) { + stbi__uint16 *result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_16bit(&s, x, y, comp, req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, -(int)(s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; } -STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - stbi__uint16 *result; - if (!f) return (stbi_us *) stbi__errpuc("can't fopen", "Unable to open file"); - result = stbi_load_from_file_16(f,x,y,comp,req_comp); - fclose(f); - return result; +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, + int req_comp) { + FILE *f = stbi__fopen(filename, "rb"); + stbi__uint16 *result; + if (!f) return (stbi_us *)stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f, x, y, comp, req_comp); + fclose(f); + return result; } + #endif //! STBI_NO_STDIO -#endif //!STBI_NO_STDIO - -STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, + int *x, int *y, int *channels_in_file, + int desired_channels) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, + desired_channels); } -STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); - return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, + int *channels_in_file, + int desired_channels) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, + desired_channels); } -STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp, int req_comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); } -STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_8bit(&s, x, y, comp, req_comp); } -#ifndef STBI_NO_GIF -STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp) -{ - unsigned char *result; - stbi__context s; - stbi__start_mem(&s,buffer,len); + #ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, + int **delays, int *x, int *y, int *z, + int *comp, int req_comp) { + unsigned char *result; + stbi__context s; + stbi__start_mem(&s, buffer, len); - result = (unsigned char*) stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); - if (stbi__vertically_flip_on_load) { - stbi__vertical_flip_slices( result, *x, *y, *z, *comp ); - } + result = + (unsigned char *)stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices(result, *x, *y, *z, *comp); + } - return result; + return result; } -#endif + #endif -#ifndef STBI_NO_LINEAR -static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) -{ - unsigned char *data; - #ifndef STBI_NO_HDR - if (stbi__hdr_test(s)) { - stbi__result_info ri; - float *hdr_data = stbi__hdr_load(s,x,y,comp,req_comp, &ri); - if (hdr_data) - stbi__float_postprocess(hdr_data,x,y,comp,req_comp); - return hdr_data; - } - #endif - data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); - if (data) - return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); - return stbi__errpf("unknown image type", "Image not of any known type, or corrupt"); + #ifndef STBI_NO_LINEAR +static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, + int req_comp) { + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + stbi__result_info ri; + float *hdr_data = stbi__hdr_load(s, x, y, comp, req_comp, &ri); + if (hdr_data) stbi__float_postprocess(hdr_data, x, y, comp, req_comp); + return hdr_data; + } + #endif + data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); + if (data) return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return stbi__errpf("unknown image type", + "Image not of any known type, or corrupt"); } -STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp, int req_comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, + void *user, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -#ifndef STBI_NO_STDIO -STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) -{ - float *result; - FILE *f = stbi__fopen(filename, "rb"); - if (!f) return stbi__errpf("can't fopen", "Unable to open file"); - result = stbi_loadf_from_file(f,x,y,comp,req_comp); - fclose(f); - return result; + #ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, + int req_comp) { + float *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) return stbi__errpf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f, x, y, comp, req_comp); + fclose(f); + return result; } -STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) -{ - stbi__context s; - stbi__start_file(&s,f); - return stbi__loadf_main(&s,x,y,comp,req_comp); +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, + int req_comp) { + stbi__context s; + stbi__start_file(&s, f); + return stbi__loadf_main(&s, x, y, comp, req_comp); } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -#endif // !STBI_NO_LINEAR + #endif // !STBI_NO_LINEAR // these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is // defined, for API simplicity; if STBI_NO_LINEAR is defined, it always // reports false! -STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) -{ - #ifndef STBI_NO_HDR - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__hdr_test(&s); - #else - STBI_NOTUSED(buffer); - STBI_NOTUSED(len); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) { + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif } -#ifndef STBI_NO_STDIO -STBIDEF int stbi_is_hdr (char const *filename) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result=0; - if (f) { - result = stbi_is_hdr_from_file(f); - fclose(f); - } - return result; + #ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr(char const *filename) { + FILE *f = stbi__fopen(filename, "rb"); + int result = 0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; } -STBIDEF int stbi_is_hdr_from_file(FILE *f) -{ - #ifndef STBI_NO_HDR - long pos = ftell(f); - int res; - stbi__context s; - stbi__start_file(&s,f); - res = stbi__hdr_test(&s); - fseek(f, pos, SEEK_SET); - return res; - #else - STBI_NOTUSED(f); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_file(FILE *f) { + #ifndef STBI_NO_HDR + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s, f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; + #else + STBI_NOTUSED(f); + return 0; + #endif } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) -{ - #ifndef STBI_NO_HDR - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); - return stbi__hdr_test(&s); - #else - STBI_NOTUSED(clbk); - STBI_NOTUSED(user); - return 0; - #endif +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, + void *user) { + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; + #endif } -#ifndef STBI_NO_LINEAR -static float stbi__l2h_gamma=2.2f, stbi__l2h_scale=1.0f; + #ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma = 2.2f, stbi__l2h_scale = 1.0f; -STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } -STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } -#endif +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { + stbi__l2h_gamma = gamma; +} +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { + stbi__l2h_scale = scale; +} + #endif -static float stbi__h2l_gamma_i=1.0f/2.2f, stbi__h2l_scale_i=1.0f; - -STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1/gamma; } -STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1/scale; } +static float stbi__h2l_gamma_i = 1.0f / 2.2f, stbi__h2l_scale_i = 1.0f; +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { + stbi__h2l_gamma_i = 1 / gamma; +} +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { + stbi__h2l_scale_i = 1 / scale; +} ////////////////////////////////////////////////////////////////////////////// // // Common code used by all image loaders // -enum -{ - STBI__SCAN_load=0, - STBI__SCAN_type, - STBI__SCAN_header -}; +enum { STBI__SCAN_load = 0, STBI__SCAN_type, STBI__SCAN_header }; -static void stbi__refill_buffer(stbi__context *s) -{ - int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); - s->callback_already_read += (int) (s->img_buffer - s->img_buffer_original); - if (n == 0) { - // at end of file, treat same as if from memory, but need to handle case - // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file - s->read_from_callbacks = 0; - s->img_buffer = s->buffer_start; - s->img_buffer_end = s->buffer_start+1; - *s->img_buffer = 0; - } else { - s->img_buffer = s->buffer_start; - s->img_buffer_end = s->buffer_start + n; - } +static void stbi__refill_buffer(stbi__context *s) { + int n = (s->io.read)(s->io_user_data, (char *)s->buffer_start, s->buflen); + s->callback_already_read += (int)(s->img_buffer - s->img_buffer_original); + if (n == 0) { + // at end of file, treat same as if from memory, but need to handle case + // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file + s->read_from_callbacks = 0; + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + 1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } } -stbi_inline static stbi_uc stbi__get8(stbi__context *s) -{ - if (s->img_buffer < s->img_buffer_end) - return *s->img_buffer++; - if (s->read_from_callbacks) { - stbi__refill_buffer(s); - return *s->img_buffer++; - } - return 0; +stbi_inline static stbi_uc stbi__get8(stbi__context *s) { + if (s->img_buffer < s->img_buffer_end) return *s->img_buffer++; + if (s->read_from_callbacks) { + stbi__refill_buffer(s); + return *s->img_buffer++; + } + return 0; } -#if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else -stbi_inline static int stbi__at_eof(stbi__context *s) -{ - if (s->io.read) { - if (!(s->io.eof)(s->io_user_data)) return 0; - // if feof() is true, check if buffer = end - // special case: we've only got the special 0 character at the end - if (s->read_from_callbacks == 0) return 1; - } + #if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && \ + defined(STBI_NO_PNM) + // nothing + #else +stbi_inline static int stbi__at_eof(stbi__context *s) { + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } - return s->img_buffer >= s->img_buffer_end; + return s->img_buffer >= s->img_buffer_end; } -#endif + #endif -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) -// nothing -#else -static void stbi__skip(stbi__context *s, int n) -{ - if (n == 0) return; // already there! - if (n < 0) { + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && \ + defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && \ + defined(STBI_NO_PIC) + // nothing + #else +static void stbi__skip(stbi__context *s, int n) { + if (n == 0) return; // already there! + if (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); return; - } - if (s->io.read) { - int blen = (int) (s->img_buffer_end - s->img_buffer); - if (blen < n) { - s->img_buffer = s->img_buffer_end; - (s->io.skip)(s->io_user_data, n - blen); - return; - } - } - s->img_buffer += n; + } + } + s->img_buffer += n; } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && defined(STBI_NO_PNM) -// nothing -#else -static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) -{ - if (s->io.read) { - int blen = (int) (s->img_buffer_end - s->img_buffer); - if (blen < n) { - int res, count; + #if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && \ + defined(STBI_NO_PNM) + // nothing + #else +static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) { + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { + int res, count; - memcpy(buffer, s->img_buffer, blen); + memcpy(buffer, s->img_buffer, blen); - count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen); - res = (count == (n-blen)); - s->img_buffer = s->img_buffer_end; - return res; - } - } + count = (s->io.read)(s->io_user_data, (char *)buffer + blen, n - blen); + res = (count == (n - blen)); + s->img_buffer = s->img_buffer_end; + return res; + } + } - if (s->img_buffer+n <= s->img_buffer_end) { - memcpy(buffer, s->img_buffer, n); - s->img_buffer += n; - return 1; - } else - return 0; + if (s->img_buffer + n <= s->img_buffer_end) { + memcpy(buffer, s->img_buffer, n); + s->img_buffer += n; + return 1; + } else + return 0; } -#endif + #endif -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) -// nothing -#else -static int stbi__get16be(stbi__context *s) -{ - int z = stbi__get8(s); - return (z << 8) + stbi__get8(s); + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && \ + defined(STBI_NO_PIC) + // nothing + #else +static int stbi__get16be(stbi__context *s) { + int z = stbi__get8(s); + return (z << 8) + stbi__get8(s); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) -// nothing -#else -static stbi__uint32 stbi__get32be(stbi__context *s) -{ - stbi__uint32 z = stbi__get16be(s); - return (z << 16) + stbi__get16be(s); + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) + // nothing + #else +static stbi__uint32 stbi__get32be(stbi__context *s) { + stbi__uint32 z = stbi__get16be(s); + return (z << 16) + stbi__get16be(s); } -#endif + #endif -#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) -// nothing -#else -static int stbi__get16le(stbi__context *s) -{ - int z = stbi__get8(s); - return z + (stbi__get8(s) << 8); + #if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) + // nothing + #else +static int stbi__get16le(stbi__context *s) { + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); } -#endif + #endif -#ifndef STBI_NO_BMP -static stbi__uint32 stbi__get32le(stbi__context *s) -{ - stbi__uint32 z = stbi__get16le(s); - return z + (stbi__get16le(s) << 16); + #ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context *s) { + stbi__uint32 z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); } -#endif + #endif -#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + #define STBI__BYTECAST(x) \ + ((stbi_uc)((x)&255)) // truncate int to byte without warnings -#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else + #if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && \ + defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && \ + defined(STBI_NO_PIC) && defined(STBI_NO_PNM) + // nothing + #else ////////////////////////////////////////////////////////////////////////////// // // generic converter from built-in img_n to req_comp @@ -1682,171 +1742,295 @@ static stbi__uint32 stbi__get32le(stbi__context *s) // assume data buffer is malloced, so malloc a new one and free that one // only failure mode is malloc failing -static stbi_uc stbi__compute_y(int r, int g, int b) -{ - return (stbi_uc) (((r*77) + (g*150) + (29*b)) >> 8); +static stbi_uc stbi__compute_y(int r, int g, int b) { + return (stbi_uc)(((r * 77) + (g * 150) + (29 * b)) >> 8); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) -// nothing -#else -static unsigned char *stbi__convert_format(unsigned char *data, int img_n, int req_comp, unsigned int x, unsigned int y) -{ - int i,j; - unsigned char *good; + #if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && \ + defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && \ + defined(STBI_NO_PNM) + // nothing + #else +static unsigned char *stbi__convert_format(unsigned char *data, int img_n, + int req_comp, unsigned int x, + unsigned int y) { + int i, j; + unsigned char *good; - if (req_comp == img_n) return data; - STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); - good = (unsigned char *) stbi__malloc_mad3(req_comp, x, y, 0); - if (good == NULL) { - STBI_FREE(data); - return stbi__errpuc("outofmem", "Out of memory"); - } + good = (unsigned char *)stbi__malloc_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } - for (j=0; j < (int) y; ++j) { - unsigned char *src = data + j * x * img_n ; - unsigned char *dest = good + j * x * req_comp; + for (j = 0; j < (int)y; ++j) { + unsigned char *src = data + j * x * img_n; + unsigned char *dest = good + j * x * req_comp; - #define STBI__COMBO(a,b) ((a)*8+(b)) - #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) - // convert source image with img_n components to one with req_comp components; - // avoid switch per pixel, so use switch per scanline and massive macros - switch (STBI__COMBO(img_n, req_comp)) { - STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=255; } break; - STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=255; } break; - STBI__CASE(2,1) { dest[0]=src[0]; } break; - STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; - STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=255; } break; - STBI__CASE(3,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; - STBI__CASE(3,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = 255; } break; - STBI__CASE(4,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; - STBI__CASE(4,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = src[3]; } break; - STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; - default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return stbi__errpuc("unsupported", "Unsupported format conversion"); + #define STBI__COMBO(a, b) ((a)*8 + (b)) + #define STBI__CASE(a, b) \ + case STBI__COMBO(a, b): \ + for (i = x - 1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp + // components; avoid switch per pixel, so use switch per scanline and + // massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { + dest[0] = src[0]; + dest[1] = 255; } - #undef STBI__CASE - } + break; + STBI__CASE(1, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(1, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = 255; + } + break; + STBI__CASE(2, 1) { + dest[0] = src[0]; + } + break; + STBI__CASE(2, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(2, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = src[1]; + } + break; + STBI__CASE(3, 4) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + dest[3] = 255; + } + break; + STBI__CASE(3, 1) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + } + break; + STBI__CASE(3, 2) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + dest[1] = 255; + } + break; + STBI__CASE(4, 1) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + } + break; + STBI__CASE(4, 2) { + dest[0] = stbi__compute_y(src[0], src[1], src[2]); + dest[1] = src[3]; + } + break; + STBI__CASE(4, 3) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + } + break; + default: + STBI_ASSERT(0); + STBI_FREE(data); + STBI_FREE(good); + return stbi__errpuc("unsupported", "Unsupported format conversion"); + } + #undef STBI__CASE + } - STBI_FREE(data); - return good; + STBI_FREE(data); + return good; } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) -// nothing -#else -static stbi__uint16 stbi__compute_y_16(int r, int g, int b) -{ - return (stbi__uint16) (((r*77) + (g*150) + (29*b)) >> 8); + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) + // nothing + #else +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) { + return (stbi__uint16)(((r * 77) + (g * 150) + (29 * b)) >> 8); } -#endif + #endif -#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) -// nothing -#else -static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, int req_comp, unsigned int x, unsigned int y) -{ - int i,j; - stbi__uint16 *good; + #if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) + // nothing + #else +static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, + int req_comp, unsigned int x, + unsigned int y) { + int i, j; + stbi__uint16 *good; - if (req_comp == img_n) return data; - STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); - good = (stbi__uint16 *) stbi__malloc(req_comp * x * y * 2); - if (good == NULL) { - STBI_FREE(data); - return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); - } + good = (stbi__uint16 *)stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16 *)stbi__errpuc("outofmem", "Out of memory"); + } - for (j=0; j < (int) y; ++j) { - stbi__uint16 *src = data + j * x * img_n ; - stbi__uint16 *dest = good + j * x * req_comp; + for (j = 0; j < (int)y; ++j) { + stbi__uint16 *src = data + j * x * img_n; + stbi__uint16 *dest = good + j * x * req_comp; - #define STBI__COMBO(a,b) ((a)*8+(b)) - #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) - // convert source image with img_n components to one with req_comp components; - // avoid switch per pixel, so use switch per scanline and massive macros - switch (STBI__COMBO(img_n, req_comp)) { - STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=0xffff; } break; - STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=0xffff; } break; - STBI__CASE(2,1) { dest[0]=src[0]; } break; - STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; - STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; - STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=0xffff; } break; - STBI__CASE(3,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; - STBI__CASE(3,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = 0xffff; } break; - STBI__CASE(4,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; - STBI__CASE(4,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = src[3]; } break; - STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; - default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return (stbi__uint16*) stbi__errpuc("unsupported", "Unsupported format conversion"); + #define STBI__COMBO(a, b) ((a)*8 + (b)) + #define STBI__CASE(a, b) \ + case STBI__COMBO(a, b): \ + for (i = x - 1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp + // components; avoid switch per pixel, so use switch per scanline and + // massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { + dest[0] = src[0]; + dest[1] = 0xffff; } - #undef STBI__CASE - } + break; + STBI__CASE(1, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(1, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = 0xffff; + } + break; + STBI__CASE(2, 1) { + dest[0] = src[0]; + } + break; + STBI__CASE(2, 3) { + dest[0] = dest[1] = dest[2] = src[0]; + } + break; + STBI__CASE(2, 4) { + dest[0] = dest[1] = dest[2] = src[0]; + dest[3] = src[1]; + } + break; + STBI__CASE(3, 4) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + dest[3] = 0xffff; + } + break; + STBI__CASE(3, 1) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + } + break; + STBI__CASE(3, 2) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + dest[1] = 0xffff; + } + break; + STBI__CASE(4, 1) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + } + break; + STBI__CASE(4, 2) { + dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); + dest[1] = src[3]; + } + break; + STBI__CASE(4, 3) { + dest[0] = src[0]; + dest[1] = src[1]; + dest[2] = src[2]; + } + break; + default: + STBI_ASSERT(0); + STBI_FREE(data); + STBI_FREE(good); + return (stbi__uint16 *)stbi__errpuc("unsupported", + "Unsupported format conversion"); + } + #undef STBI__CASE + } - STBI_FREE(data); - return good; + STBI_FREE(data); + return good; } -#endif + #endif -#ifndef STBI_NO_LINEAR -static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) -{ - int i,k,n; - float *output; - if (!data) return NULL; - output = (float *) stbi__malloc_mad4(x, y, comp, sizeof(float), 0); - if (output == NULL) { STBI_FREE(data); return stbi__errpf("outofmem", "Out of memory"); } - // compute number of non-alpha components - if (comp & 1) n = comp; else n = comp-1; - for (i=0; i < x*y; ++i) { - for (k=0; k < n; ++k) { - output[i*comp + k] = (float) (pow(data[i*comp+k]/255.0f, stbi__l2h_gamma) * stbi__l2h_scale); - } - } - if (n < comp) { - for (i=0; i < x*y; ++i) { - output[i*comp + n] = data[i*comp + n]/255.0f; - } - } - STBI_FREE(data); - return output; + #ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) { + int i, k, n; + float *output; + if (!data) return NULL; + output = (float *)stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { + STBI_FREE(data); + return stbi__errpf("outofmem", "Out of memory"); + } + // compute number of non-alpha components + if (comp & 1) + n = comp; + else + n = comp - 1; + for (i = 0; i < x * y; ++i) { + for (k = 0; k < n; ++k) { + output[i * comp + k] = + (float)(pow(data[i * comp + k] / 255.0f, stbi__l2h_gamma) * + stbi__l2h_scale); + } + } + if (n < comp) { + for (i = 0; i < x * y; ++i) { + output[i * comp + n] = data[i * comp + n] / 255.0f; + } + } + STBI_FREE(data); + return output; } -#endif + #endif -#ifndef STBI_NO_HDR -#define stbi__float2int(x) ((int) (x)) -static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) -{ - int i,k,n; - stbi_uc *output; - if (!data) return NULL; - output = (stbi_uc *) stbi__malloc_mad3(x, y, comp, 0); - if (output == NULL) { STBI_FREE(data); return stbi__errpuc("outofmem", "Out of memory"); } - // compute number of non-alpha components - if (comp & 1) n = comp; else n = comp-1; - for (i=0; i < x*y; ++i) { - for (k=0; k < n; ++k) { - float z = (float) pow(data[i*comp+k]*stbi__h2l_scale_i, stbi__h2l_gamma_i) * 255 + 0.5f; - if (z < 0) z = 0; - if (z > 255) z = 255; - output[i*comp + k] = (stbi_uc) stbi__float2int(z); - } - if (k < comp) { - float z = data[i*comp+k] * 255 + 0.5f; - if (z < 0) z = 0; - if (z > 255) z = 255; - output[i*comp + k] = (stbi_uc) stbi__float2int(z); - } - } - STBI_FREE(data); - return output; + #ifndef STBI_NO_HDR + #define stbi__float2int(x) ((int)(x)) +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) { + int i, k, n; + stbi_uc *output; + if (!data) return NULL; + output = (stbi_uc *)stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } + // compute number of non-alpha components + if (comp & 1) + n = comp; + else + n = comp - 1; + for (i = 0; i < x * y; ++i) { + for (k = 0; k < n; ++k) { + float z = (float)pow(data[i * comp + k] * stbi__h2l_scale_i, + stbi__h2l_gamma_i) * + 255 + + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i * comp + k] = (stbi_uc)stbi__float2int(z); + } + if (k < comp) { + float z = data[i * comp + k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i * comp + k] = (stbi_uc)stbi__float2int(z); + } + } + STBI_FREE(data); + return output; } -#endif + #endif ////////////////////////////////////////////////////////////////////////////// // @@ -1869,2103 +2053,2259 @@ static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) // - some SIMD kernels for common paths on targets with SSE2/NEON // - uses a lot of intermediate memory, could cache poorly -#ifndef STBI_NO_JPEG + #ifndef STBI_NO_JPEG -// huffman decoding acceleration -#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + // huffman decoding acceleration + #define FAST_BITS 9 // larger handles more cases; smaller stomps less cache -typedef struct -{ - stbi_uc fast[1 << FAST_BITS]; - // weirdly, repacking this into AoS is a 10% speed loss, instead of a win - stbi__uint16 code[256]; - stbi_uc values[256]; - stbi_uc size[257]; - unsigned int maxcode[18]; - int delta[17]; // old 'firstsymbol' - old 'firstcode' +typedef struct { + stbi_uc fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + stbi__uint16 code[256]; + stbi_uc values[256]; + stbi_uc size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' } stbi__huffman; -typedef struct -{ - stbi__context *s; - stbi__huffman huff_dc[4]; - stbi__huffman huff_ac[4]; - stbi__uint16 dequant[4][64]; - stbi__int16 fast_ac[4][1 << FAST_BITS]; +typedef struct { + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; -// sizes for components, interleaved MCUs - int img_h_max, img_v_max; - int img_mcu_x, img_mcu_y; - int img_mcu_w, img_mcu_h; + // sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; -// definition of jpeg image component - struct - { - int id; - int h,v; - int tq; - int hd,ha; - int dc_pred; + // definition of jpeg image component + struct { + int id; + int h, v; + int tq; + int hd, ha; + int dc_pred; - int x,y,w2,h2; - stbi_uc *data; - void *raw_data, *raw_coeff; - stbi_uc *linebuf; - short *coeff; // progressive only - int coeff_w, coeff_h; // number of 8x8 coefficient blocks - } img_comp[4]; + int x, y, w2, h2; + stbi_uc *data; + void *raw_data, *raw_coeff; + stbi_uc *linebuf; + short *coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } img_comp[4]; - stbi__uint32 code_buffer; // jpeg entropy-coded buffer - int code_bits; // number of valid bits - unsigned char marker; // marker seen while filling entropy buffer - int nomore; // flag if we saw a marker so must stop + stbi__uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop - int progressive; - int spec_start; - int spec_end; - int succ_high; - int succ_low; - int eob_run; - int jfif; - int app14_color_transform; // Adobe APP14 tag - int rgb; + int progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; - int scan_n, order[4]; - int restart_interval, todo; + int scan_n, order[4]; + int restart_interval, todo; -// kernels - void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); - void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step); - stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs); + // kernels + void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, + const stbi_uc *pcb, const stbi_uc *pcr, int count, + int step); + stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs); } stbi__jpeg; -static int stbi__build_huffman(stbi__huffman *h, int *count) -{ - int i,j,k=0; - unsigned int code; - // build size list for each symbol (from JPEG spec) - for (i=0; i < 16; ++i) - for (j=0; j < count[i]; ++j) - h->size[k++] = (stbi_uc) (i+1); - h->size[k] = 0; +static int stbi__build_huffman(stbi__huffman *h, int *count) { + int i, j, k = 0; + unsigned int code; + // build size list for each symbol (from JPEG spec) + for (i = 0; i < 16; ++i) + for (j = 0; j < count[i]; ++j) + h->size[k++] = (stbi_uc)(i + 1); + h->size[k] = 0; - // compute actual symbols (from jpeg spec) - code = 0; - k = 0; - for(j=1; j <= 16; ++j) { - // compute delta to add to code to compute symbol id - h->delta[j] = k - code; - if (h->size[k] == j) { - while (h->size[k] == j) - h->code[k++] = (stbi__uint16) (code++); - if (code-1 >= (1u << j)) return stbi__err("bad code lengths","Corrupt JPEG"); - } - // compute largest code + 1 for this size, preshifted as needed later - h->maxcode[j] = code << (16-j); - code <<= 1; - } - h->maxcode[j] = 0xffffffff; + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for (j = 1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (stbi__uint16)(code++); + if (code - 1 >= (1u << j)) + return stbi__err("bad code lengths", "Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16 - j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; - // build non-spec acceleration table; 255 is flag for not-accelerated - memset(h->fast, 255, 1 << FAST_BITS); - for (i=0; i < k; ++i) { - int s = h->size[i]; - if (s <= FAST_BITS) { - int c = h->code[i] << (FAST_BITS-s); - int m = 1 << (FAST_BITS-s); - for (j=0; j < m; ++j) { - h->fast[c+j] = (stbi_uc) i; - } + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i = 0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS - s); + int m = 1 << (FAST_BITS - s); + for (j = 0; j < m; ++j) { + h->fast[c + j] = (stbi_uc)i; } - } - return 1; + } + } + return 1; } // build a table that decodes both magnitude and value of small ACs in // one go. -static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) -{ - int i; - for (i=0; i < (1 << FAST_BITS); ++i) { - stbi_uc fast = h->fast[i]; - fast_ac[i] = 0; - if (fast < 255) { - int rs = h->values[fast]; - int run = (rs >> 4) & 15; - int magbits = rs & 15; - int len = h->size[fast]; +static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) { + int i; + for (i = 0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; - if (magbits && len + magbits <= FAST_BITS) { - // magnitude code followed by receive_extend code - int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); - int m = 1 << (magbits - 1); - if (k < m) k += (~0U << magbits) + 1; - // if the result is small enough, we can fit it in fast_ac table - if (k >= -128 && k <= 127) - fast_ac[i] = (stbi__int16) ((k * 256) + (run * 16) + (len + magbits)); - } + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16)((k * 256) + (run * 16) + (len + magbits)); } - } + } + } } -static void stbi__grow_buffer_unsafe(stbi__jpeg *j) -{ - do { - unsigned int b = j->nomore ? 0 : stbi__get8(j->s); - if (b == 0xff) { - int c = stbi__get8(j->s); - while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes - if (c != 0) { - j->marker = (unsigned char) c; - j->nomore = 1; - return; - } +static void stbi__grow_buffer_unsafe(stbi__jpeg *j) { + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) + c = stbi__get8(j->s); // consume fill bytes + if (c != 0) { + j->marker = (unsigned char)c; + j->nomore = 1; + return; } - j->code_buffer |= b << (24 - j->code_bits); - j->code_bits += 8; - } while (j->code_bits <= 24); + } + j->code_buffer |= b << (24 - j->code_bits); + j->code_bits += 8; + } while (j->code_bits <= 24); } // (1 << n) - 1 -static const stbi__uint32 stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; +static const stbi__uint32 stbi__bmask[17] = { + 0, 1, 3, 7, 15, 31, 63, 127, 255, + 511, 1023, 2047, 4095, 8191, 16383, 32767, 65535}; // decode a jpeg huffman value from the bitstream -stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) -{ - unsigned int temp; - int c,k; +stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) { + unsigned int temp; + int c, k; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - // look at the top FAST_BITS and determine what symbol ID it is, - // if the code is <= FAST_BITS - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - k = h->fast[c]; - if (k < 255) { - int s = h->size[k]; - if (s > j->code_bits) - return -1; - j->code_buffer <<= s; - j->code_bits -= s; - return h->values[k]; - } + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } - // naive test is to shift the code_buffer down so k bits are - // valid, then test against maxcode. To speed this up, we've - // preshifted maxcode left so that it has (16-k) 0s at the - // end; in other words, regardless of the number of bits, it - // wants to be compared against something shifted to have 16; - // that way we don't need to shift inside the loop. - temp = j->code_buffer >> 16; - for (k=FAST_BITS+1 ; ; ++k) - if (temp < h->maxcode[k]) - break; - if (k == 17) { - // error! code not found - j->code_bits -= 16; - return -1; - } + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k = FAST_BITS + 1;; ++k) + if (temp < h->maxcode[k]) break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } - if (k > j->code_bits) - return -1; + if (k > j->code_bits) return -1; - // convert the huffman code to the symbol id - c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; - STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & + stbi__bmask[h->size[c]]) == h->code[c]); - // convert the id to a symbol - j->code_bits -= k; - j->code_buffer <<= k; - return h->values[c]; + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; } // bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); +stbi_inline static int stbi__extend_receive(stbi__jpeg *j, int n) { + unsigned int k; + int sgn; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); - sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB - k = stbi_lrot(j->code_buffer, n); - if (n < 0 || n >= (int) (sizeof(stbi__bmask)/sizeof(*stbi__bmask))) return 0; - j->code_buffer = k & ~stbi__bmask[n]; - k &= stbi__bmask[n]; - j->code_bits -= n; - return k + (stbi__jbias[n] & ~sgn); + sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB + k = stbi_lrot(j->code_buffer, n); + if (n < 0 || n >= (int)(sizeof(stbi__bmask) / sizeof(*stbi__bmask))) return 0; + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & ~sgn); } // get some unsigned bits -stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) -{ - unsigned int k; - if (j->code_bits < n) stbi__grow_buffer_unsafe(j); - k = stbi_lrot(j->code_buffer, n); - j->code_buffer = k & ~stbi__bmask[n]; - k &= stbi__bmask[n]; - j->code_bits -= n; - return k; +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) { + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; } -stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) -{ - unsigned int k; - if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); - k = j->code_buffer; - j->code_buffer <<= 1; - --j->code_bits; - return k & 0x80000000; +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) { + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; } // given a value that's at position X in the zigzag stream, // where does it appear in the 8x8 matrix coded as row-major? -static const stbi_uc stbi__jpeg_dezigzag[64+15] = -{ - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 32, 25, 18, 11, 4, 5, - 12, 19, 26, 33, 40, 48, 41, 34, - 27, 20, 13, 6, 7, 14, 21, 28, - 35, 42, 49, 56, 57, 50, 43, 36, - 29, 22, 15, 23, 30, 37, 44, 51, - 58, 59, 52, 45, 38, 31, 39, 46, - 53, 60, 61, 54, 47, 55, 62, 63, - // let corrupt input sample past end - 63, 63, 63, 63, 63, 63, 63, 63, - 63, 63, 63, 63, 63, 63, 63 -}; +static const stbi_uc stbi__jpeg_dezigzag[64 + 15] = { + 0, 1, 8, 16, 9, 2, 3, 10, 17, 24, 32, 25, 18, 11, 4, 5, 12, 19, 26, 33, 40, + 48, 41, 34, 27, 20, 13, 6, 7, 14, 21, 28, 35, 42, 49, 56, 57, 50, 43, 36, + 29, 22, 15, 23, 30, 37, 44, 51, 58, 59, 52, 45, 38, 31, 39, 46, 53, 60, 61, + 54, 47, 55, 62, 63, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63, 63}; // decode one 64-entry block-- -static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, stbi__int16 *fac, int b, stbi__uint16 *dequant) -{ - int diff,dc,k; - int t; +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], + stbi__huffman *hdc, stbi__huffman *hac, + stbi__int16 *fac, int b, + stbi__uint16 *dequant) { + int diff, dc, k; + int t; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - t = stbi__jpeg_huff_decode(j, hdc); - if (t < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); - // 0 all the ac values now so we can do it 32-bits at a time - memset(data,0,64*sizeof(data[0])); + // 0 all the ac values now so we can do it 32-bits at a time + memset(data, 0, 64 * sizeof(data[0])); - diff = t ? stbi__extend_receive(j, t) : 0; - dc = j->img_comp[b].dc_pred + diff; - j->img_comp[b].dc_pred = dc; - data[0] = (short) (dc * dequant[0]); + diff = t ? stbi__extend_receive(j, t) : 0; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc * dequant[0]); - // decode AC components, see JPEG spec - k = 1; - do { - unsigned int zig; - int c,r,s; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - r = fac[c]; - if (r) { // fast-AC path - k += (r >> 4) & 15; // run - s = r & 15; // combined length - j->code_buffer <<= s; - j->code_bits -= s; - // decode into unzigzag'd location - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) ((r >> 8) * dequant[zig]); + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) * dequant[zig]); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; } else { - int rs = stbi__jpeg_huff_decode(j, hac); - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (rs != 0xf0) break; // end block - k += 16; - } else { - k += r; - // decode into unzigzag'd location - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) (stbi__extend_receive(j,s) * dequant[zig]); - } + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) * dequant[zig]); } - } while (k < 64); - return 1; + } + } while (k < 64); + return 1; } -static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], stbi__huffman *hdc, int b) -{ - int diff,dc; - int t; - if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], + stbi__huffman *hdc, int b) { + int diff, dc; + int t; + if (j->spec_end != 0) + return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - if (j->succ_high == 0) { - // first scan for DC coefficient, must be first - memset(data,0,64*sizeof(data[0])); // 0 all the ac values now - t = stbi__jpeg_huff_decode(j, hdc); - if (t == -1) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - diff = t ? stbi__extend_receive(j, t) : 0; + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data, 0, 64 * sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + if (t == -1) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + diff = t ? stbi__extend_receive(j, t) : 0; - dc = j->img_comp[b].dc_pred + diff; - j->img_comp[b].dc_pred = dc; - data[0] = (short) (dc << j->succ_low); - } else { - // refinement scan for DC coefficient - if (stbi__jpeg_get_bit(j)) - data[0] += (short) (1 << j->succ_low); - } - return 1; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc << j->succ_low); + } else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) data[0] += (short)(1 << j->succ_low); + } + return 1; } // @OPTIMIZE: store non-zigzagged during the decode passes, // and only de-zigzag when dequantizing -static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], stbi__huffman *hac, stbi__int16 *fac) -{ - int k; - if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], + stbi__huffman *hac, + stbi__int16 *fac) { + int k; + if (j->spec_start == 0) + return stbi__err("can't merge dc and ac", "Corrupt JPEG"); - if (j->succ_high == 0) { - int shift = j->succ_low; + if (j->succ_high == 0) { + int shift = j->succ_low; - if (j->eob_run) { - --j->eob_run; - return 1; + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) << shift); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) << shift); + } } + } while (k <= j->spec_end); + } else { + // refinement scan for these AC coefficients + short bit = (short)(1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } + } else { k = j->spec_start; do { - unsigned int zig; - int c,r,s; - if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); - c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); - r = fac[c]; - if (r) { // fast-AC path - k += (r >> 4) & 15; // run - s = r & 15; // combined length - j->code_buffer <<= s; - j->code_bits -= s; - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) ((r >> 8) << shift); - } else { - int rs = stbi__jpeg_huff_decode(j, hac); - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (r < 15) { - j->eob_run = (1 << r); - if (r) - j->eob_run += stbi__jpeg_get_bits(j, r); - --j->eob_run; - break; - } - k += 16; - } else { - k += r; - zig = stbi__jpeg_dezigzag[k++]; - data[zig] = (short) (stbi__extend_receive(j,s) << shift); + int r, s; + int rs = stbi__jpeg_huff_decode( + j, hac); // @OPTIMIZE see if we can use the fast path here, + // advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short *p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } else { + if (r == 0) { + *p = (short)s; + break; } - } + --r; + } + } } while (k <= j->spec_end); - } else { - // refinement scan for these AC coefficients - - short bit = (short) (1 << j->succ_low); - - if (j->eob_run) { - --j->eob_run; - for (k = j->spec_start; k <= j->spec_end; ++k) { - short *p = &data[stbi__jpeg_dezigzag[k]]; - if (*p != 0) - if (stbi__jpeg_get_bit(j)) - if ((*p & bit)==0) { - if (*p > 0) - *p += bit; - else - *p -= bit; - } - } - } else { - k = j->spec_start; - do { - int r,s; - int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh - if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); - s = rs & 15; - r = rs >> 4; - if (s == 0) { - if (r < 15) { - j->eob_run = (1 << r) - 1; - if (r) - j->eob_run += stbi__jpeg_get_bits(j, r); - r = 64; // force end of block - } else { - // r=15 s=0 should write 16 0s, so we just do - // a run of 15 0s and then write s (which is 0), - // so we don't have to do anything special here - } - } else { - if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); - // sign bit - if (stbi__jpeg_get_bit(j)) - s = bit; - else - s = -bit; - } - - // advance by r - while (k <= j->spec_end) { - short *p = &data[stbi__jpeg_dezigzag[k++]]; - if (*p != 0) { - if (stbi__jpeg_get_bit(j)) - if ((*p & bit)==0) { - if (*p > 0) - *p += bit; - else - *p -= bit; - } - } else { - if (r == 0) { - *p = (short) s; - break; - } - --r; - } - } - } while (k <= j->spec_end); - } - } - return 1; + } + } + return 1; } // take a -128..127 value and stbi__clamp it and convert to 0..255 -stbi_inline static stbi_uc stbi__clamp(int x) -{ - // trick to use a single test to catch both cases - if ((unsigned int) x > 255) { - if (x < 0) return 0; - if (x > 255) return 255; - } - return (stbi_uc) x; +stbi_inline static stbi_uc stbi__clamp(int x) { + // trick to use a single test to catch both cases + if ((unsigned int)x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (stbi_uc)x; } -#define stbi__f2f(x) ((int) (((x) * 4096 + 0.5))) -#define stbi__fsh(x) ((x) * 4096) + #define stbi__f2f(x) ((int)(((x)*4096 + 0.5))) + #define stbi__fsh(x) ((x)*4096) -// derived from jidctint -- DCT_ISLOW -#define STBI__IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7) \ - int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \ - p2 = s2; \ - p3 = s6; \ - p1 = (p2+p3) * stbi__f2f(0.5411961f); \ - t2 = p1 + p3*stbi__f2f(-1.847759065f); \ - t3 = p1 + p2*stbi__f2f( 0.765366865f); \ - p2 = s0; \ - p3 = s4; \ - t0 = stbi__fsh(p2+p3); \ - t1 = stbi__fsh(p2-p3); \ - x0 = t0+t3; \ - x3 = t0-t3; \ - x1 = t1+t2; \ - x2 = t1-t2; \ - t0 = s7; \ - t1 = s5; \ - t2 = s3; \ - t3 = s1; \ - p3 = t0+t2; \ - p4 = t1+t3; \ - p1 = t0+t3; \ - p2 = t1+t2; \ - p5 = (p3+p4)*stbi__f2f( 1.175875602f); \ - t0 = t0*stbi__f2f( 0.298631336f); \ - t1 = t1*stbi__f2f( 2.053119869f); \ - t2 = t2*stbi__f2f( 3.072711026f); \ - t3 = t3*stbi__f2f( 1.501321110f); \ - p1 = p5 + p1*stbi__f2f(-0.899976223f); \ - p2 = p5 + p2*stbi__f2f(-2.562915447f); \ - p3 = p3*stbi__f2f(-1.961570560f); \ - p4 = p4*stbi__f2f(-0.390180644f); \ - t3 += p1+p4; \ - t2 += p2+p3; \ - t1 += p2+p4; \ - t0 += p1+p3; + // derived from jidctint -- DCT_ISLOW + #define STBI__IDCT_1D(s0, s1, s2, s3, s4, s5, s6, s7) \ + int t0, t1, t2, t3, p1, p2, p3, p4, p5, x0, x1, x2, x3; \ + p2 = s2; \ + p3 = s6; \ + p1 = (p2 + p3) * stbi__f2f(0.5411961f); \ + t2 = p1 + p3 * stbi__f2f(-1.847759065f); \ + t3 = p1 + p2 * stbi__f2f(0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = stbi__fsh(p2 + p3); \ + t1 = stbi__fsh(p2 - p3); \ + x0 = t0 + t3; \ + x3 = t0 - t3; \ + x1 = t1 + t2; \ + x2 = t1 - t2; \ + t0 = s7; \ + t1 = s5; \ + t2 = s3; \ + t3 = s1; \ + p3 = t0 + t2; \ + p4 = t1 + t3; \ + p1 = t0 + t3; \ + p2 = t1 + t2; \ + p5 = (p3 + p4) * stbi__f2f(1.175875602f); \ + t0 = t0 * stbi__f2f(0.298631336f); \ + t1 = t1 * stbi__f2f(2.053119869f); \ + t2 = t2 * stbi__f2f(3.072711026f); \ + t3 = t3 * stbi__f2f(1.501321110f); \ + p1 = p5 + p1 * stbi__f2f(-0.899976223f); \ + p2 = p5 + p2 * stbi__f2f(-2.562915447f); \ + p3 = p3 * stbi__f2f(-1.961570560f); \ + p4 = p4 * stbi__f2f(-0.390180644f); \ + t3 += p1 + p4; \ + t2 += p2 + p3; \ + t1 += p2 + p4; \ + t0 += p1 + p3; -static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) -{ - int i,val[64],*v=val; - stbi_uc *o; - short *d = data; +static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) { + int i, val[64], *v = val; + stbi_uc *o; + short *d = data; - // columns - for (i=0; i < 8; ++i,++d, ++v) { - // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing - if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 - && d[40]==0 && d[48]==0 && d[56]==0) { - // no shortcut 0 seconds - // (1|2|3|4|5|6|7)==0 0 seconds - // all separate -0.047 seconds - // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds - int dcterm = d[0]*4; - v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; - } else { - STBI__IDCT_1D(d[ 0],d[ 8],d[16],d[24],d[32],d[40],d[48],d[56]) - // constants scaled things up by 1<<12; let's bring them back - // down, but keep 2 extra bits of precision - x0 += 512; x1 += 512; x2 += 512; x3 += 512; - v[ 0] = (x0+t3) >> 10; - v[56] = (x0-t3) >> 10; - v[ 8] = (x1+t2) >> 10; - v[48] = (x1-t2) >> 10; - v[16] = (x2+t1) >> 10; - v[40] = (x2-t1) >> 10; - v[24] = (x3+t0) >> 10; - v[32] = (x3-t0) >> 10; - } - } + // columns + for (i = 0; i < 8; ++i, ++d, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[8] == 0 && d[16] == 0 && d[24] == 0 && d[32] == 0 && d[40] == 0 && + d[48] == 0 && d[56] == 0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0] * 4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[0], d[8], d[16], d[24], d[32], d[40], d[48], d[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; + x1 += 512; + x2 += 512; + x3 += 512; + v[0] = (x0 + t3) >> 10; + v[56] = (x0 - t3) >> 10; + v[8] = (x1 + t2) >> 10; + v[48] = (x1 - t2) >> 10; + v[16] = (x2 + t1) >> 10; + v[40] = (x2 - t1) >> 10; + v[24] = (x3 + t0) >> 10; + v[32] = (x3 - t0) >> 10; + } + } - for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { - // no fast case since the first 1D IDCT spread components out - STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) - // constants scaled things up by 1<<12, plus we had 1<<2 from first - // loop, plus horizontal and vertical each scale by sqrt(8) so together - // we've got an extra 1<<3, so 1<<17 total we need to remove. - // so we want to round that, which means adding 0.5 * 1<<17, - // aka 65536. Also, we'll end up with -128 to 127 that we want - // to encode as 0..255 by adding 128, so we'll add that before the shift - x0 += 65536 + (128<<17); - x1 += 65536 + (128<<17); - x2 += 65536 + (128<<17); - x3 += 65536 + (128<<17); - // tried computing the shifts into temps, or'ing the temps to see - // if any were out of range, but that was slower - o[0] = stbi__clamp((x0+t3) >> 17); - o[7] = stbi__clamp((x0-t3) >> 17); - o[1] = stbi__clamp((x1+t2) >> 17); - o[6] = stbi__clamp((x1-t2) >> 17); - o[2] = stbi__clamp((x2+t1) >> 17); - o[5] = stbi__clamp((x2-t1) >> 17); - o[3] = stbi__clamp((x3+t0) >> 17); - o[4] = stbi__clamp((x3-t0) >> 17); - } + for (i = 0, v = val, o = out; i < 8; ++i, v += 8, o += out_stride) { + // no fast case since the first 1D IDCT spread components out + STBI__IDCT_1D(v[0], v[1], v[2], v[3], v[4], v[5], v[6], v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128 << 17); + x1 += 65536 + (128 << 17); + x2 += 65536 + (128 << 17); + x3 += 65536 + (128 << 17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = stbi__clamp((x0 + t3) >> 17); + o[7] = stbi__clamp((x0 - t3) >> 17); + o[1] = stbi__clamp((x1 + t2) >> 17); + o[6] = stbi__clamp((x1 - t2) >> 17); + o[2] = stbi__clamp((x2 + t1) >> 17); + o[5] = stbi__clamp((x2 - t1) >> 17); + o[3] = stbi__clamp((x3 + t0) >> 17); + o[4] = stbi__clamp((x3 - t0) >> 17); + } } -#ifdef STBI_SSE2 + #ifdef STBI_SSE2 // sse2 integer IDCT. not the fastest possible implementation but it // produces bit-identical results to the generic C version so it's // fully "transparent". -static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) -{ - // This is constructed to match our regular (generic) integer IDCT exactly. - __m128i row0, row1, row2, row3, row4, row5, row6, row7; - __m128i tmp; +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) { + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; - // dot product constant: even elems=x, odd elems=y - #define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + // dot product constant: even elems=x, odd elems=y + #define dct_const(x, y) \ + _mm_setr_epi16((x), (y), (x), (y), (x), (y), (x), (y)) - // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) - // out(1) = c1[even]*x + c1[odd]*y - #define dct_rot(out0,out1, x,y,c0,c1) \ - __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ - __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ - __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ - __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ - __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ - __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) + // out(1) = c1[even]*x + c1[odd]*y + #define dct_rot(out0, out1, x, y, c0, c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x), (y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x), (y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) - // out = in << 12 (in 16-bit, out 32-bit) - #define dct_widen(out, in) \ - __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ - __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + // out = in << 12 (in 16-bit, out 32-bit) + #define dct_widen(out, in) \ + __m128i out##_l = \ + _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = \ + _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) - // wide add - #define dct_wadd(out, a, b) \ - __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ - __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + // wide add + #define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) - // wide sub - #define dct_wsub(out, a, b) \ - __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ - __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + // wide sub + #define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) - // butterfly a/b, add bias, then shift by "s" and pack - #define dct_bfly32o(out0, out1, a,b,bias,s) \ - { \ - __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ - __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ - dct_wadd(sum, abiased, b); \ - dct_wsub(dif, abiased, b); \ - out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ - out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ - } + // butterfly a/b, add bias, then shift by "s" and pack + #define dct_bfly32o(out0, out1, a, b, bias, s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), \ + _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), \ + _mm_srai_epi32(dif_h, s)); \ + } - // 8-bit interleave step (for transposes) - #define dct_interleave8(a, b) \ - tmp = a; \ - a = _mm_unpacklo_epi8(a, b); \ - b = _mm_unpackhi_epi8(tmp, b) + // 8-bit interleave step (for transposes) + #define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) - // 16-bit interleave step (for transposes) - #define dct_interleave16(a, b) \ - tmp = a; \ - a = _mm_unpacklo_epi16(a, b); \ - b = _mm_unpackhi_epi16(tmp, b) + // 16-bit interleave step (for transposes) + #define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) - #define dct_pass(bias,shift) \ - { \ - /* even part */ \ - dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ - __m128i sum04 = _mm_add_epi16(row0, row4); \ - __m128i dif04 = _mm_sub_epi16(row0, row4); \ - dct_widen(t0e, sum04); \ - dct_widen(t1e, dif04); \ - dct_wadd(x0, t0e, t3e); \ - dct_wsub(x3, t0e, t3e); \ - dct_wadd(x1, t1e, t2e); \ - dct_wsub(x2, t1e, t2e); \ - /* odd part */ \ - dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ - dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ - __m128i sum17 = _mm_add_epi16(row1, row7); \ - __m128i sum35 = _mm_add_epi16(row3, row5); \ - dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ - dct_wadd(x4, y0o, y4o); \ - dct_wadd(x5, y1o, y5o); \ - dct_wadd(x6, y2o, y5o); \ - dct_wadd(x7, y3o, y4o); \ - dct_bfly32o(row0,row7, x0,x7,bias,shift); \ - dct_bfly32o(row1,row6, x1,x6,bias,shift); \ - dct_bfly32o(row2,row5, x2,x5,bias,shift); \ - dct_bfly32o(row3,row4, x3,x4,bias,shift); \ - } + #define dct_pass(bias, shift) \ + { \ + /* even part */ \ + dct_rot(t2e, t3e, row2, row6, rot0_0, rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o, y2o, row7, row3, rot2_0, rot2_1); \ + dct_rot(y1o, y3o, row5, row1, rot3_0, rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o, y5o, sum17, sum35, rot1_0, rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0, row7, x0, x7, bias, shift); \ + dct_bfly32o(row1, row6, x1, x6, bias, shift); \ + dct_bfly32o(row2, row5, x2, x5, bias, shift); \ + dct_bfly32o(row3, row4, x3, x4, bias, shift); \ + } - __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); - __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f( 0.765366865f), stbi__f2f(0.5411961f)); - __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); - __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); - __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f( 0.298631336f), stbi__f2f(-1.961570560f)); - __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f( 3.072711026f)); - __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f( 2.053119869f), stbi__f2f(-0.390180644f)); - __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f( 1.501321110f)); + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), + stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f(0.765366865f), + stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), + stbi__f2f(1.175875602f)); + __m128i rot1_1 = + dct_const(stbi__f2f(1.175875602f), + stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f(0.298631336f), + stbi__f2f(-1.961570560f)); + __m128i rot2_1 = + dct_const(stbi__f2f(-1.961570560f), + stbi__f2f(-1.961570560f) + stbi__f2f(3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f(2.053119869f), + stbi__f2f(-0.390180644f)); + __m128i rot3_1 = + dct_const(stbi__f2f(-0.390180644f), + stbi__f2f(-0.390180644f) + stbi__f2f(1.501321110f)); - // rounding biases in column/row passes, see stbi__idct_block for explanation. - __m128i bias_0 = _mm_set1_epi32(512); - __m128i bias_1 = _mm_set1_epi32(65536 + (128<<17)); + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128 << 17)); - // load - row0 = _mm_load_si128((const __m128i *) (data + 0*8)); - row1 = _mm_load_si128((const __m128i *) (data + 1*8)); - row2 = _mm_load_si128((const __m128i *) (data + 2*8)); - row3 = _mm_load_si128((const __m128i *) (data + 3*8)); - row4 = _mm_load_si128((const __m128i *) (data + 4*8)); - row5 = _mm_load_si128((const __m128i *) (data + 5*8)); - row6 = _mm_load_si128((const __m128i *) (data + 6*8)); - row7 = _mm_load_si128((const __m128i *) (data + 7*8)); + // load + row0 = _mm_load_si128((const __m128i *)(data + 0 * 8)); + row1 = _mm_load_si128((const __m128i *)(data + 1 * 8)); + row2 = _mm_load_si128((const __m128i *)(data + 2 * 8)); + row3 = _mm_load_si128((const __m128i *)(data + 3 * 8)); + row4 = _mm_load_si128((const __m128i *)(data + 4 * 8)); + row5 = _mm_load_si128((const __m128i *)(data + 5 * 8)); + row6 = _mm_load_si128((const __m128i *)(data + 6 * 8)); + row7 = _mm_load_si128((const __m128i *)(data + 7 * 8)); - // column pass - dct_pass(bias_0, 10); + // column pass + dct_pass(bias_0, 10); - { - // 16bit 8x8 transpose pass 1 - dct_interleave16(row0, row4); - dct_interleave16(row1, row5); - dct_interleave16(row2, row6); - dct_interleave16(row3, row7); + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); - // transpose pass 2 - dct_interleave16(row0, row2); - dct_interleave16(row1, row3); - dct_interleave16(row4, row6); - dct_interleave16(row5, row7); + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); - // transpose pass 3 - dct_interleave16(row0, row1); - dct_interleave16(row2, row3); - dct_interleave16(row4, row5); - dct_interleave16(row6, row7); - } + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } - // row pass - dct_pass(bias_1, 17); + // row pass + dct_pass(bias_1, 17); - { - // pack - __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 - __m128i p1 = _mm_packus_epi16(row2, row3); - __m128i p2 = _mm_packus_epi16(row4, row5); - __m128i p3 = _mm_packus_epi16(row6, row7); + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); - // 8bit 8x8 transpose pass 1 - dct_interleave8(p0, p2); // a0e0a1e1... - dct_interleave8(p1, p3); // c0g0c1g1... + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... - // transpose pass 2 - dct_interleave8(p0, p1); // a0c0e0g0... - dct_interleave8(p2, p3); // b0d0f0h0... + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... - // transpose pass 3 - dct_interleave8(p0, p2); // a0b0c0d0... - dct_interleave8(p1, p3); // a4b4c4d4... + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... - // store - _mm_storel_epi64((__m128i *) out, p0); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p2); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p1); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; - _mm_storel_epi64((__m128i *) out, p3); out += out_stride; - _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p3, 0x4e)); - } + // store + _mm_storel_epi64((__m128i *)out, p0); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p0, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p2); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p2, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p1); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p1, 0x4e)); + out += out_stride; + _mm_storel_epi64((__m128i *)out, p3); + out += out_stride; + _mm_storel_epi64((__m128i *)out, _mm_shuffle_epi32(p3, 0x4e)); + } -#undef dct_const -#undef dct_rot -#undef dct_widen -#undef dct_wadd -#undef dct_wsub -#undef dct_bfly32o -#undef dct_interleave8 -#undef dct_interleave16 -#undef dct_pass + #undef dct_const + #undef dct_rot + #undef dct_widen + #undef dct_wadd + #undef dct_wsub + #undef dct_bfly32o + #undef dct_interleave8 + #undef dct_interleave16 + #undef dct_pass } -#endif // STBI_SSE2 + #endif // STBI_SSE2 -#ifdef STBI_NEON + #ifdef STBI_NEON // NEON integer IDCT. should produce bit-identical // results to the generic C version. -static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) -{ - int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) { + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; - int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); - int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); - int16x4_t rot0_2 = vdup_n_s16(stbi__f2f( 0.765366865f)); - int16x4_t rot1_0 = vdup_n_s16(stbi__f2f( 1.175875602f)); - int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); - int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); - int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); - int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); - int16x4_t rot3_0 = vdup_n_s16(stbi__f2f( 0.298631336f)); - int16x4_t rot3_1 = vdup_n_s16(stbi__f2f( 2.053119869f)); - int16x4_t rot3_2 = vdup_n_s16(stbi__f2f( 3.072711026f)); - int16x4_t rot3_3 = vdup_n_s16(stbi__f2f( 1.501321110f)); + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f(0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f(1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f(0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f(2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f(3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f(1.501321110f)); -#define dct_long_mul(out, inq, coeff) \ - int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ - int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + #define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) -#define dct_long_mac(out, acc, inq, coeff) \ - int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ - int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + #define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) -#define dct_widen(out, inq) \ - int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ - int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + #define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) -// wide add -#define dct_wadd(out, a, b) \ - int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ - int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + // wide add + #define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) -// wide sub -#define dct_wsub(out, a, b) \ - int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ - int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + // wide sub + #define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) -// butterfly a/b, then shift using "shiftop" by "s" and pack -#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ - { \ - dct_wadd(sum, a, b); \ - dct_wsub(dif, a, b); \ - out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ - out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ - } + // butterfly a/b, then shift using "shiftop" by "s" and pack + #define dct_bfly32o(out0, out1, a, b, shiftop, s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } -#define dct_pass(shiftop, shift) \ - { \ - /* even part */ \ - int16x8_t sum26 = vaddq_s16(row2, row6); \ - dct_long_mul(p1e, sum26, rot0_0); \ - dct_long_mac(t2e, p1e, row6, rot0_1); \ - dct_long_mac(t3e, p1e, row2, rot0_2); \ - int16x8_t sum04 = vaddq_s16(row0, row4); \ - int16x8_t dif04 = vsubq_s16(row0, row4); \ - dct_widen(t0e, sum04); \ - dct_widen(t1e, dif04); \ - dct_wadd(x0, t0e, t3e); \ - dct_wsub(x3, t0e, t3e); \ - dct_wadd(x1, t1e, t2e); \ - dct_wsub(x2, t1e, t2e); \ - /* odd part */ \ - int16x8_t sum15 = vaddq_s16(row1, row5); \ - int16x8_t sum17 = vaddq_s16(row1, row7); \ - int16x8_t sum35 = vaddq_s16(row3, row5); \ - int16x8_t sum37 = vaddq_s16(row3, row7); \ - int16x8_t sumodd = vaddq_s16(sum17, sum35); \ - dct_long_mul(p5o, sumodd, rot1_0); \ - dct_long_mac(p1o, p5o, sum17, rot1_1); \ - dct_long_mac(p2o, p5o, sum35, rot1_2); \ - dct_long_mul(p3o, sum37, rot2_0); \ - dct_long_mul(p4o, sum15, rot2_1); \ - dct_wadd(sump13o, p1o, p3o); \ - dct_wadd(sump24o, p2o, p4o); \ - dct_wadd(sump23o, p2o, p3o); \ - dct_wadd(sump14o, p1o, p4o); \ - dct_long_mac(x4, sump13o, row7, rot3_0); \ - dct_long_mac(x5, sump24o, row5, rot3_1); \ - dct_long_mac(x6, sump23o, row3, rot3_2); \ - dct_long_mac(x7, sump14o, row1, rot3_3); \ - dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ - dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ - dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ - dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ - } + #define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0, row7, x0, x7, shiftop, shift); \ + dct_bfly32o(row1, row6, x1, x6, shiftop, shift); \ + dct_bfly32o(row2, row5, x2, x5, shiftop, shift); \ + dct_bfly32o(row3, row4, x3, x4, shiftop, shift); \ + } - // load - row0 = vld1q_s16(data + 0*8); - row1 = vld1q_s16(data + 1*8); - row2 = vld1q_s16(data + 2*8); - row3 = vld1q_s16(data + 3*8); - row4 = vld1q_s16(data + 4*8); - row5 = vld1q_s16(data + 5*8); - row6 = vld1q_s16(data + 6*8); - row7 = vld1q_s16(data + 7*8); + // load + row0 = vld1q_s16(data + 0 * 8); + row1 = vld1q_s16(data + 1 * 8); + row2 = vld1q_s16(data + 2 * 8); + row3 = vld1q_s16(data + 3 * 8); + row4 = vld1q_s16(data + 4 * 8); + row5 = vld1q_s16(data + 5 * 8); + row6 = vld1q_s16(data + 6 * 8); + row7 = vld1q_s16(data + 7 * 8); - // add DC bias - row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); - // column pass - dct_pass(vrshrn_n_s32, 10); + // column pass + dct_pass(vrshrn_n_s32, 10); - // 16bit 8x8 transpose - { -// these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. -// whether compilers actually get this is another story, sadly. -#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } -#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } -#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + // 16bit 8x8 transpose + { + // these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. + // whether compilers actually get this is another story, sadly. + #define dct_trn16(x, y) \ + { \ + int16x8x2_t t = vtrnq_s16(x, y); \ + x = t.val[0]; \ + y = t.val[1]; \ + } + #define dct_trn32(x, y) \ + { \ + int32x4x2_t t = \ + vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); \ + x = vreinterpretq_s16_s32(t.val[0]); \ + y = vreinterpretq_s16_s32(t.val[1]); \ + } + #define dct_trn64(x, y) \ + { \ + int16x8_t x0 = x; \ + int16x8_t y0 = y; \ + x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); \ + y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); \ + } - // pass 1 - dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 - dct_trn16(row2, row3); - dct_trn16(row4, row5); - dct_trn16(row6, row7); + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); - // pass 2 - dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 - dct_trn32(row1, row3); - dct_trn32(row4, row6); - dct_trn32(row5, row7); + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); - // pass 3 - dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 - dct_trn64(row1, row5); - dct_trn64(row2, row6); - dct_trn64(row3, row7); + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); -#undef dct_trn16 -#undef dct_trn32 -#undef dct_trn64 - } + #undef dct_trn16 + #undef dct_trn32 + #undef dct_trn64 + } - // row pass - // vrshrn_n_s32 only supports shifts up to 16, we need - // 17. so do a non-rounding shift of 16 first then follow - // up with a rounding shift by 1. - dct_pass(vshrn_n_s32, 16); + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); - { - // pack and round - uint8x8_t p0 = vqrshrun_n_s16(row0, 1); - uint8x8_t p1 = vqrshrun_n_s16(row1, 1); - uint8x8_t p2 = vqrshrun_n_s16(row2, 1); - uint8x8_t p3 = vqrshrun_n_s16(row3, 1); - uint8x8_t p4 = vqrshrun_n_s16(row4, 1); - uint8x8_t p5 = vqrshrun_n_s16(row5, 1); - uint8x8_t p6 = vqrshrun_n_s16(row6, 1); - uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); - // again, these can translate into one instruction, but often don't. -#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } -#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } -#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + // again, these can translate into one instruction, but often don't. + #define dct_trn8_8(x, y) \ + { \ + uint8x8x2_t t = vtrn_u8(x, y); \ + x = t.val[0]; \ + y = t.val[1]; \ + } + #define dct_trn8_16(x, y) \ + { \ + uint16x4x2_t t = \ + vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); \ + x = vreinterpret_u8_u16(t.val[0]); \ + y = vreinterpret_u8_u16(t.val[1]); \ + } + #define dct_trn8_32(x, y) \ + { \ + uint32x2x2_t t = \ + vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); \ + x = vreinterpret_u8_u32(t.val[0]); \ + y = vreinterpret_u8_u32(t.val[1]); \ + } - // sadly can't use interleaved stores here since we only write - // 8 bytes to each scan line! + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! - // 8x8 8-bit transpose pass 1 - dct_trn8_8(p0, p1); - dct_trn8_8(p2, p3); - dct_trn8_8(p4, p5); - dct_trn8_8(p6, p7); + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); - // pass 2 - dct_trn8_16(p0, p2); - dct_trn8_16(p1, p3); - dct_trn8_16(p4, p6); - dct_trn8_16(p5, p7); + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); - // pass 3 - dct_trn8_32(p0, p4); - dct_trn8_32(p1, p5); - dct_trn8_32(p2, p6); - dct_trn8_32(p3, p7); + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); - // store - vst1_u8(out, p0); out += out_stride; - vst1_u8(out, p1); out += out_stride; - vst1_u8(out, p2); out += out_stride; - vst1_u8(out, p3); out += out_stride; - vst1_u8(out, p4); out += out_stride; - vst1_u8(out, p5); out += out_stride; - vst1_u8(out, p6); out += out_stride; - vst1_u8(out, p7); + // store + vst1_u8(out, p0); + out += out_stride; + vst1_u8(out, p1); + out += out_stride; + vst1_u8(out, p2); + out += out_stride; + vst1_u8(out, p3); + out += out_stride; + vst1_u8(out, p4); + out += out_stride; + vst1_u8(out, p5); + out += out_stride; + vst1_u8(out, p6); + out += out_stride; + vst1_u8(out, p7); -#undef dct_trn8_8 -#undef dct_trn8_16 -#undef dct_trn8_32 - } + #undef dct_trn8_8 + #undef dct_trn8_16 + #undef dct_trn8_32 + } -#undef dct_long_mul -#undef dct_long_mac -#undef dct_widen -#undef dct_wadd -#undef dct_wsub -#undef dct_bfly32o -#undef dct_pass + #undef dct_long_mul + #undef dct_long_mac + #undef dct_widen + #undef dct_wadd + #undef dct_wsub + #undef dct_bfly32o + #undef dct_pass } -#endif // STBI_NEON + #endif // STBI_NEON -#define STBI__MARKER_none 0xff + #define STBI__MARKER_none 0xff // if there's a pending marker from the entropy stream, return that // otherwise, fetch from the stream and get a marker. if there's no // marker, return 0xff, which is never a valid marker value -static stbi_uc stbi__get_marker(stbi__jpeg *j) -{ - stbi_uc x; - if (j->marker != STBI__MARKER_none) { x = j->marker; j->marker = STBI__MARKER_none; return x; } - x = stbi__get8(j->s); - if (x != 0xff) return STBI__MARKER_none; - while (x == 0xff) - x = stbi__get8(j->s); // consume repeated 0xff fill bytes - return x; +static stbi_uc stbi__get_marker(stbi__jpeg *j) { + stbi_uc x; + if (j->marker != STBI__MARKER_none) { + x = j->marker; + j->marker = STBI__MARKER_none; + return x; + } + x = stbi__get8(j->s); + if (x != 0xff) return STBI__MARKER_none; + while (x == 0xff) + x = stbi__get8(j->s); // consume repeated 0xff fill bytes + return x; } -// in each scan, we'll have scan_n components, and the order -// of the components is specified by order[] -#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + // in each scan, we'll have scan_n components, and the order + // of the components is specified by order[] + #define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) // after a restart interval, stbi__jpeg_reset the entropy decoder and // the dc prediction -static void stbi__jpeg_reset(stbi__jpeg *j) -{ - j->code_bits = 0; - j->code_buffer = 0; - j->nomore = 0; - j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = j->img_comp[3].dc_pred = 0; - j->marker = STBI__MARKER_none; - j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; - j->eob_run = 0; - // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, - // since we don't even allow 1<<30 pixels +static void stbi__jpeg_reset(stbi__jpeg *j) { + j->code_bits = 0; + j->code_buffer = 0; + j->nomore = 0; + j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = + j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels } -static int stbi__parse_entropy_coded_data(stbi__jpeg *z) -{ - stbi__jpeg_reset(z); - if (!z->progressive) { - if (z->scan_n == 1) { - int i,j; - STBI_SIMD_ALIGN(short, data[64]); - int n = z->order[0]; - // non-interleaved data, we just need to process one block at a time, - // in trivial scanline order - // number of blocks to do just depends on how many actual "pixels" this - // component has, independent of interleaved MCU blocking and such - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); - // every data block is an MCU, so countdown the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - // if it's NOT a restart, then just bail, so we get corrupt data - // rather than no data - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } else { // interleaved - int i,j,k,x,y; - STBI_SIMD_ALIGN(short, data[64]); - for (j=0; j < z->img_mcu_y; ++j) { - for (i=0; i < z->img_mcu_x; ++i) { - // scan an interleaved mcu... process scan_n components in order - for (k=0; k < z->scan_n; ++k) { - int n = z->order[k]; - // scan out an mcu's worth of this component; that's just determined - // by the basic H and V specified for the component - for (y=0; y < z->img_comp[n].v; ++y) { - for (x=0; x < z->img_comp[n].h; ++x) { - int x2 = (i*z->img_comp[n].h + x)*8; - int y2 = (j*z->img_comp[n].v + y)*8; - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data); - } - } - } - // after all interleaved components, that's an interleaved MCU, - // so now count down the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) { + stbi__jpeg_reset(z); + if (!z->progressive) { + if (z->scan_n == 1) { + int i, j; + STBI_SIMD_ALIGN(short, data[64]); + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc + z->img_comp[n].hd, + z->huff_ac + ha, z->fast_ac[ha], n, + z->dequant[z->img_comp[n].tq])) + return 0; + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, + z->img_comp[n].w2, data); + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } } - } else { - if (z->scan_n == 1) { - int i,j; - int n = z->order[0]; - // non-interleaved data, we just need to process one block at a time, - // in trivial scanline order - // number of blocks to do just depends on how many actual "pixels" this - // component has, independent of interleaved MCU blocking and such - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); - if (z->spec_start == 0) { - if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) - return 0; - } else { - int ha = z->img_comp[n].ha; - if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) - return 0; - } - // every data block is an MCU, so countdown the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } else { // interleaved - int i,j,k,x,y; - for (j=0; j < z->img_mcu_y; ++j) { - for (i=0; i < z->img_mcu_x; ++i) { - // scan an interleaved mcu... process scan_n components in order - for (k=0; k < z->scan_n; ++k) { - int n = z->order[k]; - // scan out an mcu's worth of this component; that's just determined - // by the basic H and V specified for the component - for (y=0; y < z->img_comp[n].v; ++y) { - for (x=0; x < z->img_comp[n].h; ++x) { - int x2 = (i*z->img_comp[n].h + x); - int y2 = (j*z->img_comp[n].v + y); - short *data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); - if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) - return 0; - } - } - } - // after all interleaved components, that's an interleaved MCU, - // so now count down the restart interval - if (--z->todo <= 0) { - if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); - if (!STBI__RESTART(z->marker)) return 1; - stbi__jpeg_reset(z); - } - } - } - return 1; - } - } -} - -static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) -{ - int i; - for (i=0; i < 64; ++i) - data[i] *= dequant[i]; -} - -static void stbi__jpeg_finish(stbi__jpeg *z) -{ - if (z->progressive) { - // dequantize and idct the data - int i,j,n; - for (n=0; n < z->s->img_n; ++n) { - int w = (z->img_comp[n].x+7) >> 3; - int h = (z->img_comp[n].y+7) >> 3; - for (j=0; j < h; ++j) { - for (i=0; i < w; ++i) { - short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); - stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); - z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); - } - } - } - } -} - -static int stbi__process_marker(stbi__jpeg *z, int m) -{ - int L; - switch (m) { - case STBI__MARKER_none: // no marker found - return stbi__err("expected marker","Corrupt JPEG"); - - case 0xDD: // DRI - specify restart interval - if (stbi__get16be(z->s) != 4) return stbi__err("bad DRI len","Corrupt JPEG"); - z->restart_interval = stbi__get16be(z->s); - return 1; - - case 0xDB: // DQT - define quantization table - L = stbi__get16be(z->s)-2; - while (L > 0) { - int q = stbi__get8(z->s); - int p = q >> 4, sixteen = (p != 0); - int t = q & 15,i; - if (p != 0 && p != 1) return stbi__err("bad DQT type","Corrupt JPEG"); - if (t > 3) return stbi__err("bad DQT table","Corrupt JPEG"); - - for (i=0; i < 64; ++i) - z->dequant[t][stbi__jpeg_dezigzag[i]] = (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); - L -= (sixteen ? 129 : 65); - } - return L==0; - - case 0xC4: // DHT - define huffman table - L = stbi__get16be(z->s)-2; - while (L > 0) { - stbi_uc *v; - int sizes[16],i,n=0; - int q = stbi__get8(z->s); - int tc = q >> 4; - int th = q & 15; - if (tc > 1 || th > 3) return stbi__err("bad DHT header","Corrupt JPEG"); - for (i=0; i < 16; ++i) { - sizes[i] = stbi__get8(z->s); - n += sizes[i]; - } - L -= 17; - if (tc == 0) { - if (!stbi__build_huffman(z->huff_dc+th, sizes)) return 0; - v = z->huff_dc[th].values; - } else { - if (!stbi__build_huffman(z->huff_ac+th, sizes)) return 0; - v = z->huff_ac[th].values; - } - for (i=0; i < n; ++i) - v[i] = stbi__get8(z->s); - if (tc != 0) - stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); - L -= n; - } - return L==0; - } - - // check for comment block or APP blocks - if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { - L = stbi__get16be(z->s); - if (L < 2) { - if (m == 0xFE) - return stbi__err("bad COM len","Corrupt JPEG"); - else - return stbi__err("bad APP len","Corrupt JPEG"); - } - L -= 2; - - if (m == 0xE0 && L >= 5) { // JFIF APP0 segment - static const unsigned char tag[5] = {'J','F','I','F','\0'}; - int ok = 1; - int i; - for (i=0; i < 5; ++i) - if (stbi__get8(z->s) != tag[i]) - ok = 0; - L -= 5; - if (ok) - z->jfif = 1; - } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment - static const unsigned char tag[6] = {'A','d','o','b','e','\0'}; - int ok = 1; - int i; - for (i=0; i < 6; ++i) - if (stbi__get8(z->s) != tag[i]) - ok = 0; - L -= 6; - if (ok) { - stbi__get8(z->s); // version - stbi__get16be(z->s); // flags0 - stbi__get16be(z->s); // flags1 - z->app14_color_transform = stbi__get8(z->s); // color transform - L -= 6; - } - } - - stbi__skip(z->s, L); return 1; - } + } else { // interleaved + int i, j, k, x, y; + STBI_SIMD_ALIGN(short, data[64]); + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k = 0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y = 0; y < z->img_comp[n].v; ++y) { + for (x = 0; x < z->img_comp[n].h; ++x) { + int x2 = (i * z->img_comp[n].h + x) * 8; + int y2 = (j * z->img_comp[n].v + y) * 8; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, + z->huff_dc + z->img_comp[n].hd, + z->huff_ac + ha, z->fast_ac[ha], n, + z->dequant[z->img_comp[n].tq])) + return 0; + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * y2 + x2, + z->img_comp[n].w2, data); + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } else { + if (z->scan_n == 1) { + int i, j; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short *data = + z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc( + z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], + z->fast_ac[ha])) + return 0; + } + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i, j, k, x, y; + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k = 0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y = 0; y < z->img_comp[n].v; ++y) { + for (x = 0; x < z->img_comp[n].h; ++x) { + int x2 = (i * z->img_comp[n].h + x); + int y2 = (j * z->img_comp[n].v + y); + short *data = z->img_comp[n].coeff + + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc( + z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} - return stbi__err("unknown marker","Corrupt JPEG"); +static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) { + int i; + for (i = 0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg *z) { + if (z->progressive) { + // dequantize and idct the data + int i, j, n; + for (n = 0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short *data = + z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel( + z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, + z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg *z, int m) { + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker", "Corrupt JPEG"); + + case 0xDD: // DRI - specify restart interval + if (stbi__get16be(z->s) != 4) + return stbi__err("bad DRI len", "Corrupt JPEG"); + z->restart_interval = stbi__get16be(z->s); + return 1; + + case 0xDB: // DQT - define quantization table + L = stbi__get16be(z->s) - 2; + while (L > 0) { + int q = stbi__get8(z->s); + int p = q >> 4, sixteen = (p != 0); + int t = q & 15, i; + if (p != 0 && p != 1) return stbi__err("bad DQT type", "Corrupt JPEG"); + if (t > 3) return stbi__err("bad DQT table", "Corrupt JPEG"); + + for (i = 0; i < 64; ++i) + z->dequant[t][stbi__jpeg_dezigzag[i]] = + (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 65); + } + return L == 0; + + case 0xC4: // DHT - define huffman table + L = stbi__get16be(z->s) - 2; + while (L > 0) { + stbi_uc *v; + int sizes[16], i, n = 0; + int q = stbi__get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) + return stbi__err("bad DHT header", "Corrupt JPEG"); + for (i = 0; i < 16; ++i) { + sizes[i] = stbi__get8(z->s); + n += sizes[i]; + } + L -= 17; + if (tc == 0) { + if (!stbi__build_huffman(z->huff_dc + th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!stbi__build_huffman(z->huff_ac + th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i = 0; i < n; ++i) + v[i] = stbi__get8(z->s); + if (tc != 0) stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L == 0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len", "Corrupt JPEG"); + else + return stbi__err("bad APP len", "Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = {'J', 'F', 'I', 'F', '\0'}; + int ok = 1; + int i; + for (i = 0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) ok = 0; + L -= 5; + if (ok) z->jfif = 1; + } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = {'A', 'd', 'o', 'b', 'e', '\0'}; + int ok = 1; + int i; + for (i = 0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker", "Corrupt JPEG"); } // after we see SOS -static int stbi__process_scan_header(stbi__jpeg *z) -{ - int i; - int Ls = stbi__get16be(z->s); - z->scan_n = stbi__get8(z->s); - if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return stbi__err("bad SOS component count","Corrupt JPEG"); - if (Ls != 6+2*z->scan_n) return stbi__err("bad SOS len","Corrupt JPEG"); - for (i=0; i < z->scan_n; ++i) { - int id = stbi__get8(z->s), which; - int q = stbi__get8(z->s); - for (which = 0; which < z->s->img_n; ++which) - if (z->img_comp[which].id == id) - break; - if (which == z->s->img_n) return 0; // no match - z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff","Corrupt JPEG"); - z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff","Corrupt JPEG"); - z->order[i] = which; - } +static int stbi__process_scan_header(stbi__jpeg *z) { + int i; + int Ls = stbi__get16be(z->s); + z->scan_n = stbi__get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int)z->s->img_n) + return stbi__err("bad SOS component count", "Corrupt JPEG"); + if (Ls != 6 + 2 * z->scan_n) return stbi__err("bad SOS len", "Corrupt JPEG"); + for (i = 0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; + if (z->img_comp[which].hd > 3) + return stbi__err("bad DC huff", "Corrupt JPEG"); + z->img_comp[which].ha = q & 15; + if (z->img_comp[which].ha > 3) + return stbi__err("bad AC huff", "Corrupt JPEG"); + z->order[i] = which; + } - { - int aa; - z->spec_start = stbi__get8(z->s); - z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 - aa = stbi__get8(z->s); - z->succ_high = (aa >> 4); - z->succ_low = (aa & 15); - if (z->progressive) { - if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) - return stbi__err("bad SOS", "Corrupt JPEG"); - } else { - if (z->spec_start != 0) return stbi__err("bad SOS","Corrupt JPEG"); - if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS","Corrupt JPEG"); - z->spec_end = 63; - } - } + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || + z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } else { + if (z->spec_start != 0) return stbi__err("bad SOS", "Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) + return stbi__err("bad SOS", "Corrupt JPEG"); + z->spec_end = 63; + } + } - return 1; + return 1; } -static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) -{ - int i; - for (i=0; i < ncomp; ++i) { - if (z->img_comp[i].raw_data) { - STBI_FREE(z->img_comp[i].raw_data); - z->img_comp[i].raw_data = NULL; - z->img_comp[i].data = NULL; - } - if (z->img_comp[i].raw_coeff) { - STBI_FREE(z->img_comp[i].raw_coeff); - z->img_comp[i].raw_coeff = 0; - z->img_comp[i].coeff = 0; - } - if (z->img_comp[i].linebuf) { - STBI_FREE(z->img_comp[i].linebuf); - z->img_comp[i].linebuf = NULL; - } - } - return why; -} - -static int stbi__process_frame_header(stbi__jpeg *z, int scan) -{ - stbi__context *s = z->s; - int Lf,p,i,q, h_max=1,v_max=1,c; - Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len","Corrupt JPEG"); // JPEG - p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline - s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG - s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); // JPEG requires - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - c = stbi__get8(s); - if (c != 3 && c != 1 && c != 4) return stbi__err("bad component count","Corrupt JPEG"); - s->img_n = c; - for (i=0; i < c; ++i) { +static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) { + int i; + for (i = 0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; z->img_comp[i].data = NULL; - z->img_comp[i].linebuf = NULL; - } - - if (Lf != 8+3*s->img_n) return stbi__err("bad SOF len","Corrupt JPEG"); - - z->rgb = 0; - for (i=0; i < s->img_n; ++i) { - static const unsigned char rgb[3] = { 'R', 'G', 'B' }; - z->img_comp[i].id = stbi__get8(s); - if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) - ++z->rgb; - q = stbi__get8(s); - z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H","Corrupt JPEG"); - z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V","Corrupt JPEG"); - z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ","Corrupt JPEG"); - } - - if (scan != STBI__SCAN_load) return 1; - - if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) return stbi__err("too large", "Image too large to decode"); - - for (i=0; i < s->img_n; ++i) { - if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; - if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; - } - - // compute interleaved mcu info - z->img_h_max = h_max; - z->img_v_max = v_max; - z->img_mcu_w = h_max * 8; - z->img_mcu_h = v_max * 8; - // these sizes can't be more than 17 bits - z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; - z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; - - for (i=0; i < s->img_n; ++i) { - // number of effective pixels (e.g. for non-interleaved MCU) - z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; - z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; - // to simplify generation, we'll allocate enough memory to decode - // the bogus oversized data from using interleaved MCUs and their - // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't - // discard the extra data until colorspace conversion - // - // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) - // so these muls can't overflow with 32-bit ints (which we require) - z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; - z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; - z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); z->img_comp[i].linebuf = NULL; - z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); - if (z->img_comp[i].raw_data == NULL) - return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); - // align blocks for idct using mmx/sse - z->img_comp[i].data = (stbi_uc*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); - if (z->progressive) { - // w2, h2 are multiples of 8 (see above) - z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; - z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; - z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); - if (z->img_comp[i].raw_coeff == NULL) - return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); - z->img_comp[i].coeff = (short*) (((size_t) z->img_comp[i].raw_coeff + 15) & ~15); - } - } - - return 1; + } + } + return why; } -// use comparisons since in some cases we handle more than one case (e.g. SOF) -#define stbi__DNL(x) ((x) == 0xdc) -#define stbi__SOI(x) ((x) == 0xd8) -#define stbi__EOI(x) ((x) == 0xd9) -#define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) -#define stbi__SOS(x) ((x) == 0xda) +static int stbi__process_frame_header(stbi__jpeg *z, int scan) { + stbi__context *s = z->s; + int Lf, p, i, q, h_max = 1, v_max = 1, c; + Lf = stbi__get16be(s); + if (Lf < 11) return stbi__err("bad SOF len", "Corrupt JPEG"); // JPEG + p = stbi__get8(s); + if (p != 8) + return stbi__err("only 8-bit", + "JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = stbi__get16be(s); + if (s->img_y == 0) + return stbi__err( + "no header height", + "JPEG format not supported: delayed height"); // Legal, but we don't + // handle it--but neither + // does IJG + s->img_x = stbi__get16be(s); + if (s->img_x == 0) + return stbi__err("0 width", "Corrupt JPEG"); // JPEG requires + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + c = stbi__get8(s); + if (c != 3 && c != 1 && c != 4) + return stbi__err("bad component count", "Corrupt JPEG"); + s->img_n = c; + for (i = 0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } -#define stbi__SOF_progressive(x) ((x) == 0xc2) + if (Lf != 8 + 3 * s->img_n) return stbi__err("bad SOF len", "Corrupt JPEG"); -static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) -{ - int m; - z->jfif = 0; - z->app14_color_transform = -1; // valid values are 0,1,2 - z->marker = STBI__MARKER_none; // initialize cached marker to empty - m = stbi__get_marker(z); - if (!stbi__SOI(m)) return stbi__err("no SOI","Corrupt JPEG"); - if (scan == STBI__SCAN_type) return 1; - m = stbi__get_marker(z); - while (!stbi__SOF(m)) { - if (!stbi__process_marker(z,m)) return 0; + z->rgb = 0; + for (i = 0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = {'R', 'G', 'B'}; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); + if (!z->img_comp[i].h || z->img_comp[i].h > 4) + return stbi__err("bad H", "Corrupt JPEG"); + z->img_comp[i].v = q & 15; + if (!z->img_comp[i].v || z->img_comp[i].v > 4) + return stbi__err("bad V", "Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); + if (z->img_comp[i].tq > 3) return stbi__err("bad TQ", "Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) + return stbi__err("too large", "Image too large to decode"); + + for (i = 0; i < s->img_n; ++i) { + if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; + if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + // these sizes can't be more than 17 bits + z->img_mcu_x = (s->img_x + z->img_mcu_w - 1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h - 1) / z->img_mcu_h; + + for (i = 0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max - 1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max - 1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked + // earlier) so these muls can't overflow with 32-bit ints (which we require) + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = + stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i + 1, + stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = + (stbi_uc *)(((size_t)z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3( + z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components( + z, i + 1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = + (short *)(((size_t)z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + + // use comparisons since in some cases we handle more than one case (e.g. + // SOF) + #define stbi__DNL(x) ((x) == 0xdc) + #define stbi__SOI(x) ((x) == 0xd8) + #define stbi__EOI(x) ((x) == 0xd9) + #define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) + #define stbi__SOS(x) ((x) == 0xda) + + #define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) { + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI", "Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z, m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); m = stbi__get_marker(z); - while (m == STBI__MARKER_none) { - // some files have extra padding after their blocks, so ok, we'll scan - if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); - m = stbi__get_marker(z); - } - } - z->progressive = stbi__SOF_progressive(m); - if (!stbi__process_frame_header(z, scan)) return 0; - return 1; + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; } // decode image to YCbCr format -static int stbi__decode_jpeg_image(stbi__jpeg *j) -{ - int m; - for (m = 0; m < 4; m++) { - j->img_comp[m].raw_data = NULL; - j->img_comp[m].raw_coeff = NULL; - } - j->restart_interval = 0; - if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; - m = stbi__get_marker(j); - while (!stbi__EOI(m)) { - if (stbi__SOS(m)) { - if (!stbi__process_scan_header(j)) return 0; - if (!stbi__parse_entropy_coded_data(j)) return 0; - if (j->marker == STBI__MARKER_none ) { - // handle 0s at the end of image data from IP Kamera 9060 - while (!stbi__at_eof(j->s)) { - int x = stbi__get8(j->s); - if (x == 255) { - j->marker = stbi__get8(j->s); - break; - } - } - // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 - } - } else if (stbi__DNL(m)) { - int Ld = stbi__get16be(j->s); - stbi__uint32 NL = stbi__get16be(j->s); - if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); - if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); - } else { - if (!stbi__process_marker(j, m)) return 0; +static int stbi__decode_jpeg_image(stbi__jpeg *j) { + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none) { + // handle 0s at the end of image data from IP Kamera 9060 + while (!stbi__at_eof(j->s)) { + int x = stbi__get8(j->s); + if (x == 255) { + j->marker = stbi__get8(j->s); + break; + } + } + // if we reach eof without hitting a marker, stbi__get_marker() below + // will fail and we'll eventually return 0 } - m = stbi__get_marker(j); - } - if (j->progressive) - stbi__jpeg_finish(j); - return 1; + } else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + } else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + if (j->progressive) stbi__jpeg_finish(j); + return 1; } // static jfif-centered resampling (across block boundaries) typedef stbi_uc *(*resample_row_func)(stbi_uc *out, stbi_uc *in0, stbi_uc *in1, - int w, int hs); + int w, int hs); -#define stbi__div4(x) ((stbi_uc) ((x) >> 2)) + #define stbi__div4(x) ((stbi_uc)((x) >> 2)) -static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - STBI_NOTUSED(out); - STBI_NOTUSED(in_far); - STBI_NOTUSED(w); - STBI_NOTUSED(hs); - return in_near; +static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, + int w, int hs) { + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; } -static stbi_uc* stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate two samples vertically for every one in input - int i; - STBI_NOTUSED(hs); - for (i=0; i < w; ++i) - out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); - return out; +static stbi_uc *stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i = 0; i < w; ++i) + out[i] = stbi__div4(3 * in_near[i] + in_far[i] + 2); + return out; } -static stbi_uc* stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate two samples horizontally for every one in input - int i; - stbi_uc *input = in_near; +static stbi_uc *stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate two samples horizontally for every one in input + int i; + stbi_uc *input = in_near; - if (w == 1) { - // if only one sample, can't do any interpolation - out[0] = out[1] = input[0]; - return out; - } + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } - out[0] = input[0]; - out[1] = stbi__div4(input[0]*3 + input[1] + 2); - for (i=1; i < w-1; ++i) { - int n = 3*input[i]+2; - out[i*2+0] = stbi__div4(n+input[i-1]); - out[i*2+1] = stbi__div4(n+input[i+1]); - } - out[i*2+0] = stbi__div4(input[w-2]*3 + input[w-1] + 2); - out[i*2+1] = input[w-1]; + out[0] = input[0]; + out[1] = stbi__div4(input[0] * 3 + input[1] + 2); + for (i = 1; i < w - 1; ++i) { + int n = 3 * input[i] + 2; + out[i * 2 + 0] = stbi__div4(n + input[i - 1]); + out[i * 2 + 1] = stbi__div4(n + input[i + 1]); + } + out[i * 2 + 0] = stbi__div4(input[w - 2] * 3 + input[w - 1] + 2); + out[i * 2 + 1] = input[w - 1]; - STBI_NOTUSED(in_far); - STBI_NOTUSED(hs); + STBI_NOTUSED(in_far); + STBI_NOTUSED(hs); - return out; + return out; } -#define stbi__div16(x) ((stbi_uc) ((x) >> 4)) + #define stbi__div16(x) ((stbi_uc)((x) >> 4)) -static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate 2x2 samples for every one in input - int i,t0,t1; - if (w == 1) { - out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); - return out; - } +static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate 2x2 samples for every one in input + int i, t0, t1; + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } - t1 = 3*in_near[0] + in_far[0]; - out[0] = stbi__div4(t1+2); - for (i=1; i < w; ++i) { - t0 = t1; - t1 = 3*in_near[i]+in_far[i]; - out[i*2-1] = stbi__div16(3*t0 + t1 + 8); - out[i*2 ] = stbi__div16(3*t1 + t0 + 8); - } - out[w*2-1] = stbi__div4(t1+2); + t1 = 3 * in_near[0] + in_far[0]; + out[0] = stbi__div4(t1 + 2); + for (i = 1; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); - STBI_NOTUSED(hs); + STBI_NOTUSED(hs); - return out; + return out; } -#if defined(STBI_SSE2) || defined(STBI_NEON) -static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // need to generate 2x2 samples for every one in input - int i=0,t0,t1; + #if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // need to generate 2x2 samples for every one in input + int i = 0, t0, t1; - if (w == 1) { - out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); - return out; - } + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } - t1 = 3*in_near[0] + in_far[0]; - // process groups of 8 pixels for as long as we can. - // note we can't handle the last pixel in a row in this loop - // because we need to handle the filter boundary conditions. - for (; i < ((w-1) & ~7); i += 8) { -#if defined(STBI_SSE2) - // load and perform the vertical filtering pass - // this uses 3*x + y = 4*x + (y - x) - __m128i zero = _mm_setzero_si128(); - __m128i farb = _mm_loadl_epi64((__m128i *) (in_far + i)); - __m128i nearb = _mm_loadl_epi64((__m128i *) (in_near + i)); - __m128i farw = _mm_unpacklo_epi8(farb, zero); - __m128i nearw = _mm_unpacklo_epi8(nearb, zero); - __m128i diff = _mm_sub_epi16(farw, nearw); - __m128i nears = _mm_slli_epi16(nearw, 2); - __m128i curr = _mm_add_epi16(nears, diff); // current row + t1 = 3 * in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w - 1) & ~7); i += 8) { + #if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i *)(in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i *)(in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row - // horizontal filter works the same based on shifted vers of current - // row. "prev" is current row shifted right by 1 pixel; we need to - // insert the previous pixel value (from t1). - // "next" is current row shifted left by 1 pixel, with first pixel - // of next block of 8 pixels added in. - __m128i prv0 = _mm_slli_si128(curr, 2); - __m128i nxt0 = _mm_srli_si128(curr, 2); - __m128i prev = _mm_insert_epi16(prv0, t1, 0); - __m128i next = _mm_insert_epi16(nxt0, 3*in_near[i+8] + in_far[i+8], 7); + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = + _mm_insert_epi16(nxt0, 3 * in_near[i + 8] + in_far[i + 8], 7); - // horizontal filter, polyphase implementation since it's convenient: - // even pixels = 3*cur + prev = cur*4 + (prev - cur) - // odd pixels = 3*cur + next = cur*4 + (next - cur) - // note the shared term. - __m128i bias = _mm_set1_epi16(8); - __m128i curs = _mm_slli_epi16(curr, 2); - __m128i prvd = _mm_sub_epi16(prev, curr); - __m128i nxtd = _mm_sub_epi16(next, curr); - __m128i curb = _mm_add_epi16(curs, bias); - __m128i even = _mm_add_epi16(prvd, curb); - __m128i odd = _mm_add_epi16(nxtd, curb); + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); - // interleave even and odd pixels, then undo scaling. - __m128i int0 = _mm_unpacklo_epi16(even, odd); - __m128i int1 = _mm_unpackhi_epi16(even, odd); - __m128i de0 = _mm_srli_epi16(int0, 4); - __m128i de1 = _mm_srli_epi16(int1, 4); + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); - // pack and write output - __m128i outv = _mm_packus_epi16(de0, de1); - _mm_storeu_si128((__m128i *) (out + i*2), outv); -#elif defined(STBI_NEON) - // load and perform the vertical filtering pass - // this uses 3*x + y = 4*x + (y - x) - uint8x8_t farb = vld1_u8(in_far + i); - uint8x8_t nearb = vld1_u8(in_near + i); - int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); - int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); - int16x8_t curr = vaddq_s16(nears, diff); // current row + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i *)(out + i * 2), outv); + #elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row - // horizontal filter works the same based on shifted vers of current - // row. "prev" is current row shifted right by 1 pixel; we need to - // insert the previous pixel value (from t1). - // "next" is current row shifted left by 1 pixel, with first pixel - // of next block of 8 pixels added in. - int16x8_t prv0 = vextq_s16(curr, curr, 7); - int16x8_t nxt0 = vextq_s16(curr, curr, 1); - int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); - int16x8_t next = vsetq_lane_s16(3*in_near[i+8] + in_far[i+8], nxt0, 7); + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = + vsetq_lane_s16(3 * in_near[i + 8] + in_far[i + 8], nxt0, 7); - // horizontal filter, polyphase implementation since it's convenient: - // even pixels = 3*cur + prev = cur*4 + (prev - cur) - // odd pixels = 3*cur + next = cur*4 + (next - cur) - // note the shared term. - int16x8_t curs = vshlq_n_s16(curr, 2); - int16x8_t prvd = vsubq_s16(prev, curr); - int16x8_t nxtd = vsubq_s16(next, curr); - int16x8_t even = vaddq_s16(curs, prvd); - int16x8_t odd = vaddq_s16(curs, nxtd); + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); - // undo scaling and round, then store with even/odd phases interleaved - uint8x8x2_t o; - o.val[0] = vqrshrun_n_s16(even, 4); - o.val[1] = vqrshrun_n_s16(odd, 4); - vst2_u8(out + i*2, o); -#endif + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i * 2, o); + #endif - // "previous" value for next iter - t1 = 3*in_near[i+7] + in_far[i+7]; - } + // "previous" value for next iter + t1 = 3 * in_near[i + 7] + in_far[i + 7]; + } - t0 = t1; - t1 = 3*in_near[i] + in_far[i]; - out[i*2] = stbi__div16(3*t1 + t0 + 8); + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); - for (++i; i < w; ++i) { - t0 = t1; - t1 = 3*in_near[i]+in_far[i]; - out[i*2-1] = stbi__div16(3*t0 + t1 + 8); - out[i*2 ] = stbi__div16(3*t1 + t0 + 8); - } - out[w*2-1] = stbi__div4(t1+2); + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); - STBI_NOTUSED(hs); + STBI_NOTUSED(hs); - return out; + return out; } -#endif + #endif -static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) -{ - // resample with nearest-neighbor - int i,j; - STBI_NOTUSED(in_far); - for (i=0; i < w; ++i) - for (j=0; j < hs; ++j) - out[i*hs+j] = in_near[i]; - return out; +static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, + stbi_uc *in_far, int w, int hs) { + // resample with nearest-neighbor + int i, j; + STBI_NOTUSED(in_far); + for (i = 0; i < w; ++i) + for (j = 0; j < hs; ++j) + out[i * hs + j] = in_near[i]; + return out; } -// this is a reduced-precision calculation of YCbCr-to-RGB introduced -// to make sure the code produces the same results in both SIMD and scalar -#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) -static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step) -{ - int i; - for (i=0; i < count; ++i) { - int y_fixed = (y[i] << 20) + (1<<19); // rounding - int r,g,b; - int cr = pcr[i] - 128; - int cb = pcb[i] - 128; - r = y_fixed + cr* stbi__float2fixed(1.40200f); - g = y_fixed + (cr*-stbi__float2fixed(0.71414f)) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); - b = y_fixed + cb* stbi__float2fixed(1.77200f); - r >>= 20; - g >>= 20; - b >>= 20; - if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } - if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } - if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } - out[0] = (stbi_uc)r; - out[1] = (stbi_uc)g; - out[2] = (stbi_uc)b; - out[3] = 255; - out += step; - } + // this is a reduced-precision calculation of YCbCr-to-RGB introduced + // to make sure the code produces the same results in both SIMD and scalar + #define stbi__float2fixed(x) (((int)((x)*4096.0f + 0.5f)) << 8) +static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, + const stbi_uc *pcb, const stbi_uc *pcr, + int count, int step) { + int i; + for (i = 0; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + (cr * -stbi__float2fixed(0.71414f)) + + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { + if (r < 0) + r = 0; + else + r = 255; + } + if ((unsigned)g > 255) { + if (g < 0) + g = 0; + else + g = 255; + } + if ((unsigned)b > 255) { + if (b < 0) + b = 0; + else + b = 255; + } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } } -#if defined(STBI_SSE2) || defined(STBI_NEON) -static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, stbi_uc const *pcb, stbi_uc const *pcr, int count, int step) -{ - int i = 0; + #if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, + stbi_uc const *pcb, stbi_uc const *pcr, + int count, int step) { + int i = 0; -#ifdef STBI_SSE2 - // step == 3 is pretty ugly on the final interleave, and i'm not convinced - // it's useful in practice (you wouldn't use it for textures, for example). - // so just accelerate step == 4 case. - if (step == 4) { - // this is a fairly straightforward implementation and not super-optimized. - __m128i signflip = _mm_set1_epi8(-0x80); - __m128i cr_const0 = _mm_set1_epi16( (short) ( 1.40200f*4096.0f+0.5f)); - __m128i cr_const1 = _mm_set1_epi16( - (short) ( 0.71414f*4096.0f+0.5f)); - __m128i cb_const0 = _mm_set1_epi16( - (short) ( 0.34414f*4096.0f+0.5f)); - __m128i cb_const1 = _mm_set1_epi16( (short) ( 1.77200f*4096.0f+0.5f)); - __m128i y_bias = _mm_set1_epi8((char) (unsigned char) 128); - __m128i xw = _mm_set1_epi16(255); // alpha channel + #ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16((short)(1.40200f * 4096.0f + 0.5f)); + __m128i cr_const1 = _mm_set1_epi16(-(short)(0.71414f * 4096.0f + 0.5f)); + __m128i cb_const0 = _mm_set1_epi16(-(short)(0.34414f * 4096.0f + 0.5f)); + __m128i cb_const1 = _mm_set1_epi16((short)(1.77200f * 4096.0f + 0.5f)); + __m128i y_bias = _mm_set1_epi8((char)(unsigned char)128); + __m128i xw = _mm_set1_epi16(255); // alpha channel - for (; i+7 < count; i += 8) { - // load - __m128i y_bytes = _mm_loadl_epi64((__m128i *) (y+i)); - __m128i cr_bytes = _mm_loadl_epi64((__m128i *) (pcr+i)); - __m128i cb_bytes = _mm_loadl_epi64((__m128i *) (pcb+i)); - __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 - __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + for (; i + 7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i *)(y + i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i *)(pcr + i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i *)(pcb + i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 - // unpack to short (and left-shift cr, cb by 8) - __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); - __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); - __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); - // color transform - __m128i yws = _mm_srli_epi16(yw, 4); - __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); - __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); - __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); - __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); - __m128i rws = _mm_add_epi16(cr0, yws); - __m128i gwt = _mm_add_epi16(cb0, yws); - __m128i bws = _mm_add_epi16(yws, cb1); - __m128i gws = _mm_add_epi16(gwt, cr1); + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); - // descale - __m128i rw = _mm_srai_epi16(rws, 4); - __m128i bw = _mm_srai_epi16(bws, 4); - __m128i gw = _mm_srai_epi16(gws, 4); + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); - // back to byte, set up for transpose - __m128i brb = _mm_packus_epi16(rw, bw); - __m128i gxb = _mm_packus_epi16(gw, xw); + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); - // transpose to interleave channels - __m128i t0 = _mm_unpacklo_epi8(brb, gxb); - __m128i t1 = _mm_unpackhi_epi8(brb, gxb); - __m128i o0 = _mm_unpacklo_epi16(t0, t1); - __m128i o1 = _mm_unpackhi_epi16(t0, t1); + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); - // store - _mm_storeu_si128((__m128i *) (out + 0), o0); - _mm_storeu_si128((__m128i *) (out + 16), o1); - out += 32; - } - } -#endif + // store + _mm_storeu_si128((__m128i *)(out + 0), o0); + _mm_storeu_si128((__m128i *)(out + 16), o1); + out += 32; + } + } + #endif -#ifdef STBI_NEON - // in this version, step=3 support would be easy to add. but is there demand? - if (step == 4) { - // this is a fairly straightforward implementation and not super-optimized. - uint8x8_t signflip = vdup_n_u8(0x80); - int16x8_t cr_const0 = vdupq_n_s16( (short) ( 1.40200f*4096.0f+0.5f)); - int16x8_t cr_const1 = vdupq_n_s16( - (short) ( 0.71414f*4096.0f+0.5f)); - int16x8_t cb_const0 = vdupq_n_s16( - (short) ( 0.34414f*4096.0f+0.5f)); - int16x8_t cb_const1 = vdupq_n_s16( (short) ( 1.77200f*4096.0f+0.5f)); + #ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16((short)(1.40200f * 4096.0f + 0.5f)); + int16x8_t cr_const1 = vdupq_n_s16(-(short)(0.71414f * 4096.0f + 0.5f)); + int16x8_t cb_const0 = vdupq_n_s16(-(short)(0.34414f * 4096.0f + 0.5f)); + int16x8_t cb_const1 = vdupq_n_s16((short)(1.77200f * 4096.0f + 0.5f)); - for (; i+7 < count; i += 8) { - // load - uint8x8_t y_bytes = vld1_u8(y + i); - uint8x8_t cr_bytes = vld1_u8(pcr + i); - uint8x8_t cb_bytes = vld1_u8(pcb + i); - int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); - int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + for (; i + 7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); - // expand to s16 - int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); - int16x8_t crw = vshll_n_s8(cr_biased, 7); - int16x8_t cbw = vshll_n_s8(cb_biased, 7); + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); - // color transform - int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); - int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); - int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); - int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); - int16x8_t rws = vaddq_s16(yws, cr0); - int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); - int16x8_t bws = vaddq_s16(yws, cb1); + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); - // undo scaling, round, convert to byte - uint8x8x4_t o; - o.val[0] = vqrshrun_n_s16(rws, 4); - o.val[1] = vqrshrun_n_s16(gws, 4); - o.val[2] = vqrshrun_n_s16(bws, 4); - o.val[3] = vdup_n_u8(255); + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); - // store, interleaving r/g/b/a - vst4_u8(out, o); - out += 8*4; - } - } -#endif + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8 * 4; + } + } + #endif - for (; i < count; ++i) { - int y_fixed = (y[i] << 20) + (1<<19); // rounding - int r,g,b; - int cr = pcr[i] - 128; - int cb = pcb[i] - 128; - r = y_fixed + cr* stbi__float2fixed(1.40200f); - g = y_fixed + cr*-stbi__float2fixed(0.71414f) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); - b = y_fixed + cb* stbi__float2fixed(1.77200f); - r >>= 20; - g >>= 20; - b >>= 20; - if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } - if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } - if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } - out[0] = (stbi_uc)r; - out[1] = (stbi_uc)g; - out[2] = (stbi_uc)b; - out[3] = 255; - out += step; - } + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + cr * -stbi__float2fixed(0.71414f) + + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { + if (r < 0) + r = 0; + else + r = 255; + } + if ((unsigned)g > 255) { + if (g < 0) + g = 0; + else + g = 255; + } + if ((unsigned)b > 255) { + if (b < 0) + b = 0; + else + b = 255; + } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } } -#endif + #endif // set up the kernels -static void stbi__setup_jpeg(stbi__jpeg *j) -{ - j->idct_block_kernel = stbi__idct_block; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; +static void stbi__setup_jpeg(stbi__jpeg *j) { + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; -#ifdef STBI_SSE2 - if (stbi__sse2_available()) { - j->idct_block_kernel = stbi__idct_simd; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; - } -#endif + #ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } + #endif -#ifdef STBI_NEON - j->idct_block_kernel = stbi__idct_simd; - j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; - j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; -#endif + #ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + #endif } // clean up the temporary component buffers -static void stbi__cleanup_jpeg(stbi__jpeg *j) -{ - stbi__free_jpeg_components(j, j->s->img_n, 0); +static void stbi__cleanup_jpeg(stbi__jpeg *j) { + stbi__free_jpeg_components(j, j->s->img_n, 0); } -typedef struct -{ - resample_row_func resample; - stbi_uc *line0,*line1; - int hs,vs; // expansion factor in each axis - int w_lores; // horizontal pixels pre-expansion - int ystep; // how far through vertical expansion we are - int ypos; // which pre-expansion row we're on +typedef struct { + resample_row_func resample; + stbi_uc *line0, *line1; + int hs, vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on } stbi__resample; // fast 0..255 * 0..255 => 0..255 rounded multiplication -static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) -{ - unsigned int t = x*y + 128; - return (stbi_uc) ((t + (t >>8)) >> 8); +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) { + unsigned int t = x * y + 128; + return (stbi_uc)((t + (t >> 8)) >> 8); } -static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) -{ - int n, decode_n, is_rgb; - z->s->img_n = 0; // make stbi__cleanup_jpeg safe +static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, + int *comp, int req_comp) { + int n, decode_n, is_rgb; + z->s->img_n = 0; // make stbi__cleanup_jpeg safe - // validate req_comp - if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + // validate req_comp + if (req_comp < 0 || req_comp > 4) + return stbi__errpuc("bad req_comp", "Internal error"); - // load a jpeg image from whichever source, but leave in YCbCr format - if (!stbi__decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + // load a jpeg image from whichever source, but leave in YCbCr format + if (!stbi__decode_jpeg_image(z)) { + stbi__cleanup_jpeg(z); + return NULL; + } - // determine actual number of components to generate - n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; - is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + is_rgb = z->s->img_n == 3 && + (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); - if (z->s->img_n == 3 && n < 3 && !is_rgb) - decode_n = 1; - else - decode_n = z->s->img_n; + if (z->s->img_n == 3 && n < 3 && !is_rgb) + decode_n = 1; + else + decode_n = z->s->img_n; - // resample and color-convert - { - int k; - unsigned int i,j; - stbi_uc *output; - stbi_uc *coutput[4] = { NULL, NULL, NULL, NULL }; + // resample and color-convert + { + int k; + unsigned int i, j; + stbi_uc *output; + stbi_uc *coutput[4] = {NULL, NULL, NULL, NULL}; - stbi__resample res_comp[4]; + stbi__resample res_comp[4]; - for (k=0; k < decode_n; ++k) { - stbi__resample *r = &res_comp[k]; + for (k = 0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; - // allocate line buffer big enough for upsampling off the edges - // with upsample factor of 4 - z->img_comp[k].linebuf = (stbi_uc *) stbi__malloc(z->s->img_x + 3); - if (!z->img_comp[k].linebuf) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } - - r->hs = z->img_h_max / z->img_comp[k].h; - r->vs = z->img_v_max / z->img_comp[k].v; - r->ystep = r->vs >> 1; - r->w_lores = (z->s->img_x + r->hs-1) / r->hs; - r->ypos = 0; - r->line0 = r->line1 = z->img_comp[k].data; - - if (r->hs == 1 && r->vs == 1) r->resample = resample_row_1; - else if (r->hs == 1 && r->vs == 2) r->resample = stbi__resample_row_v_2; - else if (r->hs == 2 && r->vs == 1) r->resample = stbi__resample_row_h_2; - else if (r->hs == 2 && r->vs == 2) r->resample = z->resample_row_hv_2_kernel; - else r->resample = stbi__resample_row_generic; + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (stbi_uc *)stbi__malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { + stbi__cleanup_jpeg(z); + return stbi__errpuc("outofmem", "Out of memory"); } - // can't error after this so, this is safe - output = (stbi_uc *) stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); - if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + r->hs = z->img_h_max / z->img_comp[k].h; + r->vs = z->img_v_max / z->img_comp[k].v; + r->ystep = r->vs >> 1; + r->w_lores = (z->s->img_x + r->hs - 1) / r->hs; + r->ypos = 0; + r->line0 = r->line1 = z->img_comp[k].data; - // now go ahead and resample - for (j=0; j < z->s->img_y; ++j) { - stbi_uc *out = output + n * z->s->img_x * j; - for (k=0; k < decode_n; ++k) { - stbi__resample *r = &res_comp[k]; - int y_bot = r->ystep >= (r->vs >> 1); - coutput[k] = r->resample(z->img_comp[k].linebuf, - y_bot ? r->line1 : r->line0, - y_bot ? r->line0 : r->line1, - r->w_lores, r->hs); - if (++r->ystep >= r->vs) { - r->ystep = 0; - r->line0 = r->line1; - if (++r->ypos < z->img_comp[k].y) - r->line1 += z->img_comp[k].w2; - } - } - if (n >= 3) { - stbi_uc *y = coutput[0]; - if (z->s->img_n == 3) { - if (is_rgb) { - for (i=0; i < z->s->img_x; ++i) { - out[0] = y[i]; - out[1] = coutput[1][i]; - out[2] = coutput[2][i]; - out[3] = 255; - out += n; - } - } else { - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - } - } else if (z->s->img_n == 4) { - if (z->app14_color_transform == 0) { // CMYK - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - out[0] = stbi__blinn_8x8(coutput[0][i], m); - out[1] = stbi__blinn_8x8(coutput[1][i], m); - out[2] = stbi__blinn_8x8(coutput[2][i], m); - out[3] = 255; - out += n; - } - } else if (z->app14_color_transform == 2) { // YCCK - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - out[0] = stbi__blinn_8x8(255 - out[0], m); - out[1] = stbi__blinn_8x8(255 - out[1], m); - out[2] = stbi__blinn_8x8(255 - out[2], m); - out += n; - } - } else { // YCbCr + alpha? Ignore the fourth channel for now - z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); - } - } else - for (i=0; i < z->s->img_x; ++i) { - out[0] = out[1] = out[2] = y[i]; - out[3] = 255; // not used if n==3 - out += n; - } - } else { - if (is_rgb) { - if (n == 1) - for (i=0; i < z->s->img_x; ++i) - *out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); - else { - for (i=0; i < z->s->img_x; ++i, out += 2) { - out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); - out[1] = 255; - } - } - } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { - for (i=0; i < z->s->img_x; ++i) { - stbi_uc m = coutput[3][i]; - stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); - stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); - stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); - out[0] = stbi__compute_y(r, g, b); - out[1] = 255; - out += n; - } - } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { - for (i=0; i < z->s->img_x; ++i) { - out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); - out[1] = 255; - out += n; - } - } else { - stbi_uc *y = coutput[0]; - if (n == 1) - for (i=0; i < z->s->img_x; ++i) out[i] = y[i]; - else - for (i=0; i < z->s->img_x; ++i) { *out++ = y[i]; *out++ = 255; } - } - } - } + if (r->hs == 1 && r->vs == 1) + r->resample = resample_row_1; + else if (r->hs == 1 && r->vs == 2) + r->resample = stbi__resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) + r->resample = stbi__resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) + r->resample = z->resample_row_hv_2_kernel; + else + r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc *)stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); + if (!output) { stbi__cleanup_jpeg(z); - *out_x = z->s->img_x; - *out_y = z->s->img_y; - if (comp) *comp = z->s->img_n >= 3 ? 3 : 1; // report original components, not output - return output; - } + return stbi__errpuc("outofmem", "Out of memory"); + } + + // now go ahead and resample + for (j = 0; j < z->s->img_y; ++j) { + stbi_uc *out = output + n * z->s->img_x * j; + for (k = 0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + int y_bot = r->ystep >= (r->vs >> 1); + coutput[k] = + r->resample(z->img_comp[k].linebuf, y_bot ? r->line1 : r->line0, + y_bot ? r->line0 : r->line1, r->w_lores, r->hs); + if (++r->ystep >= r->vs) { + r->ystep = 0; + r->line0 = r->line1; + if (++r->ypos < z->img_comp[k].y) r->line1 += z->img_comp[k].w2; + } + } + if (n >= 3) { + stbi_uc *y = coutput[0]; + if (z->s->img_n == 3) { + if (is_rgb) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + } + } else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, + n); + } + } else + for (i = 0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + if (is_rgb) { + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + *out++ = + stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i = 0; i < z->s->img_x; ++i, out += 2) { + out[0] = + stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + out += n; + } + } else { + stbi_uc *y = coutput[0]; + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + out[i] = y[i]; + else + for (i = 0; i < z->s->img_x; ++i) { + *out++ = y[i]; + *out++ = 255; + } + } + } + } + stbi__cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) + *comp = + z->s->img_n >= 3 ? 3 : 1; // report original components, not output + return output; + } } -static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - unsigned char* result; - stbi__jpeg* j = (stbi__jpeg*) stbi__malloc(sizeof(stbi__jpeg)); - STBI_NOTUSED(ri); - j->s = s; - stbi__setup_jpeg(j); - result = load_jpeg_image(j, x,y,comp,req_comp); - STBI_FREE(j); - return result; +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + unsigned char *result; + stbi__jpeg *j = (stbi__jpeg *)stbi__malloc(sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x, y, comp, req_comp); + STBI_FREE(j); + return result; } -static int stbi__jpeg_test(stbi__context *s) -{ - int r; - stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); - j->s = s; - stbi__setup_jpeg(j); - r = stbi__decode_jpeg_header(j, STBI__SCAN_type); - stbi__rewind(s); - STBI_FREE(j); - return r; +static int stbi__jpeg_test(stbi__context *s) { + int r; + stbi__jpeg *j = (stbi__jpeg *)stbi__malloc(sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; } -static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) -{ - if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { - stbi__rewind( j->s ); - return 0; - } - if (x) *x = j->s->img_x; - if (y) *y = j->s->img_y; - if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; - return 1; +static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) { + if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { + stbi__rewind(j->s); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; + return 1; } -static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) -{ - int result; - stbi__jpeg* j = (stbi__jpeg*) (stbi__malloc(sizeof(stbi__jpeg))); - j->s = s; - result = stbi__jpeg_info_raw(j, x, y, comp); - STBI_FREE(j); - return result; +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) { + int result; + stbi__jpeg *j = (stbi__jpeg *)(stbi__malloc(sizeof(stbi__jpeg))); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; } -#endif + #endif // public domain zlib decode v0.2 Sean Barrett 2006-11-18 // simple implementation @@ -3974,86 +4314,83 @@ static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) // performance // - fast huffman -#ifndef STBI_NO_ZLIB + #ifndef STBI_NO_ZLIB -// fast-way is faster to check than jpeg huffman, but slow way is slower -#define STBI__ZFAST_BITS 9 // accelerate all cases in default tables -#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) + // fast-way is faster to check than jpeg huffman, but slow way is slower + #define STBI__ZFAST_BITS 9 // accelerate all cases in default tables + #define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) // zlib-style huffman encoding // (jpegs packs from left, zlib from right, so can't share code) -typedef struct -{ - stbi__uint16 fast[1 << STBI__ZFAST_BITS]; - stbi__uint16 firstcode[16]; - int maxcode[17]; - stbi__uint16 firstsymbol[16]; - stbi_uc size[288]; - stbi__uint16 value[288]; +typedef struct { + stbi__uint16 fast[1 << STBI__ZFAST_BITS]; + stbi__uint16 firstcode[16]; + int maxcode[17]; + stbi__uint16 firstsymbol[16]; + stbi_uc size[288]; + stbi__uint16 value[288]; } stbi__zhuffman; -stbi_inline static int stbi__bitreverse16(int n) -{ - n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); - n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); - n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); - n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); +stbi_inline static int stbi__bitreverse16(int n) { + n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); + n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); + n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); + n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); return n; } -stbi_inline static int stbi__bit_reverse(int v, int bits) -{ - STBI_ASSERT(bits <= 16); - // to bit reverse n bits, reverse 16 and shift - // e.g. 11 bits, bit reverse and shift away 5 - return stbi__bitreverse16(v) >> (16-bits); +stbi_inline static int stbi__bit_reverse(int v, int bits) { + STBI_ASSERT(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return stbi__bitreverse16(v) >> (16 - bits); } -static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int num) -{ - int i,k=0; - int code, next_code[16], sizes[17]; +static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, + int num) { + int i, k = 0; + int code, next_code[16], sizes[17]; - // DEFLATE spec for generating codes - memset(sizes, 0, sizeof(sizes)); - memset(z->fast, 0, sizeof(z->fast)); - for (i=0; i < num; ++i) - ++sizes[sizelist[i]]; - sizes[0] = 0; - for (i=1; i < 16; ++i) - if (sizes[i] > (1 << i)) - return stbi__err("bad sizes", "Corrupt PNG"); - code = 0; - for (i=1; i < 16; ++i) { - next_code[i] = code; - z->firstcode[i] = (stbi__uint16) code; - z->firstsymbol[i] = (stbi__uint16) k; - code = (code + sizes[i]); - if (sizes[i]) - if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt PNG"); - z->maxcode[i] = code << (16-i); // preshift for inner loop - code <<= 1; - k += sizes[i]; - } - z->maxcode[16] = 0x10000; // sentinel - for (i=0; i < num; ++i) { - int s = sizelist[i]; - if (s) { - int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; - stbi__uint16 fastv = (stbi__uint16) ((s << 9) | i); - z->size [c] = (stbi_uc ) s; - z->value[c] = (stbi__uint16) i; - if (s <= STBI__ZFAST_BITS) { - int j = stbi__bit_reverse(next_code[s],s); - while (j < (1 << STBI__ZFAST_BITS)) { - z->fast[j] = fastv; - j += (1 << s); - } - } - ++next_code[s]; + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 0, sizeof(z->fast)); + for (i = 0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i = 1; i < 16; ++i) + if (sizes[i] > (1 << i)) return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i = 1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (stbi__uint16)code; + z->firstsymbol[i] = (stbi__uint16)k; + code = (code + sizes[i]); + if (sizes[i]) + if (code - 1 >= (1 << i)) + return stbi__err("bad codelengths", "Corrupt PNG"); + z->maxcode[i] = code << (16 - i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i = 0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + stbi__uint16 fastv = (stbi__uint16)((s << 9) | i); + z->size[c] = (stbi_uc)s; + z->value[c] = (stbi__uint16)i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s], s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (1 << s); + } } - } - return 1; + ++next_code[s]; + } + } + return 1; } // zlib-from-memory implementation for PNG reading @@ -4062,277 +4399,291 @@ static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int // we require PNG read all the IDATs and combine them into a single // memory buffer -typedef struct -{ - stbi_uc *zbuffer, *zbuffer_end; - int num_bits; - stbi__uint32 code_buffer; +typedef struct { + stbi_uc *zbuffer, *zbuffer_end; + int num_bits; + stbi__uint32 code_buffer; - char *zout; - char *zout_start; - char *zout_end; - int z_expandable; + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; - stbi__zhuffman z_length, z_distance; + stbi__zhuffman z_length, z_distance; } stbi__zbuf; -stbi_inline static int stbi__zeof(stbi__zbuf *z) -{ - return (z->zbuffer >= z->zbuffer_end); +stbi_inline static int stbi__zeof(stbi__zbuf *z) { + return (z->zbuffer >= z->zbuffer_end); } -stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) -{ - return stbi__zeof(z) ? 0 : *z->zbuffer++; +stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) { + return stbi__zeof(z) ? 0 : *z->zbuffer++; } -static void stbi__fill_bits(stbi__zbuf *z) -{ - do { - if (z->code_buffer >= (1U << z->num_bits)) { - z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ - return; - } - z->code_buffer |= (unsigned int) stbi__zget8(z) << z->num_bits; - z->num_bits += 8; - } while (z->num_bits <= 24); +static void stbi__fill_bits(stbi__zbuf *z) { + do { + if (z->code_buffer >= (1U << z->num_bits)) { + z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ + return; + } + z->code_buffer |= (unsigned int)stbi__zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); } -stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) -{ - unsigned int k; - if (z->num_bits < n) stbi__fill_bits(z); - k = z->code_buffer & ((1 << n) - 1); - z->code_buffer >>= n; - z->num_bits -= n; - return k; +stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) { + unsigned int k; + if (z->num_bits < n) stbi__fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; } -static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) -{ - int b,s,k; - // not resolved by fast table, so compute it the slow way - // use jpeg approach, which requires MSbits at top - k = stbi__bit_reverse(a->code_buffer, 16); - for (s=STBI__ZFAST_BITS+1; ; ++s) - if (k < z->maxcode[s]) - break; - if (s >= 16) return -1; // invalid code! - // code size is s, so: - b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; - if (b >= sizeof (z->size)) return -1; // some data was corrupt somewhere! - if (z->size[b] != s) return -1; // was originally an assert, but report failure instead. - a->code_buffer >>= s; - a->num_bits -= s; - return z->value[b]; +static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) { + int b, s, k; + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = stbi__bit_reverse(a->code_buffer, 16); + for (s = STBI__ZFAST_BITS + 1;; ++s) + if (k < z->maxcode[s]) break; + if (s >= 16) return -1; // invalid code! + // code size is s, so: + b = (k >> (16 - s)) - z->firstcode[s] + z->firstsymbol[s]; + if (b >= sizeof(z->size)) return -1; // some data was corrupt somewhere! + if (z->size[b] != s) + return -1; // was originally an assert, but report failure instead. + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; } -stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) -{ - int b,s; - if (a->num_bits < 16) { - if (stbi__zeof(a)) { - return -1; /* report error for unexpected end of data. */ - } - stbi__fill_bits(a); - } - b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; - if (b) { - s = b >> 9; - a->code_buffer >>= s; - a->num_bits -= s; - return b & 511; - } - return stbi__zhuffman_decode_slowpath(a, z); +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) { + int b, s; + if (a->num_bits < 16) { + if (stbi__zeof(a)) { + return -1; /* report error for unexpected end of data. */ + } + stbi__fill_bits(a); + } + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); } -static int stbi__zexpand(stbi__zbuf *z, char *zout, int n) // need to make room for n bytes +static int stbi__zexpand(stbi__zbuf *z, char *zout, + int n) // need to make room for n bytes { - char *q; - unsigned int cur, limit, old_limit; - z->zout = zout; - if (!z->z_expandable) return stbi__err("output buffer limit","Corrupt PNG"); - cur = (unsigned int) (z->zout - z->zout_start); - limit = old_limit = (unsigned) (z->zout_end - z->zout_start); - if (UINT_MAX - cur < (unsigned) n) return stbi__err("outofmem", "Out of memory"); - while (cur + n > limit) { - if(limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); - limit *= 2; - } - q = (char *) STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); - STBI_NOTUSED(old_limit); - if (q == NULL) return stbi__err("outofmem", "Out of memory"); - z->zout_start = q; - z->zout = q + cur; - z->zout_end = q + limit; - return 1; + char *q; + unsigned int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit", "Corrupt PNG"); + cur = (unsigned int)(z->zout - z->zout_start); + limit = old_limit = (unsigned)(z->zout_end - z->zout_start); + if (UINT_MAX - cur < (unsigned)n) + return stbi__err("outofmem", "Out of memory"); + while (cur + n > limit) { + if (limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); + limit *= 2; + } + q = (char *)STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_limit); + if (q == NULL) return stbi__err("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; } static const int stbi__zlength_base[31] = { - 3,4,5,6,7,8,9,10,11,13, - 15,17,19,23,27,31,35,43,51,59, - 67,83,99,115,131,163,195,227,258,0,0 }; + 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, + 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0}; -static const int stbi__zlength_extra[31]= -{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 }; +static const int stbi__zlength_extra[31] = {0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, + 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, + 4, 4, 5, 5, 5, 5, 0, 0, 0}; -static const int stbi__zdist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193, -257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0}; +static const int stbi__zdist_base[32] = { + 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, + 49, 65, 97, 129, 193, 257, 385, 513, 769, 1025, 1537, + 2049, 3073, 4097, 6145, 8193, 12289, 16385, 24577, 0, 0}; -static const int stbi__zdist_extra[32] = -{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; +static const int stbi__zdist_extra[32] = {0, 0, 0, 0, 1, 1, 2, 2, 3, 3, + 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, + 9, 9, 10, 10, 11, 11, 12, 12, 13, 13}; -static int stbi__parse_huffman_block(stbi__zbuf *a) -{ - char *zout = a->zout; - for(;;) { - int z = stbi__zhuffman_decode(a, &a->z_length); - if (z < 256) { - if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); // error in huffman codes - if (zout >= a->zout_end) { - if (!stbi__zexpand(a, zout, 1)) return 0; - zout = a->zout; - } - *zout++ = (char) z; +static int stbi__parse_huffman_block(stbi__zbuf *a) { + char *zout = a->zout; + for (;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) + return stbi__err("bad huffman code", + "Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char)z; + } else { + stbi_uc *p; + int len, dist; + if (z == 256) { + a->zout = zout; + return 1; + } + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) + len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0) return stbi__err("bad huffman code", "Corrupt PNG"); + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) + return stbi__err("bad dist", "Corrupt PNG"); + if (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc *)(zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { + do + *zout++ = v; + while (--len); + } } else { - stbi_uc *p; - int len,dist; - if (z == 256) { - a->zout = zout; - return 1; - } - z -= 257; - len = stbi__zlength_base[z]; - if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); - z = stbi__zhuffman_decode(a, &a->z_distance); - if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); - dist = stbi__zdist_base[z]; - if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); - if (zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); - if (zout + len > a->zout_end) { - if (!stbi__zexpand(a, zout, len)) return 0; - zout = a->zout; - } - p = (stbi_uc *) (zout - dist); - if (dist == 1) { // run of one byte; common in images. - stbi_uc v = *p; - if (len) { do *zout++ = v; while (--len); } - } else { - if (len) { do *zout++ = *p++; while (--len); } - } + if (len) { + do + *zout++ = *p++; + while (--len); + } } - } + } + } } -static int stbi__compute_huffman_codes(stbi__zbuf *a) -{ - static const stbi_uc length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; - stbi__zhuffman z_codelength; - stbi_uc lencodes[286+32+137];//padding for maximum single op - stbi_uc codelength_sizes[19]; - int i,n; +static int stbi__compute_huffman_codes(stbi__zbuf *a) { + static const stbi_uc length_dezigzag[19] = { + 16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; + stbi__zhuffman z_codelength; + stbi_uc lencodes[286 + 32 + 137]; // padding for maximum single op + stbi_uc codelength_sizes[19]; + int i, n; - int hlit = stbi__zreceive(a,5) + 257; - int hdist = stbi__zreceive(a,5) + 1; - int hclen = stbi__zreceive(a,4) + 4; - int ntot = hlit + hdist; + int hlit = stbi__zreceive(a, 5) + 257; + int hdist = stbi__zreceive(a, 5) + 1; + int hclen = stbi__zreceive(a, 4) + 4; + int ntot = hlit + hdist; - memset(codelength_sizes, 0, sizeof(codelength_sizes)); - for (i=0; i < hclen; ++i) { - int s = stbi__zreceive(a,3); - codelength_sizes[length_dezigzag[i]] = (stbi_uc) s; - } - if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i = 0; i < hclen; ++i) { + int s = stbi__zreceive(a, 3); + codelength_sizes[length_dezigzag[i]] = (stbi_uc)s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; - n = 0; - while (n < ntot) { - int c = stbi__zhuffman_decode(a, &z_codelength); - if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); - if (c < 16) - lencodes[n++] = (stbi_uc) c; - else { - stbi_uc fill = 0; - if (c == 16) { - c = stbi__zreceive(a,2)+3; - if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); - fill = lencodes[n-1]; - } else if (c == 17) { - c = stbi__zreceive(a,3)+3; - } else if (c == 18) { - c = stbi__zreceive(a,7)+11; - } else { - return stbi__err("bad codelengths", "Corrupt PNG"); - } - if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); - memset(lencodes+n, fill, c); - n += c; + n = 0; + while (n < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc)c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a, 2) + 3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n - 1]; + } else if (c == 17) { + c = stbi__zreceive(a, 3) + 3; + } else if (c == 18) { + c = stbi__zreceive(a, 7) + 11; + } else { + return stbi__err("bad codelengths", "Corrupt PNG"); } - } - if (n != ntot) return stbi__err("bad codelengths","Corrupt PNG"); - if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; - if (!stbi__zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; - return 1; + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes + n, fill, c); + n += c; + } + } + if (n != ntot) return stbi__err("bad codelengths", "Corrupt PNG"); + if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, lencodes + hlit, hdist)) return 0; + return 1; } -static int stbi__parse_uncompressed_block(stbi__zbuf *a) -{ - stbi_uc header[4]; - int len,nlen,k; - if (a->num_bits & 7) - stbi__zreceive(a, a->num_bits & 7); // discard - // drain the bit-packed data into header - k = 0; - while (a->num_bits > 0) { - header[k++] = (stbi_uc) (a->code_buffer & 255); // suppress MSVC run-time check - a->code_buffer >>= 8; - a->num_bits -= 8; - } - if (a->num_bits < 0) return stbi__err("zlib corrupt","Corrupt PNG"); - // now fill header the normal way - while (k < 4) - header[k++] = stbi__zget8(a); - len = header[1] * 256 + header[0]; - nlen = header[3] * 256 + header[2]; - if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt","Corrupt PNG"); - if (a->zbuffer + len > a->zbuffer_end) return stbi__err("read past buffer","Corrupt PNG"); - if (a->zout + len > a->zout_end) - if (!stbi__zexpand(a, a->zout, len)) return 0; - memcpy(a->zout, a->zbuffer, len); - a->zbuffer += len; - a->zout += len; - return 1; +static int stbi__parse_uncompressed_block(stbi__zbuf *a) { + stbi_uc header[4]; + int len, nlen, k; + if (a->num_bits & 7) stbi__zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = + (stbi_uc)(a->code_buffer & 255); // suppress MSVC run-time check + a->code_buffer >>= 8; + a->num_bits -= 8; + } + if (a->num_bits < 0) return stbi__err("zlib corrupt", "Corrupt PNG"); + // now fill header the normal way + while (k < 4) + header[k++] = stbi__zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt", "Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) + return stbi__err("read past buffer", "Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!stbi__zexpand(a, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; } -static int stbi__parse_zlib_header(stbi__zbuf *a) -{ - int cmf = stbi__zget8(a); - int cm = cmf & 15; - /* int cinfo = cmf >> 4; */ - int flg = stbi__zget8(a); - if (stbi__zeof(a)) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec - if ((cmf*256+flg) % 31 != 0) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec - if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png - if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); // DEFLATE required for png - // window = 1 << (8 + cinfo)... but who cares, we fully buffer output - return 1; +static int stbi__parse_zlib_header(stbi__zbuf *a) { + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if (stbi__zeof(a)) + return stbi__err("bad zlib header", "Corrupt PNG"); // zlib spec + if ((cmf * 256 + flg) % 31 != 0) + return stbi__err("bad zlib header", "Corrupt PNG"); // zlib spec + if (flg & 32) + return stbi__err("no preset dict", + "Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) + return stbi__err("bad compression", + "Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; } -static const stbi_uc stbi__zdefault_length[288] = -{ - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 -}; -static const stbi_uc stbi__zdefault_distance[32] = -{ - 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 -}; +static const stbi_uc stbi__zdefault_length[288] = { + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, + 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 8}; +static const stbi_uc stbi__zdefault_distance[32] = { + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5}; /* Init algorithm: { @@ -4346,119 +4697,123 @@ Init algorithm: } */ -static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) -{ - int final, type; - if (parse_header) - if (!stbi__parse_zlib_header(a)) return 0; - a->num_bits = 0; - a->code_buffer = 0; - do { - final = stbi__zreceive(a,1); - type = stbi__zreceive(a,2); - if (type == 0) { - if (!stbi__parse_uncompressed_block(a)) return 0; - } else if (type == 3) { - return 0; +static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) { + int final, type; + if (parse_header) + if (!stbi__parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + do { + final = stbi__zreceive(a, 1); + type = stbi__zreceive(a, 2); + if (type == 0) { + if (!stbi__parse_uncompressed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + // use fixed code lengths + if (!stbi__zbuild_huffman(&a->z_length, stbi__zdefault_length, 288)) + return 0; + if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) + return 0; } else { - if (type == 1) { - // use fixed code lengths - if (!stbi__zbuild_huffman(&a->z_length , stbi__zdefault_length , 288)) return 0; - if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) return 0; - } else { - if (!stbi__compute_huffman_codes(a)) return 0; - } - if (!stbi__parse_huffman_block(a)) return 0; + if (!stbi__compute_huffman_codes(a)) return 0; } - } while (!final); - return 1; + if (!stbi__parse_huffman_block(a)) return 0; + } + } while (!final); + return 1; } -static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, int parse_header) -{ - a->zout_start = obuf; - a->zout = obuf; - a->zout_end = obuf + olen; - a->z_expandable = exp; +static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, + int parse_header) { + a->zout_start = obuf; + a->zout = obuf; + a->zout_end = obuf + olen; + a->z_expandable = exp; - return stbi__parse_zlib(a, parse_header); + return stbi__parse_zlib(a, parse_header); } -STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(initial_size); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer + len; - if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, + int initial_size, int *outlen) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) -{ - return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, + int *outlen) { + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); } -STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(initial_size); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer + len; - if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, + int len, + int initial_size, + int *outlen, + int parse_header) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) -{ - stbi__zbuf a; - a.zbuffer = (stbi_uc *) ibuffer; - a.zbuffer_end = (stbi_uc *) ibuffer + ilen; - if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) - return (int) (a.zout - a.zout_start); - else - return -1; +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, + char const *ibuffer, int ilen) { + stbi__zbuf a; + a.zbuffer = (stbi_uc *)ibuffer; + a.zbuffer_end = (stbi_uc *)ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) + return (int)(a.zout - a.zout_start); + else + return -1; } -STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) -{ - stbi__zbuf a; - char *p = (char *) stbi__malloc(16384); - if (p == NULL) return NULL; - a.zbuffer = (stbi_uc *) buffer; - a.zbuffer_end = (stbi_uc *) buffer+len; - if (stbi__do_zlib(&a, p, 16384, 1, 0)) { - if (outlen) *outlen = (int) (a.zout - a.zout_start); - return a.zout_start; - } else { - STBI_FREE(a.zout_start); - return NULL; - } +STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, + int *outlen) { + stbi__zbuf a; + char *p = (char *)stbi__malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *)buffer; + a.zbuffer_end = (stbi_uc *)buffer + len; + if (stbi__do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int)(a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } } -STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) -{ - stbi__zbuf a; - a.zbuffer = (stbi_uc *) ibuffer; - a.zbuffer_end = (stbi_uc *) ibuffer + ilen; - if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) - return (int) (a.zout - a.zout_start); - else - return -1; +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, + const char *ibuffer, int ilen) { + stbi__zbuf a; + a.zbuffer = (stbi_uc *)ibuffer; + a.zbuffer_end = (stbi_uc *)ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) + return (int)(a.zout - a.zout_start); + else + return -1; } -#endif + #endif // public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 // simple implementation @@ -4470,713 +4825,827 @@ STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char // performance // - uses stb_zlib, a PD zlib implementation with fast huffman decoding -#ifndef STBI_NO_PNG -typedef struct -{ - stbi__uint32 length; - stbi__uint32 type; + #ifndef STBI_NO_PNG +typedef struct { + stbi__uint32 length; + stbi__uint32 type; } stbi__pngchunk; -static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) -{ - stbi__pngchunk c; - c.length = stbi__get32be(s); - c.type = stbi__get32be(s); - return c; +static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) { + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; } -static int stbi__check_png_header(stbi__context *s) -{ - static const stbi_uc png_sig[8] = { 137,80,78,71,13,10,26,10 }; - int i; - for (i=0; i < 8; ++i) - if (stbi__get8(s) != png_sig[i]) return stbi__err("bad png sig","Not a PNG"); - return 1; +static int stbi__check_png_header(stbi__context *s) { + static const stbi_uc png_sig[8] = {137, 80, 78, 71, 13, 10, 26, 10}; + int i; + for (i = 0; i < 8; ++i) + if (stbi__get8(s) != png_sig[i]) + return stbi__err("bad png sig", "Not a PNG"); + return 1; } -typedef struct -{ - stbi__context *s; - stbi_uc *idata, *expanded, *out; - int depth; +typedef struct { + stbi__context *s; + stbi_uc *idata, *expanded, *out; + int depth; } stbi__png; - enum { - STBI__F_none=0, - STBI__F_sub=1, - STBI__F_up=2, - STBI__F_avg=3, - STBI__F_paeth=4, - // synthetic filters used for first scanline to avoid needing a dummy row of 0s - STBI__F_avg_first, - STBI__F_paeth_first + STBI__F_none = 0, + STBI__F_sub = 1, + STBI__F_up = 2, + STBI__F_avg = 3, + STBI__F_paeth = 4, + // synthetic filters used for first scanline to avoid needing a dummy row of + // 0s + STBI__F_avg_first, + STBI__F_paeth_first }; -static stbi_uc first_row_filter[5] = -{ - STBI__F_none, - STBI__F_sub, - STBI__F_none, - STBI__F_avg_first, - STBI__F_paeth_first -}; +static stbi_uc first_row_filter[5] = {STBI__F_none, STBI__F_sub, STBI__F_none, + STBI__F_avg_first, STBI__F_paeth_first}; -static int stbi__paeth(int a, int b, int c) -{ - int p = a + b - c; - int pa = abs(p-a); - int pb = abs(p-b); - int pc = abs(p-c); - if (pa <= pb && pa <= pc) return a; - if (pb <= pc) return b; - return c; +static int stbi__paeth(int a, int b, int c) { + int p = a + b - c; + int pa = abs(p - a); + int pb = abs(p - b); + int pc = abs(p - c); + if (pa <= pb && pa <= pc) return a; + if (pb <= pc) return b; + return c; } -static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; +static const stbi_uc stbi__depth_scale_table[9] = {0, 0xff, 0x55, 0, 0x11, + 0, 0, 0, 0x01}; // create the png data from post-deflated data -static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y, int depth, int color) -{ - int bytes = (depth == 16? 2 : 1); - stbi__context *s = a->s; - stbi__uint32 i,j,stride = x*out_n*bytes; - stbi__uint32 img_len, img_width_bytes; - int k; - int img_n = s->img_n; // copy it into a local for later +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, + stbi__uint32 raw_len, int out_n, + stbi__uint32 x, stbi__uint32 y, int depth, + int color) { + int bytes = (depth == 16 ? 2 : 1); + stbi__context *s = a->s; + stbi__uint32 i, j, stride = x * out_n * bytes; + stbi__uint32 img_len, img_width_bytes; + int k; + int img_n = s->img_n; // copy it into a local for later - int output_bytes = out_n*bytes; - int filter_bytes = img_n*bytes; - int width = x; + int output_bytes = out_n * bytes; + int filter_bytes = img_n * bytes; + int width = x; - STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); - a->out = (stbi_uc *) stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into - if (!a->out) return stbi__err("outofmem", "Out of memory"); + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n + 1); + a->out = (stbi_uc *)stbi__malloc_mad3( + x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); - if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); - img_width_bytes = (((img_n * x * depth) + 7) >> 3); - img_len = (img_width_bytes + 1) * y; + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) + return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; - // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, - // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), - // so just check for raw_len < img_len always. - if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + // we used to check for exact match between raw_len and img_len on + // non-interlaced PNGs, but issue #276 reported a PNG in the wild that had + // extra data at the end (all zeros), so just check for raw_len < img_len + // always. + if (raw_len < img_len) return stbi__err("not enough pixels", "Corrupt PNG"); - for (j=0; j < y; ++j) { - stbi_uc *cur = a->out + stride*j; - stbi_uc *prior; - int filter = *raw++; + for (j = 0; j < y; ++j) { + stbi_uc *cur = a->out + stride * j; + stbi_uc *prior; + int filter = *raw++; - if (filter > 4) - return stbi__err("invalid filter","Corrupt PNG"); + if (filter > 4) return stbi__err("invalid filter", "Corrupt PNG"); - if (depth < 8) { - if (img_width_bytes > x) return stbi__err("invalid width","Corrupt PNG"); - cur += x*out_n - img_width_bytes; // store output to the rightmost img_len bytes, so we can decode in place - filter_bytes = 1; - width = img_width_bytes; + if (depth < 8) { + if (img_width_bytes > x) return stbi__err("invalid width", "Corrupt PNG"); + cur += + x * out_n - img_width_bytes; // store output to the rightmost img_len + // bytes, so we can decode in place + filter_bytes = 1; + width = img_width_bytes; + } + prior = cur - stride; // bugfix: need to compute this after 'cur +=' + // computation above + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // handle first byte explicitly + for (k = 0; k < filter_bytes; ++k) { + switch (filter) { + case STBI__F_none: + cur[k] = raw[k]; + break; + case STBI__F_sub: + cur[k] = raw[k]; + break; + case STBI__F_up: + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + break; + case STBI__F_avg: + cur[k] = STBI__BYTECAST(raw[k] + (prior[k] >> 1)); + break; + case STBI__F_paeth: + cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0, prior[k], 0)); + break; + case STBI__F_avg_first: + cur[k] = raw[k]; + break; + case STBI__F_paeth_first: + cur[k] = raw[k]; + break; } - prior = cur - stride; // bugfix: need to compute this after 'cur +=' computation above + } - // if first row, use special filter that doesn't sample previous row - if (j == 0) filter = first_row_filter[filter]; - - // handle first byte explicitly - for (k=0; k < filter_bytes; ++k) { - switch (filter) { - case STBI__F_none : cur[k] = raw[k]; break; - case STBI__F_sub : cur[k] = raw[k]; break; - case STBI__F_up : cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; - case STBI__F_avg : cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); break; - case STBI__F_paeth : cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0,prior[k],0)); break; - case STBI__F_avg_first : cur[k] = raw[k]; break; - case STBI__F_paeth_first: cur[k] = raw[k]; break; - } + if (depth == 8) { + if (img_n != out_n) cur[img_n] = 255; // first pixel + raw += img_n; + cur += out_n; + prior += out_n; + } else if (depth == 16) { + if (img_n != out_n) { + cur[filter_bytes] = 255; // first pixel top byte + cur[filter_bytes + 1] = 255; // first pixel bottom byte } + raw += filter_bytes; + cur += output_bytes; + prior += output_bytes; + } else { + raw += 1; + cur += 1; + prior += 1; + } - if (depth == 8) { - if (img_n != out_n) - cur[img_n] = 255; // first pixel - raw += img_n; - cur += out_n; - prior += out_n; - } else if (depth == 16) { - if (img_n != out_n) { - cur[filter_bytes] = 255; // first pixel top byte - cur[filter_bytes+1] = 255; // first pixel bottom byte - } - raw += filter_bytes; - cur += output_bytes; - prior += output_bytes; - } else { - raw += 1; - cur += 1; - prior += 1; + // this is a little gross, so that we don't switch per-pixel or + // per-component + if (depth < 8 || img_n == out_n) { + int nk = (width - 1) * filter_bytes; + #define STBI__CASE(f) \ + case f: \ + for (k = 0; k < nk; ++k) + switch (filter) { + // "none" filter turns into a memcpy here; make that explicit. + case STBI__F_none: + memcpy(cur, raw, nk); + break; + STBI__CASE(STBI__F_sub) { + cur[k] = STBI__BYTECAST(raw[k] + cur[k - filter_bytes]); + } + break; + STBI__CASE(STBI__F_up) { + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + } + break; + STBI__CASE(STBI__F_avg) { + cur[k] = STBI__BYTECAST(raw[k] + + ((prior[k] + cur[k - filter_bytes]) >> 1)); + } + break; + STBI__CASE(STBI__F_paeth) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - filter_bytes], prior[k], + prior[k - filter_bytes])); + } + break; + STBI__CASE(STBI__F_avg_first) { + cur[k] = STBI__BYTECAST(raw[k] + (cur[k - filter_bytes] >> 1)); + } + break; + STBI__CASE(STBI__F_paeth_first) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - filter_bytes], 0, 0)); + } + break; } - - // this is a little gross, so that we don't switch per-pixel or per-component - if (depth < 8 || img_n == out_n) { - int nk = (width - 1)*filter_bytes; - #define STBI__CASE(f) \ - case f: \ - for (k=0; k < nk; ++k) - switch (filter) { - // "none" filter turns into a memcpy here; make that explicit. - case STBI__F_none: memcpy(cur, raw, nk); break; - STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); } break; - STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; - STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); } break; - STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],prior[k],prior[k-filter_bytes])); } break; - STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); } break; - STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],0,0)); } break; - } - #undef STBI__CASE - raw += nk; - } else { - STBI_ASSERT(img_n+1 == out_n); - #define STBI__CASE(f) \ - case f: \ - for (i=x-1; i >= 1; --i, cur[filter_bytes]=255,raw+=filter_bytes,cur+=output_bytes,prior+=output_bytes) \ - for (k=0; k < filter_bytes; ++k) - switch (filter) { - STBI__CASE(STBI__F_none) { cur[k] = raw[k]; } break; - STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k- output_bytes]); } break; - STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; - STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k- output_bytes])>>1)); } break; - STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],prior[k],prior[k- output_bytes])); } break; - STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k- output_bytes] >> 1)); } break; - STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],0,0)); } break; - } - #undef STBI__CASE - - // the loop above sets the high byte of the pixels' alpha, but for - // 16 bit png files we also need the low byte set. we'll do that here. - if (depth == 16) { - cur = a->out + stride*j; // start at the beginning of the row again - for (i=0; i < x; ++i,cur+=output_bytes) { - cur[filter_bytes+1] = 255; - } - } + #undef STBI__CASE + raw += nk; + } else { + STBI_ASSERT(img_n + 1 == out_n); + #define STBI__CASE(f) \ + case f: \ + for (i = x - 1; i >= 1; --i, cur[filter_bytes] = 255, \ + raw += filter_bytes, cur += output_bytes, prior += output_bytes) \ + for (k = 0; k < filter_bytes; ++k) + switch (filter) { + STBI__CASE(STBI__F_none) { + cur[k] = raw[k]; + } + break; + STBI__CASE(STBI__F_sub) { + cur[k] = STBI__BYTECAST(raw[k] + cur[k - output_bytes]); + } + break; + STBI__CASE(STBI__F_up) { + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + } + break; + STBI__CASE(STBI__F_avg) { + cur[k] = STBI__BYTECAST(raw[k] + + ((prior[k] + cur[k - output_bytes]) >> 1)); + } + break; + STBI__CASE(STBI__F_paeth) { + cur[k] = STBI__BYTECAST(raw[k] + + stbi__paeth(cur[k - output_bytes], prior[k], + prior[k - output_bytes])); + } + break; + STBI__CASE(STBI__F_avg_first) { + cur[k] = STBI__BYTECAST(raw[k] + (cur[k - output_bytes] >> 1)); + } + break; + STBI__CASE(STBI__F_paeth_first) { + cur[k] = + STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - output_bytes], 0, 0)); + } + break; } - } + #undef STBI__CASE - // we make a separate pass to expand bits to pixels; for performance, - // this could run two scanlines behind the above code, so it won't - // intefere with filtering but will still be in the cache. - if (depth < 8) { - for (j=0; j < y; ++j) { - stbi_uc *cur = a->out + stride*j; - stbi_uc *in = a->out + stride*j + x*out_n - img_width_bytes; - // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common 8-bit path optimal at minimal cost for 1/2/4-bit - // png guarante byte alignment, if width is not multiple of 8/4/2 we'll decode dummy trailing data that will be skipped in the later loop - stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range - - // note that the final byte might overshoot and write more data than desired. - // we can allocate enough data that this never writes out of memory, but it - // could also overwrite the next scanline. can it overwrite non-empty data - // on the next scanline? yes, consider 1-pixel-wide scanlines with 1-bit-per-pixel. - // so we need to explicitly clamp the final ones - - if (depth == 4) { - for (k=x*img_n; k >= 2; k-=2, ++in) { - *cur++ = scale * ((*in >> 4) ); - *cur++ = scale * ((*in ) & 0x0f); - } - if (k > 0) *cur++ = scale * ((*in >> 4) ); - } else if (depth == 2) { - for (k=x*img_n; k >= 4; k-=4, ++in) { - *cur++ = scale * ((*in >> 6) ); - *cur++ = scale * ((*in >> 4) & 0x03); - *cur++ = scale * ((*in >> 2) & 0x03); - *cur++ = scale * ((*in ) & 0x03); - } - if (k > 0) *cur++ = scale * ((*in >> 6) ); - if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); - if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); - } else if (depth == 1) { - for (k=x*img_n; k >= 8; k-=8, ++in) { - *cur++ = scale * ((*in >> 7) ); - *cur++ = scale * ((*in >> 6) & 0x01); - *cur++ = scale * ((*in >> 5) & 0x01); - *cur++ = scale * ((*in >> 4) & 0x01); - *cur++ = scale * ((*in >> 3) & 0x01); - *cur++ = scale * ((*in >> 2) & 0x01); - *cur++ = scale * ((*in >> 1) & 0x01); - *cur++ = scale * ((*in ) & 0x01); - } - if (k > 0) *cur++ = scale * ((*in >> 7) ); - if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); - if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); - if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); - if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); - if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); - if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); - } - if (img_n != out_n) { - int q; - // insert alpha = 255 - cur = a->out + stride*j; - if (img_n == 1) { - for (q=x-1; q >= 0; --q) { - cur[q*2+1] = 255; - cur[q*2+0] = cur[q]; - } - } else { - STBI_ASSERT(img_n == 3); - for (q=x-1; q >= 0; --q) { - cur[q*4+3] = 255; - cur[q*4+2] = cur[q*3+2]; - cur[q*4+1] = cur[q*3+1]; - cur[q*4+0] = cur[q*3+0]; - } - } - } + // the loop above sets the high byte of the pixels' alpha, but for + // 16 bit png files we also need the low byte set. we'll do that here. + if (depth == 16) { + cur = a->out + stride * j; // start at the beginning of the row again + for (i = 0; i < x; ++i, cur += output_bytes) { + cur[filter_bytes + 1] = 255; + } } - } else if (depth == 16) { - // force the image data from big-endian to platform-native. - // this is done in a separate pass due to the decoding relying - // on the data being untouched, but could probably be done - // per-line during decode if care is taken. - stbi_uc *cur = a->out; - stbi__uint16 *cur16 = (stbi__uint16*)cur; + } + } - for(i=0; i < x*y*out_n; ++i,cur16++,cur+=2) { - *cur16 = (cur[0] << 8) | cur[1]; + // we make a separate pass to expand bits to pixels; for performance, + // this could run two scanlines behind the above code, so it won't + // intefere with filtering but will still be in the cache. + if (depth < 8) { + for (j = 0; j < y; ++j) { + stbi_uc *cur = a->out + stride * j; + stbi_uc *in = a->out + stride * j + x * out_n - img_width_bytes; + // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common + // 8-bit path optimal at minimal cost for 1/2/4-bit png guarante byte + // alignment, if width is not multiple of 8/4/2 we'll decode dummy + // trailing data that will be skipped in the later loop + stbi_uc scale = (color == 0) + ? stbi__depth_scale_table[depth] + : 1; // scale grayscale values to 0..255 range + + // note that the final byte might overshoot and write more data than + // desired. we can allocate enough data that this never writes out of + // memory, but it could also overwrite the next scanline. can it overwrite + // non-empty data on the next scanline? yes, consider 1-pixel-wide + // scanlines with 1-bit-per-pixel. so we need to explicitly clamp the + // final ones + + if (depth == 4) { + for (k = x * img_n; k >= 2; k -= 2, ++in) { + *cur++ = scale * ((*in >> 4)); + *cur++ = scale * ((*in) & 0x0f); + } + if (k > 0) *cur++ = scale * ((*in >> 4)); + } else if (depth == 2) { + for (k = x * img_n; k >= 4; k -= 4, ++in) { + *cur++ = scale * ((*in >> 6)); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in) & 0x03); + } + if (k > 0) *cur++ = scale * ((*in >> 6)); + if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); + } else if (depth == 1) { + for (k = x * img_n; k >= 8; k -= 8, ++in) { + *cur++ = scale * ((*in >> 7)); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in) & 0x01); + } + if (k > 0) *cur++ = scale * ((*in >> 7)); + if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); } - } + if (img_n != out_n) { + int q; + // insert alpha = 255 + cur = a->out + stride * j; + if (img_n == 1) { + for (q = x - 1; q >= 0; --q) { + cur[q * 2 + 1] = 255; + cur[q * 2 + 0] = cur[q]; + } + } else { + STBI_ASSERT(img_n == 3); + for (q = x - 1; q >= 0; --q) { + cur[q * 4 + 3] = 255; + cur[q * 4 + 2] = cur[q * 3 + 2]; + cur[q * 4 + 1] = cur[q * 3 + 1]; + cur[q * 4 + 0] = cur[q * 3 + 0]; + } + } + } + } + } else if (depth == 16) { + // force the image data from big-endian to platform-native. + // this is done in a separate pass due to the decoding relying + // on the data being untouched, but could probably be done + // per-line during decode if care is taken. + stbi_uc *cur = a->out; + stbi__uint16 *cur16 = (stbi__uint16 *)cur; - return 1; + for (i = 0; i < x * y * out_n; ++i, cur16++, cur += 2) { + *cur16 = (cur[0] << 8) | cur[1]; + } + } + + return 1; } -static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) -{ - int bytes = (depth == 16 ? 2 : 1); - int out_bytes = out_n * bytes; - stbi_uc *final; - int p; - if (!interlaced) - return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); +static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, + stbi__uint32 image_data_len, int out_n, + int depth, int color, int interlaced) { + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, + a->s->img_x, a->s->img_y, depth, color); - // de-interlacing - final = (stbi_uc *) stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); - for (p=0; p < 7; ++p) { - int xorig[] = { 0,4,0,2,0,1,0 }; - int yorig[] = { 0,0,4,0,2,0,1 }; - int xspc[] = { 8,8,4,4,2,2,1 }; - int yspc[] = { 8,8,8,4,4,2,2 }; - int i,j,x,y; - // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 - x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; - y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; - if (x && y) { - stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; - if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { - STBI_FREE(final); - return 0; - } - for (j=0; j < y; ++j) { - for (i=0; i < x; ++i) { - int out_y = j*yspc[p]+yorig[p]; - int out_x = i*xspc[p]+xorig[p]; - memcpy(final + out_y*a->s->img_x*out_bytes + out_x*out_bytes, - a->out + (j*x+i)*out_bytes, out_bytes); - } - } - STBI_FREE(a->out); - image_data += img_len; - image_data_len -= img_len; + // de-interlacing + final = (stbi_uc *)stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + for (p = 0; p < 7; ++p) { + int xorig[] = {0, 4, 0, 2, 0, 1, 0}; + int yorig[] = {0, 0, 4, 0, 2, 0, 1}; + int xspc[] = {8, 8, 4, 4, 2, 2, 1}; + int yspc[] = {8, 8, 8, 4, 4, 2, 2}; + int i, j, x, y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p] - 1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p] - 1) / yspc[p]; + if (x && y) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, + y, depth, color)) { + STBI_FREE(final); + return 0; } - } - a->out = final; + for (j = 0; j < y; ++j) { + for (i = 0; i < x; ++i) { + int out_y = j * yspc[p] + yorig[p]; + int out_x = i * xspc[p] + xorig[p]; + memcpy(final + out_y * a->s->img_x * out_bytes + out_x * out_bytes, + a->out + (j * x + i) * out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; - return 1; + return 1; } -static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi_uc *p = z->out; +static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; - // compute color-based transparency, assuming we've - // already got 255 as the alpha value in the output - STBI_ASSERT(out_n == 2 || out_n == 4); + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); - if (out_n == 2) { - for (i=0; i < pixel_count; ++i) { - p[1] = (p[0] == tc[0] ? 0 : 255); - p += 2; - } - } else { - for (i=0; i < pixel_count; ++i) { - if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) - p[3] = 0; - p += 4; - } - } - return 1; + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 255); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) p[3] = 0; + p += 4; + } + } + return 1; } -static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], int out_n) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi__uint16 *p = (stbi__uint16*) z->out; +static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], + int out_n) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16 *p = (stbi__uint16 *)z->out; - // compute color-based transparency, assuming we've - // already got 65535 as the alpha value in the output - STBI_ASSERT(out_n == 2 || out_n == 4); + // compute color-based transparency, assuming we've + // already got 65535 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); - if (out_n == 2) { - for (i = 0; i < pixel_count; ++i) { - p[1] = (p[0] == tc[0] ? 0 : 65535); - p += 2; - } - } else { - for (i = 0; i < pixel_count; ++i) { - if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) - p[3] = 0; - p += 4; - } - } - return 1; + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 65535); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) p[3] = 0; + p += 4; + } + } + return 1; } -static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, int pal_img_n) -{ - stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; - stbi_uc *p, *temp_out, *orig = a->out; +static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, + int pal_img_n) { + stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; + stbi_uc *p, *temp_out, *orig = a->out; - p = (stbi_uc *) stbi__malloc_mad2(pixel_count, pal_img_n, 0); - if (p == NULL) return stbi__err("outofmem", "Out of memory"); + p = (stbi_uc *)stbi__malloc_mad2(pixel_count, pal_img_n, 0); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); - // between here and free(out) below, exitting would leak - temp_out = p; + // between here and free(out) below, exitting would leak + temp_out = p; - if (pal_img_n == 3) { - for (i=0; i < pixel_count; ++i) { - int n = orig[i]*4; - p[0] = palette[n ]; - p[1] = palette[n+1]; - p[2] = palette[n+2]; - p += 3; - } - } else { - for (i=0; i < pixel_count; ++i) { - int n = orig[i]*4; - p[0] = palette[n ]; - p[1] = palette[n+1]; - p[2] = palette[n+2]; - p[3] = palette[n+3]; - p += 4; - } - } - STBI_FREE(a->out); - a->out = temp_out; + if (pal_img_n == 3) { + for (i = 0; i < pixel_count; ++i) { + int n = orig[i] * 4; + p[0] = palette[n]; + p[1] = palette[n + 1]; + p[2] = palette[n + 2]; + p += 3; + } + } else { + for (i = 0; i < pixel_count; ++i) { + int n = orig[i] * 4; + p[0] = palette[n]; + p[1] = palette[n + 1]; + p[2] = palette[n + 2]; + p[3] = palette[n + 3]; + p += 4; + } + } + STBI_FREE(a->out); + a->out = temp_out; - STBI_NOTUSED(len); + STBI_NOTUSED(len); - return 1; + return 1; } static int stbi__unpremultiply_on_load = 0; static int stbi__de_iphone_flag = 0; -STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) -{ - stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; +STBIDEF void stbi_set_unpremultiply_on_load( + int flag_true_if_should_unpremultiply) { + stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; } -STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) -{ - stbi__de_iphone_flag = flag_true_if_should_convert; +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) { + stbi__de_iphone_flag = flag_true_if_should_convert; } -static void stbi__de_iphone(stbi__png *z) -{ - stbi__context *s = z->s; - stbi__uint32 i, pixel_count = s->img_x * s->img_y; - stbi_uc *p = z->out; +static void stbi__de_iphone(stbi__png *z) { + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; - if (s->img_out_n == 3) { // convert bgr to rgb - for (i=0; i < pixel_count; ++i) { - stbi_uc t = p[0]; - p[0] = p[2]; - p[2] = t; - p += 3; + if (s->img_out_n == 3) { // convert bgr to rgb + for (i = 0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } else { + STBI_ASSERT(s->img_out_n == 4); + if (stbi__unpremultiply_on_load) { + // convert bgr to rgb and unpremultiply + for (i = 0; i < pixel_count; ++i) { + stbi_uc a = p[3]; + stbi_uc t = p[0]; + if (a) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = (t * 255 + half) / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; } - } else { - STBI_ASSERT(s->img_out_n == 4); - if (stbi__unpremultiply_on_load) { - // convert bgr to rgb and unpremultiply - for (i=0; i < pixel_count; ++i) { - stbi_uc a = p[3]; - stbi_uc t = p[0]; - if (a) { - stbi_uc half = a / 2; - p[0] = (p[2] * 255 + half) / a; - p[1] = (p[1] * 255 + half) / a; - p[2] = ( t * 255 + half) / a; - } else { - p[0] = p[2]; - p[2] = t; - } - p += 4; - } - } else { - // convert bgr to rgb - for (i=0; i < pixel_count; ++i) { - stbi_uc t = p[0]; - p[0] = p[2]; - p[2] = t; - p += 4; - } + } else { + // convert bgr to rgb + for (i = 0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; } - } + } + } } -#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + #define STBI__PNG_TYPE(a, b, c, d) \ + (((unsigned)(a) << 24) + ((unsigned)(b) << 16) + ((unsigned)(c) << 8) + \ + (unsigned)(d)) -static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) -{ - stbi_uc palette[1024], pal_img_n=0; - stbi_uc has_trans=0, tc[3]={0}; - stbi__uint16 tc16[3]; - stbi__uint32 ioff=0, idata_limit=0, i, pal_len=0; - int first=1,k,interlace=0, color=0, is_iphone=0; - stbi__context *s = z->s; +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) { + stbi_uc palette[1024], pal_img_n = 0; + stbi_uc has_trans = 0, tc[3] = {0}; + stbi__uint16 tc16[3]; + stbi__uint32 ioff = 0, idata_limit = 0, i, pal_len = 0; + int first = 1, k, interlace = 0, color = 0, is_iphone = 0; + stbi__context *s = z->s; - z->expanded = NULL; - z->idata = NULL; - z->out = NULL; + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; - if (!stbi__check_png_header(s)) return 0; + if (!stbi__check_png_header(s)) return 0; - if (scan == STBI__SCAN_type) return 1; + if (scan == STBI__SCAN_type) return 1; - for (;;) { - stbi__pngchunk c = stbi__get_chunk_header(s); - switch (c.type) { - case STBI__PNG_TYPE('C','g','B','I'): - is_iphone = 1; - stbi__skip(s, c.length); - break; - case STBI__PNG_TYPE('I','H','D','R'): { - int comp,filter; - if (!first) return stbi__err("multiple IHDR","Corrupt PNG"); - first = 0; - if (c.length != 13) return stbi__err("bad IHDR len","Corrupt PNG"); - s->img_x = stbi__get32be(s); - s->img_y = stbi__get32be(s); - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only","PNG not supported: 1/2/4/8/16-bit only"); - color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype","Corrupt PNG"); - if (color == 3 && z->depth == 16) return stbi__err("bad ctype","Corrupt PNG"); - if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype","Corrupt PNG"); - comp = stbi__get8(s); if (comp) return stbi__err("bad comp method","Corrupt PNG"); - filter= stbi__get8(s); if (filter) return stbi__err("bad filter method","Corrupt PNG"); - interlace = stbi__get8(s); if (interlace>1) return stbi__err("bad interlace method","Corrupt PNG"); - if (!s->img_x || !s->img_y) return stbi__err("0-pixel image","Corrupt PNG"); - if (!pal_img_n) { - s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); - if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); - if (scan == STBI__SCAN_header) return 1; - } else { - // if paletted, then pal_n is our final components, and - // img_n is # components to decompress/filter. - s->img_n = 1; - if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); - // if SCAN_header, have to scan to see if we have a tRNS - } - break; - } + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C', 'g', 'B', 'I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I', 'H', 'D', 'R'): { + int comp, filter; + if (!first) return stbi__err("multiple IHDR", "Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len", "Corrupt PNG"); + s->img_x = stbi__get32be(s); + s->img_y = stbi__get32be(s); + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + z->depth = stbi__get8(s); + if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && + z->depth != 16) + return stbi__err("1/2/4/8/16-bit only", + "PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); + if (color > 6) return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3 && z->depth == 16) + return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3) + pal_img_n = 3; + else if (color & 1) + return stbi__err("bad ctype", "Corrupt PNG"); + comp = stbi__get8(s); + if (comp) return stbi__err("bad comp method", "Corrupt PNG"); + filter = stbi__get8(s); + if (filter) return stbi__err("bad filter method", "Corrupt PNG"); + interlace = stbi__get8(s); + if (interlace > 1) + return stbi__err("bad interlace method", "Corrupt PNG"); + if (!s->img_x || !s->img_y) + return stbi__err("0-pixel image", "Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) + return stbi__err("too large", "Image too large to decode"); + if (scan == STBI__SCAN_header) return 1; + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) + return stbi__err("too large", "Corrupt PNG"); + // if SCAN_header, have to scan to see if we have a tRNS + } + break; + } - case STBI__PNG_TYPE('P','L','T','E'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (c.length > 256*3) return stbi__err("invalid PLTE","Corrupt PNG"); - pal_len = c.length / 3; - if (pal_len * 3 != c.length) return stbi__err("invalid PLTE","Corrupt PNG"); - for (i=0; i < pal_len; ++i) { - palette[i*4+0] = stbi__get8(s); - palette[i*4+1] = stbi__get8(s); - palette[i*4+2] = stbi__get8(s); - palette[i*4+3] = 255; - } - break; - } + case STBI__PNG_TYPE('P', 'L', 'T', 'E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256 * 3) return stbi__err("invalid PLTE", "Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) + return stbi__err("invalid PLTE", "Corrupt PNG"); + for (i = 0; i < pal_len; ++i) { + palette[i * 4 + 0] = stbi__get8(s); + palette[i * 4 + 1] = stbi__get8(s); + palette[i * 4 + 2] = stbi__get8(s); + palette[i * 4 + 3] = 255; + } + break; + } - case STBI__PNG_TYPE('t','R','N','S'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (z->idata) return stbi__err("tRNS after IDAT","Corrupt PNG"); - if (pal_img_n) { - if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } - if (pal_len == 0) return stbi__err("tRNS before PLTE","Corrupt PNG"); - if (c.length > pal_len) return stbi__err("bad tRNS len","Corrupt PNG"); - pal_img_n = 4; - for (i=0; i < c.length; ++i) - palette[i*4+3] = stbi__get8(s); - } else { - if (!(s->img_n & 1)) return stbi__err("tRNS with alpha","Corrupt PNG"); - if (c.length != (stbi__uint32) s->img_n*2) return stbi__err("bad tRNS len","Corrupt PNG"); - has_trans = 1; - if (z->depth == 16) { - for (k = 0; k < s->img_n; ++k) tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is - } else { - for (k = 0; k < s->img_n; ++k) tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger - } - } - break; - } - - case STBI__PNG_TYPE('I','D','A','T'): { - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); - if (scan == STBI__SCAN_header) { s->img_n = pal_img_n; return 1; } - if ((int)(ioff + c.length) < (int)ioff) return 0; - if (ioff + c.length > idata_limit) { - stbi__uint32 idata_limit_old = idata_limit; - stbi_uc *p; - if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; - while (ioff + c.length > idata_limit) - idata_limit *= 2; - STBI_NOTUSED(idata_limit_old); - p = (stbi_uc *) STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); - z->idata = p; - } - if (!stbi__getn(s, z->idata+ioff,c.length)) return stbi__err("outofdata","Corrupt PNG"); - ioff += c.length; - break; - } - - case STBI__PNG_TYPE('I','E','N','D'): { - stbi__uint32 raw_len, bpl; - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if (scan != STBI__SCAN_load) return 1; - if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); - // initial guess for decoded data size to avoid unnecessary reallocs - bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component - raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; - z->expanded = (stbi_uc *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, raw_len, (int *) &raw_len, !is_iphone); - if (z->expanded == NULL) return 0; // zlib should set error - STBI_FREE(z->idata); z->idata = NULL; - if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) - s->img_out_n = s->img_n+1; - else - s->img_out_n = s->img_n; - if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, z->depth, color, interlace)) return 0; - if (has_trans) { - if (z->depth == 16) { - if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; - } else { - if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; - } - } - if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) - stbi__de_iphone(z); - if (pal_img_n) { - // pal_img_n == 3 or 4 - s->img_n = pal_img_n; // record the actual colors we had - s->img_out_n = pal_img_n; - if (req_comp >= 3) s->img_out_n = req_comp; - if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) - return 0; - } else if (has_trans) { - // non-paletted image with tRNS -> source image has (constant) alpha - ++s->img_n; - } - STBI_FREE(z->expanded); z->expanded = NULL; - // end of PNG chunk, read and skip CRC - stbi__get32be(s); + case STBI__PNG_TYPE('t', 'R', 'N', 'S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT", "Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { + s->img_n = 4; return 1; - } - - default: - // if critical, fail - if (first) return stbi__err("first not IHDR", "Corrupt PNG"); - if ((c.type & (1 << 29)) == 0) { - #ifndef STBI_NO_FAILURE_STRINGS - // not threadsafe - static char invalid_chunk[] = "XXXX PNG chunk not known"; - invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); - invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); - invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); - invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); - #endif - return stbi__err(invalid_chunk, "PNG not supported: unknown PNG chunk type"); - } - stbi__skip(s, c.length); - break; + } + if (pal_len == 0) return stbi__err("tRNS before PLTE", "Corrupt PNG"); + if (c.length > pal_len) + return stbi__err("bad tRNS len", "Corrupt PNG"); + pal_img_n = 4; + for (i = 0; i < c.length; ++i) + palette[i * 4 + 3] = stbi__get8(s); + } else { + if (!(s->img_n & 1)) + return stbi__err("tRNS with alpha", "Corrupt PNG"); + if (c.length != (stbi__uint32)s->img_n * 2) + return stbi__err("bad tRNS len", "Corrupt PNG"); + has_trans = 1; + if (z->depth == 16) { + for (k = 0; k < s->img_n; ++k) + tc16[k] = + (stbi__uint16)stbi__get16be(s); // copy the values as-is + } else { + for (k = 0; k < s->img_n; ++k) + tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * + stbi__depth_scale_table[z->depth]; // non 8-bit images + // will be larger + } + } + break; } - // end of PNG chunk, read and skip CRC - stbi__get32be(s); - } + + case STBI__PNG_TYPE('I', 'D', 'A', 'T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE", "Corrupt PNG"); + if (scan == STBI__SCAN_header) { + s->img_n = pal_img_n; + return 1; + } + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = idata_limit; + stbi_uc *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc *)STBI_REALLOC_SIZED(z->idata, idata_limit_old, + idata_limit); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata + ioff, c.length)) + return stbi__err("outofdata", "Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I', 'E', 'N', 'D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT", "Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc *)stbi_zlib_decode_malloc_guesssize_headerflag( + (char *)z->idata, ioff, raw_len, (int *)&raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_FREE(z->idata); + z->idata = NULL; + if ((req_comp == s->img_n + 1 && req_comp != 3 && !pal_img_n) || + has_trans) + s->img_out_n = s->img_n + 1; + else + s->img_out_n = s->img_n; + if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, + z->depth, color, interlace)) + return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } else { + if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; + } + } + if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) + stbi__de_iphone(z); + if (pal_img_n) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_FREE(z->expanded); + z->expanded = NULL; + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + return 1; + } + + default: + // if critical, fail + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + static char invalid_chunk[] = "XXXX PNG chunk not known"; + invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); + invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); + invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); + invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); + #endif + return stbi__err(invalid_chunk, + "PNG not supported: unknown PNG chunk type"); + } + stbi__skip(s, c.length); + break; + } + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + } } -static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, stbi__result_info *ri) -{ - void *result=NULL; - if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); - if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { - if (p->depth <= 8) - ri->bits_per_channel = 8; - else if (p->depth == 16) - ri->bits_per_channel = 16; +static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, + stbi__result_info *ri) { + void *result = NULL; + if (req_comp < 0 || req_comp > 4) + return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth <= 8) + ri->bits_per_channel = 8; + else if (p->depth == 16) + ri->bits_per_channel = 16; + else + return stbi__errpuc("bad bits_per_channel", + "PNG not supported: unsupported color depth"); + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char *)result, p->s->img_out_n, + req_comp, p->s->img_x, p->s->img_y); else - return stbi__errpuc("bad bits_per_channel", "PNG not supported: unsupported color depth"); - result = p->out; - p->out = NULL; - if (req_comp && req_comp != p->s->img_out_n) { - if (ri->bits_per_channel == 8) - result = stbi__convert_format((unsigned char *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); - else - result = stbi__convert_format16((stbi__uint16 *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); - p->s->img_out_n = req_comp; - if (result == NULL) return result; - } - *x = p->s->img_x; - *y = p->s->img_y; - if (n) *n = p->s->img_n; - } - STBI_FREE(p->out); p->out = NULL; - STBI_FREE(p->expanded); p->expanded = NULL; - STBI_FREE(p->idata); p->idata = NULL; + result = stbi__convert_format16((stbi__uint16 *)result, p->s->img_out_n, + req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n) *n = p->s->img_n; + } + STBI_FREE(p->out); + p->out = NULL; + STBI_FREE(p->expanded); + p->expanded = NULL; + STBI_FREE(p->idata); + p->idata = NULL; - return result; + return result; } -static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi__png p; - p.s = s; - return stbi__do_png(&p, x,y,comp,req_comp, ri); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi__png p; + p.s = s; + return stbi__do_png(&p, x, y, comp, req_comp, ri); } -static int stbi__png_test(stbi__context *s) -{ - int r; - r = stbi__check_png_header(s); - stbi__rewind(s); - return r; +static int stbi__png_test(stbi__context *s) { + int r; + r = stbi__check_png_header(s); + stbi__rewind(s); + return r; } -static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) -{ - if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { - stbi__rewind( p->s ); - return 0; - } - if (x) *x = p->s->img_x; - if (y) *y = p->s->img_y; - if (comp) *comp = p->s->img_n; - return 1; +static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) { + if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { + stbi__rewind(p->s); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; } -static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) -{ - stbi__png p; - p.s = s; - return stbi__png_info_raw(&p, x, y, comp); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) { + stbi__png p; + p.s = s; + return stbi__png_info_raw(&p, x, y, comp); } static int stbi__png_is16(stbi__context *s) @@ -5191,958 +5660,1015 @@ static int stbi__png_is16(stbi__context *s) } return 1; } -#endif + #endif // Microsoft/Windows BMP image -#ifndef STBI_NO_BMP -static int stbi__bmp_test_raw(stbi__context *s) -{ - int r; - int sz; - if (stbi__get8(s) != 'B') return 0; - if (stbi__get8(s) != 'M') return 0; - stbi__get32le(s); // discard filesize - stbi__get16le(s); // discard reserved - stbi__get16le(s); // discard reserved - stbi__get32le(s); // discard data offset - sz = stbi__get32le(s); - r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); - return r; + #ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context *s) { + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + stbi__get32le(s); // discard data offset + sz = stbi__get32le(s); + r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); + return r; } -static int stbi__bmp_test(stbi__context *s) -{ - int r = stbi__bmp_test_raw(s); - stbi__rewind(s); - return r; +static int stbi__bmp_test(stbi__context *s) { + int r = stbi__bmp_test_raw(s); + stbi__rewind(s); + return r; } - // returns 0..31 for the highest set bit -static int stbi__high_bit(unsigned int z) -{ - int n=0; - if (z == 0) return -1; - if (z >= 0x10000) { n += 16; z >>= 16; } - if (z >= 0x00100) { n += 8; z >>= 8; } - if (z >= 0x00010) { n += 4; z >>= 4; } - if (z >= 0x00004) { n += 2; z >>= 2; } - if (z >= 0x00002) { n += 1;/* >>= 1;*/ } - return n; +static int stbi__high_bit(unsigned int z) { + int n = 0; + if (z == 0) return -1; + if (z >= 0x10000) { + n += 16; + z >>= 16; + } + if (z >= 0x00100) { + n += 8; + z >>= 8; + } + if (z >= 0x00010) { + n += 4; + z >>= 4; + } + if (z >= 0x00004) { + n += 2; + z >>= 2; + } + if (z >= 0x00002) { n += 1; /* >>= 1;*/ } + return n; } -static int stbi__bitcount(unsigned int a) -{ - a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 - a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 - a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits - a = (a + (a >> 8)); // max 16 per 8 bits - a = (a + (a >> 16)); // max 32 per 8 bits - return a & 0xff; +static int stbi__bitcount(unsigned int a) { + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; } // extract an arbitrarily-aligned N-bit value (N=bits) // from v, and then make it 8-bits long and fractionally // extend it to full full range. -static int stbi__shiftsigned(unsigned int v, int shift, int bits) -{ - static unsigned int mul_table[9] = { +static int stbi__shiftsigned(unsigned int v, int shift, int bits) { + static unsigned int mul_table[9] = { 0, - 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, - 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, - }; - static unsigned int shift_table[9] = { - 0, 0,0,1,0,2,4,6,0, - }; - if (shift < 0) - v <<= -shift; - else - v >>= shift; - STBI_ASSERT(v < 256); - v >>= (8-bits); - STBI_ASSERT(bits >= 0 && bits <= 8); - return (int) ((unsigned) v * mul_table[bits]) >> shift_table[bits]; + 0xff /*0b11111111*/, + 0x55 /*0b01010101*/, + 0x49 /*0b01001001*/, + 0x11 /*0b00010001*/, + 0x21 /*0b00100001*/, + 0x41 /*0b01000001*/, + 0x81 /*0b10000001*/, + 0x01 /*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0, 0, 1, 0, 2, 4, 6, 0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v < 256); + v >>= (8 - bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int)((unsigned)v * mul_table[bits]) >> shift_table[bits]; } -typedef struct -{ - int bpp, offset, hsz; - unsigned int mr,mg,mb,ma, all_a; - int extra_read; +typedef struct { + int bpp, offset, hsz; + unsigned int mr, mg, mb, ma, all_a; + int extra_read; } stbi__bmp_data; -static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) -{ - int hsz; - if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') return stbi__errpuc("not BMP", "Corrupt BMP"); - stbi__get32le(s); // discard filesize - stbi__get16le(s); // discard reserved - stbi__get16le(s); // discard reserved - info->offset = stbi__get32le(s); - info->hsz = hsz = stbi__get32le(s); - info->mr = info->mg = info->mb = info->ma = 0; - info->extra_read = 14; +static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) { + int hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') + return stbi__errpuc("not BMP", "Corrupt BMP"); + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + info->extra_read = 14; - if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); + if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); - if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); - if (hsz == 12) { - s->img_x = stbi__get16le(s); - s->img_y = stbi__get16le(s); - } else { - s->img_x = stbi__get32le(s); - s->img_y = stbi__get32le(s); - } - if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); - info->bpp = stbi__get16le(s); - if (hsz != 12) { - int compress = stbi__get32le(s); - if (compress == 1 || compress == 2) return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); - stbi__get32le(s); // discard sizeof - stbi__get32le(s); // discard hres - stbi__get32le(s); // discard vres - stbi__get32le(s); // discard colorsused - stbi__get32le(s); // discard max important - if (hsz == 40 || hsz == 56) { - if (hsz == 56) { - stbi__get32le(s); - stbi__get32le(s); - stbi__get32le(s); - stbi__get32le(s); - } - if (info->bpp == 16 || info->bpp == 32) { - if (compress == 0) { - if (info->bpp == 32) { - info->mr = 0xffu << 16; - info->mg = 0xffu << 8; - info->mb = 0xffu << 0; - info->ma = 0xffu << 24; - info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 - } else { - info->mr = 31u << 10; - info->mg = 31u << 5; - info->mb = 31u << 0; - } - } else if (compress == 3) { - info->mr = stbi__get32le(s); - info->mg = stbi__get32le(s); - info->mb = stbi__get32le(s); - info->extra_read += 12; - // not documented, but generated by photoshop and handled by mspaint - if (info->mr == info->mg && info->mg == info->mb) { - // ?!?!? - return stbi__errpuc("bad BMP", "bad BMP"); - } - } else - return stbi__errpuc("bad BMP", "bad BMP"); - } - } else { - int i; - if (hsz != 108 && hsz != 124) + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) + return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = stbi__get16le(s); + s->img_y = stbi__get16le(s); + } else { + s->img_x = stbi__get32le(s); + s->img_y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int compress = stbi__get32le(s); + if (compress == 1 || compress == 2) + return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); + stbi__get32le(s); // discard sizeof + stbi__get32le(s); // discard hres + stbi__get32le(s); // discard vres + stbi__get32le(s); // discard colorsused + stbi__get32le(s); // discard max important + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + } + if (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha + // channel but it was all 0 + } else { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } + } else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->extra_read += 12; + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? return stbi__errpuc("bad BMP", "bad BMP"); - info->mr = stbi__get32le(s); - info->mg = stbi__get32le(s); - info->mb = stbi__get32le(s); - info->ma = stbi__get32le(s); - stbi__get32le(s); // discard color space - for (i=0; i < 12; ++i) - stbi__get32le(s); // discard color space parameters - if (hsz == 124) { - stbi__get32le(s); // discard rendering intent - stbi__get32le(s); // discard offset of profile data - stbi__get32le(s); // discard size of profile data - stbi__get32le(s); // discard reserved - } + } + } else + return stbi__errpuc("bad BMP", "bad BMP"); } - } - return (void *) 1; -} - - -static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *out; - unsigned int mr=0,mg=0,mb=0,ma=0, all_a; - stbi_uc pal[256][4]; - int psize=0,i,j,width; - int flip_vertically, pad, target; - stbi__bmp_data info; - STBI_NOTUSED(ri); - - info.all_a = 255; - if (stbi__bmp_parse_header(s, &info) == NULL) - return NULL; // error code already set - - flip_vertically = ((int) s->img_y) > 0; - s->img_y = abs((int) s->img_y); - - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - - mr = info.mr; - mg = info.mg; - mb = info.mb; - ma = info.ma; - all_a = info.all_a; - - if (info.hsz == 12) { - if (info.bpp < 24) - psize = (info.offset - info.extra_read - 24) / 3; - } else { - if (info.bpp < 16) - psize = (info.offset - info.extra_read - info.hsz) >> 2; - } - if (psize == 0) { - STBI_ASSERT(info.offset == s->callback_already_read + (int) (s->img_buffer - s->img_buffer_original)); - if (info.offset != s->callback_already_read + (s->img_buffer - s->buffer_start)) { - return stbi__errpuc("bad offset", "Corrupt BMP"); - } - } - - if (info.bpp == 24 && ma == 0xff000000) - s->img_n = 3; - else - s->img_n = ma ? 4 : 3; - if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 - target = req_comp; - else - target = s->img_n; // if they want monochrome, we'll post-convert - - // sanity-check size - if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) - return stbi__errpuc("too large", "Corrupt BMP"); - - out = (stbi_uc *) stbi__malloc_mad3(target, s->img_x, s->img_y, 0); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - if (info.bpp < 16) { - int z=0; - if (psize == 0 || psize > 256) { STBI_FREE(out); return stbi__errpuc("invalid", "Corrupt BMP"); } - for (i=0; i < psize; ++i) { - pal[i][2] = stbi__get8(s); - pal[i][1] = stbi__get8(s); - pal[i][0] = stbi__get8(s); - if (info.hsz != 12) stbi__get8(s); - pal[i][3] = 255; - } - stbi__skip(s, info.offset - info.extra_read - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); - if (info.bpp == 1) width = (s->img_x + 7) >> 3; - else if (info.bpp == 4) width = (s->img_x + 1) >> 1; - else if (info.bpp == 8) width = s->img_x; - else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } - pad = (-width)&3; - if (info.bpp == 1) { - for (j=0; j < (int) s->img_y; ++j) { - int bit_offset = 7, v = stbi__get8(s); - for (i=0; i < (int) s->img_x; ++i) { - int color = (v>>bit_offset)&0x1; - out[z++] = pal[color][0]; - out[z++] = pal[color][1]; - out[z++] = pal[color][2]; - if (target == 4) out[z++] = 255; - if (i+1 == (int) s->img_x) break; - if((--bit_offset) < 0) { - bit_offset = 7; - v = stbi__get8(s); - } - } - stbi__skip(s, pad); - } - } else { - for (j=0; j < (int) s->img_y; ++j) { - for (i=0; i < (int) s->img_x; i += 2) { - int v=stbi__get8(s),v2=0; - if (info.bpp == 4) { - v2 = v & 15; - v >>= 4; - } - out[z++] = pal[v][0]; - out[z++] = pal[v][1]; - out[z++] = pal[v][2]; - if (target == 4) out[z++] = 255; - if (i+1 == (int) s->img_x) break; - v = (info.bpp == 8) ? stbi__get8(s) : v2; - out[z++] = pal[v][0]; - out[z++] = pal[v][1]; - out[z++] = pal[v][2]; - if (target == 4) out[z++] = 255; - } - stbi__skip(s, pad); - } - } - } else { - int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0; - int z = 0; - int easy=0; - stbi__skip(s, info.offset - info.extra_read - info.hsz); - if (info.bpp == 24) width = 3 * s->img_x; - else if (info.bpp == 16) width = 2*s->img_x; - else /* bpp = 32 and pad = 0 */ width=0; - pad = (-width) & 3; - if (info.bpp == 24) { - easy = 1; - } else if (info.bpp == 32) { - if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) - easy = 2; - } - if (!easy) { - if (!mr || !mg || !mb) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } - // right shift amt to put high bit in position #7 - rshift = stbi__high_bit(mr)-7; rcount = stbi__bitcount(mr); - gshift = stbi__high_bit(mg)-7; gcount = stbi__bitcount(mg); - bshift = stbi__high_bit(mb)-7; bcount = stbi__bitcount(mb); - ashift = stbi__high_bit(ma)-7; acount = stbi__bitcount(ma); - if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } - } - for (j=0; j < (int) s->img_y; ++j) { - if (easy) { - for (i=0; i < (int) s->img_x; ++i) { - unsigned char a; - out[z+2] = stbi__get8(s); - out[z+1] = stbi__get8(s); - out[z+0] = stbi__get8(s); - z += 3; - a = (easy == 2 ? stbi__get8(s) : 255); - all_a |= a; - if (target == 4) out[z++] = a; - } - } else { - int bpp = info.bpp; - for (i=0; i < (int) s->img_x; ++i) { - stbi__uint32 v = (bpp == 16 ? (stbi__uint32) stbi__get16le(s) : stbi__get32le(s)); - unsigned int a; - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); - out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); - a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); - all_a |= a; - if (target == 4) out[z++] = STBI__BYTECAST(a); - } - } - stbi__skip(s, pad); - } - } - - // if alpha channel is all 0s, replace with all 255s - if (target == 4 && all_a == 0) - for (i=4*s->img_x*s->img_y-1; i >= 0; i -= 4) - out[i] = 255; - - if (flip_vertically) { - stbi_uc t; - for (j=0; j < (int) s->img_y>>1; ++j) { - stbi_uc *p1 = out + j *s->img_x*target; - stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; - for (i=0; i < (int) s->img_x*target; ++i) { - t = p1[i]; p1[i] = p2[i]; p2[i] = t; - } - } - } - - if (req_comp && req_comp != target) { - out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } - - *x = s->img_x; - *y = s->img_y; - if (comp) *comp = s->img_n; - return out; -} -#endif - -// Targa Truevision - TGA -// by Jonathan Dummer -#ifndef STBI_NO_TGA -// returns STBI_rgb or whatever, 0 on error -static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) -{ - // only RGB or RGBA (incl. 16bit) or grey allowed - if (is_rgb16) *is_rgb16 = 0; - switch(bits_per_pixel) { - case 8: return STBI_grey; - case 16: if(is_grey) return STBI_grey_alpha; - // fallthrough - case 15: if(is_rgb16) *is_rgb16 = 1; - return STBI_rgb; - case 24: // fallthrough - case 32: return bits_per_pixel/8; - default: return 0; - } -} - -static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) -{ - int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; - int sz, tga_colormap_type; - stbi__get8(s); // discard Offset - tga_colormap_type = stbi__get8(s); // colormap type - if( tga_colormap_type > 1 ) { - stbi__rewind(s); - return 0; // only RGB or indexed allowed - } - tga_image_type = stbi__get8(s); // image type - if ( tga_colormap_type == 1 ) { // colormapped (paletted) image - if (tga_image_type != 1 && tga_image_type != 9) { - stbi__rewind(s); - return 0; - } - stbi__skip(s,4); // skip index of first colormap entry and number of entries - sz = stbi__get8(s); // check bits per palette color entry - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) { - stbi__rewind(s); - return 0; - } - stbi__skip(s,4); // skip image x and y origin - tga_colormap_bpp = sz; - } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE - if ( (tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11) ) { - stbi__rewind(s); - return 0; // only RGB or grey allowed, +/- RLE - } - stbi__skip(s,9); // skip colormap specification and image x/y origin - tga_colormap_bpp = 0; - } - tga_w = stbi__get16le(s); - if( tga_w < 1 ) { - stbi__rewind(s); - return 0; // test width - } - tga_h = stbi__get16le(s); - if( tga_h < 1 ) { - stbi__rewind(s); - return 0; // test height - } - tga_bits_per_pixel = stbi__get8(s); // bits per pixel - stbi__get8(s); // ignore alpha bits - if (tga_colormap_bpp != 0) { - if((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { - // when using a colormap, tga_bits_per_pixel is the size of the indexes - // I don't think anything but 8 or 16bit indexes makes sense - stbi__rewind(s); - return 0; - } - tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); } else { - tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + int i; + if (hsz != 108 && hsz != 124) return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->ma = stbi__get32le(s); + stbi__get32le(s); // discard color space + for (i = 0; i < 12; ++i) + stbi__get32le(s); // discard color space parameters + if (hsz == 124) { + stbi__get32le(s); // discard rendering intent + stbi__get32le(s); // discard offset of profile data + stbi__get32le(s); // discard size of profile data + stbi__get32le(s); // discard reserved + } } - if(!tga_comp) { + } + return (void *)1; +} + +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *out; + unsigned int mr = 0, mg = 0, mb = 0, ma = 0, all_a; + stbi_uc pal[256][4]; + int psize = 0, i, j, width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int)s->img_y) > 0; + s->img_y = abs((int)s->img_y); + + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) psize = (info.offset - info.extra_read - 24) / 3; + } else { + if (info.bpp < 16) psize = (info.offset - info.extra_read - info.hsz) >> 2; + } + if (psize == 0) { + STBI_ASSERT(info.offset == + s->callback_already_read + + (int)(s->img_buffer - s->img_buffer_original)); + if (info.offset != + s->callback_already_read + (s->img_buffer - s->buffer_start)) { + return stbi__errpuc("bad offset", "Corrupt BMP"); + } + } + + if (info.bpp == 24 && ma == 0xff000000) + s->img_n = 3; + else + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc *)stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z = 0; + if (psize == 0 || psize > 256) { + STBI_FREE(out); + return stbi__errpuc("invalid", "Corrupt BMP"); + } + for (i = 0; i < psize; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + if (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - info.extra_read - info.hsz - + psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) + width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) + width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) + width = s->img_x; + else { + STBI_FREE(out); + return stbi__errpuc("bad bpp", "Corrupt BMP"); + } + pad = (-width) & 3; + if (info.bpp == 1) { + for (j = 0; j < (int)s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i = 0; i < (int)s->img_x; ++i) { + int color = (v >> bit_offset) & 0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if (target == 4) out[z++] = 255; + if (i + 1 == (int)s->img_x) break; + if ((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } else { + for (j = 0; j < (int)s->img_y; ++j) { + for (i = 0; i < (int)s->img_x; i += 2) { + int v = stbi__get8(s), v2 = 0; + if (info.bpp == 4) { + v2 = v & 15; + v >>= 4; + } + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + if (i + 1 == (int)s->img_x) break; + v = (info.bpp == 8) ? stbi__get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + stbi__skip(s, pad); + } + } + } else { + int rshift = 0, gshift = 0, bshift = 0, ashift = 0, rcount = 0, gcount = 0, + bcount = 0, acount = 0; + int z = 0; + int easy = 0; + stbi__skip(s, info.offset - info.extra_read - info.hsz); + if (info.bpp == 24) + width = 3 * s->img_x; + else if (info.bpp == 16) + width = 2 * s->img_x; + else /* bpp = 32 and pad = 0 */ + width = 0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { + STBI_FREE(out); + return stbi__errpuc("bad masks", "Corrupt BMP"); + } + // right shift amt to put high bit in position #7 + rshift = stbi__high_bit(mr) - 7; + rcount = stbi__bitcount(mr); + gshift = stbi__high_bit(mg) - 7; + gcount = stbi__bitcount(mg); + bshift = stbi__high_bit(mb) - 7; + bcount = stbi__bitcount(mb); + ashift = stbi__high_bit(ma) - 7; + acount = stbi__bitcount(ma); + if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { + STBI_FREE(out); + return stbi__errpuc("bad masks", "Corrupt BMP"); + } + } + for (j = 0; j < (int)s->img_y; ++j) { + if (easy) { + for (i = 0; i < (int)s->img_x; ++i) { + unsigned char a; + out[z + 2] = stbi__get8(s); + out[z + 1] = stbi__get8(s); + out[z + 0] = stbi__get8(s); + z += 3; + a = (easy == 2 ? stbi__get8(s) : 255); + all_a |= a; + if (target == 4) out[z++] = a; + } + } else { + int bpp = info.bpp; + for (i = 0; i < (int)s->img_x; ++i) { + stbi__uint32 v = + (bpp == 16 ? (stbi__uint32)stbi__get16le(s) : stbi__get32le(s)); + unsigned int a; + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); + a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i = 4 * s->img_x * s->img_y - 1; i >= 0; i -= 4) + out[i] = 255; + + if (flip_vertically) { + stbi_uc t; + for (j = 0; j < (int)s->img_y >> 1; ++j) { + stbi_uc *p1 = out + j * s->img_x * target; + stbi_uc *p2 = out + (s->img_y - 1 - j) * s->img_x * target; + for (i = 0; i < (int)s->img_x * target; ++i) { + t = p1[i]; + p1[i] = p2[i]; + p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} + #endif + + // Targa Truevision - TGA + // by Jonathan Dummer + #ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int *is_rgb16) { + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16) *is_rgb16 = 0; + switch (bits_per_pixel) { + case 8: + return STBI_grey; + case 16: + if (is_grey) return STBI_grey_alpha; + // fallthrough + case 15: + if (is_rgb16) *is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: + return bits_per_pixel / 8; + default: + return 0; + } +} + +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) { + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, + tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if (tga_colormap_type > 1) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if (tga_colormap_type == 1) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { stbi__rewind(s); return 0; } - if (x) *x = tga_w; - if (y) *y = tga_h; - if (comp) *comp = tga_comp; - return 1; // seems to have passed everything + stbi__skip(s, + 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 4); // skip image x and y origin + tga_colormap_bpp = sz; + } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ((tga_image_type != 2) && (tga_image_type != 3) && + (tga_image_type != 10) && (tga_image_type != 11)) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s, 9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + tga_w = stbi__get16le(s); + if (tga_w < 1) { + stbi__rewind(s); + return 0; // test width + } + tga_h = stbi__get16le(s); + if (tga_h < 1) { + stbi__rewind(s); + return 0; // test height + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } else { + tga_comp = stbi__tga_get_comp( + tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), + NULL); + } + if (!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp; + return 1; // seems to have passed everything } -static int stbi__tga_test(stbi__context *s) -{ - int res = 0; - int sz, tga_color_type; - stbi__get8(s); // discard Offset - tga_color_type = stbi__get8(s); // color type - if ( tga_color_type > 1 ) goto errorEnd; // only RGB or indexed allowed - sz = stbi__get8(s); // image type - if ( tga_color_type == 1 ) { // colormapped (paletted) image - if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 - stbi__skip(s,4); // skip index of first colormap entry and number of entries - sz = stbi__get8(s); // check bits per palette color entry - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; - stbi__skip(s,4); // skip image x and y origin - } else { // "normal" image w/o colormap - if ( (sz != 2) && (sz != 3) && (sz != 10) && (sz != 11) ) goto errorEnd; // only RGB or grey allowed, +/- RLE - stbi__skip(s,9); // skip colormap specification and image x/y origin - } - if ( stbi__get16le(s) < 1 ) goto errorEnd; // test width - if ( stbi__get16le(s) < 1 ) goto errorEnd; // test height - sz = stbi__get8(s); // bits per pixel - if ( (tga_color_type == 1) && (sz != 8) && (sz != 16) ) goto errorEnd; // for colormapped images, bpp is size of an index - if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; +static int stbi__tga_test(stbi__context *s) { + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if (tga_color_type > 1) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if (tga_color_type == 1) { // colormapped (paletted) image + if (sz != 1 && sz != 9) + goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s, + 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) + goto errorEnd; + stbi__skip(s, 4); // skip image x and y origin + } else { // "normal" image w/o colormap + if ((sz != 2) && (sz != 3) && (sz != 10) && (sz != 11)) + goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s, 9); // skip colormap specification and image x/y origin + } + if (stbi__get16le(s) < 1) goto errorEnd; // test width + if (stbi__get16le(s) < 1) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ((tga_color_type == 1) && (sz != 8) && (sz != 16)) + goto errorEnd; // for colormapped images, bpp is size of an index + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) + goto errorEnd; - res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + res = 1; // if we got this far, everything's good and we can return 1 instead + // of 0 errorEnd: - stbi__rewind(s); - return res; + stbi__rewind(s); + return res; } // read 16bit value and convert to 24bit RGB -static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc* out) -{ - stbi__uint16 px = (stbi__uint16)stbi__get16le(s); - stbi__uint16 fiveBitMask = 31; - // we have 3 channels with 5bits each - int r = (px >> 10) & fiveBitMask; - int g = (px >> 5) & fiveBitMask; - int b = px & fiveBitMask; - // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later - out[0] = (stbi_uc)((r * 255)/31); - out[1] = (stbi_uc)((g * 255)/31); - out[2] = (stbi_uc)((b * 255)/31); +static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc *out) { + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be + // swapped later + out[0] = (stbi_uc)((r * 255) / 31); + out[1] = (stbi_uc)((g * 255) / 31); + out[2] = (stbi_uc)((b * 255) / 31); - // some people claim that the most significant bit might be used for alpha - // (possibly if an alpha-bit is set in the "image descriptor byte") - // but that only made 16bit test images completely translucent.. - // so let's treat all 15 and 16bit TGAs as RGB with no alpha. + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. } -static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - // read in the TGA header stuff - int tga_offset = stbi__get8(s); - int tga_indexed = stbi__get8(s); - int tga_image_type = stbi__get8(s); - int tga_is_RLE = 0; - int tga_palette_start = stbi__get16le(s); - int tga_palette_len = stbi__get16le(s); - int tga_palette_bits = stbi__get8(s); - int tga_x_origin = stbi__get16le(s); - int tga_y_origin = stbi__get16le(s); - int tga_width = stbi__get16le(s); - int tga_height = stbi__get16le(s); - int tga_bits_per_pixel = stbi__get8(s); - int tga_comp, tga_rgb16=0; - int tga_inverted = stbi__get8(s); - // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) - // image data - unsigned char *tga_data; - unsigned char *tga_palette = NULL; - int i, j; - unsigned char raw_data[4] = {0}; - int RLE_count = 0; - int RLE_repeating = 0; - int read_next_pixel = 1; - STBI_NOTUSED(ri); - STBI_NOTUSED(tga_x_origin); // @TODO - STBI_NOTUSED(tga_y_origin); // @TODO +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + // read in the TGA header stuff + int tga_offset = stbi__get8(s); + int tga_indexed = stbi__get8(s); + int tga_image_type = stbi__get8(s); + int tga_is_RLE = 0; + int tga_palette_start = stbi__get16le(s); + int tga_palette_len = stbi__get16le(s); + int tga_palette_bits = stbi__get8(s); + int tga_x_origin = stbi__get16le(s); + int tga_y_origin = stbi__get16le(s); + int tga_width = stbi__get16le(s); + int tga_height = stbi__get16le(s); + int tga_bits_per_pixel = stbi__get8(s); + int tga_comp, tga_rgb16 = 0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused + // (useless?) + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = {0}; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + STBI_NOTUSED(tga_x_origin); // @TODO + STBI_NOTUSED(tga_y_origin); // @TODO - if (tga_height > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (tga_width > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (tga_height > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (tga_width > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - // do a tiny bit of precessing - if ( tga_image_type >= 8 ) - { - tga_image_type -= 8; - tga_is_RLE = 1; - } - tga_inverted = 1 - ((tga_inverted >> 5) & 1); + // do a tiny bit of precessing + if (tga_image_type >= 8) { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); - // If I'm paletted, then I'll use the number of bits from the palette - if ( tga_indexed ) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); - else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + // If I'm paletted, then I'll use the number of bits from the palette + if (tga_indexed) + tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), + &tga_rgb16); - if(!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency - return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + if (!tga_comp) // shouldn't really happen, stbi__tga_test() should have + // ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); - // tga info - *x = tga_width; - *y = tga_height; - if (comp) *comp = tga_comp; + // tga info + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; - if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) - return stbi__errpuc("too large", "Corrupt TGA"); + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); - tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); - if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + tga_data = + (unsigned char *)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); - // skip to the data's starting position (offset usually = 0) - stbi__skip(s, tga_offset ); + // skip to the data's starting position (offset usually = 0) + stbi__skip(s, tga_offset); - if ( !tga_indexed && !tga_is_RLE && !tga_rgb16 ) { - for (i=0; i < tga_height; ++i) { - int row = tga_inverted ? tga_height -i - 1 : i; - stbi_uc *tga_row = tga_data + row*tga_width*tga_comp; - stbi__getn(s, tga_row, tga_width * tga_comp); + if (!tga_indexed && !tga_is_RLE && !tga_rgb16) { + for (i = 0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height - i - 1 : i; + stbi_uc *tga_row = tga_data + row * tga_width * tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } else { + // do I need to load a palette? + if (tga_indexed) { + if (tga_palette_len == 0) { /* you have to have at least one entry! */ + STBI_FREE(tga_data); + return stbi__errpuc("bad palette", "Corrupt TGA"); } - } else { - // do I need to load a palette? - if ( tga_indexed) - { - if (tga_palette_len == 0) { /* you have to have at least one entry! */ - STBI_FREE(tga_data); - return stbi__errpuc("bad palette", "Corrupt TGA"); - } - // any data to skip? (offset usually = 0) - stbi__skip(s, tga_palette_start ); - // load the palette - tga_palette = (unsigned char*)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); - if (!tga_palette) { - STBI_FREE(tga_data); - return stbi__errpuc("outofmem", "Out of memory"); - } - if (tga_rgb16) { - stbi_uc *pal_entry = tga_palette; - STBI_ASSERT(tga_comp == STBI_rgb); - for (i=0; i < tga_palette_len; ++i) { - stbi__tga_read_rgb16(s, pal_entry); - pal_entry += tga_comp; - } - } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { - STBI_FREE(tga_data); - STBI_FREE(tga_palette); - return stbi__errpuc("bad palette", "Corrupt TGA"); - } + // any data to skip? (offset usually = 0) + stbi__skip(s, tga_palette_start); + // load the palette + tga_palette = + (unsigned char *)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); } - // load the data - for (i=0; i < tga_width * tga_height; ++i) - { - // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? - if ( tga_is_RLE ) - { - if ( RLE_count == 0 ) - { - // yep, get the next byte as a RLE command - int RLE_cmd = stbi__get8(s); - RLE_count = 1 + (RLE_cmd & 127); - RLE_repeating = RLE_cmd >> 7; - read_next_pixel = 1; - } else if ( !RLE_repeating ) - { - read_next_pixel = 1; - } - } else - { - read_next_pixel = 1; - } - // OK, if I need to read a pixel, do it now - if ( read_next_pixel ) - { - // load however much data we did have - if ( tga_indexed ) - { - // read in index, then perform the lookup - int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); - if ( pal_idx >= tga_palette_len ) { - // invalid index - pal_idx = 0; - } - pal_idx *= tga_comp; - for (j = 0; j < tga_comp; ++j) { - raw_data[j] = tga_palette[pal_idx+j]; - } - } else if(tga_rgb16) { - STBI_ASSERT(tga_comp == STBI_rgb); - stbi__tga_read_rgb16(s, raw_data); - } else { - // read in the data raw - for (j = 0; j < tga_comp; ++j) { - raw_data[j] = stbi__get8(s); - } - } - // clear the reading flag for the next pixel - read_next_pixel = 0; - } // end of reading a pixel - - // copy data - for (j = 0; j < tga_comp; ++j) - tga_data[i*tga_comp+j] = raw_data[j]; - - // in case we're in RLE mode, keep counting down - --RLE_count; + if (tga_rgb16) { + stbi_uc *pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i = 0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_FREE(tga_palette); + return stbi__errpuc("bad palette", "Corrupt TGA"); } - // do I need to invert the image? - if ( tga_inverted ) - { - for (j = 0; j*2 < tga_height; ++j) - { - int index1 = j * tga_width * tga_comp; - int index2 = (tga_height - 1 - j) * tga_width * tga_comp; - for (i = tga_width * tga_comp; i > 0; --i) - { - unsigned char temp = tga_data[index1]; - tga_data[index1] = tga_data[index2]; - tga_data[index2] = temp; - ++index1; - ++index2; - } - } + } + // load the data + for (i = 0; i < tga_width * tga_height; ++i) { + // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? + if (tga_is_RLE) { + if (RLE_count == 0) { + // yep, get the next byte as a RLE command + int RLE_cmd = stbi__get8(s); + RLE_count = 1 + (RLE_cmd & 127); + RLE_repeating = RLE_cmd >> 7; + read_next_pixel = 1; + } else if (!RLE_repeating) { + read_next_pixel = 1; + } + } else { + read_next_pixel = 1; } - // clear my palette, if I had one - if ( tga_palette != NULL ) - { - STBI_FREE( tga_palette ); + // OK, if I need to read a pixel, do it now + if (read_next_pixel) { + // load however much data we did have + if (tga_indexed) { + // read in index, then perform the lookup + int pal_idx = + (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if (pal_idx >= tga_palette_len) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx + j]; + } + } else if (tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } else { + // read in the data raw + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = stbi__get8(s); + } + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + + // copy data + for (j = 0; j < tga_comp; ++j) + tga_data[i * tga_comp + j] = raw_data[j]; + + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if (tga_inverted) { + for (j = 0; j * 2 < tga_height; ++j) { + int index1 = j * tga_width * tga_comp; + int index2 = (tga_height - 1 - j) * tga_width * tga_comp; + for (i = tga_width * tga_comp; i > 0; --i) { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } } - } + } + // clear my palette, if I had one + if (tga_palette != NULL) { STBI_FREE(tga_palette); } + } - // swap RGB - if the source data was RGB16, it already is in the right order - if (tga_comp >= 3 && !tga_rgb16) - { - unsigned char* tga_pixel = tga_data; - for (i=0; i < tga_width * tga_height; ++i) - { - unsigned char temp = tga_pixel[0]; - tga_pixel[0] = tga_pixel[2]; - tga_pixel[2] = temp; - tga_pixel += tga_comp; - } - } + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) { + unsigned char *tga_pixel = tga_data; + for (i = 0; i < tga_width * tga_height; ++i) { + unsigned char temp = tga_pixel[0]; + tga_pixel[0] = tga_pixel[2]; + tga_pixel[2] = temp; + tga_pixel += tga_comp; + } + } - // convert to target component count - if (req_comp && req_comp != tga_comp) - tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + // convert to target component count + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, + tga_height); - // the things I do to get rid of an error message, and yet keep - // Microsoft's C compilers happy... [8^( - tga_palette_start = tga_palette_len = tga_palette_bits = - tga_x_origin = tga_y_origin = 0; - STBI_NOTUSED(tga_palette_start); - // OK, done - return tga_data; + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = tga_x_origin = + tga_y_origin = 0; + STBI_NOTUSED(tga_palette_start); + // OK, done + return tga_data; } -#endif + #endif // ************************************************************************************************* -// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, +// tweaked by STB -#ifndef STBI_NO_PSD -static int stbi__psd_test(stbi__context *s) -{ - int r = (stbi__get32be(s) == 0x38425053); - stbi__rewind(s); - return r; + #ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s) { + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; } -static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) -{ - int count, nleft, len; +static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) { + int count, nleft, len; - count = 0; - while ((nleft = pixelCount - count) > 0) { - len = stbi__get8(s); - if (len == 128) { - // No-op. - } else if (len < 128) { - // Copy next len+1 bytes literally. - len++; - if (len > nleft) return 0; // corrupt data - count += len; - while (len) { - *p = stbi__get8(s); - p += 4; - len--; - } - } else if (len > 128) { - stbi_uc val; - // Next -len+1 bytes in the dest are replicated from next source byte. - // (Interpret len as a negative 8-bit int.) - len = 257 - len; - if (len > nleft) return 0; // corrupt data - val = stbi__get8(s); - count += len; - while (len) { - *p = val; - p += 4; - len--; - } + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + count += len; + while (len) { + *p = stbi__get8(s); + p += 4; + len--; } - } + } else if (len > 128) { + stbi_uc val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } - return 1; + return 1; } -static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) -{ - int pixelCount; - int channelCount, compression; - int channel, i; - int bitdepth; - int w,h; - stbi_uc *out; - STBI_NOTUSED(ri); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri, int bpc) { + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w, h; + stbi_uc *out; + STBI_NOTUSED(ri); - // Check identifier - if (stbi__get32be(s) != 0x38425053) // "8BPS" - return stbi__errpuc("not PSD", "Corrupt PSD image"); + // Check identifier + if (stbi__get32be(s) != 0x38425053) // "8BPS" + return stbi__errpuc("not PSD", "Corrupt PSD image"); - // Check file type version. - if (stbi__get16be(s) != 1) - return stbi__errpuc("wrong version", "Unsupported version of PSD image"); + // Check file type version. + if (stbi__get16be(s) != 1) + return stbi__errpuc("wrong version", "Unsupported version of PSD image"); - // Skip 6 reserved bytes. - stbi__skip(s, 6 ); + // Skip 6 reserved bytes. + stbi__skip(s, 6); - // Read the number of channels (R, G, B, A, etc). - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) - return stbi__errpuc("wrong channel count", "Unsupported number of channels in PSD image"); + // Read the number of channels (R, G, B, A, etc). + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) + return stbi__errpuc("wrong channel count", + "Unsupported number of channels in PSD image"); - // Read the rows and columns of the image. - h = stbi__get32be(s); - w = stbi__get32be(s); + // Read the rows and columns of the image. + h = stbi__get32be(s); + w = stbi__get32be(s); - if (h > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (w > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (h > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (w > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - // Make sure the depth is 8 bits. - bitdepth = stbi__get16be(s); - if (bitdepth != 8 && bitdepth != 16) - return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 bit"); + // Make sure the depth is 8 bits. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", + "PSD bit depth is not 8 or 16 bit"); - // Make sure the color mode is RGB. - // Valid options are: - // 0: Bitmap - // 1: Grayscale - // 2: Indexed color - // 3: RGB color - // 4: CMYK color - // 7: Multichannel - // 8: Duotone - // 9: Lab color - if (stbi__get16be(s) != 3) - return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (stbi__get16be(s) != 3) + return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); - // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) - stbi__skip(s,stbi__get32be(s) ); + // Skip the Mode Data. (It's the palette for indexed color; other info for + // other modes.) + stbi__skip(s, stbi__get32be(s)); - // Skip the image resources. (resolution, pen tool paths, etc) - stbi__skip(s, stbi__get32be(s) ); + // Skip the image resources. (resolution, pen tool paths, etc) + stbi__skip(s, stbi__get32be(s)); - // Skip the reserved data. - stbi__skip(s, stbi__get32be(s) ); + // Skip the reserved data. + stbi__skip(s, stbi__get32be(s)); - // Find out if the data is compressed. - // Known values: - // 0: no compression - // 1: RLE compressed - compression = stbi__get16be(s); - if (compression > 1) - return stbi__errpuc("bad compression", "PSD has an unknown compression format"); + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = stbi__get16be(s); + if (compression > 1) + return stbi__errpuc("bad compression", + "PSD has an unknown compression format"); - // Check size - if (!stbi__mad3sizes_valid(4, w, h, 0)) - return stbi__errpuc("too large", "Corrupt PSD"); + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); - // Create the destination image. + // Create the destination image. - if (!compression && bitdepth == 16 && bpc == 16) { - out = (stbi_uc *) stbi__malloc_mad3(8, w, h, 0); - ri->bits_per_channel = 16; - } else - out = (stbi_uc *) stbi__malloc(4 * w*h); + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc *)stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } else + out = (stbi_uc *)stbi__malloc(4 * w * h); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - pixelCount = w*h; + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + pixelCount = w * h; - // Initialize the data to zero. - //memset( out, 0, pixelCount * 4 ); + // Initialize the data to zero. + // memset( out, 0, pixelCount * 4 ); - // Finally, the image data. - if (compression) { - // RLE as used by .PSD and .TIFF - // Loop until you get the number of unpacked bytes you are expecting: - // Read the next source byte into n. - // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. - // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. - // Else if n is 128, noop. - // Endloop + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes + // literally. Else if n is between -127 and -1 inclusive, copy the next + // byte -n+1 times. Else if n is 128, noop. + // Endloop - // The RLE-compressed data is preceded by a 2-byte data count for each row in the data, - // which we're going to just skip. - stbi__skip(s, h * channelCount * 2 ); + // The RLE-compressed data is preceded by a 2-byte data count for each row + // in the data, which we're going to just skip. + stbi__skip(s, h * channelCount * 2); - // Read the RLE data by channel. - for (channel = 0; channel < 4; channel++) { - stbi_uc *p; + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; - p = out+channel; - if (channel >= channelCount) { - // Fill this channel with default data. - for (i = 0; i < pixelCount; i++, p += 4) - *p = (channel == 3 ? 255 : 0); - } else { - // Read the RLE data. - if (!stbi__psd_decode_rle(s, p, pixelCount)) { - STBI_FREE(out); - return stbi__errpuc("corrupt", "bad RLE data"); - } - } - } - - } else { - // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) - // where each channel consists of an 8-bit (or 16-bit) value for each pixel in the image. - - // Read the data by channel. - for (channel = 0; channel < 4; channel++) { - if (channel >= channelCount) { - // Fill this channel with default data. - if (bitdepth == 16 && bpc == 16) { - stbi__uint16 *q = ((stbi__uint16 *) out) + channel; - stbi__uint16 val = channel == 3 ? 65535 : 0; - for (i = 0; i < pixelCount; i++, q += 4) - *q = val; - } else { - stbi_uc *p = out+channel; - stbi_uc val = channel == 3 ? 255 : 0; - for (i = 0; i < pixelCount; i++, p += 4) - *p = val; - } - } else { - if (ri->bits_per_channel == 16) { // output bpc - stbi__uint16 *q = ((stbi__uint16 *) out) + channel; - for (i = 0; i < pixelCount; i++, q += 4) - *q = (stbi__uint16) stbi__get16be(s); - } else { - stbi_uc *p = out+channel; - if (bitdepth == 16) { // input bpc - for (i = 0; i < pixelCount; i++, p += 4) - *p = (stbi_uc) (stbi__get16be(s) >> 8); - } else { - for (i = 0; i < pixelCount; i++, p += 4) - *p = stbi__get8(s); - } - } - } - } - } - - // remove weird white matte from PSD - if (channelCount >= 4) { - if (ri->bits_per_channel == 16) { - for (i=0; i < w*h; ++i) { - stbi__uint16 *pixel = (stbi__uint16 *) out + 4*i; - if (pixel[3] != 0 && pixel[3] != 65535) { - float a = pixel[3] / 65535.0f; - float ra = 1.0f / a; - float inv_a = 65535.0f * (1 - ra); - pixel[0] = (stbi__uint16) (pixel[0]*ra + inv_a); - pixel[1] = (stbi__uint16) (pixel[1]*ra + inv_a); - pixel[2] = (stbi__uint16) (pixel[2]*ra + inv_a); - } - } + p = out + channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++, p += 4) + *p = (channel == 3 ? 255 : 0); } else { - for (i=0; i < w*h; ++i) { - unsigned char *pixel = out + 4*i; - if (pixel[3] != 0 && pixel[3] != 255) { - float a = pixel[3] / 255.0f; - float ra = 1.0f / a; - float inv_a = 255.0f * (1 - ra); - pixel[0] = (unsigned char) (pixel[0]*ra + inv_a); - pixel[1] = (unsigned char) (pixel[1]*ra + inv_a); - pixel[2] = (unsigned char) (pixel[2]*ra + inv_a); - } - } + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } } - } + } - // convert to desired output format - if (req_comp && req_comp != 4) { - if (ri->bits_per_channel == 16) - out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, 4, req_comp, w, h); - else - out = stbi__convert_format(out, 4, req_comp, w, h); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } + } else { + // We're at the raw image data. It's each channel in order (Red, Green, + // Blue, Alpha, ...) where each channel consists of an 8-bit (or 16-bit) + // value for each pixel in the image. - if (comp) *comp = 4; - *y = h; - *x = w; + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16 *q = ((stbi__uint16 *)out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + *q = val; + } else { + stbi_uc *p = out + channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + *p = val; + } + } else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16 *q = ((stbi__uint16 *)out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + *q = (stbi__uint16)stbi__get16be(s); + } else { + stbi_uc *p = out + channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + *p = (stbi_uc)(stbi__get16be(s) >> 8); + } else { + for (i = 0; i < pixelCount; i++, p += 4) + *p = stbi__get8(s); + } + } + } + } + } - return out; + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i = 0; i < w * h; ++i) { + stbi__uint16 *pixel = (stbi__uint16 *)out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16)(pixel[0] * ra + inv_a); + pixel[1] = (stbi__uint16)(pixel[1] * ra + inv_a); + pixel[2] = (stbi__uint16)(pixel[2] * ra + inv_a); + } + } + } else { + for (i = 0; i < w * h; ++i) { + unsigned char *pixel = out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char)(pixel[0] * ra + inv_a); + pixel[1] = (unsigned char)(pixel[1] * ra + inv_a); + pixel[2] = (unsigned char)(pixel[2] * ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc *)stbi__convert_format16((stbi__uint16 *)out, 4, req_comp, + w, h); + else + out = stbi__convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + if (comp) *comp = 4; + *y = h; + *x = w; + + return out; } -#endif + #endif // ************************************************************************************************* // Softimage PIC loader @@ -6151,1130 +6677,1139 @@ static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req // See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format // See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ -#ifndef STBI_NO_PIC -static int stbi__pic_is4(stbi__context *s,const char *str) -{ - int i; - for (i=0; i<4; ++i) - if (stbi__get8(s) != (stbi_uc)str[i]) - return 0; + #ifndef STBI_NO_PIC +static int stbi__pic_is4(stbi__context *s, const char *str) { + int i; + for (i = 0; i < 4; ++i) + if (stbi__get8(s) != (stbi_uc)str[i]) return 0; - return 1; + return 1; } -static int stbi__pic_test_core(stbi__context *s) -{ - int i; +static int stbi__pic_test_core(stbi__context *s) { + int i; - if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) - return 0; + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) return 0; - for(i=0;i<84;++i) - stbi__get8(s); + for (i = 0; i < 84; ++i) + stbi__get8(s); - if (!stbi__pic_is4(s,"PICT")) - return 0; + if (!stbi__pic_is4(s, "PICT")) return 0; - return 1; + return 1; } -typedef struct -{ - stbi_uc size,type,channel; +typedef struct { + stbi_uc size, type, channel; } stbi__pic_packet; -static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) -{ - int mask=0x80, i; +static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) { + int mask = 0x80, i; - for (i=0; i<4; ++i, mask>>=1) { - if (channel & mask) { - if (stbi__at_eof(s)) return stbi__errpuc("bad file","PIC file too short"); - dest[i]=stbi__get8(s); - } - } + for (i = 0; i < 4; ++i, mask >>= 1) { + if (channel & mask) { + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "PIC file too short"); + dest[i] = stbi__get8(s); + } + } - return dest; + return dest; } -static void stbi__copyval(int channel,stbi_uc *dest,const stbi_uc *src) -{ - int mask=0x80,i; +static void stbi__copyval(int channel, stbi_uc *dest, const stbi_uc *src) { + int mask = 0x80, i; - for (i=0;i<4; ++i, mask>>=1) - if (channel&mask) - dest[i]=src[i]; + for (i = 0; i < 4; ++i, mask >>= 1) + if (channel & mask) dest[i] = src[i]; } -static stbi_uc *stbi__pic_load_core(stbi__context *s,int width,int height,int *comp, stbi_uc *result) -{ - int act_comp=0,num_packets=0,y,chained; - stbi__pic_packet packets[10]; +static stbi_uc *stbi__pic_load_core(stbi__context *s, int width, int height, + int *comp, stbi_uc *result) { + int act_comp = 0, num_packets = 0, y, chained; + stbi__pic_packet packets[10]; - // this will (should...) cater for even some bizarre stuff like having data - // for the same channel in multiple packets. - do { - stbi__pic_packet *packet; + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + stbi__pic_packet *packet; - if (num_packets==sizeof(packets)/sizeof(packets[0])) - return stbi__errpuc("bad format","too many packets"); + if (num_packets == sizeof(packets) / sizeof(packets[0])) + return stbi__errpuc("bad format", "too many packets"); - packet = &packets[num_packets++]; + packet = &packets[num_packets++]; - chained = stbi__get8(s); - packet->size = stbi__get8(s); - packet->type = stbi__get8(s); - packet->channel = stbi__get8(s); + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); - act_comp |= packet->channel; + act_comp |= packet->channel; - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (reading packets)"); - if (packet->size != 8) return stbi__errpuc("bad format","packet isn't 8bpp"); - } while (chained); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "file too short (reading packets)"); + if (packet->size != 8) + return stbi__errpuc("bad format", "packet isn't 8bpp"); + } while (chained); - *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? - for(y=0; ytype) { - default: - return stbi__errpuc("bad format","packet has bad compression type"); + switch (packet->type) { + default: + return stbi__errpuc("bad format", "packet has bad compression type"); - case 0: {//uncompressed - int x; + case 0: { // uncompressed + int x; - for(x=0;xchannel,dest)) - return 0; - break; + for (x = 0; x < width; ++x, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) return 0; + break; + } + + case 1: // Pure RLE + { + int left = width, i; + + while (left > 0) { + stbi_uc count, value[4]; + + count = stbi__get8(s); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", + "file too short (pure read count)"); + + if (count > left) count = (stbi_uc)left; + + if (!stbi__readval(s, packet->channel, value)) return 0; + + for (i = 0; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + left -= count; + } + } break; + + case 2: { // Mixed RLE + int left = width; + while (left > 0) { + int count = stbi__get8(s), i; + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", + "file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + + if (count == 128) + count = stbi__get16be(s); + else + count -= 127; + if (count > left) + return stbi__errpuc("bad file", "scanline overrun"); + + if (!stbi__readval(s, packet->channel, value)) return 0; + + for (i = 0; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + } else { // Raw + ++count; + if (count > left) + return stbi__errpuc("bad file", "scanline overrun"); + + for (i = 0; i < count; ++i, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) return 0; } - - case 1://Pure RLE - { - int left=width, i; - - while (left>0) { - stbi_uc count,value[4]; - - count=stbi__get8(s); - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pure read count)"); - - if (count > left) - count = (stbi_uc) left; - - if (!stbi__readval(s,packet->channel,value)) return 0; - - for(i=0; ichannel,dest,value); - left -= count; - } - } - break; - - case 2: {//Mixed RLE - int left=width; - while (left>0) { - int count = stbi__get8(s), i; - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (mixed read count)"); - - if (count >= 128) { // Repeated - stbi_uc value[4]; - - if (count==128) - count = stbi__get16be(s); - else - count -= 127; - if (count > left) - return stbi__errpuc("bad file","scanline overrun"); - - if (!stbi__readval(s,packet->channel,value)) - return 0; - - for(i=0;ichannel,dest,value); - } else { // Raw - ++count; - if (count>left) return stbi__errpuc("bad file","scanline overrun"); - - for(i=0;ichannel,dest)) - return 0; - } - left-=count; - } - break; - } - } + left -= count; + } + break; + } } - } + } + } - return result; + return result; } -static void *stbi__pic_load(stbi__context *s,int *px,int *py,int *comp,int req_comp, stbi__result_info *ri) -{ - stbi_uc *result; - int i, x,y, internal_comp; - STBI_NOTUSED(ri); +static void *stbi__pic_load(stbi__context *s, int *px, int *py, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *result; + int i, x, y, internal_comp; + STBI_NOTUSED(ri); - if (!comp) comp = &internal_comp; + if (!comp) comp = &internal_comp; - for (i=0; i<92; ++i) - stbi__get8(s); + for (i = 0; i < 92; ++i) + stbi__get8(s); - x = stbi__get16be(s); - y = stbi__get16be(s); + x = stbi__get16be(s); + y = stbi__get16be(s); - if (y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pic header)"); - if (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC image too large to decode"); + if (stbi__at_eof(s)) + return stbi__errpuc("bad file", "file too short (pic header)"); + if (!stbi__mad3sizes_valid(x, y, 4, 0)) + return stbi__errpuc("too large", "PIC image too large to decode"); - stbi__get32be(s); //skip `ratio' - stbi__get16be(s); //skip `fields' - stbi__get16be(s); //skip `pad' + stbi__get32be(s); // skip `ratio' + stbi__get16be(s); // skip `fields' + stbi__get16be(s); // skip `pad' - // intermediate buffer is RGBA - result = (stbi_uc *) stbi__malloc_mad3(x, y, 4, 0); - memset(result, 0xff, x*y*4); + // intermediate buffer is RGBA + result = (stbi_uc *)stbi__malloc_mad3(x, y, 4, 0); + memset(result, 0xff, x * y * 4); - if (!stbi__pic_load_core(s,x,y,comp, result)) { - STBI_FREE(result); - result=0; - } - *px = x; - *py = y; - if (req_comp == 0) req_comp = *comp; - result=stbi__convert_format(result,4,req_comp,x,y); + if (!stbi__pic_load_core(s, x, y, comp, result)) { + STBI_FREE(result); + result = 0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result = stbi__convert_format(result, 4, req_comp, x, y); - return result; + return result; } -static int stbi__pic_test(stbi__context *s) -{ - int r = stbi__pic_test_core(s); - stbi__rewind(s); - return r; +static int stbi__pic_test(stbi__context *s) { + int r = stbi__pic_test_core(s); + stbi__rewind(s); + return r; } -#endif + #endif // ************************************************************************************************* // GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb -#ifndef STBI_NO_GIF -typedef struct -{ - stbi__int16 prefix; - stbi_uc first; - stbi_uc suffix; + #ifndef STBI_NO_GIF +typedef struct { + stbi__int16 prefix; + stbi_uc first; + stbi_uc suffix; } stbi__gif_lzw; -typedef struct -{ - int w,h; - stbi_uc *out; // output buffer (always 4 components) - stbi_uc *background; // The current "background" as far as a gif is concerned - stbi_uc *history; - int flags, bgindex, ratio, transparent, eflags; - stbi_uc pal[256][4]; - stbi_uc lpal[256][4]; - stbi__gif_lzw codes[8192]; - stbi_uc *color_table; - int parse, step; - int lflags; - int start_x, start_y; - int max_x, max_y; - int cur_x, cur_y; - int line_size; - int delay; +typedef struct { + int w, h; + stbi_uc *out; // output buffer (always 4 components) + stbi_uc *background; // The current "background" as far as a gif is concerned + stbi_uc *history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + stbi_uc *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; + int delay; } stbi__gif; -static int stbi__gif_test_raw(stbi__context *s) -{ - int sz; - if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') return 0; - sz = stbi__get8(s); - if (sz != '9' && sz != '7') return 0; - if (stbi__get8(s) != 'a') return 0; - return 1; +static int stbi__gif_test_raw(stbi__context *s) { + int sz; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || + stbi__get8(s) != '8') + return 0; + sz = stbi__get8(s); + if (sz != '9' && sz != '7') return 0; + if (stbi__get8(s) != 'a') return 0; + return 1; } -static int stbi__gif_test(stbi__context *s) -{ - int r = stbi__gif_test_raw(s); - stbi__rewind(s); - return r; +static int stbi__gif_test(stbi__context *s) { + int r = stbi__gif_test_raw(s); + stbi__rewind(s); + return r; } -static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], int num_entries, int transp) -{ - int i; - for (i=0; i < num_entries; ++i) { - pal[i][2] = stbi__get8(s); - pal[i][1] = stbi__get8(s); - pal[i][0] = stbi__get8(s); - pal[i][3] = transp == i ? 0 : 255; - } +static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], + int num_entries, int transp) { + int i; + for (i = 0; i < num_entries; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + pal[i][3] = transp == i ? 0 : 255; + } } -static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, int is_info) -{ - stbi_uc version; - if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') - return stbi__err("not GIF", "Corrupt GIF"); +static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, + int is_info) { + stbi_uc version; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || + stbi__get8(s) != '8') + return stbi__err("not GIF", "Corrupt GIF"); - version = stbi__get8(s); - if (version != '7' && version != '9') return stbi__err("not GIF", "Corrupt GIF"); - if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); + version = stbi__get8(s); + if (version != '7' && version != '9') + return stbi__err("not GIF", "Corrupt GIF"); + if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); - stbi__g_failure_reason = ""; - g->w = stbi__get16le(s); - g->h = stbi__get16le(s); - g->flags = stbi__get8(s); - g->bgindex = stbi__get8(s); - g->ratio = stbi__get8(s); - g->transparent = -1; + stbi__g_failure_reason = ""; + g->w = stbi__get16le(s); + g->h = stbi__get16le(s); + g->flags = stbi__get8(s); + g->bgindex = stbi__get8(s); + g->ratio = stbi__get8(s); + g->transparent = -1; - if (g->w > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); - if (g->h > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (g->w > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); + if (g->h > STBI_MAX_DIMENSIONS) + return stbi__err("too large", "Very large image (corrupt?)"); - if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + if (comp != 0) + *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the + // comments - if (is_info) return 1; + if (is_info) return 1; - if (g->flags & 0x80) - stbi__gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + if (g->flags & 0x80) + stbi__gif_parse_colortable(s, g->pal, 2 << (g->flags & 7), -1); - return 1; + return 1; } -static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) -{ - stbi__gif* g = (stbi__gif*) stbi__malloc(sizeof(stbi__gif)); - if (!stbi__gif_header(s, g, comp, 1)) { - STBI_FREE(g); - stbi__rewind( s ); - return 0; - } - if (x) *x = g->w; - if (y) *y = g->h; - STBI_FREE(g); - return 1; +static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) { + stbi__gif *g = (stbi__gif *)stbi__malloc(sizeof(stbi__gif)); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind(s); + return 0; + } + if (x) *x = g->w; + if (y) *y = g->h; + STBI_FREE(g); + return 1; } -static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) -{ - stbi_uc *p, *c; - int idx; +static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) { + stbi_uc *p, *c; + int idx; - // recurse to decode the prefixes, since the linked-list is backwards, - // and working backwards through an interleaved image would be nasty - if (g->codes[code].prefix >= 0) - stbi__out_gif_code(g, g->codes[code].prefix); + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) stbi__out_gif_code(g, g->codes[code].prefix); - if (g->cur_y >= g->max_y) return; + if (g->cur_y >= g->max_y) return; - idx = g->cur_x + g->cur_y; - p = &g->out[idx]; - g->history[idx / 4] = 1; + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; - c = &g->color_table[g->codes[code].suffix * 4]; - if (c[3] > 128) { // don't render transparent pixels; - p[0] = c[2]; - p[1] = c[1]; - p[2] = c[0]; - p[3] = c[3]; - } - g->cur_x += 4; + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; - if (g->cur_x >= g->max_x) { - g->cur_x = g->start_x; - g->cur_y += g->step; + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; - while (g->cur_y >= g->max_y && g->parse > 0) { - g->step = (1 << g->parse) * g->line_size; - g->cur_y = g->start_y + (g->step >> 1); - --g->parse; + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) { + stbi_uc lzw_cs; + stbi__int32 len, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw *p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc)init_code; + g->codes[init_code].suffix = (stbi_uc)init_code; + } + + // support no starting clear code + avail = clear + 2; + oldcode = -1; + + len = 0; + for (;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = stbi__get8(s); // start new block + if (len == 0) return g->out; } - } + --len; + bits |= (stbi__int32)stbi__get8(s) << valid_bits; + valid_bits += 8; + } else { + stbi__int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + stbi__skip(s, len); + while ((len = stbi__get8(s)) > 0) + stbi__skip(s, len); + return g->out; + } else if (code <= avail) { + if (first) { return stbi__errpuc("no clear code", "Corrupt GIF"); } + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 8192) { + return stbi__errpuc("too many codes", "Corrupt GIF"); + } + + p->prefix = (stbi__int16)oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + + stbi__out_gif_code(g, (stbi__uint16)code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + } + } + } } -static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) -{ - stbi_uc lzw_cs; - stbi__int32 len, init_code; - stbi__uint32 first; - stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; - stbi__gif_lzw *p; +// this function is designed to support animated gifs, although stb_image +// doesn't support it two back is the image from two frames ago, used for a very +// specific disposal format +static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, + int req_comp, stbi_uc *two_back) { + int dispose; + int first_frame; + int pi; + int pcount; + STBI_NOTUSED(req_comp); - lzw_cs = stbi__get8(s); - if (lzw_cs > 12) return NULL; - clear = 1 << lzw_cs; - first = 1; - codesize = lzw_cs + 1; - codemask = (1 << codesize) - 1; - bits = 0; - valid_bits = 0; - for (init_code = 0; init_code < clear; init_code++) { - g->codes[init_code].prefix = -1; - g->codes[init_code].first = (stbi_uc) init_code; - g->codes[init_code].suffix = (stbi_uc) init_code; - } + // on first frame, any non-written pixels get the background colour + // (non-transparent) + first_frame = 0; + if (g->out == 0) { + if (!stbi__gif_header(s, g, comp, 0)) + return 0; // stbi__g_failure_reason set by stbi__gif_header + if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) + return stbi__errpuc("too large", "GIF image is too large"); + pcount = g->w * g->h; + g->out = (stbi_uc *)stbi__malloc(4 * pcount); + g->background = (stbi_uc *)stbi__malloc(4 * pcount); + g->history = (stbi_uc *)stbi__malloc(pcount); + if (!g->out || !g->background || !g->history) + return stbi__errpuc("outofmem", "Out of memory"); - // support no starting clear code - avail = clear+2; - oldcode = -1; + // image is treated as "transparent" at the start - ie, nothing overwrites + // the current background; background colour is only used for pixels that + // are not rendered first frame, after that "background" color refers to the + // color that was there the previous frame. + memset(g->out, 0x00, 4 * pcount); + memset(g->background, 0x00, + 4 * pcount); // state of the background (starts transparent) + memset(g->history, 0x00, + pcount); // pixels that were affected previous frame + first_frame = 1; + } else { + // second frame - how do we dispose of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; - len = 0; - for(;;) { - if (valid_bits < codesize) { - if (len == 0) { - len = stbi__get8(s); // start new block - if (len == 0) - return g->out; - } - --len; - bits |= (stbi__int32) stbi__get8(s) << valid_bits; - valid_bits += 8; - } else { - stbi__int32 code = bits & codemask; - bits >>= codesize; - valid_bits -= codesize; - // @OPTIMIZE: is there some way we can accelerate the non-clear path? - if (code == clear) { // clear code - codesize = lzw_cs + 1; - codemask = (1 << codesize) - 1; - avail = clear + 2; - oldcode = -1; - first = 0; - } else if (code == clear + 1) { // end of stream code + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to + // the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { memcpy(&g->out[pi * 4], &two_back[pi * 4], 4); } + } + } else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy(&g->out[pi * 4], &g->background[pi * 4], 4); + } + } + } else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy(g->background, g->out, 4 * g->w * g->h); + } + + // clear my history; + memset(g->history, 0x00, + g->w * g->h); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + case 0x2C: /* Image Descriptor */ + { + stbi__int32 x, y, w, h; + stbi_uc *o; + + x = stbi__get16le(s); + y = stbi__get16le(s); + w = stbi__get16le(s); + h = stbi__get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + // if the width of the specified rectangle is 0, that means + // we may not see *any* pixels or the image is malformed; + // to make sure this is caught, move the current y down to + // max_y (which is what out_gif_code checks). + if (w == 0) g->cur_y = g->max_y; + + g->lflags = stbi__get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi__gif_parse_colortable(s, g->lpal, 2 << (g->lflags & 7), + g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (stbi_uc *)g->lpal; + } else if (g->flags & 0x80) { + g->color_table = (stbi_uc *)g->pal; + } else + return stbi__errpuc("missing color table", "Corrupt GIF"); + + o = stbi__process_gif_raster(s, g); + if (!o) return NULL; + + // if this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = + 255; // just in case it was made transparent, undo that; It + // will be reset next frame if need be; + memcpy(&g->out[pi * 4], &g->pal[g->bgindex], 4); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = + 10 * + stbi__get16le( + s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { g->pal[g->transparent][3] = 255; } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { g->pal[g->transparent][3] = 0; } + } else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } else { stbi__skip(s, len); - while ((len = stbi__get8(s)) > 0) - stbi__skip(s,len); - return g->out; - } else if (code <= avail) { - if (first) { - return stbi__errpuc("no clear code", "Corrupt GIF"); - } - - if (oldcode >= 0) { - p = &g->codes[avail++]; - if (avail > 8192) { - return stbi__errpuc("too many codes", "Corrupt GIF"); - } - - p->prefix = (stbi__int16) oldcode; - p->first = g->codes[oldcode].first; - p->suffix = (code == avail) ? p->first : g->codes[code].first; - } else if (code == avail) - return stbi__errpuc("illegal code in raster", "Corrupt GIF"); - - stbi__out_gif_code(g, (stbi__uint16) code); - - if ((avail & codemask) == 0 && avail <= 0x0FFF) { - codesize++; - codemask = (1 << codesize) - 1; - } - - oldcode = code; - } else { - return stbi__errpuc("illegal code in raster", "Corrupt GIF"); - } - } - } -} - -// this function is designed to support animated gifs, although stb_image doesn't support it -// two back is the image from two frames ago, used for a very specific disposal format -static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, int req_comp, stbi_uc *two_back) -{ - int dispose; - int first_frame; - int pi; - int pcount; - STBI_NOTUSED(req_comp); - - // on first frame, any non-written pixels get the background colour (non-transparent) - first_frame = 0; - if (g->out == 0) { - if (!stbi__gif_header(s, g, comp,0)) return 0; // stbi__g_failure_reason set by stbi__gif_header - if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) - return stbi__errpuc("too large", "GIF image is too large"); - pcount = g->w * g->h; - g->out = (stbi_uc *) stbi__malloc(4 * pcount); - g->background = (stbi_uc *) stbi__malloc(4 * pcount); - g->history = (stbi_uc *) stbi__malloc(pcount); - if (!g->out || !g->background || !g->history) - return stbi__errpuc("outofmem", "Out of memory"); - - // image is treated as "transparent" at the start - ie, nothing overwrites the current background; - // background colour is only used for pixels that are not rendered first frame, after that "background" - // color refers to the color that was there the previous frame. - memset(g->out, 0x00, 4 * pcount); - memset(g->background, 0x00, 4 * pcount); // state of the background (starts transparent) - memset(g->history, 0x00, pcount); // pixels that were affected previous frame - first_frame = 1; - } else { - // second frame - how do we dispose of the previous one? - dispose = (g->eflags & 0x1C) >> 2; - pcount = g->w * g->h; - - if ((dispose == 3) && (two_back == 0)) { - dispose = 2; // if I don't have an image to revert back to, default to the old background - } - - if (dispose == 3) { // use previous graphic - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi]) { - memcpy( &g->out[pi * 4], &two_back[pi * 4], 4 ); - } - } - } else if (dispose == 2) { - // restore what was changed last frame to background before that frame; - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi]) { - memcpy( &g->out[pi * 4], &g->background[pi * 4], 4 ); - } - } - } else { - // This is a non-disposal case eithe way, so just - // leave the pixels as is, and they will become the new background - // 1: do not dispose - // 0: not specified. - } - - // background is what out is after the undoing of the previou frame; - memcpy( g->background, g->out, 4 * g->w * g->h ); - } - - // clear my history; - memset( g->history, 0x00, g->w * g->h ); // pixels that were affected previous frame - - for (;;) { - int tag = stbi__get8(s); - switch (tag) { - case 0x2C: /* Image Descriptor */ - { - stbi__int32 x, y, w, h; - stbi_uc *o; - - x = stbi__get16le(s); - y = stbi__get16le(s); - w = stbi__get16le(s); - h = stbi__get16le(s); - if (((x + w) > (g->w)) || ((y + h) > (g->h))) - return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); - - g->line_size = g->w * 4; - g->start_x = x * 4; - g->start_y = y * g->line_size; - g->max_x = g->start_x + w * 4; - g->max_y = g->start_y + h * g->line_size; - g->cur_x = g->start_x; - g->cur_y = g->start_y; - - // if the width of the specified rectangle is 0, that means - // we may not see *any* pixels or the image is malformed; - // to make sure this is caught, move the current y down to - // max_y (which is what out_gif_code checks). - if (w == 0) - g->cur_y = g->max_y; - - g->lflags = stbi__get8(s); - - if (g->lflags & 0x40) { - g->step = 8 * g->line_size; // first interlaced spacing - g->parse = 3; - } else { - g->step = g->line_size; - g->parse = 0; - } - - if (g->lflags & 0x80) { - stbi__gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); - g->color_table = (stbi_uc *) g->lpal; - } else if (g->flags & 0x80) { - g->color_table = (stbi_uc *) g->pal; - } else - return stbi__errpuc("missing color table", "Corrupt GIF"); - - o = stbi__process_gif_raster(s, g); - if (!o) return NULL; - - // if this was the first frame, - pcount = g->w * g->h; - if (first_frame && (g->bgindex > 0)) { - // if first frame, any pixel not drawn to gets the background color - for (pi = 0; pi < pcount; ++pi) { - if (g->history[pi] == 0) { - g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; - memcpy( &g->out[pi * 4], &g->pal[g->bgindex], 4 ); - } - } - } - - return o; - } - - case 0x21: // Comment Extension. - { - int len; - int ext = stbi__get8(s); - if (ext == 0xF9) { // Graphic Control Extension. - len = stbi__get8(s); - if (len == 4) { - g->eflags = stbi__get8(s); - g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. - - // unset old transparent - if (g->transparent >= 0) { - g->pal[g->transparent][3] = 255; - } - if (g->eflags & 0x01) { - g->transparent = stbi__get8(s); - if (g->transparent >= 0) { - g->pal[g->transparent][3] = 0; - } - } else { - // don't need transparent - stbi__skip(s, 1); - g->transparent = -1; - } - } else { - stbi__skip(s, len); - break; - } - } - while ((len = stbi__get8(s)) != 0) { - stbi__skip(s, len); - } break; - } - - case 0x3B: // gif stream termination code - return (stbi_uc *) s; // using '1' causes warning on some compilers - - default: - return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } + while ((len = stbi__get8(s)) != 0) { + stbi__skip(s, len); + } + break; } - } + + case 0x3B: // gif stream termination code + return (stbi_uc *)s; // using '1' causes warning on some compilers + + default: + return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } } -static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp) -{ - if (stbi__gif_test(s)) { - int layers = 0; - stbi_uc *u = 0; - stbi_uc *out = 0; - stbi_uc *two_back = 0; - stbi__gif g; - int stride; - int out_size = 0; - int delays_size = 0; - memset(&g, 0, sizeof(g)); - if (delays) { - *delays = 0; +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, + int *z, int *comp, int req_comp) { + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc *u = 0; + stbi_uc *out = 0; + stbi_uc *two_back = 0; + stbi__gif g; + int stride; + int out_size = 0; + int delays_size = 0; + memset(&g, 0, sizeof(g)); + if (delays) { *delays = 0; } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc *)s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + void *tmp = + (stbi_uc *)STBI_REALLOC_SIZED(out, out_size, layers * stride); + if (NULL == tmp) { + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + return stbi__errpuc("outofmem", "Out of memory"); + } else { + out = (stbi_uc *)tmp; + out_size = layers * stride; + } + + if (delays) { + *delays = (int *)STBI_REALLOC_SIZED(*delays, delays_size, + sizeof(int) * layers); + delays_size = layers * sizeof(int); + } + } else { + out = (stbi_uc *)stbi__malloc(layers * stride); + out_size = layers * stride; + if (delays) { + *delays = (int *)stbi__malloc(layers * sizeof(int)); + delays_size = layers * sizeof(int); + } + } + memcpy(out + ((layers - 1) * stride), u, stride); + if (layers >= 2) { two_back = out - 2 * stride; } + + if (delays) { (*delays)[layers - 1U] = g.delay; } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + STBI_NOTUSED(ri); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc *)s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } else if (g.out) { + // if there was an error and we allocated an image buffer, free it! + STBI_FREE(g.out); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + return u; +} + +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) { + return stbi__gif_info_raw(s, x, y, comp); +} + #endif + + // ************************************************************************************************* + // Radiance RGBE HDR loader + // originally by Nicolas Schulz + #ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context *s, const char *signature) { + int i; + for (i = 0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context *s) { + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if (!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + stbi__rewind(s); + } + return r; +} + + #define STBI__HDR_BUFLEN 1024 +static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) { + int len = 0; + char c = '\0'; + + c = (char)stbi__get8(z); + + while (!stbi__at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == STBI__HDR_BUFLEN - 1) { + // flush to end of line + while (!stbi__at_eof(z) && stbi__get8(z) != '\n') + ; + break; + } + c = (char)stbi__get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) { + if (input[3] != 0) { + float f1; + // Exponent + f1 = (float)ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: + output[3] = 1; /* fallthrough */ + case 3: + output[0] = output[1] = output[2] = 0; + break; + case 2: + output[1] = 1; /* fallthrough */ + case 1: + output[0] = 0; + break; + } + } +} + +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1, c2, z; + const char *headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s, buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && + strcmp(headerToken, "#?RGBE") != 0) + return stbi__errpf("not HDR", "Corrupt HDR image"); + + // Parse header + for (;;) { + token = stbi__hdr_gettoken(s, buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) + return stbi__errpf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = stbi__hdr_gettoken(s, buffer); + if (strncmp(token, "-Y ", 3)) + return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = (int)strtol(token, &token, 10); + while (*token == ' ') + ++token; + if (strncmp(token, "+X ", 3)) + return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = (int)strtol(token, NULL, 10); + + if (height > STBI_MAX_DIMENSIONS) + return stbi__errpf("too large", "Very large image (corrupt?)"); + if (width > STBI_MAX_DIMENSIONS) + return stbi__errpf("too large", "Very large image (corrupt?)"); + + *x = width; + *y = height; + + if (comp) *comp = 3; + if (req_comp == 0) req_comp = 3; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = + (float *)stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) return stbi__errpf("outofmem", "Out of memory"); + + // Load image data + // image data is stored as some number of sca + if (width < 8 || width >= 32768) { + // Read flat data + for (j = 0; j < height; ++j) { + for (i = 0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + stbi__getn(s, rgbe, 4); + stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, + req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = stbi__get8(s); + c2 = stbi__get8(s); + len = stbi__get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a + // decoded pixel (note this can't be a valid pixel--one of RGB must be + // >= 128) + stbi_uc rgbe[4]; + rgbe[0] = (stbi_uc)c1; + rgbe[1] = (stbi_uc)c2; + rgbe[2] = (stbi_uc)len; + rgbe[3] = (stbi_uc)stbi__get8(s); + stbi__hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); + } + if (scanline == NULL) { + scanline = (stbi_uc *)stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } } - do { - u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); - if (u == (stbi_uc *) s) u = 0; // end of animated gif marker - - if (u) { - *x = g.w; - *y = g.h; - ++layers; - stride = g.w * g.h * 4; - - if (out) { - void *tmp = (stbi_uc*) STBI_REALLOC_SIZED( out, out_size, layers * stride ); - if (NULL == tmp) { - STBI_FREE(g.out); - STBI_FREE(g.history); - STBI_FREE(g.background); - return stbi__errpuc("outofmem", "Out of memory"); - } - else { - out = (stbi_uc*) tmp; - out_size = layers * stride; - } - - if (delays) { - *delays = (int*) STBI_REALLOC_SIZED( *delays, delays_size, sizeof(int) * layers ); - delays_size = layers * sizeof(int); - } - } else { - out = (stbi_uc*)stbi__malloc( layers * stride ); - out_size = layers * stride; - if (delays) { - *delays = (int*) stbi__malloc( layers * sizeof(int) ); - delays_size = layers * sizeof(int); - } + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if (count > nleft) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("corrupt", "bad RLE data in HDR"); } - memcpy( out + ((layers - 1) * stride), u, stride ); - if (layers >= 2) { - two_back = out - 2 * stride; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + if (count > nleft) { + STBI_FREE(hdr_data); + STBI_FREE(scanline); + return stbi__errpf("corrupt", "bad RLE data in HDR"); } - - if (delays) { - (*delays)[layers - 1U] = g.delay; - } - } - } while (u != 0); - - // free temp buffer; - STBI_FREE(g.out); - STBI_FREE(g.history); - STBI_FREE(g.background); - - // do the final conversion after loading everything; - if (req_comp && req_comp != 4) - out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); - - *z = layers; - return out; - } else { - return stbi__errpuc("not GIF", "Image was not as a gif type."); - } -} - -static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *u = 0; - stbi__gif g; - memset(&g, 0, sizeof(g)); - STBI_NOTUSED(ri); - - u = stbi__gif_load_next(s, &g, comp, req_comp, 0); - if (u == (stbi_uc *) s) u = 0; // end of animated gif marker - if (u) { - *x = g.w; - *y = g.h; - - // moved conversion to after successful load so that the same - // can be done for multiple frames. - if (req_comp && req_comp != 4) - u = stbi__convert_format(u, 4, req_comp, g.w, g.h); - } else if (g.out) { - // if there was an error and we allocated an image buffer, free it! - STBI_FREE(g.out); - } - - // free buffers needed for multiple frame loading; - STBI_FREE(g.history); - STBI_FREE(g.background); - - return u; -} - -static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) -{ - return stbi__gif_info_raw(s,x,y,comp); -} -#endif - -// ************************************************************************************************* -// Radiance RGBE HDR loader -// originally by Nicolas Schulz -#ifndef STBI_NO_HDR -static int stbi__hdr_test_core(stbi__context *s, const char *signature) -{ - int i; - for (i=0; signature[i]; ++i) - if (stbi__get8(s) != signature[i]) - return 0; - stbi__rewind(s); - return 1; -} - -static int stbi__hdr_test(stbi__context* s) -{ - int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); - stbi__rewind(s); - if(!r) { - r = stbi__hdr_test_core(s, "#?RGBE\n"); - stbi__rewind(s); - } - return r; -} - -#define STBI__HDR_BUFLEN 1024 -static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) -{ - int len=0; - char c = '\0'; - - c = (char) stbi__get8(z); - - while (!stbi__at_eof(z) && c != '\n') { - buffer[len++] = c; - if (len == STBI__HDR_BUFLEN-1) { - // flush to end of line - while (!stbi__at_eof(z) && stbi__get8(z) != '\n') - ; - break; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = stbi__get8(s); + } + } } - c = (char) stbi__get8(z); - } + for (i = 0; i < width; ++i) + stbi__hdr_convert(hdr_data + (j * width + i) * req_comp, + scanline + i * 4, req_comp); + } + if (scanline) STBI_FREE(scanline); + } - buffer[len] = 0; - return buffer; + return hdr_data; } -static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) -{ - if ( input[3] != 0 ) { - float f1; - // Exponent - f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); - if (req_comp <= 2) - output[0] = (input[0] + input[1] + input[2]) * f1 / 3; - else { - output[0] = input[0] * f1; - output[1] = input[1] * f1; - output[2] = input[2] * f1; - } - if (req_comp == 2) output[1] = 1; - if (req_comp == 4) output[3] = 1; - } else { - switch (req_comp) { - case 4: output[3] = 1; /* fallthrough */ - case 3: output[0] = output[1] = output[2] = 0; - break; - case 2: output[1] = 1; /* fallthrough */ - case 1: output[0] = 0; - break; - } - } +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) { + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 0) { + stbi__rewind(s); + return 0; + } + + for (;;) { + token = stbi__hdr_gettoken(s, buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi__rewind(s); + return 0; + } + token = stbi__hdr_gettoken(s, buffer); + if (strncmp(token, "-Y ", 3)) { + stbi__rewind(s); + return 0; + } + token += 3; + *y = (int)strtol(token, &token, 10); + while (*token == ' ') + ++token; + if (strncmp(token, "+X ", 3)) { + stbi__rewind(s); + return 0; + } + token += 3; + *x = (int)strtol(token, NULL, 10); + *comp = 3; + return 1; +} + #endif // STBI_NO_HDR + + #ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) { + void *p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + stbi__rewind(s); + if (p == NULL) return 0; + if (x) *x = s->img_x; + if (y) *y = s->img_y; + if (comp) { + if (info.bpp == 24 && info.ma == 0xff000000) + *comp = 3; + else + *comp = info.ma ? 4 : 3; + } + return 1; +} + #endif + + #ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) { + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind(s); + return 0; + } + *y = stbi__get32be(s); + *x = stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind(s); + return 0; + } + *comp = 4; + return 1; } -static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - char buffer[STBI__HDR_BUFLEN]; - char *token; - int valid = 0; - int width, height; - stbi_uc *scanline; - float *hdr_data; - int len; - unsigned char count, value; - int i, j, k, c1,c2, z; - const char *headerToken; - STBI_NOTUSED(ri); - - // Check identifier - headerToken = stbi__hdr_gettoken(s,buffer); - if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 0) - return stbi__errpf("not HDR", "Corrupt HDR image"); - - // Parse header - for(;;) { - token = stbi__hdr_gettoken(s,buffer); - if (token[0] == 0) break; - if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; - } - - if (!valid) return stbi__errpf("unsupported format", "Unsupported HDR format"); - - // Parse width and height - // can't use sscanf() if we're not using stdio! - token = stbi__hdr_gettoken(s,buffer); - if (strncmp(token, "-Y ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); - token += 3; - height = (int) strtol(token, &token, 10); - while (*token == ' ') ++token; - if (strncmp(token, "+X ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); - token += 3; - width = (int) strtol(token, NULL, 10); - - if (height > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); - if (width > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); - - *x = width; - *y = height; - - if (comp) *comp = 3; - if (req_comp == 0) req_comp = 3; - - if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) - return stbi__errpf("too large", "HDR image is too large"); - - // Read data - hdr_data = (float *) stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); - if (!hdr_data) - return stbi__errpf("outofmem", "Out of memory"); - - // Load image data - // image data is stored as some number of sca - if ( width < 8 || width >= 32768) { - // Read flat data - for (j=0; j < height; ++j) { - for (i=0; i < width; ++i) { - stbi_uc rgbe[4]; - main_decode_loop: - stbi__getn(s, rgbe, 4); - stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); - } - } - } else { - // Read RLE-encoded data - scanline = NULL; - - for (j = 0; j < height; ++j) { - c1 = stbi__get8(s); - c2 = stbi__get8(s); - len = stbi__get8(s); - if (c1 != 2 || c2 != 2 || (len & 0x80)) { - // not run-length encoded, so we have to actually use THIS data as a decoded - // pixel (note this can't be a valid pixel--one of RGB must be >= 128) - stbi_uc rgbe[4]; - rgbe[0] = (stbi_uc) c1; - rgbe[1] = (stbi_uc) c2; - rgbe[2] = (stbi_uc) len; - rgbe[3] = (stbi_uc) stbi__get8(s); - stbi__hdr_convert(hdr_data, rgbe, req_comp); - i = 1; - j = 0; - STBI_FREE(scanline); - goto main_decode_loop; // yes, this makes no sense - } - len <<= 8; - len |= stbi__get8(s); - if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } - if (scanline == NULL) { - scanline = (stbi_uc *) stbi__malloc_mad2(width, 4, 0); - if (!scanline) { - STBI_FREE(hdr_data); - return stbi__errpf("outofmem", "Out of memory"); - } - } - - for (k = 0; k < 4; ++k) { - int nleft; - i = 0; - while ((nleft = width - i) > 0) { - count = stbi__get8(s); - if (count > 128) { - // Run - value = stbi__get8(s); - count -= 128; - if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } - for (z = 0; z < count; ++z) - scanline[i++ * 4 + k] = value; - } else { - // Dump - if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } - for (z = 0; z < count; ++z) - scanline[i++ * 4 + k] = stbi__get8(s); - } - } - } - for (i=0; i < width; ++i) - stbi__hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); - } - if (scanline) - STBI_FREE(scanline); - } - - return hdr_data; +static int stbi__psd_is16(stbi__context *s) { + int channelCount, depth; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind(s); + return 0; + } + (void)stbi__get32be(s); + (void)stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind(s); + return 0; + } + return 1; } + #endif -static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) -{ - char buffer[STBI__HDR_BUFLEN]; - char *token; - int valid = 0; - int dummy; + #ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) { + int act_comp = 0, num_packets = 0, chained, dummy; + stbi__pic_packet packets[10]; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; - if (stbi__hdr_test(s) == 0) { - stbi__rewind( s ); - return 0; - } + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } - for(;;) { - token = stbi__hdr_gettoken(s,buffer); - if (token[0] == 0) break; - if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; - } + stbi__skip(s, 88); - if (!valid) { - stbi__rewind( s ); - return 0; - } - token = stbi__hdr_gettoken(s,buffer); - if (strncmp(token, "-Y ", 3)) { - stbi__rewind( s ); - return 0; - } - token += 3; - *y = (int) strtol(token, &token, 10); - while (*token == ' ') ++token; - if (strncmp(token, "+X ", 3)) { - stbi__rewind( s ); - return 0; - } - token += 3; - *x = (int) strtol(token, NULL, 10); - *comp = 3; - return 1; -} -#endif // STBI_NO_HDR + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind(s); + return 0; + } + if ((*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi__rewind(s); + return 0; + } -#ifndef STBI_NO_BMP -static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) -{ - void *p; - stbi__bmp_data info; + stbi__skip(s, 8); - info.all_a = 255; - p = stbi__bmp_parse_header(s, &info); - stbi__rewind( s ); - if (p == NULL) - return 0; - if (x) *x = s->img_x; - if (y) *y = s->img_y; - if (comp) { - if (info.bpp == 24 && info.ma == 0xff000000) - *comp = 3; - else - *comp = info.ma ? 4 : 3; - } - return 1; -} -#endif + do { + stbi__pic_packet *packet; -#ifndef STBI_NO_PSD -static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) -{ - int channelCount, dummy, depth; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; - if (stbi__get32be(s) != 0x38425053) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 1) { - stbi__rewind( s ); - return 0; - } - stbi__skip(s, 6); - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) { - stbi__rewind( s ); - return 0; - } - *y = stbi__get32be(s); - *x = stbi__get32be(s); - depth = stbi__get16be(s); - if (depth != 8 && depth != 16) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 3) { - stbi__rewind( s ); - return 0; - } - *comp = 4; - return 1; -} + if (num_packets == sizeof(packets) / sizeof(packets[0])) return 0; -static int stbi__psd_is16(stbi__context *s) -{ - int channelCount, depth; - if (stbi__get32be(s) != 0x38425053) { - stbi__rewind( s ); - return 0; - } - if (stbi__get16be(s) != 1) { - stbi__rewind( s ); - return 0; - } - stbi__skip(s, 6); - channelCount = stbi__get16be(s); - if (channelCount < 0 || channelCount > 16) { - stbi__rewind( s ); - return 0; - } - (void) stbi__get32be(s); - (void) stbi__get32be(s); - depth = stbi__get16be(s); - if (depth != 16) { - stbi__rewind( s ); - return 0; - } - return 1; -} -#endif + packet = &packets[num_packets++]; + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + act_comp |= packet->channel; -#ifndef STBI_NO_PIC -static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) -{ - int act_comp=0,num_packets=0,chained,dummy; - stbi__pic_packet packets[10]; - - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; - - if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) { + if (stbi__at_eof(s)) { stbi__rewind(s); return 0; - } - - stbi__skip(s, 88); - - *x = stbi__get16be(s); - *y = stbi__get16be(s); - if (stbi__at_eof(s)) { - stbi__rewind( s); + } + if (packet->size != 8) { + stbi__rewind(s); return 0; - } - if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { - stbi__rewind( s ); - return 0; - } + } + } while (chained); - stbi__skip(s, 8); + *comp = (act_comp & 0x10 ? 4 : 3); - do { - stbi__pic_packet *packet; - - if (num_packets==sizeof(packets)/sizeof(packets[0])) - return 0; - - packet = &packets[num_packets++]; - chained = stbi__get8(s); - packet->size = stbi__get8(s); - packet->type = stbi__get8(s); - packet->channel = stbi__get8(s); - act_comp |= packet->channel; - - if (stbi__at_eof(s)) { - stbi__rewind( s ); - return 0; - } - if (packet->size != 8) { - stbi__rewind( s ); - return 0; - } - } while (chained); - - *comp = (act_comp & 0x10 ? 4 : 3); - - return 1; + return 1; } -#endif + #endif // ************************************************************************************************* // Portable Gray Map and Portable Pixel Map loader @@ -7288,259 +7823,248 @@ static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) // Does not support ASCII image data (formats P2 and P3) // Does not support 16-bit-per-channel -#ifndef STBI_NO_PNM + #ifndef STBI_NO_PNM -static int stbi__pnm_test(stbi__context *s) -{ - char p, t; - p = (char) stbi__get8(s); - t = (char) stbi__get8(s); - if (p != 'P' || (t != '5' && t != '6')) { - stbi__rewind( s ); - return 0; - } - return 1; +static int stbi__pnm_test(stbi__context *s) { + char p, t; + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + return 1; } -static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) -{ - stbi_uc *out; - STBI_NOTUSED(ri); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, + int req_comp, stbi__result_info *ri) { + stbi_uc *out; + STBI_NOTUSED(ri); - if (!stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n)) - return 0; + if (!stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n)) + return 0; - if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); - if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (s->img_y > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) + return stbi__errpuc("too large", "Very large image (corrupt?)"); - *x = s->img_x; - *y = s->img_y; - if (comp) *comp = s->img_n; + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; - if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) - return stbi__errpuc("too large", "PNM too large"); + if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "PNM too large"); - out = (stbi_uc *) stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); - if (!out) return stbi__errpuc("outofmem", "Out of memory"); - stbi__getn(s, out, s->img_n * s->img_x * s->img_y); + out = (stbi_uc *)stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + stbi__getn(s, out, s->img_n * s->img_x * s->img_y); - if (req_comp && req_comp != s->img_n) { - out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); - if (out == NULL) return out; // stbi__convert_format frees input on failure - } - return out; + if (req_comp && req_comp != s->img_n) { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; } -static int stbi__pnm_isspace(char c) -{ - return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +static int stbi__pnm_isspace(char c) { + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || + c == '\r'; } -static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) -{ - for (;;) { - while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) - *c = (char) stbi__get8(s); +static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) { + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + *c = (char)stbi__get8(s); - if (stbi__at_eof(s) || *c != '#') - break; + if (stbi__at_eof(s) || *c != '#') break; - while (!stbi__at_eof(s) && *c != '\n' && *c != '\r' ) - *c = (char) stbi__get8(s); - } + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r') + *c = (char)stbi__get8(s); + } } -static int stbi__pnm_isdigit(char c) -{ - return c >= '0' && c <= '9'; +static int stbi__pnm_isdigit(char c) { + return c >= '0' && c <= '9'; } -static int stbi__pnm_getinteger(stbi__context *s, char *c) -{ - int value = 0; +static int stbi__pnm_getinteger(stbi__context *s, char *c) { + int value = 0; - while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { - value = value*10 + (*c - '0'); - *c = (char) stbi__get8(s); - } + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value * 10 + (*c - '0'); + *c = (char)stbi__get8(s); + } - return value; + return value; } -static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) -{ - int maxv, dummy; - char c, p, t; +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) { + int maxv, dummy; + char c, p, t; - if (!x) x = &dummy; - if (!y) y = &dummy; - if (!comp) comp = &dummy; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; - stbi__rewind(s); + stbi__rewind(s); - // Get identifier - p = (char) stbi__get8(s); - t = (char) stbi__get8(s); - if (p != 'P' || (t != '5' && t != '6')) { - stbi__rewind(s); - return 0; - } + // Get identifier + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } - *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + *comp = + (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm - c = (char) stbi__get8(s); - stbi__pnm_skip_whitespace(s, &c); + c = (char)stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); - *x = stbi__pnm_getinteger(s, &c); // read width - stbi__pnm_skip_whitespace(s, &c); + *x = stbi__pnm_getinteger(s, &c); // read width + stbi__pnm_skip_whitespace(s, &c); - *y = stbi__pnm_getinteger(s, &c); // read height - stbi__pnm_skip_whitespace(s, &c); + *y = stbi__pnm_getinteger(s, &c); // read height + stbi__pnm_skip_whitespace(s, &c); - maxv = stbi__pnm_getinteger(s, &c); // read max value + maxv = stbi__pnm_getinteger(s, &c); // read max value - if (maxv > 255) - return stbi__err("max value > 255", "PPM image not 8-bit"); - else - return 1; + if (maxv > 255) + return stbi__err("max value > 255", "PPM image not 8-bit"); + else + return 1; } -#endif + #endif -static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) -{ - #ifndef STBI_NO_JPEG - if (stbi__jpeg_info(s, x, y, comp)) return 1; - #endif +static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) { + #ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PNG - if (stbi__png_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_GIF - if (stbi__gif_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_BMP - if (stbi__bmp_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PSD - if (stbi__psd_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PIC - if (stbi__pic_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_PNM - if (stbi__pnm_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; + #endif - #ifndef STBI_NO_HDR - if (stbi__hdr_info(s, x, y, comp)) return 1; - #endif + #ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; + #endif - // test tga last because it's a crappy test! - #ifndef STBI_NO_TGA - if (stbi__tga_info(s, x, y, comp)) - return 1; - #endif - return stbi__err("unknown image type", "Image not of any known type, or corrupt"); + // test tga last because it's a crappy test! + #ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) return 1; + #endif + return stbi__err("unknown image type", + "Image not of any known type, or corrupt"); } -static int stbi__is_16_main(stbi__context *s) -{ - #ifndef STBI_NO_PNG - if (stbi__png_is16(s)) return 1; - #endif +static int stbi__is_16_main(stbi__context *s) { + #ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; + #endif - #ifndef STBI_NO_PSD - if (stbi__psd_is16(s)) return 1; - #endif + #ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; + #endif - return 0; + return 0; } -#ifndef STBI_NO_STDIO -STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result; - if (!f) return stbi__err("can't fopen", "Unable to open file"); - result = stbi_info_from_file(f, x, y, comp); - fclose(f); - return result; + #ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) { + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; } -STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) -{ - int r; - stbi__context s; - long pos = ftell(f); - stbi__start_file(&s, f); - r = stbi__info_main(&s,x,y,comp); - fseek(f,pos,SEEK_SET); - return r; +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) { + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__info_main(&s, x, y, comp); + fseek(f, pos, SEEK_SET); + return r; } -STBIDEF int stbi_is_16_bit(char const *filename) -{ - FILE *f = stbi__fopen(filename, "rb"); - int result; - if (!f) return stbi__err("can't fopen", "Unable to open file"); - result = stbi_is_16_bit_from_file(f); - fclose(f); - return result; +STBIDEF int stbi_is_16_bit(char const *filename) { + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; } -STBIDEF int stbi_is_16_bit_from_file(FILE *f) -{ - int r; - stbi__context s; - long pos = ftell(f); - stbi__start_file(&s, f); - r = stbi__is_16_main(&s); - fseek(f,pos,SEEK_SET); - return r; +STBIDEF int stbi_is_16_bit_from_file(FILE *f) { + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + fseek(f, pos, SEEK_SET); + return r; } -#endif // !STBI_NO_STDIO + #endif // !STBI_NO_STDIO -STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__info_main(&s,x,y,comp); +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, + int *y, int *comp) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__info_main(&s, x, y, comp); } -STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); - return stbi__info_main(&s,x,y,comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, + int *x, int *y, int *comp) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)c, user); + return stbi__info_main(&s, x, y, comp); } -STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) -{ - stbi__context s; - stbi__start_mem(&s,buffer,len); - return stbi__is_16_main(&s); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) { + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__is_16_main(&s); } -STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user) -{ - stbi__context s; - stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); - return stbi__is_16_main(&s); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, + void *user) { + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)c, user); + return stbi__is_16_main(&s); } -#endif // STB_IMAGE_IMPLEMENTATION +#endif // STB_IMAGE_IMPLEMENTATION /* revision history: - 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs - 2.19 (2018-02-11) fix warning - 2.18 (2018-01-30) fix warnings - 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and + platform ifdefs 2.19 (2018-02-11) fix warning 2.18 (2018-01-30) fix + warnings 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug 1-bit BMP *_is_16_bit api avoid warnings @@ -7555,13 +8079,11 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user warning fixes; disable run-time SSE detection on gcc; uniform handling of optional "return" values; thread-safe initialization of zlib tables - 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs - 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now - 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes - 2.11 (2016-04-02) allocate large structures on the stack - remove white matting for transparent PSD - fix reported channel count for PNG & BMP - re-enable SSE2 in non-gcc 64-bit + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet + JPGs 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now 2.12 + (2016-04-02) fix typo in 2.11 PSD fix that caused crashes 2.11 (2016-04-02) + allocate large structures on the stack remove white matting for transparent + PSD fix reported channel count for PNG & BMP re-enable SSE2 in non-gcc 64-bit support RGB-formatted JPEG read 16-bit PNGs (only as 8-bit) 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED @@ -7569,11 +8091,9 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 16-bit-per-pixel TGA (not bit-per-component) info() for TGA could break due to .hdr handling info() for BMP to shares code instead of sloppy parse - can use STBI_REALLOC_SIZED if allocator doesn't support realloc - code cleanup - 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA - 2.07 (2015-09-13) fix compiler warnings - partial animated GIF support + can use STBI_REALLOC_SIZED if allocator doesn't support + realloc code cleanup 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD + as RGBA 2.07 (2015-09-13) fix compiler warnings partial animated GIF support limited 16-bpc PSD support #ifdef unused functions bug with < 92 byte PIC,PNM,HDR,TGA @@ -7584,23 +8104,18 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user stbi_set_flip_vertically_on_load (nguillemot) fix NEON support; fix mingw support 2.02 (2015-01-19) fix incorrect assert, fix warning - 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 - 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG - 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) - progressive JPEG (stb) - PGM/PPM support (Ken Miller) - STBI_MALLOC,STBI_REALLOC,STBI_FREE + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit + without -msse2 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG 2.00 + (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) progressive + JPEG (stb) PGM/PPM support (Ken Miller) STBI_MALLOC,STBI_REALLOC,STBI_FREE GIF bugfix -- seemingly never worked STBI_NO_*, STBI_ONLY_* 1.48 (2014-12-14) fix incorrectly-named assert() - 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) - optimize PNG (ryg) - fix bug in interlaced PNG with user-specified channel count (stb) - 1.46 (2014-08-26) - fix broken tRNS chunk (colorkey-style transparency) in non-paletted PNG - 1.45 (2014-08-16) - fix MSVC-ARM internal compiler error by wrapping malloc - 1.44 (2014-08-07) + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar + Cornut & stb) optimize PNG (ryg) fix bug in interlaced PNG with + user-specified channel count (stb) 1.46 (2014-08-26) fix broken tRNS chunk + (colorkey-style transparency) in non-paletted PNG 1.45 (2014-08-16) fix + MSVC-ARM internal compiler error by wrapping malloc 1.44 (2014-08-07) various warning fixes from Ronny Chevalier 1.43 (2014-07-15) fix MSVC-only compiler problem in code changed in 1.42 @@ -7609,73 +8124,48 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user fixes to stbi__cleanup_jpeg path added STBI_ASSERT to avoid requiring assert.h 1.41 (2014-06-25) - fix search&replace from 1.36 that messed up comments/error messages - 1.40 (2014-06-22) - fix gcc struct-initialization warning - 1.39 (2014-06-15) - fix to TGA optimization when req_comp != number of components in TGA; - fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my test suite) - add support for BMP version 5 (more ignored fields) - 1.38 (2014-06-06) - suppress MSVC warnings on integer casts truncating values - fix accidental rename of 'skip' field of I/O - 1.37 (2014-06-04) - remove duplicate typedef - 1.36 (2014-06-03) - convert to header file single-file library - if de-iphone isn't set, load iphone images color-swapped instead of returning NULL - 1.35 (2014-05-27) - various warnings - fix broken STBI_SIMD path - fix bug where stbi_load_from_file no longer left file pointer in correct place - fix broken non-easy path for 32-bit BMP (possibly never used) - TGA optimization by Arseny Kapoulkine - 1.34 (unknown) - use STBI_NOTUSED in stbi__resample_row_generic(), fix one more leak in tga failure case - 1.33 (2011-07-14) - make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements - 1.32 (2011-07-13) - support for "info" function for all supported filetypes (SpartanJ) - 1.31 (2011-06-20) - a few more leak fixes, bug in PNG handling (SpartanJ) - 1.30 (2011-06-11) - added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + fix search&replace from 1.36 that messed up comments/error + messages 1.40 (2014-06-22) fix gcc struct-initialization warning 1.39 + (2014-06-15) fix to TGA optimization when req_comp != number of components in + TGA; fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my + test suite) add support for BMP version 5 (more ignored fields) 1.38 + (2014-06-06) suppress MSVC warnings on integer casts truncating values fix + accidental rename of 'skip' field of I/O 1.37 (2014-06-04) remove duplicate + typedef 1.36 (2014-06-03) convert to header file single-file library if + de-iphone isn't set, load iphone images color-swapped instead of returning + NULL 1.35 (2014-05-27) various warnings fix broken STBI_SIMD path fix bug + where stbi_load_from_file no longer left file pointer in correct place fix + broken non-easy path for 32-bit BMP (possibly never used) TGA optimization by + Arseny Kapoulkine 1.34 (unknown) use STBI_NOTUSED in + stbi__resample_row_generic(), fix one more leak in tga failure case 1.33 + (2011-07-14) make stbi_is_hdr work in STBI_NO_HDR (as specified), minor + compiler-friendly improvements 1.32 (2011-07-13) support for "info" function + for all supported filetypes (SpartanJ) 1.31 (2011-06-20) a few more leak + fixes, bug in PNG handling (SpartanJ) 1.30 (2011-06-11) added ability to + load files via callbacks to accomidate custom input streams (Ben Wenger) removed deprecated format-specific test/load functions - removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway - error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) - fix inefficiency in decoding 32-bit BMP (David Woo) - 1.29 (2010-08-16) - various warning fixes from Aurelien Pocheville - 1.28 (2010-08-01) - fix bug in GIF palette transparency (SpartanJ) - 1.27 (2010-08-01) - cast-to-stbi_uc to fix warnings - 1.26 (2010-07-24) - fix bug in file buffering for PNG reported by SpartanJ - 1.25 (2010-07-17) - refix trans_data warning (Won Chun) - 1.24 (2010-07-12) - perf improvements reading from files on platforms with lock-heavy fgetc() - minor perf improvements for jpeg - deprecated type-specific functions so we'll get feedback if they're needed - attempt to fix trans_data warning (Won Chun) - 1.23 fixed bug in iPhone support - 1.22 (2010-07-10) - removed image *writing* support - stbi_info support from Jetro Lauha - GIF support from Jean-Marc Lienher + removed support for installable file formats (stbi_loader) -- + would have been broken for IO callbacks anyway error cases in bmp and tga + give messages and don't leak (Raymond Barbiero, grisha) fix inefficiency in + decoding 32-bit BMP (David Woo) 1.29 (2010-08-16) various warning fixes from + Aurelien Pocheville 1.28 (2010-08-01) fix bug in GIF palette transparency + (SpartanJ) 1.27 (2010-08-01) cast-to-stbi_uc to fix warnings 1.26 + (2010-07-24) fix bug in file buffering for PNG reported by SpartanJ 1.25 + (2010-07-17) refix trans_data warning (Won Chun) 1.24 (2010-07-12) perf + improvements reading from files on platforms with lock-heavy fgetc() minor + perf improvements for jpeg deprecated type-specific functions so we'll get + feedback if they're needed attempt to fix trans_data warning (Won Chun) 1.23 + fixed bug in iPhone support 1.22 (2010-07-10) removed image *writing* + support stbi_info support from Jetro Lauha GIF support from Jean-Marc Lienher iPhone PNG-extensions from James Brown - warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. Janez (U+017D)emva) - 1.21 fix use of 'stbi_uc' in header (reported by jon blow) - 1.20 added support for Softimage PIC, by Tom Seddon - 1.19 bug in interlaced PNG corruption check (found by ryg) - 1.18 (2008-08-02) - fix a threading bug (local mutable static) - 1.17 support interlaced PNG - 1.16 major bugfix - stbi__convert_format converted one too many pixels - 1.15 initialize some fields for thread safety - 1.14 fix threadsafe conversion bug - header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. + Janez (U+017D)emva) 1.21 fix use of 'stbi_uc' in header (reported by jon + blow) 1.20 added support for Softimage PIC, by Tom Seddon 1.19 bug in + interlaced PNG corruption check (found by ryg) 1.18 (2008-08-02) fix a + threading bug (local mutable static) 1.17 support interlaced PNG 1.16 + major bugfix - stbi__convert_format converted one too many pixels 1.15 + initialize some fields for thread safety 1.14 fix threadsafe conversion + bug header-file-only version (#define STBI_HEADER_FILE_ONLY before including) 1.13 threadsafe 1.12 const qualifiers in the API 1.11 Support installable IDCT, colorspace conversion routines @@ -7685,15 +8175,14 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz 1.07 attempt to fix C++ warning/errors again 1.06 attempt to fix C++ warning/errors again - 1.05 fix TGA loading to return correct *comp and use good luminance calc - 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free - 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR - 1.02 support for (subset of) HDR files, float interface for preferred access to them - 1.01 fix bug: possible bug in handling right-side up bmps... not sure - fix bug: the stbi__bmp_load() and stbi__tga_load() functions didn't work at all - 1.00 interface to zlib that skips zlib header - 0.99 correct handling of alpha in palette - 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 1.05 fix TGA loading to return correct *comp and use good luminance + calc 1.04 default float alpha is 1, not 255; use 'void *' for + stbi_image_free 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR 1.02 support + for (subset of) HDR files, float interface for preferred access to them 1.01 + fix bug: possible bug in handling right-side up bmps... not sure fix bug: the + stbi__bmp_load() and stbi__tga_load() functions didn't work at all 1.00 + interface to zlib that skips zlib header 0.99 correct handling of alpha in + palette 0.98 TGA loader by lonesock; dynamically add loaders (untested) 0.97 jpeg errors on too large a file; also catch another malloc failure 0.96 fix detection of invalid v value - particleman@mollyrocket forum 0.95 during header scan, seek to markers in case of padding @@ -7706,8 +8195,8 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user 0.60 fix compiling as c++ 0.59 fix warnings: merge Dave Moore's -Wall fixes 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian - 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available - 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but + less than 16 available 0.56 fix bug: zlib uncompressed mode len vs. nlen 0.55 fix bug: restart_interval not initialized to 0 0.54 allow NULL for 'int *comp' 0.53 fix bug in png 3->4; speedup png decoding @@ -7718,7 +8207,6 @@ STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user first released version */ - /* ------------------------------------------------------------------------------ This software is available under 2 licenses -- choose whichever you prefer. diff --git a/fuzzers/python_qemu/fuzz.c b/fuzzers/python_qemu/fuzz.c index 70afd693f1..5643db8c1b 100644 --- a/fuzzers/python_qemu/fuzz.c +++ b/fuzzers/python_qemu/fuzz.c @@ -3,14 +3,11 @@ #include int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size) { - // printf("Got %ld bytes.\n", Size); - if (Size >= 4 && *(uint32_t*)Data == 0xaabbccdd) - abort(); + // printf("Got %ld bytes.\n", Size); + if (Size >= 4 && *(uint32_t *)Data == 0xaabbccdd) { abort(); } } int main() { - - char buf [10] = {0}; + char buf[10] = {0}; LLVMFuzzerTestOneInput(buf, 10); - } diff --git a/fuzzers/qemu_launcher/harness.cc b/fuzzers/qemu_launcher/harness.cc index 46a1a0ca09..f7e3d03a3e 100644 --- a/fuzzers/qemu_launcher/harness.cc +++ b/fuzzers/qemu_launcher/harness.cc @@ -23,40 +23,38 @@ #define PNG_INTERNAL #include "png.h" -#define PNG_CLEANUP \ - if(png_handler.png_ptr) \ - { \ - if (png_handler.row_ptr) \ - png_free(png_handler.png_ptr, png_handler.row_ptr); \ - if (png_handler.end_info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - &png_handler.end_info_ptr); \ - else if (png_handler.info_ptr) \ - png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr,\ - nullptr); \ - else \ - png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ - png_handler.png_ptr = nullptr; \ - png_handler.row_ptr = nullptr; \ - png_handler.info_ptr = nullptr; \ - png_handler.end_info_ptr = nullptr; \ +#define PNG_CLEANUP \ + if (png_handler.png_ptr) { \ + if (png_handler.row_ptr) \ + png_free(png_handler.png_ptr, png_handler.row_ptr); \ + if (png_handler.end_info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + &png_handler.end_info_ptr); \ + else if (png_handler.info_ptr) \ + png_destroy_read_struct(&png_handler.png_ptr, &png_handler.info_ptr, \ + nullptr); \ + else \ + png_destroy_read_struct(&png_handler.png_ptr, nullptr, nullptr); \ + png_handler.png_ptr = nullptr; \ + png_handler.row_ptr = nullptr; \ + png_handler.info_ptr = nullptr; \ + png_handler.end_info_ptr = nullptr; \ } struct BufState { - const uint8_t* data; - size_t bytes_left; + const uint8_t *data; + size_t bytes_left; }; struct PngObjectHandler { - png_infop info_ptr = nullptr; + png_infop info_ptr = nullptr; png_structp png_ptr = nullptr; - png_infop end_info_ptr = nullptr; - png_voidp row_ptr = nullptr; - BufState* buf_state = nullptr; + png_infop end_info_ptr = nullptr; + png_voidp row_ptr = nullptr; + BufState *buf_state = nullptr; ~PngObjectHandler() { - if (row_ptr) - png_free(png_ptr, row_ptr); + if (row_ptr) { png_free(png_ptr, row_ptr); } if (end_info_ptr) png_destroy_read_struct(&png_ptr, &info_ptr, &end_info_ptr); else if (info_ptr) @@ -68,10 +66,8 @@ struct PngObjectHandler { }; void user_read_data(png_structp png_ptr, png_bytep data, size_t length) { - BufState* buf_state = static_cast(png_get_io_ptr(png_ptr)); - if (length > buf_state->bytes_left) { - png_error(png_ptr, "read error"); - } + BufState *buf_state = static_cast(png_get_io_ptr(png_ptr)); + if (length > buf_state->bytes_left) { png_error(png_ptr, "read error"); } memcpy(data, buf_state->data, length); buf_state->bytes_left -= length; buf_state->data += length; @@ -82,10 +78,8 @@ static const int kPngHeaderSize = 8; // Entry point for LibFuzzer. // Roughly follows the libpng book example: // http://www.libpng.org/pub/png/book/chapter13.html -extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { - if (size < kPngHeaderSize) { - return 0; - } +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { + if (size < kPngHeaderSize) { return 0; } std::vector v(data, data + size); if (png_sig_cmp(v.data(), 0, kPngHeaderSize)) { @@ -99,11 +93,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_handler.info_ptr = nullptr; png_handler.end_info_ptr = nullptr; - png_handler.png_ptr = png_create_read_struct - (PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); - if (!png_handler.png_ptr) { - return 0; - } + png_handler.png_ptr = + png_create_read_struct(PNG_LIBPNG_VER_STRING, nullptr, nullptr, nullptr); + if (!png_handler.png_ptr) { return 0; } png_handler.info_ptr = png_create_info_struct(png_handler.png_ptr); if (!png_handler.info_ptr) { @@ -144,12 +136,12 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { } png_uint_32 width, height; - int bit_depth, color_type, interlace_type, compression_type; - int filter_type; + int bit_depth, color_type, interlace_type, compression_type; + int filter_type; - if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, - &height, &bit_depth, &color_type, &interlace_type, - &compression_type, &filter_type)) { + if (!png_get_IHDR(png_handler.png_ptr, png_handler.info_ptr, &width, &height, + &bit_depth, &color_type, &interlace_type, &compression_type, + &filter_type)) { PNG_CLEANUP return 0; } @@ -158,11 +150,11 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { if (width && height > 100000000 / width) { PNG_CLEANUP #ifdef HAS_DUMMY_CRASH - #ifdef __aarch64__ - asm volatile (".word 0xf7f0a000\n"); - #else - asm("ud2"); - #endif + #ifdef __aarch64__ + asm volatile(".word 0xf7f0a000\n"); + #else + asm("ud2"); + #endif #endif return 0; } @@ -178,9 +170,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { png_read_update_info(png_handler.png_ptr, png_handler.info_ptr); - png_handler.row_ptr = png_malloc( - png_handler.png_ptr, png_get_rowbytes(png_handler.png_ptr, - png_handler.info_ptr)); + png_handler.row_ptr = + png_malloc(png_handler.png_ptr, + png_get_rowbytes(png_handler.png_ptr, png_handler.info_ptr)); for (int pass = 0; pass < passes; ++pass) { for (png_uint_32 y = 0; y < height; ++y) { @@ -196,8 +188,6 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) { } int main() { - - uint8_t buf [10] = {0}; + uint8_t buf[10] = {0}; LLVMFuzzerTestOneInput(buf, 10); - } diff --git a/libafl_cc/src/afl-coverage-pass.cc b/libafl_cc/src/afl-coverage-pass.cc index 4960c6e34c..92bc44cacc 100644 --- a/libafl_cc/src/afl-coverage-pass.cc +++ b/libafl_cc/src/afl-coverage-pass.cc @@ -173,10 +173,11 @@ PreservedAnalyses AFLCoverage::run(Module &M, ModuleAnalysisManager &MAM) { #else bool AFLCoverage::runOnModule(Module &M) { #endif - if (Ctx && DumpCFG) + if (Ctx && DumpCFG) { FATAL( "Does not support dumping CFG with full context sensitive coverage " "enabled."); + } LLVMContext &C = M.getContext(); IntegerType *Int8Ty = IntegerType::getInt8Ty(C); @@ -201,9 +202,10 @@ bool AFLCoverage::runOnModule(Module &M) { if ((ptr = getenv("AFL_MAP_SIZE")) || (ptr = getenv("AFL_MAPSIZE"))) { map_size = atoi(ptr); - if (map_size < 8 || map_size > (1 << 29)) + if (map_size < 8 || map_size > (1 << 29)) { FATAL("illegal AFL_MAP_SIZE %u, must be between 2^3 and 2^30", - map_size); if (map_size % 8) map_size = (((map_size >> 3) + 1) << 3); + } + map_size); if (map_size % 8) {map_size = (((map_size >> 3) + 1) << 3);} } @@ -211,8 +213,9 @@ bool AFLCoverage::runOnModule(Module &M) { /* Decide instrumentation ratio */ - if (!InstRatio || InstRatio > 100) + if (!InstRatio || InstRatio > 100) { FATAL("Bad value of the instrumentation ratio (must be between 1 and 100)"); + } unsigned PrevLocSize = 0; unsigned PrevCallerSize = 0; @@ -224,25 +227,28 @@ bool AFLCoverage::runOnModule(Module &M) { /* Decide previous location vector size (must be a power of two) */ VectorType *PrevLocTy = NULL; - if (Ngram && (Ngram < 2 || Ngram > NGRAM_SIZE_MAX)) + if (Ngram && (Ngram < 2 || Ngram > NGRAM_SIZE_MAX)) { FATAL( "Bad value of the Ngram size (must be between 2 and NGRAM_SIZE_MAX " "(%u))", NGRAM_SIZE_MAX); + } - if (Ngram) + if (Ngram) { PrevLocSize = Ngram - 1; - else + } else { PrevLocSize = 1; + } /* Decide K-Ctx vector size (must be a power of two) */ VectorType *PrevCallerTy = NULL; - if (CtxK > CTX_MAX_K) + if (CtxK > CTX_MAX_K) { FATAL( "Bad value of K for K-context sensitivity (must be between 1 and " "CTX_MAX_K (%u))", CTX_MAX_K); + } if (CtxK == 1) { CtxK = 0; @@ -381,11 +387,13 @@ bool AFLCoverage::runOnModule(Module &M) { SmallVector PrevLocShuffle = {UndefValue::get(Int32Ty)}; - for (unsigned I = 0; I < PrevLocSize - 1; ++I) + for (unsigned I = 0; I < PrevLocSize - 1; ++I) { PrevLocShuffle.push_back(ConstantInt::get(Int32Ty, I)); + } - for (int I = PrevLocSize; I < PrevLocVecSize; ++I) + for (int I = PrevLocSize; I < PrevLocVecSize; ++I) { PrevLocShuffle.push_back(ConstantInt::get(Int32Ty, PrevLocSize)); + } Constant *PrevLocShuffleMask = ConstantVector::get(PrevLocShuffle); @@ -393,11 +401,13 @@ bool AFLCoverage::runOnModule(Module &M) { SmallVector PrevCallerShuffle = {UndefValue::get(Int32Ty)}; if (CtxK) { - for (unsigned I = 0; I < PrevCallerSize - 1; ++I) + for (unsigned I = 0; I < PrevCallerSize - 1; ++I) { PrevCallerShuffle.push_back(ConstantInt::get(Int32Ty, I)); + } - for (int I = PrevCallerSize; I < PrevCallerVecSize; ++I) + for (int I = PrevCallerSize; I < PrevCallerVecSize; ++I) { PrevCallerShuffle.push_back(ConstantInt::get(Int32Ty, PrevCallerSize)); + } PrevCallerShuffleMask = ConstantVector::get(PrevCallerShuffle); } @@ -424,7 +434,7 @@ bool AFLCoverage::runOnModule(Module &M) { // if (!isInInstrumentList(&F)) { continue; } if (F.size() < function_minimum_size) { continue; } - if (DumpCFG) entry_bb[F.getName()] = &F.getEntryBlock(); + if (DumpCFG) { entry_bb[F.getName()] = &F.getEntryBlock(); } std::list todo; for (auto &BB : F) { @@ -453,10 +463,9 @@ bool AFLCoverage::runOnModule(Module &M) { // local variable on the stack LoadInst *PrevCtxLoad = IRB.CreateLoad( #if LLVM_VERSION_MAJOR >= 14 - IRB.getInt32Ty(), + IRB.getInt32Ty(), #endif - AFLContext - ); + AFLContext); PrevCtxLoad->setMetadata(M.getMDKindID("nosanitize"), MDNode::get(C, None)); PrevCtx = PrevCtxLoad; @@ -465,7 +474,7 @@ bool AFLCoverage::runOnModule(Module &M) { // does the function have calls? and is any of the calls larger than one // basic block? for (auto &BB_2 : F) { - if (has_calls) break; + if (has_calls) { break; } for (auto &IN : BB_2) { CallInst *callInst = nullptr; if ((callInst = dyn_cast(&IN))) { @@ -500,7 +509,7 @@ bool AFLCoverage::runOnModule(Module &M) { #endif { - if (Ctx) NewCtx = IRB.CreateXor(PrevCtx, NewCtx); + if (Ctx) { NewCtx = IRB.CreateXor(PrevCtx, NewCtx); } StoreInst *StoreCtx = IRB.CreateStore(NewCtx, AFLContext); StoreCtx->setMetadata(M.getMDKindID("nosanitize"), MDNode::get(C, None)); @@ -508,13 +517,13 @@ bool AFLCoverage::runOnModule(Module &M) { } } - if (RandBelow(100) >= InstRatio) continue; + if (RandBelow(100) >= InstRatio) { continue; } /* Make up cur_loc */ // cur_loc++; cur_loc = RandBelow(map_size); - if (DumpCFG) bb_to_cur_loc[&BB] = cur_loc; + if (DumpCFG) { bb_to_cur_loc[&BB] = cur_loc; } /* There is a problem with Ubuntu 18.04 and llvm 6.0 (see issue #63). The inline function successors() is not inlined and also not found at runtime :-( As I am unable to detect Ubuntu18.04 heree, the next best thing is to @@ -531,7 +540,7 @@ bool AFLCoverage::runOnModule(Module &M) { BasicBlock *Pred = *PI; int count = 0; - if (more_than_one == -1) more_than_one = 0; + if (more_than_one == -1) { more_than_one = 0; } // fprintf(stderr, " %p=>", Pred); for (succ_iterator SI = succ_begin(Pred), E = succ_end(Pred); SI != E; @@ -540,11 +549,11 @@ bool AFLCoverage::runOnModule(Module &M) { // if (count > 0) // fprintf(stderr, "|"); - if (Succ != NULL) count++; + if (Succ != NULL) { count++; } // fprintf(stderr, "%p", Succ); } - if (count > 1) more_than_one = 1; + if (count > 1) { more_than_one = 1; } } // fprintf(stderr, " == %d\n", more_than_one); @@ -783,7 +792,7 @@ bool AFLCoverage::runOnModule(Module &M) { cfg += formatv("->{0}\n", bb_to_cur_loc[*bb_successor]).str(); } } - if (Debug) errs() << "CFG: \n" << cfg; + if (Debug) { errs() << "CFG: \n" << cfg; } if (cfg.size() > 0 && write(fd, cfg.c_str(), cfg.length()) <= 0) FATAL("Failed to dump CFG.\n"); } diff --git a/libafl_cc/src/autotokens-pass.cc b/libafl_cc/src/autotokens-pass.cc index 5681eeac3c..b42ab47762 100644 --- a/libafl_cc/src/autotokens-pass.cc +++ b/libafl_cc/src/autotokens-pass.cc @@ -210,8 +210,9 @@ void dict2file(int fd, uint8_t *mem, uint32_t len) { line[j] = 0; strcat(line, "\"\n"); - if (write(fd, line, strlen(line)) <= 0) + if (write(fd, line, strlen(line)) <= 0) { FATAL("Could not write to the dictionary file"); + } fsync(fd); } @@ -248,7 +249,7 @@ bool AutoTokensPass::runOnModule(Module &M) { /* Instrument all the things! */ for (auto &F : M) { - if (isIgnoreFunction(&F)) continue; + if (isIgnoreFunction(&F)) { continue; } /* Some implementation notes. * @@ -294,8 +295,8 @@ bool AutoTokensPass::runOnModule(Module &M) { if (ilen && ilen->uge(0xffffffffffffffff) == false) { uint64_t val2 = 0, val = ilen->getZExtValue(); uint32_t len = 0; - if (val > 0x10000 && val < 0xffffffff) len = 4; - if (val > 0x100000001 && val < 0xffffffffffffffff) len = 8; + if (val > 0x10000 && val < 0xffffffff) { len = 4; } + if (val > 0x100000001 && val < 0xffffffffffffffff) { len = 8; } if (len) { auto c = cmpInst->getPredicate(); @@ -309,7 +310,7 @@ bool AutoTokensPass::runOnModule(Module &M) { // signed comparison and it is a negative constant if ((len == 4 && (val & 80000000)) || (len == 8 && (val & 8000000000000000))) { - if ((val & 0xffff) != 1) val2 = val - 1; + if ((val & 0xffff) != 1) { val2 = val - 1; } break; } diff --git a/libafl_cc/src/cmplog-routines-pass.cc b/libafl_cc/src/cmplog-routines-pass.cc index 341eb960d5..4ddd4f3a37 100644 --- a/libafl_cc/src/cmplog-routines-pass.cc +++ b/libafl_cc/src/cmplog-routines-pass.cc @@ -270,7 +270,7 @@ bool CmpLogRoutines::hookRtns(Module &M) { /* iterate over all functions, bbs and instruction and add suitable calls */ for (auto &F : M) { - if (isIgnoreFunction(&F)) continue; + if (isIgnoreFunction(&F)) { continue; } for (auto &BB : F) { for (auto &IN : BB) { @@ -278,8 +278,8 @@ bool CmpLogRoutines::hookRtns(Module &M) { if ((callInst = dyn_cast(&IN))) { Function *Callee = callInst->getCalledFunction(); - if (!Callee) continue; - if (callInst->getCallingConv() != llvm::CallingConv::C) continue; + if (!Callee) { continue; } + if (callInst->getCallingConv() != llvm::CallingConv::C) { continue; } FunctionType *FT = Callee->getFunctionType(); @@ -350,8 +350,9 @@ bool CmpLogRoutines::hookRtns(Module &M) { } if (!calls.size() && !gccStdStd.size() && !gccStdC.size() && - !llvmStdStd.size() && !llvmStdC.size()) + !llvmStdStd.size() && !llvmStdC.size()) { return false; + } for (auto &callInst : calls) { Value *v1P = callInst->getArgOperand(0), *v2P = callInst->getArgOperand(1); diff --git a/libafl_cc/src/common-llvm.h b/libafl_cc/src/common-llvm.h index b2bc5fd206..26134e1c58 100644 --- a/libafl_cc/src/common-llvm.h +++ b/libafl_cc/src/common-llvm.h @@ -20,18 +20,22 @@ typedef long double max_align_t; #endif #ifdef USE_NEW_PM -#include "llvm/Passes/PassPlugin.h" -#include "llvm/Passes/PassBuilder.h" -#include "llvm/IR/PassManager.h" + #include "llvm/Passes/PassPlugin.h" + #include "llvm/Passes/PassBuilder.h" + #include "llvm/IR/PassManager.h" #else -#include "llvm/IR/LegacyPassManager.h" -#include "llvm/Transforms/IPO/PassManagerBuilder.h" + #include "llvm/IR/LegacyPassManager.h" + #include "llvm/Transforms/IPO/PassManagerBuilder.h" #endif -#define FATAL(...) do { fprintf(stderr, "FATAL: " __VA_ARGS__); exit(1); } while (0) +#define FATAL(...) \ + do { \ + fprintf(stderr, "FATAL: " __VA_ARGS__); \ + exit(1); \ + } while (0) static uint32_t RandBelow(uint32_t max) { - return (uint32_t)rand() % (max +1); + return (uint32_t)rand() % (max + 1); } /* needed up to 3.9.0 */ @@ -39,7 +43,6 @@ static uint32_t RandBelow(uint32_t max) { (LLVM_VERSION_MINOR < 9 || \ (LLVM_VERSION_MINOR == 9 && LLVM_VERSION_PATCH < 1)) static uint64_t PowerOf2Ceil(unsigned in) { - uint64_t in64 = in - 1; in64 |= (in64 >> 1); in64 |= (in64 >> 2); @@ -48,8 +51,7 @@ static uint64_t PowerOf2Ceil(unsigned in) { in64 |= (in64 >> 16); in64 |= (in64 >> 32); return in64 + 1; - } #endif -#endif // LIBAFL_COMMON_LLVM_H +#endif // LIBAFL_COMMON_LLVM_H diff --git a/libafl_cc/src/coverage-accounting-pass.cc b/libafl_cc/src/coverage-accounting-pass.cc index 1ffa2af94d..3ba16e0997 100644 --- a/libafl_cc/src/coverage-accounting-pass.cc +++ b/libafl_cc/src/coverage-accounting-pass.cc @@ -89,7 +89,7 @@ namespace { SECURITY_SENSITIVE_FUNCS(StringRef) bool isSecuritySensitiveFunction(Function *F) { - if (!F) return 0; + if (!F) { return 0; } auto func_name = F->getName(); for (auto name : securitySensitiveFunctions) { if (func_name.contains(name)) { @@ -231,21 +231,23 @@ bool AFLCoverage::runOnModule(Module &M) { BasicBlock::iterator IP = BB.getFirstInsertionPt(); IRBuilder<> IRB(&(*IP)); - if (RandBelow(100) >= InstRatio) continue; + if (RandBelow(100) >= InstRatio) { continue; } // Start with 1 to implicitly track edge coverage too uint32_t MemCnt = 1; for (auto &I : BB) { switch (granularity) { - case BB_GRAN: - if (I.mayReadFromMemory() || I.mayWriteToMemory()) ++MemCnt; + case BB_GRAN: { + if (I.mayReadFromMemory() || I.mayWriteToMemory()) { ++MemCnt; } break; - case FUNC_GRAN: + } + case FUNC_GRAN: { if (auto *C = dyn_cast(&I)) { auto F = C->getCalledFunction(); MemCnt += isSecuritySensitiveFunction(F); } break; + } } } /* Make up cur_loc */ diff --git a/libafl_cc/src/no-link-rt.c b/libafl_cc/src/no-link-rt.c index 4125fcd47e..08c069bd45 100644 --- a/libafl_cc/src/no-link-rt.c +++ b/libafl_cc/src/no-link-rt.c @@ -1,36 +1,39 @@ #include -uint8_t* __afl_area_ptr; -uint8_t* __afl_acc_memop_ptr; +uint8_t *__afl_area_ptr; +uint8_t *__afl_acc_memop_ptr; -void __libafl_targets_cmplog_instructions(uintptr_t k, uint8_t shape, uint64_t arg1, uint64_t arg2) { - (void)k; - (void)shape; - (void)arg1; - (void)arg2; +void __libafl_targets_cmplog_instructions(uintptr_t k, uint8_t shape, + uint64_t arg1, uint64_t arg2) { + (void)k; + (void)shape; + (void)arg1; + (void)arg2; } void __cmplog_rtn_hook(uint8_t *ptr1, uint8_t *ptr2) { - (void)ptr1; - (void)ptr2; + (void)ptr1; + (void)ptr2; } void __cmplog_rtn_gcc_stdstring_cstring(uint8_t *stdstring, uint8_t *cstring) { - (void)stdstring; - (void)cstring; + (void)stdstring; + (void)cstring; } -void __cmplog_rtn_gcc_stdstring_stdstring(uint8_t *stdstring1, uint8_t *stdstring2) { - (void)stdstring1; - (void)stdstring2; +void __cmplog_rtn_gcc_stdstring_stdstring(uint8_t *stdstring1, + uint8_t *stdstring2) { + (void)stdstring1; + (void)stdstring2; } void __cmplog_rtn_llvm_stdstring_cstring(uint8_t *stdstring, uint8_t *cstring) { - (void)stdstring; - (void)cstring; + (void)stdstring; + (void)cstring; } -void __cmplog_rtn_llvm_stdstring_stdstring(uint8_t *stdstring1, uint8_t *stdstring2) { - (void)stdstring1; - (void)stdstring2; +void __cmplog_rtn_llvm_stdstring_stdstring(uint8_t *stdstring1, + uint8_t *stdstring2) { + (void)stdstring1; + (void)stdstring2; } diff --git a/libafl_frida/src/gettls.c b/libafl_frida/src/gettls.c index ddf30b7de7..5fe8ca9cf5 100644 --- a/libafl_frida/src/gettls.c +++ b/libafl_frida/src/gettls.c @@ -1,9 +1,9 @@ #ifdef _MSC_VER -__declspec( thread ) int i = 0; +__declspec(thread) int i = 0; #else __thread int i = 0; #endif -void * tls_ptr() { - return (void*)&i; +void *tls_ptr() { + return (void *)&i; } diff --git a/libafl_qemu/libqasan/dlmalloc.c b/libafl_qemu/libqasan/dlmalloc.c index aff58ad57c..878c99c957 100644 --- a/libafl_qemu/libqasan/dlmalloc.c +++ b/libafl_qemu/libqasan/dlmalloc.c @@ -533,7 +533,7 @@ MAX_RELEASE_CHECK_RATE default: 4095 unless not HAVE_MMAP /* Version identifier to allow people to support multiple versions */ #ifndef DLMALLOC_VERSION #define DLMALLOC_VERSION 20806 - #endif /* DLMALLOC_VERSION */ + #endif /* DLMALLOC_VERSION */ #ifndef DLMALLOC_EXPORT #define DLMALLOC_EXPORT extern @@ -542,12 +542,12 @@ MAX_RELEASE_CHECK_RATE default: 4095 unless not HAVE_MMAP #ifndef WIN32 #ifdef _WIN32 #define WIN32 1 - #endif /* _WIN32 */ + #endif /* _WIN32 */ #ifdef _WIN32_WCE #define LACKS_FCNTL_H #define WIN32 1 - #endif /* _WIN32_WCE */ - #endif /* WIN32 */ + #endif /* _WIN32_WCE */ + #endif /* WIN32 */ #ifdef WIN32 #define WIN32_LEAN_AND_MEAN #include @@ -564,15 +564,15 @@ MAX_RELEASE_CHECK_RATE default: 4095 unless not HAVE_MMAP #define LACKS_SCHED_H #ifndef MALLOC_FAILURE_ACTION #define MALLOC_FAILURE_ACTION - #endif /* MALLOC_FAILURE_ACTION */ + #endif /* MALLOC_FAILURE_ACTION */ #ifndef MMAP_CLEARS - #ifdef _WIN32_WCE /* WINCE reportedly does not clear */ + #ifdef _WIN32_WCE /* WINCE reportedly does not clear */ #define MMAP_CLEARS 0 #else #define MMAP_CLEARS 1 - #endif /* _WIN32_WCE */ - #endif /*MMAP_CLEARS */ - #endif /* WIN32 */ + #endif /* _WIN32_WCE */ + #endif /*MMAP_CLEARS */ + #endif /* WIN32 */ #if defined(DARWIN) || defined(_DARWIN) /* Mac OSX docs advise not to use sbrk; it seems better to use mmap */ @@ -583,21 +583,21 @@ MAX_RELEASE_CHECK_RATE default: 4095 unless not HAVE_MMAP #ifndef MALLOC_ALIGNMENT #define MALLOC_ALIGNMENT ((size_t)16U) #endif - #endif /* HAVE_MORECORE */ - #endif /* DARWIN */ + #endif /* HAVE_MORECORE */ + #endif /* DARWIN */ #ifndef LACKS_SYS_TYPES_H - #include /* For size_t */ - #endif /* LACKS_SYS_TYPES_H */ + #include /* For size_t */ + #endif /* LACKS_SYS_TYPES_H */ /* The maximum possible size_t value has all bits set */ #define MAX_SIZE_T (~(size_t)0) - #ifndef USE_LOCKS /* ensure true if spin or recursive locks set */ + #ifndef USE_LOCKS /* ensure true if spin or recursive locks set */ #define USE_LOCKS \ ((defined(USE_SPIN_LOCKS) && USE_SPIN_LOCKS != 0) || \ (defined(USE_RECURSIVE_LOCKS) && USE_RECURSIVE_LOCKS != 0)) - #endif /* USE_LOCKS */ + #endif /* USE_LOCKS */ #if USE_LOCKS /* Spin locks for gcc >= 4.1, older gcc on x86, MSC >= 1310 */ #if ((defined(__GNUC__) && \ @@ -606,125 +606,125 @@ MAX_RELEASE_CHECK_RATE default: 4095 unless not HAVE_MMAP (defined(_MSC_VER) && _MSC_VER >= 1310)) #ifndef USE_SPIN_LOCKS #define USE_SPIN_LOCKS 1 - #endif /* USE_SPIN_LOCKS */ + #endif /* USE_SPIN_LOCKS */ #elif USE_SPIN_LOCKS #error "USE_SPIN_LOCKS defined without implementation" - #endif /* ... locks available... */ + #endif /* ... locks available... */ #elif !defined(USE_SPIN_LOCKS) #define USE_SPIN_LOCKS 0 - #endif /* USE_LOCKS */ + #endif /* USE_LOCKS */ #ifndef ONLY_MSPACES #define ONLY_MSPACES 0 - #endif /* ONLY_MSPACES */ + #endif /* ONLY_MSPACES */ #ifndef MSPACES #if ONLY_MSPACES #define MSPACES 1 - #else /* ONLY_MSPACES */ + #else /* ONLY_MSPACES */ #define MSPACES 0 - #endif /* ONLY_MSPACES */ - #endif /* MSPACES */ + #endif /* ONLY_MSPACES */ + #endif /* MSPACES */ #ifndef MALLOC_ALIGNMENT #define MALLOC_ALIGNMENT ((size_t)(2 * sizeof(void *))) - #endif /* MALLOC_ALIGNMENT */ + #endif /* MALLOC_ALIGNMENT */ #ifndef FOOTERS #define FOOTERS 0 - #endif /* FOOTERS */ + #endif /* FOOTERS */ #ifndef ABORT #define ABORT abort() - #endif /* ABORT */ + #endif /* ABORT */ #ifndef ABORT_ON_ASSERT_FAILURE #define ABORT_ON_ASSERT_FAILURE 1 - #endif /* ABORT_ON_ASSERT_FAILURE */ + #endif /* ABORT_ON_ASSERT_FAILURE */ #ifndef PROCEED_ON_ERROR #define PROCEED_ON_ERROR 0 - #endif /* PROCEED_ON_ERROR */ + #endif /* PROCEED_ON_ERROR */ #ifndef INSECURE #define INSECURE 0 - #endif /* INSECURE */ + #endif /* INSECURE */ #ifndef MALLOC_INSPECT_ALL #define MALLOC_INSPECT_ALL 0 - #endif /* MALLOC_INSPECT_ALL */ + #endif /* MALLOC_INSPECT_ALL */ #ifndef HAVE_MMAP #define HAVE_MMAP 1 - #endif /* HAVE_MMAP */ + #endif /* HAVE_MMAP */ #ifndef MMAP_CLEARS #define MMAP_CLEARS 1 - #endif /* MMAP_CLEARS */ + #endif /* MMAP_CLEARS */ #ifndef HAVE_MREMAP #ifdef linux #define HAVE_MREMAP 1 - #define _GNU_SOURCE /* Turns on mremap() definition */ - #else /* linux */ + #define _GNU_SOURCE /* Turns on mremap() definition */ + #else /* linux */ #define HAVE_MREMAP 0 - #endif /* linux */ - #endif /* HAVE_MREMAP */ + #endif /* linux */ + #endif /* HAVE_MREMAP */ #ifndef MALLOC_FAILURE_ACTION #define MALLOC_FAILURE_ACTION errno = ENOMEM; - #endif /* MALLOC_FAILURE_ACTION */ + #endif /* MALLOC_FAILURE_ACTION */ #ifndef HAVE_MORECORE #if ONLY_MSPACES #define HAVE_MORECORE 0 - #else /* ONLY_MSPACES */ + #else /* ONLY_MSPACES */ #define HAVE_MORECORE 1 - #endif /* ONLY_MSPACES */ - #endif /* HAVE_MORECORE */ + #endif /* ONLY_MSPACES */ + #endif /* HAVE_MORECORE */ #if !HAVE_MORECORE #define MORECORE_CONTIGUOUS 0 - #else /* !HAVE_MORECORE */ + #else /* !HAVE_MORECORE */ #define MORECORE_DEFAULT sbrk #ifndef MORECORE_CONTIGUOUS #define MORECORE_CONTIGUOUS 1 - #endif /* MORECORE_CONTIGUOUS */ - #endif /* HAVE_MORECORE */ + #endif /* MORECORE_CONTIGUOUS */ + #endif /* HAVE_MORECORE */ #ifndef DEFAULT_GRANULARITY #if (MORECORE_CONTIGUOUS || defined(WIN32)) #define DEFAULT_GRANULARITY (0) /* 0 means to compute in init_mparams */ - #else /* MORECORE_CONTIGUOUS */ + #else /* MORECORE_CONTIGUOUS */ #define DEFAULT_GRANULARITY ((size_t)64U * (size_t)1024U) - #endif /* MORECORE_CONTIGUOUS */ - #endif /* DEFAULT_GRANULARITY */ + #endif /* MORECORE_CONTIGUOUS */ + #endif /* DEFAULT_GRANULARITY */ #ifndef DEFAULT_TRIM_THRESHOLD #ifndef MORECORE_CANNOT_TRIM #define DEFAULT_TRIM_THRESHOLD \ ((size_t)2U * (size_t)1024U * (size_t)1024U) - #else /* MORECORE_CANNOT_TRIM */ + #else /* MORECORE_CANNOT_TRIM */ #define DEFAULT_TRIM_THRESHOLD MAX_SIZE_T - #endif /* MORECORE_CANNOT_TRIM */ - #endif /* DEFAULT_TRIM_THRESHOLD */ + #endif /* MORECORE_CANNOT_TRIM */ + #endif /* DEFAULT_TRIM_THRESHOLD */ #ifndef DEFAULT_MMAP_THRESHOLD #if HAVE_MMAP #define DEFAULT_MMAP_THRESHOLD ((size_t)256U * (size_t)1024U) - #else /* HAVE_MMAP */ + #else /* HAVE_MMAP */ #define DEFAULT_MMAP_THRESHOLD MAX_SIZE_T - #endif /* HAVE_MMAP */ - #endif /* DEFAULT_MMAP_THRESHOLD */ + #endif /* HAVE_MMAP */ + #endif /* DEFAULT_MMAP_THRESHOLD */ #ifndef MAX_RELEASE_CHECK_RATE #if HAVE_MMAP #define MAX_RELEASE_CHECK_RATE 4095 #else #define MAX_RELEASE_CHECK_RATE MAX_SIZE_T - #endif /* HAVE_MMAP */ - #endif /* MAX_RELEASE_CHECK_RATE */ + #endif /* HAVE_MMAP */ + #endif /* MAX_RELEASE_CHECK_RATE */ #ifndef USE_BUILTIN_FFS #define USE_BUILTIN_FFS 0 - #endif /* USE_BUILTIN_FFS */ + #endif /* USE_BUILTIN_FFS */ #ifndef USE_DEV_RANDOM #define USE_DEV_RANDOM 0 - #endif /* USE_DEV_RANDOM */ + #endif /* USE_DEV_RANDOM */ #ifndef NO_MALLINFO #define NO_MALLINFO 0 - #endif /* NO_MALLINFO */ + #endif /* NO_MALLINFO */ #ifndef MALLINFO_FIELD_TYPE #define MALLINFO_FIELD_TYPE size_t - #endif /* MALLINFO_FIELD_TYPE */ + #endif /* MALLINFO_FIELD_TYPE */ #ifndef NO_MALLOC_STATS #define NO_MALLOC_STATS 0 - #endif /* NO_MALLOC_STATS */ + #endif /* NO_MALLOC_STATS */ #ifndef NO_SEGMENT_TRAVERSAL #define NO_SEGMENT_TRAVERSAL 0 - #endif /* NO_SEGMENT_TRAVERSAL */ + #endif /* NO_SEGMENT_TRAVERSAL */ /* mallopt tuning options. SVID/XPG defines four standard parameter @@ -769,30 +769,28 @@ MAX_RELEASE_CHECK_RATE default: 4095 unless not HAVE_MMAP #ifdef HAVE_USR_INCLUDE_MALLOC_H #include "/usr/include/malloc.h" - #else /* HAVE_USR_INCLUDE_MALLOC_H */ + #else /* HAVE_USR_INCLUDE_MALLOC_H */ #ifndef STRUCT_MALLINFO_DECLARED /* HP-UX (and others?) redefines mallinfo unless _STRUCT_MALLINFO is * defined */ #define _STRUCT_MALLINFO #define STRUCT_MALLINFO_DECLARED 1 struct mallinfo { - - MALLINFO_FIELD_TYPE arena; /* non-mmapped space allocated from system */ - MALLINFO_FIELD_TYPE ordblks; /* number of free chunks */ - MALLINFO_FIELD_TYPE smblks; /* always 0 */ - MALLINFO_FIELD_TYPE hblks; /* always 0 */ - MALLINFO_FIELD_TYPE hblkhd; /* space in mmapped regions */ - MALLINFO_FIELD_TYPE usmblks; /* maximum total allocated space */ - MALLINFO_FIELD_TYPE fsmblks; /* always 0 */ - MALLINFO_FIELD_TYPE uordblks; /* total allocated space */ - MALLINFO_FIELD_TYPE fordblks; /* total free space */ - MALLINFO_FIELD_TYPE keepcost; /* releasable (via malloc_trim) space */ - + MALLINFO_FIELD_TYPE arena; /* non-mmapped space allocated from system */ + MALLINFO_FIELD_TYPE ordblks; /* number of free chunks */ + MALLINFO_FIELD_TYPE smblks; /* always 0 */ + MALLINFO_FIELD_TYPE hblks; /* always 0 */ + MALLINFO_FIELD_TYPE hblkhd; /* space in mmapped regions */ + MALLINFO_FIELD_TYPE usmblks; /* maximum total allocated space */ + MALLINFO_FIELD_TYPE fsmblks; /* always 0 */ + MALLINFO_FIELD_TYPE uordblks; /* total allocated space */ + MALLINFO_FIELD_TYPE fordblks; /* total free space */ + MALLINFO_FIELD_TYPE keepcost; /* releasable (via malloc_trim) space */ }; - #endif /* STRUCT_MALLINFO_DECLARED */ - #endif /* HAVE_USR_INCLUDE_MALLOC_H */ - #endif /* NO_MALLINFO */ + #endif /* STRUCT_MALLINFO_DECLARED */ + #endif /* HAVE_USR_INCLUDE_MALLOC_H */ + #endif /* NO_MALLINFO */ /* Try to persuade compilers to inline. The most critical functions for @@ -822,7 +820,7 @@ extern "C" { #ifndef FORCEINLINE #define FORCEINLINE inline #endif - #endif /* __cplusplus */ + #endif /* __cplusplus */ #ifndef FORCEINLINE #define FORCEINLINE #endif @@ -854,7 +852,7 @@ extern "C" { #define dlindependent_calloc independent_calloc #define dlindependent_comalloc independent_comalloc #define dlbulk_free bulk_free - #endif /* USE_DL_PREFIX */ + #endif /* USE_DL_PREFIX */ /* malloc(size_t n) @@ -1067,7 +1065,7 @@ DLMALLOC_EXPORT void dlmalloc_inspect_all(void (*handler)(void *, void *, size_t, void *), void *arg); - #endif /* MALLOC_INSPECT_ALL */ + #endif /* MALLOC_INSPECT_ALL */ #if !NO_MALLINFO /* @@ -1093,7 +1091,7 @@ DLMALLOC_EXPORT void dlmalloc_inspect_all(void (*handler)(void *, void *, thus be inaccurate. */ DLMALLOC_EXPORT struct mallinfo dlmallinfo(void); - #endif /* NO_MALLINFO */ + #endif /* NO_MALLINFO */ /* independent_calloc(size_t n_elements, size_t element_size, void* chunks[]); @@ -1287,7 +1285,7 @@ DLMALLOC_EXPORT void dlmalloc_stats(void); */ size_t dlmalloc_usable_size(void *); - #endif /* ONLY_MSPACES */ + #endif /* ONLY_MSPACES */ #if MSPACES @@ -1390,7 +1388,7 @@ DLMALLOC_EXPORT void *mspace_memalign(mspace msp, size_t alignment, */ DLMALLOC_EXPORT void **mspace_independent_calloc(mspace msp, size_t n_elements, size_t elem_size, - void * chunks[]); + void *chunks[]); /* mspace_independent_comalloc behaves as independent_comalloc, but @@ -1399,7 +1397,7 @@ DLMALLOC_EXPORT void **mspace_independent_calloc(mspace msp, size_t n_elements, DLMALLOC_EXPORT void **mspace_independent_comalloc(mspace msp, size_t n_elements, size_t sizes[], - void * chunks[]); + void *chunks[]); /* mspace_footprint() returns the number of bytes obtained from the @@ -1419,7 +1417,7 @@ DLMALLOC_EXPORT size_t mspace_max_footprint(mspace msp); the given space. */ DLMALLOC_EXPORT struct mallinfo mspace_mallinfo(mspace msp); - #endif /* NO_MALLINFO */ + #endif /* NO_MALLINFO */ /* malloc_usable_size(void* p) behaves the same as malloc_usable_size; @@ -1443,13 +1441,13 @@ DLMALLOC_EXPORT int mspace_trim(mspace msp, size_t pad); */ DLMALLOC_EXPORT int mspace_mallopt(int, int); - #endif /* MSPACES */ + #endif /* MSPACES */ #ifdef __cplusplus -} /* end of extern "C" */ +} /* end of extern "C" */ - #endif /* __cplusplus */ + #endif /* __cplusplus */ /* ======================================================================== @@ -1464,93 +1462,92 @@ DLMALLOC_EXPORT int mspace_mallopt(int, int); /*------------------------------ internal #includes ---------------------- */ #ifdef _MSC_VER - #pragma warning(disable : 4146) /* no "unsigned" warnings */ - #endif /* _MSC_VER */ + #pragma warning(disable : 4146) /* no "unsigned" warnings */ + #endif /* _MSC_VER */ #if !NO_MALLOC_STATS - #include /* for printing in malloc_stats */ - #endif /* NO_MALLOC_STATS */ + #include /* for printing in malloc_stats */ + #endif /* NO_MALLOC_STATS */ #ifndef LACKS_ERRNO_H - #include /* for MALLOC_FAILURE_ACTION */ - #endif /* LACKS_ERRNO_H */ + #include /* for MALLOC_FAILURE_ACTION */ + #endif /* LACKS_ERRNO_H */ #ifdef DEBUG #if ABORT_ON_ASSERT_FAILURE #undef assert #define assert(x) \ if (!(x)) ABORT - #else /* ABORT_ON_ASSERT_FAILURE */ + #else /* ABORT_ON_ASSERT_FAILURE */ #include - #endif /* ABORT_ON_ASSERT_FAILURE */ - #else /* DEBUG */ + #endif /* ABORT_ON_ASSERT_FAILURE */ + #else /* DEBUG */ #ifndef assert #define assert(x) #endif #define DEBUG 0 - #endif /* DEBUG */ + #endif /* DEBUG */ #if !defined(WIN32) && !defined(LACKS_TIME_H) - #include /* for magic initialization */ - #endif /* WIN32 */ + #include /* for magic initialization */ + #endif /* WIN32 */ #ifndef LACKS_STDLIB_H - #include /* for abort() */ - #endif /* LACKS_STDLIB_H */ + #include /* for abort() */ + #endif /* LACKS_STDLIB_H */ #ifndef LACKS_STRING_H - #include /* for memset etc */ - #endif /* LACKS_STRING_H */ + #include /* for memset etc */ + #endif /* LACKS_STRING_H */ #if USE_BUILTIN_FFS #ifndef LACKS_STRINGS_H - #include /* for ffs */ - #endif /* LACKS_STRINGS_H */ - #endif /* USE_BUILTIN_FFS */ + #include /* for ffs */ + #endif /* LACKS_STRINGS_H */ + #endif /* USE_BUILTIN_FFS */ #if HAVE_MMAP #ifndef LACKS_SYS_MMAN_H /* On some versions of linux, mremap decl in mman.h needs __USE_GNU set */ #if (defined(linux) && !defined(__USE_GNU)) #define __USE_GNU 1 - #include /* for mmap */ + #include /* for mmap */ #undef __USE_GNU #else - #include /* for mmap */ - #endif /* linux */ - #endif /* LACKS_SYS_MMAN_H */ + #include /* for mmap */ + #endif /* linux */ + #endif /* LACKS_SYS_MMAN_H */ #ifndef LACKS_FCNTL_H #include - #endif /* LACKS_FCNTL_H */ - #endif /* HAVE_MMAP */ + #endif /* LACKS_FCNTL_H */ + #endif /* HAVE_MMAP */ #ifndef LACKS_UNISTD_H - #include /* for sbrk, sysconf */ - #else /* LACKS_UNISTD_H */ + #include /* for sbrk, sysconf */ + #else /* LACKS_UNISTD_H */ #if !defined(__FreeBSD__) && !defined(__OpenBSD__) && !defined(__NetBSD__) extern void *sbrk(ptrdiff_t); - #endif /* FreeBSD etc */ - #endif /* LACKS_UNISTD_H */ + #endif /* FreeBSD etc */ + #endif /* LACKS_UNISTD_H */ /* Declarations for locking */ #if USE_LOCKS #ifndef WIN32 - #if defined(__SVR4) && defined(__sun) /* solaris */ + #if defined(__SVR4) && defined(__sun) /* solaris */ #include #elif !defined(LACKS_SCHED_H) #include - #endif /* solaris or LACKS_SCHED_H */ + #endif /* solaris or LACKS_SCHED_H */ #if (defined(USE_RECURSIVE_LOCKS) && USE_RECURSIVE_LOCKS != 0) || \ !USE_SPIN_LOCKS #include - #endif /* USE_RECURSIVE_LOCKS ... */ + #endif /* USE_RECURSIVE_LOCKS ... */ #elif defined(_MSC_VER) #ifndef _M_AMD64 /* These are already defined on AMD64 builds */ #ifdef __cplusplus extern "C" { - #endif /* __cplusplus */ + #endif /* __cplusplus */ LONG __cdecl _InterlockedCompareExchange(LONG volatile *Dest, LONG Exchange, LONG Comp); LONG __cdecl _InterlockedExchange(LONG volatile *Target, LONG Value); #ifdef __cplusplus - } - #endif /* __cplusplus */ - #endif /* _M_AMD64 */ + #endif /* __cplusplus */ + #endif /* _M_AMD64 */ #pragma intrinsic(_InterlockedCompareExchange) #pragma intrinsic(_InterlockedExchange) #define interlockedcompareexchange _InterlockedCompareExchange @@ -1559,9 +1556,9 @@ LONG __cdecl _InterlockedExchange(LONG volatile *Target, LONG Value); #define interlockedcompareexchange(a, b, c) \ __sync_val_compare_and_swap(a, c, b) #define interlockedexchange __sync_lock_test_and_set - #endif /* Win32 */ - #else /* USE_LOCKS */ - #endif /* USE_LOCKS */ + #endif /* Win32 */ + #else /* USE_LOCKS */ + #endif /* USE_LOCKS */ #ifndef LOCK_AT_FORK #define LOCK_AT_FORK 0 @@ -1569,29 +1566,28 @@ LONG __cdecl _InterlockedExchange(LONG volatile *Target, LONG Value); /* Declarations for bit scanning on win32 */ #if defined(_MSC_VER) && _MSC_VER >= 1300 - #ifndef BitScanForward /* Try to avoid pulling in WinNT.h */ + #ifndef BitScanForward /* Try to avoid pulling in WinNT.h */ #ifdef __cplusplus extern "C" { - #endif /* __cplusplus */ + #endif /* __cplusplus */ unsigned char _BitScanForward(unsigned long *index, unsigned long mask); unsigned char _BitScanReverse(unsigned long *index, unsigned long mask); #ifdef __cplusplus - } - #endif /* __cplusplus */ + #endif /* __cplusplus */ #define BitScanForward _BitScanForward #define BitScanReverse _BitScanReverse #pragma intrinsic(_BitScanForward) #pragma intrinsic(_BitScanReverse) - #endif /* BitScanForward */ - #endif /* defined(_MSC_VER) && _MSC_VER>=1300 */ + #endif /* BitScanForward */ + #endif /* defined(_MSC_VER) && _MSC_VER>=1300 */ #ifndef WIN32 #ifndef malloc_getpagesize - #ifdef _SC_PAGESIZE /* some SVR4 systems omit an underscore */ + #ifdef _SC_PAGESIZE /* some SVR4 systems omit an underscore */ #ifndef _SC_PAGE_SIZE #define _SC_PAGE_SIZE _SC_PAGESIZE #endif @@ -1603,7 +1599,7 @@ unsigned char _BitScanReverse(unsigned long *index, unsigned long mask); extern size_t getpagesize(); #define malloc_getpagesize getpagesize() #else - #ifdef WIN32 /* use supplied emulation of getpagesize */ + #ifdef WIN32 /* use supplied emulation of getpagesize */ #define malloc_getpagesize getpagesize() #else #ifndef LACKS_SYS_PARAM_H @@ -1624,7 +1620,7 @@ extern size_t getpagesize(); #else #ifdef PAGESIZE #define malloc_getpagesize PAGESIZE - #else /* just guess */ + #else /* just guess */ #define malloc_getpagesize ((size_t)4096U) #endif #endif @@ -1676,7 +1672,7 @@ extern size_t getpagesize(); /* MORECORE and MMAP must return MFAIL on failure */ #define MFAIL ((void *)(MAX_SIZE_T)) - #define CMFAIL ((char *)(MFAIL)) /* defined for convenience */ + #define CMFAIL ((char *)(MFAIL)) /* defined for convenience */ #if HAVE_MMAP @@ -1684,70 +1680,63 @@ extern size_t getpagesize(); #define MMAP_PROT (PROT_READ | PROT_WRITE) #if !defined(MAP_ANONYMOUS) && defined(MAP_ANON) #define MAP_ANONYMOUS MAP_ANON - #endif /* MAP_ANON */ + #endif /* MAP_ANON */ #ifdef MAP_ANONYMOUS #define MMAP_FLAGS (MAP_PRIVATE | MAP_ANONYMOUS) static FORCEINLINE void *unixmmap(size_t size) { - void *result; result = mmap(0, size, MMAP_PROT, MMAP_FLAGS, -1, 0); if (result == MFAIL) return MFAIL; return result; - } static FORCEINLINE int unixmunmap(void *ptr, size_t size) { - int result; result = munmap(ptr, size); if (result != 0) return result; return result; - } #define MMAP_DEFAULT(s) unixmmap(s) #define MUNMAP_DEFAULT(a, s) unixmunmap((a), (s)) - #else /* MAP_ANONYMOUS */ + #else /* MAP_ANONYMOUS */ /* Nearly all versions of mmap support MAP_ANONYMOUS, so the following is unlikely to be needed, but is supplied just in case. */ #define MMAP_FLAGS (MAP_PRIVATE) -static int dev_zero_fd = -1; /* Cached file descriptor for /dev/zero. */ +static int dev_zero_fd = -1; /* Cached file descriptor for /dev/zero. */ #define MMAP_DEFAULT(s) \ ((dev_zero_fd < 0) \ ? (dev_zero_fd = open("/dev/zero", O_RDWR), \ mmap(0, (s), MMAP_PROT, MMAP_FLAGS, dev_zero_fd, 0)) \ : mmap(0, (s), MMAP_PROT, MMAP_FLAGS, dev_zero_fd, 0)) #define MUNMAP_DEFAULT(a, s) munmap((a), (s)) - #endif /* MAP_ANONYMOUS */ + #endif /* MAP_ANONYMOUS */ #define DIRECT_MMAP_DEFAULT(s) MMAP_DEFAULT(s) - #else /* WIN32 */ + #else /* WIN32 */ /* Win32 MMAP via VirtualAlloc */ static FORCEINLINE void *win32mmap(size_t size) { - void *ptr; ptr = VirtualAlloc(0, size, MEM_RESERVE | MEM_COMMIT, PAGE_READWRITE); if (ptr == 0) return MFAIL; return ptr; - } /* For direct MMAP, use MEM_TOP_DOWN to minimize interference */ static FORCEINLINE void *win32direct_mmap(size_t size) { - void *ptr; ptr = VirtualAlloc(0, size, MEM_RESERVE | MEM_COMMIT | MEM_TOP_DOWN, @@ -1755,17 +1744,14 @@ static FORCEINLINE void *win32direct_mmap(size_t size) { if (ptr == 0) return MFAIL; return ptr; - } /* This function supports releasing coalesed segments */ static FORCEINLINE int win32munmap(void *ptr, size_t size) { - MEMORY_BASIC_INFORMATION minfo; char *cptr = (char *)ptr; while (size) { - if (VirtualQuery(cptr, &minfo, sizeof(minfo)) == 0) return -1; if (minfo.BaseAddress != cptr || minfo.AllocationBase != cptr || minfo.State != MEM_COMMIT || minfo.RegionSize > size) @@ -1773,38 +1759,34 @@ static FORCEINLINE int win32munmap(void *ptr, size_t size) { if (VirtualFree(cptr, 0, MEM_RELEASE) == 0) return -1; cptr += minfo.RegionSize; size -= minfo.RegionSize; - } return 0; - } #define MMAP_DEFAULT(s) win32mmap(s) #define MUNMAP_DEFAULT(a, s) win32munmap((a), (s)) #define DIRECT_MMAP_DEFAULT(s) win32direct_mmap(s) - #endif /* WIN32 */ - #endif /* HAVE_MMAP */ + #endif /* WIN32 */ + #endif /* HAVE_MMAP */ #if HAVE_MREMAP #ifndef WIN32 static FORCEINLINE void *dlmremap(void *old_address, size_t old_size, size_t new_size, int flags) { - void *result; result = mremap(old_address, old_size, new_size, flags); if (result == MFAIL) return MFAIL; return result; - } #define MREMAP_DEFAULT(addr, osz, nsz, mv) \ dlmremap((addr), (osz), (nsz), (mv)) - #endif /* WIN32 */ - #endif /* HAVE_MREMAP */ + #endif /* WIN32 */ + #endif /* HAVE_MREMAP */ /** * Define CALL_MORECORE @@ -1812,12 +1794,12 @@ static FORCEINLINE void *dlmremap(void *old_address, size_t old_size, #if HAVE_MORECORE #ifdef MORECORE #define CALL_MORECORE(S) MORECORE(S) - #else /* MORECORE */ + #else /* MORECORE */ #define CALL_MORECORE(S) MORECORE_DEFAULT(S) - #endif /* MORECORE */ - #else /* HAVE_MORECORE */ + #endif /* MORECORE */ + #else /* HAVE_MORECORE */ #define CALL_MORECORE(S) MFAIL - #endif /* HAVE_MORECORE */ + #endif /* HAVE_MORECORE */ /** * Define CALL_MMAP/CALL_MUNMAP/CALL_DIRECT_MMAP @@ -1827,20 +1809,20 @@ static FORCEINLINE void *dlmremap(void *old_address, size_t old_size, #ifdef MMAP #define CALL_MMAP(s) MMAP(s) - #else /* MMAP */ + #else /* MMAP */ #define CALL_MMAP(s) MMAP_DEFAULT(s) - #endif /* MMAP */ + #endif /* MMAP */ #ifdef MUNMAP #define CALL_MUNMAP(a, s) MUNMAP((a), (s)) - #else /* MUNMAP */ + #else /* MUNMAP */ #define CALL_MUNMAP(a, s) MUNMAP_DEFAULT((a), (s)) - #endif /* MUNMAP */ + #endif /* MUNMAP */ #ifdef DIRECT_MMAP #define CALL_DIRECT_MMAP(s) DIRECT_MMAP(s) - #else /* DIRECT_MMAP */ + #else /* DIRECT_MMAP */ #define CALL_DIRECT_MMAP(s) DIRECT_MMAP_DEFAULT(s) - #endif /* DIRECT_MMAP */ - #else /* HAVE_MMAP */ + #endif /* DIRECT_MMAP */ + #else /* HAVE_MMAP */ #define USE_MMAP_BIT (SIZE_T_ZERO) #define MMAP(s) MFAIL @@ -1849,7 +1831,7 @@ static FORCEINLINE void *dlmremap(void *old_address, size_t old_size, #define CALL_DIRECT_MMAP(s) DIRECT_MMAP(s) #define CALL_MMAP(s) MMAP(s) #define CALL_MUNMAP(a, s) MUNMAP((a), (s)) - #endif /* HAVE_MMAP */ + #endif /* HAVE_MMAP */ /** * Define CALL_MREMAP @@ -1857,13 +1839,13 @@ static FORCEINLINE void *dlmremap(void *old_address, size_t old_size, #if HAVE_MMAP && HAVE_MREMAP #ifdef MREMAP #define CALL_MREMAP(addr, osz, nsz, mv) MREMAP((addr), (osz), (nsz), (mv)) - #else /* MREMAP */ + #else /* MREMAP */ #define CALL_MREMAP(addr, osz, nsz, mv) \ MREMAP_DEFAULT((addr), (osz), (nsz), (mv)) - #endif /* MREMAP */ - #else /* HAVE_MMAP && HAVE_MREMAP */ + #endif /* MREMAP */ + #else /* HAVE_MMAP && HAVE_MREMAP */ #define CALL_MREMAP(addr, osz, nsz, mv) MFAIL - #endif /* HAVE_MMAP && HAVE_MREMAP */ + #endif /* HAVE_MMAP && HAVE_MREMAP */ /* mstate bit set if continguous morecore disabled or failed */ #define USE_NONCONTIGUOUS_BIT (4U) @@ -1932,7 +1914,6 @@ static FORCEINLINE void *dlmremap(void *old_address, size_t old_size, #elif (defined(__GNUC__) && (defined(__i386__) || defined(__x86_64__))) /* Custom spin locks for older gcc on x86 */ static FORCEINLINE int x86_cas_lock(int *sl) { - int ret; int val = 1; int cmp = 0; @@ -1941,11 +1922,9 @@ static FORCEINLINE int x86_cas_lock(int *sl) { : "r"(val), "m"(*(sl)), "0"(cmp) : "memory", "cc"); return ret; - } static FORCEINLINE void x86_clear_lock(int *sl) { - assert(*sl != 0); int prev = 0; int ret; @@ -1953,44 +1932,39 @@ static FORCEINLINE void x86_clear_lock(int *sl) { : "=r"(ret) : "m"(*(sl)), "0"(prev) : "memory"); - } #define CAS_LOCK(sl) x86_cas_lock(sl) #define CLEAR_LOCK(sl) x86_clear_lock(sl) - #else /* Win32 MSC */ + #else /* Win32 MSC */ #define CAS_LOCK(sl) interlockedexchange((volatile LONG *)sl, (LONG)1) #define CLEAR_LOCK(sl) interlockedexchange((volatile LONG *)sl, (LONG)0) - #endif /* ... gcc spins locks ... */ + #endif /* ... gcc spins locks ... */ /* How to yield for a spin lock */ #define SPINS_PER_YIELD 63 #if defined(_MSC_VER) - #define SLEEP_EX_DURATION 50 /* delay for yield/sleep */ + #define SLEEP_EX_DURATION 50 /* delay for yield/sleep */ #define SPIN_LOCK_YIELD SleepEx(SLEEP_EX_DURATION, FALSE) - #elif defined(__SVR4) && defined(__sun) /* solaris */ + #elif defined(__SVR4) && defined(__sun) /* solaris */ #define SPIN_LOCK_YIELD thr_yield(); #elif !defined(LACKS_SCHED_H) #define SPIN_LOCK_YIELD sched_yield(); #else #define SPIN_LOCK_YIELD - #endif /* ... yield ... */ + #endif /* ... yield ... */ #if !defined(USE_RECURSIVE_LOCKS) || USE_RECURSIVE_LOCKS == 0 /* Plain spin locks use single word (embedded in malloc_states) */ static int spin_acquire_lock(int *sl) { - int spins = 0; while (*(volatile int *)sl != 0 || CAS_LOCK(sl)) { - if ((++spins & SPINS_PER_YIELD) == 0) { SPIN_LOCK_YIELD; } - } return 0; - } #define MLOCK_T int @@ -2001,7 +1975,7 @@ static int spin_acquire_lock(int *sl) { #define DESTROY_LOCK(sl) (0) static MLOCK_T malloc_global_mutex = 0; - #else /* USE_RECURSIVE_LOCKS */ + #else /* USE_RECURSIVE_LOCKS */ /* types for lock owners */ #ifdef WIN32 #define THREAD_ID_T DWORD @@ -2019,74 +1993,54 @@ static MLOCK_T malloc_global_mutex = 0; #endif struct malloc_recursive_lock { - int sl; unsigned int c; THREAD_ID_T threadid; - }; #define MLOCK_T struct malloc_recursive_lock static MLOCK_T malloc_global_mutex = {0, 0, (THREAD_ID_T)0}; static FORCEINLINE void recursive_release_lock(MLOCK_T *lk) { - assert(lk->sl != 0); if (--lk->c == 0) { CLEAR_LOCK(&lk->sl); } - } static FORCEINLINE int recursive_acquire_lock(MLOCK_T *lk) { - THREAD_ID_T mythreadid = CURRENT_THREAD; int spins = 0; for (;;) { - if (*((volatile int *)(&lk->sl)) == 0) { - if (!CAS_LOCK(&lk->sl)) { - lk->threadid = mythreadid; lk->c = 1; return 0; - } } else if (EQ_OWNER(lk->threadid, mythreadid)) { - ++lk->c; return 0; - } if ((++spins & SPINS_PER_YIELD) == 0) { SPIN_LOCK_YIELD; } - } - } static FORCEINLINE int recursive_try_lock(MLOCK_T *lk) { - THREAD_ID_T mythreadid = CURRENT_THREAD; if (*((volatile int *)(&lk->sl)) == 0) { - if (!CAS_LOCK(&lk->sl)) { - lk->threadid = mythreadid; lk->c = 1; return 1; - } } else if (EQ_OWNER(lk->threadid, mythreadid)) { - ++lk->c; return 1; - } return 0; - } #define RELEASE_LOCK(lk) recursive_release_lock(lk) @@ -2095,9 +2049,9 @@ static FORCEINLINE int recursive_try_lock(MLOCK_T *lk) { #define INITIAL_LOCK(lk) \ ((lk)->threadid = (THREAD_ID_T)0, (lk)->sl = 0, (lk)->c = 0) #define DESTROY_LOCK(lk) (0) - #endif /* USE_RECURSIVE_LOCKS */ + #endif /* USE_RECURSIVE_LOCKS */ - #elif defined(WIN32) /* Win32 critical sections */ + #elif defined(WIN32) /* Win32 critical sections */ #define MLOCK_T CRITICAL_SECTION #define ACQUIRE_LOCK(lk) (EnterCriticalSection(lk), 0) #define RELEASE_LOCK(lk) LeaveCriticalSection(lk) @@ -2112,28 +2066,22 @@ static volatile LONG malloc_global_mutex_status; /* Use spin loop to initialize global lock */ static void init_malloc_global_mutex() { - for (;;) { - long stat = malloc_global_mutex_status; if (stat > 0) return; /* transition to < 0 while initializing, then to > 0) */ if (stat == 0 && interlockedcompareexchange(&malloc_global_mutex_status, (LONG)-1, (LONG)0) == 0) { - InitializeCriticalSection(&malloc_global_mutex); interlockedexchange(&malloc_global_mutex_status, (LONG)1); return; - } SleepEx(0, FALSE); - } - } - #else /* pthreads-based locks */ + #else /* pthreads-based locks */ #define MLOCK_T pthread_mutex_t #define ACQUIRE_LOCK(lk) pthread_mutex_lock(lk) #define RELEASE_LOCK(lk) pthread_mutex_unlock(lk) @@ -2150,12 +2098,11 @@ extern int pthread_mutexattr_setkind_np __P((pthread_mutexattr_t * __attr, #define PTHREAD_MUTEX_RECURSIVE PTHREAD_MUTEX_RECURSIVE_NP #define pthread_mutexattr_settype(x, y) \ pthread_mutexattr_setkind_np(x, y) - #endif /* USE_RECURSIVE_LOCKS ... */ + #endif /* USE_RECURSIVE_LOCKS ... */ static MLOCK_T malloc_global_mutex = PTHREAD_MUTEX_INITIALIZER; static int pthread_init_lock(MLOCK_T *lk) { - pthread_mutexattr_t attr; if (pthread_mutexattr_init(&attr)) return 1; #if defined(USE_RECURSIVE_LOCKS) && USE_RECURSIVE_LOCKS != 0 @@ -2164,10 +2111,9 @@ static int pthread_init_lock(MLOCK_T *lk) { if (pthread_mutex_init(lk, &attr)) return 1; if (pthread_mutexattr_destroy(&attr)) return 1; return 0; - } - #endif /* ... lock types ... */ + #endif /* ... lock types ... */ /* Common code for all lock types */ #define USE_LOCK_BIT (2U) @@ -2180,7 +2126,7 @@ static int pthread_init_lock(MLOCK_T *lk) { #define RELEASE_MALLOC_GLOBAL_LOCK() RELEASE_LOCK(&malloc_global_mutex); #endif - #endif /* USE_LOCKS */ + #endif /* USE_LOCKS */ /* ----------------------- Chunk representations ------------------------ */ @@ -2320,19 +2266,17 @@ static int pthread_init_lock(MLOCK_T *lk) { */ struct malloc_chunk { - - size_t prev_foot; /* Size of previous chunk (if free). */ - size_t head; /* Size and inuse bits. */ - struct malloc_chunk *fd; /* double links -- used only if free. */ + size_t prev_foot; /* Size of previous chunk (if free). */ + size_t head; /* Size and inuse bits. */ + struct malloc_chunk *fd; /* double links -- used only if free. */ struct malloc_chunk *bk; - }; typedef struct malloc_chunk mchunk; typedef struct malloc_chunk *mchunkptr; -typedef struct malloc_chunk *sbinptr; /* The type of bins of chunks */ -typedef unsigned int bindex_t; /* Described below */ -typedef unsigned int binmap_t; /* Described below */ +typedef struct malloc_chunk *sbinptr; /* The type of bins of chunks */ +typedef unsigned int bindex_t; /* Described below */ +typedef unsigned int binmap_t; /* Described below */ typedef unsigned int flag_t; /* The type of various bit flag sets */ /* ------------------- Chunks sizes and alignments ----------------------- */ @@ -2341,9 +2285,9 @@ typedef unsigned int flag_t; /* The type of various bit flag sets */ #if FOOTERS #define CHUNK_OVERHEAD (TWO_SIZE_T_SIZES) - #else /* FOOTERS */ + #else /* FOOTERS */ #define CHUNK_OVERHEAD (SIZE_T_SIZE) - #endif /* FOOTERS */ + #endif /* FOOTERS */ /* MMapped chunks need a second word of overhead ... */ #define MMAP_CHUNK_OVERHEAD (TWO_SIZE_T_SIZES) @@ -2432,9 +2376,9 @@ typedef unsigned int flag_t; /* The type of various bit flag sets */ /* Return true if malloced space is not necessarily cleared */ #if MMAP_CLEARS #define calloc_must_clear(p) (!is_mmapped(p)) - #else /* MMAP_CLEARS */ + #else /* MMAP_CLEARS */ #define calloc_must_clear(p) (1) - #endif /* MMAP_CLEARS */ + #endif /* MMAP_CLEARS */ /* ---------------------- Overlaid data structures ----------------------- */ @@ -2528,7 +2472,6 @@ nextchunk-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ */ struct malloc_tree_chunk { - /* The first four fields must be compatible with malloc_chunk */ size_t prev_foot; size_t head; @@ -2538,12 +2481,11 @@ struct malloc_tree_chunk { struct malloc_tree_chunk *child[2]; struct malloc_tree_chunk *parent; bindex_t index; - }; typedef struct malloc_tree_chunk tchunk; typedef struct malloc_tree_chunk *tchunkptr; -typedef struct malloc_tree_chunk *tbinptr; /* The type of bins of trees */ +typedef struct malloc_tree_chunk *tbinptr; /* The type of bins of trees */ /* A little helper macro for trees */ #define leftmost_child(t) ((t)->child[0] != 0 ? (t)->child[0] : (t)->child[1]) @@ -2606,12 +2548,10 @@ typedef struct malloc_tree_chunk *tbinptr; /* The type of bins of trees */ */ struct malloc_segment { - - char * base; /* base address */ - size_t size; /* allocated size */ - struct malloc_segment *next; /* ptr to next segment */ - flag_t sflags; /* mmap and extern flag */ - + char *base; /* base address */ + size_t size; /* allocated size */ + struct malloc_segment *next; /* ptr to next segment */ + flag_t sflags; /* mmap and extern flag */ }; #define is_mmapped_segment(S) ((S)->sflags & USE_MMAP_BIT) @@ -2718,12 +2658,11 @@ typedef struct malloc_segment *msegmentptr; #define MAX_SMALL_REQUEST (MAX_SMALL_SIZE - CHUNK_ALIGN_MASK - CHUNK_OVERHEAD) struct malloc_state { - binmap_t smallmap; binmap_t treemap; size_t dvsize; size_t topsize; - char * least_addr; + char *least_addr; mchunkptr dv; mchunkptr top; size_t trim_check; @@ -2733,15 +2672,14 @@ struct malloc_state { tbinptr treebins[NTREEBINS]; size_t footprint; size_t max_footprint; - size_t footprint_limit; /* zero means no limit */ + size_t footprint_limit; /* zero means no limit */ flag_t mflags; #if USE_LOCKS - MLOCK_T mutex; /* locate lock among fields that rarely change */ - #endif /* USE_LOCKS */ + MLOCK_T mutex; /* locate lock among fields that rarely change */ + #endif /* USE_LOCKS */ msegment seg; - void * extp; /* Unused but available for extensions */ + void *extp; /* Unused but available for extensions */ size_t exts; - }; typedef struct malloc_state *mstate; @@ -2756,14 +2694,12 @@ typedef struct malloc_state *mstate; */ struct malloc_params { - size_t magic; size_t page_size; size_t granularity; size_t mmap_threshold; size_t trim_threshold; flag_t default_mflags; - }; static struct malloc_params mparams; @@ -2778,7 +2714,7 @@ static struct malloc_state _gm_; #define gm (&_gm_) #define is_global(M) ((M) == &_gm_) - #endif /* !ONLY_MSPACES */ + #endif /* !ONLY_MSPACES */ #define is_initialized(M) ((M)->top != 0) @@ -2840,35 +2776,27 @@ static struct malloc_state _gm_; /* Return segment holding given address */ static msegmentptr segment_holding(mstate m, char *addr) { - msegmentptr sp = &m->seg; for (;;) { - if (addr >= sp->base && addr < sp->base + sp->size) return sp; if ((sp = sp->next) == 0) return 0; - } - } /* Return true if segment contains a segment link */ static int has_segment_link(mstate m, msegmentptr ss) { - msegmentptr sp = &m->seg; for (;;) { - if ((char *)sp >= ss->base && (char *)sp < ss->base + ss->size) return 1; if ((sp = sp->next) == 0) return 0; - } - } #ifndef MORECORE_CANNOT_TRIM #define should_trim(M, s) ((s) > (M)->trim_check) - #else /* MORECORE_CANNOT_TRIM */ + #else /* MORECORE_CANNOT_TRIM */ #define should_trim(M, s) (0) - #endif /* MORECORE_CANNOT_TRIM */ + #endif /* MORECORE_CANNOT_TRIM */ /* TOP_FOOT_SIZE is padding at the end of a segment, including space @@ -2891,21 +2819,19 @@ static int has_segment_link(mstate m, msegmentptr ss) { #define PREACTION(M) ((use_lock(M)) ? ACQUIRE_LOCK(&(M)->mutex) : 0) #define POSTACTION(M) \ { \ - \ if (use_lock(M)) RELEASE_LOCK(&(M)->mutex); \ - \ } - #else /* USE_LOCKS */ + #else /* USE_LOCKS */ #ifndef PREACTION #define PREACTION(M) (0) - #endif /* PREACTION */ + #endif /* PREACTION */ #ifndef POSTACTION #define POSTACTION(M) - #endif /* POSTACTION */ + #endif /* POSTACTION */ - #endif /* USE_LOCKS */ + #endif /* USE_LOCKS */ /* CORRUPTION_ERROR_ACTION is triggered upon detected bad addresses. @@ -2926,17 +2852,17 @@ static void reset_on_error(mstate m); #define CORRUPTION_ERROR_ACTION(m) reset_on_error(m) #define USAGE_ERROR_ACTION(m, p) - #else /* PROCEED_ON_ERROR */ + #else /* PROCEED_ON_ERROR */ #ifndef CORRUPTION_ERROR_ACTION #define CORRUPTION_ERROR_ACTION(m) ABORT - #endif /* CORRUPTION_ERROR_ACTION */ + #endif /* CORRUPTION_ERROR_ACTION */ #ifndef USAGE_ERROR_ACTION #define USAGE_ERROR_ACTION(m, p) ABORT - #endif /* USAGE_ERROR_ACTION */ + #endif /* USAGE_ERROR_ACTION */ - #endif /* PROCEED_ON_ERROR */ + #endif /* PROCEED_ON_ERROR */ /* -------------------------- Debugging setup ---------------------------- */ @@ -2949,7 +2875,7 @@ static void reset_on_error(mstate m); #define check_malloc_state(M) #define check_top_chunk(M, P) - #else /* DEBUG */ + #else /* DEBUG */ #define check_free_chunk(M, P) do_check_free_chunk(M, P) #define check_inuse_chunk(M, P) do_check_inuse_chunk(M, P) #define check_top_chunk(M, P) do_check_top_chunk(M, P) @@ -2969,7 +2895,7 @@ static void do_check_smallbin(mstate m, bindex_t i); static void do_check_malloc_state(mstate m); static int bin_find(mstate m, mchunkptr x); static size_t traverse_and_check(mstate m); - #endif /* DEBUG */ + #endif /* DEBUG */ /* ---------------------------- Indexing Bins ---------------------------- */ @@ -2986,70 +2912,56 @@ static size_t traverse_and_check(mstate m); #if defined(__GNUC__) && (defined(__i386__) || defined(__x86_64__)) #define compute_tree_index(S, I) \ { \ - \ unsigned int X = S >> TREEBIN_SHIFT; \ if (X == 0) \ I = 0; \ else if (X > 0xFFFF) \ I = NTREEBINS - 1; \ else { \ - \ unsigned int K = (unsigned)sizeof(X) * __CHAR_BIT__ - 1 - \ (unsigned)__builtin_clz(X); \ I = (bindex_t)((K << 1) + ((S >> (K + (TREEBIN_SHIFT - 1)) & 1))); \ - \ } \ - \ } #elif defined(__INTEL_COMPILER) #define compute_tree_index(S, I) \ { \ - \ size_t X = S >> TREEBIN_SHIFT; \ if (X == 0) \ I = 0; \ else if (X > 0xFFFF) \ I = NTREEBINS - 1; \ else { \ - \ unsigned int K = _bit_scan_reverse(X); \ I = (bindex_t)((K << 1) + ((S >> (K + (TREEBIN_SHIFT - 1)) & 1))); \ - \ } \ - \ } #elif defined(_MSC_VER) && _MSC_VER >= 1300 #define compute_tree_index(S, I) \ { \ - \ size_t X = S >> TREEBIN_SHIFT; \ if (X == 0) \ I = 0; \ else if (X > 0xFFFF) \ I = NTREEBINS - 1; \ else { \ - \ unsigned int K; \ _BitScanReverse((DWORD *)&K, (DWORD)X); \ I = (bindex_t)((K << 1) + ((S >> (K + (TREEBIN_SHIFT - 1)) & 1))); \ - \ } \ - \ } - #else /* GNUC */ + #else /* GNUC */ #define compute_tree_index(S, I) \ { \ - \ size_t X = S >> TREEBIN_SHIFT; \ if (X == 0) \ I = 0; \ else if (X > 0xFFFF) \ I = NTREEBINS - 1; \ else { \ - \ unsigned int Y = (unsigned int)X; \ unsigned int N = ((Y - 0x100) >> 16) & 8; \ unsigned int K = (((Y <<= N) - 0x1000) >> 16) & 4; \ @@ -3057,11 +2969,9 @@ static size_t traverse_and_check(mstate m); N += K = (((Y <<= K) - 0x4000) >> 16) & 2; \ K = 14 - N + ((Y <<= K) >> 15); \ I = (K << 1) + ((S >> (K + (TREEBIN_SHIFT - 1)) & 1)); \ - \ } \ - \ } - #endif /* GNUC */ + #endif /* GNUC */ /* Bit representing maximum resolved size in a treebin at i */ #define bit_for_tree_index(i) \ @@ -3107,31 +3017,25 @@ static size_t traverse_and_check(mstate m); #if defined(__GNUC__) && (defined(__i386__) || defined(__x86_64__)) #define compute_bit2idx(X, I) \ { \ - \ unsigned int J; \ J = __builtin_ctz(X); \ I = (bindex_t)J; \ - \ } #elif defined(__INTEL_COMPILER) #define compute_bit2idx(X, I) \ { \ - \ unsigned int J; \ J = _bit_scan_forward(X); \ I = (bindex_t)J; \ - \ } #elif defined(_MSC_VER) && _MSC_VER >= 1300 #define compute_bit2idx(X, I) \ { \ - \ unsigned int J; \ _BitScanForward((DWORD *)&J, X); \ I = (bindex_t)J; \ - \ } #elif USE_BUILTIN_FFS @@ -3140,7 +3044,6 @@ static size_t traverse_and_check(mstate m); #else #define compute_bit2idx(X, I) \ { \ - \ unsigned int Y = X - 1; \ unsigned int K = Y >> (16 - 4) & 16; \ unsigned int N = K; \ @@ -3154,9 +3057,8 @@ static size_t traverse_and_check(mstate m); N += K = Y >> (1 - 0) & 1; \ Y >>= K; \ I = (bindex_t)(N + Y); \ - \ } - #endif /* GNUC */ + #endif /* GNUC */ /* ----------------------- Runtime Check Support ------------------------- */ @@ -3196,30 +3098,30 @@ static size_t traverse_and_check(mstate m); /* Check if p has its pinuse bit on */ #define ok_pinuse(p) pinuse(p) - #else /* !INSECURE */ + #else /* !INSECURE */ #define ok_address(M, a) (1) #define ok_next(b, n) (1) #define ok_inuse(p) (1) #define ok_pinuse(p) (1) - #endif /* !INSECURE */ + #endif /* !INSECURE */ #if (FOOTERS && !INSECURE) /* Check if (alleged) mstate m has expected magic field */ #define ok_magic(M) ((M)->magic == mparams.magic) - #else /* (FOOTERS && !INSECURE) */ + #else /* (FOOTERS && !INSECURE) */ #define ok_magic(M) (1) - #endif /* (FOOTERS && !INSECURE) */ + #endif /* (FOOTERS && !INSECURE) */ /* In gcc, use __builtin_expect to minimize impact of checks */ #if !INSECURE #if defined(__GNUC__) && __GNUC__ >= 3 #define RTCHECK(e) __builtin_expect(e, 1) - #else /* GNUC */ + #else /* GNUC */ #define RTCHECK(e) (e) - #endif /* GNUC */ - #else /* !INSECURE */ + #endif /* GNUC */ + #else /* !INSECURE */ #define RTCHECK(e) (1) - #endif /* !INSECURE */ + #endif /* !INSECURE */ /* macros to set up inuse chunks with or without footers */ @@ -3243,7 +3145,7 @@ static size_t traverse_and_check(mstate m); #define set_size_and_pinuse_of_inuse_chunk(M, p, s) \ ((p)->head = (s | PINUSE_BIT | CINUSE_BIT)) - #else /* FOOTERS */ + #else /* FOOTERS */ /* Set foot of inuse chunk to be xor of mstate and seed */ #define mark_inuse_foot(M, p, s) \ @@ -3267,41 +3169,33 @@ static size_t traverse_and_check(mstate m); #define set_size_and_pinuse_of_inuse_chunk(M, p, s) \ ((p)->head = (s | PINUSE_BIT | CINUSE_BIT), mark_inuse_foot(M, p, s)) - #endif /* !FOOTERS */ + #endif /* !FOOTERS */ /* ---------------------------- setting mparams -------------------------- */ #if LOCK_AT_FORK static void pre_fork(void) { - ACQUIRE_LOCK(&(gm)->mutex); - } static void post_fork_parent(void) { - RELEASE_LOCK(&(gm)->mutex); - } static void post_fork_child(void) { - INITIAL_LOCK(&(gm)->mutex); - } - #endif /* LOCK_AT_FORK */ + #endif /* LOCK_AT_FORK */ /* Initialize mparams */ static int init_mparams(void) { - #ifdef NEED_GLOBAL_LOCK_INIT if (malloc_global_mutex_status <= 0) init_malloc_global_mutex(); #endif ACQUIRE_MALLOC_GLOBAL_LOCK(); if (mparams.magic == 0) { - size_t magic; size_t psize; size_t gsize; @@ -3309,19 +3203,17 @@ static int init_mparams(void) { #ifndef WIN32 psize = malloc_getpagesize; gsize = ((DEFAULT_GRANULARITY != 0) ? DEFAULT_GRANULARITY : psize); - #else /* WIN32 */ + #else /* WIN32 */ { - SYSTEM_INFO system_info; GetSystemInfo(&system_info); psize = system_info.dwPageSize; gsize = ((DEFAULT_GRANULARITY != 0) ? DEFAULT_GRANULARITY : system_info.dwAllocationGranularity); - } - #endif /* WIN32 */ + #endif /* WIN32 */ /* Sanity-check configuration: size_t must be unsigned and as wide as pointer type. @@ -3342,10 +3234,10 @@ static int init_mparams(void) { mparams.trim_threshold = DEFAULT_TRIM_THRESHOLD; #if MORECORE_CONTIGUOUS mparams.default_mflags = USE_LOCK_BIT | USE_MMAP_BIT; - #else /* MORECORE_CONTIGUOUS */ + #else /* MORECORE_CONTIGUOUS */ mparams.default_mflags = USE_LOCK_BIT | USE_MMAP_BIT | USE_NONCONTIGUOUS_BIT; - #endif /* MORECORE_CONTIGUOUS */ + #endif /* MORECORE_CONTIGUOUS */ #if !ONLY_MSPACES /* Set up lock for main malloc area */ @@ -3357,20 +3249,18 @@ static int init_mparams(void) { #endif { - #if USE_DEV_RANDOM int fd; unsigned char buf[sizeof(size_t)]; /* Try to use /dev/urandom, else fall back on using time */ if ((fd = open("/dev/urandom", O_RDONLY)) >= 0 && read(fd, buf, sizeof(buf)) == sizeof(buf)) { - magic = *((size_t *)buf); close(fd); } else - #endif /* USE_DEV_RANDOM */ + #endif /* USE_DEV_RANDOM */ #ifdef WIN32 magic = (size_t)(GetTickCount() ^ (size_t)0x55555555U); #elif defined(LACKS_TIME_H) @@ -3378,34 +3268,28 @@ static int init_mparams(void) { #else magic = (size_t)(time(0) ^ (size_t)0x55555555U); #endif - magic |= (size_t)8U; /* ensure nonzero */ - magic &= ~(size_t)7U; /* improve chances of fault for bad values */ + magic |= (size_t)8U; /* ensure nonzero */ + magic &= ~(size_t)7U; /* improve chances of fault for bad values */ /* Until memory modes commonly available, use volatile-write */ (*(volatile size_t *)(&(mparams.magic))) = magic; - } - } RELEASE_MALLOC_GLOBAL_LOCK(); return 1; - } /* support for mallopt */ static int change_mparam(int param_number, int value) { - size_t val; ensure_initialization(); val = (value == -1) ? MAX_SIZE_T : (size_t)value; switch (param_number) { - case M_TRIM_THRESHOLD: mparams.trim_threshold = val; return 1; case M_GRANULARITY: if (val >= mparams.page_size && ((val & (val - 1)) == 0)) { - mparams.granularity = val; return 1; @@ -3417,9 +3301,7 @@ static int change_mparam(int param_number, int value) { return 1; default: return 0; - } - } #if DEBUG @@ -3427,17 +3309,14 @@ static int change_mparam(int param_number, int value) { /* Check properties of any chunk, whether free, inuse, mmapped etc */ static void do_check_any_chunk(mstate m, mchunkptr p) { - assert((is_aligned(chunk2mem(p))) || (p->head == FENCEPOST_HEAD)); assert(ok_address(m, p)); - } /* Check properties of top chunk */ static void do_check_top_chunk(mstate m, mchunkptr p) { - msegmentptr sp = segment_holding(m, (char *)p); - size_t sz = p->head & ~INUSE_BITS; /* third-lowest bit can be set! */ + size_t sz = p->head & ~INUSE_BITS; /* third-lowest bit can be set! */ assert(sp != 0); assert((is_aligned(chunk2mem(p))) || (p->head == FENCEPOST_HEAD)); assert(ok_address(m, p)); @@ -3446,12 +3325,10 @@ static void do_check_top_chunk(mstate m, mchunkptr p) { assert(sz == ((sp->base + sp->size) - (char *)p) - TOP_FOOT_SIZE); assert(pinuse(p)); assert(!pinuse(chunk_plus_offset(p, sz))); - } /* Check properties of (inuse) mmapped chunks */ static void do_check_mmapped_chunk(mstate m, mchunkptr p) { - size_t sz = chunksize(p); size_t len = (sz + (p->prev_foot) + MMAP_FOOT_PAD); assert(is_mmapped(p)); @@ -3462,24 +3339,20 @@ static void do_check_mmapped_chunk(mstate m, mchunkptr p) { assert((len & (mparams.page_size - SIZE_T_ONE)) == 0); assert(chunk_plus_offset(p, sz)->head == FENCEPOST_HEAD); assert(chunk_plus_offset(p, sz + SIZE_T_SIZE)->head == 0); - } /* Check properties of inuse chunks */ static void do_check_inuse_chunk(mstate m, mchunkptr p) { - do_check_any_chunk(m, p); assert(is_inuse(p)); assert(next_pinuse(p)); /* If not pinuse and not mmapped, previous chunk has OK offset */ assert(is_mmapped(p) || pinuse(p) || next_chunk(prev_chunk(p)) == p); if (is_mmapped(p)) do_check_mmapped_chunk(m, p); - } /* Check properties of free chunks */ static void do_check_free_chunk(mstate m, mchunkptr p) { - size_t sz = chunksize(p); mchunkptr next = chunk_plus_offset(p, sz); do_check_any_chunk(m, p); @@ -3487,9 +3360,7 @@ static void do_check_free_chunk(mstate m, mchunkptr p) { assert(!next_pinuse(p)); assert(!is_mmapped(p)); if (p != m->dv && p != m->top) { - if (sz >= MIN_CHUNK_SIZE) { - assert((sz & CHUNK_ALIGN_MASK) == 0); assert(is_aligned(chunk2mem(p))); assert(next->prev_foot == sz); @@ -3498,19 +3369,15 @@ static void do_check_free_chunk(mstate m, mchunkptr p) { assert(p->fd->bk == p); assert(p->bk->fd == p); - } else /* markers are always of size SIZE_T_SIZE */ + } else /* markers are always of size SIZE_T_SIZE */ assert(sz == SIZE_T_SIZE); - } - } /* Check properties of malloced chunks at the point they are malloced */ static void do_check_malloced_chunk(mstate m, void *mem, size_t s) { - if (mem != 0) { - mchunkptr p = mem2chunk(mem); size_t sz = p->head & ~INUSE_BITS; do_check_inuse_chunk(m, p); @@ -3519,14 +3386,11 @@ static void do_check_malloced_chunk(mstate m, void *mem, size_t s) { assert(sz >= s); /* unless mmapped, size is less than MIN_CHUNK_SIZE more than request */ assert(is_mmapped(p) || sz < (s + MIN_CHUNK_SIZE)); - } - } /* Check a tree and its subtrees. */ static void do_check_tree(mstate m, tchunkptr t) { - tchunkptr head = 0; tchunkptr u = t; bindex_t tindex = t->index; @@ -3538,7 +3402,7 @@ static void do_check_tree(mstate m, tchunkptr t) { assert(tsize >= minsize_for_tree_index(idx)); assert((idx == NTREEBINS - 1) || (tsize < minsize_for_tree_index((idx + 1)))); - do { /* traverse through chain of same-sized nodes */ + do { /* traverse through chain of same-sized nodes */ do_check_any_chunk(m, ((mchunkptr)u)); assert(u->index == tindex); assert(chunksize(u) == tsize); @@ -3547,39 +3411,30 @@ static void do_check_tree(mstate m, tchunkptr t) { assert(u->fd->bk == u); assert(u->bk->fd == u); if (u->parent == 0) { - assert(u->child[0] == 0); assert(u->child[1] == 0); } else { - - assert(head == 0); /* only one node on chain has parent */ + assert(head == 0); /* only one node on chain has parent */ head = u; assert(u->parent != u); assert(u->parent->child[0] == u || u->parent->child[1] == u || *((tbinptr *)(u->parent)) == u); if (u->child[0] != 0) { - assert(u->child[0]->parent == u); assert(u->child[0] != u); do_check_tree(m, u->child[0]); - } if (u->child[1] != 0) { - assert(u->child[1]->parent == u); assert(u->child[1] != u); do_check_tree(m, u->child[1]); - } if (u->child[0] != 0 && u->child[1] != 0) { - assert(chunksize(u->child[0]) < chunksize(u->child[1])); - } - } u = u->fd; @@ -3587,31 +3442,25 @@ static void do_check_tree(mstate m, tchunkptr t) { } while (u != t); assert(head != 0); - } /* Check all the chunks in a treebin. */ static void do_check_treebin(mstate m, bindex_t i) { - - tbinptr * tb = treebin_at(m, i); + tbinptr *tb = treebin_at(m, i); tchunkptr t = *tb; int empty = (m->treemap & (1U << i)) == 0; if (t == 0) assert(empty); if (!empty) do_check_tree(m, t); - } /* Check all the chunks in a smallbin. */ static void do_check_smallbin(mstate m, bindex_t i) { - sbinptr b = smallbin_at(m, i); mchunkptr p = b->bk; unsigned int empty = (m->smallmap & (1U << i)) == 0; if (p == b) assert(empty); if (!empty) { - for (; p != b; p = p->bk) { - size_t size = chunksize(p); mchunkptr q; /* each chunk claims to be free */ @@ -3622,113 +3471,83 @@ static void do_check_smallbin(mstate m, bindex_t i) { /* chunk is followed by an inuse chunk */ q = next_chunk(p); if (q->head != FENCEPOST_HEAD) do_check_inuse_chunk(m, q); - } - } - } /* Find x in a bin. Used in other check functions. */ static int bin_find(mstate m, mchunkptr x) { - size_t size = chunksize(x); if (is_small(size)) { - bindex_t sidx = small_index(size); sbinptr b = smallbin_at(m, sidx); if (smallmap_is_marked(m, sidx)) { - mchunkptr p = b; do { - if (p == x) return 1; } while ((p = p->fd) != b); - } } else { - bindex_t tidx; compute_tree_index(size, tidx); if (treemap_is_marked(m, tidx)) { - tchunkptr t = *treebin_at(m, tidx); size_t sizebits = size << leftshift_for_tree_index(tidx); while (t != 0 && chunksize(t) != size) { - t = t->child[(sizebits >> (SIZE_T_BITSIZE - SIZE_T_ONE)) & 1]; sizebits <<= 1; - } if (t != 0) { - tchunkptr u = t; do { - if (u == (tchunkptr)x) return 1; } while ((u = u->fd) != t); - } - } - } return 0; - } /* Traverse each chunk and check it; return total */ static size_t traverse_and_check(mstate m) { - size_t sum = 0; if (is_initialized(m)) { - msegmentptr s = &m->seg; sum += m->topsize + TOP_FOOT_SIZE; while (s != 0) { - mchunkptr q = align_as_chunk(s->base); mchunkptr lastq = 0; assert(pinuse(q)); while (segment_holds(s, q) && q != m->top && q->head != FENCEPOST_HEAD) { - sum += chunksize(q); if (is_inuse(q)) { - assert(!bin_find(m, q)); do_check_inuse_chunk(m, q); } else { - assert(q == m->dv || bin_find(m, q)); - assert(lastq == 0 || is_inuse(lastq)); /* Not 2 consecutive free */ + assert(lastq == 0 || is_inuse(lastq)); /* Not 2 consecutive free */ do_check_free_chunk(m, q); - } lastq = q; q = next_chunk(q); - } s = s->next; - } - } return sum; - } /* Check all properties of malloc_state. */ static void do_check_malloc_state(mstate m) { - bindex_t i; size_t total; /* check bins */ @@ -3737,67 +3556,55 @@ static void do_check_malloc_state(mstate m) { for (i = 0; i < NTREEBINS; ++i) do_check_treebin(m, i); - if (m->dvsize != 0) { /* check dv chunk */ + if (m->dvsize != 0) { /* check dv chunk */ do_check_any_chunk(m, m->dv); assert(m->dvsize == chunksize(m->dv)); assert(m->dvsize >= MIN_CHUNK_SIZE); assert(bin_find(m, m->dv) == 0); - } - if (m->top != 0) { /* check top chunk */ + if (m->top != 0) { /* check top chunk */ do_check_top_chunk(m, m->top); /*assert(m->topsize == chunksize(m->top)); redundant */ assert(m->topsize > 0); assert(bin_find(m, m->top) == 0); - } total = traverse_and_check(m); assert(total <= m->footprint); assert(m->footprint <= m->max_footprint); - } - #endif /* DEBUG */ + #endif /* DEBUG */ /* ----------------------------- statistics ------------------------------ */ #if !NO_MALLINFO static struct mallinfo internal_mallinfo(mstate m) { - struct mallinfo nm = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; ensure_initialization(); if (!PREACTION(m)) { - check_malloc_state(m); if (is_initialized(m)) { - - size_t nfree = SIZE_T_ONE; /* top always free */ + size_t nfree = SIZE_T_ONE; /* top always free */ size_t mfree = m->topsize + TOP_FOOT_SIZE; size_t sum = mfree; msegmentptr s = &m->seg; while (s != 0) { - mchunkptr q = align_as_chunk(s->base); while (segment_holds(s, q) && q != m->top && q->head != FENCEPOST_HEAD) { - size_t sz = chunksize(q); sum += sz; if (!is_inuse(q)) { - mfree += sz; ++nfree; - } q = next_chunk(q); - } s = s->next; - } nm.arena = sum; @@ -3807,63 +3614,50 @@ static struct mallinfo internal_mallinfo(mstate m) { nm.uordblks = m->footprint - mfree; nm.fordblks = mfree; nm.keepcost = m->topsize; - } POSTACTION(m); - } return nm; - } - #endif /* !NO_MALLINFO */ + #endif /* !NO_MALLINFO */ #if !NO_MALLOC_STATS static void internal_malloc_stats(mstate m) { - ensure_initialization(); if (!PREACTION(m)) { - size_t maxfp = 0; size_t fp = 0; size_t used = 0; check_malloc_state(m); if (is_initialized(m)) { - msegmentptr s = &m->seg; maxfp = m->max_footprint; fp = m->footprint; used = fp - (m->topsize + TOP_FOOT_SIZE); while (s != 0) { - mchunkptr q = align_as_chunk(s->base); while (segment_holds(s, q) && q != m->top && q->head != FENCEPOST_HEAD) { - if (!is_inuse(q)) used -= chunksize(q); q = next_chunk(q); - } s = s->next; - } - } - POSTACTION(m); /* drop lock */ + POSTACTION(m); /* drop lock */ fprintf(stderr, "max system bytes = %10lu\n", (unsigned long)(maxfp)); fprintf(stderr, "system bytes = %10lu\n", (unsigned long)(fp)); fprintf(stderr, "in use bytes = %10lu\n", (unsigned long)(used)); - } - } - #endif /* NO_MALLOC_STATS */ + #endif /* NO_MALLOC_STATS */ /* ----------------------- Operations on smallbins ----------------------- */ @@ -3877,7 +3671,6 @@ static void internal_malloc_stats(mstate m) { /* Link a free chunk into a smallbin */ #define insert_small_chunk(M, P, S) \ { \ - \ bindex_t I = small_index(S); \ mchunkptr B = smallbin_at(M, I); \ mchunkptr F = B; \ @@ -3887,21 +3680,17 @@ static void internal_malloc_stats(mstate m) { else if (RTCHECK(ok_address(M, B->fd))) \ F = B->fd; \ else { \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ B->fd = P; \ F->bk = P; \ P->fd = F; \ P->bk = B; \ - \ } /* Unlink a chunk from a smallbin */ #define unlink_small_chunk(M, P, S) \ { \ - \ mchunkptr F = P->fd; \ mchunkptr B = P->bk; \ bindex_t I = small_index(S); \ @@ -3910,74 +3699,55 @@ static void internal_malloc_stats(mstate m) { assert(chunksize(P) == small_index2size(I)); \ if (RTCHECK(F == smallbin_at(M, I) || \ (ok_address(M, F) && F->bk == P))) { \ - \ if (B == F) { \ - \ clear_smallmap(M, I); \ \ } else if (RTCHECK(B == smallbin_at(M, I) || \ \ - \ (ok_address(M, B) && B->fd == P))) { \ - \ F->bk = B; \ B->fd = F; \ \ } else { \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ \ } else { \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ - \ } /* Unlink the first chunk from a smallbin */ #define unlink_first_small_chunk(M, B, P, I) \ { \ - \ mchunkptr F = P->fd; \ assert(P != B); \ assert(P != F); \ assert(chunksize(P) == small_index2size(I)); \ if (B == F) { \ - \ clear_smallmap(M, I); \ \ } else if (RTCHECK(ok_address(M, F) && F->bk == P)) { \ - \ F->bk = B; \ B->fd = F; \ \ } else { \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ - \ } /* Replace dv node, binning the old one */ /* Used only when dvsize known to be small */ #define replace_dv(M, P, S) \ { \ - \ size_t DVS = M->dvsize; \ assert(is_small(DVS)); \ if (DVS != 0) { \ - \ mchunkptr DV = M->dv; \ insert_small_chunk(M, DV, DVS); \ - \ } \ M->dvsize = S; \ M->dv = P; \ - \ } /* ------------------------- Operations on trees ------------------------- */ @@ -3985,7 +3755,6 @@ static void internal_malloc_stats(mstate m) { /* Insert chunk into tree */ #define insert_large_chunk(M, X, S) \ { \ - \ tbinptr *H; \ bindex_t I; \ compute_tree_index(S, I); \ @@ -3993,44 +3762,35 @@ static void internal_malloc_stats(mstate m) { X->index = I; \ X->child[0] = X->child[1] = 0; \ if (!treemap_is_marked(M, I)) { \ - \ mark_treemap(M, I); \ *H = X; \ X->parent = (tchunkptr)H; \ X->fd = X->bk = X; \ \ } else { \ - \ tchunkptr T = *H; \ size_t K = S << leftshift_for_tree_index(I); \ for (;;) { \ - \ if (chunksize(T) != S) { \ - \ tchunkptr *C = \ &(T->child[(K >> (SIZE_T_BITSIZE - SIZE_T_ONE)) & 1]); \ K <<= 1; \ if (*C != 0) \ T = *C; \ else if (RTCHECK(ok_address(M, C))) { \ - \ *C = X; \ X->parent = T; \ X->fd = X->bk = X; \ break; \ \ } else { \ - \ CORRUPTION_ERROR_ACTION(M); \ break; \ - \ } \ \ } else { \ - \ tchunkptr F = T->fd; \ if (RTCHECK(ok_address(M, T) && ok_address(M, F))) { \ - \ T->fd = F->bk = X; \ X->fd = F; \ X->bk = T; \ @@ -4038,18 +3798,12 @@ static void internal_malloc_stats(mstate m) { break; \ \ } else { \ - \ CORRUPTION_ERROR_ACTION(M); \ break; \ - \ } \ - \ } \ - \ } \ - \ } \ - \ } /* @@ -4071,57 +3825,41 @@ static void internal_malloc_stats(mstate m) { #define unlink_large_chunk(M, X) \ { \ - \ tchunkptr XP = X->parent; \ tchunkptr R; \ if (X->bk != X) { \ - \ tchunkptr F = X->fd; \ R = X->bk; \ if (RTCHECK(ok_address(M, F) && F->bk == X && R->fd == X)) { \ - \ F->bk = R; \ R->fd = F; \ \ } else { \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ \ } else { \ - \ tchunkptr *RP; \ if (((R = *(RP = &(X->child[1]))) != 0) || \ ((R = *(RP = &(X->child[0]))) != 0)) { \ - \ tchunkptr *CP; \ while ((*(CP = &(R->child[1])) != 0) || \ (*(CP = &(R->child[0])) != 0)) { \ - \ R = *(RP = CP); \ - \ } \ if (RTCHECK(ok_address(M, RP))) \ *RP = 0; \ else { \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ - \ } \ - \ } \ if (XP != 0) { \ - \ tbinptr *H = treebin_at(M, X->index); \ if (X == *H) { \ - \ if ((*H = R) == 0) clear_treemap(M, X->index); \ \ } else if (RTCHECK(ok_address(M, XP))) { \ - \ if (XP->child[0] == X) \ XP->child[0] = R; \ else \ @@ -4129,68 +3867,49 @@ static void internal_malloc_stats(mstate m) { \ } else \ \ - \ CORRUPTION_ERROR_ACTION(M); \ if (R != 0) { \ - \ if (RTCHECK(ok_address(M, R))) { \ - \ tchunkptr C0, C1; \ R->parent = XP; \ if ((C0 = X->child[0]) != 0) { \ - \ if (RTCHECK(ok_address(M, C0))) { \ - \ R->child[0] = C0; \ C0->parent = R; \ \ } else \ \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ if ((C1 = X->child[1]) != 0) { \ - \ if (RTCHECK(ok_address(M, C1))) { \ - \ R->child[1] = C1; \ C1->parent = R; \ \ } else \ \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ \ } else \ \ - \ CORRUPTION_ERROR_ACTION(M); \ - \ } \ - \ } \ - \ } /* Relays to large vs small bin operations */ #define insert_chunk(M, P, S) \ if (is_small(S)) insert_small_chunk(M, P, S) else { \ - \ tchunkptr TP = (tchunkptr)(P); \ insert_large_chunk(M, TP, S); \ - \ } #define unlink_chunk(M, P, S) \ if (is_small(S)) unlink_small_chunk(M, P, S) else { \ - \ tchunkptr TP = (tchunkptr)(P); \ unlink_large_chunk(M, TP); \ - \ } /* Relays to internal calls to malloc/free from realloc, memalign etc */ @@ -4198,7 +3917,7 @@ static void internal_malloc_stats(mstate m) { #if ONLY_MSPACES #define internal_malloc(m, b) mspace_malloc(m, b) #define internal_free(m, mem) mspace_free(m, mem); - #else /* ONLY_MSPACES */ + #else /* ONLY_MSPACES */ #if MSPACES #define internal_malloc(m, b) \ ((m == gm) ? dlmalloc(b) : mspace_malloc(m, b)) @@ -4207,11 +3926,11 @@ static void internal_malloc_stats(mstate m) { dlfree(mem); \ else \ mspace_free(m, mem); - #else /* MSPACES */ + #else /* MSPACES */ #define internal_malloc(m, b) dlmalloc(b) #define internal_free(m, mem) dlfree(mem) - #endif /* MSPACES */ - #endif /* ONLY_MSPACES */ + #endif /* MSPACES */ + #endif /* ONLY_MSPACES */ /* ----------------------- Direct-mmapping chunks ----------------------- */ @@ -4225,19 +3944,15 @@ static void internal_malloc_stats(mstate m) { /* Malloc using mmap */ static void *mmap_alloc(mstate m, size_t nb) { - size_t mmsize = mmap_align(nb + SIX_SIZE_T_SIZES + CHUNK_ALIGN_MASK); if (m->footprint_limit != 0) { - size_t fp = m->footprint + mmsize; if (fp <= m->footprint || fp > m->footprint_limit) return 0; - } - if (mmsize > nb) { /* Check for wrap around 0 */ + if (mmsize > nb) { /* Check for wrap around 0 */ char *mm = (char *)(CALL_DIRECT_MMAP(mmsize)); if (mm != CMFAIL) { - size_t offset = align_offset(chunk2mem(mm)); size_t psize = mmsize - offset - MMAP_FOOT_PAD; mchunkptr p = (mchunkptr)(mm + offset); @@ -4253,35 +3968,29 @@ static void *mmap_alloc(mstate m, size_t nb) { assert(is_aligned(chunk2mem(p))); check_mmapped_chunk(m, p); return chunk2mem(p); - } - } return 0; - } /* Realloc using mmap */ static mchunkptr mmap_resize(mstate m, mchunkptr oldp, size_t nb, int flags) { - size_t oldsize = chunksize(oldp); - (void)flags; /* placate people compiling -Wunused */ - if (is_small(nb)) /* Can't shrink mmap regions below small size */ + (void)flags; /* placate people compiling -Wunused */ + if (is_small(nb)) /* Can't shrink mmap regions below small size */ return 0; /* Keep old chunk if big enough but not too big */ if (oldsize >= nb + SIZE_T_SIZE && (oldsize - nb) <= (mparams.granularity << 1)) return oldp; else { - size_t offset = oldp->prev_foot; size_t oldmmsize = oldsize + offset + MMAP_FOOT_PAD; size_t newmmsize = mmap_align(nb + SIX_SIZE_T_SIZES + CHUNK_ALIGN_MASK); - char * cp = + char *cp = (char *)CALL_MREMAP((char *)oldp - offset, oldmmsize, newmmsize, flags); if (cp != CMFAIL) { - mchunkptr newp = (mchunkptr)(cp + offset); size_t psize = newmmsize - offset - MMAP_FOOT_PAD; newp->head = psize; @@ -4294,20 +4003,16 @@ static mchunkptr mmap_resize(mstate m, mchunkptr oldp, size_t nb, int flags) { m->max_footprint = m->footprint; check_mmapped_chunk(m, newp); return newp; - } - } return 0; - } /* -------------------------- mspace management -------------------------- */ /* Initialize top chunk and its size */ static void init_top(mstate m, mchunkptr p, size_t psize) { - /* Ensure alignment */ size_t offset = align_offset(chunk2mem(p)); p = (mchunkptr)((char *)p + offset); @@ -4318,29 +4023,23 @@ static void init_top(mstate m, mchunkptr p, size_t psize) { p->head = psize | PINUSE_BIT; /* set size of fake trailing chunk holding overhead space only once */ chunk_plus_offset(p, psize)->head = TOP_FOOT_SIZE; - m->trim_check = mparams.trim_threshold; /* reset on each update */ - + m->trim_check = mparams.trim_threshold; /* reset on each update */ } /* Initialize bins for a new mstate that is otherwise zeroed out */ static void init_bins(mstate m) { - /* Establish circular links for smallbins */ bindex_t i; for (i = 0; i < NSMALLBINS; ++i) { - sbinptr bin = smallbin_at(m, i); bin->fd = bin->bk = bin; - } - } #if PROCEED_ON_ERROR /* default corruption action */ static void reset_on_error(mstate m) { - int i; ++malloc_corruption_error_count; /* Reinitialize fields to forget about all memory */ @@ -4353,14 +4052,12 @@ static void reset_on_error(mstate m) { for (i = 0; i < NTREEBINS; ++i) *treebin_at(m, i) = 0; init_bins(m); - } - #endif /* PROCEED_ON_ERROR */ + #endif /* PROCEED_ON_ERROR */ /* Allocate chunk and prepend remainder with chunk in successor base. */ static void *prepend_alloc(mstate m, char *newbase, char *oldbase, size_t nb) { - mchunkptr p = align_as_chunk(newbase); mchunkptr oldfirst = align_as_chunk(oldbase); size_t psize = (char *)oldfirst - (char *)p; @@ -4374,52 +4071,44 @@ static void *prepend_alloc(mstate m, char *newbase, char *oldbase, size_t nb) { /* consolidate remainder with first chunk of old base */ if (oldfirst == m->top) { - size_t tsize = m->topsize += qsize; m->top = q; q->head = tsize | PINUSE_BIT; check_top_chunk(m, q); } else if (oldfirst == m->dv) { - size_t dsize = m->dvsize += qsize; m->dv = q; set_size_and_pinuse_of_free_chunk(q, dsize); } else { - if (!is_inuse(oldfirst)) { - size_t nsize = chunksize(oldfirst); unlink_chunk(m, oldfirst, nsize); oldfirst = chunk_plus_offset(oldfirst, nsize); qsize += nsize; - } set_free_with_pinuse(q, qsize, oldfirst); insert_chunk(m, q, qsize); check_free_chunk(m, q); - } check_malloced_chunk(m, chunk2mem(p), nb); return chunk2mem(p); - } /* Add a segment to hold a new noncontiguous region */ static void add_segment(mstate m, char *tbase, size_t tsize, flag_t mmapped) { - /* Determine locations and sizes of segment, fenceposts, old top */ - char * old_top = (char *)m->top; + char *old_top = (char *)m->top; msegmentptr oldsp = segment_holding(m, old_top); - char * old_end = oldsp->base + oldsp->size; + char *old_end = oldsp->base + oldsp->size; size_t ssize = pad_request(sizeof(struct malloc_segment)); - char * rawsp = old_end - (ssize + FOUR_SIZE_T_SIZES + CHUNK_ALIGN_MASK); + char *rawsp = old_end - (ssize + FOUR_SIZE_T_SIZES + CHUNK_ALIGN_MASK); size_t offset = align_offset(chunk2mem(rawsp)); - char * asp = rawsp + offset; - char * csp = (asp < (old_top + MIN_CHUNK_SIZE)) ? old_top : asp; + char *asp = rawsp + offset; + char *csp = (asp < (old_top + MIN_CHUNK_SIZE)) ? old_top : asp; mchunkptr sp = (mchunkptr)csp; msegmentptr ss = (msegmentptr)(chunk2mem(sp)); mchunkptr tnext = chunk_plus_offset(sp, ssize); @@ -4432,7 +4121,7 @@ static void add_segment(mstate m, char *tbase, size_t tsize, flag_t mmapped) { /* Set up segment record */ assert(is_aligned(ss)); set_size_and_pinuse_of_inuse_chunk(m, sp, ssize); - *ss = m->seg; /* Push current record */ + *ss = m->seg; /* Push current record */ m->seg.base = tbase; m->seg.size = tsize; m->seg.sflags = mmapped; @@ -4440,7 +4129,6 @@ static void add_segment(mstate m, char *tbase, size_t tsize, flag_t mmapped) { /* Insert trailing fenceposts */ for (;;) { - mchunkptr nextp = chunk_plus_offset(p, SIZE_T_SIZE); p->head = FENCEPOST_HEAD; ++nfences; @@ -4448,53 +4136,44 @@ static void add_segment(mstate m, char *tbase, size_t tsize, flag_t mmapped) { p = nextp; else break; - } assert(nfences >= 2); /* Insert the rest of old top into a bin as an ordinary free chunk */ if (csp != old_top) { - mchunkptr q = (mchunkptr)old_top; size_t psize = csp - old_top; mchunkptr tn = chunk_plus_offset(q, psize); set_free_with_pinuse(q, psize, tn); insert_chunk(m, q, psize); - } check_top_chunk(m, m->top); - } /* -------------------------- System allocation -------------------------- */ /* Get memory from system using MORECORE or MMAP */ static void *sys_alloc(mstate m, size_t nb) { - - char * tbase = CMFAIL; + char *tbase = CMFAIL; size_t tsize = 0; flag_t mmap_flag = 0; - size_t asize; /* allocation size */ + size_t asize; /* allocation size */ ensure_initialization(); /* Directly map large chunks, but only if already initialized */ if (use_mmap(m) && nb >= mparams.mmap_threshold && m->topsize != 0) { - void *mem = mmap_alloc(m, nb); if (mem != 0) return mem; - } asize = granularity_align(nb + SYS_ALLOC_PADDING); - if (asize <= nb) return 0; /* wraparound */ + if (asize <= nb) return 0; /* wraparound */ if (m->footprint_limit != 0) { - size_t fp = m->footprint + asize; if (fp <= m->footprint || fp > m->footprint_limit) return 0; - } /* @@ -4520,99 +4199,78 @@ static void *sys_alloc(mstate m, size_t nb) { */ if (MORECORE_CONTIGUOUS && !use_noncontiguous(m)) { - - char * br = CMFAIL; - size_t ssize = asize; /* sbrk call size */ + char *br = CMFAIL; + size_t ssize = asize; /* sbrk call size */ msegmentptr ss = (m->top == 0) ? 0 : segment_holding(m, (char *)m->top); ACQUIRE_MALLOC_GLOBAL_LOCK(); - if (ss == 0) { /* First time through or recovery */ + if (ss == 0) { /* First time through or recovery */ char *base = (char *)CALL_MORECORE(0); if (base != CMFAIL) { - size_t fp; /* Adjust to end on a page boundary */ if (!is_page_aligned(base)) ssize += (page_align((size_t)base) - (size_t)base); - fp = m->footprint + ssize; /* recheck limits */ + fp = m->footprint + ssize; /* recheck limits */ if (ssize > nb && ssize < HALF_MAX_SIZE_T && (m->footprint_limit == 0 || (fp > m->footprint && fp <= m->footprint_limit)) && (br = (char *)(CALL_MORECORE(ssize))) == base) { - tbase = base; tsize = ssize; - } - } } else { - /* Subtract out existing available top space from MORECORE request. */ ssize = granularity_align(nb - m->topsize + SYS_ALLOC_PADDING); /* Use mem here only if it did continuously extend old space */ if (ssize < HALF_MAX_SIZE_T && (br = (char *)(CALL_MORECORE(ssize))) == ss->base + ss->size) { - tbase = br; tsize = ssize; - } - } - if (tbase == CMFAIL) { /* Cope with partial failure */ - if (br != CMFAIL) { /* Try to use/extend the space we did get */ + if (tbase == CMFAIL) { /* Cope with partial failure */ + if (br != CMFAIL) { /* Try to use/extend the space we did get */ if (ssize < HALF_MAX_SIZE_T && ssize < nb + SYS_ALLOC_PADDING) { - size_t esize = granularity_align(nb + SYS_ALLOC_PADDING - ssize); if (esize < HALF_MAX_SIZE_T) { - char *end = (char *)CALL_MORECORE(esize); if (end != CMFAIL) ssize += esize; - else { /* Can't use; try to release */ + else { /* Can't use; try to release */ (void)CALL_MORECORE(-ssize); br = CMFAIL; - } - } - } - } - if (br != CMFAIL) { /* Use the space we did get */ + if (br != CMFAIL) { /* Use the space we did get */ tbase = br; tsize = ssize; } else - disable_contiguous(m); /* Don't try contiguous path in the future */ - + disable_contiguous(m); /* Don't try contiguous path in the future */ } RELEASE_MALLOC_GLOBAL_LOCK(); - } - if (HAVE_MMAP && tbase == CMFAIL) { /* Try MMAP */ + if (HAVE_MMAP && tbase == CMFAIL) { /* Try MMAP */ char *mp = (char *)(CALL_MMAP(asize)); if (mp != CMFAIL) { - tbase = mp; tsize = asize; mmap_flag = USE_MMAP_BIT; - } - } - if (HAVE_MORECORE && tbase == CMFAIL) { /* Try noncontiguous MORECORE */ + if (HAVE_MORECORE && tbase == CMFAIL) { /* Try noncontiguous MORECORE */ if (asize < HALF_MAX_SIZE_T) { - char *br = CMFAIL; char *end = CMFAIL; ACQUIRE_MALLOC_GLOBAL_LOCK(); @@ -4620,27 +4278,20 @@ static void *sys_alloc(mstate m, size_t nb) { end = (char *)(CALL_MORECORE(0)); RELEASE_MALLOC_GLOBAL_LOCK(); if (br != CMFAIL && end != CMFAIL && br < end) { - size_t ssize = end - br; if (ssize > nb + TOP_FOOT_SIZE) { - tbase = br; tsize = ssize; - } - } - } - } if (tbase != CMFAIL) { - if ((m->footprint += tsize) > m->max_footprint) m->max_footprint = m->footprint; - if (!is_initialized(m)) { /* first-time initialization */ + if (!is_initialized(m)) { /* first-time initialization */ if (m->least_addr == 0 || tbase < m->least_addr) m->least_addr = tbase; m->seg.base = tbase; m->seg.size = tsize; @@ -4658,13 +4309,11 @@ static void *sys_alloc(mstate m, size_t nb) { /* Offset top by embedded malloc_state */ mchunkptr mn = next_chunk(mem2chunk(m)); init_top(m, mn, (size_t)((tbase + tsize) - (char *)mn) - TOP_FOOT_SIZE); - } } else { - /* Try to merge with an existing segment */ msegmentptr sp = &m->seg; /* Only consider most recent segment if traversal suppressed */ @@ -4672,19 +4321,17 @@ static void *sys_alloc(mstate m, size_t nb) { sp = (NO_SEGMENT_TRAVERSAL) ? 0 : sp->next; if (sp != 0 && !is_extern_segment(sp) && (sp->sflags & USE_MMAP_BIT) == mmap_flag && - segment_holds(sp, m->top)) { /* append */ + segment_holds(sp, m->top)) { /* append */ sp->size += tsize; init_top(m, m->top, m->topsize + tsize); } else { - if (tbase < m->least_addr) m->least_addr = tbase; sp = &m->seg; while (sp != 0 && sp->base != tbase + tsize) sp = (NO_SEGMENT_TRAVERSAL) ? 0 : sp->next; if (sp != 0 && !is_extern_segment(sp) && (sp->sflags & USE_MMAP_BIT) == mmap_flag) { - char *oldbase = sp->base; sp->base = tbase; sp->size += tsize; @@ -4693,12 +4340,10 @@ static void *sys_alloc(mstate m, size_t nb) { } else add_segment(m, tbase, tsize, mmap_flag); - } - } - if (nb < m->topsize) { /* Allocate from new or extended top space */ + if (nb < m->topsize) { /* Allocate from new or extended top space */ size_t rsize = m->topsize -= nb; mchunkptr p = m->top; mchunkptr r = m->top = chunk_plus_offset(p, nb); @@ -4707,74 +4352,59 @@ static void *sys_alloc(mstate m, size_t nb) { check_top_chunk(m, m->top); check_malloced_chunk(m, chunk2mem(p), nb); return chunk2mem(p); - } - } MALLOC_FAILURE_ACTION; return 0; - } /* ----------------------- system deallocation -------------------------- */ /* Unmap and unlink any mmapped segments that don't contain used chunks */ static size_t release_unused_segments(mstate m) { - size_t released = 0; int nsegs = 0; msegmentptr pred = &m->seg; msegmentptr sp = pred->next; while (sp != 0) { - - char * base = sp->base; + char *base = sp->base; size_t size = sp->size; msegmentptr next = sp->next; ++nsegs; if (is_mmapped_segment(sp) && !is_extern_segment(sp)) { - mchunkptr p = align_as_chunk(base); size_t psize = chunksize(p); /* Can unmap if first chunk holds entire segment and not pinned */ if (!is_inuse(p) && (char *)p + psize >= base + size - TOP_FOOT_SIZE) { - tchunkptr tp = (tchunkptr)p; assert(segment_holds(sp, (char *)sp)); if (p == m->dv) { - m->dv = 0; m->dvsize = 0; } else { - unlink_large_chunk(m, tp); - } if (CALL_MUNMAP(base, size) == 0) { - released += size; m->footprint -= size; /* unlink obsoleted record */ sp = pred; sp->next = next; - } else { /* back out if cannot unmap */ + } else { /* back out if cannot unmap */ insert_large_chunk(m, tp, psize); - } - } - } - if (NO_SEGMENT_TRAVERSAL) /* scan only first segment */ + if (NO_SEGMENT_TRAVERSAL) /* scan only first segment */ break; pred = sp; sp = next; - } /* Reset check counter */ @@ -4782,19 +4412,15 @@ static size_t release_unused_segments(mstate m) { ? (size_t)nsegs : (size_t)MAX_RELEASE_CHECK_RATE); return released; - } static int sys_trim(mstate m, size_t pad) { - size_t released = 0; ensure_initialization(); if (pad < MAX_REQUEST && is_initialized(m)) { - - pad += TOP_FOOT_SIZE; /* ensure enough room for segment overhead */ + pad += TOP_FOOT_SIZE; /* ensure enough room for segment overhead */ if (m->topsize > pad) { - /* Shrink top space in granularity-size units, keeping at least one */ size_t unit = mparams.granularity; size_t extra = @@ -4802,58 +4428,43 @@ static int sys_trim(mstate m, size_t pad) { msegmentptr sp = segment_holding(m, (char *)m->top); if (!is_extern_segment(sp)) { - if (is_mmapped_segment(sp)) { - if (HAVE_MMAP && sp->size >= extra && - !has_segment_link(m, sp)) { /* can't shrink if pinned */ + !has_segment_link(m, sp)) { /* can't shrink if pinned */ size_t newsize = sp->size - extra; - (void)newsize; /* placate people compiling -Wunused-variable */ + (void)newsize; /* placate people compiling -Wunused-variable */ /* Prefer mremap, fall back to munmap */ if ((CALL_MREMAP(sp->base, sp->size, newsize, 0) != MFAIL) || (CALL_MUNMAP(sp->base + newsize, extra) == 0)) { - released = extra; - } - } } else if (HAVE_MORECORE) { - - if (extra >= HALF_MAX_SIZE_T) /* Avoid wrapping negative */ + if (extra >= HALF_MAX_SIZE_T) /* Avoid wrapping negative */ extra = (HALF_MAX_SIZE_T) + SIZE_T_ONE - unit; ACQUIRE_MALLOC_GLOBAL_LOCK(); { - /* Make sure end of memory is where we last set it. */ char *old_br = (char *)(CALL_MORECORE(0)); if (old_br == sp->base + sp->size) { - char *rel_br = (char *)(CALL_MORECORE(-extra)); char *new_br = (char *)(CALL_MORECORE(0)); if (rel_br != CMFAIL && new_br < old_br) released = old_br - new_br; - } - } RELEASE_MALLOC_GLOBAL_LOCK(); - } - } if (released != 0) { - sp->size -= released; m->footprint -= released; init_top(m, m->top, m->topsize - released); check_top_chunk(m, m->top); - } - } /* Unmap any unused mmapped segments */ @@ -4861,215 +4472,165 @@ static int sys_trim(mstate m, size_t pad) { /* On failure, disable autotrim to avoid repeated failed future calls */ if (released == 0 && m->topsize > m->trim_check) m->trim_check = MAX_SIZE_T; - } return (released != 0) ? 1 : 0; - } /* Consolidate and bin a chunk. Differs from exported versions of free mainly in that the chunk need not be marked as inuse. */ static void dispose_chunk(mstate m, mchunkptr p, size_t psize) { - mchunkptr next = chunk_plus_offset(p, psize); if (!pinuse(p)) { - mchunkptr prev; size_t prevsize = p->prev_foot; if (is_mmapped(p)) { - psize += prevsize + MMAP_FOOT_PAD; if (CALL_MUNMAP((char *)p - prevsize, psize) == 0) m->footprint -= psize; return; - } prev = chunk_minus_offset(p, prevsize); psize += prevsize; p = prev; - if (RTCHECK(ok_address(m, prev))) { /* consolidate backward */ + if (RTCHECK(ok_address(m, prev))) { /* consolidate backward */ if (p != m->dv) { - unlink_chunk(m, p, prevsize); } else if ((next->head & INUSE_BITS) == INUSE_BITS) { - m->dvsize = psize; set_free_with_pinuse(p, psize, next); return; - } } else { - CORRUPTION_ERROR_ACTION(m); return; - } - } if (RTCHECK(ok_address(m, next))) { - - if (!cinuse(next)) { /* consolidate forward */ + if (!cinuse(next)) { /* consolidate forward */ if (next == m->top) { - size_t tsize = m->topsize += psize; m->top = p; p->head = tsize | PINUSE_BIT; if (p == m->dv) { - m->dv = 0; m->dvsize = 0; - } return; } else if (next == m->dv) { - size_t dsize = m->dvsize += psize; m->dv = p; set_size_and_pinuse_of_free_chunk(p, dsize); return; } else { - size_t nsize = chunksize(next); psize += nsize; unlink_chunk(m, next, nsize); set_size_and_pinuse_of_free_chunk(p, psize); if (p == m->dv) { - m->dvsize = psize; return; - } - } } else { - set_free_with_pinuse(p, psize, next); - } insert_chunk(m, p, psize); } else { - CORRUPTION_ERROR_ACTION(m); - } - } /* ---------------------------- malloc --------------------------- */ /* allocate a large request from the best fitting chunk in a treebin */ static void *tmalloc_large(mstate m, size_t nb) { - tchunkptr v = 0; - size_t rsize = -nb; /* Unsigned negation */ + size_t rsize = -nb; /* Unsigned negation */ tchunkptr t; bindex_t idx; compute_tree_index(nb, idx); if ((t = *treebin_at(m, idx)) != 0) { - /* Traverse tree for this bin looking for node with size == nb */ size_t sizebits = nb << leftshift_for_tree_index(idx); - tchunkptr rst = 0; /* The deepest untaken right subtree */ + tchunkptr rst = 0; /* The deepest untaken right subtree */ for (;;) { - tchunkptr rt; size_t trem = chunksize(t) - nb; if (trem < rsize) { - v = t; if ((rsize = trem) == 0) break; - } rt = t->child[1]; t = t->child[(sizebits >> (SIZE_T_BITSIZE - SIZE_T_ONE)) & 1]; if (rt != 0 && rt != t) rst = rt; if (t == 0) { - - t = rst; /* set t to least subtree holding sizes > nb */ + t = rst; /* set t to least subtree holding sizes > nb */ break; - } sizebits <<= 1; - } - } - if (t == 0 && v == 0) { /* set t to root of next non-empty treebin */ + if (t == 0 && v == 0) { /* set t to root of next non-empty treebin */ binmap_t leftbits = left_bits(idx2bit(idx)) & m->treemap; if (leftbits != 0) { - bindex_t i; binmap_t leastbit = least_bit(leftbits); compute_bit2idx(leastbit, i); t = *treebin_at(m, i); - } - } - while (t != 0) { /* find smallest of tree or subtree */ + while (t != 0) { /* find smallest of tree or subtree */ size_t trem = chunksize(t) - nb; if (trem < rsize) { - rsize = trem; v = t; - } t = leftmost_child(t); - } /* If dv is a better fit, return 0 so malloc will use it */ if (v != 0 && rsize < (size_t)(m->dvsize - nb)) { - - if (RTCHECK(ok_address(m, v))) { /* split */ + if (RTCHECK(ok_address(m, v))) { /* split */ mchunkptr r = chunk_plus_offset(v, nb); assert(chunksize(v) == rsize + nb); if (RTCHECK(ok_next(v, r))) { - unlink_large_chunk(m, v); if (rsize < MIN_CHUNK_SIZE) set_inuse_and_pinuse(m, v, (rsize + nb)); else { - set_size_and_pinuse_of_inuse_chunk(m, v, nb); set_size_and_pinuse_of_free_chunk(r, rsize); insert_chunk(m, r, rsize); - } return chunk2mem(v); - } - } CORRUPTION_ERROR_ACTION(m); - } return 0; - } /* allocate a small request from the best fitting chunk in a treebin */ static void *tmalloc_small(mstate m, size_t nb) { - tchunkptr t, v; size_t rsize; bindex_t i; @@ -5079,49 +4640,37 @@ static void *tmalloc_small(mstate m, size_t nb) { rsize = chunksize(t) - nb; while ((t = leftmost_child(t)) != 0) { - size_t trem = chunksize(t) - nb; if (trem < rsize) { - rsize = trem; v = t; - } - } if (RTCHECK(ok_address(m, v))) { - mchunkptr r = chunk_plus_offset(v, nb); assert(chunksize(v) == rsize + nb); if (RTCHECK(ok_next(v, r))) { - unlink_large_chunk(m, v); if (rsize < MIN_CHUNK_SIZE) set_inuse_and_pinuse(m, v, (rsize + nb)); else { - set_size_and_pinuse_of_inuse_chunk(m, v, nb); set_size_and_pinuse_of_free_chunk(r, rsize); replace_dv(m, r, rsize); - } return chunk2mem(v); - } - } CORRUPTION_ERROR_ACTION(m); return 0; - } #if !ONLY_MSPACES void *dlmalloc(size_t bytes) { - /* Basic algorithm: If a small request (< 256 bytes minus per-chunk overhead): @@ -5146,24 +4695,22 @@ void *dlmalloc(size_t bytes) { */ #if USE_LOCKS - ensure_initialization(); /* initialize in sys_alloc if not using locks */ + ensure_initialization(); /* initialize in sys_alloc if not using locks */ #endif if (!PREACTION(gm)) { - - void * mem; + void *mem; size_t nb; if (bytes <= MAX_SMALL_REQUEST) { - bindex_t idx; binmap_t smallbits; nb = (bytes < MIN_REQUEST) ? MIN_CHUNK_SIZE : pad_request(bytes); idx = small_index(nb); smallbits = gm->smallmap >> idx; - if ((smallbits & 0x3U) != 0) { /* Remainderless fit to a smallbin. */ + if ((smallbits & 0x3U) != 0) { /* Remainderless fit to a smallbin. */ mchunkptr b, p; - idx += ~smallbits & 1; /* Uses next bin if idx empty */ + idx += ~smallbits & 1; /* Uses next bin if idx empty */ b = smallbin_at(gm, idx); p = b->fd; assert(chunksize(p) == small_index2size(idx)); @@ -5176,8 +4723,7 @@ void *dlmalloc(size_t bytes) { } else if (nb > gm->dvsize) { - - if (smallbits != 0) { /* Use chunk in next nonempty smallbin */ + if (smallbits != 0) { /* Use chunk in next nonempty smallbin */ mchunkptr b, p, r; size_t rsize; bindex_t i; @@ -5193,12 +4739,10 @@ void *dlmalloc(size_t bytes) { if (SIZE_T_SIZE != 4 && rsize < MIN_CHUNK_SIZE) set_inuse_and_pinuse(gm, p, small_index2size(i)); else { - set_size_and_pinuse_of_inuse_chunk(gm, p, nb); r = chunk_plus_offset(p, nb); set_size_and_pinuse_of_free_chunk(r, rsize); replace_dv(gm, r, rsize); - } mem = chunk2mem(p); @@ -5208,46 +4752,37 @@ void *dlmalloc(size_t bytes) { } else if (gm->treemap != 0 && (mem = tmalloc_small(gm, nb)) != 0) { - check_malloced_chunk(gm, mem, nb); goto postaction; - } - } } else if (bytes >= MAX_REQUEST) nb = MAX_SIZE_T; /* Too big to allocate. Force failure (in sys alloc) */ else { - nb = pad_request(bytes); if (gm->treemap != 0 && (mem = tmalloc_large(gm, nb)) != 0) { - check_malloced_chunk(gm, mem, nb); goto postaction; - } - } if (nb <= gm->dvsize) { - size_t rsize = gm->dvsize - nb; mchunkptr p = gm->dv; - if (rsize >= MIN_CHUNK_SIZE) { /* split dv */ + if (rsize >= MIN_CHUNK_SIZE) { /* split dv */ mchunkptr r = gm->dv = chunk_plus_offset(p, nb); gm->dvsize = rsize; set_size_and_pinuse_of_free_chunk(r, rsize); set_size_and_pinuse_of_inuse_chunk(gm, p, nb); - } else { /* exhaust dv */ + } else { /* exhaust dv */ size_t dvs = gm->dvsize; gm->dvsize = 0; gm->dv = 0; set_inuse_and_pinuse(gm, p, dvs); - } mem = chunk2mem(p); @@ -5256,7 +4791,7 @@ void *dlmalloc(size_t bytes) { } - else if (nb < gm->topsize) { /* Split top */ + else if (nb < gm->topsize) { /* Split top */ size_t rsize = gm->topsize -= nb; mchunkptr p = gm->top; mchunkptr r = gm->top = chunk_plus_offset(p, nb); @@ -5266,7 +4801,6 @@ void *dlmalloc(size_t bytes) { check_top_chunk(gm, gm->top); check_malloced_chunk(gm, mem, nb); goto postaction; - } mem = sys_alloc(gm, nb); @@ -5274,17 +4808,14 @@ void *dlmalloc(size_t bytes) { postaction: POSTACTION(gm); return mem; - } return 0; - } /* ---------------------------- free --------------------------- */ void dlfree(void *mem) { - /* Consolidate freed chunks with preceeding or succeeding bordering free chunks, if they exist, and then place in a bin. Intermixed @@ -5292,101 +4823,79 @@ void dlfree(void *mem) { */ if (mem != 0) { - mchunkptr p = mem2chunk(mem); #if FOOTERS mstate fm = get_mstate_for(p); if (!ok_magic(fm)) { - USAGE_ERROR_ACTION(fm, p); return; - } - #else /* FOOTERS */ + #else /* FOOTERS */ #define fm gm - #endif /* FOOTERS */ + #endif /* FOOTERS */ if (!PREACTION(fm)) { - check_inuse_chunk(fm, p); if (RTCHECK(ok_address(fm, p) && ok_inuse(p))) { - size_t psize = chunksize(p); mchunkptr next = chunk_plus_offset(p, psize); if (!pinuse(p)) { - size_t prevsize = p->prev_foot; if (is_mmapped(p)) { - psize += prevsize + MMAP_FOOT_PAD; if (CALL_MUNMAP((char *)p - prevsize, psize) == 0) fm->footprint -= psize; goto postaction; } else { - mchunkptr prev = chunk_minus_offset(p, prevsize); psize += prevsize; p = prev; - if (RTCHECK(ok_address(fm, prev))) { /* consolidate backward */ + if (RTCHECK(ok_address(fm, prev))) { /* consolidate backward */ if (p != fm->dv) { - unlink_chunk(fm, p, prevsize); } else if ((next->head & INUSE_BITS) == INUSE_BITS) { - fm->dvsize = psize; set_free_with_pinuse(p, psize, next); goto postaction; - } } else goto erroraction; - } - } if (RTCHECK(ok_next(p, next) && ok_pinuse(next))) { - - if (!cinuse(next)) { /* consolidate forward */ + if (!cinuse(next)) { /* consolidate forward */ if (next == fm->top) { - size_t tsize = fm->topsize += psize; fm->top = p; p->head = tsize | PINUSE_BIT; if (p == fm->dv) { - fm->dv = 0; fm->dvsize = 0; - } if (should_trim(fm, tsize)) sys_trim(fm, 0); goto postaction; } else if (next == fm->dv) { - size_t dsize = fm->dvsize += psize; fm->dv = p; set_size_and_pinuse_of_free_chunk(p, dsize); goto postaction; } else { - size_t nsize = chunksize(next); psize += nsize; unlink_chunk(fm, next, nsize); set_size_and_pinuse_of_free_chunk(p, psize); if (p == fm->dv) { - fm->dvsize = psize; goto postaction; - } - } } else @@ -5394,95 +4903,78 @@ void dlfree(void *mem) { set_free_with_pinuse(p, psize, next); if (is_small(psize)) { - insert_small_chunk(fm, p, psize); check_free_chunk(fm, p); } else { - tchunkptr tp = (tchunkptr)p; insert_large_chunk(fm, tp, psize); check_free_chunk(fm, p); if (--fm->release_checks == 0) release_unused_segments(fm); - } goto postaction; - } - } erroraction: USAGE_ERROR_ACTION(fm, p); postaction: POSTACTION(fm); - } - } #if !FOOTERS #undef fm - #endif /* FOOTERS */ - + #endif /* FOOTERS */ } void *dlcalloc(size_t n_elements, size_t elem_size) { - - void * mem; + void *mem; size_t req = 0; if (n_elements != 0) { - req = n_elements * elem_size; if (((n_elements | elem_size) & ~(size_t)0xffff) && (req / n_elements != elem_size)) - req = MAX_SIZE_T; /* force downstream failure on overflow */ - + req = MAX_SIZE_T; /* force downstream failure on overflow */ } mem = dlmalloc(req); if (mem != 0 && calloc_must_clear(mem2chunk(mem))) __builtin_memset(mem, 0, req); return mem; - } - #endif /* !ONLY_MSPACES */ + #endif /* !ONLY_MSPACES */ /* ------------ Internal support for realloc, memalign, etc -------------- */ /* Try to realloc; only in-place unless can_move true */ static mchunkptr try_realloc_chunk(mstate m, mchunkptr p, size_t nb, int can_move) { - mchunkptr newp = 0; size_t oldsize = chunksize(p); mchunkptr next = chunk_plus_offset(p, oldsize); if (RTCHECK(ok_address(m, p) && ok_inuse(p) && ok_next(p, next) && ok_pinuse(next))) { - if (is_mmapped(p)) { - newp = mmap_resize(m, p, nb, can_move); - } else if (oldsize >= nb) { /* already big enough */ + } else if (oldsize >= nb) { /* already big enough */ size_t rsize = oldsize - nb; - if (rsize >= MIN_CHUNK_SIZE) { /* split off remainder */ + if (rsize >= MIN_CHUNK_SIZE) { /* split off remainder */ mchunkptr r = chunk_plus_offset(p, nb); set_inuse(m, p, nb); set_inuse(m, r, rsize); dispose_chunk(m, r, rsize); - } newp = p; - } else if (next == m->top) { /* extend into top */ + } else if (next == m->top) { /* extend into top */ if (oldsize + m->topsize > nb) { - size_t newsize = oldsize + m->topsize; size_t newtopsize = newsize - nb; mchunkptr newtop = chunk_plus_offset(p, nb); @@ -5491,17 +4983,14 @@ static mchunkptr try_realloc_chunk(mstate m, mchunkptr p, size_t nb, m->top = newtop; m->topsize = newtopsize; newp = p; - } - } else if (next == m->dv) { /* extend into dv */ + } else if (next == m->dv) { /* extend into dv */ size_t dvs = m->dvsize; if (oldsize + dvs >= nb) { - size_t dsize = oldsize + dvs - nb; if (dsize >= MIN_CHUNK_SIZE) { - mchunkptr r = chunk_plus_offset(p, nb); mchunkptr n = chunk_plus_offset(r, dsize); set_inuse(m, p, nb); @@ -5510,86 +4999,69 @@ static mchunkptr try_realloc_chunk(mstate m, mchunkptr p, size_t nb, m->dvsize = dsize; m->dv = r; - } else { /* exhaust dv */ + } else { /* exhaust dv */ size_t newsize = oldsize + dvs; set_inuse(m, p, newsize); m->dvsize = 0; m->dv = 0; - } newp = p; - } - } else if (!cinuse(next)) { /* extend into next free chunk */ + } else if (!cinuse(next)) { /* extend into next free chunk */ size_t nextsize = chunksize(next); if (oldsize + nextsize >= nb) { - size_t rsize = oldsize + nextsize - nb; unlink_chunk(m, next, nextsize); if (rsize < MIN_CHUNK_SIZE) { - size_t newsize = oldsize + nextsize; set_inuse(m, p, newsize); } else { - mchunkptr r = chunk_plus_offset(p, nb); set_inuse(m, p, nb); set_inuse(m, r, rsize); dispose_chunk(m, r, rsize); - } newp = p; - } - } } else { - USAGE_ERROR_ACTION(m, chunk2mem(p)); - } return newp; - } static void *internal_memalign(mstate m, size_t alignment, size_t bytes) { - void *mem = 0; - if (alignment < MIN_CHUNK_SIZE) /* must be at least a minimum chunk size */ + if (alignment < MIN_CHUNK_SIZE) /* must be at least a minimum chunk size */ alignment = MIN_CHUNK_SIZE; if ((alignment & (alignment - SIZE_T_ONE)) != 0) { /* Ensure a power of 2 */ size_t a = MALLOC_ALIGNMENT << 1; while (a < alignment) a <<= 1; alignment = a; - } if (bytes >= MAX_REQUEST - alignment) { - - if (m != 0) { /* Test isn't needed but avoids compiler warning */ + if (m != 0) { /* Test isn't needed but avoids compiler warning */ MALLOC_FAILURE_ACTION; - } } else { - size_t nb = request2size(bytes); size_t req = nb + alignment + MIN_CHUNK_SIZE - CHUNK_OVERHEAD; mem = internal_malloc(m, req); if (mem != 0) { - mchunkptr p = mem2chunk(mem); if (PREACTION(m)) return 0; - if ((((size_t)(mem)) & (alignment - 1)) != 0) { /* misaligned */ + if ((((size_t)(mem)) & (alignment - 1)) != 0) { /* misaligned */ /* Find an aligned spot inside chunk. Since we need to give back leading space in a chunk of at least MIN_CHUNK_SIZE, if @@ -5598,45 +5070,40 @@ static void *internal_memalign(mstate m, size_t alignment, size_t bytes) { We've allocated enough total room so that this is always possible. */ - char * br = (char *)mem2chunk((size_t)( - ((size_t)((char *)mem + alignment - SIZE_T_ONE)) & -alignment)); - char * pos = ((size_t)(br - (char *)(p)) >= MIN_CHUNK_SIZE) + char *br = (char *)mem2chunk( + (size_t)(((size_t)((char *)mem + alignment - SIZE_T_ONE)) & + -alignment)); + char *pos = ((size_t)(br - (char *)(p)) >= MIN_CHUNK_SIZE) ? br : br + alignment; mchunkptr newp = (mchunkptr)pos; size_t leadsize = pos - (char *)(p); size_t newsize = chunksize(p) - leadsize; - if (is_mmapped(p)) { /* For mmapped chunks, just adjust offset */ + if (is_mmapped(p)) { /* For mmapped chunks, just adjust offset */ newp->prev_foot = p->prev_foot + leadsize; newp->head = newsize; - } else { /* Otherwise, give back leader, use the rest */ + } else { /* Otherwise, give back leader, use the rest */ set_inuse(m, newp, newsize); set_inuse(m, p, leadsize); dispose_chunk(m, p, leadsize); - } p = newp; - } /* Give back spare room at the end */ if (!is_mmapped(p)) { - size_t size = chunksize(p); if (size > nb + MIN_CHUNK_SIZE) { - size_t remainder_size = size - nb; mchunkptr remainder = chunk_plus_offset(p, nb); set_inuse(m, p, nb); set_inuse(m, remainder, remainder_size); dispose_chunk(m, remainder, remainder_size); - } - } mem = chunk2mem(p); @@ -5644,13 +5111,10 @@ static void *internal_memalign(mstate m, size_t alignment, size_t bytes) { assert(((size_t)mem & (alignment - 1)) == 0); check_inuse_chunk(m, p); POSTACTION(m); - } - } return mem; - } /* @@ -5662,48 +5126,43 @@ static void *internal_memalign(mstate m, size_t alignment, size_t bytes) { */ static void **ialloc(mstate m, size_t n_elements, size_t *sizes, int opts, void *chunks[]) { - - size_t element_size; /* chunksize of each element, if all same */ - size_t contents_size; /* total size of elements */ - size_t array_size; /* request size of pointer array */ - void * mem; /* malloced aggregate space */ - mchunkptr p; /* corresponding chunk */ - size_t remainder_size; /* remaining bytes while splitting */ - void ** marray; /* either "chunks" or malloced ptr array */ - mchunkptr array_chunk; /* chunk for malloced ptr array */ - flag_t was_enabled; /* to disable mmap */ + size_t element_size; /* chunksize of each element, if all same */ + size_t contents_size; /* total size of elements */ + size_t array_size; /* request size of pointer array */ + void *mem; /* malloced aggregate space */ + mchunkptr p; /* corresponding chunk */ + size_t remainder_size; /* remaining bytes while splitting */ + void **marray; /* either "chunks" or malloced ptr array */ + mchunkptr array_chunk; /* chunk for malloced ptr array */ + flag_t was_enabled; /* to disable mmap */ size_t size; size_t i; ensure_initialization(); /* compute array length, if needed */ if (chunks != 0) { - - if (n_elements == 0) return chunks; /* nothing to do */ + if (n_elements == 0) return chunks; /* nothing to do */ marray = chunks; array_size = 0; } else { - /* if empty req, must still return chunk representing empty array */ if (n_elements == 0) return (void **)internal_malloc(m, 0); marray = 0; array_size = request2size(n_elements * (sizeof(void *))); - } /* compute total element size */ - if (opts & 0x1) { /* all-same-size */ + if (opts & 0x1) { /* all-same-size */ element_size = request2size(*sizes); contents_size = n_elements * element_size; - } else { /* add up all the sizes */ + } else { /* add up all the sizes */ element_size = 0; contents_size = 0; for (i = 0; i != n_elements; ++i) contents_size += request2size(sizes[i]); - } size = contents_size + array_size; @@ -5725,30 +5184,25 @@ static void **ialloc(mstate m, size_t n_elements, size_t *sizes, int opts, assert(!is_mmapped(p)); - if (opts & 0x2) { /* optionally clear the elements */ + if (opts & 0x2) { /* optionally clear the elements */ __builtin_memset((size_t *)mem, 0, remainder_size - SIZE_T_SIZE - array_size); - } /* If not provided, allocate the pointer array as final part of chunk */ if (marray == 0) { - size_t array_chunk_size; array_chunk = chunk_plus_offset(p, contents_size); array_chunk_size = remainder_size - contents_size; marray = (void **)(chunk2mem(array_chunk)); set_size_and_pinuse_of_inuse_chunk(m, array_chunk, array_chunk_size); remainder_size = contents_size; - } /* split out elements */ for (i = 0;; ++i) { - marray[i] = chunk2mem(p); if (i != n_elements - 1) { - if (element_size != 0) size = element_size; else @@ -5757,41 +5211,33 @@ static void **ialloc(mstate m, size_t n_elements, size_t *sizes, int opts, set_size_and_pinuse_of_inuse_chunk(m, p, size); p = chunk_plus_offset(p, size); - } else { /* the final element absorbs any overallocation slop */ + } else { /* the final element absorbs any overallocation slop */ set_size_and_pinuse_of_inuse_chunk(m, p, remainder_size); break; - } - } #if DEBUG if (marray != chunks) { - /* final element must have exactly exhausted chunk */ if (element_size != 0) { - assert(remainder_size == element_size); } else { - assert(remainder_size == request2size(sizes[i])); - } check_inuse_chunk(m, mem2chunk(marray)); - } for (i = 0; i != n_elements; ++i) check_inuse_chunk(m, mem2chunk(marray[i])); - #endif /* DEBUG */ + #endif /* DEBUG */ POSTACTION(m); return marray; - } /* Try to free all pointers in the given array. @@ -5802,36 +5248,28 @@ static void **ialloc(mstate m, size_t n_elements, size_t *sizes, int opts, with ialloc or the array is sorted. */ static size_t internal_bulk_free(mstate m, void *array[], size_t nelem) { - size_t unfreed = 0; if (!PREACTION(m)) { - void **a; void **fence = &(array[nelem]); for (a = array; a != fence; ++a) { - void *mem = *a; if (mem != 0) { - mchunkptr p = mem2chunk(mem); size_t psize = chunksize(p); #if FOOTERS if (get_mstate_for(p) != m) { - ++unfreed; continue; - } #endif check_inuse_chunk(m, p); *a = 0; if (RTCHECK(ok_address(m, p) && ok_inuse(p))) { - - void ** b = a + 1; /* try to merge with next chunk */ + void **b = a + 1; /* try to merge with next chunk */ mchunkptr next = next_chunk(p); if (b != fence && *b == chunk2mem(next)) { - size_t newsize = chunksize(next) + psize; set_inuse(m, p, newsize); *b = chunk2mem(p); @@ -5841,23 +5279,17 @@ static size_t internal_bulk_free(mstate m, void *array[], size_t nelem) { dispose_chunk(m, p, psize); } else { - CORRUPTION_ERROR_ACTION(m); break; - } - } - } if (should_trim(m, m->topsize)) sys_trim(m, 0); POSTACTION(m); - } return unfreed; - } /* Traversal */ @@ -5865,360 +5297,271 @@ static size_t internal_bulk_free(mstate m, void *array[], size_t nelem) { static void internal_inspect_all(mstate m, void (*handler)(void *start, void *end, size_t used_bytes, - void * callback_arg), + void *callback_arg), void *arg) { - if (is_initialized(m)) { - mchunkptr top = m->top; msegmentptr s; for (s = &m->seg; s != 0; s = s->next) { - mchunkptr q = align_as_chunk(s->base); while (segment_holds(s, q) && q->head != FENCEPOST_HEAD) { - mchunkptr next = next_chunk(q); size_t sz = chunksize(q); size_t used; - void * start; + void *start; if (is_inuse(q)) { - - used = sz - CHUNK_OVERHEAD; /* must not be mmapped */ + used = sz - CHUNK_OVERHEAD; /* must not be mmapped */ start = chunk2mem(q); } else { - used = 0; - if (is_small(sz)) { /* offset by possible bookkeeping */ + if (is_small(sz)) { /* offset by possible bookkeeping */ start = (void *)((char *)q + sizeof(struct malloc_chunk)); } else { - start = (void *)((char *)q + sizeof(struct malloc_tree_chunk)); - } - } - if (start < (void *)next) /* skip if all space is bookkeeping */ + if (start < (void *)next) /* skip if all space is bookkeeping */ handler(start, next, used, arg); if (q == top) break; q = next; - } - } - } - } - #endif /* MALLOC_INSPECT_ALL */ + #endif /* MALLOC_INSPECT_ALL */ /* ------------------ Exported realloc, memalign, etc -------------------- */ #if !ONLY_MSPACES void *dlrealloc(void *oldmem, size_t bytes) { - void *mem = 0; if (oldmem == 0) { - mem = dlmalloc(bytes); } else if (bytes >= MAX_REQUEST) { - MALLOC_FAILURE_ACTION; } - #ifdef REALLOC_ZERO_BYTES_FREES else if (bytes == 0) { dlfree(oldmem); } - - #endif /* REALLOC_ZERO_BYTES_FREES */ + #endif /* REALLOC_ZERO_BYTES_FREES */ else { size_t nb = request2size(bytes); mchunkptr oldp = mem2chunk(oldmem); #if !FOOTERS mstate m = gm; - #else /* FOOTERS */ + #else /* FOOTERS */ mstate m = get_mstate_for(oldp); if (!ok_magic(m)) { - USAGE_ERROR_ACTION(m, oldmem); return 0; - } - #endif /* FOOTERS */ + #endif /* FOOTERS */ if (!PREACTION(m)) { - mchunkptr newp = try_realloc_chunk(m, oldp, nb, 1); POSTACTION(m); if (newp != 0) { - check_inuse_chunk(m, newp); mem = chunk2mem(newp); } else { - mem = internal_malloc(m, bytes); if (mem != 0) { - size_t oc = chunksize(oldp) - overhead_for(oldp); __builtin_memcpy(mem, oldmem, (oc < bytes) ? oc : bytes); internal_free(m, oldmem); - } - } - } - } return mem; - } void *dlrealloc_in_place(void *oldmem, size_t bytes) { - void *mem = 0; if (oldmem != 0) { - if (bytes >= MAX_REQUEST) { - MALLOC_FAILURE_ACTION; } else { - size_t nb = request2size(bytes); mchunkptr oldp = mem2chunk(oldmem); #if !FOOTERS mstate m = gm; - #else /* FOOTERS */ + #else /* FOOTERS */ mstate m = get_mstate_for(oldp); if (!ok_magic(m)) { - USAGE_ERROR_ACTION(m, oldmem); return 0; - } - #endif /* FOOTERS */ + #endif /* FOOTERS */ if (!PREACTION(m)) { - mchunkptr newp = try_realloc_chunk(m, oldp, nb, 0); POSTACTION(m); if (newp == oldp) { - check_inuse_chunk(m, newp); mem = oldmem; - } - } - } - } return mem; - } void *dlmemalign(size_t alignment, size_t bytes) { - if (alignment <= MALLOC_ALIGNMENT) { return dlmalloc(bytes); } return internal_memalign(gm, alignment, bytes); - } int dlposix_memalign(void **pp, size_t alignment, size_t bytes) { - void *mem = 0; if (alignment == MALLOC_ALIGNMENT) mem = dlmalloc(bytes); else { - size_t d = alignment / sizeof(void *); size_t r = alignment % sizeof(void *); if (r != 0 || d == 0 || (d & (d - SIZE_T_ONE)) != 0) return EINVAL; else if (bytes <= MAX_REQUEST - alignment) { - if (alignment < MIN_CHUNK_SIZE) alignment = MIN_CHUNK_SIZE; mem = internal_memalign(gm, alignment, bytes); - } - } if (mem == 0) return ENOMEM; else { - *pp = mem; return 0; - } - } void *dlvalloc(size_t bytes) { - size_t pagesz; ensure_initialization(); pagesz = mparams.page_size; return dlmemalign(pagesz, bytes); - } void *dlpvalloc(size_t bytes) { - size_t pagesz; ensure_initialization(); pagesz = mparams.page_size; return dlmemalign(pagesz, (bytes + pagesz - SIZE_T_ONE) & ~(pagesz - SIZE_T_ONE)); - } void **dlindependent_calloc(size_t n_elements, size_t elem_size, void *chunks[]) { - - size_t sz = elem_size; /* serves as 1-element array */ + size_t sz = elem_size; /* serves as 1-element array */ return ialloc(gm, n_elements, &sz, 3, chunks); - } void **dlindependent_comalloc(size_t n_elements, size_t sizes[], void *chunks[]) { - return ialloc(gm, n_elements, sizes, 0, chunks); - } size_t dlbulk_free(void *array[], size_t nelem) { - return internal_bulk_free(gm, array, nelem); - } #if MALLOC_INSPECT_ALL void dlmalloc_inspect_all(void (*handler)(void *start, void *end, size_t used_bytes, - void * callback_arg), + void *callback_arg), void *arg) { - ensure_initialization(); if (!PREACTION(gm)) { - internal_inspect_all(gm, handler, arg); POSTACTION(gm); - } - } - #endif /* MALLOC_INSPECT_ALL */ + #endif /* MALLOC_INSPECT_ALL */ int dlmalloc_trim(size_t pad) { - int result = 0; ensure_initialization(); if (!PREACTION(gm)) { - result = sys_trim(gm, pad); POSTACTION(gm); - } return result; - } size_t dlmalloc_footprint(void) { - return gm->footprint; - } size_t dlmalloc_max_footprint(void) { - return gm->max_footprint; - } size_t dlmalloc_footprint_limit(void) { - size_t maf = gm->footprint_limit; return maf == 0 ? MAX_SIZE_T : maf; - } size_t dlmalloc_set_footprint_limit(size_t bytes) { - - size_t result; /* invert sense of 0 */ - if (bytes == 0) result = granularity_align(1); /* Use minimal size */ + size_t result; /* invert sense of 0 */ + if (bytes == 0) result = granularity_align(1); /* Use minimal size */ if (bytes == MAX_SIZE_T) - result = 0; /* disable */ + result = 0; /* disable */ else result = granularity_align(bytes); return gm->footprint_limit = result; - } #if !NO_MALLINFO struct mallinfo dlmallinfo(void) { - return internal_mallinfo(gm); - } - #endif /* NO_MALLINFO */ + #endif /* NO_MALLINFO */ #if !NO_MALLOC_STATS void dlmalloc_stats() { - internal_malloc_stats(gm); - } - #endif /* NO_MALLOC_STATS */ + #endif /* NO_MALLOC_STATS */ int dlmallopt(int param_number, int value) { - return change_mparam(param_number, value); - } size_t dlmalloc_usable_size(void *mem) { - if (mem != 0) { - mchunkptr p = mem2chunk(mem); if (is_inuse(p)) return chunksize(p) - overhead_for(p); - } return 0; - } - #endif /* !ONLY_MSPACES */ + #endif /* !ONLY_MSPACES */ /* ----------------------------- user mspaces ---------------------------- */ #if MSPACES static mstate init_user_mstate(char *tbase, size_t tsize) { - size_t msize = pad_request(sizeof(struct malloc_state)); mchunkptr mn; mchunkptr msp = align_as_chunk(tbase); @@ -6239,108 +5582,83 @@ static mstate init_user_mstate(char *tbase, size_t tsize) { init_top(m, mn, (size_t)((tbase + tsize) - (char *)mn) - TOP_FOOT_SIZE); check_top_chunk(m, m->top); return m; - } mspace create_mspace(size_t capacity, int locked) { - mstate m = 0; size_t msize; ensure_initialization(); msize = pad_request(sizeof(struct malloc_state)); if (capacity < (size_t) - (msize + TOP_FOOT_SIZE + mparams.page_size)) { - size_t rs = ((capacity == 0) ? mparams.granularity : (capacity + TOP_FOOT_SIZE + msize)); size_t tsize = granularity_align(rs); - char * tbase = (char *)(CALL_MMAP(tsize)); + char *tbase = (char *)(CALL_MMAP(tsize)); if (tbase != CMFAIL) { - m = init_user_mstate(tbase, tsize); m->seg.sflags = USE_MMAP_BIT; set_lock(m, locked); - } - } return (mspace)m; - } mspace create_mspace_with_base(void *base, size_t capacity, int locked) { - mstate m = 0; size_t msize; ensure_initialization(); msize = pad_request(sizeof(struct malloc_state)); if (capacity > msize + TOP_FOOT_SIZE && capacity < (size_t) - (msize + TOP_FOOT_SIZE + mparams.page_size)) { - m = init_user_mstate((char *)base, capacity); m->seg.sflags = EXTERN_BIT; set_lock(m, locked); - } return (mspace)m; - } int mspace_track_large_chunks(mspace msp, int enable) { - int ret = 0; mstate ms = (mstate)msp; if (!PREACTION(ms)) { - if (!use_mmap(ms)) { ret = 1; } if (!enable) { - enable_mmap(ms); } else { - disable_mmap(ms); - } POSTACTION(ms); - } return ret; - } size_t destroy_mspace(mspace msp) { - size_t freed = 0; mstate ms = (mstate)msp; if (ok_magic(ms)) { - msegmentptr sp = &ms->seg; - (void)DESTROY_LOCK(&ms->mutex); /* destroy before unmapped */ + (void)DESTROY_LOCK(&ms->mutex); /* destroy before unmapped */ while (sp != 0) { - - char * base = sp->base; + char *base = sp->base; size_t size = sp->size; flag_t flag = sp->sflags; - (void)base; /* placate people compiling -Wunused-variable */ + (void)base; /* placate people compiling -Wunused-variable */ sp = sp->next; if ((flag & USE_MMAP_BIT) && !(flag & EXTERN_BIT) && CALL_MUNMAP(base, size) == 0) freed += size; - } } else { - USAGE_ERROR_ACTION(ms, ms); - } return freed; - } /* @@ -6349,30 +5667,25 @@ size_t destroy_mspace(mspace msp) { */ void *mspace_malloc(mspace msp, size_t bytes) { - mstate ms = (mstate)msp; if (!ok_magic(ms)) { - USAGE_ERROR_ACTION(ms, ms); return 0; - } if (!PREACTION(ms)) { - - void * mem; + void *mem; size_t nb; if (bytes <= MAX_SMALL_REQUEST) { - bindex_t idx; binmap_t smallbits; nb = (bytes < MIN_REQUEST) ? MIN_CHUNK_SIZE : pad_request(bytes); idx = small_index(nb); smallbits = ms->smallmap >> idx; - if ((smallbits & 0x3U) != 0) { /* Remainderless fit to a smallbin. */ + if ((smallbits & 0x3U) != 0) { /* Remainderless fit to a smallbin. */ mchunkptr b, p; - idx += ~smallbits & 1; /* Uses next bin if idx empty */ + idx += ~smallbits & 1; /* Uses next bin if idx empty */ b = smallbin_at(ms, idx); p = b->fd; assert(chunksize(p) == small_index2size(idx)); @@ -6385,8 +5698,7 @@ void *mspace_malloc(mspace msp, size_t bytes) { } else if (nb > ms->dvsize) { - - if (smallbits != 0) { /* Use chunk in next nonempty smallbin */ + if (smallbits != 0) { /* Use chunk in next nonempty smallbin */ mchunkptr b, p, r; size_t rsize; bindex_t i; @@ -6402,12 +5714,10 @@ void *mspace_malloc(mspace msp, size_t bytes) { if (SIZE_T_SIZE != 4 && rsize < MIN_CHUNK_SIZE) set_inuse_and_pinuse(ms, p, small_index2size(i)); else { - set_size_and_pinuse_of_inuse_chunk(ms, p, nb); r = chunk_plus_offset(p, nb); set_size_and_pinuse_of_free_chunk(r, rsize); replace_dv(ms, r, rsize); - } mem = chunk2mem(p); @@ -6417,46 +5727,37 @@ void *mspace_malloc(mspace msp, size_t bytes) { } else if (ms->treemap != 0 && (mem = tmalloc_small(ms, nb)) != 0) { - check_malloced_chunk(ms, mem, nb); goto postaction; - } - } } else if (bytes >= MAX_REQUEST) nb = MAX_SIZE_T; /* Too big to allocate. Force failure (in sys alloc) */ else { - nb = pad_request(bytes); if (ms->treemap != 0 && (mem = tmalloc_large(ms, nb)) != 0) { - check_malloced_chunk(ms, mem, nb); goto postaction; - } - } if (nb <= ms->dvsize) { - size_t rsize = ms->dvsize - nb; mchunkptr p = ms->dv; - if (rsize >= MIN_CHUNK_SIZE) { /* split dv */ + if (rsize >= MIN_CHUNK_SIZE) { /* split dv */ mchunkptr r = ms->dv = chunk_plus_offset(p, nb); ms->dvsize = rsize; set_size_and_pinuse_of_free_chunk(r, rsize); set_size_and_pinuse_of_inuse_chunk(ms, p, nb); - } else { /* exhaust dv */ + } else { /* exhaust dv */ size_t dvs = ms->dvsize; ms->dvsize = 0; ms->dv = 0; set_inuse_and_pinuse(ms, p, dvs); - } mem = chunk2mem(p); @@ -6465,7 +5766,7 @@ void *mspace_malloc(mspace msp, size_t bytes) { } - else if (nb < ms->topsize) { /* Split top */ + else if (nb < ms->topsize) { /* Split top */ size_t rsize = ms->topsize -= nb; mchunkptr p = ms->top; mchunkptr r = ms->top = chunk_plus_offset(p, nb); @@ -6475,7 +5776,6 @@ void *mspace_malloc(mspace msp, size_t bytes) { check_top_chunk(ms, ms->top); check_malloced_chunk(ms, mem, nb); goto postaction; - } mem = sys_alloc(ms, nb); @@ -6483,112 +5783,87 @@ void *mspace_malloc(mspace msp, size_t bytes) { postaction: POSTACTION(ms); return mem; - } return 0; - } void mspace_free(mspace msp, void *mem) { - if (mem != 0) { - mchunkptr p = mem2chunk(mem); #if FOOTERS mstate fm = get_mstate_for(p); - (void)msp; /* placate people compiling -Wunused */ - #else /* FOOTERS */ + (void)msp; /* placate people compiling -Wunused */ + #else /* FOOTERS */ mstate fm = (mstate)msp; - #endif /* FOOTERS */ + #endif /* FOOTERS */ if (!ok_magic(fm)) { - USAGE_ERROR_ACTION(fm, p); return; - } if (!PREACTION(fm)) { - check_inuse_chunk(fm, p); if (RTCHECK(ok_address(fm, p) && ok_inuse(p))) { - size_t psize = chunksize(p); mchunkptr next = chunk_plus_offset(p, psize); if (!pinuse(p)) { - size_t prevsize = p->prev_foot; if (is_mmapped(p)) { - psize += prevsize + MMAP_FOOT_PAD; if (CALL_MUNMAP((char *)p - prevsize, psize) == 0) fm->footprint -= psize; goto postaction; } else { - mchunkptr prev = chunk_minus_offset(p, prevsize); psize += prevsize; p = prev; - if (RTCHECK(ok_address(fm, prev))) { /* consolidate backward */ + if (RTCHECK(ok_address(fm, prev))) { /* consolidate backward */ if (p != fm->dv) { - unlink_chunk(fm, p, prevsize); } else if ((next->head & INUSE_BITS) == INUSE_BITS) { - fm->dvsize = psize; set_free_with_pinuse(p, psize, next); goto postaction; - } } else goto erroraction; - } - } if (RTCHECK(ok_next(p, next) && ok_pinuse(next))) { - - if (!cinuse(next)) { /* consolidate forward */ + if (!cinuse(next)) { /* consolidate forward */ if (next == fm->top) { - size_t tsize = fm->topsize += psize; fm->top = p; p->head = tsize | PINUSE_BIT; if (p == fm->dv) { - fm->dv = 0; fm->dvsize = 0; - } if (should_trim(fm, tsize)) sys_trim(fm, 0); goto postaction; } else if (next == fm->dv) { - size_t dsize = fm->dvsize += psize; fm->dv = p; set_size_and_pinuse_of_free_chunk(p, dsize); goto postaction; } else { - size_t nsize = chunksize(next); psize += nsize; unlink_chunk(fm, next, nsize); set_size_and_pinuse_of_free_chunk(p, psize); if (p == fm->dv) { - fm->dvsize = psize; goto postaction; - } - } } else @@ -6596,228 +5871,171 @@ void mspace_free(mspace msp, void *mem) { set_free_with_pinuse(p, psize, next); if (is_small(psize)) { - insert_small_chunk(fm, p, psize); check_free_chunk(fm, p); } else { - tchunkptr tp = (tchunkptr)p; insert_large_chunk(fm, tp, psize); check_free_chunk(fm, p); if (--fm->release_checks == 0) release_unused_segments(fm); - } goto postaction; - } - } erroraction: USAGE_ERROR_ACTION(fm, p); postaction: POSTACTION(fm); - } - } - } void *mspace_calloc(mspace msp, size_t n_elements, size_t elem_size) { - - void * mem; + void *mem; size_t req = 0; mstate ms = (mstate)msp; if (!ok_magic(ms)) { - USAGE_ERROR_ACTION(ms, ms); return 0; - } if (n_elements != 0) { - req = n_elements * elem_size; if (((n_elements | elem_size) & ~(size_t)0xffff) && (req / n_elements != elem_size)) - req = MAX_SIZE_T; /* force downstream failure on overflow */ - + req = MAX_SIZE_T; /* force downstream failure on overflow */ } mem = internal_malloc(ms, req); if (mem != 0 && calloc_must_clear(mem2chunk(mem))) __builtin_memset(mem, 0, req); return mem; - } void *mspace_realloc(mspace msp, void *oldmem, size_t bytes) { - void *mem = 0; if (oldmem == 0) { - mem = mspace_malloc(msp, bytes); } else if (bytes >= MAX_REQUEST) { - MALLOC_FAILURE_ACTION; } - #ifdef REALLOC_ZERO_BYTES_FREES else if (bytes == 0) { mspace_free(msp, oldmem); } - - #endif /* REALLOC_ZERO_BYTES_FREES */ + #endif /* REALLOC_ZERO_BYTES_FREES */ else { size_t nb = request2size(bytes); mchunkptr oldp = mem2chunk(oldmem); #if !FOOTERS mstate m = (mstate)msp; - #else /* FOOTERS */ + #else /* FOOTERS */ mstate m = get_mstate_for(oldp); if (!ok_magic(m)) { - USAGE_ERROR_ACTION(m, oldmem); return 0; - } - #endif /* FOOTERS */ + #endif /* FOOTERS */ if (!PREACTION(m)) { - mchunkptr newp = try_realloc_chunk(m, oldp, nb, 1); POSTACTION(m); if (newp != 0) { - check_inuse_chunk(m, newp); mem = chunk2mem(newp); } else { - mem = mspace_malloc(m, bytes); if (mem != 0) { - size_t oc = chunksize(oldp) - overhead_for(oldp); __builtin_memcpy(mem, oldmem, (oc < bytes) ? oc : bytes); mspace_free(m, oldmem); - } - } - } - } return mem; - } void *mspace_realloc_in_place(mspace msp, void *oldmem, size_t bytes) { - void *mem = 0; if (oldmem != 0) { - if (bytes >= MAX_REQUEST) { - MALLOC_FAILURE_ACTION; } else { - size_t nb = request2size(bytes); mchunkptr oldp = mem2chunk(oldmem); #if !FOOTERS mstate m = (mstate)msp; - #else /* FOOTERS */ + #else /* FOOTERS */ mstate m = get_mstate_for(oldp); - (void)msp; /* placate people compiling -Wunused */ + (void)msp; /* placate people compiling -Wunused */ if (!ok_magic(m)) { - USAGE_ERROR_ACTION(m, oldmem); return 0; - } - #endif /* FOOTERS */ + #endif /* FOOTERS */ if (!PREACTION(m)) { - mchunkptr newp = try_realloc_chunk(m, oldp, nb, 0); POSTACTION(m); if (newp == oldp) { - check_inuse_chunk(m, newp); mem = oldmem; - } - } - } - } return mem; - } void *mspace_memalign(mspace msp, size_t alignment, size_t bytes) { - mstate ms = (mstate)msp; if (!ok_magic(ms)) { - USAGE_ERROR_ACTION(ms, ms); return 0; - } if (alignment <= MALLOC_ALIGNMENT) return mspace_malloc(msp, bytes); return internal_memalign(ms, alignment, bytes); - } void **mspace_independent_calloc(mspace msp, size_t n_elements, size_t elem_size, void *chunks[]) { - - size_t sz = elem_size; /* serves as 1-element array */ + size_t sz = elem_size; /* serves as 1-element array */ mstate ms = (mstate)msp; if (!ok_magic(ms)) { - USAGE_ERROR_ACTION(ms, ms); return 0; - } return ialloc(ms, n_elements, &sz, 3, chunks); - } void **mspace_independent_comalloc(mspace msp, size_t n_elements, size_t sizes[], void *chunks[]) { - mstate ms = (mstate)msp; if (!ok_magic(ms)) { - USAGE_ERROR_ACTION(ms, ms); return 0; - } return ialloc(ms, n_elements, sizes, 0, chunks); - } size_t mspace_bulk_free(mspace msp, void *array[], size_t nelem) { - return internal_bulk_free((mstate)msp, array, nelem); - } #if MALLOC_INSPECT_ALL @@ -6825,177 +6043,130 @@ void mspace_inspect_all(mspace msp, void (*handler)(void *start, void *end, size_t used_bytes, void *callback_arg), void *arg) { - mstate ms = (mstate)msp; if (ok_magic(ms)) { - if (!PREACTION(ms)) { - internal_inspect_all(ms, handler, arg); POSTACTION(ms); - } } else { - USAGE_ERROR_ACTION(ms, ms); - } - } - #endif /* MALLOC_INSPECT_ALL */ + #endif /* MALLOC_INSPECT_ALL */ int mspace_trim(mspace msp, size_t pad) { - int result = 0; mstate ms = (mstate)msp; if (ok_magic(ms)) { - if (!PREACTION(ms)) { - result = sys_trim(ms, pad); POSTACTION(ms); - } } else { - USAGE_ERROR_ACTION(ms, ms); - } return result; - } #if !NO_MALLOC_STATS void mspace_malloc_stats(mspace msp) { - mstate ms = (mstate)msp; if (ok_magic(ms)) { - internal_malloc_stats(ms); } else { - USAGE_ERROR_ACTION(ms, ms); - } - } - #endif /* NO_MALLOC_STATS */ + #endif /* NO_MALLOC_STATS */ size_t mspace_footprint(mspace msp) { - size_t result = 0; mstate ms = (mstate)msp; if (ok_magic(ms)) { - result = ms->footprint; } else { - USAGE_ERROR_ACTION(ms, ms); - } return result; - } size_t mspace_max_footprint(mspace msp) { - size_t result = 0; mstate ms = (mstate)msp; if (ok_magic(ms)) { - result = ms->max_footprint; } else { - USAGE_ERROR_ACTION(ms, ms); - } return result; - } size_t mspace_footprint_limit(mspace msp) { - size_t result = 0; mstate ms = (mstate)msp; if (ok_magic(ms)) { - size_t maf = ms->footprint_limit; result = (maf == 0) ? MAX_SIZE_T : maf; } else { - USAGE_ERROR_ACTION(ms, ms); - } return result; - } size_t mspace_set_footprint_limit(mspace msp, size_t bytes) { - size_t result = 0; mstate ms = (mstate)msp; if (ok_magic(ms)) { - - if (bytes == 0) result = granularity_align(1); /* Use minimal size */ + if (bytes == 0) result = granularity_align(1); /* Use minimal size */ if (bytes == MAX_SIZE_T) - result = 0; /* disable */ + result = 0; /* disable */ else result = granularity_align(bytes); ms->footprint_limit = result; } else { - USAGE_ERROR_ACTION(ms, ms); - } return result; - } #if !NO_MALLINFO struct mallinfo mspace_mallinfo(mspace msp) { - mstate ms = (mstate)msp; if (!ok_magic(ms)) { USAGE_ERROR_ACTION(ms, ms); } return internal_mallinfo(ms); - } - #endif /* NO_MALLINFO */ + #endif /* NO_MALLINFO */ size_t mspace_usable_size(const void *mem) { - if (mem != 0) { - mchunkptr p = mem2chunk(mem); if (is_inuse(p)) return chunksize(p) - overhead_for(p); - } return 0; - } int mspace_mallopt(int param_number, int value) { - return change_mparam(param_number, value); - } - #endif /* MSPACES */ + #endif /* MSPACES */ /* -------------------- Alternative MORECORE functions ------------------- */ @@ -7092,7 +6263,7 @@ int mspace_mallopt(int param_number, int value) { void **ptr; - for (ptr = our_os_pools; ptr < &our_os_pools[MAX_POOL_ENTRIES]; ptr++) + for (ptr = our_os_pools; ptr < &our_os_pools[MAX_POOL_ENTRIES]; ptr++) { if (*ptr) { @@ -7100,6 +6271,7 @@ int mspace_mallopt(int param_number, int value) { *ptr = 0; } + } } @@ -7325,4 +6497,3 @@ History: */ #endif // __GLIBC__ - diff --git a/libafl_qemu/libqasan/hooks.c b/libafl_qemu/libqasan/hooks.c index c542521c25..c0adf97759 100644 --- a/libafl_qemu/libqasan/hooks.c +++ b/libafl_qemu/libqasan/hooks.c @@ -34,18 +34,15 @@ long (*__lq_libc_atol)(const char *); long long (*__lq_libc_atoll)(const char *); void __libqasan_init_hooks(void) { - __libqasan_init_malloc(); __lq_libc_fgets = ASSERT_DLSYM(fgets); __lq_libc_atoi = ASSERT_DLSYM(atoi); __lq_libc_atol = ASSERT_DLSYM(atol); __lq_libc_atoll = ASSERT_DLSYM(atoll); - } ssize_t write(int fd, const void *buf, size_t count) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: write(%d, %p, %zu)\n", rtv, fd, buf, count); @@ -54,11 +51,9 @@ ssize_t write(int fd, const void *buf, size_t count) { QASAN_DEBUG("\t\t = %zd\n", r); return r; - } ssize_t read(int fd, void *buf, size_t count) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: read(%d, %p, %zu)\n", rtv, fd, buf, count); @@ -67,12 +62,10 @@ ssize_t read(int fd, void *buf, size_t count) { QASAN_DEBUG("\t\t = %zd\n", r); return r; - } #ifdef __ANDROID__ size_t malloc_usable_size(const void *ptr) { - #else size_t malloc_usable_size(void *ptr) { @@ -85,11 +78,9 @@ size_t malloc_usable_size(void *ptr) { QASAN_DEBUG("\t\t = %zu\n", r); return r; - } void *malloc(size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: malloc(%zu)\n", rtv, size); @@ -97,11 +88,9 @@ void *malloc(size_t size) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *calloc(size_t nmemb, size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: calloc(%zu, %zu)\n", rtv, nmemb, size); @@ -109,11 +98,9 @@ void *calloc(size_t nmemb, size_t size) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *realloc(void *ptr, size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: realloc(%p, %zu)\n", rtv, ptr, size); @@ -121,11 +108,9 @@ void *realloc(void *ptr, size_t size) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } int posix_memalign(void **memptr, size_t alignment, size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: posix_memalign(%p, %zu, %zu)\n", rtv, memptr, alignment, @@ -134,11 +119,9 @@ int posix_memalign(void **memptr, size_t alignment, size_t size) { QASAN_DEBUG("\t\t = %d [*memptr = %p]\n", r, *memptr); return r; - } void *memalign(size_t alignment, size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memalign(%zu, %zu)\n", rtv, alignment, size); @@ -146,11 +129,9 @@ void *memalign(size_t alignment, size_t size) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *aligned_alloc(size_t alignment, size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: aligned_alloc(%zu, %zu)\n", rtv, alignment, size); @@ -158,11 +139,9 @@ void *aligned_alloc(size_t alignment, size_t size) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *valloc(size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: valloc(%zu)\n", rtv, size); @@ -170,11 +149,9 @@ void *valloc(size_t size) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *pvalloc(size_t size) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: pvalloc(%zu)\n", rtv, size); @@ -184,20 +161,16 @@ void *pvalloc(size_t size) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void free(void *ptr) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: free(%p)\n", rtv, ptr); __libqasan_free(ptr); - } char *fgets(char *s, int size, FILE *stream) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: fgets(%p, %d, %p)\n", rtv, s, size, stream); @@ -209,11 +182,9 @@ char *fgets(char *s, int size, FILE *stream) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } int memcmp(const void *s1, const void *s2, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memcmp(%p, %p, %zu)\n", rtv, s1, s2, n); @@ -223,11 +194,9 @@ int memcmp(const void *s1, const void *s2, size_t n) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } void *memcpy(void *dest, const void *src, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memcpy(%p, %p, %zu)\n", rtv, dest, src, n); @@ -237,11 +206,9 @@ void *memcpy(void *dest, const void *src, size_t n) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *mempcpy(void *dest, const void *src, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: mempcpy(%p, %p, %zu)\n", rtv, dest, src, n); @@ -251,11 +218,9 @@ void *mempcpy(void *dest, const void *src, size_t n) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *memmove(void *dest, const void *src, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memmove(%p, %p, %zu)\n", rtv, dest, src, n); @@ -265,11 +230,9 @@ void *memmove(void *dest, const void *src, size_t n) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *memset(void *s, int c, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memset(%p, %d, %zu)\n", rtv, s, c, n); @@ -278,27 +241,24 @@ void *memset(void *s, int c, size_t n) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *memchr(const void *s, int c, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memchr(%p, %d, %zu)\n", rtv, s, c, n); void *r = __libqasan_memchr(s, c, n); - if (r == NULL) + if (r == NULL) { QASAN_LOAD(s, n); - else + } else { QASAN_LOAD(s, r - s); + } QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *memrchr(const void *s, int c, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memrchr(%p, %d, %zu)\n", rtv, s, c, n); @@ -307,12 +267,10 @@ void *memrchr(const void *s, int c, size_t n) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } void *memmem(const void *haystack, size_t haystacklen, const void *needle, size_t needlelen) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: memmem(%p, %zu, %p, %zu)\n", rtv, haystack, haystacklen, @@ -323,34 +281,28 @@ void *memmem(const void *haystack, size_t haystacklen, const void *needle, QASAN_DEBUG("\t\t = %p\n", r); return r; - } #ifndef __BIONIC__ void bzero(void *s, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: bzero(%p, %zu)\n", rtv, s, n); QASAN_STORE(s, n); __libqasan_memset(s, 0, n); - } #endif void explicit_bzero(void *s, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: bzero(%p, %zu)\n", rtv, s, n); QASAN_STORE(s, n); __libqasan_memset(s, 0, n); - } int bcmp(const void *s1, const void *s2, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: bcmp(%p, %p, %zu)\n", rtv, s1, s2, n); @@ -360,11 +312,9 @@ int bcmp(const void *s1, const void *s2, size_t n) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } char *strchr(const char *s, int c) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strchr(%p, %d)\n", rtv, s, c); @@ -374,11 +324,9 @@ char *strchr(const char *s, int c) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } char *strrchr(const char *s, int c) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strrchr(%p, %d)\n", rtv, s, c); @@ -388,11 +336,9 @@ char *strrchr(const char *s, int c) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } int strcasecmp(const char *s1, const char *s2) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strcasecmp(%p, %p)\n", rtv, s1, s2); @@ -404,11 +350,9 @@ int strcasecmp(const char *s1, const char *s2) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } int strncasecmp(const char *s1, const char *s2, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strncasecmp(%p, %p, %zu)\n", rtv, s1, s2, n); @@ -420,11 +364,9 @@ int strncasecmp(const char *s1, const char *s2, size_t n) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } char *strcat(char *dest, const char *src) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strcat(%p, %p)\n", rtv, dest, src); @@ -438,11 +380,9 @@ char *strcat(char *dest, const char *src) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } int strcmp(const char *s1, const char *s2) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strcmp(%p, %p)\n", rtv, s1, s2); @@ -454,11 +394,9 @@ int strcmp(const char *s1, const char *s2) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } int strncmp(const char *s1, const char *s2, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strncmp(%p, %p, %zu)\n", rtv, s1, s2, n); @@ -470,11 +408,9 @@ int strncmp(const char *s1, const char *s2, size_t n) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } char *strcpy(char *dest, const char *src) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strcpy(%p, %p)\n", rtv, dest, src); @@ -485,11 +421,9 @@ char *strcpy(char *dest, const char *src) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } char *strncpy(char *dest, const char *src, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strncpy(%p, %p, %zu)\n", rtv, dest, src, n); @@ -497,25 +431,20 @@ char *strncpy(char *dest, const char *src, size_t n) { QASAN_STORE(dest, n); void *r; if (l < n) { - QASAN_LOAD(src, l + 1); r = __libqasan_memcpy(dest, src, l + 1); } else { - QASAN_LOAD(src, n); r = __libqasan_memcpy(dest, src, n); - } QASAN_DEBUG("\t\t = %p\n", r); return r; - } char *stpcpy(char *dest, const char *src) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: stpcpy(%p, %p)\n", rtv, dest, src); @@ -526,11 +455,9 @@ char *stpcpy(char *dest, const char *src) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } char *strdup(const char *s) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strdup(%p)\n", rtv, s); @@ -541,11 +468,9 @@ char *strdup(const char *s) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } size_t strlen(const char *s) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strlen(%p)\n", rtv, s); @@ -554,11 +479,9 @@ size_t strlen(const char *s) { QASAN_DEBUG("\t\t = %zu\n", r); return r; - } size_t strnlen(const char *s, size_t n) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strnlen(%p, %zu)\n", rtv, s, n); @@ -567,11 +490,9 @@ size_t strnlen(const char *s, size_t n) { QASAN_DEBUG("\t\t = %zu\n", r); return r; - } char *strstr(const char *haystack, const char *needle) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strstr(%p, %p)\n", rtv, haystack, needle); @@ -583,11 +504,9 @@ char *strstr(const char *haystack, const char *needle) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } char *strcasestr(const char *haystack, const char *needle) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: strcasestr(%p, %p)\n", rtv, haystack, needle); @@ -599,11 +518,9 @@ char *strcasestr(const char *haystack, const char *needle) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } int atoi(const char *nptr) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: atoi(%p)\n", rtv, nptr); @@ -613,11 +530,9 @@ int atoi(const char *nptr) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } long atol(const char *nptr) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: atol(%p)\n", rtv, nptr); @@ -627,11 +542,9 @@ long atol(const char *nptr) { QASAN_DEBUG("\t\t = %ld\n", r); return r; - } long long atoll(const char *nptr) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: atoll(%p)\n", rtv, nptr); @@ -641,11 +554,9 @@ long long atoll(const char *nptr) { QASAN_DEBUG("\t\t = %lld\n", r); return r; - } size_t wcslen(const wchar_t *s) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: wcslen(%p)\n", rtv, s); @@ -654,11 +565,9 @@ size_t wcslen(const wchar_t *s) { QASAN_DEBUG("\t\t = %zu\n", r); return r; - } wchar_t *wcscpy(wchar_t *dest, const wchar_t *src) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: wcscpy(%p, %p)\n", rtv, dest, src); @@ -669,11 +578,9 @@ wchar_t *wcscpy(wchar_t *dest, const wchar_t *src) { QASAN_DEBUG("\t\t = %p\n", r); return r; - } int wcscmp(const wchar_t *s1, const wchar_t *s2) { - void *rtv = __builtin_return_address(0); QASAN_DEBUG("%14p: wcscmp(%p, %p)\n", rtv, s1, s2); @@ -685,6 +592,4 @@ int wcscmp(const wchar_t *s1, const wchar_t *s2) { QASAN_DEBUG("\t\t = %d\n", r); return r; - } - diff --git a/libafl_qemu/libqasan/libqasan.c b/libafl_qemu/libqasan/libqasan.c index a64db10f8e..12b07d41b5 100644 --- a/libafl_qemu/libqasan/libqasan.c +++ b/libafl_qemu/libqasan/libqasan.c @@ -31,7 +31,6 @@ int __qasan_debug; int __qasan_log; void __libqasan_print_maps(void) { - int fd = open("/proc/self/maps", O_RDONLY); char buf[4096] = {0}; @@ -44,33 +43,30 @@ void __libqasan_print_maps(void) { int i; char *line = NULL; for (i = 0; i < len; i++) { - - if (!line) line = &buf[i]; - if (buf[i] == '\n') { - - buf[i] = 0; - QASAN_LOG("%s\n", line); - line = NULL; - - } - + if (!line) line = { &buf[i]; } + if (buf[i] == '\n') { + buf[i] = 0; + QASAN_LOG("%s\n", line); + line = NULL; + } +} - if (line) QASAN_LOG("%s\n", line); - QASAN_LOG("\n"); - +if (line) { QASAN_LOG("%s\n", line); } +QASAN_LOG("\n"); } int __libqasan_is_initialized = 0; __attribute__((constructor)) void __libqasan_init() { - - if (__libqasan_is_initialized) return; + if (__libqasan_is_initialized) { return; } __libqasan_is_initialized = 1; __libqasan_init_hooks(); - if (getenv("AFL_INST_LIBS") || getenv("QASAN_HOTPACH")) __libqasan_hotpatch(); + if (getenv("AFL_INST_LIBS") || getenv("QASAN_HOTPACH")) { + __libqasan_hotpatch(); + } #ifdef DEBUG __qasan_debug = getenv("QASAN_DEBUG") != NULL; @@ -82,19 +78,15 @@ __attribute__((constructor)) void __libqasan_init() { "Copyright (C) 2019-2021 Andrea Fioraldi \n"); QASAN_LOG("\n"); - if (__qasan_log) __libqasan_print_maps(); - + if (__qasan_log) { __libqasan_print_maps(); } } int __libc_start_main(int (*main)(int, char **, char **), int argc, char **argv, int (*init)(int, char **, char **), void (*fini)(void), void (*rtld_fini)(void), void *stack_end) { - typeof(&__libc_start_main) orig = dlsym(RTLD_NEXT, "__libc_start_main"); __libqasan_init(); return orig(main, argc, argv, init, fini, rtld_fini, stack_end); - } - diff --git a/libafl_qemu/libqasan/libqasan.h b/libafl_qemu/libqasan/libqasan.h index 43b7adb5a8..655dd80b62 100644 --- a/libafl_qemu/libqasan/libqasan.h +++ b/libafl_qemu/libqasan/libqasan.h @@ -43,12 +43,9 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #define QASAN_LOG(msg...) \ do { \ - \ if (__qasan_log) { \ - \ fprintf(stderr, "==%d== ", getpid()); \ fprintf(stderr, msg); \ - \ } \ \ } while (0) @@ -56,12 +53,9 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #ifdef DEBUG #define QASAN_DEBUG(msg...) \ do { \ - \ if (__qasan_debug) { \ - \ fprintf(stderr, "==%d== ", getpid()); \ fprintf(stderr, msg); \ - \ } \ \ } while (0) @@ -69,23 +63,18 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #else #define QASAN_DEBUG(msg...) \ do { \ - \ } while (0) #endif #define ASSERT_DLSYM(name) \ ({ \ - \ void *a = (void *)dlsym(RTLD_NEXT, #name); \ if (!a) { \ - \ fprintf(stderr, \ "FATAL ERROR: failed dlsym of " #name " in libqasan!\n"); \ abort(); \ - \ } \ a; \ - \ }) extern int __qasan_debug; @@ -97,19 +86,19 @@ void __libqasan_init_malloc(void); void __libqasan_hotpatch(void); size_t __libqasan_malloc_usable_size(void *ptr); -void * __libqasan_malloc(size_t size); +void *__libqasan_malloc(size_t size); void __libqasan_free(void *ptr); -void * __libqasan_calloc(size_t nmemb, size_t size); -void * __libqasan_realloc(void *ptr, size_t size); +void *__libqasan_calloc(size_t nmemb, size_t size); +void *__libqasan_realloc(void *ptr, size_t size); int __libqasan_posix_memalign(void **ptr, size_t align, size_t len); -void * __libqasan_memalign(size_t align, size_t len); -void * __libqasan_aligned_alloc(size_t align, size_t len); +void *__libqasan_memalign(size_t align, size_t len); +void *__libqasan_aligned_alloc(size_t align, size_t len); -void * __libqasan_memcpy(void *dest, const void *src, size_t n); -void * __libqasan_memmove(void *dest, const void *src, size_t n); -void * __libqasan_memset(void *s, int c, size_t n); -void * __libqasan_memchr(const void *s, int c, size_t n); -void * __libqasan_memrchr(const void *s, int c, size_t n); +void *__libqasan_memcpy(void *dest, const void *src, size_t n); +void *__libqasan_memmove(void *dest, const void *src, size_t n); +void *__libqasan_memset(void *s, int c, size_t n); +void *__libqasan_memchr(const void *s, int c, size_t n); +void *__libqasan_memrchr(const void *s, int c, size_t n); size_t __libqasan_strlen(const char *s); size_t __libqasan_strnlen(const char *s, size_t len); int __libqasan_strcmp(const char *str1, const char *str2); @@ -118,15 +107,14 @@ int __libqasan_strcasecmp(const char *str1, const char *str2); int __libqasan_strncasecmp(const char *str1, const char *str2, size_t len); int __libqasan_memcmp(const void *mem1, const void *mem2, size_t len); int __libqasan_bcmp(const void *mem1, const void *mem2, size_t len); -char * __libqasan_strstr(const char *haystack, const char *needle); -char * __libqasan_strcasestr(const char *haystack, const char *needle); -void * __libqasan_memmem(const void *haystack, size_t haystack_len, +char *__libqasan_strstr(const char *haystack, const char *needle); +char *__libqasan_strcasestr(const char *haystack, const char *needle); +void *__libqasan_memmem(const void *haystack, size_t haystack_len, const void *needle, size_t needle_len); -char * __libqasan_strchr(const char *s, int c); -char * __libqasan_strrchr(const char *s, int c); +char *__libqasan_strchr(const char *s, int c); +char *__libqasan_strrchr(const char *s, int c); size_t __libqasan_wcslen(const wchar_t *s); wchar_t *__libqasan_wcscpy(wchar_t *d, const wchar_t *s); int __libqasan_wcscmp(const wchar_t *s1, const wchar_t *s2); #endif - diff --git a/libafl_qemu/libqasan/malloc.c b/libafl_qemu/libqasan/malloc.c index 5893a4e50c..61f5835ea5 100644 --- a/libafl_qemu/libqasan/malloc.c +++ b/libafl_qemu/libqasan/malloc.c @@ -38,7 +38,6 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. (defined(__FreeBSD__) && __FreeBSD_version < 1200000) // use this hack if not C11 typedef struct { - long long __ll; long double __ld; @@ -49,21 +48,17 @@ typedef struct { #define ALLOC_ALIGN_SIZE (_Alignof(max_align_t)) struct chunk_begin { - size_t requested_size; - void * aligned_orig; // NULL if not aligned + void *aligned_orig; // NULL if not aligned struct chunk_begin *next; struct chunk_begin *prev; char redzone[REDZONE_SIZE]; - }; struct chunk_struct { - struct chunk_begin begin; char redzone[REDZONE_SIZE]; size_t prev_size_padding; - }; #ifdef __GLIBC__ @@ -80,7 +75,7 @@ static unsigned char __tmp_alloc_zone[TMP_ZONE_SIZE]; #else // From dlmalloc.c -void * dlmalloc(size_t); +void *dlmalloc(size_t); void dlfree(void *); #define backend_malloc dlmalloc #define backend_free dlfree @@ -107,13 +102,11 @@ static pthread_spinlock_t quarantine_lock; // need qasan disabled static int quarantine_push(struct chunk_begin *ck) { - if (ck->requested_size >= QUARANTINE_MAX_BYTES) return 0; if (LOCK_TRY(&quarantine_lock)) return 0; while (ck->requested_size + quarantine_bytes >= QUARANTINE_MAX_BYTES) { - struct chunk_begin *tmp = quarantine_end; quarantine_end = tmp->prev; @@ -123,7 +116,6 @@ static int quarantine_push(struct chunk_begin *ck) { backend_free(tmp->aligned_orig); else backend_free(tmp); - } ck->next = quarantine_top; @@ -133,11 +125,9 @@ static int quarantine_push(struct chunk_begin *ck) { LOCK_UNLOCK(&quarantine_lock); return 1; - } void __libqasan_init_malloc(void) { - if (__libqasan_malloc_initialized) return; #ifdef __GLIBC__ @@ -151,11 +141,9 @@ void __libqasan_init_malloc(void) { QASAN_LOG("\n"); QASAN_LOG("Allocator initialization done.\n"); QASAN_LOG("\n"); - } size_t __libqasan_malloc_usable_size(void *ptr) { - char *p = ptr; p -= sizeof(struct chunk_begin); @@ -163,13 +151,10 @@ size_t __libqasan_malloc_usable_size(void *ptr) { // to verify that ptr is a valid malloc region before we dereference it) QASAN_LOAD(p, sizeof(struct chunk_begin) - REDZONE_SIZE); return ((struct chunk_begin *)p)->requested_size; - } void *__libqasan_malloc(size_t size) { - if (!__libqasan_malloc_initialized) { - __libqasan_init_malloc(); #ifdef __GLIBC__ @@ -183,7 +168,6 @@ void *__libqasan_malloc(size_t size) { return r; #endif - } int state = QASAN_SWAP(QASAN_DISABLED); // disable qasan for this thread @@ -212,11 +196,9 @@ void *__libqasan_malloc(size_t size) { __builtin_memset(&p[1], 0xff, size); return &p[1]; - } void __libqasan_free(void *ptr) { - if (!ptr) return; #ifdef __GLIBC__ @@ -237,12 +219,10 @@ void __libqasan_free(void *ptr) { int state = QASAN_SWAP(QASAN_DISABLED); // disable qasan for this thread if (!quarantine_push(p)) { - if (p->aligned_orig) backend_free(p->aligned_orig); else backend_free(p); - } QASAN_SWAP(state); @@ -252,20 +232,16 @@ void __libqasan_free(void *ptr) { QASAN_POISON(ptr, n, ASAN_HEAP_FREED); QASAN_DEALLOC(ptr); - } void *__libqasan_calloc(size_t nmemb, size_t size) { - size *= nmemb; #ifdef __GLIBC__ if (!__libqasan_malloc_initialized) { - void *r = &__tmp_alloc_zone[__tmp_alloc_zone_idx]; __tmp_alloc_zone_idx += size; return r; - } #endif @@ -276,11 +252,9 @@ void *__libqasan_calloc(size_t nmemb, size_t size) { __builtin_memset(p, 0, size); return p; - } void *__libqasan_realloc(void *ptr, size_t size) { - char *p = __libqasan_malloc(size); if (!p) return NULL; @@ -293,17 +267,13 @@ void *__libqasan_realloc(void *ptr, size_t size) { __libqasan_free(ptr); return p; - } int __libqasan_posix_memalign(void **ptr, size_t align, size_t len) { - if ((align % 2) || (align % sizeof(void *))) return EINVAL; if (len == 0) { - *ptr = NULL; return 0; - } size_t rem = len % align; @@ -343,21 +313,17 @@ int __libqasan_posix_memalign(void **ptr, size_t align, size_t len) { *ptr = data; return 0; - } void *__libqasan_memalign(size_t align, size_t len) { - void *ret = NULL; __libqasan_posix_memalign(&ret, align, len); return ret; - } void *__libqasan_aligned_alloc(size_t align, size_t len) { - void *ret = NULL; if ((len % align)) return NULL; @@ -365,6 +331,4 @@ void *__libqasan_aligned_alloc(size_t align, size_t len) { __libqasan_posix_memalign(&ret, align, len); return ret; - } - diff --git a/libafl_qemu/libqasan/map_macro.h b/libafl_qemu/libqasan/map_macro.h index e9438dc5c1..a377f81261 100644 --- a/libafl_qemu/libqasan/map_macro.h +++ b/libafl_qemu/libqasan/map_macro.h @@ -71,4 +71,3 @@ EVAL(MAP_LIST1(f, __VA_ARGS__, ()()(), ()()(), ()()(), 0)) #endif - diff --git a/libafl_qemu/libqasan/patch.c b/libafl_qemu/libqasan/patch.c index fbc09c99e8..9dafd3d786 100644 --- a/libafl_qemu/libqasan/patch.c +++ b/libafl_qemu/libqasan/patch.c @@ -29,7 +29,6 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #ifdef __x86_64__ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { - // mov rax, dest addr[0] = 0x48; addr[1] = 0xb8; @@ -40,13 +39,11 @@ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { addr[11] = 0xe0; return &addr[12]; - } #elif __i386__ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { - // mov eax, dest addr[0] = 0xb8; *(uint8_t **)&addr[1] = dest; @@ -56,7 +53,6 @@ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { addr[6] = 0xe0; return &addr[7]; - } #elif __arm__ @@ -65,7 +61,6 @@ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { // so let's use it in our stub uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { - // ldr r12, OFF addr[0] = 0x0; addr[1] = 0xc0; @@ -82,7 +77,6 @@ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { *(uint32_t *)&addr[8] = (uint32_t)dest; return &addr[12]; - } #elif __aarch64__ @@ -91,7 +85,6 @@ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { // so let's use it in our stub uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { - // ldr x16, OFF addr[0] = 0x50; addr[1] = 0x0; @@ -108,7 +101,6 @@ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { *(uint64_t *)&addr[8] = (uint64_t)dest; return &addr[16]; - } #else @@ -120,7 +112,6 @@ uint8_t *__libqasan_patch_jump(uint8_t *addr, uint8_t *dest) { #ifdef CANNOT_HOTPATCH void __libqasan_hotpatch(void) { - } #else @@ -129,17 +120,15 @@ static void *libc_start, *libc_end; int libc_perms; static void find_libc(void) { - - FILE * fp; - char * line = NULL; + FILE *fp; + char *line = NULL; size_t len = 0; ssize_t read; fp = fopen("/proc/self/maps", "r"); - if (fp == NULL) return; + if (fp == NULL) { return; } while ((read = getline(&line, &len, fp)) != -1) { - int fields, dev_maj, dev_min, inode; uint64_t min, max, offset; char flag_r, flag_w, flag_x, flag_p; @@ -151,27 +140,23 @@ static void find_libc(void) { &min, &max, &flag_r, &flag_w, &flag_x, &flag_p, &offset, &dev_maj, &dev_min, &inode, path); - if ((fields < 10) || (fields > 11)) continue; + if ((fields < 10) || (fields > 11)) { continue; } if (flag_x == 'x' && (__libqasan_strstr(path, "/libc.so") || __libqasan_strstr(path, "/libc-"))) { - libc_start = (void *)min; libc_end = (void *)max; libc_perms = PROT_EXEC; - if (flag_w == 'w') libc_perms |= PROT_WRITE; - if (flag_r == 'r') libc_perms |= PROT_READ; + if (flag_w == 'w') { libc_perms |= PROT_WRITE; } + if (flag_r == 'r') { libc_perms |= PROT_READ; } break; - } - } free(line); fclose(fp); - } /* Why this shit? https://twitter.com/andreafioraldi/status/1227635146452541441 @@ -181,28 +166,29 @@ static void find_libc(void) { */ void __libqasan_hotpatch(void) { - find_libc(); - if (!libc_start) return; + if (!libc_start) { return; } if (mprotect(libc_start, libc_end - libc_start, - PROT_READ | PROT_WRITE | PROT_EXEC) < 0) + PROT_READ | PROT_WRITE | PROT_EXEC) < 0) { return; + } void *libc = dlopen("libc.so.6", RTLD_LAZY); #define HOTPATCH(fn) \ uint8_t *p_##fn = (uint8_t *)dlsym(libc, #fn); \ - if (p_##fn) __libqasan_patch_jump(p_##fn, (uint8_t *)&(fn)); + if (p_##fn) { __libqasan_patch_jump(p_##fn, (uint8_t *)&(fn)); } HOTPATCH(memcmp) HOTPATCH(memmove) uint8_t *p_memcpy = (uint8_t *)dlsym(libc, "memcpy"); // fuck you libc - if (p_memcpy && p_memmove != p_memcpy) + if (p_memcpy && p_memmove != p_memcpy) { __libqasan_patch_jump(p_memcpy, (uint8_t *)&memcpy); + } HOTPATCH(memchr) HOTPATCH(memrchr) @@ -236,8 +222,6 @@ void __libqasan_hotpatch(void) { #undef HOTPATCH mprotect(libc_start, libc_end - libc_start, libc_perms); - } #endif - diff --git a/libafl_qemu/libqasan/qasan.h b/libafl_qemu/libqasan/qasan.h index dea9c706ec..05b59ef5c7 100644 --- a/libafl_qemu/libqasan/qasan.h +++ b/libafl_qemu/libqasan/qasan.h @@ -76,8 +76,7 @@ enum { #include -#define QASAN_CALL0(action) \ - syscall(QASAN_FAKESYS_NR, action, NULL, NULL, NULL) +#define QASAN_CALL0(action) syscall(QASAN_FAKESYS_NR, action, NULL, NULL, NULL) #define QASAN_CALL1(action, arg1) \ syscall(QASAN_FAKESYS_NR, action, arg1, NULL, NULL) #define QASAN_CALL2(action, arg1, arg2) \ @@ -85,26 +84,19 @@ enum { #define QASAN_CALL3(action, arg1, arg2, arg3) \ syscall(QASAN_FAKESYS_NR, action, arg1, arg2, arg3) -#define QASAN_LOAD(ptr, len) \ - QASAN_CALL2(QASAN_ACTION_CHECK_LOAD, ptr, len) -#define QASAN_STORE(ptr, len) \ - QASAN_CALL2(QASAN_ACTION_CHECK_STORE, ptr, len) +#define QASAN_LOAD(ptr, len) QASAN_CALL2(QASAN_ACTION_CHECK_LOAD, ptr, len) +#define QASAN_STORE(ptr, len) QASAN_CALL2(QASAN_ACTION_CHECK_STORE, ptr, len) #define QASAN_POISON(ptr, len, poison_byte) \ QASAN_CALL3(QASAN_ACTION_POISON, ptr, len, poison_byte) #define QASAN_USER_POISON(ptr, len) \ QASAN_CALL3(QASAN_ACTION_POISON, ptr, len, ASAN_USER) -#define QASAN_UNPOISON(ptr, len) \ - QASAN_CALL2(QASAN_ACTION_UNPOISON, ptr, len) -#define QASAN_IS_POISON(ptr, len) \ - QASAN_CALL2(QASAN_ACTION_IS_POISON, ptr, len) +#define QASAN_UNPOISON(ptr, len) QASAN_CALL2(QASAN_ACTION_UNPOISON, ptr, len) +#define QASAN_IS_POISON(ptr, len) QASAN_CALL2(QASAN_ACTION_IS_POISON, ptr, len) -#define QASAN_ALLOC(start, end) \ - QASAN_CALL2(QASAN_ACTION_ALLOC, start, end) -#define QASAN_DEALLOC(ptr) \ - QASAN_CALL1(QASAN_ACTION_DEALLOC, ptr) +#define QASAN_ALLOC(start, end) QASAN_CALL2(QASAN_ACTION_ALLOC, start, end) +#define QASAN_DEALLOC(ptr) QASAN_CALL1(QASAN_ACTION_DEALLOC, ptr) -#define QASAN_SWAP(state) \ - QASAN_CALL1(QASAN_ACTION_SWAP_STATE, state) +#define QASAN_SWAP(state) QASAN_CALL1(QASAN_ACTION_SWAP_STATE, state) #endif diff --git a/libafl_qemu/libqasan/string.c b/libafl_qemu/libqasan/string.c index 4be0127908..334f3851e1 100644 --- a/libafl_qemu/libqasan/string.c +++ b/libafl_qemu/libqasan/string.c @@ -27,33 +27,30 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #include void *__libqasan_memcpy(void *dest, const void *src, size_t n) { - - unsigned char * d = dest; + unsigned char *d = dest; const unsigned char *s = src; - if (!n) return dest; + if (!n) { return dest; } while (n--) { - *d = *s; ++d; ++s; - } return dest; - } void *__libqasan_memmove(void *dest, const void *src, size_t n) { - - unsigned char * d = dest; + unsigned char *d = dest; const unsigned char *s = src; - if (!n) return dest; + if (!n) { return dest; } if (!((d + n) >= s && d <= (s + n))) // do not overlap + { return __libqasan_memcpy(dest, src, n); + } d = __libqasan_malloc(n); __libqasan_memcpy(d, src, n); @@ -62,199 +59,165 @@ void *__libqasan_memmove(void *dest, const void *src, size_t n) { __libqasan_free(d); return dest; - } void *__libqasan_memset(void *s, int c, size_t n) { - unsigned char *b = s; - while (n--) + while (n--) { *(b++) = (unsigned char)c; + } return s; - } void *__libqasan_memchr(const void *s, int c, size_t n) { - unsigned char *m = (unsigned char *)s; size_t i; - for (i = 0; i < n; ++i) - if (m[i] == (unsigned char)c) return &m[i]; + for (i = 0; i < n; ++i) { + if (m[i] == (unsigned char)c) { return &m[i]; } + } return NULL; - } void *__libqasan_memrchr(const void *s, int c, size_t n) { - unsigned char *m = (unsigned char *)s; long i; - for (i = n; i >= 0; --i) - if (m[i] == (unsigned char)c) return &m[i]; + for (i = n; i >= 0; --i) { + if (m[i] == (unsigned char)c) { return &m[i]; } + } return NULL; - } size_t __libqasan_strlen(const char *s) { - const char *i = s; while (*(i++)) ; return i - s - 1; - } size_t __libqasan_strnlen(const char *s, size_t len) { - size_t r = 0; - while (len-- && *(s++)) + while (len-- && *(s++)) { ++r; + } return r; - } int __libqasan_strcmp(const char *str1, const char *str2) { - while (1) { - const unsigned char c1 = *str1, c2 = *str2; - if (c1 != c2) return c1 - c2; - if (!c1) return 0; + if (c1 != c2) { return c1 - c2; } + if (!c1) { return 0; } str1++; str2++; - } return 0; - } int __libqasan_strncmp(const char *str1, const char *str2, size_t len) { - while (len--) { - unsigned char c1 = *str1, c2 = *str2; - if (c1 != c2) return c1 - c2; - if (!c1) return 0; + if (c1 != c2) { return c1 - c2; } + if (!c1) { return 0; } str1++; str2++; - } return 0; - } int __libqasan_strcasecmp(const char *str1, const char *str2) { - while (1) { - const unsigned char c1 = tolower(*str1), c2 = tolower(*str2); - if (c1 != c2) return c1 - c2; - if (!c1) return 0; + if (c1 != c2) { return c1 - c2; } + if (!c1) { return 0; } str1++; str2++; - } return 0; - } int __libqasan_strncasecmp(const char *str1, const char *str2, size_t len) { - while (len--) { - const unsigned char c1 = tolower(*str1), c2 = tolower(*str2); - if (c1 != c2) return c1 - c2; - if (!c1) return 0; + if (c1 != c2) { return c1 - c2; } + if (!c1) { return 0; } str1++; str2++; - } return 0; - } int __libqasan_memcmp(const void *mem1, const void *mem2, size_t len) { - const char *strmem1 = (const char *)mem1; const char *strmem2 = (const char *)mem2; while (len--) { - const unsigned char c1 = *strmem1, c2 = *strmem2; if (c1 != c2) return (c1 > c2) ? 1 : -1; strmem1++; strmem2++; - } return 0; - } int __libqasan_bcmp(const void *mem1, const void *mem2, size_t len) { - const char *strmem1 = (const char *)mem1; const char *strmem2 = (const char *)mem2; while (len--) { - int diff = *strmem1 ^ *strmem2; if (diff != 0) return 1; strmem1++; strmem2++; - } return 0; - } char *__libqasan_strstr(const char *haystack, const char *needle) { - do { - const char *n = needle; const char *h = haystack; - while (*n && *h && *n == *h) - n++, h++; + while (*n && *h && *n == *h) { + n++; + h++; + } - if (!*n) return (char *)haystack; + if (!*n) { return (char *)haystack; } } while (*(haystack++)); return 0; - } char *__libqasan_strcasestr(const char *haystack, const char *needle) { - do { - const char *n = needle; const char *h = haystack; - while (*n && *h && tolower(*n) == tolower(*h)) - n++, h++; + while (*n && *h && tolower(*n) == tolower(*h)) { + n++; + h++; + } if (!*n) return (char *)haystack; } while (*(haystack++)); return 0; - } void *__libqasan_memmem(const void *haystack, size_t haystack_len, const void *needle, size_t needle_len) { - const char *n = (const char *)needle; const char *h = (const char *)haystack; if (haystack_len < needle_len) return 0; @@ -264,76 +227,59 @@ void *__libqasan_memmem(const void *haystack, size_t haystack_len, const char *end = h + (haystack_len - needle_len); do { - if (*h == *n) { - if (memcmp(h, n, needle_len) == 0) return (void *)h; - } } while (++h <= end); return 0; - } char *__libqasan_strchr(const char *s, int c) { - - while (*s != (char)c) - if (!*s++) return 0; + while (*s != (char)c) { + if (!*s++) { return 0; } + } return (char *)s; - } char *__libqasan_strrchr(const char *s, int c) { - char *r = NULL; - do - if (*s == (char)c) r = (char *)s; - while (*s++); + do { + if (*s == (char)c) { r = (char *)s; } + } while (*s++); return r; - } size_t __libqasan_wcslen(const wchar_t *s) { - size_t len = 0; while (s[len] != L'\0') { - if (s[++len] == L'\0') return len; if (s[++len] == L'\0') return len; if (s[++len] == L'\0') return len; ++len; - } return len; - } wchar_t *__libqasan_wcscpy(wchar_t *d, const wchar_t *s) { - wchar_t *a = d; while ((*d++ = *s++)) ; return a; - } int __libqasan_wcscmp(const wchar_t *s1, const wchar_t *s2) { - wchar_t c1, c2; do { - c1 = *s1++; c2 = *s2++; - if (c2 == L'\0') return c1 - c2; + if (c2 == L'\0') { return c1 - c2; } } while (c1 == c2); return c1 < c2 ? -1 : 1; - } - diff --git a/libafl_qemu/libqasan/uninstrument.c b/libafl_qemu/libqasan/uninstrument.c index 5bf841a3c6..b3c7adf561 100644 --- a/libafl_qemu/libqasan/uninstrument.c +++ b/libafl_qemu/libqasan/uninstrument.c @@ -43,17 +43,15 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #define X_GET_FNDECL(type, name) type name #define GET_FNDECL(x) X_GET_FNDECL x -#define HOOK_UNINSTRUMENT(rettype, name, ...) \ - rettype (*__lq_libc_##name)(MAP_LIST(GET_FNTYPE, __VA_ARGS__)); \ - rettype name(MAP_LIST(GET_FNDECL, __VA_ARGS__)) { \ - \ - if (!(__lq_libc_##name)) __lq_libc_##name = ASSERT_DLSYM(name); \ - int state = QASAN_SWAP(QASAN_DISABLED); \ - rettype r = __lq_libc_##name(MAP_LIST(GET_FNPAR, __VA_ARGS__)); \ - QASAN_SWAP(state); \ - \ - return r; \ - \ +#define HOOK_UNINSTRUMENT(rettype, name, ...) \ + rettype (*__lq_libc_##name)(MAP_LIST(GET_FNTYPE, __VA_ARGS__)); \ + rettype name(MAP_LIST(GET_FNDECL, __VA_ARGS__)) { \ + if (!(__lq_libc_##name)) { __lq_libc_##name = ASSERT_DLSYM(name); } \ + int state = QASAN_SWAP(QASAN_DISABLED); \ + rettype r = __lq_libc_##name(MAP_LIST(GET_FNPAR, __VA_ARGS__)); \ + QASAN_SWAP(state); \ + \ + return r; \ } HOOK_UNINSTRUMENT(char *, getenv, (const char *, name)) @@ -80,4 +78,3 @@ HOOK_UNINSTRUMENT(int, getpwnam_r, (const char *, name), (struct passwd *, pwd), HOOK_UNINSTRUMENT(int, getpwuid_r, (uid_t, uid), (struct passwd *, pwd), (char *, buf), (size_t, buflen), (struct passwd **, result)) */ - diff --git a/libafl_qemu/src/asan-giovese.c b/libafl_qemu/src/asan-giovese.c index 333b08f916..522bcff721 100644 --- a/libafl_qemu/src/asan-giovese.c +++ b/libafl_qemu/src/asan-giovese.c @@ -50,11 +50,9 @@ extern uintptr_t guest_base; // TODO use a mutex for locking insert/delete struct alloc_tree_node { - struct rb_node rb; struct chunk_info ckinfo; target_ulong __subtree_last; - }; #define START(node) ((node)->ckinfo.start) @@ -65,54 +63,46 @@ INTERVAL_TREE_DEFINE(struct alloc_tree_node, rb, target_ulong, __subtree_last, static struct rb_root root = RB_ROOT; -struct chunk_info* asan_giovese_alloc_search(target_ulong query) { - - struct alloc_tree_node* node = alloc_tree_iter_first(&root, query, query); +struct chunk_info *asan_giovese_alloc_search(target_ulong query) { + struct alloc_tree_node *node = alloc_tree_iter_first(&root, query, query); if (node) return &node->ckinfo; return NULL; - } void asan_giovese_alloc_remove(target_ulong start, target_ulong end) { - - struct alloc_tree_node* prev_node = alloc_tree_iter_first(&root, start, end); + struct alloc_tree_node *prev_node = alloc_tree_iter_first(&root, start, end); while (prev_node) { - - struct alloc_tree_node* n = alloc_tree_iter_next(prev_node, start, end); + struct alloc_tree_node *n = alloc_tree_iter_next(prev_node, start, end); free(prev_node->ckinfo.alloc_ctx); free(prev_node->ckinfo.free_ctx); alloc_tree_remove(prev_node, &root); prev_node = n; - } - } void asan_giovese_alloc_insert(target_ulong start, target_ulong end, - struct call_context* alloc_ctx) { - + struct call_context *alloc_ctx) { asan_giovese_alloc_remove(start, end); - struct alloc_tree_node* node = calloc(sizeof(struct alloc_tree_node), 1); + struct alloc_tree_node *node = calloc(sizeof(struct alloc_tree_node), 1); node->ckinfo.start = start; node->ckinfo.end = end; node->ckinfo.alloc_ctx = alloc_ctx; alloc_tree_insert(node, &root); - } // ------------------------------------------------------------------------- // // Init // ------------------------------------------------------------------------- // -void* __ag_high_shadow = HIGH_SHADOW_ADDR; -void* __ag_low_shadow = LOW_SHADOW_ADDR; +void *__ag_high_shadow = HIGH_SHADOW_ADDR; +void *__ag_low_shadow = LOW_SHADOW_ADDR; void asan_giovese_init(void) { - #if UINTPTR_MAX == 0xffffffff - fprintf(stderr, "ERROR: Cannot allocate sanitizer shadow memory on 32 bit " - "platforms."); + fprintf(stderr, + "ERROR: Cannot allocate sanitizer shadow memory on 32 bit " + "platforms."); exit(1); #else assert(mmap(__ag_high_shadow, HIGH_SHADOW_SIZE, PROT_READ | PROT_WRITE, @@ -127,85 +117,67 @@ void asan_giovese_init(void) { MAP_PRIVATE | MAP_FIXED | MAP_NORESERVE | MAP_ANON, -1, 0) != MAP_FAILED); #endif - } // ------------------------------------------------------------------------- // // Checks // ------------------------------------------------------------------------- // -int asan_giovese_load1(void* ptr) { - +int asan_giovese_load1(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && (intptr_t)((h & 7) + 1) > k; - } -int asan_giovese_load2(void* ptr) { - +int asan_giovese_load2(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && (intptr_t)((h & 7) + 2) > k; - } -int asan_giovese_load4(void* ptr) { - +int asan_giovese_load4(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && (intptr_t)((h & 7) + 4) > k; - } -int asan_giovese_load8(void* ptr) { - +int asan_giovese_load8(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; return (*shadow_addr); - } -int asan_giovese_store1(void* ptr) { - +int asan_giovese_store1(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && (intptr_t)((h & 7) + 1) > k; - } -int asan_giovese_store2(void* ptr) { - +int asan_giovese_store2(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && (intptr_t)((h & 7) + 2) > k; - } -int asan_giovese_store4(void* ptr) { - +int asan_giovese_store4(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && (intptr_t)((h & 7) + 4) > k; - } -int asan_giovese_store8(void* ptr) { - +int asan_giovese_store8(void *ptr) { uintptr_t h = (uintptr_t)ptr; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; return (*shadow_addr); - } -int asan_giovese_loadN(void* ptr, size_t n) { - +int asan_giovese_loadN(void *ptr, size_t n) { if (!n) return 0; uintptr_t start = (uintptr_t)ptr; @@ -213,53 +185,43 @@ int asan_giovese_loadN(void* ptr, size_t n) { uintptr_t last_8 = end & ~7; if (start & 0x7) { - uintptr_t next_8 = (start & ~7) + 8; - size_t first_size = next_8 - start; + size_t first_size = next_8 - start; if (n <= first_size) { - uintptr_t h = start; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + n) > k); - } uintptr_t h = start; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; if (k != 0 && ((intptr_t)((h & 7) + first_size) > k)) return 1; start = next_8; - } while (start < last_8) { - uintptr_t h = start; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; if (*shadow_addr) return 1; start += 8; - } if (last_8 != end) { - uintptr_t h = start; size_t last_size = end - last_8; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + last_size) > k); - } return 0; - } -int asan_giovese_storeN(void* ptr, size_t n) { - +int asan_giovese_storeN(void *ptr, size_t n) { if (!n) return 0; uintptr_t start = (uintptr_t)ptr; @@ -267,53 +229,43 @@ int asan_giovese_storeN(void* ptr, size_t n) { uintptr_t last_8 = end & ~7; if (start & 0x7) { - uintptr_t next_8 = (start & ~7) + 8; - size_t first_size = next_8 - start; + size_t first_size = next_8 - start; if (n <= first_size) { - uintptr_t h = start; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + n) > k); - } uintptr_t h = start; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; if (k != 0 && ((intptr_t)((h & 7) + first_size) > k)) return 1; start = next_8; - } while (start < last_8) { - uintptr_t h = start; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; if (*shadow_addr) return 1; start += 8; - } if (last_8 != end) { - uintptr_t h = start; size_t last_size = end - last_8; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + last_size) > k); - } return 0; - } int asan_giovese_guest_loadN(target_ulong addr, size_t n) { - if (!n) return 0; target_ulong start = addr; @@ -321,53 +273,43 @@ int asan_giovese_guest_loadN(target_ulong addr, size_t n) { target_ulong last_8 = end & ~7; if (start & 0x7) { - target_ulong next_8 = (start & ~7) + 8; size_t first_size = next_8 - start; if (n <= first_size) { - uintptr_t h = (uintptr_t)AFL_G2H(start); - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + n) > k); - } uintptr_t h = (uintptr_t)AFL_G2H(start); - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; if (k != 0 && ((intptr_t)((h & 7) + first_size) > k)) return 1; start = next_8; - } while (start < last_8) { - uintptr_t h = (uintptr_t)AFL_G2H(start); - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; if (*shadow_addr) return 1; start += 8; - } if (last_8 != end) { - uintptr_t h = (uintptr_t)AFL_G2H(start); size_t last_size = end - last_8; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + last_size) > k); - } return 0; - } int asan_giovese_guest_storeN(target_ulong addr, size_t n) { - if (!n) return 0; target_ulong start = addr; @@ -375,58 +317,47 @@ int asan_giovese_guest_storeN(target_ulong addr, size_t n) { target_ulong last_8 = end & ~7; if (start & 0x7) { - target_ulong next_8 = (start & ~7) + 8; size_t first_size = next_8 - start; if (n <= first_size) { - uintptr_t h = (uintptr_t)AFL_G2H(start); - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + n) > k); - } uintptr_t h = (uintptr_t)AFL_G2H(start); - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; if (k != 0 && ((intptr_t)((h & 7) + first_size) > k)) return 1; start = next_8; - } while (start < last_8) { - uintptr_t h = (uintptr_t)AFL_G2H(start); - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; if (*shadow_addr) return 1; start += 8; - } if (last_8 != end) { - uintptr_t h = (uintptr_t)AFL_G2H(start); size_t last_size = end - last_8; - int8_t* shadow_addr = (int8_t*)(h >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(h >> 3) + SHADOW_OFFSET; int8_t k = *shadow_addr; return k != 0 && ((intptr_t)((h & 7) + last_size) > k); - } return 0; - } // ------------------------------------------------------------------------- // // Poison // ------------------------------------------------------------------------- // -int asan_giovese_poison_region(void* ptr, size_t n, - uint8_t poison_byte) { - +int asan_giovese_poison_region(void *ptr, size_t n, uint8_t poison_byte) { if (!n) return 0; uintptr_t start = (uintptr_t)ptr; @@ -434,119 +365,95 @@ int asan_giovese_poison_region(void* ptr, size_t n, uintptr_t last_8 = end & ~7; if (start & 0x7) { - target_ulong next_8 = (start & ~7) + 8; size_t first_size = next_8 - start; if (n < first_size) return 0; uintptr_t h = start; - uint8_t* shadow_addr = (uint8_t*)(h >> 3) + SHADOW_OFFSET; + uint8_t *shadow_addr = (uint8_t *)(h >> 3) + SHADOW_OFFSET; *shadow_addr = 8 - first_size; start = next_8; - } while (start < last_8) { - uintptr_t h = start; - uint8_t* shadow_addr = (uint8_t*)(h >> 3) + SHADOW_OFFSET; + uint8_t *shadow_addr = (uint8_t *)(h >> 3) + SHADOW_OFFSET; *shadow_addr = poison_byte; start += 8; - } return 1; - } -int asan_giovese_user_poison_region(void* ptr, size_t n) { - +int asan_giovese_user_poison_region(void *ptr, size_t n) { return asan_giovese_poison_region(ptr, n, ASAN_USER); - } -int asan_giovese_unpoison_region(void* ptr, size_t n) { - +int asan_giovese_unpoison_region(void *ptr, size_t n) { target_ulong start = (uintptr_t)ptr; target_ulong end = start + n; while (start < end) { - uintptr_t h = start; - uint8_t* shadow_addr = (uint8_t*)(h >> 3) + SHADOW_OFFSET; + uint8_t *shadow_addr = (uint8_t *)(h >> 3) + SHADOW_OFFSET; *shadow_addr = 0; start += 8; - } return 1; - } int asan_giovese_poison_guest_region(target_ulong addr, size_t n, uint8_t poison_byte) { - if (!n) return 0; - + target_ulong start = addr; target_ulong end = start + n; target_ulong last_8 = end & ~7; - - if (start & 0x7) { + if (start & 0x7) { target_ulong next_8 = (start & ~7) + 8; size_t first_size = next_8 - start; if (n < first_size) return 0; uintptr_t h = (uintptr_t)AFL_G2H(start); - uint8_t* shadow_addr = (uint8_t*)(h >> 3) + SHADOW_OFFSET; + uint8_t *shadow_addr = (uint8_t *)(h >> 3) + SHADOW_OFFSET; *shadow_addr = 8 - first_size; start = next_8; - } while (start < last_8) { - uintptr_t h = (uintptr_t)AFL_G2H(start); - uint8_t* shadow_addr = (uint8_t*)(h >> 3) + SHADOW_OFFSET; + uint8_t *shadow_addr = (uint8_t *)(h >> 3) + SHADOW_OFFSET; *shadow_addr = poison_byte; start += 8; - } return 1; - } int asan_giovese_user_poison_guest_region(target_ulong addr, size_t n) { - return asan_giovese_poison_guest_region(addr, n, ASAN_USER); - } int asan_giovese_unpoison_guest_region(target_ulong addr, size_t n) { - target_ulong start = addr; target_ulong end = start + n; while (start < end) { - uintptr_t h = (uintptr_t)AFL_G2H(start); - uint8_t* shadow_addr = (uint8_t*)(h >> 3) + SHADOW_OFFSET; + uint8_t *shadow_addr = (uint8_t *)(h >> 3) + SHADOW_OFFSET; *shadow_addr = 0; start += 8; - } return 1; - } - // ------------------------------------------------------------------------- // // Report // ------------------------------------------------------------------------- // @@ -576,7 +483,7 @@ int asan_giovese_unpoison_guest_region(target_ulong addr, size_t n) { // Reset #define ANSI_COLOR_RESET "\e[0m" -static const char* shadow_color_map[] = { +static const char *shadow_color_map[] = { "" /* 0x0 */, "" /* 0x1 */, @@ -833,42 +740,39 @@ static const char* shadow_color_map[] = { ANSI_COLOR_HBLU /* 0xfc */, ANSI_COLOR_HMAG /* 0xfd */, ANSI_COLOR_HYEL /* 0xfe */, - "" /* 0xff */ + "" /* 0xff */ }; -static const char* access_type_str[] = {"READ", "WRITE"}; - -static const char* poisoned_strerror(uint8_t poison_byte) { +static const char *access_type_str[] = {"READ", "WRITE"}; +static const char *poisoned_strerror(uint8_t poison_byte) { switch (poison_byte) { - case ASAN_HEAP_RZ: case ASAN_HEAP_LEFT_RZ: - case ASAN_HEAP_RIGHT_RZ: return "heap-buffer-overflow"; - case ASAN_HEAP_FREED: return "heap-use-after-free"; - + case ASAN_HEAP_RIGHT_RZ: + return "heap-buffer-overflow"; + case ASAN_HEAP_FREED: + return "heap-use-after-free"; } return "use-after-poison"; - } static int poisoned_find_error(target_ulong addr, size_t n, - target_ulong* fault_addr, - const char** err_string) { - + target_ulong *fault_addr, + const char **err_string) { target_ulong start = addr; target_ulong end = start + n; int have_partials = 0; while (start < end) { - uintptr_t rs = (uintptr_t)AFL_G2H(start); - int8_t* shadow_addr = (int8_t*)(rs >> 3) + SHADOW_OFFSET; + int8_t *shadow_addr = (int8_t *)(rs >> 3) + SHADOW_OFFSET; switch (*shadow_addr) { - - case ASAN_VALID: have_partials = 0; break; + case ASAN_VALID: + have_partials = 0; + break; case ASAN_PARTIAL1: case ASAN_PARTIAL2: case ASAN_PARTIAL3: @@ -876,49 +780,39 @@ static int poisoned_find_error(target_ulong addr, size_t n, case ASAN_PARTIAL5: case ASAN_PARTIAL6: case ASAN_PARTIAL7: { - have_partials = 1; target_ulong a = (start & ~7) + *shadow_addr; if (*fault_addr == 0 && a >= start && a < end) *fault_addr = a; break; - } default: { - if (*fault_addr == 0) *fault_addr = start; *err_string = poisoned_strerror(*shadow_addr); return 1; - } - } start += 8; - } if (have_partials) { - uintptr_t rs = (uintptr_t)AFL_G2H((end & ~7) + 8); - uint8_t* last_shadow_addr = (uint8_t*)(rs >> 3) + SHADOW_OFFSET; + uint8_t *last_shadow_addr = (uint8_t *)(rs >> 3) + SHADOW_OFFSET; *err_string = poisoned_strerror(*last_shadow_addr); return 1; - } if (*fault_addr == 0) *fault_addr = addr; *err_string = "use-after-poison"; return 1; - } -#define _MEM2SHADOW(x) ((uint8_t*)((uintptr_t)AFL_G2H(x) >> 3) + SHADOW_OFFSET) +#define _MEM2SHADOW(x) ((uint8_t *)((uintptr_t)AFL_G2H(x) >> 3) + SHADOW_OFFSET) #define _MEM2SHADOWPRINT(x) shadow_color_map[*_MEM2SHADOW(x)], *_MEM2SHADOW(x) static int print_shadow_line(target_ulong addr) { - fprintf(stderr, " 0x%012" PRIxPTR ": %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET @@ -941,13 +835,11 @@ static int print_shadow_line(target_ulong addr) { _MEM2SHADOWPRINT(addr + 120)); return 1; - } static int print_shadow_line_fault(target_ulong addr, target_ulong fault_addr) { - int i = (fault_addr - addr) / 8; - const char* format = + const char *format = "=>0x%012" PRIxPTR ": %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET @@ -958,7 +850,6 @@ static int print_shadow_line_fault(target_ulong addr, target_ulong fault_addr) { " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET "\n"; switch (i) { - case 0: format = "=>0x%012" PRIxPTR ":[%s%02x" ANSI_COLOR_RESET "]%s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET @@ -1151,7 +1042,6 @@ static int print_shadow_line_fault(target_ulong addr, target_ulong fault_addr) { "%s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET " %s%02x" ANSI_COLOR_RESET "[%s%02x" ANSI_COLOR_RESET "]\n"; break; - } fprintf(stderr, format, (uintptr_t)_MEM2SHADOW(addr), _MEM2SHADOWPRINT(addr), @@ -1165,14 +1055,12 @@ static int print_shadow_line_fault(target_ulong addr, target_ulong fault_addr) { _MEM2SHADOWPRINT(addr + 120)); return 1; - } #undef _MEM2SHADOW #undef _MEM2SHADOWPRINT static void print_shadow(target_ulong addr) { - target_ulong center = addr & ~127; print_shadow_line(center - 16 * 8 * 5); print_shadow_line(center - 16 * 8 * 4); @@ -1185,46 +1073,42 @@ static void print_shadow(target_ulong addr) { print_shadow_line(center + 16 * 8 * 3); print_shadow_line(center + 16 * 8 * 4); print_shadow_line(center + 16 * 8 * 5); - } -static void print_alloc_location_chunk(struct chunk_info* ckinfo, +static void print_alloc_location_chunk(struct chunk_info *ckinfo, target_ulong fault_addr) { - if (fault_addr >= ckinfo->start && fault_addr < ckinfo->end) fprintf(stderr, - ANSI_COLOR_HGRN - "0x" TARGET_FMT_lx " is located " TARGET_FMT_ld - " bytes inside of " TARGET_FMT_ld "-byte region [0x" - TARGET_FMT_lx ",0x" TARGET_FMT_lx ")" ANSI_COLOR_RESET "\n", + ANSI_COLOR_HGRN "0x" TARGET_FMT_lx " is located " TARGET_FMT_ld + " bytes inside of " TARGET_FMT_ld + "-byte region [0x" TARGET_FMT_lx ",0x" TARGET_FMT_lx + ")" ANSI_COLOR_RESET "\n", fault_addr, fault_addr - ckinfo->start, ckinfo->end - ckinfo->start, ckinfo->start, ckinfo->end); else if (ckinfo->start >= fault_addr) fprintf(stderr, - ANSI_COLOR_HGRN - "0x" TARGET_FMT_lx " is located " TARGET_FMT_ld - " bytes to the left of " TARGET_FMT_ld "-byte region [0x" - TARGET_FMT_lx ",0x" TARGET_FMT_lx ")" ANSI_COLOR_RESET "\n", + ANSI_COLOR_HGRN "0x" TARGET_FMT_lx " is located " TARGET_FMT_ld + " bytes to the left of " TARGET_FMT_ld + "-byte region [0x" TARGET_FMT_lx ",0x" TARGET_FMT_lx + ")" ANSI_COLOR_RESET "\n", fault_addr, ckinfo->start - fault_addr, ckinfo->end - ckinfo->start, ckinfo->start, ckinfo->end); else fprintf(stderr, - ANSI_COLOR_HGRN - "0x" TARGET_FMT_lx " is located " TARGET_FMT_ld - " bytes to the right of " TARGET_FMT_ld "-byte region [0x" - TARGET_FMT_lx ",0x" TARGET_FMT_lx ")" ANSI_COLOR_RESET "\n", + ANSI_COLOR_HGRN "0x" TARGET_FMT_lx " is located " TARGET_FMT_ld + " bytes to the right of " TARGET_FMT_ld + "-byte region [0x" TARGET_FMT_lx ",0x" TARGET_FMT_lx + ")" ANSI_COLOR_RESET "\n", fault_addr, fault_addr - ckinfo->end, ckinfo->end - ckinfo->start, ckinfo->start, ckinfo->end); if (ckinfo->free_ctx) { - fprintf(stderr, ANSI_COLOR_HMAG "freed by thread T%d here:" ANSI_COLOR_RESET "\n", ckinfo->free_ctx->tid); size_t i; for (i = 0; i < ckinfo->free_ctx->size; ++i) { - - char* printable = asan_giovese_printaddr(ckinfo->free_ctx->addresses[i]); + char *printable = asan_giovese_printaddr(ckinfo->free_ctx->addresses[i]); if (printable) fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "%s\n", i, ckinfo->free_ctx->addresses[i], printable); @@ -1249,97 +1133,85 @@ static void print_alloc_location_chunk(struct chunk_info* ckinfo, size_t i; for (i = 0; i < ckinfo->alloc_ctx->size; ++i) { - - char* printable = asan_giovese_printaddr(ckinfo->alloc_ctx->addresses[i]); + char *printable = asan_giovese_printaddr(ckinfo->alloc_ctx->addresses[i]); if (printable) fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "%s\n", i, ckinfo->alloc_ctx->addresses[i], printable); else fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "\n", i, ckinfo->alloc_ctx->addresses[i]); - } fputc('\n', stderr); - } static void print_alloc_location(target_ulong addr, target_ulong fault_addr) { - - struct chunk_info* ckinfo = asan_giovese_alloc_search(fault_addr); + struct chunk_info *ckinfo = asan_giovese_alloc_search(fault_addr); if (!ckinfo && addr != fault_addr) ckinfo = asan_giovese_alloc_search(addr); if (ckinfo) { - print_alloc_location_chunk(ckinfo, fault_addr); return; - } int i = 0; - while (!ckinfo && i < DEFAULT_REDZONE_SIZE) + while (!ckinfo && i < DEFAULT_REDZONE_SIZE) { ckinfo = asan_giovese_alloc_search(fault_addr - (i++)); + } if (ckinfo) { - print_alloc_location_chunk(ckinfo, fault_addr); return; - } i = 0; - while (!ckinfo && i < DEFAULT_REDZONE_SIZE) + while (!ckinfo && i < DEFAULT_REDZONE_SIZE) { ckinfo = asan_giovese_alloc_search(fault_addr + (i++)); + } if (ckinfo) { - print_alloc_location_chunk(ckinfo, fault_addr); return; - } fprintf(stderr, "Address 0x" TARGET_FMT_lx " is a wild pointer.\n", fault_addr); - } int asan_giovese_report_and_crash(int access_type, target_ulong addr, size_t n, target_ulong pc, target_ulong bp, target_ulong sp) { - struct call_context ctx; asan_giovese_populate_context(&ctx, pc); target_ulong fault_addr = 0; - const char* error_type; + const char *error_type; if (!poisoned_find_error(addr, n, &fault_addr, &error_type)) return 0; - + fprintf(stderr, - "=================================================================\n" - ANSI_COLOR_HRED "==%d==ERROR: " ASAN_NAME_STR ": %s on address 0x" - TARGET_FMT_lx " at pc 0x" TARGET_FMT_lx " bp 0x" TARGET_FMT_lx - " sp 0x" TARGET_FMT_lx ANSI_COLOR_RESET "\n", + "=================================================================" + "\n" ANSI_COLOR_HRED "==%d==ERROR: " ASAN_NAME_STR + ": %s on address 0x" TARGET_FMT_lx " at pc 0x" TARGET_FMT_lx + " bp 0x" TARGET_FMT_lx " sp 0x" TARGET_FMT_lx ANSI_COLOR_RESET "\n", getpid(), error_type, addr, pc, bp, sp); fprintf(stderr, - ANSI_COLOR_HBLU "%s of size %zu at 0x" TARGET_FMT_lx " thread T%d" - ANSI_COLOR_RESET "\n", + ANSI_COLOR_HBLU "%s of size %zu at 0x" TARGET_FMT_lx + " thread T%d" ANSI_COLOR_RESET "\n", access_type_str[access_type], n, addr, ctx.tid); size_t i; for (i = 0; i < ctx.size; ++i) { - - char* printable = asan_giovese_printaddr(ctx.addresses[i]); + char *printable = asan_giovese_printaddr(ctx.addresses[i]); if (printable) fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "%s\n", i, ctx.addresses[i], printable); else fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "\n", i, ctx.addresses[i]); - } fputc('\n', stderr); print_alloc_location(addr, fault_addr); - const char* printable_pc = asan_giovese_printaddr(pc); + const char *printable_pc = asan_giovese_printaddr(pc); if (!printable_pc) printable_pc = ""; fprintf(stderr, "SUMMARY: " ASAN_NAME_STR @@ -1354,9 +1226,12 @@ int asan_giovese_report_and_crash(int access_type, target_ulong addr, size_t n, "Shadow byte legend (one shadow byte represents 8 application bytes):\n" " Addressable: 00\n" " Partially addressable: 01 02 03 04 05 06 07\n" - " Heap left redzone: " ANSI_COLOR_HRED "fa" ANSI_COLOR_RESET "\n" - " Heap right redzone: " ANSI_COLOR_HRED "fb" ANSI_COLOR_RESET "\n" - " Freed heap region: " ANSI_COLOR_HMAG "fd" ANSI_COLOR_RESET "\n" + " Heap left redzone: " ANSI_COLOR_HRED "fa" ANSI_COLOR_RESET + "\n" + " Heap right redzone: " ANSI_COLOR_HRED "fb" ANSI_COLOR_RESET + "\n" + " Freed heap region: " ANSI_COLOR_HMAG "fd" ANSI_COLOR_RESET + "\n" //" Stack left redzone: " ANSI_COLOR_HRED "f1" ANSI_COLOR_RESET "\n" //" Stack mid redzone: " ANSI_COLOR_HRED "f2" ANSI_COLOR_RESET "\n" //" Stack right redzone: " ANSI_COLOR_HRED "f3" ANSI_COLOR_RESET "\n" @@ -1364,11 +1239,13 @@ int asan_giovese_report_and_crash(int access_type, target_ulong addr, size_t n, //" Stack use after scope: " ANSI_COLOR_HMAG "f8" ANSI_COLOR_RESET "\n" //" Global redzone: " ANSI_COLOR_HRED "f9" ANSI_COLOR_RESET "\n" //" Global init order: " ANSI_COLOR_HCYN "f6" ANSI_COLOR_RESET "\n" - " Poisoned by user: " ANSI_COLOR_HBLU "f7" ANSI_COLOR_RESET "\n" + " Poisoned by user: " ANSI_COLOR_HBLU "f7" ANSI_COLOR_RESET + "\n" //" Container overflow: " ANSI_COLOR_HBLU "fc" ANSI_COLOR_RESET "\n" //" Array cookie: " ANSI_COLOR_HRED "ac" ANSI_COLOR_RESET "\n" //" Intra object redzone: " ANSI_COLOR_HYEL "bb" ANSI_COLOR_RESET "\n" - " ASan internal: " ANSI_COLOR_HYEL "fe" ANSI_COLOR_RESET "\n" + " ASan internal: " ANSI_COLOR_HYEL "fe" ANSI_COLOR_RESET + "\n" //" Left alloca redzone: " ANSI_COLOR_HBLU "ca" ANSI_COLOR_RESET "\n" //" Right alloca redzone: " ANSI_COLOR_HBLU "cb" ANSI_COLOR_RESET "\n" " Shadow gap: cc\n" @@ -1377,55 +1254,35 @@ int asan_giovese_report_and_crash(int access_type, target_ulong addr, size_t n, // signal(SIGABRT, SIG_DFL); abort(); - } -static const char* singal_to_string[] = { - [SIGHUP] = "HUP", - [SIGINT] = "INT", - [SIGQUIT] = "QUIT", - [SIGILL] = "ILL", - [SIGTRAP] = "TRAP", - [SIGABRT] = "ABRT", - [SIGBUS] = "BUS", - [SIGFPE] = "FPE", - [SIGKILL] = "KILL", - [SIGUSR1] = "USR1", - [SIGSEGV] = "SEGV", - [SIGUSR2] = "USR2", - [SIGPIPE] = "PIPE", - [SIGALRM] = "ALRM", - [SIGTERM] = "TERM", +static const char *singal_to_string[] = { + [SIGHUP] = "HUP", [SIGINT] = "INT", [SIGQUIT] = "QUIT", + [SIGILL] = "ILL", [SIGTRAP] = "TRAP", [SIGABRT] = "ABRT", + [SIGBUS] = "BUS", [SIGFPE] = "FPE", [SIGKILL] = "KILL", + [SIGUSR1] = "USR1", [SIGSEGV] = "SEGV", [SIGUSR2] = "USR2", + [SIGPIPE] = "PIPE", [SIGALRM] = "ALRM", [SIGTERM] = "TERM", #ifdef SIGSTKFLT [SIGSTKFLT] = "STKFLT", #endif - [SIGCHLD] = "CHLD", - [SIGCONT] = "CONT", - [SIGSTOP] = "STOP", - [SIGTSTP] = "TSTP", - [SIGTTIN] = "TTIN", - [SIGTTOU] = "TTOU", - [SIGURG] = "URG", - [SIGXCPU] = "XCPU", - [SIGXFSZ] = "XFSZ", - [SIGVTALRM] = "VTALRM", - [SIGPROF] = "PROF", - [SIGWINCH] = "WINCH", - [SIGIO] = "IO", - [SIGPWR] = "PWR", - [SIGSYS] = "SYS", + [SIGCHLD] = "CHLD", [SIGCONT] = "CONT", [SIGSTOP] = "STOP", + [SIGTSTP] = "TSTP", [SIGTTIN] = "TTIN", [SIGTTOU] = "TTOU", + [SIGURG] = "URG", [SIGXCPU] = "XCPU", [SIGXFSZ] = "XFSZ", + [SIGVTALRM] = "VTALRM", [SIGPROF] = "PROF", [SIGWINCH] = "WINCH", + [SIGIO] = "IO", [SIGPWR] = "PWR", [SIGSYS] = "SYS", }; -int asan_giovese_deadly_signal(int signum, target_ulong addr, target_ulong pc, target_ulong bp, target_ulong sp) { - +int asan_giovese_deadly_signal(int signum, target_ulong addr, target_ulong pc, + target_ulong bp, target_ulong sp) { struct call_context ctx; asan_giovese_populate_context(&ctx, pc); - const char* error_type = singal_to_string[signum]; + const char *error_type = singal_to_string[signum]; fprintf(stderr, - ASAN_NAME_STR ":DEADLYSIGNAL\n" - "=================================================================\n" - ANSI_COLOR_HRED "==%d==ERROR: " ASAN_NAME_STR + ASAN_NAME_STR + ":DEADLYSIGNAL\n" + "=================================================================" + "\n" ANSI_COLOR_HRED "==%d==ERROR: " ASAN_NAME_STR ": %s on unknown address 0x" TARGET_FMT_lx " (pc 0x" TARGET_FMT_lx " bp 0x" TARGET_FMT_lx " sp 0x" TARGET_FMT_lx " T%d)" ANSI_COLOR_RESET "\n", @@ -1433,66 +1290,55 @@ int asan_giovese_deadly_signal(int signum, target_ulong addr, target_ulong pc, t size_t i; for (i = 0; i < ctx.size; ++i) { - - char* printable = asan_giovese_printaddr(ctx.addresses[i]); + char *printable = asan_giovese_printaddr(ctx.addresses[i]); if (printable) fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "%s\n", i, ctx.addresses[i], printable); else fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "\n", i, ctx.addresses[i]); - } - + fputc('\n', stderr); fprintf(stderr, ASAN_NAME_STR " can not provide additional info.\n"); - - const char* printable_pc = asan_giovese_printaddr(pc); + + const char *printable_pc = asan_giovese_printaddr(pc); if (!printable_pc) printable_pc = ""; - fprintf(stderr, - "SUMMARY: " ASAN_NAME_STR - ": %s\n", printable_pc); + fprintf(stderr, "SUMMARY: " ASAN_NAME_STR ": %s\n", printable_pc); fprintf(stderr, "==%d==ABORTING\n", getpid()); return signum; - } int asan_giovese_badfree(target_ulong addr, target_ulong pc) { - struct call_context ctx; asan_giovese_populate_context(&ctx, pc); - fprintf(stderr, - "=================================================================" - "\n" ANSI_COLOR_HRED "==%d==ERROR: " ASAN_NAME_STR - ": attempting free on address which was not malloc()-ed: 0x" - TARGET_FMT_lx " in thread T%d" ANSI_COLOR_RESET "\n", getpid(), addr, - ctx.tid); + fprintf( + stderr, + "=================================================================" + "\n" ANSI_COLOR_HRED "==%d==ERROR: " ASAN_NAME_STR + ": attempting free on address which was not malloc()-ed: 0x" TARGET_FMT_lx + " in thread T%d" ANSI_COLOR_RESET "\n", + getpid(), addr, ctx.tid); size_t i; for (i = 0; i < ctx.size; ++i) { - - char* printable = asan_giovese_printaddr(ctx.addresses[i]); + char *printable = asan_giovese_printaddr(ctx.addresses[i]); if (printable) fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "%s\n", i, ctx.addresses[i], printable); else fprintf(stderr, " #%zu 0x" TARGET_FMT_lx "\n", i, ctx.addresses[i]); - } - + fputc('\n', stderr); print_alloc_location(addr, addr); - - const char* printable_pc = asan_giovese_printaddr(pc); + + const char *printable_pc = asan_giovese_printaddr(pc); if (!printable_pc) printable_pc = ""; - fprintf(stderr, - "SUMMARY: " ASAN_NAME_STR - ": bad-free %s\n", printable_pc); + fprintf(stderr, "SUMMARY: " ASAN_NAME_STR ": bad-free %s\n", printable_pc); fprintf(stderr, "==%d==ABORTING\n", getpid()); - //signal(SIGABRT, SIG_DFL); + // signal(SIGABRT, SIG_DFL); abort(); - } - diff --git a/libafl_qemu/src/asan-giovese.h b/libafl_qemu/src/asan-giovese.h index f42efa8921..9db2da7921 100644 --- a/libafl_qemu/src/asan-giovese.h +++ b/libafl_qemu/src/asan-giovese.h @@ -38,12 +38,12 @@ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #define TARGET_FMT_ld "%" PRId64 #ifndef ASAN_NAME_STR -#define ASAN_NAME_STR "AddressSanitizer" + #define ASAN_NAME_STR "AddressSanitizer" #endif -#define HIGH_SHADOW_ADDR ((void*)0x02008fff7000ULL) -#define LOW_SHADOW_ADDR ((void*)0x00007fff8000ULL) -#define GAP_SHADOW_ADDR ((void*)0x00008fff7000) +#define HIGH_SHADOW_ADDR ((void *)0x02008fff7000ULL) +#define LOW_SHADOW_ADDR ((void *)0x00007fff8000ULL) +#define GAP_SHADOW_ADDR ((void *)0x00008fff7000) #define HIGH_SHADOW_SIZE (0xdfff0000fffULL) #define LOW_SHADOW_SIZE (0xfffefffULL) @@ -82,32 +82,28 @@ enum { }; struct call_context { - - target_ulong* addresses; + target_ulong *addresses; uint32_t tid; uint32_t size; - }; struct chunk_info { - target_ulong start; target_ulong end; - struct call_context* alloc_ctx; - struct call_context* free_ctx; // NULL if chunk is allocated - + struct call_context *alloc_ctx; + struct call_context *free_ctx; // NULL if chunk is allocated }; -extern void* __ag_high_shadow; -extern void* __ag_low_shadow; +extern void *__ag_high_shadow; +extern void *__ag_low_shadow; // ------------------------------------------------------------------------- // // Virtual functions, you have to implement them // ------------------------------------------------------------------------- // /////////////////////////////////////////////////////////////////////////////// -void asan_giovese_populate_context(struct call_context* ctx, target_ulong pc); -char* asan_giovese_printaddr(target_ulong addr); +void asan_giovese_populate_context(struct call_context *ctx, target_ulong pc); +char *asan_giovese_printaddr(target_ulong addr); /////////////////////////////////////////////////////////////////////////////// // ------------------------------------------------------------------------- // @@ -118,25 +114,25 @@ void asan_giovese_init(void); // this has to be fast, ptr is an host pointer -int asan_giovese_load1(void* ptr); -int asan_giovese_load2(void* ptr); -int asan_giovese_load4(void* ptr); -int asan_giovese_load8(void* ptr); -int asan_giovese_store1(void* ptr); -int asan_giovese_store2(void* ptr); -int asan_giovese_store4(void* ptr); -int asan_giovese_store8(void* ptr); -int asan_giovese_loadN(void* ptr, size_t n); -int asan_giovese_storeN(void* ptr, size_t n); +int asan_giovese_load1(void *ptr); +int asan_giovese_load2(void *ptr); +int asan_giovese_load4(void *ptr); +int asan_giovese_load8(void *ptr); +int asan_giovese_store1(void *ptr); +int asan_giovese_store2(void *ptr); +int asan_giovese_store4(void *ptr); +int asan_giovese_store8(void *ptr); +int asan_giovese_loadN(void *ptr, size_t n); +int asan_giovese_storeN(void *ptr, size_t n); int asan_giovese_guest_loadN(target_ulong addr, size_t n); int asan_giovese_guest_storeN(target_ulong addr, size_t n); -int asan_giovese_poison_region(void* ptr, size_t n, - uint8_t poison_byte); -int asan_giovese_user_poison_region(void* ptr, size_t n); -int asan_giovese_unpoison_region(void* ptr, size_t n); +int asan_giovese_poison_region(void *ptr, size_t n, uint8_t poison_byte); +int asan_giovese_user_poison_region(void *ptr, size_t n); +int asan_giovese_unpoison_region(void *ptr, size_t n); -int asan_giovese_poison_guest_region(target_ulong addr, size_t n, uint8_t poison_byte); +int asan_giovese_poison_guest_region(target_ulong addr, size_t n, + uint8_t poison_byte); int asan_giovese_user_poison_guest_region(target_ulong addr, size_t n); int asan_giovese_unpoison_guest_region(target_ulong addr, size_t n); @@ -151,10 +147,9 @@ int asan_giovese_deadly_signal(int signum, target_ulong addr, target_ulong pc, int asan_giovese_badfree(target_ulong addr, target_ulong pc); -struct chunk_info* asan_giovese_alloc_search(target_ulong query); +struct chunk_info *asan_giovese_alloc_search(target_ulong query); void asan_giovese_alloc_remove(target_ulong start, target_ulong end); void asan_giovese_alloc_insert(target_ulong start, target_ulong end, - struct call_context* alloc_ctx); + struct call_context *alloc_ctx); #endif - diff --git a/libafl_qemu/src/interval-tree/compiler.h b/libafl_qemu/src/interval-tree/compiler.h index 1c820b5a06..e9748e158d 100644 --- a/libafl_qemu/src/interval-tree/compiler.h +++ b/libafl_qemu/src/interval-tree/compiler.h @@ -1,5 +1,5 @@ -#ifndef __INT_COMPILER_H__ -#define __INT_COMPILER_H__ +#ifndef __INT_COMPILER_H__ +#define __INT_COMPILER_H__ /** * container_of - cast a member of a structure out to the containing structure @@ -9,9 +9,11 @@ * */ #ifndef container_of -#define container_of(ptr, type, member) ({ \ - const typeof( ((type *)0)->member ) *__mptr = (ptr); \ - (type *)( (char *)__mptr - offsetof(type,member) );}) + #define container_of(ptr, type, member) \ + ({ \ + const typeof(((type *)0)->member) *__mptr = (ptr); \ + (type *)((char *)__mptr - offsetof(type, member)); \ + }) #endif -#endif /* __INT_COMPILER_H__ */ +#endif /* __INT_COMPILER_H__ */ diff --git a/libafl_qemu/src/interval-tree/interval_tree_generic.h b/libafl_qemu/src/interval-tree/interval_tree_generic.h index e26c7322ce..ed8cbb754c 100644 --- a/libafl_qemu/src/interval-tree/interval_tree_generic.h +++ b/libafl_qemu/src/interval-tree/interval_tree_generic.h @@ -39,155 +39,138 @@ * (interval_tree.h) would work for you... */ -#define INTERVAL_TREE_DEFINE(ITSTRUCT, ITRB, ITTYPE, ITSUBTREE, \ - ITSTART, ITLAST, ITSTATIC, ITPREFIX) \ - \ -/* Callbacks for augmented rbtree insert and remove */ \ - \ -static inline ITTYPE ITPREFIX ## _compute_subtree_last(ITSTRUCT *node) \ -{ \ - ITTYPE max = ITLAST(node), subtree_last; \ - if (node->ITRB.rb_left) { \ - subtree_last = rb_entry(node->ITRB.rb_left, \ - ITSTRUCT, ITRB)->ITSUBTREE; \ - if (max < subtree_last) \ - max = subtree_last; \ - } \ - if (node->ITRB.rb_right) { \ - subtree_last = rb_entry(node->ITRB.rb_right, \ - ITSTRUCT, ITRB)->ITSUBTREE; \ - if (max < subtree_last) \ - max = subtree_last; \ - } \ - return max; \ -} \ - \ -RB_DECLARE_CALLBACKS(static, ITPREFIX ## _augment, ITSTRUCT, ITRB, \ - ITTYPE, ITSUBTREE, ITPREFIX ## _compute_subtree_last) \ - \ -/* Insert / remove interval nodes from the tree */ \ - \ -ITSTATIC void ITPREFIX ## _insert(ITSTRUCT *node, struct rb_root *root) \ -{ \ - struct rb_node **link = &root->rb_node, *rb_parent = NULL; \ - ITTYPE start = ITSTART(node), last = ITLAST(node); \ - ITSTRUCT *parent; \ - \ - while (*link) { \ - rb_parent = *link; \ - parent = rb_entry(rb_parent, ITSTRUCT, ITRB); \ - if (parent->ITSUBTREE < last) \ - parent->ITSUBTREE = last; \ - if (start < ITSTART(parent)) \ - link = &parent->ITRB.rb_left; \ - else \ - link = &parent->ITRB.rb_right; \ - } \ - \ - node->ITSUBTREE = last; \ - rb_link_node(&node->ITRB, rb_parent, link); \ - rb_insert_augmented(&node->ITRB, root, &ITPREFIX ## _augment); \ -} \ - \ -ITSTATIC void ITPREFIX ## _remove(ITSTRUCT *node, struct rb_root *root) \ -{ \ - rb_erase_augmented(&node->ITRB, root, &ITPREFIX ## _augment); \ -} \ - \ -/* \ - * Iterate over intervals intersecting [start;last] \ - * \ - * Note that a node's interval intersects [start;last] iff: \ - * Cond1: ITSTART(node) <= last \ - * and \ - * Cond2: start <= ITLAST(node) \ - */ \ - \ -static ITSTRUCT * \ -ITPREFIX ## _subtree_search(ITSTRUCT *node, ITTYPE start, ITTYPE last) \ -{ \ - while (true) { \ - /* \ - * Loop invariant: start <= node->ITSUBTREE \ - * (Cond2 is satisfied by one of the subtree nodes) \ - */ \ - if (node->ITRB.rb_left) { \ - ITSTRUCT *left = rb_entry(node->ITRB.rb_left, \ - ITSTRUCT, ITRB); \ - if (start <= left->ITSUBTREE) { \ - /* \ - * Some nodes in left subtree satisfy Cond2. \ - * Iterate to find the leftmost such node N. \ - * If it also satisfies Cond1, that's the \ - * match we are looking for. Otherwise, there \ - * is no matching interval as nodes to the \ - * right of N can't satisfy Cond1 either. \ - */ \ - node = left; \ - continue; \ - } \ - } \ - if (ITSTART(node) <= last) { /* Cond1 */ \ - if (start <= ITLAST(node)) /* Cond2 */ \ - return node; /* node is leftmost match */ \ - if (node->ITRB.rb_right) { \ - node = rb_entry(node->ITRB.rb_right, \ - ITSTRUCT, ITRB); \ - if (start <= node->ITSUBTREE) \ - continue; \ - } \ - } \ - return NULL; /* No match */ \ - } \ -} \ - \ -ITSTATIC ITSTRUCT * \ -ITPREFIX ## _iter_first(struct rb_root *root, ITTYPE start, ITTYPE last) \ -{ \ - ITSTRUCT *node; \ - \ - if (!root->rb_node) \ - return NULL; \ - node = rb_entry(root->rb_node, ITSTRUCT, ITRB); \ - if (node->ITSUBTREE < start) \ - return NULL; \ - return ITPREFIX ## _subtree_search(node, start, last); \ -} \ - \ -ITSTATIC ITSTRUCT * \ -ITPREFIX ## _iter_next(ITSTRUCT *node, ITTYPE start, ITTYPE last) \ -{ \ - struct rb_node *rb = node->ITRB.rb_right, *prev; \ - \ - while (true) { \ - /* \ - * Loop invariants: \ - * Cond1: ITSTART(node) <= last \ - * rb == node->ITRB.rb_right \ - * \ - * First, search right subtree if suitable \ - */ \ - if (rb) { \ - ITSTRUCT *right = rb_entry(rb, ITSTRUCT, ITRB); \ - if (start <= right->ITSUBTREE) \ - return ITPREFIX ## _subtree_search(right, \ - start, last); \ - } \ - \ - /* Move up the tree until we come from a node's left child */ \ - do { \ - rb = rb_parent(&node->ITRB); \ - if (!rb) \ - return NULL; \ - prev = &node->ITRB; \ - node = rb_entry(rb, ITSTRUCT, ITRB); \ - rb = node->ITRB.rb_right; \ - } while (prev == rb); \ - \ - /* Check if the node intersects [start;last] */ \ - if (last < ITSTART(node)) /* !Cond1 */ \ - return NULL; \ - else if (start <= ITLAST(node)) /* Cond2 */ \ - return node; \ - } \ -} +#define INTERVAL_TREE_DEFINE(ITSTRUCT, ITRB, ITTYPE, ITSUBTREE, ITSTART, \ + ITLAST, ITSTATIC, ITPREFIX) \ + \ + /* Callbacks for augmented rbtree insert and remove */ \ + \ + static inline ITTYPE ITPREFIX##_compute_subtree_last(ITSTRUCT *node) { \ + ITTYPE max = ITLAST(node), subtree_last; \ + if (node->ITRB.rb_left) { \ + subtree_last = rb_entry(node->ITRB.rb_left, ITSTRUCT, ITRB)->ITSUBTREE; \ + if (max < subtree_last) { max = subtree_last; } \ + } \ + if (node->ITRB.rb_right) { \ + subtree_last = rb_entry(node->ITRB.rb_right, ITSTRUCT, ITRB)->ITSUBTREE; \ + if (max < subtree_last) max = subtree_last; \ + } \ + return max; \ + } \ + \ + RB_DECLARE_CALLBACKS(static, ITPREFIX##_augment, ITSTRUCT, ITRB, ITTYPE, \ + ITSUBTREE, ITPREFIX##_compute_subtree_last) \ + \ + /* Insert / remove interval nodes from the tree */ \ + \ + ITSTATIC void ITPREFIX##_insert(ITSTRUCT *node, struct rb_root *root) { \ + struct rb_node **link = &root->rb_node, *rb_parent = NULL; \ + ITTYPE start = ITSTART(node), last = ITLAST(node); \ + ITSTRUCT *parent; \ + \ + while (*link) { \ + rb_parent = *link; \ + parent = rb_entry(rb_parent, ITSTRUCT, ITRB); \ + if (parent->ITSUBTREE < last) parent->ITSUBTREE = last; \ + if (start < ITSTART(parent)) \ + link = &parent->ITRB.rb_left; \ + else \ + link = &parent->ITRB.rb_right; \ + } \ + \ + node->ITSUBTREE = last; \ + rb_link_node(&node->ITRB, rb_parent, link); \ + rb_insert_augmented(&node->ITRB, root, &ITPREFIX##_augment); \ + } \ + \ + ITSTATIC void ITPREFIX##_remove(ITSTRUCT *node, struct rb_root *root) { \ + rb_erase_augmented(&node->ITRB, root, &ITPREFIX##_augment); \ + } \ + \ + /* \ + * Iterate over intervals intersecting [start;last] \ + * \ + * Note that a node's interval intersects [start;last] iff: \ + * Cond1: ITSTART(node) <= last \ + * and \ + * Cond2: start <= ITLAST(node) \ + */ \ + \ + static ITSTRUCT *ITPREFIX##_subtree_search(ITSTRUCT *node, ITTYPE start, \ + ITTYPE last) { \ + while (true) { \ + /* \ + * Loop invariant: start <= node->ITSUBTREE \ + * (Cond2 is satisfied by one of the subtree nodes) \ + */ \ + if (node->ITRB.rb_left) { \ + ITSTRUCT *left = rb_entry(node->ITRB.rb_left, ITSTRUCT, ITRB); \ + if (start <= left->ITSUBTREE) { \ + /* \ + * Some nodes in left subtree satisfy Cond2. \ + * Iterate to find the leftmost such node N. \ + * If it also satisfies Cond1, that's the \ + * match we are looking for. Otherwise, there \ + * is no matching interval as nodes to the \ + * right of N can't satisfy Cond1 either. \ + */ \ + node = left; \ + continue; \ + } \ + } \ + if (ITSTART(node) <= last) { /* Cond1 */ \ + if (start <= ITLAST(node)) /* Cond2 */ \ + return node; /* node is leftmost match */ \ + if (node->ITRB.rb_right) { \ + node = rb_entry(node->ITRB.rb_right, ITSTRUCT, ITRB); \ + if (start <= node->ITSUBTREE) continue; \ + } \ + } \ + return NULL; /* No match */ \ + } \ + } \ + \ + ITSTATIC ITSTRUCT *ITPREFIX##_iter_first(struct rb_root *root, ITTYPE start, \ + ITTYPE last) { \ + ITSTRUCT *node; \ + \ + if (!root->rb_node) return NULL; \ + node = rb_entry(root->rb_node, ITSTRUCT, ITRB); \ + if (node->ITSUBTREE < start) return NULL; \ + return ITPREFIX##_subtree_search(node, start, last); \ + } \ + \ + ITSTATIC ITSTRUCT *ITPREFIX##_iter_next(ITSTRUCT *node, ITTYPE start, \ + ITTYPE last) { \ + struct rb_node *rb = node->ITRB.rb_right, *prev; \ + \ + while (true) { \ + /* \ + * Loop invariants: \ + * Cond1: ITSTART(node) <= last \ + * rb == node->ITRB.rb_right \ + * \ + * First, search right subtree if suitable \ + */ \ + if (rb) { \ + ITSTRUCT *right = rb_entry(rb, ITSTRUCT, ITRB); \ + if (start <= right->ITSUBTREE) \ + return ITPREFIX##_subtree_search(right, start, last); \ + } \ + \ + /* Move up the tree until we come from a node's left child */ \ + do { \ + rb = rb_parent(&node->ITRB); \ + if (!rb) { return NULL; } \ + prev = &node->ITRB; \ + node = rb_entry(rb, ITSTRUCT, ITRB); \ + rb = node->ITRB.rb_right; \ + } while (prev == rb); \ + \ + /* Check if the node intersects [start;last] */ \ + if (last < ITSTART(node)) { /* !Cond1 */ \ + return NULL; \ + } else if (start <= ITLAST(node)) { /* Cond2 */ \ + return node; \ + } \ + } \ + } diff --git a/libafl_qemu/src/interval-tree/rbtree.h b/libafl_qemu/src/interval-tree/rbtree.h index da67ade044..bcb087f886 100644 --- a/libafl_qemu/src/interval-tree/rbtree.h +++ b/libafl_qemu/src/interval-tree/rbtree.h @@ -1,7 +1,7 @@ /* Red Black Trees (C) 1999 Andrea Arcangeli - + This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or @@ -26,42 +26,40 @@ See Documentation/rbtree.txt for documentation and samples. */ -#ifndef _LINUX_RBTREE_H -#define _LINUX_RBTREE_H +#ifndef _LINUX_RBTREE_H +#define _LINUX_RBTREE_H #include #include "compiler.h" struct rb_node { - unsigned long __rb_parent_color; - struct rb_node *rb_right; - struct rb_node *rb_left; + unsigned long __rb_parent_color; + struct rb_node *rb_right; + struct rb_node *rb_left; } __attribute__((aligned(sizeof(long)))); - /* The alignment might seem pointless, but allegedly CRIS needs it */ +/* The alignment might seem pointless, but allegedly CRIS needs it */ struct rb_root { - struct rb_node *rb_node; + struct rb_node *rb_node; }; +#define rb_parent(r) ((struct rb_node *)((r)->__rb_parent_color & ~3)) -#define rb_parent(r) ((struct rb_node *)((r)->__rb_parent_color & ~3)) +#define RB_ROOT \ + (struct rb_root) { \ + NULL, \ + } +#define rb_entry(ptr, type, member) container_of(ptr, type, member) -#define RB_ROOT (struct rb_root) { NULL, } -#define rb_entry(ptr, type, member) container_of(ptr, type, member) - -#define RB_EMPTY_ROOT(root) ((root)->rb_node == NULL) +#define RB_EMPTY_ROOT(root) ((root)->rb_node == NULL) /* 'empty' nodes are nodes that are known not to be inserted in an rbtree */ -#define RB_EMPTY_NODE(node) \ - ((node)->__rb_parent_color == (unsigned long)(node)) -#define RB_CLEAR_NODE(node) \ - ((node)->__rb_parent_color = (unsigned long)(node)) - +#define RB_EMPTY_NODE(node) ((node)->__rb_parent_color == (unsigned long)(node)) +#define RB_CLEAR_NODE(node) ((node)->__rb_parent_color = (unsigned long)(node)) extern void rb_insert_color(struct rb_node *, struct rb_root *); extern void rb_erase(struct rb_node *, struct rb_root *); - /* Find logical next and previous nodes in a tree */ extern struct rb_node *rb_next(const struct rb_node *); extern struct rb_node *rb_prev(const struct rb_node *); @@ -73,22 +71,22 @@ extern struct rb_node *rb_first_postorder(const struct rb_root *); extern struct rb_node *rb_next_postorder(const struct rb_node *); /* Fast replacement of a single node without remove/rebalance/add/rebalance */ -extern void rb_replace_node(struct rb_node *victim, struct rb_node *new, - struct rb_root *root); +extern void rb_replace_node(struct rb_node *victim, struct rb_node *new, + struct rb_root *root); -static inline void rb_link_node(struct rb_node * node, struct rb_node * parent, - struct rb_node ** rb_link) -{ - node->__rb_parent_color = (unsigned long)parent; - node->rb_left = node->rb_right = NULL; +static inline void rb_link_node(struct rb_node *node, struct rb_node *parent, + struct rb_node **rb_link) { + node->__rb_parent_color = (unsigned long)parent; + node->rb_left = node->rb_right = NULL; - *rb_link = node; + *rb_link = node; } -#define rb_entry_safe(ptr, type, member) \ - ({ typeof(ptr) ____ptr = (ptr); \ - ____ptr ? rb_entry(____ptr, type, member) : NULL; \ - }) +#define rb_entry_safe(ptr, type, member) \ + ({ \ + typeof(ptr) ____ptr = (ptr); \ + ____ptr ? rb_entry(____ptr, type, member) : NULL; \ + }) /** * rbtree_postorder_for_each_entry_safe - iterate over rb_root in post order of @@ -99,10 +97,13 @@ static inline void rb_link_node(struct rb_node * node, struct rb_node * parent, * @root: 'rb_root *' of the rbtree. * @field: the name of the rb_node field within 'type'. */ -#define rbtree_postorder_for_each_entry_safe(pos, n, root, field) \ - for (pos = rb_entry_safe(rb_first_postorder(root), typeof(*pos), field); \ - pos && ({ n = rb_entry_safe(rb_next_postorder(&pos->field), \ - typeof(*pos), field); 1; }); \ - pos = n) +#define rbtree_postorder_for_each_entry_safe(pos, n, root, field) \ + for (pos = rb_entry_safe(rb_first_postorder(root), typeof(*pos), field); \ + pos && ({ \ + n = rb_entry_safe(rb_next_postorder(&pos->field), typeof(*pos), \ + field); \ + 1; \ + }); \ + pos = n) -#endif /* _LINUX_RBTREE_H */ +#endif /* _LINUX_RBTREE_H */ diff --git a/libafl_qemu/src/interval-tree/rbtree.inl b/libafl_qemu/src/interval-tree/rbtree.inl index a5b0d313df..bfb49bcdbd 100644 --- a/libafl_qemu/src/interval-tree/rbtree.inl +++ b/libafl_qemu/src/interval-tree/rbtree.inl @@ -416,10 +416,12 @@ struct rb_node *rb_first(const struct rb_root *root) struct rb_node *n; n = root->rb_node; - if (!n) + if (!n) { return NULL; - while (n->rb_left) + } + while (n->rb_left) { n = n->rb_left; + } return n; } @@ -428,10 +430,12 @@ struct rb_node *rb_last(const struct rb_root *root) struct rb_node *n; n = root->rb_node; - if (!n) + if (!n) { return NULL; - while (n->rb_right) + } + while (n->rb_right) { n = n->rb_right; + } return n; } @@ -448,8 +452,9 @@ struct rb_node *rb_next(const struct rb_node *node) */ if (node->rb_right) { node = node->rb_right; - while (node->rb_left) + while (node->rb_left) { node=node->rb_left; + } return (struct rb_node *)node; } @@ -460,8 +465,9 @@ struct rb_node *rb_next(const struct rb_node *node) * parent, keep going up. First time it's a left-hand child of its * parent, said parent is our 'next' node. */ - while ((parent = rb_parent(node)) && node == parent->rb_right) + while ((parent = rb_parent(node)) && node == parent->rb_right) { node = parent; + } return parent; } @@ -479,8 +485,9 @@ struct rb_node *rb_prev(const struct rb_node *node) */ if (node->rb_left) { node = node->rb_left; - while (node->rb_right) + while (node->rb_right) { node=node->rb_right; + } return (struct rb_node *)node; } @@ -488,8 +495,9 @@ struct rb_node *rb_prev(const struct rb_node *node) * No left-hand children. Go up till we find an ancestor which * is a right-hand child of its parent. */ - while ((parent = rb_parent(node)) && node == parent->rb_left) + while ((parent = rb_parent(node)) && node == parent->rb_left) { node = parent; + } return parent; } @@ -501,10 +509,12 @@ void rb_replace_node(struct rb_node *victim, struct rb_node *new, /* Set the surrounding nodes to point to the replacement */ __rb_change_child(victim, new, parent, root); - if (victim->rb_left) + if (victim->rb_left) { rb_set_parent(victim->rb_left, new); - if (victim->rb_right) + } + if (victim->rb_right) { rb_set_parent(victim->rb_right, new); + } /* Copy the pointers/colour from the victim to the replacement */ *new = *victim; @@ -513,20 +523,24 @@ void rb_replace_node(struct rb_node *victim, struct rb_node *new, static struct rb_node *rb_left_deepest_node(const struct rb_node *node) { for (;;) { - if (node->rb_left) + if (node->rb_left) { node = node->rb_left; - else if (node->rb_right) + } + else if (node->rb_right) { node = node->rb_right; - else + } + else { return (struct rb_node *)node; + } } } struct rb_node *rb_next_postorder(const struct rb_node *node) { const struct rb_node *parent; - if (!node) + if (!node) { return NULL; + } parent = rb_parent(node); /* If we're sitting on node, we've already seen our children */ @@ -542,8 +556,9 @@ struct rb_node *rb_next_postorder(const struct rb_node *node) struct rb_node *rb_first_postorder(const struct rb_root *root) { - if (!root->rb_node) + if (!root->rb_node) { return NULL; + } return rb_left_deepest_node(root->rb_node); } diff --git a/libafl_qemu/src/interval-tree/rbtree_augmented.h b/libafl_qemu/src/interval-tree/rbtree_augmented.h index 311abb6cde..6762e7b459 100644 --- a/libafl_qemu/src/interval-tree/rbtree_augmented.h +++ b/libafl_qemu/src/interval-tree/rbtree_augmented.h @@ -39,13 +39,14 @@ */ struct rb_augment_callbacks { - void (*propagate)(struct rb_node *node, struct rb_node *stop); - void (*copy)(struct rb_node *old, struct rb_node *new); - void (*rotate)(struct rb_node *old, struct rb_node *new); + void (*propagate)(struct rb_node *node, struct rb_node *stop); + void (*copy)(struct rb_node *old, struct rb_node *new); + void (*rotate)(struct rb_node *old, struct rb_node *new); }; extern void __rb_insert_augmented(struct rb_node *node, struct rb_root *root, - void (*augment_rotate)(struct rb_node *old, struct rb_node *new)); + void (*augment_rotate)(struct rb_node *old, + struct rb_node *new)); /* * Fixup the rbtree and update the augmented information when rebalancing. * @@ -56,190 +57,180 @@ extern void __rb_insert_augmented(struct rb_node *node, struct rb_root *root, * a user provided function to update the augmented information on the * affected subtrees. */ -static inline void -rb_insert_augmented(struct rb_node *node, struct rb_root *root, - const struct rb_augment_callbacks *augment) -{ - __rb_insert_augmented(node, root, augment->rotate); +static inline void rb_insert_augmented( + struct rb_node *node, struct rb_root *root, + const struct rb_augment_callbacks *augment) { + __rb_insert_augmented(node, root, augment->rotate); } -#define RB_DECLARE_CALLBACKS(rbstatic, rbname, rbstruct, rbfield, \ - rbtype, rbaugmented, rbcompute) \ -static inline void \ -rbname ## _propagate(struct rb_node *rb, struct rb_node *stop) \ -{ \ - while (rb != stop) { \ - rbstruct *node = rb_entry(rb, rbstruct, rbfield); \ - rbtype augmented = rbcompute(node); \ - if (node->rbaugmented == augmented) \ - break; \ - node->rbaugmented = augmented; \ - rb = rb_parent(&node->rbfield); \ - } \ -} \ -static inline void \ -rbname ## _copy(struct rb_node *rb_old, struct rb_node *rb_new) \ -{ \ - rbstruct *old = rb_entry(rb_old, rbstruct, rbfield); \ - rbstruct *new = rb_entry(rb_new, rbstruct, rbfield); \ - new->rbaugmented = old->rbaugmented; \ -} \ -static void \ -rbname ## _rotate(struct rb_node *rb_old, struct rb_node *rb_new) \ -{ \ - rbstruct *old = rb_entry(rb_old, rbstruct, rbfield); \ - rbstruct *new = rb_entry(rb_new, rbstruct, rbfield); \ - new->rbaugmented = old->rbaugmented; \ - old->rbaugmented = rbcompute(old); \ -} \ -rbstatic const struct rb_augment_callbacks rbname = { \ - rbname ## _propagate, rbname ## _copy, rbname ## _rotate \ -}; +#define RB_DECLARE_CALLBACKS(rbstatic, rbname, rbstruct, rbfield, rbtype, \ + rbaugmented, rbcompute) \ + static inline void rbname##_propagate(struct rb_node *rb, \ + struct rb_node *stop) { \ + while (rb != stop) { \ + rbstruct *node = rb_entry(rb, rbstruct, rbfield); \ + rbtype augmented = rbcompute(node); \ + if (node->rbaugmented == augmented) { break; } \ + node->rbaugmented = augmented; \ + rb = rb_parent(&node->rbfield); \ + } \ + } \ + static inline void rbname##_copy(struct rb_node *rb_old, \ + struct rb_node *rb_new) { \ + rbstruct *old = rb_entry(rb_old, rbstruct, rbfield); \ + rbstruct *new = rb_entry(rb_new, rbstruct, rbfield); \ + new->rbaugmented = old->rbaugmented; \ + } \ + static void rbname##_rotate(struct rb_node *rb_old, \ + struct rb_node *rb_new) { \ + rbstruct *old = rb_entry(rb_old, rbstruct, rbfield); \ + rbstruct *new = rb_entry(rb_new, rbstruct, rbfield); \ + new->rbaugmented = old->rbaugmented; \ + old->rbaugmented = rbcompute(old); \ + } \ + rbstatic const struct rb_augment_callbacks rbname = { \ + rbname##_propagate, rbname##_copy, rbname##_rotate}; +#define RB_RED 0 +#define RB_BLACK 1 -#define RB_RED 0 -#define RB_BLACK 1 +#define __rb_parent(pc) ((struct rb_node *)(pc & ~3)) -#define __rb_parent(pc) ((struct rb_node *)(pc & ~3)) +#define __rb_color(pc) ((pc)&1) +#define __rb_is_black(pc) __rb_color(pc) +#define __rb_is_red(pc) (!__rb_color(pc)) +#define rb_color(rb) __rb_color((rb)->__rb_parent_color) +#define rb_is_red(rb) __rb_is_red((rb)->__rb_parent_color) +#define rb_is_black(rb) __rb_is_black((rb)->__rb_parent_color) -#define __rb_color(pc) ((pc) & 1) -#define __rb_is_black(pc) __rb_color(pc) -#define __rb_is_red(pc) (!__rb_color(pc)) -#define rb_color(rb) __rb_color((rb)->__rb_parent_color) -#define rb_is_red(rb) __rb_is_red((rb)->__rb_parent_color) -#define rb_is_black(rb) __rb_is_black((rb)->__rb_parent_color) - -static inline void rb_set_parent(struct rb_node *rb, struct rb_node *p) -{ - rb->__rb_parent_color = rb_color(rb) | (unsigned long)p; +static inline void rb_set_parent(struct rb_node *rb, struct rb_node *p) { + rb->__rb_parent_color = rb_color(rb) | (unsigned long)p; } -static inline void rb_set_parent_color(struct rb_node *rb, - struct rb_node *p, int color) -{ - rb->__rb_parent_color = (unsigned long)p | color; +static inline void rb_set_parent_color(struct rb_node *rb, struct rb_node *p, + int color) { + rb->__rb_parent_color = (unsigned long)p | color; } -static inline void -__rb_change_child(struct rb_node *old, struct rb_node *new, - struct rb_node *parent, struct rb_root *root) -{ - if (parent) { - if (parent->rb_left == old) - parent->rb_left = new; - else - parent->rb_right = new; - } else - root->rb_node = new; +static inline void __rb_change_child(struct rb_node *old, struct rb_node *new, + struct rb_node *parent, + struct rb_root *root) { + if (parent) { + if (parent->rb_left == old) { + parent->rb_left = new; + } else { + parent->rb_right = new; + } + } else { + root->rb_node = new; + } } extern void __rb_erase_color(struct rb_node *parent, struct rb_root *root, - void (*augment_rotate)(struct rb_node *old, struct rb_node *new)); + void (*augment_rotate)(struct rb_node *old, + struct rb_node *new)); -static inline struct rb_node * -__rb_erase_augmented(struct rb_node *node, struct rb_root *root, - const struct rb_augment_callbacks *augment) -{ - struct rb_node *child = node->rb_right, *tmp = node->rb_left; - struct rb_node *parent, *rebalance; - unsigned long pc; +static inline struct rb_node *__rb_erase_augmented( + struct rb_node *node, struct rb_root *root, + const struct rb_augment_callbacks *augment) { + struct rb_node *child = node->rb_right, *tmp = node->rb_left; + struct rb_node *parent, *rebalance; + unsigned long pc; - if (!tmp) { - /* - * Case 1: node to erase has no more than 1 child (easy!) - * - * Note that if there is one child it must be red due to 5) - * and node must be black due to 4). We adjust colors locally - * so as to bypass __rb_erase_color() later on. - */ - pc = node->__rb_parent_color; - parent = __rb_parent(pc); - __rb_change_child(node, child, parent, root); - if (child) { - child->__rb_parent_color = pc; - rebalance = NULL; - } else - rebalance = __rb_is_black(pc) ? parent : NULL; - tmp = parent; - } else if (!child) { - /* Still case 1, but this time the child is node->rb_left */ - tmp->__rb_parent_color = pc = node->__rb_parent_color; - parent = __rb_parent(pc); - __rb_change_child(node, tmp, parent, root); - rebalance = NULL; - tmp = parent; - } else { - struct rb_node *successor = child, *child2; - tmp = child->rb_left; - if (!tmp) { - /* - * Case 2: node's successor is its right child - * - * (n) (s) - * / \ / \ - * (x) (s) -> (x) (c) - * \ - * (c) - */ - parent = successor; - child2 = successor->rb_right; - augment->copy(node, successor); - } else { - /* - * Case 3: node's successor is leftmost under - * node's right child subtree - * - * (n) (s) - * / \ / \ - * (x) (y) -> (x) (y) - * / / - * (p) (p) - * / / - * (s) (c) - * \ - * (c) - */ - do { - parent = successor; - successor = tmp; - tmp = tmp->rb_left; - } while (tmp); - parent->rb_left = child2 = successor->rb_right; - successor->rb_right = child; - rb_set_parent(child, successor); - augment->copy(node, successor); - augment->propagate(parent, successor); - } + if (!tmp) { + /* + * Case 1: node to erase has no more than 1 child (easy!) + * + * Note that if there is one child it must be red due to 5) + * and node must be black due to 4). We adjust colors locally + * so as to bypass __rb_erase_color() later on. + */ + pc = node->__rb_parent_color; + parent = __rb_parent(pc); + __rb_change_child(node, child, parent, root); + if (child) { + child->__rb_parent_color = pc; + rebalance = NULL; + } else + rebalance = __rb_is_black(pc) ? parent : NULL; + tmp = parent; + } else if (!child) { + /* Still case 1, but this time the child is node->rb_left */ + tmp->__rb_parent_color = pc = node->__rb_parent_color; + parent = __rb_parent(pc); + __rb_change_child(node, tmp, parent, root); + rebalance = NULL; + tmp = parent; + } else { + struct rb_node *successor = child, *child2; + tmp = child->rb_left; + if (!tmp) { + /* + * Case 2: node's successor is its right child + * + * (n) (s) + * / \ / \ + * (x) (s) -> (x) (c) + * \ + * (c) + */ + parent = successor; + child2 = successor->rb_right; + augment->copy(node, successor); + } else { + /* + * Case 3: node's successor is leftmost under + * node's right child subtree + * + * (n) (s) + * / \ / \ + * (x) (y) -> (x) (y) + * / / + * (p) (p) + * / / + * (s) (c) + * \ + * (c) + */ + do { + parent = successor; + successor = tmp; + tmp = tmp->rb_left; + } while (tmp); + parent->rb_left = child2 = successor->rb_right; + successor->rb_right = child; + rb_set_parent(child, successor); + augment->copy(node, successor); + augment->propagate(parent, successor); + } - successor->rb_left = tmp = node->rb_left; - rb_set_parent(tmp, successor); + successor->rb_left = tmp = node->rb_left; + rb_set_parent(tmp, successor); - pc = node->__rb_parent_color; - tmp = __rb_parent(pc); - __rb_change_child(node, successor, tmp, root); - if (child2) { - successor->__rb_parent_color = pc; - rb_set_parent_color(child2, parent, RB_BLACK); - rebalance = NULL; - } else { - unsigned long pc2 = successor->__rb_parent_color; - successor->__rb_parent_color = pc; - rebalance = __rb_is_black(pc2) ? parent : NULL; - } - tmp = successor; - } + pc = node->__rb_parent_color; + tmp = __rb_parent(pc); + __rb_change_child(node, successor, tmp, root); + if (child2) { + successor->__rb_parent_color = pc; + rb_set_parent_color(child2, parent, RB_BLACK); + rebalance = NULL; + } else { + unsigned long pc2 = successor->__rb_parent_color; + successor->__rb_parent_color = pc; + rebalance = __rb_is_black(pc2) ? parent : NULL; + } + tmp = successor; + } - augment->propagate(tmp, NULL); - return rebalance; + augment->propagate(tmp, NULL); + return rebalance; } -static inline void -rb_erase_augmented(struct rb_node *node, struct rb_root *root, - const struct rb_augment_callbacks *augment) -{ - struct rb_node *rebalance = __rb_erase_augmented(node, root, augment); - if (rebalance) - __rb_erase_color(rebalance, root, augment->rotate); +static inline void rb_erase_augmented( + struct rb_node *node, struct rb_root *root, + const struct rb_augment_callbacks *augment) { + struct rb_node *rebalance = __rb_erase_augmented(node, root, augment); + if (rebalance) { __rb_erase_color(rebalance, root, augment->rotate); } } -#endif /* _LINUX_RBTREE_AUGMENTED_H */ +#endif /* _LINUX_RBTREE_AUGMENTED_H */ diff --git a/libafl_targets/src/cmplog.c b/libafl_targets/src/cmplog.c index fca6ab7fa7..eba3ad703b 100644 --- a/libafl_targets/src/cmplog.c +++ b/libafl_targets/src/cmplog.c @@ -52,25 +52,26 @@ void __libafl_targets_cmplog_instructions(uintptr_t k, uint8_t shape, uint64_t a // to avoid to call it on .text addresses static long area_is_valid(void *ptr, size_t len) { - if (!ptr || __asan_region_is_poisoned(ptr, len)) return 0; + if (!ptr || __asan_region_is_poisoned(ptr, len)) {return 0;} long valid_len; #if defined(_WIN32) - if (IsBadReadPtr(ptr, len)) return 0; + if (IsBadReadPtr(ptr, len)) {return 0;} valid_len = (long)len; #elif defined(__unix__) || (defined(__APPLE__) && defined(__MACH__)) if (!dymmy_initialized) { if ((dummy_fd[1] = open("/dev/null", O_WRONLY)) < 0) { - if (pipe(dummy_fd) < 0) + if (pipe(dummy_fd) < 0) { dummy_fd[1] = 1; + } } dymmy_initialized = 1; } valid_len = syscall(SYS_write, dummy_fd[1], ptr, len); - if (valid_len <= 0 || valid_len > (long)len) return 0; + if (valid_len <= 0 || valid_len > (long)len) {return 0;} #endif // even if the write succeed this can be a false positive if we cross @@ -98,12 +99,13 @@ static long area_is_valid(void *ptr, size_t len) { void __libafl_targets_cmplog_routines(uintptr_t k, uint8_t *ptr1, uint8_t *ptr2) { - if (!libafl_cmplog_enabled) return; + if (!libafl_cmplog_enabled) { return; } int l1, l2; if ((l1 = area_is_valid(ptr1, CMPLOG_RTN_LEN)) <= 0 || - (l2 = area_is_valid(ptr2, CMPLOG_RTN_LEN)) <= 0) + (l2 = area_is_valid(ptr2, CMPLOG_RTN_LEN)) <= 0) { return; + } int len = MIN(l1, l2); uint32_t hits; @@ -115,8 +117,9 @@ void __libafl_targets_cmplog_routines(uintptr_t k, uint8_t *ptr1, uint8_t *ptr2) hits = 0; } else { hits = libafl_cmplog_map_ptr->headers[k].hits++; - if (libafl_cmplog_map_ptr->headers[k].shape < len) + if (libafl_cmplog_map_ptr->headers[k].shape < len) { libafl_cmplog_map_ptr->headers[k].shape = len; + } } hits &= CMPLOG_MAP_RTN_H - 1; @@ -154,8 +157,8 @@ static uint8_t *get_gcc_stdstring(uint8_t *string) { // IcEEE7compareEmmPKcm static uint8_t *get_llvm_stdstring(uint8_t *string) { - // length is in: if ((string[0] & 1) == 0) uint8_t len = (string[0] >> 1); - // or: if (string[0] & 1) uint32_t *len = (uint32_t *) (string + 8); + // length is in: if ((string[0] & 1) == 0) {uint8_t len = (string[0] >> 1);} + // or: if (string[0] & 1) {uint32_t *len = (uint32_t *) (string + 8);} if (string[0] & 1) { // in memory uint8_t **ptr = (uint8_t **)(string + 16); @@ -168,9 +171,10 @@ static uint8_t *get_llvm_stdstring(uint8_t *string) { void __cmplog_rtn_gcc_stdstring_cstring(uint8_t *stdstring, uint8_t *cstring) { - if (!libafl_cmplog_enabled) return; - if (area_is_valid(stdstring, 32) <= 0) + if (!libafl_cmplog_enabled) {return;} + if (area_is_valid(stdstring, 32) <= 0) { return; + } __cmplog_rtn_hook(get_gcc_stdstring(stdstring), cstring); @@ -178,9 +182,10 @@ void __cmplog_rtn_gcc_stdstring_cstring(uint8_t *stdstring, uint8_t *cstring) { void __cmplog_rtn_gcc_stdstring_stdstring(uint8_t *stdstring1, uint8_t *stdstring2) { - if (!libafl_cmplog_enabled) return; - if (area_is_valid(stdstring1, 32) <= 0 || area_is_valid(stdstring2, 32) <= 0) + if (!libafl_cmplog_enabled) {return;} + if (area_is_valid(stdstring1, 32) <= 0 || area_is_valid(stdstring2, 32) <= 0) { return; + } __cmplog_rtn_hook(get_gcc_stdstring(stdstring1), get_gcc_stdstring(stdstring2)); @@ -189,9 +194,10 @@ void __cmplog_rtn_gcc_stdstring_stdstring(uint8_t *stdstring1, uint8_t *stdstrin void __cmplog_rtn_llvm_stdstring_cstring(uint8_t *stdstring, uint8_t *cstring) { - if (!libafl_cmplog_enabled) return; - if (area_is_valid(stdstring, 32) <= 0) + if (!libafl_cmplog_enabled) {return;} + if (area_is_valid(stdstring, 32) <= 0){ return; + } __cmplog_rtn_hook(get_llvm_stdstring(stdstring), cstring); @@ -199,9 +205,10 @@ void __cmplog_rtn_llvm_stdstring_cstring(uint8_t *stdstring, uint8_t *cstring) { void __cmplog_rtn_llvm_stdstring_stdstring(uint8_t *stdstring1, uint8_t *stdstring2) { - if (!libafl_cmplog_enabled) return; - if (area_is_valid(stdstring1, 32) <= 0 || area_is_valid(stdstring2, 32) <= 0) + if (!libafl_cmplog_enabled) {return;} + if (area_is_valid(stdstring1, 32) <= 0 || area_is_valid(stdstring2, 32) <= 0) { return; + } __cmplog_rtn_hook(get_llvm_stdstring(stdstring1), get_llvm_stdstring(stdstring2)); diff --git a/libafl_targets/src/cmplog.h b/libafl_targets/src/cmplog.h index 5dbdca185c..cfc7935bc7 100644 --- a/libafl_targets/src/cmplog.h +++ b/libafl_targets/src/cmplog.h @@ -4,56 +4,59 @@ #include "common.h" #ifndef CMPLOG_MAP_W -#define CMPLOG_MAP_W 65536 + #define CMPLOG_MAP_W 65536 #endif #ifndef CMPLOG_MAP_H -#define CMPLOG_MAP_H 32 + #define CMPLOG_MAP_H 32 #endif #define CMPLOG_RTN_LEN 32 -#define CMPLOG_MAP_RTN_H ((CMPLOG_MAP_H * sizeof(CmpLogInstruction)) / sizeof(CmpLogRoutine)) +#define CMPLOG_MAP_RTN_H \ + ((CMPLOG_MAP_H * sizeof(CmpLogInstruction)) / sizeof(CmpLogRoutine)) #define CMPLOG_KIND_INS 0 #define CMPLOG_KIND_RTN 1 typedef struct CmpLogHeader { - uint16_t hits; - uint8_t shape; - uint8_t kind; + uint16_t hits; + uint8_t shape; + uint8_t kind; } CmpLogHeader; typedef struct CmpLogInstruction { - uint64_t v0; - uint64_t v1; + uint64_t v0; + uint64_t v1; } CmpLogInstruction; typedef struct CmpLogRoutine { - uint8_t v0[CMPLOG_RTN_LEN]; - uint8_t v1[CMPLOG_RTN_LEN]; + uint8_t v0[CMPLOG_RTN_LEN]; + uint8_t v1[CMPLOG_RTN_LEN]; } CmpLogRoutine; typedef struct CmpLogMap { CmpLogHeader headers[CMPLOG_MAP_W]; union { - CmpLogInstruction operands[CMPLOG_MAP_W][CMPLOG_MAP_H]; - CmpLogRoutine routines[CMPLOG_MAP_W][CMPLOG_MAP_RTN_H]; + CmpLogInstruction operands[CMPLOG_MAP_W][CMPLOG_MAP_H]; + CmpLogRoutine routines[CMPLOG_MAP_W][CMPLOG_MAP_RTN_H]; } vals; } CmpLogMap; -extern CmpLogMap libafl_cmplog_map; -extern CmpLogMap* libafl_cmplog_map_ptr; +extern CmpLogMap libafl_cmplog_map; +extern CmpLogMap *libafl_cmplog_map_ptr; extern uint8_t libafl_cmplog_enabled; -void __libafl_targets_cmplog_instructions(uintptr_t k, uint8_t shape, uint64_t arg1, uint64_t arg2); +void __libafl_targets_cmplog_instructions(uintptr_t k, uint8_t shape, + uint64_t arg1, uint64_t arg2); -void __libafl_targets_cmplog_routines(uintptr_t k, uint8_t *ptr1, uint8_t *ptr2); +void __libafl_targets_cmplog_routines(uintptr_t k, uint8_t *ptr1, + uint8_t *ptr2); -static inline void __libafl_targets_cmplog(uintptr_t k, uint8_t shape, uint64_t arg1, uint64_t arg2) { +static inline void __libafl_targets_cmplog(uintptr_t k, uint8_t shape, + uint64_t arg1, uint64_t arg2) { + if (!libafl_cmplog_enabled) { return; } - if (!libafl_cmplog_enabled) return; - uint16_t hits; if (libafl_cmplog_map_ptr->headers[k].kind != CMPLOG_KIND_INS) { libafl_cmplog_map_ptr->headers[k].kind = CMPLOG_KIND_INS; @@ -70,7 +73,6 @@ static inline void __libafl_targets_cmplog(uintptr_t k, uint8_t shape, uint64_t hits &= CMPLOG_MAP_H - 1; libafl_cmplog_map_ptr->vals.operands[k][hits].v0 = arg1; libafl_cmplog_map_ptr->vals.operands[k][hits].v1 = arg2; - } #endif diff --git a/libafl_targets/src/common.h b/libafl_targets/src/common.h index 46e8b5ef57..a13b47f0a9 100644 --- a/libafl_targets/src/common.h +++ b/libafl_targets/src/common.h @@ -6,21 +6,20 @@ #define true 1 #define false 0 -#define STATIC_ASSERT(pred) switch(0){case 0:case pred:;} +#define STATIC_ASSERT(pred) \ + switch (0) { \ + case 0: \ + case pred:; \ + } // From https://stackoverflow.com/a/18298965 #if __STDC_VERSION__ >= 201112 && !defined __STDC_NO_THREADS__ #define THREAD_LOCAL _Thread_local -#elif defined _WIN32 && ( \ - defined _MSC_VER || \ - defined __ICL || \ - defined __DMC__ || \ - defined __BORLANDC__ ) - #define THREAD_LOCAL __declspec(thread) +#elif defined _WIN32 && (defined _MSC_VER || defined __ICL || \ + defined __DMC__ || defined __BORLANDC__) + #define THREAD_LOCAL __declspec(thread) /* note that ICC (linux) and Clang are covered by __GNUC__ */ -#elif defined __GNUC__ || \ - defined __SUNPRO_C || \ - defined __xlC__ +#elif defined __GNUC__ || defined __SUNPRO_C || defined __xlC__ #define THREAD_LOCAL __thread #endif @@ -40,119 +39,116 @@ #endif #ifdef _WIN32 - #define RETADDR (uintptr_t)_ReturnAddress() + #define RETADDR (uintptr_t) _ReturnAddress() #define EXPORT_FN __declspec(dllexport) #else - #define RETADDR (uintptr_t)__builtin_return_address(0) + #define RETADDR (uintptr_t) __builtin_return_address(0) #define EXPORT_FN #endif #ifdef __GNUC__ #define MAX(a, b) \ ({ \ - \ __typeof__(a) _a = (a); \ __typeof__(b) _b = (b); \ _a > _b ? _a : _b; \ - \ }) #define MIN(a, b) \ ({ \ - \ __typeof__(a) _a = (a); \ __typeof__(b) _b = (b); \ _a < _b ? _a : _b; \ - \ }) #define MEMCPY __builtin_memcpy #else - #define MAX(a, b) (((a) > (b)) ? (a) : (b)) + #define MAX(a, b) (((a) > (b)) ? (a) : (b)) #define MIN(a, b) (((a) < (b)) ? (a) : (b)) #define MEMCPY memcpy #endif #ifdef _WIN32 -// From Libfuzzer -// Intermediate macro to ensure the parameter is expanded before stringified. -#define STRINGIFY_(A) #A -#define STRINGIFY(A) STRINGIFY_(A) + // From Libfuzzer + // Intermediate macro to ensure the parameter is expanded before stringified. + #define STRINGIFY_(A) #A + #define STRINGIFY(A) STRINGIFY_(A) -#if _MSC_VER -// Copied from compiler-rt/lib/sanitizer_common/sanitizer_win_defs.h -#if defined(_M_IX86) || defined(__i386__) -#define WIN_SYM_PREFIX "_" -#else -#define WIN_SYM_PREFIX -#endif + #if _MSC_VER + // Copied from compiler-rt/lib/sanitizer_common/sanitizer_win_defs.h + #if defined(_M_IX86) || defined(__i386__) + #define WIN_SYM_PREFIX "_" + #else + #define WIN_SYM_PREFIX + #endif -// Declare external functions as having alternativenames, so that we can -// determine if they are not defined. -#define EXTERNAL_FUNC(Name, Default) \ - __pragma(comment(linker, "/alternatename:" WIN_SYM_PREFIX STRINGIFY( \ - Name) "=" WIN_SYM_PREFIX STRINGIFY(Default))) + // Declare external functions as having alternativenames, so that we can + // determine if they are not defined. + #define EXTERNAL_FUNC(Name, Default) \ + __pragma( \ + comment(linker, "/alternatename:" WIN_SYM_PREFIX STRINGIFY( \ + Name) "=" WIN_SYM_PREFIX STRINGIFY(Default))) -#define CHECK_WEAK_FN(Name) ((void*)Name != (void*)&Name##Def) + #define CHECK_WEAK_FN(Name) ((void *)Name != (void *)&Name##Def) -#define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - RETURN_TYPE NAME##Def FUNC_SIG; \ - EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG; \ - RETURN_TYPE NAME##Def FUNC_SIG \ + #define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + RETURN_TYPE NAME##Def FUNC_SIG; \ + EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG; \ + RETURN_TYPE NAME##Def FUNC_SIG -#define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - RETURN_TYPE (*NAME##Def) FUNC_SIG = NULL; \ - EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG -#else -// Declare external functions as weak to allow them to default to a specified -// function if not defined explicitly. We must use weak symbols because clang's -// support for alternatename is not 100%, see -// https://bugs.llvm.org/show_bug.cgi?id=40218 for more details. -#define EXTERNAL_FUNC(Name, Default) \ - __attribute__((weak, alias(STRINGIFY(Default)))) + #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + RETURN_TYPE(*NAME##Def) FUNC_SIG = NULL; \ + EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG + #else + // Declare external functions as weak to allow them to default to a + // specified function if not defined explicitly. We must use weak symbols + // because clang's support for alternatename is not 100%, see + // https://bugs.llvm.org/show_bug.cgi?id=40218 for more details. + #define EXTERNAL_FUNC(Name, Default) \ + __attribute__((weak, alias(STRINGIFY(Default)))) -#define CHECK_WEAK_FN(Name) (Name != NULL) + #define CHECK_WEAK_FN(Name) (Name != NULL) -#define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) + #define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) -#define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - RETURN_TYPE (*NAME##Def) FUNC_SIG = NULL; \ - EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG -#endif // _MSC_VER + #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + RETURN_TYPE(*NAME##Def) FUNC_SIG = NULL; \ + EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG + #endif // _MSC_VER #else -#if defined(__APPLE__) - // On Apple, weak_import and weak attrs behave differently to linux. + #if defined(__APPLE__) + // On Apple, weak_import and weak attrs behave differently to linux. - #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - __attribute__((weak, visibility("default"))) RETURN_TYPE NAME FUNC_SIG { \ - return (RETURN_TYPE) 0; \ - } + #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + __attribute__((weak, visibility("default"))) RETURN_TYPE NAME FUNC_SIG { \ + return (RETURN_TYPE)0; \ + } - #define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - __attribute__((weak, visibility("default"))) RETURN_TYPE NAME FUNC_SIG + #define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + __attribute__((weak, visibility("default"))) RETURN_TYPE NAME FUNC_SIG - // Weakly defined globals - #define EXT_VAR(NAME, TYPE) \ - TYPE __attribute__((weak, visibility("default"))) NAME + // Weakly defined globals + #define EXT_VAR(NAME, TYPE) \ + TYPE __attribute__((weak, visibility("default"))) NAME -#else + #else -#define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) + #define EXT_FUNC_IMPL(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) -// Declare these symbols as weak to allow them to be optionally defined. -#define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ - __attribute__((weak, visibility("default"))) RETURN_TYPE NAME FUNC_SIG + // Declare these symbols as weak to allow them to be optionally defined. + #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \ + __attribute__((weak, visibility("default"))) RETURN_TYPE NAME FUNC_SIG -// Weakly defined globals -#define EXT_VAR(NAME, TYPE) \ - TYPE __attribute__((weak, visibility("default"))) NAME - -#endif - -#define CHECK_WEAK_FN(Name) (Name != NULL) -#endif // _WIN32 + // Weakly defined globals + #define EXT_VAR(NAME, TYPE) \ + TYPE __attribute__((weak, visibility("default"))) NAME + + #endif + + #define CHECK_WEAK_FN(Name) (Name != NULL) +#endif // _WIN32 #endif diff --git a/libafl_targets/src/sancov_cmp.c b/libafl_targets/src/sancov_cmp.c index edee036092..a7def0a437 100644 --- a/libafl_targets/src/sancov_cmp.c +++ b/libafl_targets/src/sancov_cmp.c @@ -76,7 +76,7 @@ void __sanitizer_cov_trace_switch(uint64_t val, uint64_t *cases) { uintptr_t rt = RETADDR; - // if (!cases[1]) return; + // if (!cases[1]) {return;} for (uint64_t i = 0; i < cases[0]; i++) { diff --git a/libafl_targets/src/value_profile.h b/libafl_targets/src/value_profile.h index 8af9c7e152..ddb26ff533 100644 --- a/libafl_targets/src/value_profile.h +++ b/libafl_targets/src/value_profile.h @@ -4,7 +4,7 @@ #include "common.h" #ifndef CMP_MAP_SIZE -#define CMP_MAP_SIZE 65536 + #define CMP_MAP_SIZE 65536 #endif extern uint8_t libafl_cmp_map[CMP_MAP_SIZE]; @@ -15,28 +15,28 @@ extern uint8_t libafl_cmp_map[CMP_MAP_SIZE]; #define __builtin_popcountll __popcnt64 #endif -static void __libafl_targets_value_profile1(uintptr_t k, uint8_t arg1, uint8_t arg2) { - - libafl_cmp_map[k] = MAX(libafl_cmp_map[k], (__builtin_popcount(~(arg1 ^ arg2)))); - +static void __libafl_targets_value_profile1(uintptr_t k, uint8_t arg1, + uint8_t arg2) { + libafl_cmp_map[k] = + MAX(libafl_cmp_map[k], (__builtin_popcount(~(arg1 ^ arg2)))); } -static void __libafl_targets_value_profile2(uintptr_t k, uint16_t arg1, uint16_t arg2) { - - libafl_cmp_map[k] = MAX(libafl_cmp_map[k], (__builtin_popcount(~(arg1 ^ arg2)))); - +static void __libafl_targets_value_profile2(uintptr_t k, uint16_t arg1, + uint16_t arg2) { + libafl_cmp_map[k] = + MAX(libafl_cmp_map[k], (__builtin_popcount(~(arg1 ^ arg2)))); } -static void __libafl_targets_value_profile4(uintptr_t k, uint32_t arg1, uint32_t arg2) { - - libafl_cmp_map[k] = MAX(libafl_cmp_map[k], (__builtin_popcount(~(arg1 ^ arg2)))); - +static void __libafl_targets_value_profile4(uintptr_t k, uint32_t arg1, + uint32_t arg2) { + libafl_cmp_map[k] = + MAX(libafl_cmp_map[k], (__builtin_popcount(~(arg1 ^ arg2)))); } -static void __libafl_targets_value_profile8(uintptr_t k, uint64_t arg1, uint64_t arg2) { - - libafl_cmp_map[k] = MAX(libafl_cmp_map[k], (__builtin_popcountll(~(arg1 ^ arg2)))); - +static void __libafl_targets_value_profile8(uintptr_t k, uint64_t arg1, + uint64_t arg2) { + libafl_cmp_map[k] = + MAX(libafl_cmp_map[k], (__builtin_popcountll(~(arg1 ^ arg2)))); } #endif diff --git a/scripts/afl-persistent-config b/scripts/afl-persistent-config index 5a9f5e31f8..a62b48971d 100755 --- a/scripts/afl-persistent-config +++ b/scripts/afl-persistent-config @@ -5,7 +5,7 @@ test "$1" = "-h" -o "$1" = "-hh" && { echo 'afl-persistent-config' echo - echo $0 + echo "$0" echo echo afl-persistent-config has no command line options echo @@ -32,7 +32,7 @@ if [[ "$ANSWER" != "YES" ]]; then fi echo -PLATFORM=`uname -s` +PLATFORM=$(uname -s) # check that we're on Mac if [[ "$PLATFORM" = "Darwin" ]] ; then diff --git a/scripts/autofix.sh b/scripts/autofix.sh index eb987a6b28..5a622802f3 100755 --- a/scripts/autofix.sh +++ b/scripts/autofix.sh @@ -1,6 +1,6 @@ #!/bin/bash SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" -cd "$SCRIPT_DIR/.." +cd "$SCRIPT_DIR/.." || exit 1 # Clippy checks if [ "$1" != "--no-clean" ]; then diff --git a/scripts/clean_all.sh b/scripts/clean_all.sh index 595a6f64c2..d55514f593 100755 --- a/scripts/clean_all.sh +++ b/scripts/clean_all.sh @@ -1,7 +1,7 @@ #!/bin/bash SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" -cd "$SCRIPT_DIR/.." +cd "$SCRIPT_DIR/.." || exit 1 # TODO: This should be rewritten in rust, a Makefile, or some platform-independent language @@ -12,10 +12,11 @@ cargo clean fuzzers=$(find ./fuzzers -maxdepth 1 -type d) backtrace_fuzzers=$(find ./fuzzers/backtrace_baby_fuzzers -maxdepth 1 -type d) -for fuzzer in $(echo $fuzzers $backtrace_fuzzers); +# shellcheck disable=SC2116 +for fuzzer in $(echo "$fuzzers" "$backtrace_fuzzers"); do - pushd $fuzzer + pushd "$fuzzer" || exit 1 echo "[*] Running clean for $fuzzer" cargo clean - popd + popd || exit 1 done diff --git a/scripts/clippy.sh b/scripts/clippy.sh index 2593eed2ee..092600f3d0 100755 --- a/scripts/clippy.sh +++ b/scripts/clippy.sh @@ -1,6 +1,6 @@ #!/bin/bash SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" -cd "$SCRIPT_DIR/.." +cd "$SCRIPT_DIR/.." || exit 1 # Clippy checks if [ "$1" != "--no-clean" ]; then diff --git a/scripts/fmt_all.sh b/scripts/fmt_all.sh index f765111bcd..4161b497a4 100755 --- a/scripts/fmt_all.sh +++ b/scripts/fmt_all.sh @@ -1,7 +1,7 @@ #!/bin/bash SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" -cd "$SCRIPT_DIR/.." +cd "$SCRIPT_DIR/.." || exit 1 # TODO: This should be rewritten in rust, a Makefile, or some platform-independent language @@ -9,13 +9,20 @@ echo "Welcome to the happy fmt script. :)" echo "[*] Running fmt for the main crates" cargo fmt +echo "[*] Formatting C(pp) files" +# shellcheck disable=SC2046 +clang-format-13 -i --style=file $(find . -type f \( -name '*.cpp' -o -iname '*.hpp' -o -name '*.cc' -o -name '*.cxx' -o -name '*.cc' -o -name '*.h' \) | grep -v '/target/' | grep -v 'libpng-1\.6\.37' | grep -v 'stb_image\.h' | grep -v 'dlmalloc\.c') + + + fuzzers=$(find ./fuzzers -maxdepth 1 -type d) backtrace_fuzzers=$(find ./fuzzers/backtrace_baby_fuzzers -maxdepth 1 -type d) -for fuzzer in $(echo $fuzzers $backtrace_fuzzers); +# shellcheck disable=SC2116 +for fuzzer in $(echo "$fuzzers" "$backtrace_fuzzers"); do - pushd $fuzzer + pushd "$fuzzer" || exit 1 echo "[*] Running fmt for $fuzzer" cargo fmt --all - popd + popd || exit 1 done diff --git a/scripts/publish.sh b/scripts/publish.sh index 5953862a1c..06ee4a784d 100755 --- a/scripts/publish.sh +++ b/scripts/publish.sh @@ -1,54 +1,54 @@ #!/bin/bash SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" -cd "$SCRIPT_DIR/.." +cd "$SCRIPT_DIR/.." || exit 1 set -e cd libafl_derive -cargo publish $@ -cd .. +cargo publish "$@" +cd .. || exit 1 sleep 20 cd libafl_cc -cargo publish $@ -cd .. +cargo publish "$@" +cd .. || exit 1 sleep 20 cd libafl -cargo publish $@ -cd .. +cargo publish "$@" +cd .. || exit 1 sleep 20 cd libafl_targets -cargo publish $@ -cd .. +cargo publish "$@" +cd .. || exit 1 sleep 20 cd libafl_frida -cargo publish $@ -cd .. +cargo publish "$@" +cd .. || exit 1 sleep 20 cd libafl_qemu -cargo publish $@ -cd .. +cargo publish "$@" +cd .. || exit 1 sleep 20 cd libafl_sugar -cargo publish $@ -cd .. +cargo publish "$@" +cd .. || exit 1 sleep 20 cd libafl_concolic/symcc_libafl -cargo publish $@ -cd ../.. +cargo publish "$@" +cd ../.. || exit 1 sleep 20 @@ -60,5 +60,5 @@ if git submodule status | grep "^-">/dev/null ; then \ fi cd libafl_concolic/symcc_runtime -cargo publish $@ -cd ../.. +cargo publish "$@" +cd ../.. || exit 1 diff --git a/scripts/shmem_limits_macos.sh b/scripts/shmem_limits_macos.sh index f8a3b67f8b..b36a01578e 100755 --- a/scripts/shmem_limits_macos.sh +++ b/scripts/shmem_limits_macos.sh @@ -1,5 +1,6 @@ #!/bin/sh +# shellcheck disable=SC2016 echo "Warning: this script is not a proper fix to do LLMP fuzzing." \ 'Instead, run `afl-persistent-config` with SIP disabled.' diff --git a/scripts/test_all_fuzzers.sh b/scripts/test_all_fuzzers.sh index 4c7fa2c0a3..c077e541ca 100755 --- a/scripts/test_all_fuzzers.sh +++ b/scripts/test_all_fuzzers.sh @@ -1,7 +1,7 @@ #!/bin/bash SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" -cd "$SCRIPT_DIR/.." +cd "$SCRIPT_DIR/.." || exit 1 # TODO: This should be rewritten in rust, a Makefile, or some platform-independent language @@ -25,11 +25,11 @@ done # record time of each fuzzer declare -A time_record || (echo "declare -A not avaliable, please update your bash version to 4";exit 1) - -for fuzzer in $(echo $fuzzers $backtrace_fuzzers); +# shellcheck disable=SC2116 +for fuzzer in $(echo "$fuzzers" "$backtrace_fuzzers"); do - cd $fuzzer - start=`date +%s` + cd "$fuzzer" || exit 1 + start=$(date +%s) # Clippy checks if [ "$1" != "--no-fmt" ]; then @@ -50,15 +50,15 @@ do cargo build || exit 1 echo "[+] Done building $fuzzer" fi - end=`date +%s` + end=$(date +%s) time_record[$fuzzer]=$((end-start)) # Save disk space cargo clean - cd $libafl + cd "$libafl" || exit 1 echo "" done # print time for each fuzzer -for key in ${!time_record[@]}; do - echo "dir:"$key" time:"${time_record[$key]}; +for key in "${!time_record[@]}"; do + echo "dir: $key, time: ${time_record[$key]}"; done