llvm-for-llvmta/test/CodeGen/AMDGPU/GlobalISel/legalize-mul.mir

649 lines
35 KiB
Plaintext
Raw Permalink Normal View History

2022-04-25 10:02:23 +02:00
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=tahiti -O0 -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX6 %s
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=fiji -O0 -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX8 %s
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -O0 -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX9 %s
---
name: test_mul_s32
body: |
bb.0:
liveins: $vgpr0, $vgpr1
; GFX6-LABEL: name: test_mul_s32
; GFX6: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX6: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY]], [[COPY1]]
; GFX6: $vgpr0 = COPY [[MUL]](s32)
; GFX8-LABEL: name: test_mul_s32
; GFX8: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX8: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY]], [[COPY1]]
; GFX8: $vgpr0 = COPY [[MUL]](s32)
; GFX9-LABEL: name: test_mul_s32
; GFX9: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX9: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY]], [[COPY1]]
; GFX9: $vgpr0 = COPY [[MUL]](s32)
%0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1
%2:_(s32) = G_MUL %0, %1
$vgpr0 = COPY %2
...
---
name: test_mul_v2s32
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX6-LABEL: name: test_mul_v2s32
; GFX6: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; GFX6: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; GFX6: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<2 x s32>)
; GFX6: [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](<2 x s32>)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV2]]
; GFX6: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV3]]
; GFX6: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[MUL]](s32), [[MUL1]](s32)
; GFX6: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
; GFX8-LABEL: name: test_mul_v2s32
; GFX8: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; GFX8: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; GFX8: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<2 x s32>)
; GFX8: [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](<2 x s32>)
; GFX8: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV2]]
; GFX8: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV3]]
; GFX8: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[MUL]](s32), [[MUL1]](s32)
; GFX8: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
; GFX9-LABEL: name: test_mul_v2s32
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; GFX9: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<2 x s32>)
; GFX9: [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](<2 x s32>)
; GFX9: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV2]]
; GFX9: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV3]]
; GFX9: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[MUL]](s32), [[MUL1]](s32)
; GFX9: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<2 x s32>) = COPY $vgpr0_vgpr1
%1:_(<2 x s32>) = COPY $vgpr2_vgpr3
%2:_(<2 x s32>) = G_MUL %0, %1
$vgpr0_vgpr1 = COPY %2
...
---
name: test_mul_s64
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX6-LABEL: name: test_mul_s64
; GFX6: [[COPY:%[0-9]+]]:_(s64) = COPY $vgpr0_vgpr1
; GFX6: [[COPY1:%[0-9]+]]:_(s64) = COPY $vgpr2_vgpr3
; GFX6: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](s64)
; GFX6: [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](s64)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV2]]
; GFX6: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV2]]
; GFX6: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV3]]
; GFX6: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV2]]
; GFX6: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[MUL1]], [[MUL2]]
; GFX6: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[ADD]], [[UMULH]]
; GFX6: [[MV:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL]](s32), [[ADD1]](s32)
; GFX6: $vgpr0_vgpr1 = COPY [[MV]](s64)
; GFX8-LABEL: name: test_mul_s64
; GFX8: [[COPY:%[0-9]+]]:_(s64) = COPY $vgpr0_vgpr1
; GFX8: [[COPY1:%[0-9]+]]:_(s64) = COPY $vgpr2_vgpr3
; GFX8: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](s64)
; GFX8: [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](s64)
; GFX8: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV2]]
; GFX8: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV2]]
; GFX8: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV3]]
; GFX8: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV2]]
; GFX8: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[MUL1]], [[MUL2]]
; GFX8: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[ADD]], [[UMULH]]
; GFX8: [[MV:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL]](s32), [[ADD1]](s32)
; GFX8: $vgpr0_vgpr1 = COPY [[MV]](s64)
; GFX9-LABEL: name: test_mul_s64
; GFX9: [[COPY:%[0-9]+]]:_(s64) = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:_(s64) = COPY $vgpr2_vgpr3
; GFX9: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](s64)
; GFX9: [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](s64)
; GFX9: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV2]]
; GFX9: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV2]]
; GFX9: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV3]]
; GFX9: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV2]]
; GFX9: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[MUL1]], [[MUL2]]
; GFX9: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[ADD]], [[UMULH]]
; GFX9: [[MV:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL]](s32), [[ADD1]](s32)
; GFX9: $vgpr0_vgpr1 = COPY [[MV]](s64)
%0:_(s64) = COPY $vgpr0_vgpr1
%1:_(s64) = COPY $vgpr2_vgpr3
%2:_(s64) = G_MUL %0, %1
$vgpr0_vgpr1 = COPY %2
...
---
name: test_mul_v2s64
body: |
bb.0:
liveins: $vgpr0_vgpr1_vgpr2_vgpr3, $vgpr4_vgpr5_vgpr6_vgpr7
; GFX6-LABEL: name: test_mul_v2s64
; GFX6: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
; GFX6: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
; GFX6: [[UV:%[0-9]+]]:_(s64), [[UV1:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY]](<2 x s64>)
; GFX6: [[UV2:%[0-9]+]]:_(s64), [[UV3:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY1]](<2 x s64>)
; GFX6: [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV]](s64)
; GFX6: [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV2]](s64)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV4]], [[UV6]]
; GFX6: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV5]], [[UV6]]
; GFX6: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV4]], [[UV7]]
; GFX6: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV4]], [[UV6]]
; GFX6: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[MUL1]], [[MUL2]]
; GFX6: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[ADD]], [[UMULH]]
; GFX6: [[MV:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL]](s32), [[ADD1]](s32)
; GFX6: [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV1]](s64)
; GFX6: [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV3]](s64)
; GFX6: [[MUL3:%[0-9]+]]:_(s32) = G_MUL [[UV8]], [[UV10]]
; GFX6: [[MUL4:%[0-9]+]]:_(s32) = G_MUL [[UV9]], [[UV10]]
; GFX6: [[MUL5:%[0-9]+]]:_(s32) = G_MUL [[UV8]], [[UV11]]
; GFX6: [[UMULH1:%[0-9]+]]:_(s32) = G_UMULH [[UV8]], [[UV10]]
; GFX6: [[ADD2:%[0-9]+]]:_(s32) = G_ADD [[MUL4]], [[MUL5]]
; GFX6: [[ADD3:%[0-9]+]]:_(s32) = G_ADD [[ADD2]], [[UMULH1]]
; GFX6: [[MV1:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL3]](s32), [[ADD3]](s32)
; GFX6: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[MV]](s64), [[MV1]](s64)
; GFX6: $vgpr0_vgpr1_vgpr2_vgpr3 = COPY [[BUILD_VECTOR]](<2 x s64>)
; GFX8-LABEL: name: test_mul_v2s64
; GFX8: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
; GFX8: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
; GFX8: [[UV:%[0-9]+]]:_(s64), [[UV1:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY]](<2 x s64>)
; GFX8: [[UV2:%[0-9]+]]:_(s64), [[UV3:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY1]](<2 x s64>)
; GFX8: [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV]](s64)
; GFX8: [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV2]](s64)
; GFX8: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV4]], [[UV6]]
; GFX8: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV5]], [[UV6]]
; GFX8: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV4]], [[UV7]]
; GFX8: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV4]], [[UV6]]
; GFX8: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[MUL1]], [[MUL2]]
; GFX8: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[ADD]], [[UMULH]]
; GFX8: [[MV:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL]](s32), [[ADD1]](s32)
; GFX8: [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV1]](s64)
; GFX8: [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV3]](s64)
; GFX8: [[MUL3:%[0-9]+]]:_(s32) = G_MUL [[UV8]], [[UV10]]
; GFX8: [[MUL4:%[0-9]+]]:_(s32) = G_MUL [[UV9]], [[UV10]]
; GFX8: [[MUL5:%[0-9]+]]:_(s32) = G_MUL [[UV8]], [[UV11]]
; GFX8: [[UMULH1:%[0-9]+]]:_(s32) = G_UMULH [[UV8]], [[UV10]]
; GFX8: [[ADD2:%[0-9]+]]:_(s32) = G_ADD [[MUL4]], [[MUL5]]
; GFX8: [[ADD3:%[0-9]+]]:_(s32) = G_ADD [[ADD2]], [[UMULH1]]
; GFX8: [[MV1:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL3]](s32), [[ADD3]](s32)
; GFX8: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[MV]](s64), [[MV1]](s64)
; GFX8: $vgpr0_vgpr1_vgpr2_vgpr3 = COPY [[BUILD_VECTOR]](<2 x s64>)
; GFX9-LABEL: name: test_mul_v2s64
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
; GFX9: [[UV:%[0-9]+]]:_(s64), [[UV1:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY]](<2 x s64>)
; GFX9: [[UV2:%[0-9]+]]:_(s64), [[UV3:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY1]](<2 x s64>)
; GFX9: [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV]](s64)
; GFX9: [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV2]](s64)
; GFX9: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV4]], [[UV6]]
; GFX9: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV5]], [[UV6]]
; GFX9: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV4]], [[UV7]]
; GFX9: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV4]], [[UV6]]
; GFX9: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[MUL1]], [[MUL2]]
; GFX9: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[ADD]], [[UMULH]]
; GFX9: [[MV:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL]](s32), [[ADD1]](s32)
; GFX9: [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV1]](s64)
; GFX9: [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UV3]](s64)
; GFX9: [[MUL3:%[0-9]+]]:_(s32) = G_MUL [[UV8]], [[UV10]]
; GFX9: [[MUL4:%[0-9]+]]:_(s32) = G_MUL [[UV9]], [[UV10]]
; GFX9: [[MUL5:%[0-9]+]]:_(s32) = G_MUL [[UV8]], [[UV11]]
; GFX9: [[UMULH1:%[0-9]+]]:_(s32) = G_UMULH [[UV8]], [[UV10]]
; GFX9: [[ADD2:%[0-9]+]]:_(s32) = G_ADD [[MUL4]], [[MUL5]]
; GFX9: [[ADD3:%[0-9]+]]:_(s32) = G_ADD [[ADD2]], [[UMULH1]]
; GFX9: [[MV1:%[0-9]+]]:_(s64) = G_MERGE_VALUES [[MUL3]](s32), [[ADD3]](s32)
; GFX9: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[MV]](s64), [[MV1]](s64)
; GFX9: $vgpr0_vgpr1_vgpr2_vgpr3 = COPY [[BUILD_VECTOR]](<2 x s64>)
%0:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
%1:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
%2:_(<2 x s64>) = G_MUL %0, %1
$vgpr0_vgpr1_vgpr2_vgpr3 = COPY %2
...
---
name: test_mul_s16
body: |
bb.0:
liveins: $vgpr0, $vgpr1
; GFX6-LABEL: name: test_mul_s16
; GFX6: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX6: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX6: [[COPY2:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX6: [[COPY3:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY2]], [[COPY3]]
; GFX6: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX6: [[COPY4:%[0-9]+]]:_(s32) = COPY [[MUL]](s32)
; GFX6: [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY4]], [[C]]
; GFX6: $vgpr0 = COPY [[AND]](s32)
; GFX8-LABEL: name: test_mul_s16
; GFX8: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX8: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[COPY]](s32)
; GFX8: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[COPY1]](s32)
; GFX8: [[MUL:%[0-9]+]]:_(s16) = G_MUL [[TRUNC]], [[TRUNC1]]
; GFX8: [[ZEXT:%[0-9]+]]:_(s32) = G_ZEXT [[MUL]](s16)
; GFX8: $vgpr0 = COPY [[ZEXT]](s32)
; GFX9-LABEL: name: test_mul_s16
; GFX9: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX9: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[COPY]](s32)
; GFX9: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[COPY1]](s32)
; GFX9: [[MUL:%[0-9]+]]:_(s16) = G_MUL [[TRUNC]], [[TRUNC1]]
; GFX9: [[ZEXT:%[0-9]+]]:_(s32) = G_ZEXT [[MUL]](s16)
; GFX9: $vgpr0 = COPY [[ZEXT]](s32)
%0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1
%2:_(s16) = G_TRUNC %0
%3:_(s16) = G_TRUNC %1
%4:_(s16) = G_MUL %2, %3
%5:_(s32) = G_ZEXT %4
$vgpr0 = COPY %5
...
---
name: test_mul_v2s16
body: |
bb.0:
liveins: $vgpr0, $vgpr1
; GFX6-LABEL: name: test_mul_v2s16
; GFX6: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX6: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX6: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX6: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX6: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX6: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX6: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX6: [[COPY2:%[0-9]+]]:_(s32) = COPY [[BITCAST]](s32)
; GFX6: [[COPY3:%[0-9]+]]:_(s32) = COPY [[BITCAST1]](s32)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY2]], [[COPY3]]
; GFX6: [[COPY4:%[0-9]+]]:_(s32) = COPY [[LSHR]](s32)
; GFX6: [[COPY5:%[0-9]+]]:_(s32) = COPY [[LSHR1]](s32)
; GFX6: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[COPY4]], [[COPY5]]
; GFX6: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX6: [[COPY6:%[0-9]+]]:_(s32) = COPY [[MUL]](s32)
; GFX6: [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY6]], [[C1]]
; GFX6: [[COPY7:%[0-9]+]]:_(s32) = COPY [[MUL1]](s32)
; GFX6: [[AND1:%[0-9]+]]:_(s32) = G_AND [[COPY7]], [[C1]]
; GFX6: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; GFX6: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX6: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX6: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX8-LABEL: name: test_mul_v2s16
; GFX8: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST]](s32)
; GFX8: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[LSHR]](s32)
; GFX8: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST1]](s32)
; GFX8: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8: [[TRUNC3:%[0-9]+]]:_(s16) = G_TRUNC [[LSHR1]](s32)
; GFX8: [[MUL:%[0-9]+]]:_(s16) = G_MUL [[TRUNC]], [[TRUNC2]]
; GFX8: [[MUL1:%[0-9]+]]:_(s16) = G_MUL [[TRUNC1]], [[TRUNC3]]
; GFX8: [[ZEXT:%[0-9]+]]:_(s32) = G_ZEXT [[MUL]](s16)
; GFX8: [[ZEXT1:%[0-9]+]]:_(s32) = G_ZEXT [[MUL1]](s16)
; GFX8: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[ZEXT1]], [[C]](s32)
; GFX8: [[OR:%[0-9]+]]:_(s32) = G_OR [[ZEXT]], [[SHL]]
; GFX8: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: test_mul_v2s16
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[MUL:%[0-9]+]]:_(<2 x s16>) = G_MUL [[COPY]], [[COPY1]]
; GFX9: $vgpr0 = COPY [[MUL]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_MUL %0, %1
$vgpr0 = COPY %2
...
---
name: test_mul_v3s16
body: |
bb.0:
liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX6-LABEL: name: test_mul_v3s16
; GFX6: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX6: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX6: [[COPY2:%[0-9]+]]:_(s32) = COPY $vgpr2
; GFX6: [[COPY3:%[0-9]+]]:_(s32) = COPY $vgpr3
; GFX6: [[COPY4:%[0-9]+]]:_(s32) = COPY $vgpr4
; GFX6: [[COPY5:%[0-9]+]]:_(s32) = COPY $vgpr5
; GFX6: [[COPY6:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX6: [[COPY7:%[0-9]+]]:_(s32) = COPY [[COPY3]](s32)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY6]], [[COPY7]]
; GFX6: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[MUL]](s32)
; GFX6: [[COPY8:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX6: [[COPY9:%[0-9]+]]:_(s32) = COPY [[COPY4]](s32)
; GFX6: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[COPY8]], [[COPY9]]
; GFX6: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[MUL1]](s32)
; GFX6: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY2]](s32)
; GFX6: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY5]](s32)
; GFX6: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[COPY10]], [[COPY11]]
; GFX6: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[MUL2]](s32)
; GFX6: S_ENDPGM 0, implicit [[TRUNC]](s16), implicit [[TRUNC1]](s16), implicit [[TRUNC2]](s16)
; GFX8-LABEL: name: test_mul_v3s16
; GFX8: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX8: [[COPY2:%[0-9]+]]:_(s32) = COPY $vgpr2
; GFX8: [[COPY3:%[0-9]+]]:_(s32) = COPY $vgpr3
; GFX8: [[COPY4:%[0-9]+]]:_(s32) = COPY $vgpr4
; GFX8: [[COPY5:%[0-9]+]]:_(s32) = COPY $vgpr5
; GFX8: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[COPY]](s32)
; GFX8: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[COPY1]](s32)
; GFX8: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[COPY2]](s32)
; GFX8: [[TRUNC3:%[0-9]+]]:_(s16) = G_TRUNC [[COPY3]](s32)
; GFX8: [[TRUNC4:%[0-9]+]]:_(s16) = G_TRUNC [[COPY4]](s32)
; GFX8: [[TRUNC5:%[0-9]+]]:_(s16) = G_TRUNC [[COPY5]](s32)
; GFX8: [[MUL:%[0-9]+]]:_(s16) = G_MUL [[TRUNC]], [[TRUNC3]]
; GFX8: [[MUL1:%[0-9]+]]:_(s16) = G_MUL [[TRUNC1]], [[TRUNC4]]
; GFX8: [[MUL2:%[0-9]+]]:_(s16) = G_MUL [[TRUNC2]], [[TRUNC5]]
; GFX8: S_ENDPGM 0, implicit [[MUL]](s16), implicit [[MUL1]](s16), implicit [[MUL2]](s16)
; GFX9-LABEL: name: test_mul_v3s16
; GFX9: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX9: [[COPY2:%[0-9]+]]:_(s32) = COPY $vgpr2
; GFX9: [[COPY3:%[0-9]+]]:_(s32) = COPY $vgpr3
; GFX9: [[COPY4:%[0-9]+]]:_(s32) = COPY $vgpr4
; GFX9: [[COPY5:%[0-9]+]]:_(s32) = COPY $vgpr5
; GFX9: [[COPY6:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX9: [[COPY7:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[COPY6]](s32), [[COPY7]](s32)
; GFX9: [[COPY8:%[0-9]+]]:_(s32) = COPY [[COPY2]](s32)
; GFX9: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX9: [[COPY9:%[0-9]+]]:_(s32) = COPY [[DEF]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC1:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[COPY8]](s32), [[COPY9]](s32)
; GFX9: [[DEF1:%[0-9]+]]:_(<2 x s16>) = G_IMPLICIT_DEF
; GFX9: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY3]](s32)
; GFX9: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY4]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC2:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[COPY10]](s32), [[COPY11]](s32)
; GFX9: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY5]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC3:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[COPY12]](s32), [[DEF]](s32)
; GFX9: [[MUL:%[0-9]+]]:_(<2 x s16>) = G_MUL [[BUILD_VECTOR_TRUNC]], [[BUILD_VECTOR_TRUNC2]]
; GFX9: [[MUL1:%[0-9]+]]:_(<2 x s16>) = G_MUL [[BUILD_VECTOR_TRUNC1]], [[BUILD_VECTOR_TRUNC3]]
; GFX9: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[MUL]](<2 x s16>)
; GFX9: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST]](s32)
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[LSHR]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[MUL1]](<2 x s16>)
; GFX9: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST1]](s32)
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX9: S_ENDPGM 0, implicit [[TRUNC]](s16), implicit [[TRUNC1]](s16), implicit [[TRUNC2]](s16)
%0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1
%2:_(s32) = COPY $vgpr2
%3:_(s32) = COPY $vgpr3
%4:_(s32) = COPY $vgpr4
%5:_(s32) = COPY $vgpr5
%6:_(s16) = G_TRUNC %0
%7:_(s16) = G_TRUNC %1
%8:_(s16) = G_TRUNC %2
%9:_(s16) = G_TRUNC %3
%10:_(s16) = G_TRUNC %4
%11:_(s16) = G_TRUNC %5
%12:_(<3 x s16>) = G_BUILD_VECTOR %6, %7, %8
%13:_(<3 x s16>) = G_BUILD_VECTOR %9, %10, %11
%14:_(<3 x s16>) = G_MUL %12, %13
%15:_(s16), %16:_(s16), %17:_(s16) = G_UNMERGE_VALUES %14
S_ENDPGM 0, implicit %15, implicit %16, implicit %17
...
---
name: test_mul_v4s16
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX6-LABEL: name: test_mul_v4s16
; GFX6: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX6: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX6: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[COPY]](<4 x s16>)
; GFX6: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>)
; GFX6: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX6: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX6: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>)
; GFX6: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX6: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[COPY1]](<4 x s16>)
; GFX6: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>)
; GFX6: [[LSHR2:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32)
; GFX6: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>)
; GFX6: [[LSHR3:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST3]], [[C]](s32)
; GFX6: [[COPY2:%[0-9]+]]:_(s32) = COPY [[BITCAST]](s32)
; GFX6: [[COPY3:%[0-9]+]]:_(s32) = COPY [[BITCAST2]](s32)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY2]], [[COPY3]]
; GFX6: [[COPY4:%[0-9]+]]:_(s32) = COPY [[LSHR]](s32)
; GFX6: [[COPY5:%[0-9]+]]:_(s32) = COPY [[LSHR2]](s32)
; GFX6: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[COPY4]], [[COPY5]]
; GFX6: [[COPY6:%[0-9]+]]:_(s32) = COPY [[BITCAST1]](s32)
; GFX6: [[COPY7:%[0-9]+]]:_(s32) = COPY [[BITCAST3]](s32)
; GFX6: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[COPY6]], [[COPY7]]
; GFX6: [[COPY8:%[0-9]+]]:_(s32) = COPY [[LSHR1]](s32)
; GFX6: [[COPY9:%[0-9]+]]:_(s32) = COPY [[LSHR3]](s32)
; GFX6: [[MUL3:%[0-9]+]]:_(s32) = G_MUL [[COPY8]], [[COPY9]]
; GFX6: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX6: [[COPY10:%[0-9]+]]:_(s32) = COPY [[MUL]](s32)
; GFX6: [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY10]], [[C1]]
; GFX6: [[COPY11:%[0-9]+]]:_(s32) = COPY [[MUL1]](s32)
; GFX6: [[AND1:%[0-9]+]]:_(s32) = G_AND [[COPY11]], [[C1]]
; GFX6: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; GFX6: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX6: [[BITCAST4:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX6: [[COPY12:%[0-9]+]]:_(s32) = COPY [[MUL2]](s32)
; GFX6: [[AND2:%[0-9]+]]:_(s32) = G_AND [[COPY12]], [[C1]]
; GFX6: [[COPY13:%[0-9]+]]:_(s32) = COPY [[MUL3]](s32)
; GFX6: [[AND3:%[0-9]+]]:_(s32) = G_AND [[COPY13]], [[C1]]
; GFX6: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[AND3]], [[C]](s32)
; GFX6: [[OR1:%[0-9]+]]:_(s32) = G_OR [[AND2]], [[SHL1]]
; GFX6: [[BITCAST5:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR1]](s32)
; GFX6: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[BITCAST4]](<2 x s16>), [[BITCAST5]](<2 x s16>)
; GFX6: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>)
; GFX8-LABEL: name: test_mul_v4s16
; GFX8: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX8: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX8: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[COPY]](<4 x s16>)
; GFX8: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>)
; GFX8: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST]](s32)
; GFX8: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[LSHR]](s32)
; GFX8: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>)
; GFX8: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST1]](s32)
; GFX8: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8: [[TRUNC3:%[0-9]+]]:_(s16) = G_TRUNC [[LSHR1]](s32)
; GFX8: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[COPY1]](<4 x s16>)
; GFX8: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>)
; GFX8: [[TRUNC4:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST2]](s32)
; GFX8: [[LSHR2:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32)
; GFX8: [[TRUNC5:%[0-9]+]]:_(s16) = G_TRUNC [[LSHR2]](s32)
; GFX8: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>)
; GFX8: [[TRUNC6:%[0-9]+]]:_(s16) = G_TRUNC [[BITCAST3]](s32)
; GFX8: [[LSHR3:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST3]], [[C]](s32)
; GFX8: [[TRUNC7:%[0-9]+]]:_(s16) = G_TRUNC [[LSHR3]](s32)
; GFX8: [[MUL:%[0-9]+]]:_(s16) = G_MUL [[TRUNC]], [[TRUNC4]]
; GFX8: [[MUL1:%[0-9]+]]:_(s16) = G_MUL [[TRUNC1]], [[TRUNC5]]
; GFX8: [[MUL2:%[0-9]+]]:_(s16) = G_MUL [[TRUNC2]], [[TRUNC6]]
; GFX8: [[MUL3:%[0-9]+]]:_(s16) = G_MUL [[TRUNC3]], [[TRUNC7]]
; GFX8: [[ZEXT:%[0-9]+]]:_(s32) = G_ZEXT [[MUL]](s16)
; GFX8: [[ZEXT1:%[0-9]+]]:_(s32) = G_ZEXT [[MUL1]](s16)
; GFX8: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[ZEXT1]], [[C]](s32)
; GFX8: [[OR:%[0-9]+]]:_(s32) = G_OR [[ZEXT]], [[SHL]]
; GFX8: [[BITCAST4:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8: [[ZEXT2:%[0-9]+]]:_(s32) = G_ZEXT [[MUL2]](s16)
; GFX8: [[ZEXT3:%[0-9]+]]:_(s32) = G_ZEXT [[MUL3]](s16)
; GFX8: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[ZEXT3]], [[C]](s32)
; GFX8: [[OR1:%[0-9]+]]:_(s32) = G_OR [[ZEXT2]], [[SHL1]]
; GFX8: [[BITCAST5:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR1]](s32)
; GFX8: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[BITCAST4]](<2 x s16>), [[BITCAST5]](<2 x s16>)
; GFX8: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>)
; GFX9-LABEL: name: test_mul_v4s16
; GFX9: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX9: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[COPY]](<4 x s16>)
; GFX9: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[COPY1]](<4 x s16>)
; GFX9: [[MUL:%[0-9]+]]:_(<2 x s16>) = G_MUL [[UV]], [[UV2]]
; GFX9: [[MUL1:%[0-9]+]]:_(<2 x s16>) = G_MUL [[UV1]], [[UV3]]
; GFX9: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[MUL]](<2 x s16>), [[MUL1]](<2 x s16>)
; GFX9: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>)
%0:_(<4 x s16>) = COPY $vgpr0_vgpr1
%1:_(<4 x s16>) = COPY $vgpr2_vgpr3
%2:_(<4 x s16>) = G_MUL %0, %1
$vgpr0_vgpr1 = COPY %2
...
---
name: test_mul_s24
body: |
bb.0:
liveins: $vgpr0, $vgpr1
; GFX6-LABEL: name: test_mul_s24
; GFX6: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX6: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX6: [[COPY2:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX6: [[COPY3:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY2]], [[COPY3]]
; GFX6: [[COPY4:%[0-9]+]]:_(s32) = COPY [[MUL]](s32)
; GFX6: $vgpr0 = COPY [[COPY4]](s32)
; GFX8-LABEL: name: test_mul_s24
; GFX8: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX8: [[COPY2:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX8: [[COPY3:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX8: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY2]], [[COPY3]]
; GFX8: [[COPY4:%[0-9]+]]:_(s32) = COPY [[MUL]](s32)
; GFX8: $vgpr0 = COPY [[COPY4]](s32)
; GFX9-LABEL: name: test_mul_s24
; GFX9: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX9: [[COPY2:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX9: [[COPY3:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX9: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[COPY2]], [[COPY3]]
; GFX9: [[COPY4:%[0-9]+]]:_(s32) = COPY [[MUL]](s32)
; GFX9: $vgpr0 = COPY [[COPY4]](s32)
%0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1
%2:_(s24) = G_TRUNC %0
%3:_(s24) = G_TRUNC %1
%4:_(s24) = G_MUL %2, %3
%5:_(s32) = G_ANYEXT %4
$vgpr0 = COPY %5
...
# FIXME:
# ---
# name: test_mul_s33
# body: |
# bb.0:
# liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
# %0:_(s64) = COPY $vgpr0_vgpr1
# %1:_(s64) = COPY $vgpr2_vgpr3
# %2:_(s33) = G_TRUNC %0
# %3:_(s33) = G_TRUNC %1
# %4:_(s33) = G_MUL %2, %3
# %5:_(s64) = G_ANYEXT %4
# $vgpr0_vgpr1 = COPY %5
# ...
---
name: test_mul_s96
body: |
bb.0:
liveins: $vgpr0_vgpr1_vgpr2, $vgpr3_vgpr4_vgpr5
; GFX6-LABEL: name: test_mul_s96
; GFX6: [[COPY:%[0-9]+]]:_(s96) = COPY $vgpr0_vgpr1_vgpr2
; GFX6: [[COPY1:%[0-9]+]]:_(s96) = COPY $vgpr3_vgpr4_vgpr5
; GFX6: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](s96)
; GFX6: [[UV3:%[0-9]+]]:_(s32), [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](s96)
; GFX6: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV3]]
; GFX6: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV3]]
; GFX6: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV4]]
; GFX6: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV3]]
; GFX6: [[UADDO:%[0-9]+]]:_(s32), [[UADDO1:%[0-9]+]]:_(s1) = G_UADDO [[MUL1]], [[MUL2]]
; GFX6: [[ZEXT:%[0-9]+]]:_(s32) = G_ZEXT [[UADDO1]](s1)
; GFX6: [[UADDO2:%[0-9]+]]:_(s32), [[UADDO3:%[0-9]+]]:_(s1) = G_UADDO [[UADDO]], [[UMULH]]
; GFX6: [[ZEXT1:%[0-9]+]]:_(s32) = G_ZEXT [[UADDO3]](s1)
; GFX6: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[ZEXT]], [[ZEXT1]]
; GFX6: [[MUL3:%[0-9]+]]:_(s32) = G_MUL [[UV2]], [[UV3]]
; GFX6: [[MUL4:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV4]]
; GFX6: [[MUL5:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV5]]
; GFX6: [[UMULH1:%[0-9]+]]:_(s32) = G_UMULH [[UV1]], [[UV3]]
; GFX6: [[UMULH2:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV4]]
; GFX6: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[MUL3]], [[MUL4]]
; GFX6: [[ADD2:%[0-9]+]]:_(s32) = G_ADD [[ADD1]], [[MUL5]]
; GFX6: [[ADD3:%[0-9]+]]:_(s32) = G_ADD [[ADD2]], [[UMULH1]]
; GFX6: [[ADD4:%[0-9]+]]:_(s32) = G_ADD [[ADD3]], [[UMULH2]]
; GFX6: [[ADD5:%[0-9]+]]:_(s32) = G_ADD [[ADD4]], [[ADD]]
; GFX6: [[MV:%[0-9]+]]:_(s96) = G_MERGE_VALUES [[MUL]](s32), [[UADDO2]](s32), [[ADD5]](s32)
; GFX6: $vgpr0_vgpr1_vgpr2 = COPY [[MV]](s96)
; GFX8-LABEL: name: test_mul_s96
; GFX8: [[COPY:%[0-9]+]]:_(s96) = COPY $vgpr0_vgpr1_vgpr2
; GFX8: [[COPY1:%[0-9]+]]:_(s96) = COPY $vgpr3_vgpr4_vgpr5
; GFX8: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](s96)
; GFX8: [[UV3:%[0-9]+]]:_(s32), [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](s96)
; GFX8: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV3]]
; GFX8: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV3]]
; GFX8: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV4]]
; GFX8: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV3]]
; GFX8: [[UADDO:%[0-9]+]]:_(s32), [[UADDO1:%[0-9]+]]:_(s1) = G_UADDO [[MUL1]], [[MUL2]]
; GFX8: [[ZEXT:%[0-9]+]]:_(s32) = G_ZEXT [[UADDO1]](s1)
; GFX8: [[UADDO2:%[0-9]+]]:_(s32), [[UADDO3:%[0-9]+]]:_(s1) = G_UADDO [[UADDO]], [[UMULH]]
; GFX8: [[ZEXT1:%[0-9]+]]:_(s32) = G_ZEXT [[UADDO3]](s1)
; GFX8: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[ZEXT]], [[ZEXT1]]
; GFX8: [[MUL3:%[0-9]+]]:_(s32) = G_MUL [[UV2]], [[UV3]]
; GFX8: [[MUL4:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV4]]
; GFX8: [[MUL5:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV5]]
; GFX8: [[UMULH1:%[0-9]+]]:_(s32) = G_UMULH [[UV1]], [[UV3]]
; GFX8: [[UMULH2:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV4]]
; GFX8: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[MUL3]], [[MUL4]]
; GFX8: [[ADD2:%[0-9]+]]:_(s32) = G_ADD [[ADD1]], [[MUL5]]
; GFX8: [[ADD3:%[0-9]+]]:_(s32) = G_ADD [[ADD2]], [[UMULH1]]
; GFX8: [[ADD4:%[0-9]+]]:_(s32) = G_ADD [[ADD3]], [[UMULH2]]
; GFX8: [[ADD5:%[0-9]+]]:_(s32) = G_ADD [[ADD4]], [[ADD]]
; GFX8: [[MV:%[0-9]+]]:_(s96) = G_MERGE_VALUES [[MUL]](s32), [[UADDO2]](s32), [[ADD5]](s32)
; GFX8: $vgpr0_vgpr1_vgpr2 = COPY [[MV]](s96)
; GFX9-LABEL: name: test_mul_s96
; GFX9: [[COPY:%[0-9]+]]:_(s96) = COPY $vgpr0_vgpr1_vgpr2
; GFX9: [[COPY1:%[0-9]+]]:_(s96) = COPY $vgpr3_vgpr4_vgpr5
; GFX9: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](s96)
; GFX9: [[UV3:%[0-9]+]]:_(s32), [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY1]](s96)
; GFX9: [[MUL:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV3]]
; GFX9: [[MUL1:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV3]]
; GFX9: [[MUL2:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV4]]
; GFX9: [[UMULH:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV3]]
; GFX9: [[UADDO:%[0-9]+]]:_(s32), [[UADDO1:%[0-9]+]]:_(s1) = G_UADDO [[MUL1]], [[MUL2]]
; GFX9: [[ZEXT:%[0-9]+]]:_(s32) = G_ZEXT [[UADDO1]](s1)
; GFX9: [[UADDO2:%[0-9]+]]:_(s32), [[UADDO3:%[0-9]+]]:_(s1) = G_UADDO [[UADDO]], [[UMULH]]
; GFX9: [[ZEXT1:%[0-9]+]]:_(s32) = G_ZEXT [[UADDO3]](s1)
; GFX9: [[ADD:%[0-9]+]]:_(s32) = G_ADD [[ZEXT]], [[ZEXT1]]
; GFX9: [[MUL3:%[0-9]+]]:_(s32) = G_MUL [[UV2]], [[UV3]]
; GFX9: [[MUL4:%[0-9]+]]:_(s32) = G_MUL [[UV1]], [[UV4]]
; GFX9: [[MUL5:%[0-9]+]]:_(s32) = G_MUL [[UV]], [[UV5]]
; GFX9: [[UMULH1:%[0-9]+]]:_(s32) = G_UMULH [[UV1]], [[UV3]]
; GFX9: [[UMULH2:%[0-9]+]]:_(s32) = G_UMULH [[UV]], [[UV4]]
; GFX9: [[ADD1:%[0-9]+]]:_(s32) = G_ADD [[MUL3]], [[MUL4]]
; GFX9: [[ADD2:%[0-9]+]]:_(s32) = G_ADD [[ADD1]], [[MUL5]]
; GFX9: [[ADD3:%[0-9]+]]:_(s32) = G_ADD [[ADD2]], [[UMULH1]]
; GFX9: [[ADD4:%[0-9]+]]:_(s32) = G_ADD [[ADD3]], [[UMULH2]]
; GFX9: [[ADD5:%[0-9]+]]:_(s32) = G_ADD [[ADD4]], [[ADD]]
; GFX9: [[MV:%[0-9]+]]:_(s96) = G_MERGE_VALUES [[MUL]](s32), [[UADDO2]](s32), [[ADD5]](s32)
; GFX9: $vgpr0_vgpr1_vgpr2 = COPY [[MV]](s96)
%0:_(s96) = COPY $vgpr0_vgpr1_vgpr2
%1:_(s96) = COPY $vgpr3_vgpr4_vgpr5
%2:_(s96) = G_MUL %0, %1
$vgpr0_vgpr1_vgpr2 = COPY %2
...