llvm-for-llvmta/test/CodeGen/RISCV/rvv/vamoand-rv64.ll

1683 lines
54 KiB
LLVM

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv64 -mattr=+experimental-v,+d,+experimental-zvamo -verify-machineinstrs \
; RUN: --riscv-no-aliases < %s | FileCheck %s
declare <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i64(
<vscale x 1 x i32>*,
<vscale x 1 x i64>,
<vscale x 1 x i32>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_v_nxv1i32_nxv1i64(<vscale x 1 x i32> *%0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i32_nxv1i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei64.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i64(
<vscale x 1 x i32> *%0,
<vscale x 1 x i64> %1,
<vscale x 1 x i32> %2,
i64 %3)
ret <vscale x 1 x i32> %a
}
declare <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i64(
<vscale x 1 x i32>*,
<vscale x 1 x i64>,
<vscale x 1 x i32>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_mask_v_nxv1i32_nxv1i64(<vscale x 1 x i32> *%0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i32_nxv1i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei64.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i64(
<vscale x 1 x i32> *%0,
<vscale x 1 x i64> %1,
<vscale x 1 x i32> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i64(
<vscale x 2 x i32>*,
<vscale x 2 x i64>,
<vscale x 2 x i32>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_v_nxv2i32_nxv2i64(<vscale x 2 x i32> *%0, <vscale x 2 x i64> %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i32_nxv2i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei64.v v10, (a0), v8, v10
; CHECK-NEXT: vmv1r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i64(
<vscale x 2 x i32> *%0,
<vscale x 2 x i64> %1,
<vscale x 2 x i32> %2,
i64 %3)
ret <vscale x 2 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i64(
<vscale x 2 x i32>*,
<vscale x 2 x i64>,
<vscale x 2 x i32>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_mask_v_nxv2i32_nxv2i64(<vscale x 2 x i32> *%0, <vscale x 2 x i64> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i32_nxv2i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei64.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv1r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i64(
<vscale x 2 x i32> *%0,
<vscale x 2 x i64> %1,
<vscale x 2 x i32> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i64(
<vscale x 4 x i32>*,
<vscale x 4 x i64>,
<vscale x 4 x i32>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_v_nxv4i32_nxv4i64(<vscale x 4 x i32> *%0, <vscale x 4 x i64> %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i32_nxv4i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei64.v v12, (a0), v8, v12
; CHECK-NEXT: vmv2r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i64(
<vscale x 4 x i32> *%0,
<vscale x 4 x i64> %1,
<vscale x 4 x i32> %2,
i64 %3)
ret <vscale x 4 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i64(
<vscale x 4 x i32>*,
<vscale x 4 x i64>,
<vscale x 4 x i32>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_mask_v_nxv4i32_nxv4i64(<vscale x 4 x i32> *%0, <vscale x 4 x i64> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i32_nxv4i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei64.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv2r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i64(
<vscale x 4 x i32> *%0,
<vscale x 4 x i64> %1,
<vscale x 4 x i32> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i64(
<vscale x 8 x i32>*,
<vscale x 8 x i64>,
<vscale x 8 x i32>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_v_nxv8i32_nxv8i64(<vscale x 8 x i32> *%0, <vscale x 8 x i64> %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i32_nxv8i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei64.v v16, (a0), v8, v16
; CHECK-NEXT: vmv4r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i64(
<vscale x 8 x i32> *%0,
<vscale x 8 x i64> %1,
<vscale x 8 x i32> %2,
i64 %3)
ret <vscale x 8 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i64(
<vscale x 8 x i32>*,
<vscale x 8 x i64>,
<vscale x 8 x i32>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_mask_v_nxv8i32_nxv8i64(<vscale x 8 x i32> *%0, <vscale x 8 x i64> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i32_nxv8i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei64.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv4r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i64(
<vscale x 8 x i32> *%0,
<vscale x 8 x i64> %1,
<vscale x 8 x i32> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i32> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i64(
<vscale x 1 x i64>*,
<vscale x 1 x i64>,
<vscale x 1 x i64>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_v_nxv1i64_nxv1i64(<vscale x 1 x i64> *%0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i64_nxv1i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei64.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i64(
<vscale x 1 x i64> *%0,
<vscale x 1 x i64> %1,
<vscale x 1 x i64> %2,
i64 %3)
ret <vscale x 1 x i64> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i64(
<vscale x 1 x i64>*,
<vscale x 1 x i64>,
<vscale x 1 x i64>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64> *%0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i64_nxv1i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei64.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i64(
<vscale x 1 x i64> *%0,
<vscale x 1 x i64> %1,
<vscale x 1 x i64> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i64(
<vscale x 2 x i64>*,
<vscale x 2 x i64>,
<vscale x 2 x i64>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_v_nxv2i64_nxv2i64(<vscale x 2 x i64> *%0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i64_nxv2i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei64.v v10, (a0), v8, v10
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i64(
<vscale x 2 x i64> *%0,
<vscale x 2 x i64> %1,
<vscale x 2 x i64> %2,
i64 %3)
ret <vscale x 2 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i64(
<vscale x 2 x i64>*,
<vscale x 2 x i64>,
<vscale x 2 x i64>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_mask_v_nxv2i64_nxv2i64(<vscale x 2 x i64> *%0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i64_nxv2i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei64.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i64(
<vscale x 2 x i64> *%0,
<vscale x 2 x i64> %1,
<vscale x 2 x i64> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i64(
<vscale x 4 x i64>*,
<vscale x 4 x i64>,
<vscale x 4 x i64>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_v_nxv4i64_nxv4i64(<vscale x 4 x i64> *%0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i64_nxv4i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei64.v v12, (a0), v8, v12
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i64(
<vscale x 4 x i64> *%0,
<vscale x 4 x i64> %1,
<vscale x 4 x i64> %2,
i64 %3)
ret <vscale x 4 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i64(
<vscale x 4 x i64>*,
<vscale x 4 x i64>,
<vscale x 4 x i64>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_mask_v_nxv4i64_nxv4i64(<vscale x 4 x i64> *%0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i64_nxv4i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei64.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i64(
<vscale x 4 x i64> *%0,
<vscale x 4 x i64> %1,
<vscale x 4 x i64> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i64(
<vscale x 8 x i64>*,
<vscale x 8 x i64>,
<vscale x 8 x i64>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_v_nxv8i64_nxv8i64(<vscale x 8 x i64> *%0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i64_nxv8i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei64.v v16, (a0), v8, v16
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i64(
<vscale x 8 x i64> *%0,
<vscale x 8 x i64> %1,
<vscale x 8 x i64> %2,
i64 %3)
ret <vscale x 8 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i64(
<vscale x 8 x i64>*,
<vscale x 8 x i64>,
<vscale x 8 x i64>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_mask_v_nxv8i64_nxv8i64(<vscale x 8 x i64> *%0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i64_nxv8i64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei64.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i64(
<vscale x 8 x i64> *%0,
<vscale x 8 x i64> %1,
<vscale x 8 x i64> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i64> %a
}
declare <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i32(
<vscale x 1 x i32>*,
<vscale x 1 x i32>,
<vscale x 1 x i32>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_v_nxv1i32_nxv1i32(<vscale x 1 x i32> *%0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i32_nxv1i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei32.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i32(
<vscale x 1 x i32> *%0,
<vscale x 1 x i32> %1,
<vscale x 1 x i32> %2,
i64 %3)
ret <vscale x 1 x i32> %a
}
declare <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i32(
<vscale x 1 x i32>*,
<vscale x 1 x i32>,
<vscale x 1 x i32>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_mask_v_nxv1i32_nxv1i32(<vscale x 1 x i32> *%0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i32_nxv1i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei32.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i32(
<vscale x 1 x i32> *%0,
<vscale x 1 x i32> %1,
<vscale x 1 x i32> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i32(
<vscale x 2 x i32>*,
<vscale x 2 x i32>,
<vscale x 2 x i32>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_v_nxv2i32_nxv2i32(<vscale x 2 x i32> *%0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i32_nxv2i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei32.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i32(
<vscale x 2 x i32> *%0,
<vscale x 2 x i32> %1,
<vscale x 2 x i32> %2,
i64 %3)
ret <vscale x 2 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i32(
<vscale x 2 x i32>*,
<vscale x 2 x i32>,
<vscale x 2 x i32>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_mask_v_nxv2i32_nxv2i32(<vscale x 2 x i32> *%0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i32_nxv2i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei32.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i32(
<vscale x 2 x i32> *%0,
<vscale x 2 x i32> %1,
<vscale x 2 x i32> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i32(
<vscale x 4 x i32>*,
<vscale x 4 x i32>,
<vscale x 4 x i32>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_v_nxv4i32_nxv4i32(<vscale x 4 x i32> *%0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i32_nxv4i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei32.v v10, (a0), v8, v10
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i32(
<vscale x 4 x i32> *%0,
<vscale x 4 x i32> %1,
<vscale x 4 x i32> %2,
i64 %3)
ret <vscale x 4 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i32(
<vscale x 4 x i32>*,
<vscale x 4 x i32>,
<vscale x 4 x i32>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_mask_v_nxv4i32_nxv4i32(<vscale x 4 x i32> *%0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i32_nxv4i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei32.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i32(
<vscale x 4 x i32> *%0,
<vscale x 4 x i32> %1,
<vscale x 4 x i32> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i32(
<vscale x 8 x i32>*,
<vscale x 8 x i32>,
<vscale x 8 x i32>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_v_nxv8i32_nxv8i32(<vscale x 8 x i32> *%0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i32_nxv8i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei32.v v12, (a0), v8, v12
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i32(
<vscale x 8 x i32> *%0,
<vscale x 8 x i32> %1,
<vscale x 8 x i32> %2,
i64 %3)
ret <vscale x 8 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i32(
<vscale x 8 x i32>*,
<vscale x 8 x i32>,
<vscale x 8 x i32>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_mask_v_nxv8i32_nxv8i32(<vscale x 8 x i32> *%0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i32_nxv8i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei32.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i32(
<vscale x 8 x i32> *%0,
<vscale x 8 x i32> %1,
<vscale x 8 x i32> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i32> %a
}
declare <vscale x 16 x i32> @llvm.riscv.vamoand.nxv16i32.nxv16i32(
<vscale x 16 x i32>*,
<vscale x 16 x i32>,
<vscale x 16 x i32>,
i64);
define <vscale x 16 x i32> @intrinsic_vamoand_v_nxv16i32_nxv16i32(<vscale x 16 x i32> *%0, <vscale x 16 x i32> %1, <vscale x 16 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv16i32_nxv16i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m8,tu,mu
; CHECK-NEXT: vamoandei32.v v16, (a0), v8, v16
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 16 x i32> @llvm.riscv.vamoand.nxv16i32.nxv16i32(
<vscale x 16 x i32> *%0,
<vscale x 16 x i32> %1,
<vscale x 16 x i32> %2,
i64 %3)
ret <vscale x 16 x i32> %a
}
declare <vscale x 16 x i32> @llvm.riscv.vamoand.mask.nxv16i32.nxv16i32(
<vscale x 16 x i32>*,
<vscale x 16 x i32>,
<vscale x 16 x i32>,
<vscale x 16 x i1>,
i64);
define <vscale x 16 x i32> @intrinsic_vamoand_mask_v_nxv16i32_nxv16i32(<vscale x 16 x i32> *%0, <vscale x 16 x i32> %1, <vscale x 16 x i32> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv16i32_nxv16i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m8,tu,mu
; CHECK-NEXT: vamoandei32.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 16 x i32> @llvm.riscv.vamoand.mask.nxv16i32.nxv16i32(
<vscale x 16 x i32> *%0,
<vscale x 16 x i32> %1,
<vscale x 16 x i32> %2,
<vscale x 16 x i1> %3,
i64 %4)
ret <vscale x 16 x i32> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i32(
<vscale x 1 x i64>*,
<vscale x 1 x i32>,
<vscale x 1 x i64>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_v_nxv1i64_nxv1i32(<vscale x 1 x i64> *%0, <vscale x 1 x i32> %1, <vscale x 1 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i64_nxv1i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei32.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i32(
<vscale x 1 x i64> *%0,
<vscale x 1 x i32> %1,
<vscale x 1 x i64> %2,
i64 %3)
ret <vscale x 1 x i64> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i32(
<vscale x 1 x i64>*,
<vscale x 1 x i32>,
<vscale x 1 x i64>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_mask_v_nxv1i64_nxv1i32(<vscale x 1 x i64> *%0, <vscale x 1 x i32> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i64_nxv1i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei32.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i32(
<vscale x 1 x i64> *%0,
<vscale x 1 x i32> %1,
<vscale x 1 x i64> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i32(
<vscale x 2 x i64>*,
<vscale x 2 x i32>,
<vscale x 2 x i64>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_v_nxv2i64_nxv2i32(<vscale x 2 x i64> *%0, <vscale x 2 x i32> %1, <vscale x 2 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i64_nxv2i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei32.v v10, (a0), v8, v10
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i32(
<vscale x 2 x i64> *%0,
<vscale x 2 x i32> %1,
<vscale x 2 x i64> %2,
i64 %3)
ret <vscale x 2 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i32(
<vscale x 2 x i64>*,
<vscale x 2 x i32>,
<vscale x 2 x i64>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_mask_v_nxv2i64_nxv2i32(<vscale x 2 x i64> *%0, <vscale x 2 x i32> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i64_nxv2i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei32.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i32(
<vscale x 2 x i64> *%0,
<vscale x 2 x i32> %1,
<vscale x 2 x i64> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i32(
<vscale x 4 x i64>*,
<vscale x 4 x i32>,
<vscale x 4 x i64>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_v_nxv4i64_nxv4i32(<vscale x 4 x i64> *%0, <vscale x 4 x i32> %1, <vscale x 4 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i64_nxv4i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei32.v v12, (a0), v8, v12
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i32(
<vscale x 4 x i64> *%0,
<vscale x 4 x i32> %1,
<vscale x 4 x i64> %2,
i64 %3)
ret <vscale x 4 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i32(
<vscale x 4 x i64>*,
<vscale x 4 x i32>,
<vscale x 4 x i64>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_mask_v_nxv4i64_nxv4i32(<vscale x 4 x i64> *%0, <vscale x 4 x i32> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i64_nxv4i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei32.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i32(
<vscale x 4 x i64> *%0,
<vscale x 4 x i32> %1,
<vscale x 4 x i64> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i32(
<vscale x 8 x i64>*,
<vscale x 8 x i32>,
<vscale x 8 x i64>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_v_nxv8i64_nxv8i32(<vscale x 8 x i64> *%0, <vscale x 8 x i32> %1, <vscale x 8 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i64_nxv8i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei32.v v16, (a0), v8, v16
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i32(
<vscale x 8 x i64> *%0,
<vscale x 8 x i32> %1,
<vscale x 8 x i64> %2,
i64 %3)
ret <vscale x 8 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i32(
<vscale x 8 x i64>*,
<vscale x 8 x i32>,
<vscale x 8 x i64>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_mask_v_nxv8i64_nxv8i32(<vscale x 8 x i64> *%0, <vscale x 8 x i32> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i64_nxv8i32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei32.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i32(
<vscale x 8 x i64> *%0,
<vscale x 8 x i32> %1,
<vscale x 8 x i64> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i64> %a
}
declare <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i16(
<vscale x 1 x i32>*,
<vscale x 1 x i16>,
<vscale x 1 x i32>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_v_nxv1i32_nxv1i16(<vscale x 1 x i32> *%0, <vscale x 1 x i16> %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i32_nxv1i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei16.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i16(
<vscale x 1 x i32> *%0,
<vscale x 1 x i16> %1,
<vscale x 1 x i32> %2,
i64 %3)
ret <vscale x 1 x i32> %a
}
declare <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i16(
<vscale x 1 x i32>*,
<vscale x 1 x i16>,
<vscale x 1 x i32>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_mask_v_nxv1i32_nxv1i16(<vscale x 1 x i32> *%0, <vscale x 1 x i16> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i32_nxv1i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei16.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i16(
<vscale x 1 x i32> *%0,
<vscale x 1 x i16> %1,
<vscale x 1 x i32> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i16(
<vscale x 2 x i32>*,
<vscale x 2 x i16>,
<vscale x 2 x i32>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_v_nxv2i32_nxv2i16(<vscale x 2 x i32> *%0, <vscale x 2 x i16> %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i32_nxv2i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei16.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i16(
<vscale x 2 x i32> *%0,
<vscale x 2 x i16> %1,
<vscale x 2 x i32> %2,
i64 %3)
ret <vscale x 2 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i16(
<vscale x 2 x i32>*,
<vscale x 2 x i16>,
<vscale x 2 x i32>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_mask_v_nxv2i32_nxv2i16(<vscale x 2 x i32> *%0, <vscale x 2 x i16> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i32_nxv2i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei16.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i16(
<vscale x 2 x i32> *%0,
<vscale x 2 x i16> %1,
<vscale x 2 x i32> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i16(
<vscale x 4 x i32>*,
<vscale x 4 x i16>,
<vscale x 4 x i32>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_v_nxv4i32_nxv4i16(<vscale x 4 x i32> *%0, <vscale x 4 x i16> %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i32_nxv4i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei16.v v10, (a0), v8, v10
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i16(
<vscale x 4 x i32> *%0,
<vscale x 4 x i16> %1,
<vscale x 4 x i32> %2,
i64 %3)
ret <vscale x 4 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i16(
<vscale x 4 x i32>*,
<vscale x 4 x i16>,
<vscale x 4 x i32>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_mask_v_nxv4i32_nxv4i16(<vscale x 4 x i32> *%0, <vscale x 4 x i16> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i32_nxv4i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei16.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i16(
<vscale x 4 x i32> *%0,
<vscale x 4 x i16> %1,
<vscale x 4 x i32> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i16(
<vscale x 8 x i32>*,
<vscale x 8 x i16>,
<vscale x 8 x i32>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_v_nxv8i32_nxv8i16(<vscale x 8 x i32> *%0, <vscale x 8 x i16> %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i32_nxv8i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei16.v v12, (a0), v8, v12
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i16(
<vscale x 8 x i32> *%0,
<vscale x 8 x i16> %1,
<vscale x 8 x i32> %2,
i64 %3)
ret <vscale x 8 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i16(
<vscale x 8 x i32>*,
<vscale x 8 x i16>,
<vscale x 8 x i32>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_mask_v_nxv8i32_nxv8i16(<vscale x 8 x i32> *%0, <vscale x 8 x i16> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i32_nxv8i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei16.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i16(
<vscale x 8 x i32> *%0,
<vscale x 8 x i16> %1,
<vscale x 8 x i32> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i32> %a
}
declare <vscale x 16 x i32> @llvm.riscv.vamoand.nxv16i32.nxv16i16(
<vscale x 16 x i32>*,
<vscale x 16 x i16>,
<vscale x 16 x i32>,
i64);
define <vscale x 16 x i32> @intrinsic_vamoand_v_nxv16i32_nxv16i16(<vscale x 16 x i32> *%0, <vscale x 16 x i16> %1, <vscale x 16 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv16i32_nxv16i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m8,tu,mu
; CHECK-NEXT: vamoandei16.v v16, (a0), v8, v16
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 16 x i32> @llvm.riscv.vamoand.nxv16i32.nxv16i16(
<vscale x 16 x i32> *%0,
<vscale x 16 x i16> %1,
<vscale x 16 x i32> %2,
i64 %3)
ret <vscale x 16 x i32> %a
}
declare <vscale x 16 x i32> @llvm.riscv.vamoand.mask.nxv16i32.nxv16i16(
<vscale x 16 x i32>*,
<vscale x 16 x i16>,
<vscale x 16 x i32>,
<vscale x 16 x i1>,
i64);
define <vscale x 16 x i32> @intrinsic_vamoand_mask_v_nxv16i32_nxv16i16(<vscale x 16 x i32> *%0, <vscale x 16 x i16> %1, <vscale x 16 x i32> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv16i32_nxv16i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m8,tu,mu
; CHECK-NEXT: vamoandei16.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 16 x i32> @llvm.riscv.vamoand.mask.nxv16i32.nxv16i16(
<vscale x 16 x i32> *%0,
<vscale x 16 x i16> %1,
<vscale x 16 x i32> %2,
<vscale x 16 x i1> %3,
i64 %4)
ret <vscale x 16 x i32> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i16(
<vscale x 1 x i64>*,
<vscale x 1 x i16>,
<vscale x 1 x i64>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_v_nxv1i64_nxv1i16(<vscale x 1 x i64> *%0, <vscale x 1 x i16> %1, <vscale x 1 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i64_nxv1i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei16.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i16(
<vscale x 1 x i64> *%0,
<vscale x 1 x i16> %1,
<vscale x 1 x i64> %2,
i64 %3)
ret <vscale x 1 x i64> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i16(
<vscale x 1 x i64>*,
<vscale x 1 x i16>,
<vscale x 1 x i64>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_mask_v_nxv1i64_nxv1i16(<vscale x 1 x i64> *%0, <vscale x 1 x i16> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i64_nxv1i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei16.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i16(
<vscale x 1 x i64> *%0,
<vscale x 1 x i16> %1,
<vscale x 1 x i64> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i16(
<vscale x 2 x i64>*,
<vscale x 2 x i16>,
<vscale x 2 x i64>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_v_nxv2i64_nxv2i16(<vscale x 2 x i64> *%0, <vscale x 2 x i16> %1, <vscale x 2 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i64_nxv2i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei16.v v10, (a0), v8, v10
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i16(
<vscale x 2 x i64> *%0,
<vscale x 2 x i16> %1,
<vscale x 2 x i64> %2,
i64 %3)
ret <vscale x 2 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i16(
<vscale x 2 x i64>*,
<vscale x 2 x i16>,
<vscale x 2 x i64>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_mask_v_nxv2i64_nxv2i16(<vscale x 2 x i64> *%0, <vscale x 2 x i16> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i64_nxv2i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei16.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i16(
<vscale x 2 x i64> *%0,
<vscale x 2 x i16> %1,
<vscale x 2 x i64> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i16(
<vscale x 4 x i64>*,
<vscale x 4 x i16>,
<vscale x 4 x i64>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_v_nxv4i64_nxv4i16(<vscale x 4 x i64> *%0, <vscale x 4 x i16> %1, <vscale x 4 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i64_nxv4i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei16.v v12, (a0), v8, v12
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i16(
<vscale x 4 x i64> *%0,
<vscale x 4 x i16> %1,
<vscale x 4 x i64> %2,
i64 %3)
ret <vscale x 4 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i16(
<vscale x 4 x i64>*,
<vscale x 4 x i16>,
<vscale x 4 x i64>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_mask_v_nxv4i64_nxv4i16(<vscale x 4 x i64> *%0, <vscale x 4 x i16> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i64_nxv4i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei16.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i16(
<vscale x 4 x i64> *%0,
<vscale x 4 x i16> %1,
<vscale x 4 x i64> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i16(
<vscale x 8 x i64>*,
<vscale x 8 x i16>,
<vscale x 8 x i64>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_v_nxv8i64_nxv8i16(<vscale x 8 x i64> *%0, <vscale x 8 x i16> %1, <vscale x 8 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i64_nxv8i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei16.v v16, (a0), v8, v16
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i16(
<vscale x 8 x i64> *%0,
<vscale x 8 x i16> %1,
<vscale x 8 x i64> %2,
i64 %3)
ret <vscale x 8 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i16(
<vscale x 8 x i64>*,
<vscale x 8 x i16>,
<vscale x 8 x i64>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_mask_v_nxv8i64_nxv8i16(<vscale x 8 x i64> *%0, <vscale x 8 x i16> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i64_nxv8i16:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei16.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i16(
<vscale x 8 x i64> *%0,
<vscale x 8 x i16> %1,
<vscale x 8 x i64> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i64> %a
}
declare <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i8(
<vscale x 1 x i32>*,
<vscale x 1 x i8>,
<vscale x 1 x i32>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_v_nxv1i32_nxv1i8(<vscale x 1 x i32> *%0, <vscale x 1 x i8> %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i32_nxv1i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei8.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.nxv1i32.nxv1i8(
<vscale x 1 x i32> *%0,
<vscale x 1 x i8> %1,
<vscale x 1 x i32> %2,
i64 %3)
ret <vscale x 1 x i32> %a
}
declare <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i8(
<vscale x 1 x i32>*,
<vscale x 1 x i8>,
<vscale x 1 x i32>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i32> @intrinsic_vamoand_mask_v_nxv1i32_nxv1i8(<vscale x 1 x i32> *%0, <vscale x 1 x i8> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i32_nxv1i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,mf2,tu,mu
; CHECK-NEXT: vamoandei8.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i32> @llvm.riscv.vamoand.mask.nxv1i32.nxv1i8(
<vscale x 1 x i32> *%0,
<vscale x 1 x i8> %1,
<vscale x 1 x i32> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i8(
<vscale x 2 x i32>*,
<vscale x 2 x i8>,
<vscale x 2 x i32>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_v_nxv2i32_nxv2i8(<vscale x 2 x i32> *%0, <vscale x 2 x i8> %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i32_nxv2i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei8.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.nxv2i32.nxv2i8(
<vscale x 2 x i32> *%0,
<vscale x 2 x i8> %1,
<vscale x 2 x i32> %2,
i64 %3)
ret <vscale x 2 x i32> %a
}
declare <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i8(
<vscale x 2 x i32>*,
<vscale x 2 x i8>,
<vscale x 2 x i32>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i32> @intrinsic_vamoand_mask_v_nxv2i32_nxv2i8(<vscale x 2 x i32> *%0, <vscale x 2 x i8> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i32_nxv2i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m1,tu,mu
; CHECK-NEXT: vamoandei8.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i32> @llvm.riscv.vamoand.mask.nxv2i32.nxv2i8(
<vscale x 2 x i32> *%0,
<vscale x 2 x i8> %1,
<vscale x 2 x i32> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i8(
<vscale x 4 x i32>*,
<vscale x 4 x i8>,
<vscale x 4 x i32>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_v_nxv4i32_nxv4i8(<vscale x 4 x i32> *%0, <vscale x 4 x i8> %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i32_nxv4i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei8.v v10, (a0), v8, v10
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.nxv4i32.nxv4i8(
<vscale x 4 x i32> *%0,
<vscale x 4 x i8> %1,
<vscale x 4 x i32> %2,
i64 %3)
ret <vscale x 4 x i32> %a
}
declare <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i8(
<vscale x 4 x i32>*,
<vscale x 4 x i8>,
<vscale x 4 x i32>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i32> @intrinsic_vamoand_mask_v_nxv4i32_nxv4i8(<vscale x 4 x i32> *%0, <vscale x 4 x i8> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i32_nxv4i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m2,tu,mu
; CHECK-NEXT: vamoandei8.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i32> @llvm.riscv.vamoand.mask.nxv4i32.nxv4i8(
<vscale x 4 x i32> *%0,
<vscale x 4 x i8> %1,
<vscale x 4 x i32> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i8(
<vscale x 8 x i32>*,
<vscale x 8 x i8>,
<vscale x 8 x i32>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_v_nxv8i32_nxv8i8(<vscale x 8 x i32> *%0, <vscale x 8 x i8> %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i32_nxv8i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei8.v v12, (a0), v8, v12
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.nxv8i32.nxv8i8(
<vscale x 8 x i32> *%0,
<vscale x 8 x i8> %1,
<vscale x 8 x i32> %2,
i64 %3)
ret <vscale x 8 x i32> %a
}
declare <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i8(
<vscale x 8 x i32>*,
<vscale x 8 x i8>,
<vscale x 8 x i32>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i32> @intrinsic_vamoand_mask_v_nxv8i32_nxv8i8(<vscale x 8 x i32> *%0, <vscale x 8 x i8> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i32_nxv8i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m4,tu,mu
; CHECK-NEXT: vamoandei8.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i32> @llvm.riscv.vamoand.mask.nxv8i32.nxv8i8(
<vscale x 8 x i32> *%0,
<vscale x 8 x i8> %1,
<vscale x 8 x i32> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i32> %a
}
declare <vscale x 16 x i32> @llvm.riscv.vamoand.nxv16i32.nxv16i8(
<vscale x 16 x i32>*,
<vscale x 16 x i8>,
<vscale x 16 x i32>,
i64);
define <vscale x 16 x i32> @intrinsic_vamoand_v_nxv16i32_nxv16i8(<vscale x 16 x i32> *%0, <vscale x 16 x i8> %1, <vscale x 16 x i32> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv16i32_nxv16i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m8,tu,mu
; CHECK-NEXT: vamoandei8.v v16, (a0), v8, v16
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 16 x i32> @llvm.riscv.vamoand.nxv16i32.nxv16i8(
<vscale x 16 x i32> *%0,
<vscale x 16 x i8> %1,
<vscale x 16 x i32> %2,
i64 %3)
ret <vscale x 16 x i32> %a
}
declare <vscale x 16 x i32> @llvm.riscv.vamoand.mask.nxv16i32.nxv16i8(
<vscale x 16 x i32>*,
<vscale x 16 x i8>,
<vscale x 16 x i32>,
<vscale x 16 x i1>,
i64);
define <vscale x 16 x i32> @intrinsic_vamoand_mask_v_nxv16i32_nxv16i8(<vscale x 16 x i32> *%0, <vscale x 16 x i8> %1, <vscale x 16 x i32> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv16i32_nxv16i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e32,m8,tu,mu
; CHECK-NEXT: vamoandei8.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 16 x i32> @llvm.riscv.vamoand.mask.nxv16i32.nxv16i8(
<vscale x 16 x i32> *%0,
<vscale x 16 x i8> %1,
<vscale x 16 x i32> %2,
<vscale x 16 x i1> %3,
i64 %4)
ret <vscale x 16 x i32> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i8(
<vscale x 1 x i64>*,
<vscale x 1 x i8>,
<vscale x 1 x i64>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_v_nxv1i64_nxv1i8(<vscale x 1 x i64> *%0, <vscale x 1 x i8> %1, <vscale x 1 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv1i64_nxv1i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei8.v v9, (a0), v8, v9
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.nxv1i64.nxv1i8(
<vscale x 1 x i64> *%0,
<vscale x 1 x i8> %1,
<vscale x 1 x i64> %2,
i64 %3)
ret <vscale x 1 x i64> %a
}
declare <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i8(
<vscale x 1 x i64>*,
<vscale x 1 x i8>,
<vscale x 1 x i64>,
<vscale x 1 x i1>,
i64);
define <vscale x 1 x i64> @intrinsic_vamoand_mask_v_nxv1i64_nxv1i8(<vscale x 1 x i64> *%0, <vscale x 1 x i8> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv1i64_nxv1i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m1,tu,mu
; CHECK-NEXT: vamoandei8.v v9, (a0), v8, v9, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 1 x i64> @llvm.riscv.vamoand.mask.nxv1i64.nxv1i8(
<vscale x 1 x i64> *%0,
<vscale x 1 x i8> %1,
<vscale x 1 x i64> %2,
<vscale x 1 x i1> %3,
i64 %4)
ret <vscale x 1 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i8(
<vscale x 2 x i64>*,
<vscale x 2 x i8>,
<vscale x 2 x i64>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_v_nxv2i64_nxv2i8(<vscale x 2 x i64> *%0, <vscale x 2 x i8> %1, <vscale x 2 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv2i64_nxv2i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei8.v v10, (a0), v8, v10
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.nxv2i64.nxv2i8(
<vscale x 2 x i64> *%0,
<vscale x 2 x i8> %1,
<vscale x 2 x i64> %2,
i64 %3)
ret <vscale x 2 x i64> %a
}
declare <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i8(
<vscale x 2 x i64>*,
<vscale x 2 x i8>,
<vscale x 2 x i64>,
<vscale x 2 x i1>,
i64);
define <vscale x 2 x i64> @intrinsic_vamoand_mask_v_nxv2i64_nxv2i8(<vscale x 2 x i64> *%0, <vscale x 2 x i8> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv2i64_nxv2i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m2,tu,mu
; CHECK-NEXT: vamoandei8.v v10, (a0), v8, v10, v0.t
; CHECK-NEXT: vmv2r.v v8, v10
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 2 x i64> @llvm.riscv.vamoand.mask.nxv2i64.nxv2i8(
<vscale x 2 x i64> *%0,
<vscale x 2 x i8> %1,
<vscale x 2 x i64> %2,
<vscale x 2 x i1> %3,
i64 %4)
ret <vscale x 2 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i8(
<vscale x 4 x i64>*,
<vscale x 4 x i8>,
<vscale x 4 x i64>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_v_nxv4i64_nxv4i8(<vscale x 4 x i64> *%0, <vscale x 4 x i8> %1, <vscale x 4 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv4i64_nxv4i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei8.v v12, (a0), v8, v12
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.nxv4i64.nxv4i8(
<vscale x 4 x i64> *%0,
<vscale x 4 x i8> %1,
<vscale x 4 x i64> %2,
i64 %3)
ret <vscale x 4 x i64> %a
}
declare <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i8(
<vscale x 4 x i64>*,
<vscale x 4 x i8>,
<vscale x 4 x i64>,
<vscale x 4 x i1>,
i64);
define <vscale x 4 x i64> @intrinsic_vamoand_mask_v_nxv4i64_nxv4i8(<vscale x 4 x i64> *%0, <vscale x 4 x i8> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv4i64_nxv4i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m4,tu,mu
; CHECK-NEXT: vamoandei8.v v12, (a0), v8, v12, v0.t
; CHECK-NEXT: vmv4r.v v8, v12
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 4 x i64> @llvm.riscv.vamoand.mask.nxv4i64.nxv4i8(
<vscale x 4 x i64> *%0,
<vscale x 4 x i8> %1,
<vscale x 4 x i64> %2,
<vscale x 4 x i1> %3,
i64 %4)
ret <vscale x 4 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i8(
<vscale x 8 x i64>*,
<vscale x 8 x i8>,
<vscale x 8 x i64>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_v_nxv8i64_nxv8i8(<vscale x 8 x i64> *%0, <vscale x 8 x i8> %1, <vscale x 8 x i64> %2, i64 %3) nounwind {
; CHECK-LABEL: intrinsic_vamoand_v_nxv8i64_nxv8i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei8.v v16, (a0), v8, v16
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.nxv8i64.nxv8i8(
<vscale x 8 x i64> *%0,
<vscale x 8 x i8> %1,
<vscale x 8 x i64> %2,
i64 %3)
ret <vscale x 8 x i64> %a
}
declare <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i8(
<vscale x 8 x i64>*,
<vscale x 8 x i8>,
<vscale x 8 x i64>,
<vscale x 8 x i1>,
i64);
define <vscale x 8 x i64> @intrinsic_vamoand_mask_v_nxv8i64_nxv8i8(<vscale x 8 x i64> *%0, <vscale x 8 x i8> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
; CHECK-LABEL: intrinsic_vamoand_mask_v_nxv8i64_nxv8i8:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vsetvli a1, a1, e64,m8,tu,mu
; CHECK-NEXT: vamoandei8.v v16, (a0), v8, v16, v0.t
; CHECK-NEXT: vmv8r.v v8, v16
; CHECK-NEXT: jalr zero, 0(ra)
entry:
%a = call <vscale x 8 x i64> @llvm.riscv.vamoand.mask.nxv8i64.nxv8i8(
<vscale x 8 x i64> *%0,
<vscale x 8 x i8> %1,
<vscale x 8 x i64> %2,
<vscale x 8 x i1> %3,
i64 %4)
ret <vscale x 8 x i64> %a
}