llvm-for-llvmta/test/CodeGen/RISCV/rvv/vwredsum-rv32.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -mattr=+experimental-v -verify-machineinstrs \
; RUN:   --riscv-no-aliases < %s | FileCheck %s
declare <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv1i8(
  <vscale x 4 x i16>,
  <vscale x 1 x i8>,
  <vscale x 4 x i16>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_vs_nxv4i16_nxv1i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 1 x i8> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv4i16_nxv1i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv1i8(
    <vscale x 4 x i16> %0,
    <vscale x 1 x i8> %1,
    <vscale x 4 x i16> %2,
    i32 %3)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv1i8.nxv4i16(
  <vscale x 4 x i16>,
  <vscale x 1 x i8>,
  <vscale x 4 x i16>,
  <vscale x 1 x i1>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_mask_vs_nxv4i16_nxv1i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 1 x i8> %1, <vscale x 4 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv4i16_nxv1i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv1i8.nxv4i16(
    <vscale x 4 x i16> %0,
    <vscale x 1 x i8> %1,
    <vscale x 4 x i16> %2,
    <vscale x 1 x i1> %3,
    i32 %4)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv2i8(
  <vscale x 4 x i16>,
  <vscale x 2 x i8>,
  <vscale x 4 x i16>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_vs_nxv4i16_nxv2i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 2 x i8> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv4i16_nxv2i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv2i8(
    <vscale x 4 x i16> %0,
    <vscale x 2 x i8> %1,
    <vscale x 4 x i16> %2,
    i32 %3)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv2i8.nxv4i16(
  <vscale x 4 x i16>,
  <vscale x 2 x i8>,
  <vscale x 4 x i16>,
  <vscale x 2 x i1>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_mask_vs_nxv4i16_nxv2i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 2 x i8> %1, <vscale x 4 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv4i16_nxv2i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv2i8.nxv4i16(
    <vscale x 4 x i16> %0,
    <vscale x 2 x i8> %1,
    <vscale x 4 x i16> %2,
    <vscale x 2 x i1> %3,
    i32 %4)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv4i8(
  <vscale x 4 x i16>,
  <vscale x 4 x i8>,
  <vscale x 4 x i16>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_vs_nxv4i16_nxv4i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv4i16_nxv4i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv4i8(
    <vscale x 4 x i16> %0,
    <vscale x 4 x i8> %1,
    <vscale x 4 x i16> %2,
    i32 %3)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv4i8.nxv4i16(
  <vscale x 4 x i16>,
  <vscale x 4 x i8>,
  <vscale x 4 x i16>,
  <vscale x 4 x i1>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_mask_vs_nxv4i16_nxv4i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv4i16_nxv4i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv4i8.nxv4i16(
    <vscale x 4 x i16> %0,
    <vscale x 4 x i8> %1,
    <vscale x 4 x i16> %2,
    <vscale x 4 x i1> %3,
    i32 %4)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv8i8(
  <vscale x 4 x i16>,
  <vscale x 8 x i8>,
  <vscale x 4 x i16>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_vs_nxv4i16_nxv8i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 8 x i8> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv4i16_nxv8i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv8i8(
    <vscale x 4 x i16> %0,
    <vscale x 8 x i8> %1,
    <vscale x 4 x i16> %2,
    i32 %3)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv8i8.nxv4i16(
  <vscale x 4 x i16>,
  <vscale x 8 x i8>,
  <vscale x 4 x i16>,
  <vscale x 8 x i1>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_mask_vs_nxv4i16_nxv8i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 8 x i8> %1, <vscale x 4 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv4i16_nxv8i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv8i8.nxv4i16(
    <vscale x 4 x i16> %0,
    <vscale x 8 x i8> %1,
    <vscale x 4 x i16> %2,
    <vscale x 8 x i1> %3,
    i32 %4)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv16i8(
  <vscale x 4 x i16>,
  <vscale x 16 x i8>,
  <vscale x 4 x i16>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_vs_nxv4i16_nxv16i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 16 x i8> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv4i16_nxv16i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v10, v9
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv16i8(
    <vscale x 4 x i16> %0,
    <vscale x 16 x i8> %1,
    <vscale x 4 x i16> %2,
    i32 %3)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv16i8.nxv4i16(
  <vscale x 4 x i16>,
  <vscale x 16 x i8>,
  <vscale x 4 x i16>,
  <vscale x 16 x i1>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_mask_vs_nxv4i16_nxv16i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 16 x i8> %1, <vscale x 4 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv4i16_nxv16i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v10, v9, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv16i8.nxv4i16(
    <vscale x 4 x i16> %0,
    <vscale x 16 x i8> %1,
    <vscale x 4 x i16> %2,
    <vscale x 16 x i1> %3,
    i32 %4)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv32i8(
  <vscale x 4 x i16>,
  <vscale x 32 x i8>,
  <vscale x 4 x i16>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_vs_nxv4i16_nxv32i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 32 x i8> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv4i16_nxv32i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v12, v9
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv32i8(
    <vscale x 4 x i16> %0,
    <vscale x 32 x i8> %1,
    <vscale x 4 x i16> %2,
    i32 %3)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv32i8.nxv4i16(
  <vscale x 4 x i16>,
  <vscale x 32 x i8>,
  <vscale x 4 x i16>,
  <vscale x 32 x i1>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_mask_vs_nxv4i16_nxv32i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 32 x i8> %1, <vscale x 4 x i16> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv4i16_nxv32i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v12, v9, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv32i8.nxv4i16(
    <vscale x 4 x i16> %0,
    <vscale x 32 x i8> %1,
    <vscale x 4 x i16> %2,
    <vscale x 32 x i1> %3,
    i32 %4)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv64i8(
  <vscale x 4 x i16>,
  <vscale x 64 x i8>,
  <vscale x 4 x i16>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_vs_nxv4i16_nxv64i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 64 x i8> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv4i16_nxv64i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m8,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v16, v9
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.nxv4i16.nxv64i8(
    <vscale x 4 x i16> %0,
    <vscale x 64 x i8> %1,
    <vscale x 4 x i16> %2,
    i32 %3)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv64i8.nxv4i16(
  <vscale x 4 x i16>,
  <vscale x 64 x i8>,
  <vscale x 4 x i16>,
  <vscale x 64 x i1>,
  i32);

define <vscale x 4 x i16> @intrinsic_vwredsum_mask_vs_nxv4i16_nxv64i8_nxv4i16(<vscale x 4 x i16> %0, <vscale x 64 x i8> %1, <vscale x 4 x i16> %2, <vscale x 64 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv4i16_nxv64i8_nxv4i16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e8,m8,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v16, v9, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 4 x i16> @llvm.riscv.vwredsum.mask.nxv4i16.nxv64i8.nxv4i16(
    <vscale x 4 x i16> %0,
    <vscale x 64 x i8> %1,
    <vscale x 4 x i16> %2,
    <vscale x 64 x i1> %3,
    i32 %4)

  ret <vscale x 4 x i16> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv1i16(
  <vscale x 2 x i32>,
  <vscale x 1 x i16>,
  <vscale x 2 x i32>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_vs_nxv2i32_nxv1i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 1 x i16> %1, <vscale x 2 x i32> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv2i32_nxv1i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv1i16(
    <vscale x 2 x i32> %0,
    <vscale x 1 x i16> %1,
    <vscale x 2 x i32> %2,
    i32 %3)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv1i16.nxv2i32(
  <vscale x 2 x i32>,
  <vscale x 1 x i16>,
  <vscale x 2 x i32>,
  <vscale x 1 x i1>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_mask_vs_nxv2i32_nxv1i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 1 x i16> %1, <vscale x 2 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv2i32_nxv1i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv1i16.nxv2i32(
    <vscale x 2 x i32> %0,
    <vscale x 1 x i16> %1,
    <vscale x 2 x i32> %2,
    <vscale x 1 x i1> %3,
    i32 %4)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv2i16(
  <vscale x 2 x i32>,
  <vscale x 2 x i16>,
  <vscale x 2 x i32>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_vs_nxv2i32_nxv2i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i32> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv2i32_nxv2i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv2i16(
    <vscale x 2 x i32> %0,
    <vscale x 2 x i16> %1,
    <vscale x 2 x i32> %2,
    i32 %3)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv2i16.nxv2i32(
  <vscale x 2 x i32>,
  <vscale x 2 x i16>,
  <vscale x 2 x i32>,
  <vscale x 2 x i1>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_mask_vs_nxv2i32_nxv2i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv2i32_nxv2i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv2i16.nxv2i32(
    <vscale x 2 x i32> %0,
    <vscale x 2 x i16> %1,
    <vscale x 2 x i32> %2,
    <vscale x 2 x i1> %3,
    i32 %4)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv4i16(
  <vscale x 2 x i32>,
  <vscale x 4 x i16>,
  <vscale x 2 x i32>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_vs_nxv2i32_nxv4i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 4 x i16> %1, <vscale x 2 x i32> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv2i32_nxv4i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv4i16(
    <vscale x 2 x i32> %0,
    <vscale x 4 x i16> %1,
    <vscale x 2 x i32> %2,
    i32 %3)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv4i16.nxv2i32(
  <vscale x 2 x i32>,
  <vscale x 4 x i16>,
  <vscale x 2 x i32>,
  <vscale x 4 x i1>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_mask_vs_nxv2i32_nxv4i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 4 x i16> %1, <vscale x 2 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv2i32_nxv4i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v9, v10, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv4i16.nxv2i32(
    <vscale x 2 x i32> %0,
    <vscale x 4 x i16> %1,
    <vscale x 2 x i32> %2,
    <vscale x 4 x i1> %3,
    i32 %4)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv8i16(
  <vscale x 2 x i32>,
  <vscale x 8 x i16>,
  <vscale x 2 x i32>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_vs_nxv2i32_nxv8i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 8 x i16> %1, <vscale x 2 x i32> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv2i32_nxv8i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v10, v9
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv8i16(
    <vscale x 2 x i32> %0,
    <vscale x 8 x i16> %1,
    <vscale x 2 x i32> %2,
    i32 %3)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv8i16.nxv2i32(
  <vscale x 2 x i32>,
  <vscale x 8 x i16>,
  <vscale x 2 x i32>,
  <vscale x 8 x i1>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_mask_vs_nxv2i32_nxv8i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 8 x i16> %1, <vscale x 2 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv2i32_nxv8i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v10, v9, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv8i16.nxv2i32(
    <vscale x 2 x i32> %0,
    <vscale x 8 x i16> %1,
    <vscale x 2 x i32> %2,
    <vscale x 8 x i1> %3,
    i32 %4)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv16i16(
  <vscale x 2 x i32>,
  <vscale x 16 x i16>,
  <vscale x 2 x i32>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_vs_nxv2i32_nxv16i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 16 x i16> %1, <vscale x 2 x i32> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv2i32_nxv16i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v12, v9
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv16i16(
    <vscale x 2 x i32> %0,
    <vscale x 16 x i16> %1,
    <vscale x 2 x i32> %2,
    i32 %3)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv16i16.nxv2i32(
  <vscale x 2 x i32>,
  <vscale x 16 x i16>,
  <vscale x 2 x i32>,
  <vscale x 16 x i1>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_mask_vs_nxv2i32_nxv16i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 16 x i16> %1, <vscale x 2 x i32> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv2i32_nxv16i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m4,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v12, v9, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv16i16.nxv2i32(
    <vscale x 2 x i32> %0,
    <vscale x 16 x i16> %1,
    <vscale x 2 x i32> %2,
    <vscale x 16 x i1> %3,
    i32 %4)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv32i16(
  <vscale x 2 x i32>,
  <vscale x 32 x i16>,
  <vscale x 2 x i32>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_vs_nxv2i32_nxv32i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 32 x i16> %1, <vscale x 2 x i32> %2, i32 %3) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_vs_nxv2i32_nxv32i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m8,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v16, v9
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.nxv2i32.nxv32i16(
    <vscale x 2 x i32> %0,
    <vscale x 32 x i16> %1,
    <vscale x 2 x i32> %2,
    i32 %3)

  ret <vscale x 2 x i32> %a
}

declare <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv32i16.nxv2i32(
  <vscale x 2 x i32>,
  <vscale x 32 x i16>,
  <vscale x 2 x i32>,
  <vscale x 32 x i1>,
  i32);

define <vscale x 2 x i32> @intrinsic_vwredsum_mask_vs_nxv2i32_nxv32i16_nxv2i32(<vscale x 2 x i32> %0, <vscale x 32 x i16> %1, <vscale x 2 x i32> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
; CHECK-LABEL: intrinsic_vwredsum_mask_vs_nxv2i32_nxv32i16_nxv2i32:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e16,m8,ta,mu
; CHECK-NEXT:    vwredsum.vs v8, v16, v9, v0.t
; CHECK-NEXT:    jalr zero, 0(ra)
entry:
  %a = call <vscale x 2 x i32> @llvm.riscv.vwredsum.mask.nxv2i32.nxv32i16.nxv2i32(
    <vscale x 2 x i32> %0,
    <vscale x 32 x i16> %1,
    <vscale x 2 x i32> %2,
    <vscale x 32 x i1> %3,
    i32 %4)

  ret <vscale x 2 x i32> %a
}