; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc -mtriple=riscv32 -mattr=+v,+zvfh,+zvfbfmin -verify-machineinstrs < %s | FileCheck %s
; RUN: llc -mtriple=riscv64 -mattr=+v,+zvfh,+zvfbfmin -verify-machineinstrs < %s | FileCheck %s
; RUN: llc -mtriple=riscv32 -mattr=+v,+zvfhmin,+zvfbfmin -verify-machineinstrs < %s | FileCheck %s
; RUN: llc -mtriple=riscv64 -mattr=+v,+zvfhmin,+zvfbfmin -verify-machineinstrs < %s | FileCheck %s

define <1 x bfloat> @masked_load_v1bf16(ptr %a, <1 x i1> %mask) {
; CHECK-LABEL: masked_load_v1bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 1, e16, mf4, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <1 x bfloat> @llvm.masked.load.v1bf16(ptr %a, i32 8, <1 x i1> %mask, <1 x bfloat> poison)
  ret <1 x bfloat> %load
}

define <1 x half> @masked_load_v1f16(ptr %a, <1 x i1> %mask) {
; CHECK-LABEL: masked_load_v1f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 1, e16, mf4, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <1 x half> @llvm.masked.load.v1f16(ptr %a, i32 8, <1 x i1> %mask, <1 x half> poison)
  ret <1 x half> %load
}

define <1 x float> @masked_load_v1f32(ptr %a, <1 x i1> %mask) {
; CHECK-LABEL: masked_load_v1f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <1 x float> @llvm.masked.load.v1f32(ptr %a, i32 8, <1 x i1> %mask, <1 x float> poison)
  ret <1 x float> %load
}

define <1 x double> @masked_load_v1f64(ptr %a, <1 x i1> %mask) {
; CHECK-LABEL: masked_load_v1f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
; CHECK-NEXT:    vle64.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <1 x double> @llvm.masked.load.v1f64(ptr %a, i32 8, <1 x i1> %mask, <1 x double> poison)
  ret <1 x double> %load
}

define <2 x bfloat> @masked_load_v2bf16(ptr %a, <2 x i1> %mask) {
; CHECK-LABEL: masked_load_v2bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 2, e16, mf4, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <2 x bfloat> @llvm.masked.load.v2bf16(ptr %a, i32 8, <2 x i1> %mask, <2 x bfloat> poison)
  ret <2 x bfloat> %load
}

define <2 x half> @masked_load_v2f16(ptr %a, <2 x i1> %mask) {
; CHECK-LABEL: masked_load_v2f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 2, e16, mf4, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <2 x half> @llvm.masked.load.v2f16(ptr %a, i32 8, <2 x i1> %mask, <2 x half> poison)
  ret <2 x half> %load
}

define <2 x float> @masked_load_v2f32(ptr %a, <2 x i1> %mask) {
; CHECK-LABEL: masked_load_v2f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 2, e32, mf2, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <2 x float> @llvm.masked.load.v2f32(ptr %a, i32 8, <2 x i1> %mask, <2 x float> poison)
  ret <2 x float> %load
}

define <2 x double> @masked_load_v2f64(ptr %a, <2 x i1> %mask) {
; CHECK-LABEL: masked_load_v2f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; CHECK-NEXT:    vle64.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <2 x double> @llvm.masked.load.v2f64(ptr %a, i32 8, <2 x i1> %mask, <2 x double> poison)
  ret <2 x double> %load
}

define <4 x bfloat> @masked_load_v4bf16(ptr %a, <4 x i1> %mask) {
; CHECK-LABEL: masked_load_v4bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 4, e16, mf2, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <4 x bfloat> @llvm.masked.load.v4bf16(ptr %a, i32 8, <4 x i1> %mask, <4 x bfloat> poison)
  ret <4 x bfloat> %load
}

define <4 x half> @masked_load_v4f16(ptr %a, <4 x i1> %mask) {
; CHECK-LABEL: masked_load_v4f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 4, e16, mf2, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <4 x half> @llvm.masked.load.v4f16(ptr %a, i32 8, <4 x i1> %mask, <4 x half> poison)
  ret <4 x half> %load
}

define <4 x float> @masked_load_v4f32(ptr %a, <4 x i1> %mask) {
; CHECK-LABEL: masked_load_v4f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <4 x float> @llvm.masked.load.v4f32(ptr %a, i32 8, <4 x i1> %mask, <4 x float> poison)
  ret <4 x float> %load
}

define <4 x double> @masked_load_v4f64(ptr %a, <4 x i1> %mask) {
; CHECK-LABEL: masked_load_v4f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 4, e64, m2, ta, ma
; CHECK-NEXT:    vle64.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <4 x double> @llvm.masked.load.v4f64(ptr %a, i32 8, <4 x i1> %mask, <4 x double> poison)
  ret <4 x double> %load
}

define <8 x bfloat> @masked_load_v8bf16(ptr %a, <8 x i1> %mask) {
; CHECK-LABEL: masked_load_v8bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 8, e16, m1, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <8 x bfloat> @llvm.masked.load.v8bf16(ptr %a, i32 8, <8 x i1> %mask, <8 x bfloat> poison)
  ret <8 x bfloat> %load
}

define <8 x half> @masked_load_v8f16(ptr %a, <8 x i1> %mask) {
; CHECK-LABEL: masked_load_v8f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 8, e16, m1, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <8 x half> @llvm.masked.load.v8f16(ptr %a, i32 8, <8 x i1> %mask, <8 x half> poison)
  ret <8 x half> %load
}

define <8 x float> @masked_load_v8f32(ptr %a, <8 x i1> %mask) {
; CHECK-LABEL: masked_load_v8f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <8 x float> @llvm.masked.load.v8f32(ptr %a, i32 8, <8 x i1> %mask, <8 x float> poison)
  ret <8 x float> %load
}

define <8 x double> @masked_load_v8f64(ptr %a, <8 x i1> %mask) {
; CHECK-LABEL: masked_load_v8f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 8, e64, m4, ta, ma
; CHECK-NEXT:    vle64.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <8 x double> @llvm.masked.load.v8f64(ptr %a, i32 8, <8 x i1> %mask, <8 x double> poison)
  ret <8 x double> %load
}

define <16 x bfloat> @masked_load_v16bf16(ptr %a, <16 x i1> %mask) {
; CHECK-LABEL: masked_load_v16bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <16 x bfloat> @llvm.masked.load.v16bf16(ptr %a, i32 8, <16 x i1> %mask, <16 x bfloat> poison)
  ret <16 x bfloat> %load
}

define <16 x half> @masked_load_v16f16(ptr %a, <16 x i1> %mask) {
; CHECK-LABEL: masked_load_v16f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <16 x half> @llvm.masked.load.v16f16(ptr %a, i32 8, <16 x i1> %mask, <16 x half> poison)
  ret <16 x half> %load
}

define <16 x float> @masked_load_v16f32(ptr %a, <16 x i1> %mask) {
; CHECK-LABEL: masked_load_v16f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 16, e32, m4, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <16 x float> @llvm.masked.load.v16f32(ptr %a, i32 8, <16 x i1> %mask, <16 x float> poison)
  ret <16 x float> %load
}

define <16 x double> @masked_load_v16f64(ptr %a, <16 x i1> %mask) {
; CHECK-LABEL: masked_load_v16f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, ma
; CHECK-NEXT:    vle64.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <16 x double> @llvm.masked.load.v16f64(ptr %a, i32 8, <16 x i1> %mask, <16 x double> poison)
  ret <16 x double> %load
}

define <32 x bfloat> @masked_load_v32bf16(ptr %a, <32 x i1> %mask) {
; CHECK-LABEL: masked_load_v32bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 32
; CHECK-NEXT:    vsetvli zero, a1, e16, m4, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <32 x bfloat> @llvm.masked.load.v32bf16(ptr %a, i32 8, <32 x i1> %mask, <32 x bfloat> poison)
  ret <32 x bfloat> %load
}

define <32 x half> @masked_load_v32f16(ptr %a, <32 x i1> %mask) {
; CHECK-LABEL: masked_load_v32f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 32
; CHECK-NEXT:    vsetvli zero, a1, e16, m4, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <32 x half> @llvm.masked.load.v32f16(ptr %a, i32 8, <32 x i1> %mask, <32 x half> poison)
  ret <32 x half> %load
}

define <32 x float> @masked_load_v32f32(ptr %a, <32 x i1> %mask) {
; CHECK-LABEL: masked_load_v32f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 32
; CHECK-NEXT:    vsetvli zero, a1, e32, m8, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <32 x float> @llvm.masked.load.v32f32(ptr %a, i32 8, <32 x i1> %mask, <32 x float> poison)
  ret <32 x float> %load
}

define <32 x double> @masked_load_v32f64(ptr %a, <32 x i1> %mask) {
; CHECK-LABEL: masked_load_v32f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, ma
; CHECK-NEXT:    vle64.v v8, (a0), v0.t
; CHECK-NEXT:    vsetivli zero, 2, e8, mf4, ta, ma
; CHECK-NEXT:    vslidedown.vi v0, v0, 2
; CHECK-NEXT:    addi a0, a0, 128
; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, ma
; CHECK-NEXT:    vle64.v v16, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <32 x double> @llvm.masked.load.v32f64(ptr %a, i32 8, <32 x i1> %mask, <32 x double> poison)
  ret <32 x double> %load
}

define <64 x bfloat> @masked_load_v64bf16(ptr %a, <64 x i1> %mask) {
; CHECK-LABEL: masked_load_v64bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 64
; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <64 x bfloat> @llvm.masked.load.v64bf16(ptr %a, i32 8, <64 x i1> %mask, <64 x bfloat> poison)
  ret <64 x bfloat> %load
}

define <64 x half> @masked_load_v64f16(ptr %a, <64 x i1> %mask) {
; CHECK-LABEL: masked_load_v64f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 64
; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <64 x half> @llvm.masked.load.v64f16(ptr %a, i32 8, <64 x i1> %mask, <64 x half> poison)
  ret <64 x half> %load
}

define <64 x float> @masked_load_v64f32(ptr %a, <64 x i1> %mask) {
; CHECK-LABEL: masked_load_v64f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 32
; CHECK-NEXT:    vsetvli zero, a1, e32, m8, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    vsetivli zero, 4, e8, mf2, ta, ma
; CHECK-NEXT:    vslidedown.vi v0, v0, 4
; CHECK-NEXT:    addi a0, a0, 128
; CHECK-NEXT:    vsetvli zero, a1, e32, m8, ta, ma
; CHECK-NEXT:    vle32.v v16, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <64 x float> @llvm.masked.load.v64f32(ptr %a, i32 8, <64 x i1> %mask, <64 x float> poison)
  ret <64 x float> %load
}

define <128 x bfloat> @masked_load_v128bf16(ptr %a, <128 x i1> %mask) {
; CHECK-LABEL: masked_load_v128bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 64
; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    vsetivli zero, 8, e8, m1, ta, ma
; CHECK-NEXT:    vslidedown.vi v0, v0, 8
; CHECK-NEXT:    addi a0, a0, 128
; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, ma
; CHECK-NEXT:    vle16.v v16, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <128 x bfloat> @llvm.masked.load.v128bf16(ptr %a, i32 8, <128 x i1> %mask, <128 x bfloat> poison)
  ret <128 x bfloat> %load
}

define <128 x half> @masked_load_v128f16(ptr %a, <128 x i1> %mask) {
; CHECK-LABEL: masked_load_v128f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 64
; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, ma
; CHECK-NEXT:    vle16.v v8, (a0), v0.t
; CHECK-NEXT:    vsetivli zero, 8, e8, m1, ta, ma
; CHECK-NEXT:    vslidedown.vi v0, v0, 8
; CHECK-NEXT:    addi a0, a0, 128
; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, ma
; CHECK-NEXT:    vle16.v v16, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <128 x half> @llvm.masked.load.v128f16(ptr %a, i32 8, <128 x i1> %mask, <128 x half> poison)
  ret <128 x half> %load
}

define <7 x float> @masked_load_v7f32(ptr %a, <7 x i1> %mask) {
; CHECK-LABEL: masked_load_v7f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 7, e32, m2, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0), v0.t
; CHECK-NEXT:    ret
  %load = call <7 x float> @llvm.masked.load.v7f32(ptr %a, i32 8, <7 x i1> %mask, <7 x float> poison)
  ret <7 x float> %load
}