llvm/test/CodeGen/AArch64/sve-nontemporal-masked-ldst.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve < %s | FileCheck %s

define <4 x i32> @masked_load_v4i32(ptr %a, <4 x i1> %mask) nounwind {
; CHECK-LABEL: masked_load_v4i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ushll v0.4s, v0.4h, #0
; CHECK-NEXT:    ptrue p0.s, vl4
; CHECK-NEXT:    shl v0.4s, v0.4s, #31
; CHECK-NEXT:    cmlt v0.4s, v0.4s, #0
; CHECK-NEXT:    cmpne p0.s, p0/z, z0.s, #0
; CHECK-NEXT:    ldnt1w { z0.s }, p0/z, [x0]
; CHECK-NEXT:    // kill: def $q0 killed $q0 killed $z0
; CHECK-NEXT:    ret
  %load = call <4 x i32> @llvm.masked.load.v4i32(ptr %a, i32 1, <4 x i1> %mask, <4 x i32> undef), !nontemporal !0
  ret <4 x i32> %load
}

define void @masked_store_v4i32(<4 x i32> %x, ptr %a, <4 x i1> %mask) nounwind {
; CHECK-LABEL: masked_store_v4i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ushll v1.4s, v1.4h, #0
; CHECK-NEXT:    ptrue p0.s, vl4
; CHECK-NEXT:    // kill: def $q0 killed $q0 def $z0
; CHECK-NEXT:    shl v1.4s, v1.4s, #31
; CHECK-NEXT:    cmlt v1.4s, v1.4s, #0
; CHECK-NEXT:    cmpne p0.s, p0/z, z1.s, #0
; CHECK-NEXT:    stnt1w { z0.s }, p0, [x0]
; CHECK-NEXT:    ret
  call void @llvm.masked.store.v4i32.p0(<4 x i32> %x, ptr %a, i32 1, <4 x i1> %mask), !nontemporal !0
  ret void
}

define <4 x i32> @load_v4i32(ptr %a) nounwind {
; CHECK-LABEL: load_v4i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ldr q0, [x0]
; CHECK-NEXT:    ret
  %load = call <4 x i32> @llvm.masked.load.v4i32(ptr %a, i32 1, <4 x i1> <i1 1, i1 1, i1 1, i1 1>, <4 x i32> undef), !nontemporal !0
  ret <4 x i32> %load
}

define void @store_v4i32(<4 x i32> %x, ptr %a) nounwind {
; CHECK-LABEL: store_v4i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    mov d1, v0.d[1]
; CHECK-NEXT:    stnp d0, d1, [x0]
; CHECK-NEXT:    ret
  call void @llvm.masked.store.v4i32.p0(<4 x i32> %x, ptr %a, i32 1, <4 x i1> <i1 1, i1 1, i1 1, i1 1>), !nontemporal !0
  ret void
}

define <vscale x 4 x i32> @masked_load_nxv4i32(ptr %a, <vscale x 4 x i1> %mask) nounwind {
; CHECK-LABEL: masked_load_nxv4i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ldnt1w { z0.s }, p0/z, [x0]
; CHECK-NEXT:    ret
  %load = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32(ptr %a, i32 1, <vscale x 4 x i1> %mask, <vscale x 4 x i32> undef), !nontemporal !0
  ret <vscale x 4 x i32> %load
}

define void @masked_store_nxv4i32(<vscale x 4 x i32> %x, ptr %a, <vscale x 4 x i1> %mask) nounwind {
; CHECK-LABEL: masked_store_nxv4i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stnt1w { z0.s }, p0, [x0]
; CHECK-NEXT:    ret
  call void @llvm.masked.store.nxv4i32.p0(<vscale x 4 x i32> %x, ptr %a, i32 1, <vscale x 4 x i1> %mask), !nontemporal !0
  ret void
}

declare <vscale x 4 x i32> @llvm.masked.load.nxv4i32(ptr, i32, <vscale x 4 x i1>, <vscale x 4 x i32>)
declare void @llvm.masked.store.nxv4i32.p0(<vscale x 4 x i32>, ptr, i32, <vscale x 4 x i1>)
declare <4 x i32> @llvm.masked.load.v4i32(ptr, i32, <4 x i1>, <4 x i32>)
declare void @llvm.masked.store.v4i32.p0(<4 x i32>, ptr, i32, <4 x i1>)

!0 = !{i32 1}