llvm/llvm/test/Transforms/MemCpyOpt/memset-memcpy-to-2x-memset.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -passes=memcpyopt -S %s -verify-memoryssa | FileCheck %s

target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"

define void @test(ptr %dst1, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST1:%.*]], i8 [[C:%.*]], i64 128, i1 false)
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr align 8 [[DST2:%.*]], i8 [[C]], i64 128, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr %dst1, i8 %c, i64 128, i1 false)
  call void @llvm.memcpy.p0.p0.i64(ptr align 8 %dst2, ptr align 8 %dst1, i64 128, i1 false)
  ret void
}

define void @test_smaller_memcpy(ptr %dst1, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test_smaller_memcpy(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST1:%.*]], i8 [[C:%.*]], i64 128, i1 false)
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST2:%.*]], i8 [[C]], i64 100, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr %dst1, i8 %c, i64 128, i1 false)
  call void @llvm.memcpy.p0.p0.i64(ptr %dst2, ptr %dst1, i64 100, i1 false)
  ret void
}

define void @test_smaller_memset(ptr %dst1, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test_smaller_memset(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST1:%.*]], i8 [[C:%.*]], i64 100, i1 false)
; CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr [[DST2:%.*]], ptr [[DST1]], i64 128, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr %dst1, i8 %c, i64 100, i1 false)
  call void @llvm.memcpy.p0.p0.i64(ptr %dst2, ptr %dst1, i64 128, i1 false)
  ret void
}

define void @test_align_memset(ptr %dst1, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test_align_memset(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr align 8 [[DST1:%.*]], i8 [[C:%.*]], i64 128, i1 false)
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST2:%.*]], i8 [[C]], i64 128, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr align 8 %dst1, i8 %c, i64 128, i1 false)
  call void @llvm.memcpy.p0.p0.i64(ptr %dst2, ptr %dst1, i64 128, i1 false)
  ret void
}

define void @test_different_types(ptr %dst1, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test_different_types(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr align 8 [[DST1:%.*]], i8 [[C:%.*]], i64 128, i1 false)
; CHECK-NEXT:    call void @llvm.memset.p0.i32(ptr [[DST2:%.*]], i8 [[C]], i32 100, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr align 8 %dst1, i8 %c, i64 128, i1 false)
  call void @llvm.memcpy.p0.p0.i32(ptr %dst2, ptr %dst1, i32 100, i1 false)
  ret void
}

define void @test_different_types_2(ptr %dst1, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test_different_types_2(
; CHECK-NEXT:    call void @llvm.memset.p0.i32(ptr align 8 [[DST1:%.*]], i8 [[C:%.*]], i32 128, i1 false)
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST2:%.*]], i8 [[C]], i64 100, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i32(ptr align 8 %dst1, i8 %c, i32 128, i1 false)
  call void @llvm.memcpy.p0.p0.i64(ptr %dst2, ptr %dst1, i64 100, i1 false)
  ret void
}

define void @test_different_source_gep(ptr %dst1, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test_different_source_gep(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST1:%.*]], i8 [[C:%.*]], i64 128, i1 false)
; CHECK-NEXT:    [[P:%.*]] = getelementptr i8, ptr [[DST1]], i64 64
; CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr [[DST2:%.*]], ptr [[P]], i64 64, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr %dst1, i8 %c, i64 128, i1 false)
  ; FIXME: We could optimize this as well.
  %p = getelementptr i8, ptr %dst1, i64 64
  call void @llvm.memcpy.p0.p0.i64(ptr %dst2, ptr %p, i64 64, i1 false)
  ret void
}

define void @test_variable_size_1(ptr %dst1, i64 %dst1_size, ptr %dst2, i8 %c) {
; CHECK-LABEL: @test_variable_size_1(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST1:%.*]], i8 [[C:%.*]], i64 [[DST1_SIZE:%.*]], i1 false)
; CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr [[DST2:%.*]], ptr [[DST1]], i64 128, i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr %dst1, i8 %c, i64 %dst1_size, i1 false)
  call void @llvm.memcpy.p0.p0.i64(ptr %dst2, ptr %dst1, i64 128, i1 false)
  ret void
}

define void @test_variable_size_2(ptr %dst1, ptr %dst2, i64 %dst2_size, i8 %c) {
; CHECK-LABEL: @test_variable_size_2(
; CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr [[DST1:%.*]], i8 [[C:%.*]], i64 128, i1 false)
; CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr [[DST2:%.*]], ptr [[DST1]], i64 [[DST2_SIZE:%.*]], i1 false)
; CHECK-NEXT:    ret void
;
  call void @llvm.memset.p0.i64(ptr %dst1, i8 %c, i64 128, i1 false)
  call void @llvm.memcpy.p0.p0.i64(ptr %dst2, ptr %dst1, i64 %dst2_size, i1 false)
  ret void
}

declare void @llvm.memset.p0.i64(ptr nocapture, i8, i64, i1)
declare void @llvm.memcpy.p0.p0.i64(ptr nocapture, ptr nocapture readonly, i64, i1)
declare void @llvm.memset.p0.i32(ptr nocapture, i8, i32, i1)
declare void @llvm.memcpy.p0.p0.i32(ptr nocapture, ptr nocapture readonly, i32, i1)