X-Git-Url: http://plrg.eecs.uci.edu/git/?a=blobdiff_plain;f=test%2FCodeGen%2FX86%2Fsmall-byval-memcpy.ll;h=3c03750199cbe077a6e0b125c4c53138ce454f28;hb=915b8a9ffe3935860c662975477a63db2c2ad3a8;hp=9ec9182e5e3c209e3d93bc1c75f012c5c0549c1f;hpb=36a0947820fd4aa4b8a5fa26e3f079bdf572bc81;p=oota-llvm.git diff --git a/test/CodeGen/X86/small-byval-memcpy.ll b/test/CodeGen/X86/small-byval-memcpy.ll index 9ec9182e5e3..3c03750199c 100644 --- a/test/CodeGen/X86/small-byval-memcpy.ll +++ b/test/CodeGen/X86/small-byval-memcpy.ll @@ -1,22 +1,25 @@ -; RUN: llc < %s | not grep movs +; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core2 | FileCheck %s --check-prefix=CORE2 +; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=nehalem | FileCheck %s --check-prefix=NEHALEM +; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=btver2 | FileCheck %s --check-prefix=BTVER2 -target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128" -target triple = "i386-apple-darwin8" +declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) -define void @ccosl({ x86_fp80, x86_fp80 }* noalias sret %agg.result, { x86_fp80, x86_fp80 }* byval align 4 %z) nounwind { -entry: - %iz = alloca { x86_fp80, x86_fp80 } ; <{ x86_fp80, x86_fp80 }*> [#uses=3] - %tmp1 = getelementptr { x86_fp80, x86_fp80 }* %z, i32 0, i32 1 ; [#uses=1] - %tmp2 = load x86_fp80* %tmp1, align 16 ; [#uses=1] - %tmp3 = fsub x86_fp80 0xK80000000000000000000, %tmp2 ; [#uses=1] - %tmp4 = getelementptr { x86_fp80, x86_fp80 }* %iz, i32 0, i32 1 ; [#uses=1] - %real = getelementptr { x86_fp80, x86_fp80 }* %iz, i32 0, i32 0 ; [#uses=1] - %tmp6 = getelementptr { x86_fp80, x86_fp80 }* %z, i32 0, i32 0 ; [#uses=1] - %tmp7 = load x86_fp80* %tmp6, align 16 ; [#uses=1] - store x86_fp80 %tmp3, x86_fp80* %real, align 16 - store x86_fp80 %tmp7, x86_fp80* %tmp4, align 16 - call void @ccoshl( { x86_fp80, x86_fp80 }* noalias sret %agg.result, { x86_fp80, x86_fp80 }* byval align 4 %iz ) nounwind - ret void -} +define void @copy16bytes(i8* nocapture %a, i8* nocapture readonly %b) { + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 16, i32 1, i1 false) + ret void + + ; CHECK-LABEL: copy16bytes + ; CORE2: movq + ; CORE2-NEXT: movq + ; CORE2-NEXT: movq + ; CORE2-NEXT: movq + ; CORE2-NEXT: retq -declare void @ccoshl({ x86_fp80, x86_fp80 }* noalias sret , { x86_fp80, x86_fp80 }* byval align 4 ) nounwind + ; NEHALEM: movups + ; NEHALEM-NEXT: movups + ; NEHALEM-NEXT: retq + + ; BTVER2: movups + ; BTVER2-NEXT: movups + ; BTVER2-NEXT: retq +}