Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100751 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/test/CodeGen/X86/memcpy-2.ll b/test/CodeGen/X86/memcpy-2.ll
index 4fe32d9..17cd8e8 100644
--- a/test/CodeGen/X86/memcpy-2.ll
+++ b/test/CodeGen/X86/memcpy-2.ll
@@ -3,20 +3,20 @@
; RUN: llc < %s -mattr=-sse -mtriple=i686-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=NOSSE
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=X86-64
- %struct.ParmT = type { [25 x i8], i8, i8* }
-@.str12 = internal constant [25 x i8] c"image\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00" ; <[25 x i8]*> [#uses=1]
+@.str = internal constant [25 x i8] c"image\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00"
+@.str2 = internal constant [30 x i8] c"xxxxxxxxxxxxxxxxxxxxxxxxxxxxx\00", align 4
define void @t1(i32 %argc, i8** %argv) nounwind {
entry:
; SSE2: t1:
-; SSE2: movaps _.str12, %xmm0
+; SSE2: movaps _.str, %xmm0
; SSE2: movaps %xmm0
; SSE2: movb $0
; SSE2: movl $0
; SSE2: movl $0
; SSE1: t1:
-; SSE1: movaps _.str12, %xmm0
+; SSE1: movaps _.str, %xmm0
; SSE1: movaps %xmm0
; SSE1: movb $0
; SSE1: movl $0
@@ -32,14 +32,14 @@
; NOSSE: movl $1734438249
; X86-64: t1:
-; X86-64: movaps _.str12(%rip), %xmm0
+; X86-64: movaps _.str(%rip), %xmm0
; X86-64: movaps %xmm0
; X86-64: movb $0
; X86-64: movq $0
- %parms.i = alloca [13 x %struct.ParmT] ; <[13 x %struct.ParmT]*> [#uses=1]
- %parms1.i = getelementptr [13 x %struct.ParmT]* %parms.i, i32 0, i32 0, i32 0, i32 0 ; <i8*> [#uses=1]
- call void @llvm.memcpy.i32( i8* %parms1.i, i8* getelementptr ([25 x i8]* @.str12, i32 0, i32 0), i32 25, i32 1 ) nounwind
- unreachable
+ %tmp1 = alloca [25 x i8]
+ %tmp2 = bitcast [25 x i8]* %tmp1 to i8*
+ call void @llvm.memcpy.i32( i8* %tmp2, i8* getelementptr ([25 x i8]* @.str, i32 0, i32 0), i32 25, i32 1 ) nounwind
+ unreachable
}
;rdar://7774704
@@ -119,4 +119,49 @@
ret void
}
+define void @t4() nounwind {
+entry:
+; SSE2: t4:
+; SSE2: movw $120
+; SSE2: movl $2021161080
+; SSE2: movl $2021161080
+; SSE2: movl $2021161080
+; SSE2: movl $2021161080
+; SSE2: movl $2021161080
+; SSE2: movl $2021161080
+; SSE2: movl $2021161080
+
+; SSE1: t4:
+; SSE1: movw $120
+; SSE1: movl $2021161080
+; SSE1: movl $2021161080
+; SSE1: movl $2021161080
+; SSE1: movl $2021161080
+; SSE1: movl $2021161080
+; SSE1: movl $2021161080
+; SSE1: movl $2021161080
+
+; NOSSE: t4:
+; NOSSE: movw $120
+; NOSSE: movl $2021161080
+; NOSSE: movl $2021161080
+; NOSSE: movl $2021161080
+; NOSSE: movl $2021161080
+; NOSSE: movl $2021161080
+; NOSSE: movl $2021161080
+; NOSSE: movl $2021161080
+
+; X86-64: t4:
+; X86-64: movabsq $8680820740569200760, %rax
+; X86-64: movq %rax
+; X86-64: movq %rax
+; X86-64: movq %rax
+; X86-64: movw $120
+; X86-64: movl $2021161080
+ %tmp1 = alloca [30 x i8]
+ %tmp2 = bitcast [30 x i8]* %tmp1 to i8*
+ call void @llvm.memcpy.i32(i8* %tmp2, i8* getelementptr inbounds ([30 x i8]* @.str2, i32 0, i32 0), i32 30, i32 1)
+ unreachable
+}
+
declare void @llvm.memcpy.i32(i8* nocapture, i8* nocapture, i32, i32) nounwind