[llvm-commits] [llvm] r51062 - /llvm/trunk/lib/Target/X86/README-SSE.txt

[Chris Lattner]

Author: lattner
Date: Tue May 13 14:56:20 2008
New Revision: 51062

URL: http://llvm.org/viewvc/llvm-project?rev=51062&view=rev
Log:
add a note

Modified:
    llvm/trunk/lib/Target/X86/README-SSE.txt

Modified: llvm/trunk/lib/Target/X86/README-SSE.txt
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/README-SSE.txt?rev=51062&r1=51061&r2=51062&view=diff

==============================================================================
--- llvm/trunk/lib/Target/X86/README-SSE.txt (original)
+++ llvm/trunk/lib/Target/X86/README-SSE.txt Tue May 13 14:56:20 2008
@@ -787,5 +787,23 @@
 
 In sse4 mode, we could use insertps to make both better.
 
+Here's another testcase that could use insertps [mem]:
+
+#include <xmmintrin.h>
+extern float x2, x3;
+__m128 foo1 (float x1, float x4) {
+ return _mm_set_ps (x2, x1, x3, x4);
+}
+
+gcc mainline compiles it to:
+
+foo1:
+       insertps        $0x10, x2(%rip), %xmm0
+       insertps        $0x10, x3(%rip), %xmm1
+       movaps  %xmm1, %xmm2
+       movlhps %xmm0, %xmm2
+       movaps  %xmm2, %xmm0
+       ret
+
 //===---------------------------------------------------------------------===//