RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-03-07 11:59:09 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

add a note

Browse Source 
llvm-svn: 38507
This commit is contained in:
Chris Lattner 2007-07-10 20:03:50 +00:00
parent c0d9376230
commit abcdeef610
1 changed files with 26 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
lib/Target/X86/README-SSE.txt
View File

@ -572,3 +572,29 @@ swizzle:
ret
//===---------------------------------------------------------------------===//
This code:
#include <emmintrin.h>
__m128i test(long long i) { return _mm_cvtsi64x_si128(i); }
Should turn into a single 'movq %rdi, %xmm0' instruction. Instead, we
get this (on x86-64):
_test:
movd %rdi, %xmm1
xorps %xmm0, %xmm0
movsd %xmm1, %xmm0
ret
The LLVM IR is:
target triple = "x86_64-apple-darwin8"
define <2 x i64> @test(i64 %i) {
entry:
%tmp10 = insertelement <2 x i64> undef, i64 %i, i32 0
%tmp11 = insertelement <2 x i64> %tmp10, i64 0, i32 1
ret <2 x i64> %tmp11
}
//===---------------------------------------------------------------------===//