diff options
| author | Nadav Rotem <nadav.rotem@intel.com> | 2012-04-10 14:33:13 +0000 |
|---|---|---|
| committer | Nadav Rotem <nadav.rotem@intel.com> | 2012-04-10 14:33:13 +0000 |
| commit | 50e64cfe6e250dbe2528fc5bda75c68b04a8bc49 (patch) | |
| tree | 11f8711c3c0c8a1418507063bde7ac8b81a7826f /test/CodeGen/X86/vec_shuffle-20.ll | |
| parent | 45fb79bc54159330979bf24e4bfbdbb64bee1e2c (diff) | |
Modify the code that lowers shuffles to blends from using blendvXX to vblendXX.
blendv uses a register for the selection while vblend uses an immediate.
On sandybridge they still have the same latency and execute on the same execution ports.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154396 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/vec_shuffle-20.ll')
| -rw-r--r-- | test/CodeGen/X86/vec_shuffle-20.ll | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/test/CodeGen/X86/vec_shuffle-20.ll b/test/CodeGen/X86/vec_shuffle-20.ll index fc06b9514e..107411e351 100644 --- a/test/CodeGen/X86/vec_shuffle-20.ll +++ b/test/CodeGen/X86/vec_shuffle-20.ll @@ -1,4 +1,4 @@ -; RUN: llc < %s -o /dev/null -march=x86 -mattr=+sse2 -mtriple=i686-apple-darwin9 -stats -info-output-file - | grep asm-printer | grep 3 +; RUN: llc < %s -o /dev/null -march=x86 -mattr=+sse2 -mtriple=i686-apple-darwin9 -stats -info-output-file - | grep asm-printer | grep 2 define <4 x float> @func(<4 x float> %fp0, <4 x float> %fp1) nounwind { entry: |