diff options
author | Evan Cheng <evan.cheng@apple.com> | 2011-01-07 19:35:30 +0000 |
---|---|---|
committer | Evan Cheng <evan.cheng@apple.com> | 2011-01-07 19:35:30 +0000 |
commit | a5e1362f968568d66d76ddcdcff4ab98e203a48c (patch) | |
tree | 53e266c315432b49be8ad6f3a2d2a5873265ab53 /test/CodeGen/X86/memset64-on-x86-32.ll | |
parent | 1434f66b2e132a707e2c8ccb3350ea13fb5aa051 (diff) | |
download | llvm-a5e1362f968568d66d76ddcdcff4ab98e203a48c.tar.gz llvm-a5e1362f968568d66d76ddcdcff4ab98e203a48c.tar.bz2 llvm-a5e1362f968568d66d76ddcdcff4ab98e203a48c.tar.xz |
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123015 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/memset64-on-x86-32.ll')
-rw-r--r-- | test/CodeGen/X86/memset64-on-x86-32.ll | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/test/CodeGen/X86/memset64-on-x86-32.ll b/test/CodeGen/X86/memset64-on-x86-32.ll index 5a0e893e3b..3f069b4a1a 100644 --- a/test/CodeGen/X86/memset64-on-x86-32.ll +++ b/test/CodeGen/X86/memset64-on-x86-32.ll @@ -1,5 +1,6 @@ ; RUN: llc < %s -mtriple=i386-apple-darwin -mcpu=nehalem | grep movups | count 5 -; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core2 | grep movups | count 5 +; RUN: llc < %s -mtriple=i386-apple-darwin -mcpu=core2 | grep movl | count 20 +; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core2 | grep movq | count 10 define void @bork() nounwind { entry: |