diff options
author | Bruno Cardoso Lopes <bruno.cardoso@gmail.com> | 2011-09-15 18:27:36 +0000 |
---|---|---|
committer | Bruno Cardoso Lopes <bruno.cardoso@gmail.com> | 2011-09-15 18:27:36 +0000 |
commit | 0c4b9ff0775c8d15242f65dc4a7ab98f04bd5d68 (patch) | |
tree | 3fbcc7099f828a6bad99a757eee6e772320546d1 /test | |
parent | 41a9635292a069934f0674faca744118d4d25c5a (diff) | |
download | llvm-0c4b9ff0775c8d15242f65dc4a7ab98f04bd5d68.tar.gz llvm-0c4b9ff0775c8d15242f65dc4a7ab98f04bd5d68.tar.bz2 llvm-0c4b9ff0775c8d15242f65dc4a7ab98f04bd5d68.tar.xz |
Change all checks regarding the presence of any SSE level to always
take into consideration the presence of AVX. This change, together with
the SSEDomainFix enabled for AVX, makes AVX codegen to always (hopefully)
emit the same code as SSE for 128-bit vector ops. I don't
have a testcase for this, but AVX now beats SSE in performance for
128-bit ops in the majority of programas in the llvm testsuite
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139817 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test')
-rw-r--r-- | test/CodeGen/X86/avx-basic.ll | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/test/CodeGen/X86/avx-basic.ll b/test/CodeGen/X86/avx-basic.ll index edbdc06a1c..0a46b0828a 100644 --- a/test/CodeGen/X86/avx-basic.ll +++ b/test/CodeGen/X86/avx-basic.ll @@ -6,7 +6,7 @@ define void @zero128() nounwind ssp { entry: - ; CHECK: vxorps + ; CHECK: vpxor ; CHECK: vmovaps store <4 x float> zeroinitializer, <4 x float>* @z, align 16 ret void |