fixed to test features, not CPU model
authorSanjay Patel <spatel@rotateright.com>
Wed, 18 Mar 2015 16:07:10 +0000 (16:07 +0000)
committerSanjay Patel <spatel@rotateright.com>
Wed, 18 Mar 2015 16:07:10 +0000 (16:07 +0000)
commit4795cb202c61e76090abac711b97e58362991abe
tree448b09bd99a8a713c2886327cdb6dc9f4acbe3e8
parentd5cb4a90e56af1eb1cd41a33b1c1c421bd88dbca
fixed to test features, not CPU model

The 'vmovntdq' was only passing due to a fluke in
SandyBridge codegen that splits 32-byte stores in half,
but that meant that the test was not correctly checking
for the 32-byte store that we thought we were generating.

The lax checking in this file will be addressed in
another commit. There are bigger problems here.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232644 91177308-0d34-0410-b5e6-96231b3b80d8
test/CodeGen/X86/avx-intrinsics-x86.ll