[PATCH] [X86][AVX] 256-bit vector stack unaligned load/stores identification

[Simon Pilgrim]

Hi qcolombet, andreadb, spatel,

Under most circumstances the stack is not 32-byte aligned, resulting in the use of the vmovups/vmovupd/vmovdqu instructions when inserting ymm reloads/spills.

This minor patch adds these instructions to the isFrameLoadOpcode/isFrameStoreOpcode helpers so that these ops can be correctly identified and not be treated as folded reloads/spills.

This has also been noticed by http://llvm.org/bugs/show_bug.cgi?id=18846

http://reviews.llvm.org/D6252

Files:
  lib/Target/X86/X86InstrInfo.cpp
  test/CodeGen/X86/avx-intel-ocl.ll
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D6252.16163.patch
Type: text/x-patch
Size: 3763 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20141113/a25b9155/attachment.bin>