[PATCH] D51541: AMDGPU: Restrict extract_vector_elt combine to loads

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added a reviewer: rampitec.
Herald added subscribers: t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl.

The intention is to enable the extract_vector_elt load combine,
and doing this for other operations interferes with more
useful optimizations on vectors.

      

Handle any type of load since in principle we should do the
same combine for the various load intrinsics.


https://reviews.llvm.org/D51541

Files:
  lib/Target/AMDGPU/SIISelLowering.cpp
  test/CodeGen/AMDGPU/fexp.ll
  test/CodeGen/AMDGPU/fmax_legacy.f16.ll
  test/CodeGen/AMDGPU/fmin_legacy.f16.ll
  test/CodeGen/AMDGPU/reduction.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D51541.163530.patch
Type: text/x-patch
Size: 32527 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180831/23992083/attachment.bin>