[PATCH] D57708: AMDGPU: Don't rematerialize mov with implicit operands
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Feb 4 13:02:22 PST 2019
arsenm created this revision.
arsenm added reviewers: rampitec, cfang, scott.linder.
Herald added subscribers: t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl.
This was pulling the mov used for register indexing on gfx9 out of the loop.
https://reviews.llvm.org/D57708
Files:
lib/Target/AMDGPU/SIInstrInfo.cpp
test/CodeGen/AMDGPU/no-remat-indirect-mov.mir
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D57708.185138.patch
Type: text/x-patch
Size: 6661 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190204/f4427fbb/attachment.bin>
More information about the llvm-commits
mailing list