[PATCH] D51347: AMDGPU: Force shrinking of add/sub even if the carry is used
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Aug 28 04:56:29 PDT 2018
arsenm created this revision.
arsenm added a reviewer: rampitec.
Herald added subscribers: t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl.
The original motivating example uses a 64-bit add, so the carry
is used. Insert a copy from VCC. This may allow shrinking of
the used carry instruction. At worst, we are replacing a
mov to materialize the constant with a copy of vcc.
https://reviews.llvm.org/D51347
Files:
lib/Target/AMDGPU/SIFoldOperands.cpp
test/CodeGen/AMDGPU/fold-immediate-operand-shrink-with-carry.mir
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D51347.162829.patch
Type: text/x-patch
Size: 4964 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180828/be485b19/attachment.bin>
More information about the llvm-commits
mailing list