[PATCH] D64393: [AMDGPU] Fix DPP combiner check for exec modification
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Jul 12 07:14:36 PDT 2019
arsenm added inline comments.
================
Comment at: llvm/lib/Target/AMDGPU/SIInstrInfo.cpp:6159
+ }
+
if (I->modifiesRegister(AMDGPU::EXEC, TRI))
----------------
foad wrote:
> vpykhtin wrote:
> > foad wrote:
> > > vpykhtin wrote:
> > > > vpykhtin wrote:
> > > > > Now I recall why I used the set originally - to detect use instruction fast. Matt, is there other way to detect whether the instruction use a register without scanning its operands?
> > > > and without scanning whether an instruction is in the use list for a register.
> > > We have to scan the operands of every instruction anyway, in order to test `I->modifiesRegister(AMDGPU::EXEC, TRI)`. Your original code did that too.
> > Good point. BTW you can probably use readsRegister instead of the loop.
> `readsRegister` doesn't appear to ignore debug operands.
debug uses should only appear on the debug pseudo-instructions, which you already skipped with the ->isDebugInstr check
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D64393/new/
https://reviews.llvm.org/D64393
More information about the llvm-commits
mailing list