[PATCH] D101405: [AMDGPU] Change FLAT SADDR to VADDR form in moveToVALU
Joe Nash via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Apr 28 09:23:50 PDT 2021
Joe_Nash added inline comments.
================
Comment at: llvm/lib/Target/AMDGPU/SIInstrInfo.cpp:5037
+ int OldVAddrIdx = AMDGPU::getNamedOperandIdx(Opc, AMDGPU::OpName::vaddr);
+ assert(OldVAddrIdx >= 0);
+
----------------
rampitec wrote:
> Joe_Nash wrote:
> > I don't understand why this assert is true. Does some previous check guarantee that? Other than that LGTM, but please wait for @arsenm
> All flat global SADDR instructions have $vaddr component and other instructions cannot get to this point as getGlobalVaddrOp() will return -1. This changes in the followup D101408 which can also process flat scratch.
I see it now, thanks.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D101405/new/
https://reviews.llvm.org/D101405
More information about the llvm-commits
mailing list