[PATCH] D81070: [AMDGPU] Fixup use of StackPtrOffsetReg when not initialized
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Jun 4 14:23:47 PDT 2020
arsenm added a comment.
A small MIR test would also be helpful
================
Comment at: llvm/lib/Target/AMDGPU/SIFrameLowering.cpp:460
if (MF.getFrameInfo().hasCalls()) {
+ // Note: Spilling code in SIRegisterInfo.cpp assumes that if !hasCalls() it
----------------
hasCalls seems like the wrong check here. Should there be a stackPointerIsTracked() or something to go along with hasFP?
================
Comment at: llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp:741
MCRegister SOffset = ScratchOffsetReg;
+ bool UninitStackPtrOffset = false;
----------------
This shouldn't need to be an explicit property?
================
Comment at: llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp:785
ScratchOffsetReg = FuncInfo->getStackPtrOffsetReg();
+ UninitStackPtrOffset = !MFI.hasCalls();
+ }
----------------
Check the theoretical stackPointerIsTracked instead?
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D81070/new/
https://reviews.llvm.org/D81070
More information about the llvm-commits
mailing list