[llvm] [AMDGPU] Enable more consecutive load folding during aggressive-instcombine (PR #158036)
Matt Arsenault via llvm-commits
llvm-commits at lists.llvm.org
Fri Sep 12 02:08:49 PDT 2025
================
@@ -2090,10 +2090,16 @@ bool SITargetLowering::allowsMisalignedMemoryAccessesImpl(
if (AddrSpace == AMDGPUAS::PRIVATE_ADDRESS ||
AddrSpace == AMDGPUAS::FLAT_ADDRESS) {
bool AlignedBy4 = Alignment >= Align(4);
+ if (Subtarget->hasUnalignedScratchAccessEnabled()) {
----------------
arsenm wrote:
Can you change the title to describe this is changing reporting of fast unaligned access, and mention the consequence in the longer description
https://github.com/llvm/llvm-project/pull/158036
More information about the llvm-commits
mailing list