[PATCH] D104316: [AMDGPU] Propagate LDS align into to instructions

[Jay Foad via Phabricator via llvm-commits]

foad added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPULowerModuleLDSPass.cpp:340
+      return;
+
+    for (User *U : Ptr->users()) {
----------------
Could early-out here if Align==1 to save a lot of unnecessary work.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPULowerModuleLDSPass.cpp:346
+      }
+      if (auto *SI = dyn_cast<StoreInst>(U)) {
+        SI->setAlignment(std::max(A, SI->getAlign()));
----------------
foad wrote:
> For StoreInst, AtomicRMWInst, AtomicCmpXchgInst and GetElementPtrInst you need to check that the use is actually the "address" operand of the instruction.
Oh, I see you are already checking it for GetElementPtrInst.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D104316/new/

https://reviews.llvm.org/D104316