[PATCH] D152706: [AMDGPU] Use SSAUpdater in PromoteAlloca

[Pierre van Houtryve via Phabricator via llvm-commits]

Pierre-vh updated this revision to Diff 535656.
Pierre-vh added a comment.
This revision is now accepted and ready to land.

Fix openmp tests

In the end I didn't need to do anything complicated with lanes and such. I just had to do the same system as LoadAndStoreUpdater, which is to defer live-in loads lowering to a second pass.
Now, some dummy loads are also inserted on the first pass for instructions that must be promoted on the first pass but also need to know the current vector value.

It leads to less folding than I'd like, but it's a lot more correct.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D152706/new/

https://reviews.llvm.org/D152706

Files:
  llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
  llvm/test/CodeGen/AMDGPU/fix-frame-reg-in-custom-csr-spills.ll
  llvm/test/CodeGen/AMDGPU/promote-alloca-array-aggregate.ll
  llvm/test/CodeGen/AMDGPU/promote-alloca-globals.ll
  llvm/test/CodeGen/AMDGPU/promote-alloca-loadstores.ll
  llvm/test/CodeGen/AMDGPU/promote-alloca-memset.ll
  llvm/test/CodeGen/AMDGPU/promote-alloca-pointer-array.ll
  llvm/test/CodeGen/AMDGPU/promote-alloca-vector-to-vector.ll
  llvm/test/CodeGen/AMDGPU/sroa-before-unroll.ll
  llvm/test/CodeGen/AMDGPU/vector-alloca-bitcast.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D152706.535656.patch
Type: text/x-patch
Size: 66835 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230629/88d3cd31/attachment.bin>