[PATCH] D149773: [AMDGPU][IGLP] Add iglp_opt(1) strategy for single wave gemms

[Jeffrey Byrnes via Phabricator via llvm-commits]

jrbyrnes updated this revision to Diff 530954.
jrbyrnes marked 2 inline comments as done.
jrbyrnes added a comment.

Like you, I am mindful of the compile time costs, but not overly concerned since it is only relevant to a few blocks.

That said, the cache idea is a cheap way to get some compile time improvements, so I've implemented that feature.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D149773/new/

https://reviews.llvm.org/D149773

Files:
  llvm/lib/Target/AMDGPU/AMDGPUIGroupLP.cpp
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.iglp.opt.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.iglp.opt.single.2b.mir
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.iglp.opt.single.2c.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D149773.530954.patch
Type: text/x-patch
Size: 88048 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230613/b99b94e5/attachment.bin>