[PATCH] D127391: [AMDGPU] Update SIInsertHardClauses for GFX11

[Jay Foad via Phabricator via llvm-commits]

foad created this revision.
foad added reviewers: Joe_Nash, rampitec, arsenm, piotr.
Herald added subscribers: kosarev, jsilvanus, hsmhsm, kerbowa, hiraditya, t-tye, tpr, dstuttard, yaxunl, nhaehnle, jvesely, kzhuravl.
Herald added a project: All.
foad requested review of this revision.
Herald added subscribers: llvm-commits, wdng.
Herald added a project: LLVM.

Changes for GFX11:

- Clauses may not mix instructions of different types, and there are more types. For example image instructions with and without a sampler are now different types.
- The max size of a clause is explicitly documented as 63 instructions. Previously it was implicitly assumed to be 64. This is such a tiny difference that it does not seem worth making it conditional on the subtarget.
- It can be beneficial to clause stores as well as loads.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D127391

Files:
  llvm/lib/Target/AMDGPU/SIInsertHardClauses.cpp
  llvm/test/CodeGen/AMDGPU/hard-clauses.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D127391.435489.patch
Type: text/x-patch
Size: 52543 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220609/6e6bca5c/attachment.bin>