[PATCH] D127977: [AMDGPU] Tag GFX11 LDS loads as using strict_wqm

[Jay Foad via Phabricator via llvm-commits]

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

LDS_PARAM_LOAD and LDS_DIRECT_LOAD use EXEC per quad
(if any pixel is enabled in the quad, data is written
to all 4 pixels/threads in the quad).

Tag LDS_PARAM_LOAD and LDS_DIRECT_LOAD as using strict_wqm
to enforce this and avoid lane clobbering issues.
Note that only the instruction itself is tagged.
The implicit uses of these do not need to be set WQM.
The reduces unnecessary WQM calculation of M0.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D127977

Files:
  llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
  llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.interp.inreg.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.interp.inreg.ll
  llvm/test/CodeGen/AMDGPU/wqm-gfx11.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D127977.437555.patch
Type: text/x-patch
Size: 14922 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220616/a7b6c89c/attachment-0001.bin>