[PATCH] D151126: [AMDGPU] Don't flush vmcnt for loops with use/def pairs

[Austin Kerbow via Phabricator via llvm-commits]

kerbowa created this revision.
kerbowa added reviewers: bsaleil, rochauha, foad, nhaehnle.
Herald added subscribers: StephenFan, hiraditya, tpr, dstuttard, yaxunl, jvesely, kzhuravl, arsenm.
Herald added a project: All.
kerbowa requested review of this revision.
Herald added subscribers: llvm-commits, wdng.
Herald added a project: LLVM.

Conditions for hoisting vmcnt with flat instructions should be similar to VMEM.
If there are use/def pairs in a loop body we cannot guarantee that hosting the
waitcnt will be profitable. Better heuristics are needed to analyse whether
gains from avoiding waitcnt in loop bodys outweighs waiting for loads in the
preheader.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D151126

Files:
  llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
  llvm/test/CodeGen/AMDGPU/GlobalISel/fp64-atomics-gfx90a.ll
  llvm/test/CodeGen/AMDGPU/atomicrmw-expand.ll
  llvm/test/CodeGen/AMDGPU/atomicrmw-nand.ll
  llvm/test/CodeGen/AMDGPU/flat_atomics_i64_min_max_system.ll
  llvm/test/CodeGen/AMDGPU/flat_atomics_min_max_system.ll
  llvm/test/CodeGen/AMDGPU/fp64-atomics-gfx90a.ll
  llvm/test/CodeGen/AMDGPU/global-load-saddr-to-vaddr.ll
  llvm/test/CodeGen/AMDGPU/global-saddr-atomics-min-max-system.ll
  llvm/test/CodeGen/AMDGPU/waitcnt-vmcnt-loop.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D151126.524392.patch
Type: text/x-patch
Size: 143806 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230522/40693b3e/attachment.bin>