[PATCH] D83919: [AMDGPU][CostModel] Improve cost estimation for fused {fadd|fsub}(a,fmul(b,c))

[Daniil Fukalov via Phabricator via llvm-commits]

dfukalov created this revision.
dfukalov added a reviewer: rampitec.
Herald added subscribers: kerbowa, hiraditya, t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl, arsenm.
Herald added a project: LLVM.

If result of fmul(b,c) has one use, in almost all cases (except denormals are
IEEE) the pair of operations will be fused in one fma/mad/mac/etc.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D83919

Files:
  llvm/include/llvm/CodeGen/BasicTTIImpl.h
  llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
  llvm/test/Analysis/CostModel/AMDGPU/fused_costs.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D83919.278335.patch
Type: text/x-patch
Size: 5471 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200715/767e409e/attachment.bin>