[PATCH] D75889: AMDGPU: Use V_MAC_F32 for fmad.ftz

[Matt Arsenault via Phabricator via llvm-commits]

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

This avoids regressions in a future patch. I'm confused by the use of
the gfx9 usage legacy_mad. Was this a pointless instruction rename, or
uses fmul_legacy handling? Why is regular mac avilable in that case?


https://reviews.llvm.org/D75889

Files:
  llvm/lib/Target/AMDGPU/SIInstructions.td
  llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-amdgcn.fmad.ftz.mir
  llvm/test/CodeGen/AMDGPU/fcanonicalize-elimination.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.fmad.ftz.f16.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.fmad.ftz.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D75889.249268.patch
Type: text/x-patch
Size: 9790 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200310/73d119e8/attachment.bin>