[PATCH] D153258: AMDGPU: Optimize set_rounding if input is known to fit in 2 bits

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: AMDGPU, foad, rampitec, jhuber6, Pierre-vh, b-sumner.
Herald added subscribers: StephenFan, kerbowa, hiraditya, tpr, dstuttard, yaxunl, jvesely, kzhuravl.
Herald added a project: All.
arsenm requested review of this revision.
Herald added a subscriber: wdng.
Herald added a project: LLVM.

We don't need to figure out the weird extended rounding modes or
handle offsets to keep the lookup table in 64-bits.


https://reviews.llvm.org/D153258

Files:
  llvm/lib/Target/AMDGPU/SIISelLowering.cpp
  llvm/lib/Target/AMDGPU/SIModeRegisterDefaults.cpp
  llvm/lib/Target/AMDGPU/SIModeRegisterDefaults.h
  llvm/test/CodeGen/AMDGPU/llvm.set.rounding.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D153258.532587.patch
Type: text/x-patch
Size: 33819 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230619/515b6311/attachment.bin>