[PATCH] D152749: AMDGPU: Correct constants used in fast math log expansion

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: AMDGPU, foad, rampitec, Pierre-vh, cdevadas.
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.

The division between float constants was done with less
precision. Performing the divide in double and truncating to float
provides the same value as used in the library fast math expansion.


https://reviews.llvm.org/D152749

Files:
  llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
  llvm/test/CodeGen/AMDGPU/llvm.log10.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D152749.530655.patch
Type: text/x-patch
Size: 47056 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230612/92e1fdc0/attachment.bin>