[PATCH] D71293: AMDGPU: Generate the correct sequence of code for FDIV32 when correctly-rounded-divide-sqrt is set

[Changpeng Fang via Phabricator via llvm-commits]

cfang updated this revision to Diff 239207.
cfang marked an inline comment as done.
cfang added a comment.
Herald added a subscriber: kerbowa.

Implement rcp optimization for fdiv in AMGGPUCodegenPrepare to insert amdgcn_rcp intrinsic. For f32 type fdiv,
if fpmath metadata is unavailable, we could not do rcp optimization unless fast unsafe math is specified.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D71293/new/

https://reviews.llvm.org/D71293

Files:
  llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp
  llvm/lib/Target/AMDGPU/SIISelLowering.cpp
  llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-fdiv.ll
  llvm/test/CodeGen/AMDGPU/fdiv.ll
  llvm/test/CodeGen/AMDGPU/fdiv32-to-rcp-folding.ll
  llvm/test/CodeGen/AMDGPU/fneg-combines.ll
  llvm/test/CodeGen/AMDGPU/known-never-snan.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.rcp.ll
  llvm/test/CodeGen/AMDGPU/mul24-pass-ordering.ll
  llvm/test/CodeGen/AMDGPU/rcp-pattern.ll
  llvm/test/CodeGen/AMDGPU/rcp_iflag.ll
  llvm/test/CodeGen/AMDGPU/rsq.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D71293.239207.patch
Type: text/x-patch
Size: 48965 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200121/c8f33054/attachment.bin>