[PATCH] D126158: [MLIR][GPU] Replace fdiv on fp16 with promoted (fp32) multiplication with reciprocal plus one (conditional) Newton iteration.

[Christian Sigg via Phabricator via cfe-commits]

csigg updated this revision to Diff 432871.
csigg added a comment.

Rebase.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D126158/new/

https://reviews.llvm.org/D126158

Files:
  mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
  mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
  mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir
  mlir/test/Dialect/LLVMIR/nvvm.mlir
  mlir/test/Target/LLVMIR/nvvmir.mlir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D126158.432871.patch
Type: text/x-patch
Size: 8862 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20220530/36fadb3d/attachment-0001.bin>