[PATCH] D70951: [Matrix] Use fmuladd when lowering matrix.multiply.

[Florian Hahn via Phabricator via llvm-commits]

fhahn created this revision.
fhahn added reviewers: anemet, Gerolf, hfinkel, andrew.w.kaylor.
Herald added subscribers: tschuett, hiraditya.
Herald added a project: LLVM.

We should be able to use fmuladd instead of fmul/fadd, as this just
reduces the rounding error between operations. We could also make this
clear in the LangRef.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D70951

Files:
  llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
  llvm/test/Transforms/LowerMatrixIntrinsics/bigger-expressions-double.ll
  llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double.ll
  llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float.ll
  llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backwards-unsupported.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D70951.231861.patch
Type: text/x-patch
Size: 177482 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191203/aa8f5af3/attachment-0001.bin>