diff options
| author | Hugh Delaney <hugh.delaney@codeplay.com> | 2024-11-18 15:29:17 +0000 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-11-18 15:29:17 +0000 |
| commit | 8f8016fe66dd260b03a4d1c2b50636e36e02942b (patch) | |
| tree | 98a899a98a76a04dd743a332d09f0a85cfb29ef2 /llvm/lib/Bitcode/Reader/BitcodeReader.cpp | |
| parent | ed8019d9fbed2e6a6b08f8f73e9fa54a24f3ed52 (diff) | |
[NVPTX] Add patterns for fma.relu.{f16|f16x2|bf16|bf16x2} (#114977)
Add patterns to lower `fmaxnum(fma(a, b, c), 0)` to `fma.rn{.ftz}.relu`
for `f16`, `f16x2`, `bf16`, `bf16x2` types, when `nnan` is used.
`fma_relu` honours `NaN`, so the substitution is only made if the `fma`
is `nnan`, since `fmaxnum` returns the non NaN argument when passed a
NaN value.
This patch also removes some `bf16` ftz instructions since `FTZ` is not
supported with the `bf16` type, according to the PTX ISA docs.
Diffstat (limited to 'llvm/lib/Bitcode/Reader/BitcodeReader.cpp')
0 files changed, 0 insertions, 0 deletions