diff options
| author | ZhaoQi <zhaoqi01@loongson.cn> | 2024-11-06 09:26:28 +0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-11-06 09:26:28 +0800 |
| commit | 92be2cb08632ea38f6fbc41adfeb475ba27447dd (patch) | |
| tree | 6ab859631db94905a6e904b735cf40113734fe0a /llvm/lib/Analysis/ModuleSummaryAnalysis.cpp | |
| parent | 435e58468a1a99a4bbfad88d060abd37a9bc6928 (diff) | |
| download | llvm-92be2cb08632ea38f6fbc41adfeb475ba27447dd.zip llvm-92be2cb08632ea38f6fbc41adfeb475ba27447dd.tar.gz llvm-92be2cb08632ea38f6fbc41adfeb475ba27447dd.tar.bz2 | |
[LoongArch] Use LSX for scalar FP rounding with explicit rounding mode (#114766)
LoongArch FP base ISA only have frint.{s/d} instruction which reads the
global rounding mode. Utilize LSX for explicit rounding mode for scalar
ceil/floor/trunc/roundeven calls when -mlsx opend. It is faster than
calling the libm library functions.
Same as what gcc did:
https://gcc.gnu.org/pipermail/gcc-cvs/2023-November/394218.html
Diffstat (limited to 'llvm/lib/Analysis/ModuleSummaryAnalysis.cpp')
0 files changed, 0 insertions, 0 deletions