Repository navigation

Optimize FP64 fabs on GPUs with low FP64 performance - #12294

Merged

davebayer merged 8 commits into

NVIDIA:mainfrom

lennartvoelz:libcudacxx/feat/fp64_fabs_int_path

Nov 6, 2026

Merged

Optimize FP64 fabs on GPUs with low FP64 performance#12294
davebayer merged 8 commits into
NVIDIA:mainfrom
lennartvoelz:libcudacxx/feat/fp64_fabs_int_path

Commits on Oct 27, 2026
- Add fabs integer path for fp64 slow targets
  lennartvoelz
  committed
  a623b37
- Add tests for fabs integer device and host path; exclude fast path fp64 targets
  lennartvoelz
  committed
  94c6c30
- Loosen fabs NaN tests to only check for NaN
  lennartvoelz
  committed
  c5844b8
- Merge branch 'main' into libcudacxx/feat/fp64_fabs_int_path
  lennartvoelz
  authored
  cf44ee6
- Clean up fabs integer path
  lennartvoelz
  committed
  9664a31
Commits on Oct 28, 2026
- Merge branch 'main' into libcudacxx/feat/fp64_fabs_int_path
  lennartvoelz
  authored
  43c1c39
- Use 120 column limit for comments in fp64 fabs integer path
  lennartvoelz
  committed
  0133ad4
Commits on Nov 2, 2026
- Merge branch 'main' into libcudacxx/feat/fp64_fabs_int_path
  lennartvoelz
  authored
  56ecf2d