Files
pytorch/cmake
Jerry Mannil 202f83dc4e [ROCm][layer_norm] Use __builtin_amdgcn_rcpf(x) instead of 1.f/x (#165589)
Replace (more) exact calculation with hardware approximation.

Benefits:
Reduced code size.
Improved performance for certain scenarios.

Experiments show low reduction in precision.
Experiments show no significant performance regressions. bfloat16 as well as float16 related calculations may benefit largely from this change.

Co-author: @mhalk @amd-hhashemi

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165589
Approved by: https://github.com/jeffdaily
2025-10-17 09:12:30 +00:00
..
2025-06-17 19:25:43 +00:00