diff options
| author | Joana Cruz <Joana.Cruz@arm.com> | 2024-12-17 14:50:33 +0000 |
|---|---|---|
| committer | Wilco Dijkstra <wilco.dijkstra@arm.com> | 2025-02-27 17:07:22 +0000 |
| commit | bf2b60a56036c951a798845223a2e04cc48507e4 (patch) | |
| tree | 92e09dd18d0ed4d96b40bc2ddcc32c8281c85492 /scripts/check-execstack.awk | |
| parent | 41dc9e7c2d80bc5e886950b8a7bd21f77c9793b3 (diff) | |
| download | glibc-bf2b60a56036c951a798845223a2e04cc48507e4.tar.xz glibc-bf2b60a56036c951a798845223a2e04cc48507e4.zip | |
AArch64: Improve codegen of AdvSIMD expf family
Load the polynomial evaluation coefficients into 2 vectors and use lanewise MLAs.
Also use intrinsics instead of native operations.
expf: 3% improvement in throughput microbenchmark on Neoverse V1, exp2f: 5%,
exp10f: 13%, coshf: 14%.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
(cherry picked from commit cff9648d0b50d19cdaf685f6767add040d4e1a8e)
Diffstat (limited to 'scripts/check-execstack.awk')
0 files changed, 0 insertions, 0 deletions