llama.cpp
CUDA: consistent use of restrict + PDL for FA
#25185

Merged

CUDA: consistent use of restrict + PDL for FA #25185

JohannesGaessler merged 1 commit into ggml-org:master from JohannesGaessler:cuda-fattn-restrict-pdl

CUDA: consistent use of __restrict__ + PDL for FA

a74ffc69

JohannesGaessler requested a review 43 days ago

github-actions added ggml

github-actions added CUDA

am17an approved these changes on 2026-07-01

ORippler approved these changes on 2026-07-01

JohannesGaessler merged b820cc8e into master 43 days ago

Reviewers

ORippler

am17an

Assignees

No one assigned

Labels

ggml CUDA

Milestone

No milestone