llama.cpp
opencl: fix two issues on flash attention for Adreno a7x
#25697

Merged

opencl: fix two issues on flash attention for Adreno a7x #25697

max-krasnyansky merged 2 commits into ggml-org:master from qualcomm:pr/opencl-fa-adreno-a7x

opencl: route `sub_group_shuffle_xor` to qcom ext when KHR ext is una…

a584d461

opencl: skip FA kernels with mixed and quant types for A7x to avoid c…

582160a7

github-actions added ggml

github-actions added OpenCL

lhez marked this pull request as ready for review 1 day ago

lhez requested a review 1 day ago

lhez approved these changes on 2026-07-15

lhez requested a review from

max-krasnyansky 1 day ago

max-krasnyansky approved these changes on 2026-07-15

max-krasnyansky merged c3d47e69 into master 1 day ago

Reviewers

max-krasnyansky

lhez

Assignees

No one assigned

Labels

ggml OpenCL

Milestone

No milestone