Jeff Bolz c9a24fb932 vulkan: Support FA with any multiple of 8 head sizes (#15537) ...

The scalar FA shader already handled multiples of 8. The coopmat1 FA
shader assumed 16x16x16 and the shared memory allocations need the HSK
dimensions padded to a multiple of 16. NVIDIA's coopmat2 implementation
requires multiples of 16 for N and K, and needs the matrix dimensions
padded and loads clamped.

Store the FA pipelines in a map, indexed by the pipeline state.

2025-08-24 11:24:25 +02:00

248 KiB

Raw Blame History

View Raw