Files
composable_kernel/include/ck_tile/ops
root e8587b86c2 Fix CK-UA pipeline: s_waitcnt_vmcnt<0> in fmha_post_process
The final V tile's async load was not properly waited on before reading
from LDS: s_waitcnt_vmcnt<K_inst> allowed V_inst outstanding loads
(a no-op when K_inst == V_inst). The last loop iteration never prefetches
K, so only V is outstanding. Use s_waitcnt_vmcnt<0> unconditionally.

This partially fixes the BS32 race condition for production workloads
(maxk >= 256). A deeper pipeline race remains for very short KV
sequences (maxk < ~165, 2-5 pages) with block_size=32 at high batch.

Made-with: Cursor
2026-04-01 23:04:07 +00:00
..
2026-01-13 09:21:29 -08:00