Ye Wang e9af75800d introduce env ROCM_FLASH_ATTN_CU_NUM to control bwd group mode persistent kernel grid size

2026-05-07 15:48:17 -05:00

..

fmha_batch_prefill_kernel.hpp

[CK_TILLE] Temporarily remove batch prefill KV cache overflow asserts (#6201)

2026-04-07 20:41:24 +08:00

fmha_bwd_kernel.hpp

introduce env ROCM_FLASH_ATTN_CU_NUM to control bwd group mode persistent kernel grid size

2026-05-07 15:48:17 -05:00

fmha_fwd_appendkv_kernel.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

fmha_fwd_appendkv_tile_partitioner.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

fmha_fwd_kernel.hpp

[CK_TILE] Add LLC-aware FMHA head grouping and head-major scheduling on RDNA (#5018)

2026-03-16 21:18:34 +00:00

fmha_fwd_pagedkv_kernel.hpp

[CK_TILE][FMHA] Enable gpt-oss sink (#3490)

2026-01-14 21:32:06 +08:00

fmha_fwd_splitkv_combine_kernel.hpp

Shuffle fix for gfx950 (#3491)

2026-01-13 09:21:29 -08:00

fmha_fwd_splitkv_kernel.hpp

[CK_TILE][FMHA] Enable gpt-oss sink (#3490)

2026-01-14 21:32:06 +08:00

fmha_fwd_v3_kernel.hpp

[CK] Add FP8 per-tensor quantization support for FMHA V3 pipeline (#6051)

2026-04-07 22:19:28 +08:00