ViperEkura
/
AstrAI
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
AstrAI/csrc/kernels
History
ViperEkura b5cdea98ad refactor: remove ineffective __launch_bounds__ from prefill kernel 
- Remove MIN_BLOCKS template param and __launch_bounds__ attribute
- Profiling shows smem (not registers) is the occupancy bottleneck for D>=64, making the hint a no-op
- D=64 sees 2-4% speedup, D=128 unchanged (smem-capped at 1 block/SM)
- Update comment blocks in kernel header and both dispatch sites
- Verified correctness via standalone CUDA test (max_err ~1e-4)
 		2026-07-12 15:00:53 +08:00
..
attn_common.h feat: paged decode attention with split-KV (scalar + MMA) 2026-07-11 18:14:51 +08:00
attn_decode.cu perf: double-buffer KV pipeline and Q direct-to-register in decode 2026-07-12 14:14:54 +08:00
attn_decode_split_kv.cuh refactor: template AttentionParams, rename .cuh to .h 2026-07-11 11:03:14 +08:00
attn_decode_split_kv_mma.cuh perf: double-buffer KV pipeline and Q direct-to-register in decode 2026-07-12 14:14:54 +08:00
attn_entry_utils.cuh refactor: template AttentionParams, rename .cuh to .h 2026-07-11 11:03:14 +08:00
attn_mma_utils.cuh refactor: template AttentionParams, rename .cuh to .h 2026-07-11 11:03:14 +08:00
attn_paged_decode.cu perf: double-buffer KV pipeline and Q direct-to-register in decode 2026-07-12 14:14:54 +08:00
attn_paged_decode_split_kv.cuh feat: paged decode attention with split-KV (scalar + MMA) 2026-07-11 18:14:51 +08:00
attn_paged_decode_split_kv_mma.cuh perf: double-buffer KV pipeline and Q direct-to-register in decode 2026-07-12 14:14:54 +08:00
attn_prefill.cu refactor: remove ineffective __launch_bounds__ from prefill kernel 2026-07-12 15:00:53 +08:00
attn_prefill_split_q.cuh refactor: template AttentionParams, rename .cuh to .h 2026-07-11 11:03:14 +08:00
attn_prefill_split_q_mma.cuh refactor: remove ineffective __launch_bounds__ from prefill kernel 2026-07-12 15:00:53 +08:00