ViperEkura 29b0423c4e refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

..

gqa_common.cuh

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_decode_attn_mma.cuh

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_decode_attn.cu

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_decode_attn.cuh

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_entry_utils.cuh

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_mma_utils.cuh

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_prefill_attn_mma.cuh

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_prefill_attn.cu

refactor: deduplicate kernel code with shared MMA and entry-point helpers

2026-07-10 20:52:27 +08:00

gqa_prefill_attn.cuh

perf: reduce MMA kernel registers, switch to static smem

2026-07-10 00:39:47 +08:00