vllm-project/vllm

main

vllm

Grade	Name	Complexity	Churn	Issues
A	vllm\v1\attention\backends\mla\triton_mla.py	16	1	1
A	vllm\v1\attention\backends\mla\xpu_mla_sparse.py	2	1	-
A	vllm\v1\attention\backends\registry.py	12	1	-
A	vllm\v1\attention\backends\rocm_aiter_fa.py	51	1	2
A	vllm\v1\attention\backends\rocm_aiter_unified_attn.py	11	1	1
A	vllm\v1\attention\backends\rocm_attn.py	22	1	2
A	vllm\v1\attention\backends\short_conv_attn.py	0	1	-
A	vllm\v1\attention\backends\triton_attn.py	47	1	1
A	vllm\v1\attention\backends\triton_attn_diffkv.py	14	1	-
A	vllm\v1\attention\backends\turboquant_attn.py	41	1	1
A	vllm\v1\attention\backends\utils.py	41	1	-
A	vllm\v1\attention\ops\chunked_prefill_paged_decode.py	28	1	-
A	vllm\v1\attention\ops\common.py	27	1	-
A	vllm\v1\attention\ops\dcp_alltoall.py	31	1	1
A	vllm\v1\attention\ops\flashmla.py	9	1	-
A	vllm\v1\attention\ops\int4_per_token_head.py	34	1	3
A	vllm\v1\attention\ops\merge_attn_states.py	6	1	-
A	vllm\v1\attention\ops\paged_attn.py	0	1	-
A	vllm\v1\attention\ops\prefix_prefill.py	47	1	2
A	vllm\v1\attention\ops\rocm_aiter_mla_sparse.py	115	1	4