vLLM/tests/kernels/attention at 08582942d8af7fc58339f3a5fedaffbbc8ea9f02 - vLLM

conftest.py

[Chore] Clean up pytorch helper functions in vllm.utils (#26908 )

2025-10-18 09:48:22 -07:00

test_amx_mla.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212 )

2026-09-16 10:57:24 -07:00

test_attention_selector.py

[Bugfix][Attention] Support int4_per_token_head on non-power-of-two head sizes (#56198 )

2026-10-05 04:52:39 +00:00

test_attention.py

[Refactor] Remove dead tests code (#58916 )

2026-09-30 04:39:03 +00:00

test_cache.py

[Bugfix][SM120][MLA] Support NoPE sparse MLA (GLM-5.3-Flash) on the FlashInfer SM120 backend (#55277 )

2026-09-23 02:03:20 -07:00

test_cpu_attn.py

[Bugfix][DiffusionGemma] Read the causal mask as int32 and compile the sample step (#59992 )

2026-10-06 19:25:51 +08:00

test_cutlass_mla_decode.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212 )

2026-09-16 10:57:24 -07:00

test_deepgemm_attention.py

[KV Cache] GLM-5.3-Flash: use the generic packed KV layout; CircularBufferSpec tail with speculative ring slots (#57169 )

2026-10-09 15:48:37 -04:00

test_flash_attn.py

[MyPy][3/N] Fix MyPy errors in test groups (part 3) (#55939 )

2026-09-30 11:06:07 +00:00

test_flashinfer_mla_decode.py

[Perf][Attention] Reduce GLM sparse MLA preparation overhead (#57458 )

2026-09-21 17:24:25 -07:00

test_flashinfer_trtllm_attention.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212 )

2026-09-16 10:57:24 -07:00

test_flashinfer.py

[Attention] Allow an NVFP4 KV cache on SM90 with fa2 (#60620 )

2026-10-08 12:16:22 -07:00

test_flashmla_sparse.py

[Perf][DSv4] Fuse inverse RoPE + FP8 quant into FlashInfer sparse MLA (#58621 )

2026-09-24 23:26:50 -07:00

test_flashmla.py

[ROCM][CI] Attention test speedup (#52763 )

2026-08-18 16:40:30 -05:00

test_kimi_k3_mla_fused_epilogue.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212 )

2026-09-16 10:57:24 -07:00

test_kimi_k3_mla_key_concat_kv_cache.py

[MyPy][3/N] Fix MyPy errors in test groups (part 3) (#55939 )

2026-09-30 11:06:07 +00:00

test_lightning_attn.py

Add pydocstyle to the ruff rules (#52136 )

2026-09-16 16:44:29 +00:00

test_merge_attn_states.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212 )

2026-09-16 10:57:24 -07:00

test_mha_attn.py

[MyPy][3/N] Fix MyPy errors in test groups (part 3) (#55939 )

2026-09-30 11:06:07 +00:00

test_minimax_m3_fp8_triton_indexer.py

[Minimax-M3] Keep the native FP8 MMA in the Triton indexer scorers (#59481 )

2026-10-02 12:30:39 -04:00

test_minimax_m3_msa_cutlass_sparse_decode.py

[Perf][MiniMax-M3] MSA prefill attends with the E4M3 query; bump fmha_sm100 to 968061a (#60684 )

2026-10-11 13:40:45 +08:00

test_minimax_m3.py

[Bugfix][MiniMax-M3] Guard sparse decode sentinel indices (#54597 )

2026-10-08 18:27:57 +00:00

test_mixed_causal_attn.py

[Model] Add DiffusionGemma Support (#45163 )

2026-06-11 22:17:35 -07:00

test_mla_cross_layer_kernel_equivalence.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212 )

2026-09-16 10:57:24 -07:00

test_mla_decode_cpu.py

[Chore]:Extract math and argparse utilities to separate modules (#27188 )

2025-10-26 04:03:32 -07:00

test_pack_unpack_triton.py

[1/N][Attention] Restructure attention: move files (#31916 )

2026-01-09 13:10:24 -08:00

test_prefix_prefill.py

[Attention] Remove cascade attention (#61126 )

2026-10-12 10:41:07 -07:00

test_rocm_aiter_candidate_mask.py

[ROCm][DSV4.1][Perf] Stride the DSA decode candidate mask over the live context (#56628 )

2026-09-14 20:47:32 +00:00

test_rocm_aiter_fa.py

[MyPy][3/N] Fix MyPy errors in test groups (part 3) (#55939 )

2026-09-30 11:06:07 +00:00

test_rocm_aiter_mla_causal_verify_mask.py

Disable mypy arg-type and assignment checks in tests (#59428 )

2026-09-30 15:18:50 +00:00

test_rocm_aiter_mla_decode_metadata.py

[Bugfix][ROCm][CI] Give the AITER MLA decode metadata stub its MLA dims (#52139 )

2026-08-13 13:35:00 -05:00

test_rocm_aiter_mla_decode.py

[ROCm][CI] Add MLA decode accuracy and determinism tests (#50480 )

2026-08-05 20:35:56 -05:00

test_rocm_aiter_mla_fp8_prefill.py

Disable mypy arg-type and assignment checks in tests (#59428 )

2026-09-30 15:18:50 +00:00

test_rocm_aiter_mla_fp8_support.py

[ROCm][Spec Decode] Add Aiter MLA decode support non-causal draft block (#55966 )

2026-09-15 12:18:09 -05:00

test_rocm_aiter_mla_head_padding.py

[ROCm][MLA][DCP] Support causal multi-token verification (#51705 )

2026-08-31 10:42:28 -07:00

test_rocm_aiter_mla_op_registration.py

Add pydocstyle to the ruff rules (#52136 )

2026-09-16 16:44:29 +00:00

test_rocm_aiter_mla_sink.py

[KV Cache] GLM-5.3-Flash: use the generic packed KV layout; CircularBufferSpec tail with speculative ring slots (#57169 )

2026-10-09 15:48:37 -04:00

test_rocm_aiter_mla_sparse_metadata_sync.py

[Perf] [Feat] [ROCm] Add densemha support to ROCm AITER Sparse MLA (#49263 )

2026-10-12 15:50:28 +00:00

test_rocm_aiter_unified_attn.py

Fix ROCm AITER FP8 KV test tolerances. (#53590 )

2026-09-09 12:22:25 -05:00

test_rocm_paged_mxfp4_indexer.py

Disable mypy arg-type and assignment checks in tests (#59428 )

2026-09-30 15:18:50 +00:00

test_rocm_triton_attn_dsv4.py

[ROCm][DSv4.1][Perf] Replace the Torch prefill combine with the shared combine_topk_swa_indices kernel (#56638 )

2026-10-08 14:15:18 +08:00

test_segmented_attention.py

[ROCM][Perf] Add ROCM_SEGMENTED_ATTN attention-backend works across RDNA3/3.5/4 (#59132 )

2026-10-06 07:53:47 +00:00

test_triton_decode_attention.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212 )

2026-09-16 10:57:24 -07:00

test_triton_mla_causal_verify_flatten.py

[Bugfix][MLA] TritonMLA: fix illegal memory access on causal multi-token decode (#51065 )

2026-09-18 03:49:53 -05:00

test_triton_prefill_attention.py

[Perf][ROCm][Attention] Narrow the Triton prefill-attention KV tile on RDNA3/RDNA4 (#58225 )

2026-09-23 08:39:27 +00:00

test_triton_unified_attention_diffkv.py

[Feature][Attention] Cover a whole spec-verify in one DiffKV split-KV program (#59085 )

2026-10-10 11:55:45 +08:00

test_triton_unified_attention.py

[Bugfix][Attention] Preserve small FP8 softmax weights in Triton attention (#60156 )

2026-10-08 12:35:22 +02:00

test_trtllm_kvfp8_dequant.py

Add pydocstyle to the ruff rules (#52136 )

2026-09-16 16:44:29 +00:00

test_use_trtllm_attention.py

[Attention] Add FlashInfer XQA decode support on SM12x (#49718 )

2026-08-11 16:32:49 -07:00

test_xpu_mla_sparse.py

Add pydocstyle to the ruff rules (#52136 )

2026-09-16 16:44:29 +00:00