Shengqi Chen a10e369f06 [Build] Address CUDA arch review comments ...

Fix CUDA arch warning tests so they do not depend on importing the stable libtorch extension, which keeps the warning coverage active in lightweight CI environments and avoids mypy treating a fixture value as a base class.

Use regex for the compiled-arch parser, apply formatter output, and make the CUTLASS grouped GEMM Python support query fall back to false when the op is unavailable or unimplemented in the current build.

Signed-off-by: Shengqi Chen <harry-chen@outlook.com>

2026-06-30 20:03:28 +08:00

..

nvfp4_utils.py

Add nvfp4 support to reshape_and_cache_flash (#37332)

2026-04-17 07:28:00 -07:00

test_allspark_gemm.py

[Hardware] Replace torch.cuda.synchronize() api with torch.accelerator.synchronize (#36085)

2026-03-05 10:36:39 +00:00

test_awq_triton.py

[Kernel][Test] Extend lightning_attn and awq_triton kernel tests to XPU (#43307)

2026-06-04 14:25:59 -04:00

test_awq.py

[Refactor] Remove dead quantization code and tests (#45454)

2026-06-17 16:12:01 -04:00

test_block_fp8.py

[W8A8 Block Linear Refactor][2/N] Remove W8A8Fp8BlockLinearOp and adopt Fp8 block linear kernel selections. (#33892)

2026-04-09 08:50:39 +08:00

test_block_int8.py

[Misc] Make SchedulerConfig.max_model_len init-only (#28733)

2025-11-15 01:59:31 -08:00

test_cpu_fp8_scaled_mm.py

[CPU] Add FP8 W8A16 linear support (#41186)

2026-05-06 07:05:27 +00:00

test_cutlass_group_gemm_support.py

[Build] Address CUDA arch review comments

2026-06-30 20:03:28 +08:00

test_cutlass_scaled_mm.py

[Perf] SM90 cutlass fp8 mm supports odd M by swap_ab, 180~290% kernel performance improvement (#44572)

2026-06-13 12:05:45 -07:00

test_cutlass_w4a8_moe.py

[Hardware] Replace torch.cuda.synchronize() api with torch.accelerator.synchronize (#36085)

2026-03-05 10:36:39 +00:00

test_cutlass_w4a8.py

[CI/Build][AMD] Skip quantization kernels tests that require CUTLASS or e4m3fn when not supported by platform (#30020)

2025-12-10 02:28:37 +00:00

test_flashinfer_nvfp4_scaled_mm.py

[Kernel][Performance] Add FlashInfer cutedsl NVFP4 GEMM backend (#42235)

2026-06-22 16:17:18 -04:00

test_flashinfer_scaled_mm.py

[Platform] Deprecate seed_everything (#31659)

2026-01-04 18:34:04 -08:00

test_fp8_min_max_helper.py

[Bugfix][Hardware][AMD] Consolidate FP8 min/max values helper function (#31106)

2026-01-07 06:55:03 +00:00

test_fp8_quant_group.py

[Misc] Fix Current vLLM config is not set. warnings, assert to avoid issues in the future (#31747)

2026-01-08 15:20:49 -08:00

test_fp8_quant.py

[CI] Bump mypy version (#34950)

2026-03-04 20:55:11 +00:00

test_gptq.py

[Kernel] Add GPTQv2 format support for low-bit or asymmetric quantization, by adapting gptq_gemm (#26092)

2025-10-23 23:26:13 -04:00

test_hadacore.py

[CI/Build][AMD] Skip marlin, machete, and hadacore tests since these require _C functions not defined for ROCm (#30109)

2025-12-06 12:54:17 +08:00

test_int8_kernel.py

[Misc] Fix Current vLLM config is not set. warnings, assert to avoid issues in the future (#31747)

2026-01-08 15:20:49 -08:00

test_int8_quant.py

[Platform] Deprecate seed_everything (#31659)

2026-01-04 18:34:04 -08:00

test_machete_mm.py

[Hardware] Replace torch.cuda.device_count/current_device/set_device API (#36145)

2026-03-12 07:57:47 -07:00

test_marlin_gemm.py

[Bugfix] Fix SM121 (DGX Spark) exclusion from Marlin/CUTLASS FP8 paths (#35568)

2026-05-15 10:59:00 -07:00

test_marlin_tile_padding.py

[ROCm][CI] Skip the MoE Marlin tile-padding helper assertion (#46580)

2026-06-24 07:31:33 +00:00

test_mxfp4_qutlass.py

[Platform] Deprecate seed_everything (#31659)

2026-01-04 18:34:04 -08:00

test_mxfp4_triton_ep.py

[Perf][gpt-oss] Downgrade triton_kernels to v3.5.1 (#43135)

2026-05-20 14:13:12 -07:00

test_nvfp4_emulation.py

[Hardware][AMD][CI] Fix Kernels Quantization test timeout (#46859)

2026-06-26 15:19:16 -05:00

test_nvfp4_quant.py

[Perf] Padded nvfp4 quant kernel to remove additional copy, 2.4%~5.7% e2e performance improvement (#42774)

2026-05-18 16:38:12 -07:00

test_nvfp4_qutlass.py

[Platform] Deprecate seed_everything (#31659)

2026-01-04 18:34:04 -08:00

test_nvfp4_scaled_mm.py

[Platform] Deprecate seed_everything (#31659)

2026-01-04 18:34:04 -08:00

test_per_token_group_quant.py

[Bugfix] Fix gridDim.y overflow for large row counts (#45255)

2026-06-19 23:27:45 -04:00

test_quantized_embedding.py

[Quant] Support compressed-tensors WNA8O8Int linears and WNInt embeddings (#44340)

2026-06-04 07:40:33 -07:00

test_rdna3_compile_guards.py

[Refactor] Rename rocm_moe.py to rocm_moe_rdna.py (#45011)

2026-06-10 17:02:09 +08:00

test_rdna3_moe_w4a16.py

[ROCm][Perf] Fused MoE W4A16 HIP kernel for AMD RDNA3 (gfx1100) (#44075)

2026-06-06 15:30:39 -05:00

test_rdna3_w4a16_selection.py

[Kernel][ROCm] Native W4A16 kernel for AMD RDNA3 (gfx1100) — fp16 + bf16 (#41394)

2026-05-29 11:04:40 +00:00

test_rdna3_w4a16.py

[Kernel][ROCm] Native W4A16 kernel for AMD RDNA3 (gfx1100) — fp16 + bf16 (#41394)

2026-05-29 11:04:40 +00:00

test_rocm_compressed_tensors_w4a16.py

[Kernel][Hardware][AMD] Add TritonW4A16LinearKernel for ROCm (#37352)

2026-04-10 10:25:27 +00:00

test_rocm_skinny_gemms.py

[ROCm] Enable wvSplitK skinny GEMM kernel for RDNA4/gfx1x decode (#34709)

2026-03-20 10:11:23 -05:00

test_scaled_mm_kernel_selection.py

[OOT] Add OOT support for linear kernel. (#37989)

2026-03-31 14:33:21 +08:00

test_silu_mul_nvfp4_quant.py

[Misc] Fix Current vLLM config is not set. warnings, assert to avoid issues in the future (#31747)

2026-01-08 15:20:49 -08:00

test_triton_scaled_mm.py

[ROCm] Fix OOB During Model Warmup With ROCM_ATTN and MRV2 (#46548)

2026-06-24 10:53:21 -05:00

test_triton_w4a16.py

[Kernel][Hardware][AMD] Add TritonW4A16LinearKernel for ROCm (#37352)

2026-04-10 10:25:27 +00:00

test_w4a16_kernel_selection.py

[Kernel][Hardware][AMD] Add TritonW4A16LinearKernel for ROCm (#37352)

2026-04-10 10:25:27 +00:00