Xin Li ae268b6326

Fix Flashinfer Allreduce+Norm enable disable calculation based on fi_allreduce_fusion_max_token_num (#21325) ...

Signed-off-by: XIn Li <xinli@nvidia.com>

2025-07-22 12:42:31 -07:00

..

__init__.py

[torch.compile] avoid Dynamo guard evaluation overhead (#7898)

2024-08-28 16:10:12 -07:00

activation_quant_fusion.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

backends.py

[BugFix] VLLM_DISABLE_COMPILE_CACHE=1 should disable all reads and writes from the cache (#20942)

2025-07-15 01:26:18 +00:00

base_piecewise_backend.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

collective_fusion.py

Fix Flashinfer Allreduce+Norm enable disable calculation based on fi_allreduce_fusion_max_token_num (#21325)

2025-07-22 12:42:31 -07:00

compiler_interface.py

[BugFix] VLLM_DISABLE_COMPILE_CACHE=1 should disable all reads and writes from the cache (#20942)

2025-07-15 01:26:18 +00:00

counter.py

[BugFix] VLLM_DISABLE_COMPILE_CACHE=1 should disable all reads and writes from the cache (#20942)

2025-07-15 01:26:18 +00:00

cuda_piecewise_backend.py

[CUDA] Enable full cudagraph for FlashMLA (#18581)

2025-06-13 18:12:26 +00:00

decorators.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

fix_functionalization.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

fusion_attn.py

[Misc] Add SPDX-FileCopyrightText (#20428)

2025-07-04 07:40:42 +00:00

fusion.py

[Perf][fp8] Use CustomOp abstraction for fp8 quant for better perf (#19830)

2025-07-11 04:56:28 +00:00

fx_utils.py

[torch.compile][ROCm] Fuse quantization onto attention using a torch.compile pass (#16756)

2025-06-12 08:31:04 -07:00

inductor_pass.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

monitor.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

multi_output_match.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

noop_elimination.py

[torch.compile][ROCm] Fuse quantization onto attention using a torch.compile pass (#16756)

2025-06-12 08:31:04 -07:00

pass_manager.py

[Misc] Refactor AllReduceFusionPass. Remove parameter (#20918)

2025-07-15 06:57:40 +00:00

sequence_parallelism.py

[Feature] Support sequence parallelism for static fp8 quantization (#19181)

2025-06-23 16:09:02 -04:00

torch25_custom_graph_pass.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

vllm_inductor_pass.py

[Bugfix] Upgrade depyf to 0.19 and streamline custom pass logging (#20777)

2025-07-11 00:15:22 -07:00

wrapper.py

[cold start] replace VLLM_COMPILE_DEPYF with debug_dump_dir (#20940)

2025-07-15 13:02:17 +08:00