Zhengxu Chen 53cd7f868b

[compile] Recompile graph module during Dynamo cache loading. (#30743) ...

Signed-off-by: Zhengxu Chen <zhxchen17@fb.com>

2025-12-17 02:00:12 -08:00

..

__init__.py

…

activation_quant_fusion.py

…

backends.py

[torch.compile] Add encoder tag for compilation (#30489)

2025-12-14 18:15:11 +08:00

base_static_graph.py

…

caching.py

[compile] Recompile graph module during Dynamo cache loading. (#30743)

2025-12-17 02:00:12 -08:00

collective_fusion.py

[Bugfix] fix fuse_allreduce_rms when tp =1 (#30178)

2025-12-08 06:43:47 +00:00

compiler_interface.py

[Compile] Conditional compilation. Introduce compile_ranges (#24252)

2025-12-05 18:17:32 +00:00

counter.py

…

cuda_graph.py

Add logging for cudagraph related info (#29825)

2025-12-03 01:01:48 -08:00

decorators.py

[compile] Disable aot when eager backend is used. (#30810)

2025-12-17 01:55:56 -08:00

fix_functionalization.py

[Bugfix] Fix flashinfer ar+norm kernel not available issue (#29960)

2025-12-03 18:50:53 +00:00

fusion_attn.py

[Attention] Remove imports from vllm/attention/__init__.py (#29342)

2025-11-26 10:53:15 -07:00

fusion.py

[Bugfix] Fix fusion for VL models (#30244)

2025-12-14 21:22:37 +08:00

fx_utils.py

[Bugfix] Defunctionalize TRTLLM AR+Norm op for avoiding extra clone kernel before it (#29631)

2025-12-03 05:15:50 +00:00

inductor_pass.py

[BugFix] Fix non detected failing tests (#30277)

2025-12-09 17:57:55 +00:00

matcher_utils.py

[Bugfix] Fix fusion for VL models (#30244)

2025-12-14 21:22:37 +08:00

monitor.py

…

noop_elimination.py

Do not guard during noop elimination pass (#30095)

2025-12-05 04:10:12 +00:00

partition_rules.py

…

pass_manager.py

[Rocm][torch.compile] Adding layernorm + fp8 block quant and silu + fp8 block quant for Aiter (#25693)

2025-12-09 22:39:26 +00:00

piecewise_backend.py

[torch.compile] Add encoder tag for compilation (#30489)

2025-12-14 18:15:11 +08:00

post_cleanup.py

…

qk_norm_rope_fusion.py

[Attention] Remove imports from vllm/attention/__init__.py (#29342)

2025-11-26 10:53:15 -07:00

rocm_aiter_fusion.py

[Rocm][torch.compile] Adding layernorm + fp8 block quant and silu + fp8 block quant for Aiter (#25693)

2025-12-09 22:39:26 +00:00

sequence_parallelism.py

[Compile] Conditional compilation. Introduce compile_ranges (#24252)

2025-12-05 18:17:32 +00:00

torch25_custom_graph_pass.py

…

vllm_inductor_pass.py

…

wrapper.py

[compile] Stop one-off setting enable_aot_compile and use context manager instead. (#30503)

2025-12-11 20:28:03 +00:00