Zhengxu Chen 951dca8019 [compile] Invoke split FX graph by codegen. (#38657) ...

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

2026-04-15 21:03:41 -07:00

..

passes

fused qknorm+rope kernel optimization for SM9.0 (#37376)

2026-04-12 19:58:37 -07:00

__init__.py

…

backends.py

[compile] Invoke split FX graph by codegen. (#38657)

2026-04-15 21:03:41 -07:00

base_static_graph.py

…

caching.py

[compile] Invoke split FX graph by codegen. (#38657)

2026-04-15 21:03:41 -07:00

codegen.py

[compile] Invoke split FX graph by codegen. (#38657)

2026-04-15 21:03:41 -07:00

compiler_interface.py

[torch.compile] Allow usage of Opaque Objects in PyTorch 2.11 (#39286)

2026-04-08 23:21:10 +00:00

counter.py

…

cuda_graph.py

[Perf] Fix slow hasattr in CUDAGraphWrapper.__getattr__ (#37425)

2026-03-19 15:43:48 +08:00

decorators.py

[compile] Nest inductor cache under AOT compile dir (#39718)

2026-04-14 17:17:54 +00:00

monitor.py

…

partition_rules.py

…

piecewise_backend.py

[BugFix][Graph] fix: handle empty sym_shape_indices in PiecewiseBackend. (#39395)

2026-04-15 09:28:09 -04:00

wrapper.py

[torch.compile] Allow usage of Opaque Objects in PyTorch 2.11 (#39286)

2026-04-08 23:21:10 +00:00