pytorch

mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00

Files

Sherlock Huang c1722db0f7 [NativeRT] Make VariadicOpConverter and FuseListUnpackConverter for cpu nodes only (#159519 )

Summary:
VariadicOpConverter and FuseListUnpackConverter would introduce ops that only have CPU kernels.

Currently, the graph passes are ran if static_dispatch is enabled.

As we plan to enable static_dispatch by default, this diff add the additional check for the graph pass to only work on the node that has all the inputs/outputs on CPU.

Test Plan:
CI

Rollback Plan:

Differential Revision: D79295640

Pull Request resolved: https://github.com/pytorch/pytorch/pull/159519
Approved by: https://github.com/dolpm, https://github.com/henryoier

2025-07-31 18:17:21 +00:00

Graph.cpp

Reland D78841818 (#159216 )

2025-07-28 07:39:35 +00:00

Graph.h

Reland D78841818 (#159216 )

2025-07-28 07:39:35 +00:00

GraphPasses.cpp

[NativeRT] Strengthen matcher check for StaticDispatch kernel (#159187 )

2025-07-29 04:03:49 +00:00

GraphPasses.h

[nativert] clean up some migration side-effects (#156919 )