pytorch

mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 12:54:11 +08:00

Files

Shivam Raikundalia 1083bc749d [Memory Snapshot] Add Flag to Toggle Global and Local Callbacks for Annotations (#154932 )

Summary:
There are some cases where we want only local annotations for memory snapshot such as executing inside the cudastream callback, which cannot execute CUDA operators. Thus the cuda errors happen: Exception in RecordFunction callback: CUDA error: operation not permitted

However, we need to have an option to turn on the globally so that on-demand snapshot can get annotations. Additionally, there may be some cases in which auto-trace will also want annotations using record functions so we expose the flag to the auto-trace as well.

Test Plan:
Run MVAI executable and see that the errors go away

Rollback Plan:

Differential Revision: D75831687

Pull Request resolved: https://github.com/pytorch/pytorch/pull/154932
Approved by: https://github.com/mzzchy, https://github.com/sanrise

2025-06-04 23:15:19 +00:00

amp

[BE][Easy][17/19] enforce style for empty lines in import segments in torch/[a-c]*/ and torch/[e-n]*/ (#129769 )

2024-08-04 10:24:09 +00:00

__init__.py

[BE] Introduce torch.AcceleratorError (#152023 )

2025-06-01 21:02:43 +00:00

_gpu_trace.py

[BE][Easy] enable PYFMT for torch/[a-s]*/ (#138447 )

2024-12-23 14:04:00 +00:00

_memory_viz.py

[BE][Easy] enable PYFMT for torch/[a-s]*/ (#138447 )