fduwjj f6dfbffb3b [c10d] Add hashing as a debug feature for before and after NCCL collective call (#113238) ... For now, we use `TORCH_DISTRIBUTED_DEBUG = DETAIL` to turn a debug feature which calculate the hashing for input tensors and output results of c10d collective in NCCL. This is a debugging feature so that we can rule out the bug from c10d level. <img width="840" alt="image" src="https://github.com/pytorch/pytorch/assets/6937752/cdc70b0b-ae3c-4efd-86ff-adc5c5ba505f"> Pull Request resolved: https://github.com/pytorch/pytorch/pull/113238 Approved by: https://github.com/wconstab, https://github.com/fegin		2023-12-25 22:25:38 +00:00
..
autograd
c10d	[c10d] Add hashing as a debug feature for before and after NCCL collective call (#113238)	2023-12-25 22:25:38 +00:00
rpc	[CI] Update clang-format (#116002)	2023-12-18 14:58:46 +00:00