Will Constable ca03f36233 Change ProcessGroupNCCL default timeout to 10 min (#110947) ... Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. Pull Request resolved: https://github.com/pytorch/pytorch/pull/110947 Approved by: https://github.com/xw285cornell, https://github.com/fduwjj		2023-10-10 22:28:39 +00:00
..
aot_inductor	[aotinductor] Rename aot_runtime to aoti_runtime (#110007)	2023-09-26 00:46:54 +00:00
api	[ROCM] enable skipped test_api cpp tests (#109817)	2023-09-27 16:52:46 +00:00
c10d	Change ProcessGroupNCCL default timeout to 10 min (#110947)	2023-10-10 22:28:39 +00:00
common
dist_autograd	Remove unneeded linking in CMake targets (#109192)	2023-09-15 19:43:25 +00:00
jit	Fix optimize_for_inference to support modules that don't have a forward method (#110013)	2023-10-02 20:13:44 +00:00
lazy	Remove unneeded linking in CMake targets (#109192)	2023-09-15 19:43:25 +00:00
lite_interpreter_runtime	[BE] [3/3] Rewrite super() calls in test (#94592)	2023-02-12 22:20:53 +00:00
monitor
profiler	[Profiler] Fix SOFT_ASSERT test to not raise on debug builds (#91464)	2022-12-30 05:31:03 +00:00
rpc	Remove unneeded linking in CMake targets (#109192)	2023-09-15 19:43:25 +00:00
tensorexpr	[2/N] Move c10::variant to std::variant (#109723)	2023-09-24 02:47:43 +00:00
__init__.py