saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-15 21:00:47 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
pytorch/c10
History
Edward Yang 3df79f403e Tighten up invariants regarding StreamId. (#15125) 
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15125

I realized that it is really bad juju if you fake a StreamId
out of thin air, because in general this isn't going to work.
So, make the constructor a lot scarier.

Most "faking StreamId out of thin air" happens because someone
just wants to put something on the default stream.

Reviewed By: dzhulgakov

Differential Revision: D13432800

fbshipit-source-id: a86991d6fc1d8aa4e54e8175e5f06f90856238e6
2018-12-17 13:30:54 -08:00
..
core Move TensorImpl::CopyFrom to caffe2::Tensor (2/2) (#14858) 2018-12-13 18:41:24 -08:00
cuda Tighten up invariants regarding StreamId. (#15125) 2018-12-17 13:30:54 -08:00
hip Build c10 HIP test 2018-12-14 15:36:38 -08:00
impl Tighten up invariants regarding StreamId. (#15125) 2018-12-17 13:30:54 -08:00
macros Move numa.{h, cc} to c10/util (#15024) 2018-12-12 12:21:10 -08:00
test Tighten up invariants regarding StreamId. (#15125) 2018-12-17 13:30:54 -08:00
util Fix typo (#15045) 2018-12-13 11:13:19 -08:00
CMakeLists.txt Move numa.{h, cc} to c10/util (#15024) 2018-12-12 12:21:10 -08:00
Device.cpp Unify device argument parsing between torch and c10 2018-12-05 18:37:32 -08:00
Device.h Unify device argument parsing between torch and c10 2018-12-05 18:37:32 -08:00
DeviceGuard.h Rename c10::detail to c10::impl (#13838) 2018-11-14 07:39:37 -08:00
DeviceType.cpp Add hash functions for Stream, CUDAStream; fix Device hash function (#14191) 2018-11-27 08:32:08 -08:00
DeviceType.h Change Tensor::CopyFrom to a simple double dispatch (#14268) 2018-11-28 15:45:37 -08:00
Half-inl.h Replace at::Half non-vectorized conversions with implementations from FP16 (#14411) 2018-12-04 14:32:33 -08:00
Half.cpp Replace at::Half non-vectorized conversions with implementations from FP16 (#14411) 2018-12-04 14:32:33 -08:00
Half.h Replace at::Half non-vectorized conversions with implementations from FP16 (#14411) 2018-12-04 14:32:33 -08:00
Stream.cpp Add c10::Stream, make at::cuda::CUDAStream use it. (#13133) 2018-10-31 07:55:10 -07:00
Stream.h Tighten up invariants regarding StreamId. (#15125) 2018-12-17 13:30:54 -08:00
StreamGuard.h Rename c10::detail to c10::impl (#13838) 2018-11-14 07:39:37 -08:00