New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
add channels last support for thnn_conv2d (non-dilated) #49582
Conversation
[ghstack-poisoned]
💊 CI failures summary and remediationsAs of commit 7279dd8 (more details on the Dr. CI page and at hud.pytorch.org/pr/49582):
🕵️ 1 new failure recognized by patternsThe following CI failures do not appear to be due to upstream breakages: pytorch_linux_xenial_py3_clang5_asan_test1 (1/1)Step: "Run tests" (full log | diagnosis details | 🔁 rerun)
|
ghstack-source-id: 8044e97be67244e41383d2a4a956860c000e90a0 Pull Request resolved: #49582
Initiative of this patchDue to performance or functionality reasons, PerformanceMeasured Conv2d test configs #40610 on single core of Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz, mkldnn path is manually disabled with Notice that on grouped conv CL performs worse than Contiguous in the following table, but grouped conv will end up in mkldnn with default setting (mkldnn has much better perf with group conv than the gemm based algorithm)
|
[ghstack-poisoned]
ghstack-source-id: 420d9cdfbcf948e6695a89018643071ac1409d51 Pull Request resolved: #49582
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
ghstack-source-id: 0584ef4b7b499004aa55c1d62db34c584ee60aab Pull Request resolved: pytorch#49582
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
@VitalyFedyunin has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
rebased! |
@VitalyFedyunin has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
Differential Revision: [D26007050](https://our.internmc.facebook.com/intern/diff/D26007050) [ghstack-poisoned]
rebased! |
@VitalyFedyunin has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
@VitalyFedyunin merged this pull request in 8b88c24. |
This pull request has been reverted by b60d1b7. |
Reverting as it might be offender to Iphone perf regression, please do nothing at thing point, might be false positive, I'm investigating. |
@VitalyFedyunin The current code base in I will rewrite this one accordingly, and collect new performance data. @VitalyFedyunin |
Stack from ghstack:
Differential Revision: D26007050