Skip to content

[v1.4.0] Fix reading __cuda_array_interface__ inferred strides, add… #31450

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.

Sign up for GitHub

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

Jump to bottom
Merged
mingbowan merged 1 commit into from
Dec 20, 2019
Merged

[v1.4.0] Fix reading __cuda_array_interface__ inferred strides, add… #31450

mingbowan merged 1 commit into from
Dec 20, 2019

Conversation

@gchanan
Copy link
Contributor

@gchanan gchanan commented Dec 18, 2019

… test.

This is a simpler fix than #24947, which both fixed the bug and updated the protocol version.
This also adds a test (which the previous PR did not).

So the plan is that master (1.5) will have the new protocol version (and a test), 1.4 will have the old protocol version and the test.

@gchanan
[v1.4.0] Fix reading __cuda_array_interface__ inferred strides, add…
f7472e5 
… test.

This is a simpler fix than pytorch#24947, which both fixed the bug and updated the protocol version.
This also adds a test (which the previous PR did not).

So the plan is that master (1.5) will have the new protocol version (and a test), 1.4 will have the old protocol version and the test.
@kostmo
Copy link
Member

kostmo commented Dec 18, 2019 •
edited
Loading

💊 CircleCI build failures summary and remediations

As of commit f7472e5:

  • 2/2 failures introduced in this PR

Detailed failure analysis

One may explore the probable reasons each build failed interactively on the Dr. CI website.

🕵️ 2 new failures recognized by patterns

The following build failures don't appear to be due to upstream breakage:

See CircleCI build pytorch_xla_linux_xenial_py3_6_clang7_build (1/2)

Step: "Build" (full log | pattern match details)

Dec 18 23:22:53 RuntimeError: Failed to apply patch: /var/lib/jenkins/workspace/xla/torch_patches/X10-serialization.diff 
Dec 18 23:22:52 + python /var/lib/jenkins/workspace/xla/scripts/cond_patch.py /var/lib/jenkins/workspace/xla/scripts/../torch_patches /var/lib/jenkins/workspace/xla/scripts/../.. 
Dec 18 23:22:53 1 out of 1 hunk FAILED 
Dec 18 23:22:53 Applying patch file: /var/lib/jenkins/workspace/xla/torch_patches/X10-clip_grad.diff 
Dec 18 23:22:53 Applying patch file: /var/lib/jenkins/workspace/xla/torch_patches/X10-device_test.diff 
Dec 18 23:22:53 Applying patch file: /var/lib/jenkins/workspace/xla/torch_patches/X10-serialization.diff 
Dec 18 23:22:53 Traceback (most recent call last): 
Dec 18 23:22:53   File "/var/lib/jenkins/workspace/xla/scripts/cond_patch.py", line 67, in <module> 
Dec 18 23:22:53     patch_repo(args) 
Dec 18 23:22:53   File "/var/lib/jenkins/workspace/xla/scripts/cond_patch.py", line 49, in patch_repo 
Dec 18 23:22:53     raise RuntimeError('Failed to apply patch: {}'.format(ppath)) 
Dec 18 23:22:53 RuntimeError: Failed to apply patch: /var/lib/jenkins/workspace/xla/torch_patches/X10-serialization.diff 
Dec 18 23:22:53 + cleanup 
Dec 18 23:22:53 + retcode=1 
Dec 18 23:22:53 + set +x 
Dec 18 23:22:53 =================== sccache compilation log =================== 
Dec 18 23:22:53 =========== If your build fails, please take a look at the log above for possible reasons =========== 
Dec 18 23:22:53 Compile requests                 1 
Dec 18 23:22:53 Compile requests executed        0 
Dec 18 23:22:53 Cache hits                       0 
Dec 18 23:22:53 Cache misses                     0 
Dec 18 23:22:53 Cache timeouts                   0

See CircleCI build pytorch_linux_backward_compatibility_check_test (2/2)

Step: "Test" (full log | pattern match details)

Dec 18 23:45:59 The PR is introducing backward incompatible changes to the operator library. Please contact PyTorch team to confirm whether this change is wanted or not. 
Dec 18 23:45:59 processing existing schema:  aten::__upsample_nearest(Tensor input, int? size=None, int? scale_factor=None) -> (Tensor) 
Dec 18 23:45:59 processing existing schema:  aten::__upsample_nearest(Tensor input, int[]? size=None, int? scale_factor=None) -> (Tensor) 
Dec 18 23:45:59 processing existing schema:  aten::__interpolate(Tensor input, int? size=None, float[]? scale_factor=None, str mode="nearest", bool? align_corners=None, bool? use_scale_factor=None) -> (Tensor) 
Dec 18 23:45:59 Can NOT find backward compatible schemas after changes for schema aten::__interpolate(Tensor input, int? size=None, float[]? scale_factor=None, str mode="nearest", bool? align_corners=None, bool? use_scale_factor=None) -> (Tensor) from the following candidates: 
Dec 18 23:45:59 [ 
Dec 18 23:45:59 aten::__interpolate(Tensor input, int? size=None, float[]? scale_factor=None, str mode="nearest", bool? align_corners=None) -> (Tensor) 
Dec 18 23:45:59 	aten::__interpolate(Tensor input, int[]? size=None, float[]? scale_factor=None, str mode="nearest", bool? align_corners=None) -> (Tensor) 
Dec 18 23:45:59 	aten::__interpolate(Tensor input, int? size=None, float? scale_factor=None, str mode="nearest", bool? align_corners=None) -> (Tensor) 
Dec 18 23:45:59 	aten::__interpolate(Tensor input, int[]? size=None, float? scale_factor=None, str mode="nearest", bool? align_corners=None) -> (Tensor) 
Dec 18 23:45:59 ] 
Dec 18 23:45:59 The PR is introducing backward incompatible changes to the operator library. Please contact PyTorch team to confirm whether this change is wanted or not. 
Dec 18 23:45:59 + cleanup 
Dec 18 23:45:59 + retcode=1 
Dec 18 23:45:59 + set +x 
Dec 18 23:45:59 =================== sccache compilation log =================== 
Dec 18 23:45:59 =========== If your build fails, please take a look at the log above for possible reasons =========== 
Dec 18 23:45:59 Compile requests                 0 
Dec 18 23:45:59 Compile requests executed        0 
Dec 18 23:45:59 Cache hits                       0 
Dec 18 23:45:59 Cache misses                     0 
Dec 18 23:45:59 Cache timeouts                   0
This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker.

This comment has been revised 1 time.

@gchanan gchanan requested a review from zou3519 December 19, 2019 00:04
@mingbowan mingbowan merged commit a7d187b into pytorch:v1.4.0 Dec 20, 2019
This was referenced Dec 23, 2019
Closed
Closed
@leofang
Copy link
Contributor

leofang commented Jan 17, 2020

@nairbv Any chance you could update the release note for v1.4.0 to note this bug fix?

@nairbv
Copy link
Collaborator

nairbv commented Jan 17, 2020

@nairbv Any chance you could update the release note for v1.4.0 to note this bug fix?

Updated, let me know if you have a better description I should use: https://github.com/pytorch/pytorch/releases/tag/v1.4.0

@leofang
Copy link
Contributor

leofang commented Jan 18, 2020

Thanks @nairbv 🙏 You could use something like “Fix reading from the interface when strides is missing or None”.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@zou3519 zou3519 zou3519 approved these changes

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@gchanan @kostmo @leofang @nairbv @zou3519 @mingbowan