Skip to content

Updated cudnn_frontend submodule to v0.6 #75481

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
Closed
dzdang wants to merge 1 commit into from
Closed

Updated cudnn_frontend submodule to v0.6 #75481

dzdang wants to merge 1 commit into from

Conversation

@dzdang
Copy link
Contributor

@dzdang dzdang commented Apr 8, 2022

No description provided.

@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented Apr 8, 2022 •
edited
Loading

🔗 Helpful links

💊 CI failures summary and remediations

As of commit 145f08e (more details on the Dr. CI page):

  • 2/2 failures introduced in this PR

🕵️ 2 new failures recognized by patterns

The following CI failures do not appear to be due to upstream breakages

See GitHub Actions build pull / linux-bionic-cuda11.3-py3.7-clang9 / build (1/2)

Step: "Calculate docker image" (full log | diagnosis details | 🔁 rerun)

2022-04-11T20:18:55.6586604Z ERROR: Something h... isn't available for the merge-base of your branch 
2022-04-11T20:18:55.6515848Z ++ git rev-parse HEAD
2022-04-11T20:18:55.6527162Z + [[ 23b8414391b6a649e1a0ef56c841d5855eb42b43 = \1\4\5\f\0\8\e\1\c\d\c\c\f\c\d\9\1\1\f\4\6\3\6\0\7\6\2\9\a\f\2\9\6\4\8\c\1\7\3\3 ]]
2022-04-11T20:18:55.6529947Z ++ git merge-base HEAD 23b8414391b6a649e1a0ef56c841d5855eb42b43
2022-04-11T20:18:55.6556092Z + MERGE_BASE=8ac472910516ba23c3cc7143c5a2a4b4aba3daba
2022-04-11T20:18:55.6556539Z + git rev-parse 8ac472910516ba23c3cc7143c5a2a4b4aba3daba:.circleci/docker
2022-04-11T20:18:55.6568662Z b0a888fe39b5e7f3543e198ce8875507d832dae0
2022-04-11T20:18:55.6572460Z ++ git rev-parse 8ac472910516ba23c3cc7143c5a2a4b4aba3daba:.circleci/docker
2022-04-11T20:18:55.6584636Z + PREVIOUS_DOCKER_TAG=b0a888fe39b5e7f3543e198ce8875507d832dae0
2022-04-11T20:18:55.6585187Z + [[ b0a888fe39b5e7f3543e198ce8875507d832dae0 = \b\0\a\8\8\8\f\e\3\9\b\5\e\7\f\3\5\4\3\e\1\9\8\c\e\8\8\7\5\5\0\7\d\8\3\2\d\a\e\0 ]]
2022-04-11T20:18:55.6586122Z + echo 'ERROR: Something has gone wrong and the previous image isn'\''t available for the merge-base of your branch'
2022-04-11T20:18:55.6586604Z ERROR: Something has gone wrong and the previous image isn't available for the merge-base of your branch
2022-04-11T20:18:55.6587274Z + echo '       contact the PyTorch team to restore the original images'
2022-04-11T20:18:55.6587669Z + exit 1
2022-04-11T20:18:55.6587884Z        contact the PyTorch team to restore the original images
2022-04-11T20:18:55.6597007Z ##[error]Process completed with exit code 1.
2022-04-11T20:18:55.6947434Z Prepare all required actions
2022-04-11T20:18:55.6967147Z ##[group]Run ./.github/actions/teardown-linux
2022-04-11T20:18:55.6967341Z with:
2022-04-11T20:18:55.6967495Z env:
2022-04-11T20:18:55.6967653Z   IN_CI: 1
2022-04-11T20:18:55.6967799Z   IS_GHA: 1

See GitHub Actions build pull / linux-xenial-py3.7-gcc5.4 / test (backwards_compat, 1, 1, linux.2xlarge) (2/2)

Step: "Test" (full log | diagnosis details | 🔁 rerun)

2022-04-11T20:29:55.5060551Z The PR is introduc...m to confirm whether this change is wanted or not. 
2022-04-11T20:29:55.5048224Z processing existing schema:  text(__torch__.torch.classes.profiling.SourceRef _0) -> (str _0)
2022-04-11T20:29:55.5049462Z processing existing schema:  count(__torch__.torch.classes.profiling.InstructionStats _0) -> (int _0)
2022-04-11T20:29:55.5050762Z processing existing schema:  duration_ns(__torch__.torch.classes.profiling.InstructionStats _0) -> (int _0)
2022-04-11T20:29:55.5051775Z processing existing schema:  source(__torch__.torch.classes.profiling.SourceStats _0) -> (__torch__.torch.classes.profiling.SourceRef _0)
2022-04-11T20:29:55.5053634Z processing existing schema:  line_map(__torch__.torch.classes.profiling.SourceStats _0) -> (Dict(int, __torch__.torch.classes.profiling.InstructionStats) _0)
2022-04-11T20:29:55.5054601Z processing existing schema:  __init__(__torch__.torch.classes.profiling._ScriptProfile _0) -> (NoneType _0)
2022-04-11T20:29:55.5056075Z processing existing schema:  enable(__torch__.torch.classes.profiling._ScriptProfile _0) -> (NoneType _0)
2022-04-11T20:29:55.5056915Z processing existing schema:  disable(__torch__.torch.classes.profiling._ScriptProfile _0) -> (NoneType _0)
2022-04-11T20:29:55.5058900Z processing existing schema:  _dump_stats(__torch__.torch.classes.profiling._ScriptProfile _0) -> (__torch__.torch.classes.profiling.SourceStats[] _0)
2022-04-11T20:29:55.5059937Z processing existing schema:  __init__(__torch__.torch.classes.dist_rpc.WorkerInfo _0, str _1, int _2) -> (NoneType _0)
2022-04-11T20:29:55.5060551Z The PR is introducing backward incompatible changes to the operator library. Please contact PyTorch team to confirm whether this change is wanted or not. 
2022-04-11T20:29:55.5060563Z 
2022-04-11T20:29:55.5060667Z Broken ops: [
2022-04-11T20:29:55.5060985Z 	aten::_addmm_activation(Tensor self, Tensor mat1, Tensor mat2, *, Scalar beta=1, Scalar alpha=1, bool use_gelu=False) -> (Tensor)
2022-04-11T20:29:55.5061323Z 	aten::_addmm_activation.out(Tensor self, Tensor mat1, Tensor mat2, *, Scalar beta=1, Scalar alpha=1, bool use_gelu=False, Tensor(a!) out) -> (Tensor(a!))
2022-04-11T20:29:55.5061587Z 	aten::_sparse_csr_sum.dim_dtype(Tensor self, int[1] dim, bool keepdim=False, *, int? dtype=None) -> (Tensor)
2022-04-11T20:29:55.5061805Z 	aten::split.sizes(Tensor(a -> *) self, int[] split_size, int dim=0) -> (Tensor[])
2022-04-11T20:29:55.5062055Z 	aten::gelu_(Tensor(a!) self, *, str approximate="none") -> (Tensor(a!))
2022-04-11T20:29:55.5062119Z ]
2022-04-11T20:29:55.6098992Z + cleanup
2022-04-11T20:29:55.6099127Z + retcode=1
This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

@dzdang dzdang requested a review from jerryzh168 April 8, 2022 02:18
@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented Apr 8, 2022

@dzdang has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

facebook-github-bot pushed a commit that referenced this pull request Apr 8, 2022
@dzdang @facebook-github-bot
Updated cudnn_frontend submodule to v0.6 (#75481)
ea2ca3f 
Summary: Pull Request resolved: #75481

Reviewed By: jerryzh168

Differential Revision: D35489740

Pulled By: dzdang

fbshipit-source-id: b24a71e6774f54e0035d8e3e4b91290b17ba3330
@github-actions
Copy link
Contributor

github-actions bot commented Apr 8, 2022

Hey @dzdang.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

@dzdang dzdang added module: cudnn Related to torch.backends.cudnn, and CuDNN support release notes: cudnn labels Apr 8, 2022
@jjsjann123
Copy link
Collaborator

jjsjann123 commented Apr 8, 2022

hmmm, it's breaking my local build again :)

In file included from /raid/pytorch/cmake/../third_party/cudnn_frontend/include/cudnn_frontend_OperationGraph.h:36:
/raid/pytorch/cmake/../third_party/cudnn_frontend/include/cudnn_frontend_Operation.h:1442:5: error: non-void function does not return a value [-Werror,-Wreturn-type]
}

We are just missing some random return after the error to keep the compiler happy...

https://github.com/NVIDIA/cudnn-frontend/blob/e8e186af7142049db96314b213bd748b8d891ff7/include/cudnn_frontend_Operation.h#L1437-L1440

@dzdang
Copy link
Contributor Author

dzdang commented Apr 8, 2022

@jjsjann123 Ah I see. I'll follow up with cudnn on this. In the mean time, I can probably just revert this so it doesn't block others

@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented Apr 8, 2022

This pull request has been reverted by 771527f. To re-land this change, please open another pull request, assignthe same reviewers, fix the CI failures that caused the revert and make sure that the failing CI runs on the PR by applying the proper ciflow label (e.g., ciflow/trunk).

@jjsjann123
Copy link
Collaborator

jjsjann123 commented Apr 9, 2022

FYI: NVIDIA/cudnn-frontend@fa61199

They are also adding clang to their CI, hopefully similar issues will not happen again 🎉

@dzdang dzdang reopened this Apr 11, 2022
@dzdang dzdang closed this Apr 11, 2022
@dzdang dzdang reopened this Apr 11, 2022
@dzdang dzdang closed this Apr 11, 2022
@dzdang dzdang mentioned this pull request Apr 12, 2022
Closed
pytorchmergebot pushed a commit that referenced this pull request Apr 12, 2022
@dzdang @pytorchmergebot
updated cudnn frontend to v0.6.1
25aa251 
cudnn frontend v0.6.0 had an issue. we upgrade to cudnn frontend to v0.6.1. see reverted PR: #75481
Pull Request resolved: #75615
Approved by: https://github.com/jerryzh168
@WBobby WBobby mentioned this pull request Aug 17, 2022
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@jerryzh168 jerryzh168 jerryzh168 approved these changes

Assignees

No one assigned

Labels

cla signed module: cudnn Related to torch.backends.cudnn, and CuDNN support release notes: cudnn Reverted

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@dzdang @facebook-github-bot @jjsjann123 @jerryzh168