Skip to content
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

add the function specialization for promote with ITensorListRef #87756

Closed
wants to merge 7 commits into from
Closed

add the function specialization for promote with ITensorListRef #87756

wants to merge 7 commits into from

Conversation

leslie-fang-intel
Copy link
Collaborator

Fixes #87684
It's due to a new tensor list type is introduced as ITensorListRef. We need the function specialization for prioritize and cached_cast for this new tensor list type.

@pytorch-bot
Copy link

pytorch-bot bot commented Oct 26, 2022 •
edited

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/87756

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit a29e213:
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@leslie-fang-intel leslie-fang-intel added the ciflow/trunk Trigger trunk jobs on your pull request label Oct 26, 2022
jgong5
jgong5 reviewed Oct 26, 2022
View reviewed changes
test/test_jit_autocast.py Outdated Show resolved Hide resolved
jgong5
jgong5 requested changes Oct 26, 2022
View reviewed changes
test/test_jit_autocast.py Outdated Show resolved Hide resolved
jgong5
jgong5 requested changes Oct 27, 2022
View reviewed changes
test/test_jit_autocast.py Outdated Show resolved Hide resolved
@leslie-fang-intel leslie-fang-intel marked this pull request as ready for review October 27, 2022 04:21
@leslie-fang-intel
Copy link
Collaborator Author

@ezyang Could you take a look of this PR?

ezyang
ezyang approved these changes Oct 27, 2022
View reviewed changes
test/test_jit_autocast.py
self.assertTrue(c.dtype, torch.float32)
self.assertTrue(c2.dtype, torch.float32)
traced_graph = traced.graph_for(a, b)
self.assertTrue(any(n.kind() == "aten::to" for n in traced_graph.nodes()))
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

An easier way to write this test is use logging tensor mode and expect test on the trace

@ezyang
Copy link
Contributor

ezyang commented Oct 27, 2022

@pytorchbot merge

@pytorchmergebot
Copy link
Collaborator

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging
Check the merge workflow status
here

@pytorchmergebot
Copy link
Collaborator

Merge failed

Reason: 2 additional jobs have failed, first few of them are: trunk ,trunk / libtorch-linux-bionic-cuda11.6-py3.7-gcc7 / build

Details for Dev Infra team Raised by workflow job

@leslie-fang-intel
Copy link
Collaborator Author

There is one CI failure. The error message shows out of memory during build process of the system.
It looks irrelevant of the changes in this PR.

2022-10-27T04:46:26.5853309Z virtual memory exhausted: Cannot allocate memory
2022-10-27T04:46:26.9950032Z Makefile:73: recipe for target '/var/lib/jenkins/cpp-build/caffe2/build/nccl/obj/collectives/device/devlink.o' failed
2022-10-27T04:46:26.9961694Z make[5]: *** [/var/lib/jenkins/cpp-build/caffe2/build/nccl/obj/collectives/device/devlink.o] Error 1
2022-10-27T04:46:26.9962560Z make[5]: Leaving directory '/var/lib/jenkins/workspace/third_party/nccl/nccl/src/collectives/device'
2022-10-27T04:46:26.9963382Z Makefile:51: recipe for target '/var/lib/jenkins/cpp-build/caffe2/build/nccl/obj/collectives/device/colldevice.a' failed
2022-10-27T04:46:26.9964136Z make[4]: *** [/var/lib/jenkins/cpp-build/caffe2/build/nccl/obj/collectives/device/colldevice.a] Error 2
2022-10-27T04:46:26.9964881Z make[4]: Leaving directory '/var/lib/jenkins/workspace/third_party/nccl/nccl/src'
2022-10-27T04:46:26.9965447Z Makefile:25: recipe for target 'src.build' failed
2022-10-27T04:46:26.9977910Z make[3]: *** [src.build] Error 2
2022-10-27T04:46:26.9978680Z make[3]: Leaving directory '/var/lib/jenkins/workspace/third_party/nccl/nccl'
2022-10-27T04:46:26.9979572Z CMakeFiles/nccl_external.dir/build.make:85: recipe for target 'nccl_external-prefix/src/nccl_external-stamp/nccl_external-build' failed
2022-10-27T04:46:26.9980050Z make[2]: *** [nccl_external-prefix/src/nccl_external-stamp/nccl_external-build] Error 2
2022-10-27T04:46:26.9998276Z make[2]: Leaving directory '/var/lib/jenkins/cpp-build/caffe2/build'
2022-10-27T04:46:27.0003010Z CMakeFiles/Makefile2:2035: recipe for target 'CMakeFiles/nccl_external.dir/all' failed
2022-10-27T04:46:27.0003341Z make[1]: *** [CMakeFiles/nccl_external.dir/all] Error 2
2022-10-27T04:46:27.0003587Z make[1]: *** Waiting for unfinished jobs....

@sanchitintel sanchitintel added the intel This tag is for PR from Intel label Oct 27, 2022
@leslie-fang-intel
Copy link
Collaborator Author

@pytorchbot merge

@pytorchmergebot
Copy link
Collaborator

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging
Check the merge workflow status
here

@github-actions
Copy link

Hey @leslie-fang-intel.
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.

sgrigory pushed a commit to sgrigory/pytorch that referenced this pull request Oct 28, 2022
@leslie-fang-intel @sgrigory
add the function specialization for promote with ITensorListRef (pyto…
7e9c965 
…rch#87756)

Fixes [pytorch#87684](pytorch#87684)
It's due to a new tensor list type is introduced as `ITensorListRef`.  We need the function specialization for `prioritize` and `cached_cast` for this new tensor list type.

Pull Request resolved: pytorch#87756
Approved by: https://github.com/jgong5, https://github.com/ezyang
kulinseth pushed a commit to kulinseth/pytorch that referenced this pull request Nov 5, 2022
@leslie-fang-intel @kulinseth
add the function specialization for promote with ITensorListRef (pyto…
497cb7a 
…rch#87756)

Fixes [pytorch#87684](pytorch#87684)
It's due to a new tensor list type is introduced as `ITensorListRef`.  We need the function specialization for `prioritize` and `cached_cast` for this new tensor list type.

Pull Request resolved: pytorch#87756
Approved by: https://github.com/jgong5, https://github.com/ezyang
kulinseth pushed a commit to kulinseth/pytorch that referenced this pull request Dec 10, 2022
@leslie-fang-intel @kulinseth
add the function specialization for promote with ITensorListRef (pyto…
b447722 
…rch#87756)

Fixes [pytorch#87684](pytorch#87684)
It's due to a new tensor list type is introduced as `ITensorListRef`.  We need the function specialization for `prioritize` and `cached_cast` for this new tensor list type.

Pull Request resolved: pytorch#87756
Approved by: https://github.com/jgong5, https://github.com/ezyang
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ezyang ezyang ezyang approved these changes

@jgong5 jgong5 jgong5 approved these changes

@CaoE CaoE Awaiting requested review from CaoE

@sanchitintel sanchitintel Awaiting requested review from sanchitintel

Assignees
No one assigned
Labels
ciflow/trunk Trigger trunk jobs on your pull request intel This tag is for PR from Intel Merged open source
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Eager mode CPU Autocast implementation for torch.cat seems to be broken
6 participants
@leslie-fang-intel @ezyang @pytorchmergebot @jgong5 @sanchitintel @pytorchbot