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.

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

[GraphBolt] torch.compile() support for gb.expand_indptr. #7188

Merged
merged 11 commits into from
Mar 5, 2024

Conversation

mfbalin
Copy link
Collaborator

@mfbalin mfbalin commented Mar 2, 2024

Description

Details in this document by PyTorch developers: https://docs.google.com/document/d/1W--T6wz8IY8fOI0Vm8BF44PdBgs283QvpelJZWieQWQ/edit?usp=sharing

This PR utilizes torch>=2.2.0 features.

Checklist

Please feel free to remove inapplicable items for your PR.

  • The PR title starts with [$CATEGORY] (such as [NN], [Model], [Doc], [Feature]])
  • I've leverage the tools to beautify the python and c++ code.
  • The PR is complete and small, read the Google eng practice (CL equals to PR) to understand more about small PR. In DGL, we consider PRs with less than 200 lines of core code change are small (example, test and documentation could be exempted).
  • All changes have test coverage
  • Code is well-documented
  • To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change
  • Related issue is referred in this PR
  • If the PR is for a new model/paper, I've updated the example index here.

Changes

@mfbalin mfbalin requested a review from frozenbugs March 2, 2024 03:17
@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 2, 2024

To trigger regression tests:

  • @dgl-bot run [instance-type] [which tests] [compare-with-branch];
    For example: @dgl-bot run g4dn.4xlarge all dmlc/master or @dgl-bot run c5.9xlarge kernel,api dmlc/master

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 2, 2024

Commit ID: 71b0673

Build ID: 1

Status: ❌ CI test failed in Stage [Lint Check].

Report path: link

Full logs path: link

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 2, 2024

Commit ID: 3001f42

Build ID: 2

Status: ❌ CI test failed in Stage [Lint Check].

Report path: link

Full logs path: link

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 2, 2024

Commit ID: abc7c6c

Build ID: 3

Status: ❌ CI test failed in Stage [CPU Build].

Report path: link

Full logs path: link

@mfbalin mfbalin removed the request for review from frozenbugs March 2, 2024 03:55
@mfbalin mfbalin marked this pull request as draft March 2, 2024 03:55
@mfbalin
Copy link
Collaborator Author

mfbalin commented Mar 2, 2024

There is an issue, I hope pytorch developers can help fix it: pytorch/pytorch#121080

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 3, 2024

Commit ID: 3a55cb7

Build ID: 4

Status: ❌ CI test failed in Stage [CPU Build].

Report path: link

Full logs path: link

@mfbalin mfbalin marked this pull request as ready for review March 3, 2024 01:46
@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 3, 2024

Commit ID: a0a2b45

Build ID: 5

Status: ❌ CI test failed in Stage [CPU Build].

Report path: link

Full logs path: link

@frozenbugs
Copy link
Collaborator

@mfbalin can you fix the build?

@mfbalin
Copy link
Collaborator Author

mfbalin commented Mar 4, 2024

@mfbalin can you fix the build?

I can not, we need to start depending on Torch 2.2 instead of 2.0. The CI needs to be updated.

@frozenbugs
Copy link
Collaborator

@mfbalin can you fix the build?

I can not, we need to start depending on Torch 2.2 instead of 2.0. The CI needs to be updated.

We are deprecating Torch 1.3 (oldest Torch version: 2.0) in DGL 2.1, likely we will not move to 2.2 in near future. If it requires 2.2, let's table the change until then?

@mfbalin
Copy link
Collaborator Author

mfbalin commented Mar 4, 2024

@mfbalin can you fix the build?

I can not, we need to start depending on Torch 2.2 instead of 2.0. The CI needs to be updated.

We are deprecating Torch 1.3 (oldest Torch version: 2.0) in DGL 2.1, likely we will not move to 2.2 in near future. If it requires 2.2, let's table the change until then?

Sure. Which release of DGL do you think will move to 2.2 do you think?

@mfbalin
Copy link
Collaborator Author

mfbalin commented Mar 4, 2024

@mfbalin can you fix the build?

I can not, we need to start depending on Torch 2.2 instead of 2.0. The CI needs to be updated.

We are deprecating Torch 1.3 (oldest Torch version: 2.0) in DGL 2.1, likely we will not move to 2.2 in near future. If it requires 2.2, let's table the change until then?

Also, it may be possible to still get this PR through by adding version checks on certain parts of the code. I will see if it is easily doable. Then, maybe we won't have to wait that long.

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 4, 2024

Commit ID: e412403e361f6b0a9f6038a3808cfa71269855b6

Build ID: 6

Status: ❌ CI test failed in Stage [CPU Build (Win64)].

Report path: link

Full logs path: link

@mfbalin
Copy link
Collaborator Author

mfbalin commented Mar 4, 2024

The CI will hopefully pass now.

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 4, 2024

Commit ID: 0b55f4cabebb9a4b92c94f22f2c0e1c2434fbea4

Build ID: 7

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 4, 2024

Commit ID: c5b0bf9f27a98702bc30604105087ef8c9a47dee

Build ID: 8

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 4, 2024

Commit ID: 996b735

Build ID: 9

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 4, 2024

Commit ID: 9a2350c

Build ID: 10

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

@mfbalin
Copy link
Collaborator Author

mfbalin commented Mar 4, 2024

@frozenbugs got the CI to pass. I think it is ready to review.

@mfbalin
Copy link
Collaborator Author

mfbalin commented Mar 4, 2024

@Rhett-Ying Can we add torch 2.2 to the CI also so that the code that is enabled only for 2.2 can be tested? For now, I tested locally with a docker container that has 2.2.0a0+81ea7a4.

@dgl-bot
Copy link
Collaborator

dgl-bot commented Mar 5, 2024

Commit ID: 42210fdadb90c054ca3c89129474633a7f225705

Build ID: 11

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

@mfbalin mfbalin merged commit bc97873 into dmlc:master Mar 5, 2024
2 checks passed
@mfbalin mfbalin deleted the gb_expand_indptr_torch_compile branch March 5, 2024 05:26
This pull request was closed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants