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] Always enable prefetch before CopyTo. #7721

Merged
merged 2 commits into from
Aug 20, 2024

Conversation

mfbalin
Copy link
Collaborator

@mfbalin mfbalin commented Aug 19, 2024

Description

The only difference between the new dataloader CopyTo related code and old code replaced in #7603 is the change in this PR. I am hoping it will fix the regression we are having in the cpu-cpu case.

I don't know why disabling prefetch in the CPU only case would cause a regression.

@frozenbugs can we test if this PR fixes the regression issue for the cpu-cpu case? I can't seem to reliably verify it but I strongly suspect this PR will fix it.

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

Sorry, something went wrong.

@mfbalin mfbalin requested a review from frozenbugs August 19, 2024 18:35
@dgl-bot
Copy link
Collaborator

dgl-bot commented Aug 19, 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 Aug 19, 2024

Commit ID: a7ccc7f

Build ID: 1

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

@frozenbugs
Copy link
Collaborator

Yes, this PR fixes the issue.

'graphbolt' 'exclude_edges' 'cpu-cpu' 133.32784
'graphbolt' 'not_exclude_edges' 'cpu-cpu' 135.22002

@frozenbugs
Copy link
Collaborator

Please clean up the code and merge it.

@mfbalin
Copy link
Collaborator Author

mfbalin commented Aug 20, 2024

Yes, this PR fixes the issue.

'graphbolt' 'exclude_edges' 'cpu-cpu' 133.32784 'graphbolt' 'not_exclude_edges' 'cpu-cpu' 135.22002

Thanks for verifying. For some reason, I couldn't reproduce the regression locally.

@dgl-bot
Copy link
Collaborator

dgl-bot commented Aug 20, 2024

Commit ID: 204d7acbc774e1ab0af69b00201f7724f8da11fc

Build ID: 2

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

@mfbalin mfbalin merged commit e5a5d76 into dmlc:master Aug 20, 2024
2 checks passed
@mfbalin mfbalin deleted the gb_cpu_always_enable_prefetch branch August 20, 2024 17:46
lijialin03 pushed a commit to lijialin03/dgl that referenced this pull request Jan 6, 2025
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.

None yet

3 participants