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 --disable-gpu to testem.js for chrome #2193

Merged
merged 1 commit into from
Dec 10, 2024
Merged

add --disable-gpu to testem.js for chrome #2193

merged 1 commit into from
Dec 10, 2024
+6 −0

Conversation

mansona
Copy link
Member

@mansona mansona commented Dec 3, 2024

In a lot of our CI recently we are having issues with windows and there is often this output in the build:

Created TensorFlow Lite XNNPACK delegate for CPU.
Attempting to use a delegate that only supports static-sized tensors with a graph that has dynamic-sized tensors (tensor#58 is a dynamic-sized tensor).

From what I can see this has something to do with the fact that Chrome added tensorflow and I have also seen something to suggest that the GPU initialisation can sometimes go wrong. My theory is that adding --disable-gpu might solve this 🤷

@ef4
Copy link
Contributor

ef4 commented Dec 9, 2024

I'm guessing from the three CI runs here that it didn't help?

@patricklx
Copy link
Contributor

The failing test is unrelated.
I also tried this before with success #2117

But i also read that the error is visible when running chrome in debug mode. So it might not be because of tensor flow.
But if it fixes it

@mansona
Copy link
Member Author

mansona commented Dec 10, 2024

@ef4 so the other failures were not related to the TensorFlow warning at all. The first failure was this:

not ok 4 v2-addon-dev-watch > Watching the addon via rollup -c -w > files are correctly synced > deleting hbs file updates dist component file
  ---
  message: "Expected D:\\temp\\tmp-3516nKoVUAQf77Gq\\dist\\components\\demo.js to be modified. Latest: 1733249428004.746, previously: 1733249428004.746"
  severity: failed
  actual  : 1733249428004.746
  expected: 1733249428004.746
  stack: |
        at becomesModified (D:\a\embroider\embroider\tests\scenarios\helpers\filesystem.ts:[18](https://github.com/embroider-build/embroider/actions/runs/12145413687/job/33984909683#step:5:19):10)
        at async Object.<anonymous> (D:\a\embroider\embroider\tests\scenarios\v2-addon-dev-watch-test.ts:225:13)

which is a test that depends on change time of a file (which seems very brittle 🫠 )

The second failure was the same test but a different error:

not ok 4 v2-addon-dev-watch > Watching the addon via rollup -c -w > files are correctly synced > deleting hbs file updates dist component file
  ---
  message: "Promise rejected during \"deleting hbs file updates dist component file\": Could not load D:\\temp\\tmp-3468jRGf7bn5Y9bu\\src\\components\\demo.hbs (imported by src/components/demo.js): ENOENT: no such file or directory, open 'D:\\temp\\tmp-3468jRGf7bn5Y9bu\\src\\components\\demo.hbs'"
  severity: failed
  actual  : null
  expected: undefined
  stack: |
    Error: Could not load D:\temp\tmp-3468jRGf7bn5Y9bu\src\components\demo.hbs (imported by src/components/demo.js): ENOENT: no such file or directory, open 'D:\temp\tmp-3468jRGf7bn5Y9bu\src\components\demo.hbs'
        at async open (node:internal/fs/promises:639:25)
        at async Object.readFile (node:internal/fs/promises:1[24](https://github.com/embroider-build/embroider/actions/runs/12145413687/job/33984909683#step:5:25)2:14)
        at async D:\a\embroider\embroider\node_modules\.pnpm\[email protected]\node_modules\rollup\dist\shared\rollup.js:25766:128
        at async Queue.work (D:\a\embroider\embroider\node_modules\.pnpm\[email protected]\node_modules\rollup\dist\shared\rollup.js:[26](https://github.com/embroider-build/embroider/actions/runs/12145413687/job/33984909683#step:5:27)145:32)

I wonder if that test is just written in a way that is too flaky to be reliable 🤔

As for if the "no-gpu" actually helps anything... it's hard to say. We would need to merge it and do some sort of statistical analysis of before and after to get a good sense of if we need it 🙈

@ef4 ef4 merged commit 48f6e51 into main Dec 10, 2024
211 checks passed
@ef4 ef4 deleted the disable-gpu branch December 10, 2024 14:21
@github-actions github-actions bot mentioned this pull request Dec 10, 2024
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
internal
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@mansona @ef4 @patricklx