Skip to content

Refactor IPC tests and enable them for OS provider #690

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

Merged
merged 3 commits into from
Sep 11, 2024

Conversation

vinser52
Copy link
Contributor

@vinser52 vinser52 commented Aug 21, 2024

Description

Minor refactoring of IPC. Enabled IPC tests for the OS provider.

Checklist

  • Code compiles without errors locally
  • All tests pass locally
  • CI workflows execute properly
  • New tests added, especially if they will fail without my changes

@vinser52 vinser52 requested a review from a team as a code owner August 21, 2024 13:27
@vinser52
Copy link
Contributor Author

vinser52 commented Sep 9, 2024

@bratpiorka
As I understand, CI fails because the mmap is failed inside the os_open_ipc_handle function. According to the log, it failed not immediately but after some iterations. I guess it is because the IPC cache is not implemented yet. But locally tests are passed.
Is there a way to change CI runner parameters to allow more virtual memory to be used by the process?

@lukaszstolarczuk
Copy link
Contributor

@bratpiorka As I understand, CI fails because the mmap is failed inside the os_open_ipc_handle function. According to the log, it failed not immediately but after some iterations. I guess it is because the IPC cache is not implemented yet. But locally tests are passed. Is there a way to change CI runner parameters to allow more virtual memory to be used by the process?

The tests that failed are running on the VMs delivered to us, not defined by us. If you can change it in the system - just update our CI workflow; if you can't change it we have to update the test, I guess.

Perhaps, we could also enable this specific test only in a selected workflow - e.g. multi-numa (which has our own machine used for CI).

@vinser52 vinser52 changed the title Enable IPC tests for OS provider Refactor IPC tests and enable them for OS provider Sep 9, 2024
@vinser52
Copy link
Contributor Author

@bratpiorka As I understand, CI fails because the mmap is failed inside the os_open_ipc_handle function. According to the log, it failed not immediately but after some iterations. I guess it is because the IPC cache is not implemented yet. But locally tests are passed. Is there a way to change CI runner parameters to allow more virtual memory to be used by the process?

The tests that failed are running on the VMs delivered to us, not defined by us. If you can change it in the system - just update our CI workflow; if you can't change it we have to update the test, I guess.

Perhaps, we could also enable this specific test only in a selected workflow - e.g. multi-numa (which has our own machine used for CI).

Ok, thank you, @lukaszstolarczuk.
For now I enabled only OS provider with Disjoint pool combination. We can enable tests with a scalable pool once IPC caching is done.

@vinser52 vinser52 force-pushed the svinogra_ipc_tests branch 2 times, most recently from 3ce871a to aa5d33d Compare September 10, 2024 14:42
@vinser52 vinser52 requested a review from ldorau September 10, 2024 14:50
@bratpiorka bratpiorka self-requested a review September 11, 2024 09:18
@lukaszstolarczuk lukaszstolarczuk merged commit 40eea01 into oneapi-src:main Sep 11, 2024
66 checks passed
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