Skip to content

Disable xccl timer avoid drlm hang #1706

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 1 commit into from
May 29, 2025
Merged

Disable xccl timer avoid drlm hang #1706

merged 1 commit into from
May 29, 2025

Conversation

Chao1Han
Copy link
Contributor

No description provided.

@Copilot Copilot AI review requested due to automatic review settings May 28, 2025 08:31
Copy link
Contributor

@Copilot Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull Request Overview

This PR disables the XPUEvent timer to avoid potential hangs by disabling the elapsed_time function and using disabled event instances.

  • Re-initializes all XPUEvent objects with timing disabled
  • Comments out the call to elapsed_time and hardcodes the result to 0

@@ -56,7 +57,10 @@ class XpuTimer : public Timer {

start_event.synchronize();
end_event.synchronize();
float milliseconds = start_event.elapsed_time(end_event);

// Event elapsed_time may cause stuck, disable it for now
Copy link
Preview

Copilot AI May 28, 2025

Choose a reason for hiding this comment

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

Please add a reference to the associated bug report or a TODO comment explaining under what conditions the elapsed_time function can be re-enabled. This extra detail will help future maintainers understand the temporary workaround.

Copilot uses AI. Check for mistakes.

@Chao1Han Chao1Han added this pull request to the merge queue May 29, 2025
Merged via the queue into main with commit 729e1e4 May 29, 2025
7 checks passed
@Chao1Han Chao1Han deleted the xccl/wa_timer branch May 29, 2025 02:03
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.

2 participants