Skip to content

[MLU] fix: reuse ckpt send/recv buffers to avoid OOM - #16

Open
uqyxx wants to merge 2 commits into
verl-project:mainfrom
uqyxx:mlu-fix-ckpt-buffer-reuse
Open

uqyxx wants to merge 2 commits into
verl-project:mainfrom
uqyxx:mlu-fix-ckpt-buffer-reuse

Conversation

@uqyxx

@uqyxx uqyxx commented Sep 15, 2026

Copy link
Copy Markdown
Contributor

Summary

  1. Reuse ckpt send/recv buffers to avoid OOM
  2. set_expandable_segments(False) before buffer to avoid memory register error

Testing

  • pytest tests/ -v passes
  • Manually verified on target hardware (if applicable)

Acceptance Baseline (for new hardware adaptation PRs)

  • Ran scripts/baseline_grpo_gsm8k.sh on target hardware (8 devices)
  • Training completed all epochs without error
  • critic/rewards/mean shows clear upward trend in first 100 steps
  • Curve is consistent with NVIDIA reference

SwanLab or training log link:

Reward curve comparison (first 100 steps):

Checklist

  • Code follows the project's style and passes pre-commit checks
  • Documentation updated (if applicable)
  • No secrets or credentials included

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