xla
Fix distributed checkpoint errors when PyTorch CUDA is enabled.
#6421
Merged

Fix distributed checkpoint errors when PyTorch CUDA is enabled. #6421

ysiraichi
ysiraichi Pass `per_thread_copy_ahead=0` to avoid `_OverlappingCpuLoader`.
8bb5ada9
ysiraichi ysiraichi added xla:gpu
ysiraichi ysiraichi requested a review from jonb377 jonb377 1 year ago
ysiraichi ysiraichi requested a review from JackCaoG JackCaoG 1 year ago
jonb377
ysiraichi
ysiraichi
jonb377
jonb377 approved these changes on 2024-01-31
ysiraichi ysiraichi merged 85c66f5f into master 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone