#9837: Assign workers after performing ref count cleanup in async mode #9944
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
tt-asaigal
requested review from
arakhmati,
eyonland,
cfjchu and
xanderchin
as code owners
arakhmati
approved these changes
Jul 4, 2024
- This handles cases where a device tensor is reassigned to a host tensor
- Exposed during model cache generation which uses the following pattern:
device_tensor = device_tensor.cpu()
tt-asaigal
force-pushed
the
asaigal/cache_gen_segfault_rebased
branch
from
4558673 to
112ea79
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Ticket
#9837
Problem description
Segfault seen when generating rotation matrix cache for LLMs. Was never seen on CI, since these runners have the cache already created.
What's changed
Reference count management for tensors was erroneous. This case was not correctly supported:
device_tensor = device_tensor.cpu()The reference count for the device tensor would not get decremented post assignment.
Assign worker vector after decrementing ref count when calling the tensor copy or move assignment operators.
Checklist