fix: guard against double torch.compile #39
Merged
Merged
Conversation
When SDR runs before LER, the same model object gets torch.compile'd twice, producing a nested OptimizedModule that segfaults during the first forward pass. Skip compilation when the model is already compiled. Also eagerly tear down SDR's DataLoader workers before LER starts to prevent leaked /dev/shm semaphores. Signed-off-by: kvmto <kmato@nvidia.com>
kvmto
requested review from
bmhowe23 and
ivanbasov
and removed request for
ivanbasov
Collaborator
|
Does this fix the nightly CI failure that @ivanbasov was seeing? |
Collaborator
Never mind. I see an internal email now. |
ivanbasov
approved these changes
Apr 2, 2026
ivanbasov
pushed a commit
that referenced
this pull request
Apr 10, 2026
When SDR runs before LER, the same model object gets torch.compile'd twice, producing a nested OptimizedModule that segfaults during the first forward pass. Skip compilation when the model is already compiled. Also eagerly tear down SDR's DataLoader workers before LER starts to prevent leaked /dev/shm semaphores. Signed-off-by: kvmto <kmato@nvidia.com>
ivanbasov
pushed a commit
that referenced
this pull request
Apr 10, 2026
When SDR runs before LER, the same model object gets torch.compile'd twice, producing a nested OptimizedModule that segfaults during the first forward pass. Skip compilation when the model is already compiled. Also eagerly tear down SDR's DataLoader workers before LER starts to prevent leaked /dev/shm semaphores. Signed-off-by: kvmto <kmato@nvidia.com>
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
torch.compilewhen SDR runs before LER on the same model_is_compiled()checkTest plan
PREDECODER_DISABLE_SDR=0 PREDECODER_LER_FINAL_ONLY=0on 2x A6000 — SDR 44.96x, LER 0.003010, no segfault, no leaked semaphores