Security: Unsafe deserialization via torch.load in checkpoint loader#483
Open
tuanaiseo wants to merge 2 commits intoopen-mmlab:mainfrom
Open
Conversation
The checkpoint loader directly calls `torch.load(filepath, map_location=device)` on a path without trust validation. `torch.load` uses pickle under the hood and can execute arbitrary code when loading a malicious checkpoint. Affected files: utils.py, utilities.py Signed-off-by: tuanaiseo <221258316+tuanaiseo@users.noreply.github.com>
The checkpoint loader directly calls `torch.load(filepath, map_location=device)` on a path without trust validation. `torch.load` uses pickle under the hood and can execute arbitrary code when loading a malicious checkpoint. Affected files: utils.py, utilities.py Signed-off-by: tuanaiseo <221258316+tuanaiseo@users.noreply.github.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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
The checkpoint loader directly calls
torch.load(filepath, map_location=device)on a path without trust validation.torch.loaduses pickle under the hood and can execute arbitrary code when loading a malicious checkpoint.Severity:
highFile:
models/tta/ldm/inference_utils/utils.pySolution
Treat checkpoints as untrusted input. Prefer
safetensorsfor model weights, or enforce strict trust boundaries and signature/hash verification before loading. If using PyTorch >=2.0, consider safer loading patterns and avoid arbitrary pickled objects.Changes
models/tta/ldm/inference_utils/utils.py(modified)models/tta/picoaudio/picoaudio/audioldm/hifigan/utilities.py(modified)Testing