ExecuTorch integer overflow vulnerability
Meta Platforms, Inc ExecuTorch
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| executorch PyPI |
|
|
| org.pytorch:executorch-android Maven |
|
|
| github.com/pytorch/executorch Swift |
Meta Platforms, Inc ExecuTorch
| Product | Affected | Fixed in |
|---|---|---|
| executorch PyPI |
|
|
| org.pytorch:executorch-android Maven |
|
|
| github.com/pytorch/executorch Swift |
< 0.7.0
|
An integer overflow vulnerability in the loading of ExecuTorch models can cause objects to be placed outside their allocated memory area, potentially resulting in code execution or other undesirable effects. This issue affects ExecuTorch prior to commit 0830af8207240df8d7f35b984cdf8bc35d74fa73.
From the CVE record · facebook
An integer overflow while loading ExecuTorch models can place objects outside allocated memory. This could result in code execution or other undesirable effects when a model is loaded. The input does not specify network access, authentication, or user-interaction requirements.
Engineers using PyPI executorch, Maven org.pytorch:executorch-android, or SwiftURL github.com/pytorch/executorch versions earlier than 0.7.0 should check their deployments.