Browse / AI Framework and Platform Flaws / CVE-2022-41911
Invalid char to bool conversion when printing a tensor
- Class
- AI Framework and Platform Flaws (filed by a model that read the advisory)
- Technique
- Other weaknesses
- Affected
- tensorflow-cpu, tensorflow (PyPI)
- CVSS
- CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:N/I:N/A:H
- EPSS
- 0.6% chance of exploitation in 30 days
How it works, in short
Tensor printing casts raw char values to bool, which is undefined behaviour for values other than zero or one and crashes sanitizer builds. (Our summary, written by a model. The source text is below.)
Details
Impact
When printing a tensor, we get it's data as a const char* array (since that's the underlying storage) and then we typecast it to the element type. However, conversions from char to bool are undefined if the char is not 0 or 1, so sanitizers/fuzzers will crash.
Patches
We have patched the issue in GitHub commit 1be743703279782a357adbf9b77dcb994fe8b508.
The fix will be included in TensorFlow 2.11.0. We will also cherrypick this commit on TensorFlow 2.10.1, TensorFlow 2.9.3, and TensorFlow 2.8.4, as these are also affected and still in supported range.
For more information
Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.
Attribution
This vulnerability was discovered via internal fuzzing.
References
- https://github.com/tensorflow/tensorflow/security/advisories/GHSA-pf36-r9c6-h97j
- https://nvd.nist.gov/vuln/detail/CVE-2022-41911
- https://github.com/tensorflow/tensorflow/commit/1be743703279782a357adbf9b77dcb994fe8b508
- https://github.com/tensorflow/tensorflow
- https://github.com/tensorflow/tensorflow/blob/807cae8a807960fd7ac2313cde73a11fc15e7942/tensorflow/core/framework/tensor.cc#L1200-L1227
Source: GitHub Advisory Database, used under CC BY 4.0.