Skip to content

CVE-2022-29210 Heap buffer overflow due to incorrect hash function in TensorFlow

MediumCVSS 5.5 EPSS 0.23%
Published
Updated
Exploit status
No public exploit known
CISA SSVC
No known exploitation · automatable: no · technical impact: partial
Technology
Debian

Description

TensorFlow is an open source platform for machine learning. In version 2.8.0, the TensorKey hash function used total estimated AllocatedBytes(), which (a) is an estimate per tensor, and (b) is a very poor hash function for constants (e.g. int32_t). It also tried to access individual tensor bytes through tensor.data() of size AllocatedBytes(). This led to ASAN failures because the AllocatedBytes() is an estimate of total bytes allocated by a tensor, including any pointed-to constructs (e.g. strings), and does not refer to contiguous bytes in the .data() buffer. The discoverers could not use this byte vector anyway because types such as tstring include pointers, whereas they needed to hash the string values themselves. This issue is patched in Tensorflow versions 2.9.0 and 2.8.1.

Upgrade path

5 packages: 3 fixed.

PackageAffectedFixed inStatus
pypi tensorflow(PyPI) = 2.8.0; >= 2.8.0, < 2.8.1 2.8.1 Fixed
pypi tensorflow-cpu(PyPI) = 2.8.0; >= 2.8.0, < 2.8.1 2.8.1 Fixed
pypi tensorflow-gpu(PyPI) = 2.8.0; >= 2.8.0, < 2.8.1 2.8.1 Fixed
Google tensorflow = 2.8.0 No fix listed Versions listed
TensorFlow = == 2.8.0 No fix listed Versions listed

Affected products

ProductAffected versionsFixed in
pypi tensorflow(PyPI) >= 2.8.0, < 2.8.1 2.8.1
pypi tensorflow-cpu(PyPI) >= 2.8.0, < 2.8.1 2.8.1
pypi tensorflow-gpu(PyPI) >= 2.8.0, < 2.8.1 2.8.1
pypi tensorflow(PyPI) = 2.8.0 No fix listed
pypi tensorflow-cpu(PyPI) = 2.8.0 No fix listed
pypi tensorflow-gpu(PyPI) = 2.8.0 No fix listed
TensorFlow = == 2.8.0 No fix listed
Google tensorflow = 2.8.0 No fix listed

References