Skip to content
OpenAdvisories
Medium · 4.3 CVE-2022-23578 BIT-tensorflow-2022-23578EUVD-2022-0326GHSA-8r7c-3cm2-3h8fPYSEC-2022-142PYSEC-2022-87PYSEC-2026-3138

Memory leak in Tensorflow

Published Feb 4, 2022 · Updated Apr 22, 2025 · Assigned by GitHub_M

Vendor: TensorFlow Product: TensorFlow Package: tensorflow Package: tensorflow-cpu Package: tensorflow-gpu Ecosystem: PyPI

Description

Tensorflow is an Open Source Machine Learning Framework. If a graph node is invalid, TensorFlow can leak memory in the implementation of `ImmutableExecutorState::Initialize`. Here, we set `item->kernel` to `nullptr` but it is a simple `OpKernel*` pointer so the memory that was previously allocated to it would leak. The fix will be included in TensorFlow 2.8.0. We will also cherrypick this commit on TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3, as these are also affected and still in supported range.

Affected products

VendorProductAffectedFixed
tensorflow tensorflow >= 2.7.0, < 2.7.1, < 2.5.3, >= 2.6.0, < 2.6.3 —

Affected packages

EcosystemPackageAffectedFixedRecord
PyPI tensorflow < 2.5.3 2.5.3 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow >= 2.6.0, < 2.6.3 2.6.3 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow >= 2.7.0, < 2.7.1 2.7.1 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow < 2.5.3 · >= 2.6.0, < 2.6.3 · >= 2.7.0, < 2.7.1 2.5.3, 2.6.3, 2.7.1 PYSEC-2026-3138
PyPI tensorflow-cpu < 2.5.3 2.5.3 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow-cpu >= 2.6.0, < 2.6.3 2.6.3 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow-cpu >= 2.7.0, < 2.7.1 2.7.1 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow-cpu < 2.5.3 · >= 2.6.0, < 2.6.3 2.5.3, 2.6.3 PYSEC-2022-87
PyPI tensorflow-gpu < 2.5.3 2.5.3 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow-gpu >= 2.6.0, < 2.6.3 2.6.3 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow-gpu >= 2.7.0, < 2.7.1 2.7.1 GHSA-8r7c-3cm2-3h8f
PyPI tensorflow-gpu < 2.5.3 · >= 2.6.0, < 2.6.3 2.5.3, 2.6.3 PYSEC-2022-142

References