Skip to main content
← All vulnerabilities
CVE-2022-41909PYSEC-2026-3368BIT-TENSORFLOW-2022-41909GHSA-RJX6-V474-2CH9PYSEC-2026-1034PYSEC-2026-3238

Segfault in `CompositeTensorVariantToComponents` in Tensorflow

Published
Last modified
published
Not listed in CISA KEV
Fix
Fixed in 2.10.1, 2.9.3, 2.8.4
None on Exploit-DB, Metasploit or GitHub
  • Exploitation: poc
  • Automatable: no
  • Technical impact: partial

Summary

TensorFlow is an open source platform for machine learning. An input `encoded` that is not a valid `CompositeTensorVariant` tensor will trigger a segfault in `tf.raw_ops.CompositeTensorVariantToComponents`. We have patched the issue in GitHub commits bf594d08d377dc6a3354d9fdb494b32d45f91971 and 660ce5a89eb6766834bdc303d2ab3902aef99d3d. The fix will be included in TensorFlow 2.11. We will also cherrypick this commit on TensorFlow 2.10.1, 2.9.3, and TensorFlow 2.8.4, as these are also affected and still in supported range.

Timeline

  1. CVE published
  2. CISA SSVC assessment

    Exploitation: poc · Automatable: no · Technical impact: partial

  3. Record last updated

Affected versions

  • tensorflow tensorflow

    From 2.10.0 · fixed in 2.10.1; From 2.9.0 · fixed in 2.9.3; fixed in 2.8.4

  • tensorflow-cpuPyPI

    fixed in 2.8.4; From 2.9.0 · fixed in 2.9.3; From 2.10.0 · fixed in 2.10.1

  • tensorflowPyPI

    fixed in 2.8.4; From 2.9.0 · fixed in 2.9.3; From 2.10.0 · fixed in 2.10.1

  • tensorflowPyPI

    fixed in 2.8.4

  • tensorflowPyPI

    From 2.9.0 · fixed in 2.9.3

  • tensorflowPyPI

    From 2.10.0 · fixed in 2.10.1

  • tensorflow-cpuPyPI

    fixed in 2.8.4

  • tensorflow-gpuPyPI

    fixed in 2.8.4

  • tensorflow-cpuPyPI

    From 2.9.0 · fixed in 2.9.3

  • tensorflow-gpuPyPI

    From 2.9.0 · fixed in 2.9.3

  • tensorflow-cpuPyPI

    From 2.10.0 · fixed in 2.10.1

  • tensorflow-gpuPyPI

    From 2.10.0 · fixed in 2.10.1

  • tensorflow-gpuPyPI

    fixed in 2.8.4; From 2.9.0 · fixed in 2.9.3; From 2.10.0 · fixed in 2.10.1

References 4

GitHub · tensorflow/tensorflow