Advisory record · GHSA-5v77-j66x-4c4g

Missing validation causes denial of service via `Conv3DBackpropFilterV2`

CVSS 5.5 MediumEPSS 0.0032 · 25th percentile

Published by the GitHub Advisory Database at 24 May 2022, 22:07 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

PackageEcosystemAffectedFixed in
tensorflowPyPI<2.6.4 || >=2.7.0, <2.7.2 || >=2.8.0, <2.8.12.6.4, 2.7.2, 2.8.1
tensorflow-cpuPyPI<2.6.4 || >=2.7.0, <2.7.2 || >=2.8.0, <2.8.12.6.4, 2.7.2, 2.8.1
tensorflow-gpuPyPI<2.6.4 || >=2.7.0, <2.7.2 || >=2.8.0, <2.8.12.6.4, 2.7.2, 2.8.1
Identifiers
GHSA-5v77-j66x-4c4g · BIT-tensorflow-2022-29196 · CVE-2022-29196 · PYSEC-2026-3110 · PYSEC-2026-3278 · PYSEC-2026-953
Weakness (CWE)
CWE-20 · CWE-1284
Published
24 May 2022
Updated
13 Jul 2026

Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

TensorFlow, tensorflow-cpu and tensorflow-gpu versions before 2.6.4, from 2.7.0 before 2.7.2, and from 2.8.0 before 2.8.1 are affected. The Conv3DBackpropFilterV2 operation does not check that its filter_sizes argument is a vector, which triggers a CHECK failure. An attacker who can call this operation can crash the process, causing a denial of service. Upgrade to TensorFlow 2.6.4, 2.7.2, 2.8.1, or a later release.

Written 6 Nov 2026.

Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

Severity and exploitation

ScoreRatingCVSSVectorSource
5.5 Medium 3.1 CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H CNA (GitHub_M)
  • CISA KEV Not listed.
  • EPSS 0.0032, the 25th percentile, on 9 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

Timeline

  1. 31 Oct 2026, 14:59 UTC First seen by agentprime
  2. 31 Oct 2026, 14:59 UTC Confirmed by OSV

Sources

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.