Advisory

GHSA-m539-j985-hcr8

Also known as: BIT-tensorflow-2021-41196, CVE-2021-41196, PYSEC-2021-389, PYSEC-2021-606, PYSEC-2021-804

Summary

Crash in max_pool3d when size argument is 0 or negative

Details

Impact
The Keras pooling layers can trigger a segfault if the size of the pool is 0 or if a dimension is negative:

import tensorflow as tf

pool_size = [2, 2, 0]
layer = tf.keras.layers.MaxPooling3D(strides=1, pool_size=pool_size)
input_tensor = tf.random.uniform([3, 4, 10, 11, 12], dtype=tf.float32)
res = layer(input_tensor)

This is due to the TensorFlow's implementation of pooling operations where the values in the sliding window are not checked to be strictly positive.

Patches
We have patched the issue in GitHub commit 12b1ff82b3f26ff8de17e58703231d5a02ef1b8b (merging #51975).

The fix will be included in TensorFlow 2.7.0. We will also cherrypick this commit on TensorFlow 2.6.1, TensorFlow 2.5.2, and TensorFlow 2.4.4, 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 has been reported externally via a GitHub issue.

Severity

Medium Basis: The source's own band: MODERATE

Dates

Published
2021-11-10 19:36 UTC
Modified
2026-07-08 06:28 UTC

Affected packages and fixes

tensorflow

PyPI · All systems

Affected

  • introduced 2.6.0, fixed 2.6.1 (ecosystem)

Fixed versions

2.6.1

Fix

  • Fixed version 2.6.1
    pip install --upgrade tensorflow==2.6.1

tensorflow

PyPI · All systems

Affected

  • introduced 2.5.0, fixed 2.5.2 (ecosystem)

Fixed versions

2.5.2

Fix

  • Fixed version 2.5.2
    pip install --upgrade tensorflow==2.5.2

tensorflow

PyPI · All systems

Affected

  • introduced 0, fixed 2.4.4 (ecosystem)

Fixed versions

2.4.4

Fix

  • Fixed version 2.4.4
    pip install --upgrade tensorflow==2.4.4

tensorflow-cpu

PyPI · All systems

Affected

  • introduced 2.6.0, fixed 2.6.1 (ecosystem)

Fixed versions

2.6.1

Fix

  • Fixed version 2.6.1
    pip install --upgrade tensorflow-cpu==2.6.1

tensorflow-cpu

PyPI · All systems

Affected

  • introduced 2.5.0, fixed 2.5.2 (ecosystem)

Fixed versions

2.5.2

Fix

  • Fixed version 2.5.2
    pip install --upgrade tensorflow-cpu==2.5.2

tensorflow-cpu

PyPI · All systems

Affected

  • introduced 0, fixed 2.4.4 (ecosystem)

Fixed versions

2.4.4

Fix

  • Fixed version 2.4.4
    pip install --upgrade tensorflow-cpu==2.4.4

tensorflow-gpu

PyPI · All systems

Affected

  • introduced 2.6.0, fixed 2.6.1 (ecosystem)

Fixed versions

2.6.1

Fix

  • Fixed version 2.6.1
    pip install --upgrade tensorflow-gpu==2.6.1

tensorflow-gpu

PyPI · All systems

Affected

  • introduced 2.5.0, fixed 2.5.2 (ecosystem)

Fixed versions

2.5.2

Fix

  • Fixed version 2.5.2
    pip install --upgrade tensorflow-gpu==2.5.2

tensorflow-gpu

PyPI · All systems

Affected

  • introduced 0, fixed 2.4.4 (ecosystem)

Fixed versions

2.4.4

Fix

  • Fixed version 2.4.4
    pip install --upgrade tensorflow-gpu==2.4.4

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: GitHub Advisory Database (reviewed advisories) · CC BY 4.0

Sources & licences