Vulnerability · Published Sep 16, 2022 · Assigned by GitHub_M
CVE-2022-35982Segfault in `SparseBincount` in TensorFlow
Medium 5.9
In plain English
A medium-severity improper input validation in tensorflow from tensorflow. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 0.5% (higher than 44% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.
Official description
TensorFlow is an open source platform for machine learning. If `SparseBincount` is given inputs for `indices`, `values`, and `dense_shape` that do not make a valid sparse tensor, it results in a segfault that can be used to trigger a denial of service attack. We have patched the issue in GitHub commit 40adbe4dd15b582b0210dfbf40c243a62f5119fa. 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. There are no known workarounds for this issue.
Affected packages & fixed versions
| Ecosystem | Package | Fixed in | Advisory |
|---|---|---|---|
| PyPI | tensorflow | 2.7.2, 2.8.1, 2.9.1 | PYSEC-2026-3093 |
| PyPI | tensorflow | 2.7.2 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow | 2.8.1 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow | 2.9.1 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow-cpu | 2.8.1 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow-cpu | 2.7.2, 2.8.1, 2.9.1 | PYSEC-2026-3270 |
| PyPI | tensorflow-cpu | 2.7.2 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow-cpu | 2.9.1 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow-gpu | 2.7.2, 2.8.1, 2.9.1 | PYSEC-2026-947 |
| PyPI | tensorflow-gpu | 2.7.2 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow-gpu | 2.8.1 | GHSA-397c-5g2j-qxpv |
| PyPI | tensorflow-gpu | 2.9.1 | GHSA-397c-5g2j-qxpv |
Affected products (as reported)
| Vendor | Product | Versions |
|---|---|---|
| tensorflow | tensorflow | < 2.7.2>= 2.8.0, < 2.8.1>= 2.9.0, < 2.9.1 |
Weakness: CWE-20 · Vector: CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H