CVE-2021-41197 Crashes due to overflow and `CHECK`-fail in ops with large tensor shapes
An attacker can cause a denial of service in TensorFlow by crafting tensors with large shapes that trigger an integer overflow and subsequent crash.
- Exploited in the wild
- Not known
- Login needed
- Any signed-in user
- Attackable
- Local or same network only
- Victim action needed
- Not needed
- Fix
- Available
What is CVE-2021-41197?
CVE-2021-41197 is a medium-severity vulnerability in Google Tensorflow, published on 5 Nov 2021. It can only be exploited from the same machine or local network, with no action from the victim.
Official description
TensorFlow is an open source platform for machine learning. In affected versions TensorFlow allows tensor to have a large number of dimensions and each dimension can be as large as desired. However, the total number of elements in a tensor must fit within an `int64_t`. If an overflow occurs, `MultiplyWithoutOverflow` would return a negative result. In the majority of TensorFlow codebase this then results in a `CHECK`-failure. Newer constructs exist which return a `Status` instead of crashing the binary. This is similar to CVE-2021-29584. 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.
Weakness: CWE-190
Which versions are affected by CVE-2021-41197?
22 open source packages ship an affected version, including tensorflow, tensorflow-cpu and tensorflow-gpu. It affects 1 product.
| Package | Affected | Fixed in |
|---|---|---|
tensorflow Bitnami | All versions | 2.4.4, 2.5.2, 2.6.1 |
tensorflow PyPI | All versions | 2.4.4, 2.5.2, 2.6.1, 2.7.0 |
tensorflow PyPI | All versions | 2.4.4 |
tensorflow PyPI | All versions | 2.4.4 |
tensorflow PyPI | From 2.5.0 | 2.5.2 |
tensorflow PyPI | From 2.5.0 | 2.5.2 |
16 more packages
tensorflow PyPI | From 2.6.0 | 2.6.1 |
tensorflow PyPI | From 2.6.0 | 2.6.1 |
tensorflow-cpu PyPI | All versions | 2.4.4, 2.5.2, 2.6.1, 2.7.0 |
tensorflow-cpu PyPI | All versions | 2.4.4 |
tensorflow-cpu PyPI | All versions | 2.4.4 |
tensorflow-cpu PyPI | From 2.5.0 | 2.5.2 |
tensorflow-cpu PyPI | From 2.5.0 | 2.5.2 |
tensorflow-cpu PyPI | From 2.6.0 | 2.6.1 |
tensorflow-cpu PyPI | From 2.6.0 | 2.6.1 |
tensorflow-gpu PyPI | All versions | 2.4.4, 2.5.2, 2.6.1, 2.7.0 |
tensorflow-gpu PyPI | All versions | 2.4.4 |
tensorflow-gpu PyPI | All versions | 2.4.4 |
tensorflow-gpu PyPI | From 2.5.0 | 2.5.2 |
tensorflow-gpu PyPI | From 2.5.0 | 2.5.2 |
tensorflow-gpu PyPI | From 2.6.0 | 2.6.1 |
tensorflow-gpu PyPI | From 2.6.0 | 2.6.1 |
Products include Google Tensorflow.
How do I fix CVE-2021-41197?
Upgrade tensorflow to a fixed version for your release line: 2.4.4, 2.5.2, 2.6.1 or 2.7.0.
Is CVE-2021-41197 being exploited?
There is no known exploitation of CVE-2021-41197 in the wild and no public exploit code.
Other vulnerabilities in Google Tensorflow
- CVE-2023-25668 TensorFlow vulnerable to heap out-of-buffer read in the QuantizeAndDequantize operation
- CVE-2018-7575 Integer Overflow or Wraparound in Google TensorFlow
- CVE-2021-37678 Arbitrary code execution due to YAML deserialization
- CVE-2021-35958 Path traversal in n/a
- CVE-2020-15202 Integer truncation in Shard API usage
Sources
Also known as BIT-tensorflow-2021-41197, GHSA-prcg-wp5q-rv7p, PYSEC-2021-390, PYSEC-2021-607, PYSEC-2021-805.