AVID-2026-R0894

Description

Unitialized access in EinsumHelper::ParseEquation (CVE-2021-41201)

Details

TensorFlow is an open source platform for machine learning. In affeced versions during execution, EinsumHelper::ParseEquation() is supposed to set the flags in input_has_ellipsis vector and *output_has_ellipsis boolean to indicate whether there is ellipsis in the corresponding inputs and output. However, the code only changes these flags to true and never assigns false. This results in unitialized variable access if callers assume that EinsumHelper::ParseEquation() always sets these flags. 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.

Reason for inclusion in AVID: CVE-2021-41201 describes an uninitialized access flaw in TensorFlow’s EinsumHelper::ParseEquation, a software vulnerability in a widely-used AI framework. It affects TensorFlow versions in the AI software stack, impacting models and ML pipelines. It is a software supply-chain-relevant vulnerability (dependency/framework level) with CVSS-severity high, and there is clear evidence in the advisory and references.

References

• NVD entry
• https://github.com/tensorflow/tensorflow/security/advisories/GHSA-j86v-p27c-73fm
• https://github.com/tensorflow/tensorflow/commit/f09caa532b6e1ac8d2aa61b7832c78c5b79300c6

Affected or Relevant Artifacts

• Developer: tensorflow
• Deployer: tensorflow
• Artifact Details:

Impact

AVID Taxonomy Categorization

• Risk domains: Security
• SEP subcategories: S0100: Software Vulnerability
• Lifecycle stages: L06: Deployment

CVSS

CWE

Other information

• Report Type: Advisory
• Credits:
• Date Reported: 2021-11-05
• Version: 0.3.3
• AVID Entry