CVE-2021-37650: TensorFlow heap buffer overflow in DatasetToTFRecord raw ops

Summary

TensorFlow's implementation of `tf.raw_ops.ExperimentalDatasetToTFRecord` and `tf.raw_ops.DatasetToTFRecord` in affected versions can trigger a heap buffer overflow and segmentation fault. The code assumes every record in the dataset is a string, but performs no check, and the example given uses numeric types.

Mitigation

Patched in GitHub commit e0b6e58c328059829c3eb968136f17aa72b6c876. The fix will be included in TensorFlow 2.6.0, and the commit will be cherrypicked to TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4.

Affected packages in the Exposure Registry

Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.

• tensorflowPyPINo LLM dependency detected · 26 tracked dependents
• tensorflow-gpuPyPINo LLM dependency detected · 2 tracked dependents

Related items

• MediumCVE-2026-121936: easyVoice server-side request forgery through openaiBaseUrl in /api/v1/ttsSimilar attack · NVD/CVE Database
• MediumCVE-2026-118628: MLflow get-artifacts server-side request forgery information disclosureSimilar attack · NVD/CVE Database
• HighCVE-2026-118625: MLflow Tracking Server privilege escalation via get_endpointsSimilar attack · NVD/CVE Database
• InfoWhen trust becomes the payload: Inside the threats, etc quarterly reportSimilar attack · Elastic Security Labs
• MediumCritical Langflow flaws enable authentication bypass and RCESimilar attack · CSO Online