CVE-2021-37682

Advisory

Tensorflow-rocm version 2.3.4, 2.4.3, 2.5.1 and 2.6.0 include a fix for CVE-2021-37682:
In affected versions all TFLite operations that use quantization can be made to use unitialized values. (For example, https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/lite/kernels/depthwise_conv.cc#L198-L200). The issue stems from the fact that "quantization.params" is only valid if "quantization.type" is different that "kTfLiteNoQuantization". However, these checks are missing in large parts of the code. The Tensorflow team has patched the issue in GitHub commits 537bc7c723439b9194a358f64d871dd326c18887, 4a91f2069f7145aab6ba2d8cfe41be8a110c18a5 and 8933b8a21280696ab119b63263babdb54c298538.
https://github.com/tensorflow/tensorflow/security/advisories/GHSA-4c4g-crqm-xrxw
https://github.com/tensorflow/tensorflow/commit/4a91f2069f7145aab6ba2d8cfe41be8a110c18a5
https://github.com/tensorflow/tensorflow/commit/537bc7c723439b9194a358f64d871dd326c18887
https://github.com/tensorflow/tensorflow/commit/8933b8a21280696ab119b63263babdb54c298538

Affected package

Affected versions

Fixed versions

Vulnerability changelog

This vulnerability has no description

Resources

• https://github.com/tensorflow/tensorflow/security/advisories/GHSA-4c4g-crqm-xrxw
• https://github.com/tensorflow/tensorflow/commit/4a91f2069f7145aab6ba2d8cfe41be8a110c18a5
• https://github.com/tensorflow/tensorflow/commit/537bc7c723439b9194a358f64d871dd326c18887
• https://github.com/tensorflow/tensorflow/commit/8933b8a21280696ab119b63263babdb54c298538
• https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-37682