PT-2021-18295 · Google · Tensorflow

Yakun Zhang + 1 · Published 2021-05-14 · Updated 2024-10-31 · CVE-2021-29544

Positive Technologies (AI Summary)

CVSS v3.1

5.5

Medium

Base vector

AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Name of the Vulnerable Software and Affected Versions TensorFlow versions 2.4.2 through 2.4.x and versions prior to 2.5.0

Description An attacker can trigger a denial of service via a CHECK-fail in tf.raw ops.QuantizeAndDequantizeV4Grad. This is because the implementation does not validate the rank of the input * tensors, resulting in the tensors being passed as they are to QuantizeAndDequantizePerChannelGradientImpl. However, the vec<T> method requires the rank to be 1 and triggers a CHECK failure otherwise.

Recommendations For versions 2.4.2, update to version 2.5.0 or later to resolve the issue. For versions prior to 2.5.0, update to version 2.5.0 or later to resolve the issue. As a temporary workaround, consider restricting the use of tf.raw ops.QuantizeAndDequantizeV4Grad until a patch is available.

Exploit

Fix

Improper Check for Exceptional Conditions

Weakness Enumeration

CWE-754

Related Identifiers

BIT-TENSORFLOW-2021-29544

CVE-2021-29544

GHSA-6G85-3HM8-83F9

PYSEC-2021-181

PYSEC-2021-472

PYSEC-2021-670

Affected Products

Tensorflow

References · 15

• 🔥 https://github.com/tensorflow/tensorflow/security/advisories/GHSA-6g85-3hm8-83f9 · Exploit
• https://github.com/tensorflow/tensorflow/commit/20431e9044cf2ad3c0323c34888b192f3289af6b · Patch
• https://osv.dev/vulnerability/PYSEC-2021-670 · Vendor Advisory
• https://osv.dev/vulnerability/PYSEC-2021-472 · Vendor Advisory
• https://nvd.nist.gov/vuln/detail/CVE-2021-29544 · Security Note
• https://osv.dev/vulnerability/BIT-tensorflow-2021-29544 · Vendor Advisory
• https://osv.dev/vulnerability/PYSEC-2021-181 · Vendor Advisory
• https://osv.dev/vulnerability/CVE-2021-29544 · Vendor Advisory
• https://osv.dev/vulnerability/GHSA-6g85-3hm8-83f9 · Vendor Advisory
• https://github.com/tensorflow/tensorflow/blob/95078c145b5a7a43ee046144005f733092756ab5/tensorflow/core/kernels/quantize_and_dequantize_op.cc#L162-L163 · Note
• https://github.com/tensorflow/tensorflow/blob/95078c145b5a7a43ee046144005f733092756ab5/tensorflow/core/kernels/quantize_and_dequantize_op.h#L295-L306 · Note
• https://github.com/tensorflow/tensorflow · Note
• https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-cpu/PYSEC-2021-472.yaml · Note
• https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-gpu/PYSEC-2021-670.yaml · Note
• https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow/PYSEC-2021-181.yaml · Note