Vulnerability · Published Aug 12, 2021 · Assigned by GitHub_M
CVE-2021-37665Incomplete validation in MKL requantization in TensorFlow
High 7.8
In plain English
A high-severity improper input validation in tensorflow from tensorflow. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 0.2% (higher than 8% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.
Official description
TensorFlow is an end-to-end open source platform for machine learning. In affected versions due to incomplete validation in MKL implementation of requantization, an attacker can trigger undefined behavior via binding a reference to a null pointer or can access data outside the bounds of heap allocated arrays. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/core/kernels/mkl/mkl_requantization_range_per_channel_op.cc) does not validate the dimensions of the `input` tensor. A similar issue occurs in `MklRequantizePerChannelOp`. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/core/kernels/mkl/mkl_requantize_per_channel_op.cc) does not perform full validation for all the input arguments. We have patched the issue in GitHub commit 9e62869465573cb2d9b5053f1fa02a81fce21d69 and in the Github commit 203214568f5bc237603dbab6e1fd389f1572f5c9. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.
Affected packages & fixed versions
| Ecosystem | Package | Fixed in | Advisory |
|---|---|---|---|
| PyPI | tensorflow | 2.3.4, 2.4.3 | PYSEC-2021-287 |
| PyPI | tensorflow | 2.3.4 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow | 2.4.3 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow | 2.5.1 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow-cpu | 2.4.3 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow-cpu | 2.3.4, 2.4.3 | PYSEC-2021-578 |
| PyPI | tensorflow-cpu | 2.3.4 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow-cpu | 2.5.1 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow-gpu | 2.3.4, 2.4.3 | PYSEC-2021-776 |
| PyPI | tensorflow-gpu | 2.3.4 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow-gpu | 2.4.3 | GHSA-v82p-hv3v-p6qp |
| PyPI | tensorflow-gpu | 2.5.1 | GHSA-v82p-hv3v-p6qp |
Affected products (as reported)
| Vendor | Product | Versions |
|---|---|---|
| tensorflow | tensorflow | >= 2.5.0, < 2.5.1>= 2.4.0, < 2.4.3< 2.3.4 |
Weakness: CWE-20 · Vector: CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
References
- https://github.com/tensorflow/tensorflow/security/advisories/GHSA-v82p-hv3v-p6qp x_refsource_CONFIRM
- https://github.com/tensorflow/tensorflow/commit/203214568f5bc237603dbab6e1fd389f1572f5c9 x_refsource_MISC
- https://github.com/tensorflow/tensorflow/commit/9e62869465573cb2d9b5053f1fa02a81fce21d69 x_refsource_MISC