CVE-2020-15193
Summary
| CVE | CVE-2020-15193 |
|---|
| State | PUBLIC |
|---|
| Assigner | [email protected] |
|---|
| Source Priority | CVE Program / NVD first with legacy fallback |
|---|
| Published | 2020-09-25 19:15:00 UTC |
|---|
| Updated | 2021-11-18 17:20:00 UTC |
|---|
| Description | In Tensorflow before versions 2.2.1 and 2.3.1, the implementation of `dlpack.to_dlpack` can be made to use uninitialized memory resulting in further memory corruption. This is because the pybind11 glue code assumes that the argument is a tensor. However, there is nothing stopping users from passing in a Python object instead of a tensor. The uninitialized memory address is due to a `reinterpret_cast` Since the `PyObject` is a Python object, not a TensorFlow Tensor, the cast to `EagerTensor` fails. The issue is patched in commit 22e07fb204386768e5bcbea563641ea11f96ceb8 and is released in TensorFlow versions 2.2.1, or 2.3.1. |
|---|
NVD Known Affected Configurations (CPE 2.3)
References
| Reference | Source | Link | Tags |
|---|
| Memory corruption in dlpack.to_dlpack · Advisory · tensorflow/tensorflow · GitHub |
CONFIRM |
github.com |
Exploit, Third Party Advisory |
| Fix multiple vulnerabilities in `tf.experimental.dlpack.to_dlpack`. · tensorflow/tensorflow@22e07fb · GitHub |
MISC |
github.com |
Patch, Third Party Advisory |
| Release TensorFlow 2.3.1 · tensorflow/tensorflow · GitHub |
MISC |
github.com |
Third Party Advisory |
| [security-announce] openSUSE-SU-2020:1766-1: moderate: Security update f |
SUSE |
lists.opensuse.org |
|
| CVE Program record |
CVE.ORG |
www.cve.org |
canonical |
| NVD vulnerability detail |
NVD |
nvd.nist.gov |
canonical, analysis |
No vendor comments have been submitted for this CVE.
Legacy QID Mappings
- 981473 Python (pip) Security Update for tensorflow-gpu (GHSA-rjjg-hgv6-h69v)
