CVE-2020-15193

See a problem?
Source
https://nvd.nist.gov/vuln/detail/CVE-2020-15193
Import Source
https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2020-15193.json
JSON Data
https://api.osv.dev/v1/vulns/CVE-2020-15193
Aliases
Related
Published
2020-09-25T19:15:14Z
Modified
2024-05-23T01:27:24.463065Z
Severity
  • 7.1 (High) CVSS_V3 - CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L CVSS Calculator
Summary
[none]
Details

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.

References

Affected packages

Git / github.com/tensorflow/tensorflow

Affected ranges

Type
GIT
Repo
https://github.com/tensorflow/tensorflow
Events
Introduced
0 Unknown introduced commit / All previous commits are affected
Fixed
22e07fb204386768e5bcbea563641ea11f96ceb8

Affected versions

0.*

0.12.0-rc0
0.12.0-rc1
0.12.1
0.5.0
0.6.0

v0.*

v0.10.0
v0.10.0rc0
v0.11.0
v0.11.0rc0
v0.11.0rc1
v0.11.0rc2
v0.12.0
v0.7.0
v0.7.1
v0.8.0rc0
v0.9.0
v0.9.0rc0

v1.*

v1.0.0
v1.0.0-alpha
v1.0.0-rc0
v1.0.0-rc1
v1.0.0-rc2
v1.1.0
v1.1.0-rc0
v1.1.0-rc1
v1.1.0-rc2
v1.12.0
v1.12.0-rc0
v1.12.0-rc1
v1.12.0-rc2
v1.12.1
v1.2.0
v1.2.0-rc0
v1.2.0-rc1
v1.2.0-rc2
v1.3.0-rc0
v1.3.0-rc1
v1.5.0
v1.5.0-rc0
v1.5.0-rc1
v1.6.0
v1.6.0-rc0
v1.6.0-rc1
v1.7.0
v1.7.0-rc0
v1.7.0-rc1
v1.8.0
v1.8.0-rc0
v1.8.0-rc1
v1.9.0
v1.9.0-rc0
v1.9.0-rc1
v1.9.0-rc2