2.1 Low
CVSS2
Attack Vector
LOCAL
Attack Complexity
LOW
Authentication
NONE
Confidentiality Impact
NONE
Integrity Impact
NONE
Availability Impact
PARTIAL
AV:L/AC:L/Au:N/C:N/I:N/A:P
8.8 High
CVSS3
Attack Vector
LOCAL
Attack Complexity
LOW
Privileges Required
LOW
User Interaction
NONE
Scope
CHANGED
Confidentiality Impact
HIGH
Integrity Impact
HIGH
Availability Impact
HIGH
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H
0.0004 Low
EPSS
Percentile
12.8%
The TFG dialect of TensorFlow (MLIR) makes several assumptions about the incoming GraphDef
before converting it to the MLIR-based dialect.
If an attacker changes the SavedModel
format on disk to invalidate these assumptions and the GraphDef
is then converted to MLIR-based IR then they can cause a crash in the Python interpreter. Under certain scenarios, heap OOB read/writes are possible.
These issues have been discovered via fuzzing and it is possible that more weaknesses exist. We will patch them as they are discovered.
We have patched the issue in multiple GitHub commits and these will be included in TensorFlow 2.8.0 and TensorFlow 2.7.1, as both are affected.
Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.
CPE | Name | Operator | Version |
---|---|---|---|
tensorflow-gpu | eq | 2.7.0 | |
tensorflow-cpu | eq | 2.7.0 | |
tensorflow | eq | 2.7.0 |
2.1 Low
CVSS2
Attack Vector
LOCAL
Attack Complexity
LOW
Authentication
NONE
Confidentiality Impact
NONE
Integrity Impact
NONE
Availability Impact
PARTIAL
AV:L/AC:L/Au:N/C:N/I:N/A:P
8.8 High
CVSS3
Attack Vector
LOCAL
Attack Complexity
LOW
Privileges Required
LOW
User Interaction
NONE
Scope
CHANGED
Confidentiality Impact
HIGH
Integrity Impact
HIGH
Availability Impact
HIGH
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H
0.0004 Low
EPSS
Percentile
12.8%