CVE-2020-15200

In Tensorflow before version 2.3.1, the `RaggedCountSparseOutput` implementation does not validate that the input arguments form a valid ragged tensor. In particular, there is no validation that the values in the `splits` tensor generate a valid partitioning of the `values` tensor. Thus, the code sets up conditions to cause a heap buffer overflow. A `BatchedMap` is equivalent to a vector where each element is a hashmap. However, if the first element of `splits_values` is not 0, `batch_idx` will never be 1, hence there will be no hashmap at index 0 in `per_batch_counts`. Trying to access that in the user code results in a segmentation fault. The issue is patched in commit 3cbb917b4714766030b28eba9fb41bb97ce9ee02 and is released in TensorFlow version 2.3.1.

• CVE ID
• CVE-2020-15200

• Assigner
• security-advisories@github.com

• Vulnerability Status
• Analyzed

• Published Version
• 2020-09-25T19:15:15

• Last Modified Date
• 2021-11-18T17:24:30

• Version
• 2.0

• Vector String
• AV:N/AC:M/Au:N/C:N/I:N/A:P

• Access Vector
• NETWORK

• Access Compatibility
• MEDIUM

• Authentication
• NONE

• Confidentiality Impact
• NONE

• Integrity Impact
• NONE

• Availability Impact
• PARTIAL

• Base Score
• 4.3

• Severity
• MEDIUM

• Exploitability Score
• 8.6

• Impact Score
• 2.9

• Version
• 3.1

• Vector String
• CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H

• Attack Vector
• NETWORK

• Attack Compatibility
• HIGH

• Privileges Required
• NONE

• User Interaction
• NONE

• Scope
• UNCHANGED

• Confidentiality Impact
• NONE

• Availability Impact
• HIGH

• Base Score
• 5.9

• Base Severity
• MEDIUM

• Exploitability Score
• 2.2

• Impact Score
• 3.6