CVE-2021-29550

Division by 0 in `FractionalAvgPool`

Severity Score

5.5

*CVSS v3.1

Exploit Likelihood

*EPSS

Affected Versions

*CPE

Public Exploits

*Multiple Sources

Exploited in Wild

*KEV

Decision

*SSVC

Descriptions

TensorFlow is an end-to-end open source platform for machine learning. An attacker can cause a runtime division by zero error and denial of service in `tf.raw_ops.FractionalAvgPool`. This is because the implementation(https://github.com/tensorflow/tensorflow/blob/acc8ee69f5f46f92a3f1f11230f49c6ac266f10c/tensorflow/core/kernels/fractional_avg_pool_op.cc#L85-L89) computes a divisor quantity by dividing two user controlled values. The user controls the values of `input_size[i]` and `pooling_ratio_[i]` (via the `value.shape()` and `pooling_ratio` arguments). If the value in `input_size[i]` is smaller than the `pooling_ratio_[i]`, then the floor operation results in `output_size[i]` being 0. The `DCHECK_GT` line is a no-op outside of debug mode, so in released versions of TF this does not trigger. Later, these computed values are used as arguments(https://github.com/tensorflow/tensorflow/blob/acc8ee69f5f46f92a3f1f11230f49c6ac266f10c/tensorflow/core/kernels/fractional_avg_pool_op.cc#L96-L99) to `GeneratePoolingSequence`(https://github.com/tensorflow/tensorflow/blob/acc8ee69f5f46f92a3f1f11230f49c6ac266f10c/tensorflow/core/kernels/fractional_pool_common.cc#L100-L108). There, the first computation is a division in a modulo operation. Since `output_length` can be 0, this results in runtime crashing. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit 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.

TensorFlow es una plataforma de código abierto de extremo a extremo para el aprendizaje automático. Un atacante puede causar un error de división por cero en tiempo de ejecución y una denegación de servicio en "tf.raw_ops.FractionalAvgPool". Esto es debido a que la implementación (https://github.com/tensorflow/tensorflow/blob/acc8ee69f5f46f92a3f1f11230f49c6ac266f10c/tensorflow/core/kernels/fractional_avg_pool_op.cc#L85-L89) calcula una cantidad divisora al dividir dos valores controlados por el usuario. El usuario controla los valores de "input_size[i]" y "pooling_ratio_[i]" (por medio de los argumentos "value.shape()" y "pooling_ratio"). Si el valor en "input_size[i]" es menor que el "pooling_ratio_[i]", entonces la operación floor resulta en que "output_size[i]" sea 0. La línea "DCHECK_GT" es un no-op fuera del modo de depuración, así que en las versiones liberadas de TF no es desencadenado. Más tarde, estos valores calculados son usados como argumentos (https://github.com/tensorflow/tensorflow/blob/acc8ee69f5f46f92a3f1f11230f49c6ac266f10c/tensorflow/core/kernels/fractional_avg_pool_op.cc#L96-L99) para "GeneratePoolingSequence"(https://github.com/tensorflow/tensorflow/blob/acc8ee69f5f46f92a3f1f11230f49c6ac266f10c/tensorflow/core/kernels/fractional_pool_common.cc#L100-L108). Allí, el primer cálculo es una división en una operación de módulo. Dado que "output_length" puede ser 0, esto resulta en un bloqueo en tiempo de ejecución. La corrección será incluida en TensorFlow versión 2.5.0. También se incluirá este commit en TensorFlow versión 2.4.2, TensorFlow versión 2.3.3, TensorFlow versión 2.2.3 y TensorFlow versión 2.1.4, ya que estos también están afectados y siguen siendo compatibles

*Credits: N/A

CVSS Scores

Attack Vector

Local

Attack Complexity

Low

Privileges Required

Low

User Interaction

None

Scope

Unchanged

Confidentiality

None

Integrity

None

Availability

High

Attack Vector

Local

Attack Complexity

High

Privileges Required

Low

User Interaction

None

Scope

Unchanged

Confidentiality

None

Integrity

None

Availability

Low

Attack Vector

Local

Attack Complexity

Low

Authentication

None

Confidentiality

None

Integrity

None

Availability

Partial

* Common Vulnerability Scoring System

SSVC

Decision:-

Exploitation

Automatable

Tech. Impact

* Organization's Worst-case Scenario

Timeline

2021-03-30 CVE Reserved
2021-05-14 CVE Published
2023-03-08 EPSS Updated
2024-08-03 CVE Updated
2024-08-03 First Exploit
---------- Exploited in Wild
---------- KEV Due Date

CWE

CWE-369: Divide By Zero

CAPEC

References (2)

URL	Tag	Source

URL	Date	SRC
https://github.com/tensorflow/tensorflow/security/advisories/GHSA-f78g-q7r4-9wcv	2024-08-03

URL	Date	SRC
https://github.com/tensorflow/tensorflow/commit/548b5eaf23685d86f722233d8fbc21d0a4aecb96	2021-07-27

URL	Date	SRC

Affected Vendors, Products, and Versions

Vendor		Product				Version		Other		Status
Vendor	Product	Version	Other	Status	<-- -->	Vendor	Product	Version	Other	Status
Google Search vendor "Google"		Tensorflow Search vendor "Google" for product "Tensorflow"				< 2.1.4 Search vendor "Google" for product "Tensorflow" and version " < 2.1.4"		-		Affected
Google Search vendor "Google"		Tensorflow Search vendor "Google" for product "Tensorflow"				>= 2.2.0 < 2.2.3 Search vendor "Google" for product "Tensorflow" and version " >= 2.2.0 < 2.2.3"		-		Affected
Google Search vendor "Google"		Tensorflow Search vendor "Google" for product "Tensorflow"				>= 2.3.0 < 2.3.3 Search vendor "Google" for product "Tensorflow" and version " >= 2.3.0 < 2.3.3"		-		Affected
Google Search vendor "Google"		Tensorflow Search vendor "Google" for product "Tensorflow"				>= 2.4.0 < 2.4.2 Search vendor "Google" for product "Tensorflow" and version " >= 2.4.0 < 2.4.2"		-		Affected