Division by zero in `Conv3D`
Impact
A malicious user could trigger a division by 0 in Conv3D implementation:
import tensorflow as tf
input_tensor = tf.constant([], shape=[0, 0, 0, 0, 0], dtype=tf.float32)
filter_tensor = tf.constant([], shape=[0, 0, 0, 0, 0], dtype=tf.float32)
tf.raw_ops.Conv3D(input=input_tensor, filter=filter_tensor, strides=[1, 56, 56, 56, 1], padding='VALID', data_format='NDHWC', dilations=[1, 1, 1, 23, 1])The implementation does a modulo operation based on user controlled input:
const int64 out_depth = filter.dim_size(4);
OP_REQUIRES(context, in_depth % filter_depth == 0, ...);Thus, when filter has a 0 as the fifth element, this results in a division by 0.
Additionally, if the shape of the two tensors is not valid, an Eigen assertion can be triggered, resulting in a program crash:
import tensorflow as tf
input_tensor = tf.constant([], shape=[2, 2, 2, 2, 0], dtype=tf.float32)
filter_tensor = tf.constant([], shape=[0, 0, 2, 6, 2], dtype=tf.float32)
tf.raw_ops.Conv3D(input=input_tensor, filter=filter_tensor, strides=[1, 56, 39, 34, 1], padding='VALID', data_format='NDHWC', dilations=[1, 1, 1, 1, 1])The shape of the two tensors must follow the constraints specified in the op description.
Patches
We have patched the issue in GitHub commit 799f835a3dfa00a4d852defa29b15841eea9d64f.
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.
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.
Czego dotyczy
Kliknij technologię, aby zobaczyć wszystkie dotyczące jej podatności.
| Platforma | Paczka | Podatne wersje | Poprawka w wersji |
|---|---|---|---|
| Python (PyPI) | 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 |
| Python (PyPI) | 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 |
| Python (PyPI) | 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 |
Odnośniki
- https://github.com/tensorflow/tensorflow/security/advisories/GHSA-772p-x54p-hjrv
- https://github.com/tensorflow/tensorflow/commit/799f835a3dfa00a4d852defa29b15841eea9d64f
- https://nvd.nist.gov/vuln/detail/CVE-2021-29517
- https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-cpu/PYSEC-2021-445.yaml
- https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-gpu/PYSEC-2021-643.yaml
- https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow/PYSEC-2021-154.yaml
- Wektor CVSS
CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:L- Słabość (CWE)
CWE-369- Platforma
- Python (PyPI)
- Paczka
tensorflow, tensorflow-cpu, tensorflow-gpu- Ujawniono
- 14.05.2021
- Aktualizacja
- 11.10.2026 13:19
Źródła i licencje
OSV / GitHub Advisory Database - PYSEC-2021-154 - PyPA Advisory Database; CC BY 4.0
OSV / GitHub Advisory Database - PYSEC-2021-445 - PyPA Advisory Database; CC BY 4.0
OSV / GitHub Advisory Database - PYSEC-2021-643 - PyPA Advisory Database; CC BY 4.0
OSV / GitHub Advisory Database - GHSA-772p-x54p-hjrv - CC BY 4.0
Historia podatności: tensorflow-cpu
Wszystkie opisane podatności tej paczki (Python (PyPI)), od najnowszej. Widać, czy to jednorazowa wpadka, czy paczka, której trzeba pilnować.
- Wszystkich wpisów
- 430
- Pierwsza → ostatnia
- 04.2019 07.2024
- Bez poprawki
- 0
- Wykorzystywana w atakach
- 0
Rozkład istotności
- Krytyczna 6
- Wysoka 111
- Średnia 208
- Niska 105
Ujawnienia w latach
- 2019: 7
- 2020: 35
- 2021: 200
- 2022: 166
- 2023: 21
- 2024: 1
- 2025: 0
- 2026: 0
-
Division by zero in `Conv3D` ten wpis
Dowiedz się, które podatności dotyczą Twojej aplikacji
Po podłączeniu integracji DockRay i włączeniu raportowania oprogramowania DockRay porównuje zgłoszone wersje komponentów ze znanymi podatnościami i powiadamia o potwierdzonych dopasowaniach. Zakres zależy od obsługiwanej technologii i dostępnych biuletynów.