Integer overflow in TFLite concatentation
tensorflow
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
tensorflow
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
| tensorflow-cpu PyPI |
|
|
| tensorflow-gpu PyPI |
|
|
TensorFlow is an end-to-end open source platform for machine learning. The TFLite implementation of concatenation is vulnerable to an integer overflow issue(https://github.com/tensorflow/tensorflow/blob/7b7352a724b690b11bfaae2cd54bc3907daf6285/tensorflow/lite/kernels/concatenation.cc#L70-L76). An attacker can craft a model such that the dimensions of one of the concatenation input overflow the values of `int`. TFLite uses `int` to represent tensor dimensions, whereas TF uses `int64`. Hence, valid TF models can trigger an integer overflow when converted to TFLite format. 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.
From the CVE record · GitHub_M
TFLite concatenation can overflow an integer when a crafted model has an input dimension too large for its `int` representation. An attacker can trigger the overflow if the model is converted to TFLite or processed by the affected implementation; the input does not specify network access, authentication, or user-interaction requirements.
Engineers using PyPI tensorflow, tensorflow-cpu, or tensorflow-gpu versions below 2.1.4, from 2.2.0 to below 2.2.3, from 2.3.0 to below 2.3.3, or from 2.4.0 to below 2.4.2 should check their installations.