CHECK-fail in `QuantizeAndDequantizeV4Grad`
tensorflow
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
tensorflow
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
2.3.42.4.3| tensorflow-cpu PyPI |
|
|
| tensorflow-gpu PyPI |
|
|
TensorFlow is an end-to-end open source platform for machine learning. An attacker can trigger a denial of service via a `CHECK`-fail in `tf.raw_ops.QuantizeAndDequantizeV4Grad`. This is because the implementation does not validate the rank of the `input_*` tensors. In turn, this results in the tensors being passes as they are to `QuantizeAndDequantizePerChannelGradientImpl`. However, the `vec<T>` method, requires the rank to 1 and triggers a `CHECK` failure otherwise. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2 as this is the only other affected version.
From the CVE record · GitHub_M
TensorFlow does not validate the rank of input tensors in `tf.raw_ops.QuantizeAndDequantizeV4Grad`. An attacker can supply tensors with an unexpected rank to trigger a `CHECK` failure and cause a denial of service. The input does not specify whether exploitation requires network access, authentication, or user interaction.
Engineers using the PyPI `tensorflow`, `tensorflow-cpu`, or `tensorflow-gpu` packages should check the affected version ranges listed in the advisory.