Undefined behavior in Tensorflow
tensorflow
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
| tensorflow-cpu PyPI |
|
|
| tensorflow-gpu PyPI |
tensorflow
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
| tensorflow-cpu PyPI |
|
|
| tensorflow-gpu PyPI |
< 2.4.0
|
In Tensorflow before version 2.4.0, when the `boxes` argument of `tf.image.crop_and_resize` has a very large value, the CPU kernel implementation receives it as a C++ `nan` floating point value. Attempting to operate on this is undefined behavior which later produces a segmentation fault. The issue is patched in eccb7ec454e6617738554a255d77f08e60ee0808 and TensorFlow 2.4.0 will be released containing the patch. TensorFlow nightly packages after this commit will also have the issue resolved.
From the CVE record · GitHub_M
In TensorFlow before 2.4.0, a very large `boxes` value passed to `tf.image.crop_and_resize` can cause undefined behavior and a segmentation fault in the CPU kernel. An attacker who can control that argument could trigger a crash; the input does not establish network access or authentication requirements.
Engineers running PyPI `tensorflow`, `tensorflow-cpu`, or `tensorflow-gpu` versions before 2.4.0 should check their installations.