1## TFSA-2022-112: `CHECK` fail in `QuantizeAndDequantizeV3` 2 3### CVE Number 4CVE-2022-36026 5 6### Impact 7If `QuantizeAndDequantizeV3` is given a nonscalar `num_bits` input tensor, it results in a `CHECK` fail that can be used to trigger a denial of service attack. 8```python 9import tensorflow as tf 10 11signed_input = True 12range_given = False 13narrow_range = False 14axis = -1 15input = tf.constant(-3.5, shape=[1], dtype=tf.float32) 16input_min = tf.constant(-3.5, shape=[1], dtype=tf.float32) 17input_max = tf.constant(-3.5, shape=[1], dtype=tf.float32) 18num_bits = tf.constant([], shape=[0], dtype=tf.int32) 19tf.raw_ops.QuantizeAndDequantizeV3(input=input, input_min=input_min, input_max=input_max, num_bits=num_bits, signed_input=signed_input, range_given=range_given, narrow_range=narrow_range, axis=axis) 20``` 21 22### Patches 23We have patched the issue in GitHub commit [f3f9cb38ecfe5a8a703f2c4a8fead434ef291713](https://github.com/tensorflow/tensorflow/commit/f3f9cb38ecfe5a8a703f2c4a8fead434ef291713). 24 25The 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. 26 27 28### For more information 29Please 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. 30 31 32### Attribution 33This vulnerability has been reported by Neophytos Christou, Secure Systems Labs, Brown University. 34