CVE-2022-41887
tf.keras.losses.poisson receives a y_pred and y_true that are passed through functor::mul in BinaryOp. If the resulting dimensions overflow an int32, TensorFlow will crash due to a size mismatch during broadcast assignment.
import numpy as np
import tensorflow as tf
true_value = tf.reshape(shape=[1, 2500000000], tensor = tf.zeros(dtype=tf.bool, shape=[50000, 50000]))
pred_value = np.array([[[-2]], [[8]]], dtype = np.float64)
tf.keras.losses.poisson(y_true=true_value,y_pred=pred_value)We have patched the issue in GitHub commit c5b30379ba87cbe774b08ac50c1f6d36df4ebb7c.
The fix will be included in TensorFlow 2.11. We will also cherrypick this commit on TensorFlow 2.10.1 and 2.9.3, as these are also affected and still in supported range. However, we will not cherrypick this commit into TensorFlow 2.8.x, as it depends on Eigen behavior that changed between 2.8 and 2.9.
Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.
This vulnerability has been reported by Pattarakrit Rattankul.