cbcvebase.
CVE-2021-29547
published 2021-05-14

CVE-2021-29547: TensorFlow is an end-to-end open source platform for machine learning. An attacker can cause a segfault and denial of service via accessing data outside of…

medium5.5CVSS 3.1
AVLACLPRLUINSUCNINAH
TensorFlow is an end-to-end open source platform for machine learning. An attacker can cause a segfault and denial of service via accessing data outside of bounds in `tf.raw_ops.QuantizedBatchNormWithGlobalNormalization`. This is because the implementation(https://github.com/tensorflow/tensorflow/blob/55a97caa9e99c7f37a0bbbeb414dc55553d3ae7f/tensorflow/core/kernels/quantized_batch_norm_op.cc#L176-L189) assumes the inputs are not empty. If any of these inputs is empty, `.flat()` is an empty buffer, so accessing the element at index 0 is accessing data outside of bounds. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, as these are also affected and still in supported range.

Affected

18 ranges
VendorProductVersion rangeFixed in
debiantensorflow
googletensorflow< 2.1.42.1.4
googletensorflow>= 2.2.0 < 2.2.32.2.3
googletensorflow>= 2.3.0 < 2.3.32.3.3
googletensorflow>= 2.4.0 < 2.4.22.4.2
inteloptimization_for_tensorflow>= 0 < d6ed5bcfe1dcab9e85a4d39931bd18d99018e75bd6ed5bcfe1dcab9e85a4d39931bd18d99018e75b
inteloptimization_for_tensorflow>= 0 < 2.2.0rc02.2.0rc0
inteloptimization_for_tensorflow>= 0 < 2.1.42.1.4
inteloptimization_for_tensorflow>= 2.2.0 < 2.3.0rc02.3.0rc0
inteloptimization_for_tensorflow>= 2.2.0 < 2.2.32.2.3
inteloptimization_for_tensorflow>= 2.3.0 < 2.3.42.3.4
inteloptimization_for_tensorflow>= 2.3.0 < 2.3.32.3.3
inteloptimization_for_tensorflow>= 2.4.0 < 2.4.32.4.3
inteloptimization_for_tensorflow>= 2.4.0 < 2.4.22.4.2
tensorflowtensorflow< 2.1.42.1.4
tensorflowtensorflow
tensorflowtensorflow
tensorflowtensorflow