cbcvebase.
CVE-2020-26270
published 2020-12-10

CVE-2020-26270: In affected versions of TensorFlow running an LSTM/GRU model where the LSTM/GRU layer receives an input with zero-length results in a CHECK failure when using…

low3.3CVSS 3.1
AVLACLPRLUINSUCNINAL
In affected versions of TensorFlow running an LSTM/GRU model where the LSTM/GRU layer receives an input with zero-length results in a CHECK failure when using the CUDA backend. This can result in a query-of-death vulnerability, via denial of service, if users can control the input to the layer. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.

Affected

17 ranges
VendorProductVersion rangeFixed in
debiantensorflow——
googletensorflow< 1.15.51.15.5
googletensorflow>= 2.0.0 < 2.0.42.0.4
googletensorflow>= 2.1.0 < 2.1.32.1.3
googletensorflow>= 2.2.0 < 2.2.22.2.2
googletensorflow>= 2.3.0 < 2.3.22.3.2
inteloptimization_for_tensorflow>= 0 < 14755416e364f17fb1870882fa778c7fec7f16e314755416e364f17fb1870882fa778c7fec7f16e3
inteloptimization_for_tensorflow>= 0 < 1.15.51.15.5
inteloptimization_for_tensorflow>= 2.0.0 < 2.0.42.0.4
inteloptimization_for_tensorflow>= 2.1.0 < 2.1.32.1.3
inteloptimization_for_tensorflow>= 2.2.0 < 2.2.22.2.2
inteloptimization_for_tensorflow>= 2.3.0 < 2.3.22.3.2
tensorflowtensorflow< 1.15.51.15.5
tensorflowtensorflow——
tensorflowtensorflow——
tensorflowtensorflow——
tensorflowtensorflow——