Segfault if `tf.histogram_fixed_width` is called with NaN values in TensorFlow

Impact The implementation of tf.histogramfixedwidth is vulnerable to a crash when the values array contain NaN elements: python import tensorflow as tf import numpy as np tf.histogramfixedwidthvalues=np.nan, valuerange=1,2 The implementation assumes that all floating point operations are defined...

Heap buffer overflow due to incorrect hash function in TensorFlow

Impact The TensorKey hash function used total estimated AllocatedBytes, which a is an estimate per tensor, and b is a very poor hash function for constants e.g. int32t. It also tried to access individual tensor bytes through tensor.data of size AllocatedBytes. This led to ASAN failures because th...

Type confusion leading to `CHECK`-failure based denial of service in TensorFlow

Impact The macros that TensorFlow uses for writing assertions e.g., CHECKLT, CHECKGT, etc. have an incorrect logic when comparing sizet and int values. Due to type conversion rules, several of the macros would trigger incorrectly. Patches We have patched the issue in GitHub commit...

GHSA-F4RR-5M7V-WXCW Type confusion leading to `CHECK`-failure based denial of service in TensorFlow

Impact The macros that TensorFlow uses for writing assertions e.g., CHECKLT, CHECKGT, etc. have an incorrect logic when comparing sizet and int values. Due to type conversion rules, several of the macros would trigger incorrectly. Patches We have patched the issue in GitHub commit...

GHSA-2R2F-G8MW-9GVR Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow

Impact The implementation of tf.rawops.EditDistance has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service: python import tensorflow as tf hypothesisindices = tf.constant-1250999896764, shape=3, 3, dtype=tf.int64 hypothesisvalues =...

Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow

Impact The implementation of tf.rawops.EditDistance has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service: python import tensorflow as tf hypothesisindices = tf.constant-1250999896764, shape=3, 3, dtype=tf.int64 hypothesisvalues =...

GHSA-5WPJ-C6F7-24X8 Undefined behavior when users supply invalid resource handles

Impact Multiple TensorFlow operations misbehave in eager mode when the resource handle provided to them is invalid: python import tensorflow as tf tf.rawops.QueueIsClosedV2handle= python import tensorflow as tf tf.summary.flushwriter= In graph mode, it would have been impossible to perform these...

Undefined behavior when users supply invalid resource handles

Impact Multiple TensorFlow operations misbehave in eager mode when the resource handle provided to them is invalid: python import tensorflow as tf tf.rawops.QueueIsClosedV2handle= python import tensorflow as tf tf.summary.flushwriter= In graph mode, it would have been impossible to perform these...

GHSA-RC9W-5C64-9VQQ Missing validation results in undefined behavior in `SparseTensorDenseAdd

Impact The implementation of tf.rawops.SparseTensorDenseAdd does not fully validate the input arguments: python import tensorflow as tf aindices = tf.constant0, shape=17, 2, dtype=tf.int64 avalues = tf.constant, shape=0, dtype=tf.float32 ashape = tf.constant6, 12, shape=2, dtype=tf.int64 b =...

Missing validation results in undefined behavior in `SparseTensorDenseAdd

Impact The implementation of tf.rawops.SparseTensorDenseAdd does not fully validate the input arguments: python import tensorflow as tf aindices = tf.constant0, shape=17, 2, dtype=tf.int64 avalues = tf.constant, shape=0, dtype=tf.float32 ashape = tf.constant6, 12, shape=2, dtype=tf.int64 b =...

GHSA-54CH-GJQ5-4976 Segfault due to missing support for quantized types

Impact There is a potential for segfault / denial of service in TensorFlow by calling tf.compat.v1. ops which don't yet have support for quantized types added after migration to TF 2.x: python import numpy as np import tensorflow as tf...

Segfault due to missing support for quantized types

Impact There is a potential for segfault / denial of service in TensorFlow by calling tf.compat.v1. ops which don't yet have support for quantized types added after migration to TF 2.x: python import numpy as np import tensorflow as tf...

GHSA-HX9Q-2MX4-M4PG Missing validation causes denial of service via `Conv3DBackpropFilterV2`

Impact The implementation of tf.rawops.UnsortedSegmentJoin does not fully validate the input arguments. This results in a CHECK-failure which can be used to trigger a denial of service attack: python import tensorflow as tf tf.strings.unsortedsegmentjoin inputs='123', segmentids=0, numsegments=-1...

Missing validation causes denial of service via `Conv3DBackpropFilterV2`

Impact The implementation of tf.rawops.UnsortedSegmentJoin does not fully validate the input arguments. This results in a CHECK-failure which can be used to trigger a denial of service attack: python import tensorflow as tf tf.strings.unsortedsegmentjoin inputs='123', segmentids=0, numsegments=-1...

GHSA-JJM6-4VF7-CJH4 Integer overflow in `SpaceToBatchND`

Impact The implementation of tf.rawops.SpaceToBatchND in all backends such as XLA and handwritten kernels is vulnerable to an integer overflow: python import tensorflow as tf input = tf.constant-3.5e+35, shape=10,19,22, dtype=tf.float32 blockshape = tf.constant-1879048192, shape=2, dtype=tf.int64...

Integer overflow in `SpaceToBatchND`

Impact The implementation of tf.rawops.SpaceToBatchND in all backends such as XLA and handwritten kernels is vulnerable to an integer overflow: python import tensorflow as tf input = tf.constant-3.5e+35, shape=10,19,22, dtype=tf.float32 blockshape = tf.constant-1879048192, shape=2, dtype=tf.int64...

Denial of service in `tf.ragged.constant` due to lack of validation

Impact The implementation of tf.ragged.constant does not fully validate the input arguments. This results in a denial of service by consuming all available memory: python import tensorflow as tf tf.ragged.constantpylist=,raggedrank=8968073515812833920 Patches We have patched the issue in GitHub...

GHSA-CWPM-F78V-7M5C Denial of service in `tf.ragged.constant` due to lack of validation

Impact The implementation of tf.ragged.constant does not fully validate the input arguments. This results in a denial of service by consuming all available memory: python import tensorflow as tf tf.ragged.constantpylist=,raggedrank=8968073515812833920 Patches We have patched the issue in GitHub...

Missing validation results in undefined behavior in `QuantizedConv2D`

Impact The implementation of tf.rawops.QuantizedConv2D does not fully validate the input arguments: python import tensorflow as tf input = tf.constant1, shape=1, 2, 3, 3, dtype=tf.quint8 filter = tf.constant1, shape=1, 2, 3, 3, dtype=tf.quint8 bad args mininput = tf.constant, shape=0,...

GHSA-PQHM-4WVF-2JG8 Missing validation results in undefined behavior in `QuantizedConv2D`

Impact The implementation of tf.rawops.QuantizedConv2D does not fully validate the input arguments: python import tensorflow as tf input = tf.constant1, shape=1, 2, 3, 3, dtype=tf.quint8 filter = tf.constant1, shape=1, 2, 3, 3, dtype=tf.quint8 bad args mininput = tf.constant, shape=0,...