PYSEC-2020-324

In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, a crafted TFLite model can force a node to have as input a tensor backed by a nullptr buffer. This can be achieved by changing a buffer index in the flatbuffer serialization to convert a read-only tensor to a read-write one...

CVE-2020-15212 Out of bounds access in tensorflow-lite

In TensorFlow Lite before versions 2.2.1 and 2.3.1, models using segment sum can trigger writes outside of bounds of heap allocated buffers by inserting negative elements in the segment ids tensor. Users having access to segmentidsdata can alter outputindex and then write to outside of outputdata...

CVE-2020-15212

CVE-2020-15212 affects TensorFlow Lite: models using segment_sum can trigger writes outside heap buffers when negative segment_ids are present, allowing out-of-bounds writes and potential memory corruption. The issue is fixed in TensorFlow 2.2.1 and 2.3.1 (commit 204945b19e44b57906c9344c0d00120ee...

CVE-2020-15212

In TensorFlow Lite before versions 2.2.1 and 2.3.1, models using segment sum can trigger writes outside of bounds of heap allocated buffers by inserting negative elements in the segment ids tensor. Users having access to segmentidsdata can alter outputindex and then write to outside of outputdata...

CVE-2020-15213

CVE-2020-15213 affects TensorFlow Lite prior to 2.2.1 and 2.3.1. The vulnerability arises in the segment_sum implementation: models that use segment sums can trigger a denial of service by causing an out-of-memory allocation. The root cause is that code uses the last element of the segment_ids te...

CVE-2020-15213

In TensorFlow Lite before versions 2.2.1 and 2.3.1, models using segment sum can trigger a denial of service by causing an out of memory allocation in the implementation of segment sum. Since code uses the last element of the tensor holding them to determine the dimensionality of output tensor,...

CVE-2020-15213 Denial of service in tensorflow-lite

In TensorFlow Lite before versions 2.2.1 and 2.3.1, models using segment sum can trigger a denial of service by causing an out of memory allocation in the implementation of segment sum. Since code uses the last element of the tensor holding them to determine the dimensionality of output tensor,...

CVE-2020-15214

CVE-2020-15214 affects TensorFlow Lite prior to 2.2.1 and 2.3.1. A write-out-of-bounds can occur when segment IDs are not sorted in segment_sum, due to memory allocation based on the last segment-id element, causing segmentation faults and potential memory corruption. The issue is patched in comm...

CVE-2020-15214 Out of bounds write in tensorflow-lite

In TensorFlow Lite before versions 2.2.1 and 2.3.1, models using segment sum can trigger a write out bounds / segmentation fault if the segment ids are not sorted. Code assumes that the segment ids are in increasing order, using the last element of the tensor holding them to determine the...

CVE-2020-15207

In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, to mimic Python's indexing with negative values, TFLite uses ResolveAxis to convert negative values to positive indices. However, the only check that the converted index is now valid is only present in debug builds. If the...

CVE-2020-15207

CVE-2020-15207 affects TensorFlow Lite: negative indexing support uses ResolveAxis and only debug builds validate the converted index, allowing out-of-bounds access that can cause segfaults/data corruption. Affected: TensorFlow Lite before 1.15.4, 2.0.3, 2.1.2, 2.2.1, 2.3.1. Root cause: insuffici...

CVE-2020-15208

The CVE-2020-15208 issue affects TensorFlow Lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1, and 2.3.1. A debug-only DCHECK used to determine the common tensor dimension returns the first tensor’s size, which can be larger than the second tensor’s, allowing reads/writes outside bounds. This is a...

CVE-2020-15208

In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, when determining the common dimension size of two tensors, TFLite uses a DCHECK which is no-op outside of debug compilation modes. Since the function always returns the dimension of the first tensor, malicious attackers can...

CVE-2020-15208 Data corruption in tensorflow-lite

In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, when determining the common dimension size of two tensors, TFLite uses a DCHECK which is no-op outside of debug compilation modes. Since the function always returns the dimension of the first tensor, malicious attackers can...

CVE-2020-15209

Observation: CVE-2020-15209 affects TensorFlow Lite. A crafted TFLite flatbuffer can flip a tensor’s buffer index, turning a read-only tensor into read-write, which the runtime may treat as writable and initialize with a null buffer, causing a null pointer dereference. The issue has a concrete ro...

CVE-2020-15210 Segmentation fault in tensorflow-lite

In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, if a TFLite saved model uses the same tensor as both input and output of an operator, then, depending on the operator, we can observe a segmentation fault or just memory corruption. We have patched the issue in d58c96946b a...

CVE-2020-15210

CVE-2020-15210 affects TensorFlow/TFLite where a saved model reuses the same tensor as input and output for an operator, causing a segmentation fault or memory corruption depending on the operator. The issue has a patch in commit d58c96946b2880991d63d1dacacb32f0a4dfa453 and is addressed in patch ...

CVE-2020-15211 Out of bounds access in tensorflow-lite

In TensorFlow Lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, saved models in the flatbuffer format use a double indexing scheme: a model has a set of subgraphs, each subgraph has a set of operators and each operator has a set of input/output tensors. The flatbuffer format uses indice...

CVE-2020-15211

CVE-2020-15211 : In TensorFlow Lite (before 1.15.4, 2.0.3, 2.1.2, 2.2.1, 2.3.1), a negative -1 tensor index used for optional inputs can be treated as a valid index during validation, allowing out-of-bounds reads/writes in some operators. The root cause is the double indexing scheme for tensors i...

CVE-2020-15211

In TensorFlow Lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, saved models in the flatbuffer format use a double indexing scheme: a model has a set of subgraphs, each subgraph has a set of operators and each operator has a set of input/output tensors. The flatbuffer format uses indice...