CVE-2020-15202

CVE-2020-15202 : TensorFlow Shard API truncation bug affects multiple releases (1.15.4, 2.0.3, 2.1.2, 2.2.1, 2.3.1). A lambda taking int/int32 instead of int64 in work-parallelization can cause integer truncation, leading to segfaults, out-of-bounds reads/writes, stack overflows, or data corrupti...

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

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-2020-15203

CVE-2020-15203 is a TensorFlow format-string vulnerability in tf.strings.as_string triggered by the fill argument. The issue can cause segmentation faults and is fixed in TensorFlow releases 1.15.4, 2.0.3, 2.1.2, 2.2.1, and 2.3.1 (commit 33be22c65d86256e6826666662e40dbdfe70ee83). Public reference...

CVE-2020-26266

CVE-2020-26266 (TensorFlow) arises from use of uninitialized Eigen quantized floating point types during code execution, triggered by saved-model handling. Affected TensorFlow versions include 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0; fixes are in those same branches as indicated. The issue ...

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-2020-15211

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-2020-15205

CVE-2020-15205 affects TensorFlow: the data_splits parameter of tf.raw_ops.StringNGrams lacks validation, allowing crafted input that can cause heap overflow and memory leakage, potentially leaking memory contents and aiding ASLR defeat. Affected TF versions include 1.15.4 and 2.x releases up to ...

CVE-2020-15206

CVE-2020-15206 affects TensorFlow: changing SavedModel protocol buffers and required key names can cause segfaults and data corruption while loading models, leading to a denial of service in inference deployments. The vulnerability was addressed with fixes committed in TF, and TensorFlow versions...

CVE-2020-15195

TensorFlow vulnerability CVE-2020-15195: SparseFillEmptyRowsGrad uses a double indexing pattern where reverse_index_map(i) can reference grad_values out of bounds, causing a heap-based buffer overflow. Affected releases include 1.15.4, 2.0.3, 2.1.2, 2.2.1, and 2.3.1. A fix was committed (390611e0...

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-2020-15194

CVE-2020-15194 (TensorFlow) affects TensorFlow before 1.15.4, 2.0.3, 2.1.2, 2.2.1, or 2.3.1. The SparseFillEmptyRowsGrad implementation has incomplete validation of argument shapes; while reverse_index_map_t is validated, grad_values_t is not, enabling an attacker to pass a bad grad_values_t and ...

CVE-2020-15190

TensorFlow CVE-2020-15190 is a vulnerability in tf.raw_ops.Switch where, in eager mode, the runtime binds a reference to a nullptr when one of the two outputs is undefined. This causes undefined behavior and can segfault when compiled with -fsanitize=null. The issue affects TensorFlow versions 1....

CVE-2020-15191

CVE-2020-15191 is a TensorFlow vulnerability affecting dlpack.to_dlpack. In TF versions prior to 2.2.1 and 2.3.1, passing an invalid argument to to_dlpack allows the code to bind references to null pointers due to improper status handling, leading to undefined behavior when compiled with -fsaniti...

CVE-2020-15204

CVE-2020-15204 affects TensorFlow in eager mode where a missing session_state leads to a null pointer dereference in tf.raw_ops.GetSessionHandle/GetSessionHandleV2, causing a segmentation fault (denial of service). The issue is fixed in commit 9a133d73ae4b4664d22bd1aa6d654fec13c52ee1 and releases...

CVE-2020-15192

CVE-2020-15192 affects TensorFlow and is a memory leak in the dlpack.to_dlpack path when a list of strings is passed. Root cause: the status argument used during validation failures isn’t checked, allowing a potentially failing status to be ignored and leading to memory leakage. Affected: TensorF...

CVE-2020-15193

CVE-2020-15193 in TensorFlow arises from dlpack.to_dlpack handling where a non-tensor Python object can lead to uninitialized memory and memory corruption due to an improper reinterpret_cast in pybind11 glue code. The issue is fixed in commit 22e07fb204386768e5bcbea563641ea11f96ceb8 and released ...

CVE-2020-5215

CVE-2020-5215 affects TensorFlow before 1.15.2 and 2.0.1, where converting a Python string to tf.float16 in eager mode can trigger a segmentation fault. The issue arises because format checks for this use case exist only in graph mode, potentially enabling denial of service during inference/train...

CVE-2020-26268

CVE-2020-26268 affects TensorFlow: tf.raw_ops.ImmutableConst can crash Python when mapping a file to a non-integral tensor type, due to an allocator not returning an opaque handle. The issue may trigger a segmentation fault if the memory area is large enough; a check prevents the fault if the fil...

CVE-2020-15266

In TensorFlow, CVE-2020-15266 affects the tf.image.crop_and_resize path: when boxes contains a very large value in versions before 2.4.0, the CPU kernel reads it as a C++ NaN and causes undefined behavior leading to a segmentation fault. The issue is fixed in commit eccb7ec454e6617738554a255d77f0...

CVE-2020-26267

CVE-2020-26267 affects TensorFlow where tf.raw_ops.DataFormatVecPermute does not validate src_format and dst_format, allowing uninitialized memory accesses, out-of-bounds reads, or crashes by assuming a NHWC permutation. Affected releases include various TensorFlow branches; fixes are published i...

CVE-2020-15200

CVE-2020-15200 affects TensorFlow before 2.3.1. The RaggedCountSparseOutput path does not validate that the input ragged tensor is well-formed, specifically not validating that the splits form a valid partition of values. This can set up conditions that lead to a heap-based buffer overflow and, i...

CVE-2020-26270

CVE-2020-26270 affects TensorFlow: when an LSTM/GRU layer receives a zero-length input, the CUDA backend triggers a CHECK failure leading to a denial-of-service (query-of-death). Public sources consistently describe this as a vulnerability in affected TensorFlow builds, with fixes implemented in ...

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-2020-15198

CVE-2020-15198 affects TensorFlow up to 2.3.0: SparseCountSparseOutput may access heap buffers out of bounds due to missing validation that indices and values shapes match in a sparse tensor. This root cause enables a heap buffer overflow in pre-2.3.1 builds. A fix was committed (3cbb917b47147660...

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-2020-15265

TensorFlow before 2.4.0 is affected by CVE-2020-15265: passing an invalid axis to tf.quantization.quantize_and_dequantize can trigger a segfault due to a DCHECK in dim_size, potentially crashing the C++ kernel. The issue is patched in commit eccb7ec454e6617738554a255d77f08e60ee0808 and TensorFlow...

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-2020-15197

TensorFlow prior to 2.3.1 is affected by CVE-2020-15197 due to a validation gap in SparseCountSparseOutput: the indices tensor is not checked to be rank 2, though code treats it as a matrix. This can allow crafted input sparse tensors to cause a CHECK failure and crash, enabling denial of service...

CVE-2020-15199

Summary: TensorFlow prior to 2.3.1 contains a bug in RaggedCountSparseOutput where input ragged tensors are not validated for proper splits; an empty or single-element splits can trigger a SIGABRT due to an initialization bound. Root cause: lack of validation in RaggedCountSparseOutput when formi...

CVE-2020-15201

CVE-2020-15201 : TensorFlow before 2.3.1 contains a bounds-checking flaw in RaggedCountSparseOutput, where input ragged-tensor validation is missing. Specifically, values in the splits tensor may not form a valid partitioning of values, risking a heap-based buffer overflow if split_values does no...

CVE-2018-21233

TensorFlow

CVE-2020-15196

CVE-2020-15196 affects TensorFlow 2.3.0: SparseCountSparseOutput and RaggedCountSparseOutput do not validate that the weights tensor has the same shape as the data, allowing reads beyond the weights buffer when fewer weights are provided. This heap-based overflow is mitigated in TensorFlow 2.3.1 ...

CVE-2020-26271

CVE-2020-26271 : TensorFlow contains a heap out-of-bounds access in MakeEdge when wiring graph edges, caused by missing bounds checks on indices into arrays of tensor data. This can lead to uninitialized memory access and, in certain cases, leakage of library addresses. The description notes the ...

CVE-2020-26269

CVE-2020-26269 affects TensorFlow (release candidate 2.4.0rc* and related master branch changes) and describes an access out-of-bounds read in the general filesystem-globbing path matching implementation. The root cause is that invariants and preconditions assumed by the parallel GetMatchingPaths...