~ / search / GHSA-828c-5j5q-vrjq

GHSA-828c-5j5q-vrjq

TensorFlow vulnerable to null-dereference in `mlir::tfg::GraphDefImporter::ConvertNodeDef`

Source: GitHub Advisory Database (reviewed) Moderate CVE-2022-36013

Packages

EcosystemPackageAffectedFixed in
PyPI tensorflow < 2.7.2 2.7.2
PyPI tensorflow >= 2.8.0, < 2.8.1 2.8.1
PyPI tensorflow >= 2.9.0, < 2.9.1 2.9.1
PyPI tensorflow-cpu < 2.7.2 2.7.2
PyPI tensorflow-cpu >= 2.8.0, < 2.8.1 2.8.1
PyPI tensorflow-cpu >= 2.9.0, < 2.9.1 2.9.1
PyPI tensorflow-gpu < 2.7.2 2.7.2
PyPI tensorflow-gpu >= 2.8.0, < 2.8.1 2.8.1
PyPI tensorflow-gpu >= 2.9.0, < 2.9.1 2.9.1

Details

As written in the advisory (Markdown source, shown as plain text).

### Impact When [`mlir::tfg::GraphDefImporter::ConvertNodeDef`](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/ir/importexport/graphdef_import.cc) tries to convert NodeDefs without an op name, it crashes. ```cpp Status GraphDefImporter::ConvertNodeDef(OpBuilder &builder, ConversionState &s, const NodeDef &node) { VLOG(4) << "Importing: " << node.name(); OperationState state(ConvertLocation(node), absl::StrCat("tfg.", node.op())); // The GraphImporter does light shape inference, but here we will defer all of // that to the shape inference pass. const OpDef *op_def; const OpRegistrationData *op_reg_data = nullptr; if ((op_reg_data = registry_.LookUp(node.op()))) { op_def = &op_reg_data->op_def; } else { auto it = function_op_defs_.find(node.op()); if (it == function_op_defs_.end()) return InvalidArgument("Unable to find OpDef for ", node.op()); op_def = it->second; } ``` `node.op().empty()` cannot be empty. ### Patches We have patched the issue in GitHub commit [a0f0b9a21c9270930457095092f558fbad4c03e5](https://github.com/tensorflow/tensorflow/commit/a0f0b9a21c9270930457095092f558fbad4c03e5). The fix will be included in TensorFlow 2.10.0. We will also cherrypick this commit on TensorFlow 2.9.1, TensorFlow 2.8.1, and TensorFlow 2.7.2, as these are also affected and still in supported range. ### For more information Please consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.

References

External links from the advisory. They lead to third-party sites.