Out of bounds read in Tensorflow

Impact

The TFG dialect of TensorFlow (MLIR) makes several assumptions about the incoming GraphDef before converting it to the MLIR-based dialect.

If an attacker changes the SavedModel format on disk to invalidate these assumptions and the GraphDef is then converted to MLIR-based IR then they can cause a crash in the Python interpreter. Under certain scenarios, heap OOB read/writes are possible.

These issues have been discovered via fuzzing and it is possible that more weaknesses exist. We will patch them as they are discovered.

Patches

We have patched the issue in multiple GitHub commits and these will be included in TensorFlow 2.8.0 and TensorFlow 2.7.1, as both are affected.

For more information

Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.