Out of bounds read in Tensorflow
tensorflow
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
| tensorflow-cpu PyPI |
|
|
| tensorflow-gpu PyPI |
tensorflow
| Product | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI |
|
|
| tensorflow-cpu PyPI |
|
|
| tensorflow-gpu PyPI |
≥ 2.7.0 · < 2.7.1
|
Tensorflow is an Open Source Machine Learning Framework. 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.
From the CVE record · GitHub_M
TensorFlow’s TFG dialect makes assumptions about a GraphDef when converting it to MLIR-based IR. An attacker who can alter a SavedModel on disk can cause a Python interpreter crash when that conversion occurs; heap out-of-bounds reads and writes are also possible in some scenarios.
Teams using PyPI tensorflow, tensorflow-cpu, or tensorflow-gpu versions >=2.7.0 and <2.7.1 should check their installations.