Hugging Face's logo

luckilytsh
/
executorch-backend-delegate-data-vector-oob-poc

ExecuTorch
security
poc
Model card Files
xet
 Community

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

Harmless PoC: ExecuTorch Program::get_backend_delegate_data vector length OOB

Program::get_backend_delegate_data indexes Program.backend_delegate_data[index] using a forged FlatBuffer vector length with no bound against the program buffer. Under default Program::Verification::Minimal, loading a malicious .pte and calling Method::load (FakeBackend INLINE delegate) triggers ASan heap-buffer-overflow READ at program.cpp:547.

Distinct from packaged values/chains/instructions, READY operators/args, READY ExecutionPlan.delegates/compile_specs, READY constant/mutable offsets, MethodMeta sizes, #16704/#19904 null backend_delegate_data, Vela, VGF, Vulkan.

Files

  • poc_backend_delegate_data_asan.cpp, asan.log
  • method_deps_stub.cpp (FakeBackend + Method link stubs)
  • NOTES.md
./poc_backend_delegate_data_asan
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support