Instructions to use SantanamoDon/tensorrt-asan-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- TensorRT
How to use SantanamoDon/tensorrt-asan-poc with TensorRT:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Malicious TensorRT Engine PoC (ASAN Heap-Buffer-Overflow)
WARNING: DO NOT LOAD THIS MODEL IN A PRODUCTION ENVIRONMENT.
This repository contains a maliciously crafted .engine file designed strictly for security research and vulnerability disclosure via Huntr.
Vulnerability Details
Loading this .engine file via nvinfer1::builder::deserializeEngine triggers a heap-buffer-overflow due to an integer overflow during tensor dimension metadata allocation.
Usage (Proof of Concept)
To reproduce the crash safely, run this model inside an isolated Docker container containing an AddressSanitizer-instrumented build of the TensorRT parser.
#include "NvInfer.h"
#include <iostream>
int main() {
// DO NOT RUN OUTSIDE OF ISOLATED TESTING ENVIRONMENTS
// Loading this file will trigger memory corruption.
}
This model was generated mathematically using OSS-Fuzz-Gen to bypass static header validation.
- Downloads last month
- -