NVIDIA Public Bug Bounty

{
  "id": "13e4f5c9-5d05-4b55-b957-b036854c6ec6",
  "name": "NVIDIA Public Bug Bounty",
  "company_handle": "nvidia",
  "handle": "nvidiapublicbugbounty",
  "url": "https://www.intigriti.com/programs/nvidia/nvidiapublicbugbounty/detail",
  "status": "open",
  "confidentiality_level": "public",
  "tacRequired": false,
  "twoFactorRequired": false,
  "min_bounty": {
    "value": 150,
    "currency": "USD"
  },
  "max_bounty": {
    "value": 15000,
    "currency": "USD"
  },
  "targets": {
    "in_scope": [
      {
        "type": "other",
        "endpoint": " new CDI-based architecture",
        "description": "Refer to github documentation: https://github.com/cncf-tags/container-device-interface",
        "impact": "Tier 1"
      },
      {
        "type": "other",
        "endpoint": "All Other NVIDIA Container Toolkit Assets",
        "description": null,
        "impact": "Tier 2"
      },
      {
        "type": "other",
        "endpoint": "libNVVM API",
        "description": "**libNVVM API** provides an interface for generating PTX code from both binary and text NVVM IR inputs. Compatible input can be generated by tools and libraries that produce LLVM 7.0 IR and bitcode.\n\nMore information: https://docs.nvidia.com/cuda/libnvvm-api/index.html",
        "impact": "Tier 1"
      },
      {
        "type": "other",
        "endpoint": "Nsight Systems",
        "description": "**NSIGHT Systems CLI** provides a simple interface to collect on a target without using the GUI. The collected data can then be copied to any system and analyzed later.\n\nThe CLI is distributed in the Target directory of the standard Nsight Systems download package.\n\nIf you wish to run the CLI without root (recommended mode), you will want to install in a directory where you have full access.\n\nMore information: https://docs.nvidia.com/nsight-systems/UserGuide/index.html\n\n**In Scope:**\n* Security testing for Nsight Systems should be limited to the CUDA Toolkit on x86 desktop platforms.\n\n* Ensure that many Nsight Systems commands, which require super-user privileges, cannot be overridden through privilege escalation during security testing.",
        "impact": "Tier 1"
      },
      {
        "type": "other",
        "endpoint": "NVIDIA Nsight Developer Tools",
        "description": "**NVIDIA Nsight Developer Tools** are a powerful set of libraries, SDKs, and developer tools spanning across desktop and mobile targets that enable developers to build, debug, profile, and develop software that utilizes the latest accelerated computing hardware.\n\nMore information about all dev tools: https://developer.nvidia.com/tools-overview\n\n---\n\n**Scoping focus:**\n\n*Debugger Security:*\nLook for Hardware Misconfiguration:\nEnsure debuggers do not allow hardware to be put in a bad state.\nThis is critical in cloud environments to prevent one container from affecting other users or the entire machine.\n\n*Virtualization Security:*\nInvestigate any actions that could lead to escaping virtualization.\nEscaping virtualization can compromise the security of the entire system.\n\n*User Permissions:*\nCheck that only super users have the rights to control or affect processes belonging to other users.\nRelated article, see overview point 2: https://developer.nvidia.com/nvidia-development-tools-solutions-err_nvgpuctrperm-permission-issue-performance-counters\n\n*Context Isolation:*\nFocus on ensuring that profilers maintain strict context isolation.\nVerify that profilers do not interfere with or access other applications during context switches.\n\nTest for vulnerabilities that could allow a profiler to escape its designated context and affect other processes.\n\nBy concentrating on these areas, security researchers can help identify and mitigate potential security risks associated with profilers.\n\n**Further resources about specific dev tools**\n\n**CUDA-GDB** is the NVIDIA tool for debugging CUDA applications on Linux and QNX systems.: https://docs.nvidia.com/cuda/cuda-gdb/index.html\n\n**Nsight Compute** is an interactive kernel profiler for CUDA applications. It provides detailed performance metrics and API debugging via a user interface and command line tool.: https://docs.nvidia.com/nsight-compute/index.html\n\n**Nsight Eclipse Plugins** offers full-featured IDE that provides an all-in-one integrated environment to edit, build, debug and profile CUDA-C applications.: https://docs.nvidia.com/cuda/nsightee-plugins-install-guide/index.html\n\n**NVIDIA® Nsight™ Eclipse Edition** is a unified CPU plus GPU integrated development environment (IDE) for developing CUDA® applications on Linux and Mac OS X for the x86, and ARM platforms. It is designed to help developers on all stages of the software development process.: https://docs.nvidia.com/cuda/nsight-eclipse-plugins-guide/index.html\n\n**Compute Sanitizer** is a functional correctness checking suite included in the CUDA toolkit: https://docs.nvidia.com/compute-sanitizer/ComputeSanitizer/index.html\n\n**The Nsight Systems CLI** provides a simple interface to collect on a target without using the GUI. The collected data can then be copied to any system and analyzed later.: https://docs.nvidia.com/nsight-systems/UserGuide/index.html",
        "impact": "Tier 1"
      },
      {
        "type": "other",
        "endpoint": "NVRTC library",
        "description": "**NVRTC** is a runtime compilation library for CUDA C++. It accepts CUDA C++ source code in character string form and creates handles that can be used to obtain the PTX. The PTX string generated by NVRTC can be loaded by cuModuleLoadData and cuModuleLoadDataEx, and linked with other modules by using the nvJitLink library or using cuLinkAddData of the CUDA Driver API. This facility can often provide optimizations and performance not possible in a purely offline static compilation.\n\nMore information: https://docs.nvidia.com/cuda/nvrtc/",
        "impact": "Tier 1"
      },
      {
        "type": "other",
        "endpoint": "CUDA Libraries",
        "description": "**Scoping guidance:**\n* Only test most CUDA toolkit major/minor release, example: version 13.1. Download link: https://developer.nvidia.com/cuda-downloads\n* Test nvJPEG most recent version included in CUDA toolkit\n* Test nvCOMP most recent version, download link: https://developer.nvidia.com/nvcomp-downloads\n\n**Out of scope:**\nAll libraries and versions not explicitly mentioned above\n\nDeprecated platforms: maxwell, pascal and volta",
        "impact": "Tier 2"
      },
      {
        "type": "other",
        "endpoint": "NVCC",
        "description": "**NVCC is the CUDA Compiler Driver**\n\nMore information: https://docs.nvidia.com/cuda/cuda-compiler-driver-nvcc/index.html",
        "impact": "Tier 2"
      },
      {
        "type": "other",
        "endpoint": "nvJitLink APIs",
        "description": "**nvJitLink** The JIT Link APIs are a set of APIs which can be used at runtime to link together GPU devide code.\n\nMore information:  https://docs.nvidia.com/cuda/nvjitlink/index.html",
        "impact": "Tier 2"
      },
      {
        "type": "url",
        "endpoint": "CUDA Runtime APIs",
        "description": "Documentation: https://docs.nvidia.com/cuda/cuda-driver-api/index.html",
        "impact": "Tier 3"
      },
      {
        "type": "other",
        "endpoint": "CUDA Driver APIs",
        "description": "Documentation: https://docs.nvidia.com/cuda/cuda-runtime-api/index.html",
        "impact": "Tier 3"
      }
    ],
    "out_of_scope": []
  }
}