NVIDIA's CUDA C++ GUIDELINES
Comply with the guidelines for effective safety-critical programming and ensure your CUDA C++ code meets the requirements for highly reliable software.
Prefer a live demo or want to speak to one of our Solution Engineers? Get in touch.
What are the CUDA C++ Guidelines?
NVIDIA's CUDA C++ Guidelines for robust and safety-critical programming outline recommended coding practices and guidelines that improve the robustness and safety of software developed in CUDA C++. They are intended for developers working on safety-critical and high-reliability applications.
The guidelines were developed to enforce practices that will increase the robustness of software that uses CUDA. They fill a gap by covering the aspects of the CUDA C++ programming language that differ from the ISO C++ programming language which CUDA C++ is based on.
Axivion for CUDA helps you follow these guidelines with ease. The coding guideline checker also supports standards like MISRA, CERT, and CWE.
Learning by Doing
Understand the CUDA C++ Guidelines while developing. Detailed explanations of the rules will be shown to you if you violate them. Next time you will know better.
Part of Your Daily Workflow
No need to change tools or IDE. Axivion for CUDA seamlessly integrates into your CI/DevOps environment.
Save Time by Automating Checks
Don't waste time with manual checks. Rely on comprehensive and reliable automated checks to comply with the CUDA C++ Guidelines.