There are two APIs available in CUDA programming: Driver API and runtime API. I cannot distinguish them. Why should we make them co-exist; why not integrate them into one? Where are they located in execution: in device environment or in host environment? Thanks.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Runtime API? Driver API? | 2 | 6134 | May 7, 2008 | |
| Difference between runtime and driver CUDA | 0 | 3186 | June 5, 2010 | |
| Runtime API vs Driver API which one is faster | 6 | 1981 | March 27, 2009 | |
| cuda driver api vs. runtime api | 1 | 1005 | January 14, 2020 | |
| 'CUDA runtime API' term explain meaning of the term in order to translate into other languag | 4 | 1606 | December 25, 2009 | |
| Mixing runtime and device API Any downside? | 0 | 3404 | August 6, 2008 | |
| What is the difference between runtime and driver API? | 8 | 16062 | August 28, 2016 | |
| Runtime & Driver API differences 11.5 -> 11.6? | 1 | 447 | January 27, 2022 | |
| runtime API and driver API mutually exclusive? according to 0.8.1 SDK not really | 5 | 3557 | February 13, 2008 | |
| CUDA Driver API and CUDA runtime mutually exclusive | 1 | 885 | July 30, 2009 |