At first glance, nvprof seems to be just a GUI-less version of the graphical profiling features available in the NVIDIA Visual Profiler and NSight Eclipse edition. nvprof is a command-line profiler available for Linux, Windows, and OS X. GPU can be slower because of clocking reasons.CUDA 5 added a powerful new tool to the CUDA Toolkit: nvprof. GPU is often a waste of resources for these tasks and it is better to offload them to the VIC. Is it possible to collapse operations into a single kernel?
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |