Skip to content

Latest commit

 

History

History
3 lines (3 loc) · 466 Bytes

File metadata and controls

3 lines (3 loc) · 466 Bytes

This example shows how to measure the time taken by a CUDA kernel using events. In CUDA, you can measure the time it takes to execute a CUDA kernel using the CUDA API functions for timing. The typical approach involves using the cudaEvent_t type to create events before and after the kernel execution, and then calculating the time difference between these events. The source code for this exercise is here.