Expand description

Functions that work over warps of threads.

Warps in CUDA are groups of 32 threads that are dispatched together inside of thread blocks and execute in SIMT fashion.

Functions

Returns the thread’s lane within its warp. This value ranges from 0 to WARP_SIZE - 1 (WARP_SIZE is 32 on all architectures currently).

Synchronizes all of the threads inside of this warp according to mask.