Hi everyone, I am trying to use shuffle in a Cuda kernel, can someone explain to me how to use CUDA.shfl_sync, CUDA.shfl_up_sync, CUDA.shfl_down_sync, CUDA.shfl_xor_sync with a simple example…? It’s not clear in the documentation.

[image] Using CUDA Warp-Level Primitives | NVIDIA Technical Blog NVIDIA GPUs execute groups of threads known as warps in SIMT (Single Instruction, Multiple Thread) fashion. Many CUDA programs achieve high performance by taking advantage of warp execution.

Understanding Shuffle in CUDA

General Usage Performance

Gnimuc November 14, 2022, 1:43pm 2

1 Like

Topic		Replies	Views
How to understand MapReduce GPU	1	720	August 31, 2021
Sequence of warp and how to avoid divergence when folding shared memory in a reduction kernel GPU cuda	3	1226	July 19, 2018
CUDAnative , performance drop after several timesteps Performance question , gpu , cudanative , cuda	5	1193	September 8, 2019
Examples with atomic operations using CUDA.jl GPU	6	2981	September 15, 2020
Most efficient way of _waiting_ for GPU results? GPU	20	3123	January 31, 2019

Understanding Shuffle in CUDA

Related topics