I am trying to get atomic to work. I am trying to produce an MWE so the code by itself doesn’t make any sense. I have a buffer which I am trying to add 1.0 for every thread in block that gets run. using CUDA CUDA.allowscalar(false) function atleast2_gpu_v1!(buffer) i = threadIdx().x j = …

[image] xiaodai: j = 0.0 j = Float32(0.0) this fixes it.

How to use `@atomic` with CUDA?

xiaodai October 5, 2020, 3:21am 2

j = Float32(0.0) this fixes it.

Topic		Replies	Views
CUDA.@atomic causes type instability? GPU question	5	646	November 29, 2021
Atomic operations stop working when upgrading to CUDA.jl version 3.3 GPU question	4	1090	June 17, 2021
Kernel Error: kernel returns a value of type `Union{}` GPU	3	1050	October 14, 2019
Create a simple CUDA.sum kernel GPU	3	2026	January 3, 2021
GPU Compilation Error: KernelError: kernel returns a value of type `Union{}` General Usage gpu , cuda	3	840	December 4, 2021