Hi, I’m trying to pass an object with a wrapped CuArray to a kernel: import CUDA struct MyStruct arrayField::CUDA.CuArray{Float32, 1, Nothing} end function f!(obj::MyStruct) id = CUDA.threadIdx().x if id <= length(a) obj.arrayField[id] = id end nothing end myObj = MyS…

CuArray is the host-representation of a CUDA array, you need CuDeviceArray. That conversion happens automatically for known types, but you need to register your own Adapt.jl conversions for custom types.

Passing a wrapped array to a kernel

Specific Domains GPU

emiwar May 27, 2020, 1:41pm 3

I see, thanks!

Topic		Replies	Views
Kernel Compilation error- KernelError: recursion is currently not supported GPU	8	1346	November 8, 2019
Arrays of arrays and arrays of structures in CUDA kernels cause random errors GPU gpu , cuda	21	2907	October 21, 2021
Store CuArrays on a mutable struct? GPU	5	1465	July 2, 2018
Composite types array, and composite type with array fields GPU cudanative , cuda	5	1165	December 20, 2019
Passing array of pointers to CUDA kernel GPU question , cudanative , memory-allocation	2	3633	January 20, 2020

Passing a wrapped array to a kernel

Related topics