Hey, I’m trying to write a program which launches many kernels, but the launches depend on results of earlier kernels. Sometimes the result means a kernel should not be launched at all. They can be distributed across many streams. Currently, I store the results of a kernel in a CuArray called c, c…

unsafe_wrap gives me the rest of the functionality I need I think. TBD if the performance is what I’m looking for using CUDA function kernel(cuarray) cuarray[1] += 3 return nothing end unified = CUDA.Mem.alloc(CUDA.Mem.Unified, 4) cuptr = convert(CuPtr{UInt32}, unified) ptr = convert(Ptr{…

Faster small CUDA memory transfers (UnifiedMem?)

Specific Domains GPU

ToucheSir August 31, 2020, 7:15pm 2

There was https://github.com/JuliaGPU/CuArrays.jl/pull/336 which would’ve made the interface much friendlier, but it doesn’t seem to have been moved to CUDA.jl…

Topic		Replies	Views
Allocating different arrays on multiple GPUs GPU	8	1083	September 30, 2021
Using Unified Memory with oversubscription GPU	3	759	March 4, 2022
Efficiency when handling jobs larger than VRAM GPU	3	1088	January 15, 2019
Multiple CPU threads that use Unified Memory gives a Bus Error GPU	5	666	April 23, 2021
CuArrays/CUDAnative/CUDAdrv equivalent to cudaMemcpy3DAsync? GPU question	10	1366	July 15, 2019

Faster small CUDA memory transfers (UnifiedMem?)

Related topics