JuliaGPU/CUDA.jl

Consider maximizing grid utilization

Open

#1 321 ouverte le 13 janv. 2022

Voir sur GitHub
 (1 commentaire) (0 réactions) (0 assignés)Julia (274 forks)batch import
good first issueperformance

Métriques du dépôt

Stars
 (1 408 stars)
Métriques de merge PR
 (Merge moyen 5j 5h) (16 PRs mergées en 30 j)

Description

We currently maximize block utilization (taking the max threads), which may leave SMs underutilized. We should consider first selecting an optimal amount of blocks, before maximizing the thread could:

    config = launch_configuration(kernel.fun)
    threads = min(length(ps), config.threads)
    # XXX: this kernel performs much better with all blocks active
    blocks = max(cld(length(ps), threads), config.blocks)
    threads = cld(length(ps), blocks)

I'm sure this will lead to some kernels performing worse, though, but it's probably a good thing to test.

Guide contributeur