good first issueperformance
Métricas do repositório
- Stars
- (1.408 stars)
- Métricas de merge de PR
- (Mesclagem média 5d 5h) (16 fundiu PRs em 30d)
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.