Skip to content
This repository was archived by the owner on Apr 24, 2022. It is now read-only.

CUDA defaults meaning? #318

Closed
kkkrackpot opened this issue Sep 14, 2017 · 3 comments
Closed

CUDA defaults meaning? #318

kkkrackpot opened this issue Sep 14, 2017 · 3 comments

Comments

@kkkrackpot
Copy link

Hi,

    --cuda-block-size Set the CUDA block work size. Default is 128
    --cuda-grid-size Set the CUDA grid size. Default is 8192

What exactly these defaults mean?

A block has 128 threads, and the grid has 8192 such blocks (i.e. 128*8129 threads totally)?

I'm not that good at CUDA, sorry...

Regards,
Alex

@cedricwalter
Copy link

you dont want --cuda-block-size too high, this is the number of registers available to each thread,you want it just big enough to do its work so you can put more streams in parallell

@kkkrackpot
Copy link
Author

@cedricwalter by more streams in parallel you mean --cuda-streams option?

@DeadManWalkingTO
Copy link
Contributor

I think this issue can be closed.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants