Comments | brad: "Generic AIO by scheduling stacks"

brad

"Generic AIO by scheduling stacks"

Jan 30, 2007 14:55

Zach Brown just posted to lkml (a few minutes ago) ...

[PATCH 0 of 4] Generic AIO by scheduling stacks

It's a syscall to submit syscalls to run async. Then another syscall to async gather the results of the submitted syscalls as they complete. One of the most wonderful things I've seen in awhile! Any syscall!

And I'm especially happy that ( Read more... )

tech, linux

Comments 6

hawk January 31 2007, 01:15:58 UTC

Zach Brown from Aloha? He's back in Portland, yeah? I need to have a beer with that man...

brad January 31 2007, 01:55:23 UTC

Not sure if he's from Aloha... the name doesn't ring a bell as somebody I knew.

hawk January 31 2007, 23:58:19 UTC

Yup. Same guy. He was a year ahead of me. Vanished off to Canada for a while, got married, and now I believe they live in Portland. I see him now and again, far less often than I'd like. If you haven't met him I should try to introduce you over the holidays. Or introduce yourself, for that matter. He's a great guy, very chill, and obviously a solid geek. His father is also sort of famous.

dossy January 31 2007, 01:57:02 UTC

Kernel-space threading?

"Fibrils"? Didn't the Windows folks call these "fibers"?

As Linus said--the design is obvious, but the odds of a correct implementation are low. Threads are hard to get right. Userland AIO via kernel threads is great in principle, but I'll be surprised to see it done right in practice.

Still, I'd rather a relatively clued kernel hacker try to write the code than your J. Average Codemonkey try to do it in userspace with pthreads.

eqe January 31 2007, 03:24:03 UTC

Zach makes my brane explode.

ext_9734 January 31 2007, 05:54:38 UTC

Dude. This freaking hot. Hot hot hot hot hot!

It makes me want to start on httpd 3.0... async everything. mmmmm.