lkml.org 
[lkml]   [1999]   [Oct]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: Task queue makes slow device even slower
Date
Hello!

> Hmm, I'm not sure if net_bh is the answer: the soft timers paper describes
> soft timer checks basically whenever we run BHs, plus soft timer checks on
> system call return (funny, I thought we did this ?),

Linux really does it (arch/i386/kernel/entry.S).

> and on CPU idle.

Linux does it (sys_idle()).

BHs are really perfect soft timer, you need only raise some BH bit
and hook a handler.

> Unless I've overlooked something, TBF and CBQ are only invoked if a)
....
> So net_bh basically slows down when we're waiting. Correct ?

Exactly.

> Is this roughly the scenario you've tested ?

Yes. The problem is that when one process is running not making
syscalls, the generator stops completely falling back to timer.
It means that pure soft timer is inappropriate for real time tasks
(including one, which this thread was started from). Variable resolution
timer is the only solution and, after it will be implemented,
soft timers will be fully useless.

Actually, in softnet I removed all the traces of this "softtimer"
approach. 8)8) qdiscs are kicked only by events on scheduled device
and by real time timer.

Alexey

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:54    [W:0.049 / U:0.088 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site