lkml.org 
[lkml]   [1999]   [Jul]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Can't sleep less than 20 ms
On Fri, 9 Jul 1999, Jamie Lokier wrote:
> Matti Aarnio wrote:
> > > I'd rather see:
> > >
> > > - slow HZ -- low interrupt load
> > > - accurate timers on demand
> > > - *precise* timing with accuracy of hardware
> >
> > RTC based interrupts ? Highly accurate periodic interrupts,
> > not re-sceduling based "regular by luck"...
>
> I'd like "regular by design".

:)

> Note that RTC interrupts are just as inaccurate for user space
> apps than 8253-based accurate timers -- the scheduling overhead
> and lack of guarantee is still present.

- RTC interrupts can be delivered on any 'wanted' time,
not just on 10ms marks
- the lack of guarantee isn't that bad for 99% of applications
(multimedia and related stuff)
- using a decent timer will improve multimedia performance
- having a decent timer will/might allow us to better pinpoint
the cause of the lack of guarantee
- the scheduling overhead is very small already

Maybe we want a special 'multimedia' scheduling mode with these
properties:
- able to ask for rescheduling on RTC interrupts
- CPU time is twice as expensive
- priority is higher, but the process gets less CPU to hog

This way, multimedia programmers are encouraged to code
their program with one 'worker' thread and one 'output'
thread -- with some buffering in between to catch most
I/O delays etc...

For real RT or low-latency work, you will need to work
on an unloaded computer anyway.

Rik -- Open Source: you deserve to be in control of your data.
+-------------------------------------------------------------------+
| Le Reseau netwerksystemen BV: http://www.reseau.nl/ |
| Linux Memory Management site: http://www.linux.eu.org/Linux-MM/ |
| Nederlandse Linux documentatie: http://www.nl.linux.org/ |
+-------------------------------------------------------------------+


-
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 13:52    [W:0.111 / U:0.104 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site