Re: 1Khz timer interrupt

[Milan Svoboda <msvoboda@xxxxxxxxxxxxxxx>]

Hello Jonathan,

linux-arm-kernel-bounces@xxxxxxxxxxxxxxxxxxxxxx wrote on 07/22/2008 
10:49:19 PM:

> Remy Bohmer wrote:
> 
> My current (temporary) solution was to implementa realtime
> clock driver which registers one clock per timer (with a
> global one to detect missed interrupts). It's 
> clunky but it
> works. (see pxa27x-rtc sent to rtc mailing list 
> rtc-linux@xxxxxxxxxxxxxxxx a couple of days 
> ago.) I'd definitely
> be interested in a more general approach to this.
> 

My temporary solution was the same (for ixp465). I started with
the driver that implements a realtime clock driver that registers itself 
to
the posix clocks. I also did an rtc like interface (character device)
and noticed that this character interface has a lower latency
times. But it may be due older version of Linux kernel I have to use.
So I still use this character interface...

> Anyone else interested enough to work out a general set of
> requirements?
> 

Sure, I'm. I think that there is real need to utilize available
hardware clocks on platforms that provide more than one hardware clock...

Milan


> --
> Jonathan Cameron
> 
> 
> -------------------------------------------------------------------
> List admin: http://lists.arm.linux.org.
> uk/mailman/listinfo/linux-arm-kernel
> FAQ:        http://www.arm.linux.org.
> uk/mailinglists/faq.php
> Etiquette:  http://www.arm.linux.org.
> uk/mailinglists/etiquette.php


-------------------------------------------------------------------
List admin: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:        http://www.arm.linux.org.uk/mailinglists/faq.php
Etiquette:  http://www.arm.linux.org.uk/mailinglists/etiquette.php