Re: IOAPIC and 8254

From: Lyle Coder (x_coder@hotmail.com)
Date: Thu Mar 02 2000 - 15:56:30 EST


Im sorry, what I actually meant was....
So on an MP system, does Linux use the LOCAL APIC or the 8254 as the actual
timer tick source... ie... which device actually generates the timer ticks
(IRQ0)? And what vector does it generate?

Best Wishes,
Lyle

>From: Chris Wedgwood <cw@f00f.org>
>To: "Maciej W. Rozycki" <macro@ds2.pg.gda.pl>
>CC: Lyle Coder <x_coder@hotmail.com>, linux-kernel@vger.rutgers.edu
>Subject: Re: IOAPIC and 8254
>Date: Fri, 3 Mar 2000 09:12:05 +1300
>
>On Thu, Mar 02, 2000 at 07:55:29PM +0100, Maciej W. Rozycki wrote:
>
> > As a source of the system time tick. We could probably use the RTC
> > periodic interrupt instead, but HZ got fixed at 100 on i386 long time
>ago
> > and the best approximation we can get from the RTC is 128 ;-) as it only
> > allows powers of 2. Alphas use 1024 for HZ so they do not suffer from
> > this problem.
>
>I theory, it should be pretty safe to change the default value of HZ
>except many userspace programs probably 100 as an assumption and may
>give funny results.
>
>People do increase HZ to much higher values (10000?) to be better
>control over some kernel features such as CBQ.
>
>
>-cw

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com

-
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/



This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:13 EST