lkml.org 
[lkml]   [1998]   [Aug]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Linux Interrupt Latency benchmark v4 avaible -> detected irq bottleneck before do_bottom_half()
On Mon, 31 Aug 1998, Alexander Kjeldaas wrote:

>Btw, I ran your module (v3) for 14 hours on a PII233 and got 3.9us min
>and 20.3max. The mean calculated by the module couldn't be trusted
>because of the integer arithmetic, but from other tests with fewer

Yes it can' t be trusted too much.

>interrupts, I think it's very close to the minimum time.

If it' s not precise it doesn' t mean that it has to be very close to the
minimum time per default.

To decrease the max irq latency there are few things to do. You have to
cut the timer irq handler duration (that seems the only thing that harm
here). I don' t think it' s safe to run the irq handler with irq enabled.

If you want to try simply change this line in arch/i386/kernel/time.c
from:

static struct irqaction irq0 = { timer_interrupt, SA_INTERRUPT, 0, "timer", NULL, NULL};

to:

static struct irqaction irq0 = { timer_interrupt, 0, 0, "timer", NULL, NULL};

I am not going to try it ;-).

Andrea[s] Arcangeli


-
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.altern.org/andrebalsa/doc/lkml-faq.html

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.334 / U:0.032 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site