Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 22 Apr 2004 22:51:24 +0200
From:      Eirik Oeverby <ltning@anduin.net>
To:        John Baldwin <jhb@FreeBSD.org>
Cc:        current@freebsd.org
Subject:   Re: SMP weirdness in -CURRENT
Message-ID:  <4088304C.6000108@anduin.net>
In-Reply-To: <200404220957.51856.jhb@FreeBSD.org>
References:  <200404201311.i3KDBY7C071099@anduin.net> <408524D6.9010603@bis.midco.net> <40862C93.7040803@anduin.net> <200404220957.51856.jhb@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi,

I don't *have* anything on IRQ 20.
Here's the output of vmstat -i after >2 days uptime:

[root@ranger] ~# vmstat -i
interrupt                          total       rate
irq0: clk                       22888351         99
irq1: atkbd0                       20530          0
irq8: rtc                       29298041        128
irq9: acpi0                         3902          0
irq11: cbb0 cbb1+++               203530          0
irq12: psm0                       117066          0
irq13: npx0                            1          0
irq14: ata0                       313747          1
irq15: ata1                           46          0
Total                           52845214        230
[root@ranger] ~# uptime
10:50PM  up 2 days, 15:36, 5 users, load averages: 0.14, 0.14, 0.11

Note that this is *without* the suggested loader.conf option.


/Eirik


John Baldwin wrote:
> On Wednesday 21 April 2004 04:10 am, Eirik Oeverby wrote:
> 
>>Hi,
>>
>>Yes I'm using ACPI. I assume that without ACPI things wouldn't work
>>overly well at all..?
>>When was this fix committed? I have a fairly recent kernel, too, see below.
> 
> 
> do a 'vmstat -i' to see if you are getting an interrupt storm.  If you have a 
> very high count of interrupts on IRQ 20, add the 'hw.acpi.force_sci_lo' 
> option to loader.conf as the previous poster said.
> 



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4088304C.6000108>