Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 11 Mar 2004 13:03:34 -0500
From:      John Baldwin <jhb@FreeBSD.org>
To:        Alexander Leidinger <Alexander@Leidinger.net>
Cc:        Putinas Piliponis <putinas.piliponis@icnspot.net>
Subject:   Re: still spurious interrupts with ICH5 SATA
Message-ID:  <200403111303.34855.jhb@FreeBSD.org>
In-Reply-To: <20040311163758.0fbc172e@Magellan.Leidinger.net>
References:  <001801c40745$043449d0$1e64a8c0@spotripoli.local> <40503B8A.7040105@DeepCore.dk> <20040311163758.0fbc172e@Magellan.Leidinger.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday 11 March 2004 10:37 am, Alexander Leidinger wrote:
> On Thu, 11 Mar 2004 11:12:26 +0100
>
> S=F8ren Schmidt <sos@DeepCore.dk> wrote:
> > Putinas Piliponis wrote:
> > > Hi all,
> > > I upgraded from 5.2.1 to todays current and I get this:
> > > ( my mobo is Asus P4P800 Deluxe )
> > > If I boot with hint acpi disabled - I get panic
> > > If I boot with apic disabled - doesn't make difference
> > > HTT is disabled in bios, and SMP not compiled in kernel.
> > >
> > > If I try to boot in verbose mode, I continues repeating this
> > > ata3: spurious interupt - status=3D0x50 error=3D0x00
> > > ( really alot I even cannot get dmesg output from booting, because th=
is
> > > ata3: blabla fill ups all the available space for msg ) but it still
> > > continues booting.
> >
> > Getting spurious interrupts does not nessesarily mean that something is
> > wrong, they will show up if you have shared irq's... If that is not the
> > problem it most likelu is a problem with interrupt routing or semialr
> > resource messups so that the interrupt is not properly ack'ed to the
> > device causing interrupt storms...
>
> I think the last time this was discussed, jhb (CCed) said it's maybe a
> problem in the SMP handling in the ata code...
>
> John, do I remember this correctly?

Well, I don't know where the bug lies, but I doubt it has to do with interr=
upt=20
routing.  Based on the vmstat -i output, none of the devices on the system=
=20
are storming (or even close to storming) so I don't think the routing can b=
e=20
at fault here.  It does seem that all the reports I see now of problems wit=
h=20
interrupts (other than a problem with storm on irq 20 with acpi0 when it is=
=20
not shared with any other devices) involve ata(4).

=2D-=20
John Baldwin <jhb@FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =3D  http://www.FreeBSD.org



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200403111303.34855.jhb>