spurious 8259A interrupt : IRQ7 + motherboard

Matthew Burgess ca9mbu at eos.sunderland.ac.uk
Fri Jan 17 03:36:35 PST 2003


""Sequeira, Tony"" <Tony.Sequeira at Edexcel.org.uk> wrote in message
news:92E04269C4EED411BD430008C716A2320802103F at nymph.internal.edexcel.org.uk.
..
> > -----Original Message-----
> > From: Matthew Burgess [mailto:ca9mbu at eos.sunderland.ac.uk]
> > Sent: 17 January 2003 10:59 am
> > To: lfs-support at linuxfromscratch.org
> > Subject: Re: spurious 8259A interrupt : IRQ7 + motherboard
> >
> >
> > "Ian Molton" <spyro at f2s.com> wrote in message
> > news:20030117084045.2867b491.spyro at f2s.com...
> > > On Fri, 17 Jan 2003 02:48:37 +0000 (UTC)
> > > jrh29 at po.cwru.edu (Justin Hibbits) wrote:
> > >
> > > > > I think thats what the 'spurious' is supposed to mean, dude ;-)
> > > > >
> > > >
> > > > I know that.  But the kernel should silently ignore
> > interrupts from an
> > > > uninitialized device, and if it doesn't, they should fix it.
> > >
> > > Im no expert on the X86, but isnt IRQ 7 a multiplexed IRQ?
> > >
> > > If so, its not possible to mask it whilst one device is unmasked.
> > >
> > > >  btw, it
> > > > may depend on the kernel too.  I use 2.4.18, and it works like a
> > > > charm, no noise whatsoever.  Hell, it may just be I'm
> > lucky at this.
> > >
> > > probably :)
> >
> > Ian, your knowledge of this sounds far beyond mine, but I
> > seem to remember
> > having this problem a while back with my Athlon XP rig.  I
> > did a search on
> > google and was lead to believe it was the inclusion of one of
> > the "APIC"
> > options in the kernel config that caused this particular
> > message to fire off
> > (try "Local APIC support" or similar).  Disabling support for
> > this had no
> > ill effects, and removed the message once and for all.
> >
> > Regards,
> >
> > Matt.
>
> That is odd.  My old 486 without APIC support of any
> kind in the kernel comes up with this.  Only at system boot
> time and only when scanning the IDE drives and always on
> drive 3.  Kernel 2.4.18.  The thing is I had a disk failure
> on that very drive only days after I upgraded and this
> message appeared, but this may be coincidental, as the machine
> and drives are very old.  Another drive, and total rebuild
> of the system still shows the same symptoms.
>
Well, just found this from google
https://listman.redhat.com/pipermail/enigma-list/2002-October/017511.html

---begin quote---
"A device with part number 8259A, which is a programmable interrrupt
controller used with many devices to interface with the system bus, has
generated a unexpected Interrrupt.  Asked for service without a real
request.  ie. Noise.........

what lies around this line in the log file may give you a clue as to what
device is actually generating the interrrupt. If it only happens at boot
time, it is possible that the device driver has a bug in it's initialization
routine that generates a spurious interrrupt to the OS. It should be
reported appropriately.

If it happens all the time while the system is running, you have a bad
hardware problem or something is loose or improperly connected.  First ID
the real source of the interrupt. The 8259A could be on many devices or it
may be on the MB. (more than one present.)"
--end quote---

So this I think explains why disabling local APIC support on my machine
cured "my" problem, but wouldn't necessarily solve anyone elses, it just
happened to be the APIC which had/was using the 8259A device. - maybe.

Matt.




-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-support' in the subject header of the message



More information about the lfs-support mailing list