From owner-freebsd-hackers  Sat Nov 16 02:39:35 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id CAA20323
          for hackers-outgoing; Sat, 16 Nov 1996 02:39:35 -0800 (PST)
Received: from Octopussy.MI.Uni-Koeln.DE (Octopussy.MI.Uni-Koeln.DE [134.95.166.20])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id CAA20317
          for <hackers@freebsd.org>; Sat, 16 Nov 1996 02:39:29 -0800 (PST)
Received: from x14.mi.uni-koeln.de (annexr2-44.slip.Uni-Koeln.DE) by Octopussy.MI.Uni-Koeln.DE with SMTP id AA05042
  (5.67b/IDA-1.5 for <hackers@freebsd.org>); Sat, 16 Nov 1996 11:39:19 +0100
Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.2/8.6.9) id LAA00640; Sat, 16 Nov 1996 11:39:06 +0100 (MET)
Message-Id: <199611161039.LAA00640@x14.mi.uni-koeln.de>
Date: Sat, 16 Nov 1996 11:39:05 +0100
From: se@freebsd.org (Stefan Esser)
To: mrcpu@cdsnet.net (Jaye Mathisen)
Cc: hackers@freebsd.org
Subject: Re: IRQ sharing on PCI?
References: <Pine.NEB.3.95.961115163716.2837T-100000@mail.cdsnet.net>
X-Mailer: Mutt 0.49-PL10
Mime-Version: 1.0
In-Reply-To: <Pine.NEB.3.95.961115163716.2837T-100000@mail.cdsnet.net>; from Jaye Mathisen on Nov 15, 1996 16:38:56 -0800
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On Nov 15, mrcpu@cdsnet.net (Jaye Mathisen) wrote:
> 
> 
> Do we support this?
> 
> I noticed on 1 of my boxes when I rebooted it the following:
> 
> tible display device> rev 76 int a irq 11 on pci0:8
> de0 <Digital DC21140 Fast Ethernet> rev 18 int a irq 11 on pci0:9
> de0: DC21140 [10-100Mb/s] pass 1.2 Ethernet address 00:00:c0:4b:10:ed
> de0: enabling 10baseT UTP port
> ahc0 <Adaptec 2940 SCSI host adapter> rev 0 int a irq 11 on pci0:11
> 
> 
> Notice that they're all IRQ11, which I thought seemed odd.

Why ?

> None of my other Pentiums do it, they all get different IRQ's.

This depends on the PCI BIOS, which is free 
to assign any IRQ to any PCI Int line.
PCI requires shared interrupts to work, since
there are far less real interrupt request 
inputs in a typical system, than independent 
PCI Int lines.

> Anyway, any idea appreciated, I suppose it's entirely possible that it's
> just a reporting problem, as opposed to anything actually wrong.

If it really bothers you (there is in fact 
some overhead caused by the polling of both
the EThernet and SCSI chip, if one of them
signaled an interrupt), then you'll have to
look into your PCI BIOS setup. That is where
the PCI interrupts are assigned to IRQs. 

The chip set will activate the PCI Int to
ISA IRQ routing at POST, and will put the
IRQ choosen in some PCI config space register
of each PCI chip. That is the place where the
driver finds it. The driver has to accept 
whatever is there, since the interrupt routing
is different for each chip set, and only the 
BIOS knows how to do it right :)

Regards, STefan

> 
>