Subject: Re: SoundBlaster 64 PCI and eap driver
To: Nathan J. Williams <nathanw@MIT.EDU>
From: Manuel Bouyer <>
List: port-i386
Date: 11/16/2000 15:49:39
On Thu, Nov 16, 2000 at 09:44:13AM -0500, Nathan J. Williams wrote:
> It's absolutely a software bug (probably the driver, or possibly but
> less likely, a low-level interrupt-handling code bug). PCI devices and
> their drivers *must* be able to share interrupts. 
> There's no requirement in PCI that interrupts be distinct; it would be
> completely legitimate for all PCI devices in a system to share a
> single interrupt (and I think I've seen people post here about
> machines, particularly laptops, set up this way).

Yes, however, by design some PCI boards makes it very hard to handle
shared IRQ because there's no easy way to know if the board has generated
the interrupt or not (especially I have one PCI IDE controller in mind).

Manuel Bouyer, LIP6, Universite Paris VI.