Subject: Fwd: Re: isp driver issues in SMP on NetBSD (fwd)
To: Thomas T. Thai <firstname.lastname@example.org>
From: Matthew Jacob <email@example.com>
Date: 11/15/2004 10:24:17
I believe that we can drop tech-smp from this list- I don't believe
that this is an SMP issue. The change I suggested just ducks around
what is apparently some issue with config_interrupts not working
correctly for some configurations- the config_interrupts hook turned on
using interrupts for the rest of mailbox (housekeeping type) commands-
like fibre channel SAN evaluation. Apparently the interrupts weren't
actually getting through after the config_interrupts hook was executed.
The easiest solution is to live with polling for these types of
commands. FreeBSD also broke things by insisting on an architecture
from the new GEOM code that wouldn't allow a driver to sleep, tra la.
The other issue here- a 'phantom' isp, sounds more like some kind of
wierd configuration issue just with your box. If you want to just run,
compile in the option that turns of 12160 support.
> Matthew Jacob wrote:
> > I missed a lot of details (as I really don't pay much attention to
> the *BSDs
> > these days), but is this 'extra' isp the Ultra3 part? If so, it's
> > not really a 12160 but part of some RAID unit- not really a true
> 12160 part
> > (there are checks in OpenBSD and FreeBSD to catch just this
> occurrence- can't
> > remember if this is also true for NetBSD)
> Yes it's the Ultra3 part. Originally there were only two phamtom isp
> (Chan A and B) isp1 and isp2. Now there is a 3rd one, isp3. isp0
> being the
> real fibre card.
> isp0 at pci1 dev 9 function 0: QLogic FC-AL HBA
> isp0: interrupting at ioapic1 pin 7 (irq 10)
> isp1 at pci2 dev 1 function 0: QLogic Dual Channel Ultra-3 Wide SCSI
> isp1: interrupting at ioapic1 pin 5 (irq 5)
> I only have a built-in raid:
> ahc2 at pci0 dev 5 function 1: Adaptec aic7899 Ultra160 SCSI adapter
> ahc2: interrupting at ioapic1 pin 2 (irq 10)
> and the QLA-2200F. There is no other Qlogic Ultra-3 Wide.
> The original problem of getting the RAID array detected by the isp
> device is
> resolve per your suggestion. However, I'm not sure if that may have
> other issues to crop up.
Do you Yahoo!?
Check out the new Yahoo! Front Page.