> What do you do about reserve/release then?

That's the point about not doing a bus reset!!  You don't want to clear them
just by rebooting a system!

But anyway, PERSISTENT RESERVE IN/OUT is where folks are going on this anyway.

> 
> > Perhaps the GENERIC kernel can do the bus reset in this case. Which is what we
> > have, I guess. If I explicitly set SCSI_DELAY in my config file, we don't do
> > settling (easy enough to do). But we also don't encourage HBAs to *not* issue
> > bus resets with abandon. It's hard in such environments to believe that w/o a
> > lot of work NetBSD can be in a multi-initiator environment.
> 
> Well, a 2 second reset delay is o.k. for disks, but some RAID boxes can
> take upwards of 60 seconds to complete initialization after being reset.
> 
> The problem is not really the bus reset and the associated reset
> delay.  The problem is that a reset and reset delay is issued to each and
> every scsi bus that's being probed before the devices are probed, not when
> the HBA driver is attached.
> 
> If issued resets when the HBAs were attached we would only require a
> single reset delay.  (The next step would be multi-threaded probing....)

I still think that in the scenario of equipment you mention above, you don't
want to do resets if you can help it.

-matt