tech-smp note: this thread orginally started in -current under subject 
heading "isp driver issues", however, I think it's more an smp issue, so 
I've brought it here instead.

Matthew Jacob wrote:

>
> I'm pretty sure that Bill can help. You might also try the suggestion 
> I made on -current.
>
I wanted to follow up and let you know that your suggestion did work.

    "disabling the code in isp_netbsd.c:isp_config_interrupts that sets
    isp->isp_osinfo.no_mbox_ints to zero (i.e., leave it set to one) and
    see if that helps."

However now there is an extra isp device listed during bootup. Also, 
dmesg is misbehaving. Instead of the normal reports, all it lists are:

: {4} dmesg |less
in RESPONSE Queue (type 0x0) @ idx 62 (next 63) nlooked 44
isp1: Unhandled Response Type 0x0
isp1: Not RESPONSE in RESPONSE Queue (type 0x0) @ idx 63 (next 0) nlooked 45
... last two lines repeat many times ...
... once in a while it would say "isp1: Polled Mailbox Command (0x15) 
Timeout" ...
isp1: Polled Mailbox Command (0x15) Timeout
boot device: ld0
root on ld0a dumps on ld0b
root file system type: ffs
cpu1: CPU 0 running
cpu2: CPU 1 running
wsdisplay0: screen 1 added (80x25, vt100 emulation)
wsdisplay0: screen 2 added (80x25, vt100 emulation)
wsdisplay0: screen 3 added (80x25, vt100 emulation)
wsdisplay0: screen 4 added (80x25, vt100 emulation)

isp1 and isp2 isn't the qla2200F, but some other Qlogic wide SCSI 
(channel A and B) NetBSD claims I have but I don't. isp0 is the fibre 
card. With the change above, I now also have isp3 on bootup. If I do, 
"scsictl /dev/scsibus0 scan any any" (that's the fibre bus) then nothing 
is shown because it's detected already. However if I do, ""scsictl 
/dev/scsibus1 scan any any," (which is the phantom wide SCSI channel), 
then the system reports, in an endless loop, on the main console"

isp1: Unhandled Response Type 0x0
isp1: Not RESPONSE in RESPONSE Queue (type 0x0) @ idx 63 (next 0) nlooked 45

Regards,
Thomas