During non-intensive SCSI i/o (isp0) basically just attempting to
read or write a file, for instance writing out a large tar file, I'm
seeing the following:

Mar 14 12:19:32 droog /netbsd: stray 6600 irq 47
Mar 14 12:20:28 droog /netbsd: stray 6600 irq 47
Mar 14 12:21:09 droog /netbsd: stray 6600 irq 47
Mar 14 12:21:09 droog /netbsd: stray 6600 irq 47; stopped logging

At that point the devices attached to the isp0 can be accessed intermittenly
without errors (for instance, there might be a pause of up to 10 or
so seconds while doing an 'ls' on a filesystem attached to the isp0

Or, my tar command's status of:

load: 3.95  cmd: tar 5027 [pager_map] 0.31u 5.76s 6% 608k
load: 0.76  cmd: gzip 5028 [biowait] 12.77u 0.82s 0% 56k
load: 0.88  cmd: tar 5027 [pager_map] 0.32u 6.19s 0% 216k

NetBSD droog 1.6.1_RC2 NetBSD 1.6.1_RC2 (sdf) #0: Fri Mar 14 03:16:48 UTC 2003

sp0 at pci0 dev 17 function 0: QLogic 1080 Ultra-2 Wide SCSI HBA
isp0: interrupting at dec 6600 irq 47
scsibus0 at isp0: 16 targets, 8 luns per target

The machine is a DS10L 21264