It would seem that 

cd1 at scsibus0 target 4 lun 0: <HP, CD-Writer+ 9200, 1.0e> SCSI4 5/cdrom removable

thinks it's target 0 every now and then? Is that possible? End result: timeouts

but at least no apparent data corruption as opposed to the tagged queueing
with quantum atlas II.

Cheers,

Patrick

sd0(ahc0:0:0): BDR message in message buffer
sd0(ahc0:0:0): SCB 1c - timed out in Data-out phase, SEQADDR == 0x111
SCSIRATE == 0xf
sd0(ahc0:0:0): no longer in timeout, status = 0
ahc0: Issued Channel A Bus Reset. 2 SCBs aborted
ahc0: target 2 synchronous at 20.0MHz, offset = 0xf
ahc0: target 0 synchronous at 20.0MHz, offset = 0xf
ahc0: target 0 synchronous at 20.0MHz, offset = 0xf