scsi errors, how dangerous could it be?

[Alexey Chetroi <debian@twilight.telco.md>]

 Hello All,

 I have some messages in /var/log/syslog which attracted my attention.
it about scsi errors which I don't fully understand:
Dec  1 07:40:30 ial kernel: scsi0: PCI error Interrupt at seqaddr = 0x9
Dec  1 07:40:30 ial kernel: scsi0: Data Parity Error Detected during address or
write data phase

 These message appear time to time. I've tried to google for this
problem, but haven't figured anything, beside that: 1) problem exists :(
2) it also happens to bsd folks. 
 I'm running fresh woody install, kernel-2.4.18-686, on Asus P2B motherboard
whith integrated scsi controller. Because of this I cannot dedicate a single
IRQ to this controller:
ial:/etc/raddb# cat /proc/interrupts
           CPU0
  0:    6548460          XT-PIC  timer
  1:          2          XT-PIC  keyboard
  2:          0          XT-PIC  cascade
  5:   15754026          XT-PIC  mxser
  8:          3          XT-PIC  rtc
 10:   13550691          XT-PIC  mxser
 12:    1901848          XT-PIC  eth0
 14:     482894          XT-PIC  aic7xxx, usb-uhci
 15:         17          XT-PIC  ide1
NMI:          0
LOC:    6548426
ERR:          0
MIS:          0
ial:/etc/raddb#

Is this a problem? Bios setting disables IRQ for USB, but usb-uhci still uses
it. Could USB be disabled without recompiling kernel?
Motherboard is dual-pentium ready, but only one processor is installed.
Should I use kernel-image-*-smp or not?

 Any help is apreciated. Thank you in advance.

-- 

  Best regards,
  Alexey Chetroi

---
Smile... Tomorrow will be worse.   (c) Murphy's law