In discussion with Lashi ([EMAIL PROTECTED]) after my earlier posting, I've been shown another way I can lock up my system. Occasionally removing the g_NCR5380 module causes a kernel panic. Its not completely consistent in my case and seems most likely to occur when scanimage has been stopped mid-scan. Included is the results of running my kerenl oops through ksymoops. Lashi reports a similiar problem. > Output of ksymoops > > Unable to handle kernel paging request at virtual address c886e714 > current->tss.cr3 = 00101000, %cr3 = 00101000 > *pde = 07d11063 > EFLAGS: 0010096 > eax: c09c32fc ebx: c886e714 ecx: c886e714 edx: 0000008e > esi: c09c32fc edi: 00000001 ebp: c01d5f48 esp: c01d5f44 > ds: 0018 es: 0018 ss: 0018 > Call Trace: [<c01177c5>] [<c010a24d>] [<c0109f20>] [<c0107885>] [<c0106000>] >[<c01078a8>] [<c0109084>] > [<c0106000>] [<c010607b>] [<c0106000>] [<c010607b>] [<c0106000>] >[<c0100176>] > Code: 8b 1b 8b 41 08 89 c2 2b 15 04 38 1c c0 81 fa ff 00 00 00 77 > > Trace: c01177c5 <do_bottom_half+49/64> > Trace: c010a24d <do_IRQ+39/40> > Trace: c0109f20 <common_interrupt+18/20> > Trace: c0107885 <cpu_idle+5d/6c> > Trace: c0106000 <get_options+0/74> > Trace: c01078a8 <sys_idle+14/24> > Trace: c0109084 <system_call+34/38> > Trace: c0106000 <get_options+0/74> > Code: 00000000 Before first symbol 00000000 <_IP>: <=== > Code: 00000000 Before first symbol 0: 8b 1b movl >(%ebx),%ebx <=== > Code: 00000002 Before first symbol 2: 8b 41 08 movl >0x8(%ecx),%eax > Code: 00000005 Before first symbol 5: 89 c2 movl >%eax,%edx > Code: 00000007 Before first symbol 7: 2b 15 04 38 1c subl >0xc01c3804,%edx > Code: 0000000c Before first symbol c: c0 > Code: 0000000d Before first symbol d: 81 fa ff 00 00 cmpl >$0xff,%edx > Code: 00000012 Before first symbol 12: 00 > Code: 00000013 Before first symbol 13: 77 00 ja > 00000015 Before first symbol > > Aiee, killing interrupt handler > kernel panic: Attempting to kill the idle task! > In swapper task - not syncing I've taken a look at the g_NCR5380 driver without any blinding flashes of enlightenment. Does anyone have any pointers to specs on this card? I suspect I might have a better chance of understanding the code if I actaully knew what it was trying to achieve in the first place. -- Neil Muller [EMAIL PROTECTED] - To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to [EMAIL PROTECTED]
