This page is dedicated to random notes about hardware I've used with Linux. I'm not trying hard to link it to things, but am counting on those wonderful full-text search engines to make it useful to people. I know when I get a random problem with hardware, I try typing terms into a search engine.
Jul 1 01:26:27 eusa kernel: scsi : aborting command due to timeout : pid 17, sc si0, channel 0, id 3, lun 0 Test Unit Ready 00 00 00 00 00 Jul 1 01:26:57 eusa kernel: scsi : aborting command due to timeout : pid 17, sc si0, channel 0, id 3, lun 0 Test Unit Ready 00 00 00 00 00 Jul 1 01:26:57 eusa kernel: (scsi0:0:15:7) Invalid SCB during SEQINT 0x71, SCB_ TAG 255. Jul 1 01:27:27 eusa kernel: scsi : aborting command due to timeout : pid 17, sc si0, channel 0, id 3, lun 0 Request Sense 00 00 00 10 00 Jul 1 01:27:29 eusa kernel: SCSI host 0 abort (pid 17) timed out - resetting Jul 1 01:27:29 eusa kernel: SCSI bus is being reset for host 0 channel 0. Jul 1 01:28:00 eusa kernel: SCSI host 0 abort (pid 17) timed out - resetting Jul 1 01:28:00 eusa kernel: SCSI bus is being reset for host 0 channel 0. Jul 1 01:28:01 eusa kernel: SCSI host 0 channel 0 reset (pid 17) timed out - tr ying harder Jul 1 01:28:01 eusa kernel: SCSI bus is being reset for host 0 channel 0. Jul 1 01:28:01 eusa kernel: (scsi0:0:3:0) Performing Domain validation. Jul 1 01:28:32 eusa kernel: SCSI host 0 abort (pid 17) timed out - resetting Jul 1 01:28:32 eusa kernel: SCSI bus is being reset for host 0 channel 0. Jul 1 01:28:32 eusa kernel: (scsi0:0:3:0) Successfully completed Domain validation. panic: Unable to handle kernel NULL pointer dereference at virtual address 00000004 current->tss.cr3 = 00101000, %cr = 00101000 *pde = 00000000 Oops: 0000 CPU: 0 EIP: 0010:[(more or less--in my case it said "trying harder").] EFLAGS: 00010046 eax: 00001064 ebx: c7fe3400 ecx: c009a878 edx: 0000001f esi: c009a878 edi: 00000000 ebp: c8000000 esp: c0229e94 ds: 0018 es: 0018 ss: 0018 Process swapper (pid: 0, process nr: 0, stackpage = c0229000) Stack: c009a878 0000000b c0235dbe c0235dff 0000003b 000000ff c01ffbe0 c01cd61d c009a878 c7fe3400 00000003 c009a878 00000286 c0229f68 0000000b c01c7e58 c009a878 00000004 c009a878 00000286 c0229f68 0000000b 00000001 00000004 Call Trace: [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] Code: 8b 47 04 31 db 31 d2 8a 58 40 8a 50 42 c1 e2 03 09 d3 8d 96 Aiee, killing interrupt handler Kernel panic: Attempted to kill the idle task! In swapper task - not syncing
The work-around is to up the reset delay value (kernel parameter CONFIG_AIC7XXX_RESET_DELAY). In my case, 20 (seconds) worked sometimes; 45 seems to work consistently (of course, I don't reboot often with Linux :-).
Clearly this is a hack around some kind of problem in the firmware.
My home PC (with an EFA P6BX motherboard) has an WinBond W83781D chip for temperature monitoring. Basically this is a clone of a more popular chip (the LM78/LM79). It's supposed to be compatible, but doesn't seem to be.
LM specs ar at National (see the web page or pdf). The Winbond W83781D specs are at Winbond (web page, data sheet in pdf).
Unfortunately, I haven't yet gotten around to writing the driver for it. If anyone has a driver or hears of one one, please let me know.