cannot boot system built today

Brandon Fosdick bfoz at terrandev.com
Sun Sep 5 13:18:45 PDT 2004


I finally found some spare time to track this problem down.

I did a binary search of RELENG_4 from March 2004 to May 2004 and found 
that the last version that boots properly is dated 28 Mar 2004 @ 
11:45am. The version from the same day @ noon displays the familiar boot 
hang. The only commit in those 15 minutes is at 11:50 and appears to be 
a fairly extenive update to the firewire subsystem.

That seemed odd to me so I built a kernel from the noon (non-working) 
source with firewire disabled and it booted just fine. I started playing 
around and found that a non-working firewire-enabled kernel will boot 
normally with my iPod plugged in.  In fact it will continue to boot 
normally even after removing the iPod. Disconnecting the computer's 
power cable from the wall causes the familiar boot hang to return. 
However, if the iPod is connected after the computer hangs, it will 
proceed to boot properly.

So something is definately wrong with the firewire code, but I don't 
have the expertise to fix it. I'll keep poking to see what I can find, 
but I could really use the help of someone more knowledgeable. For now 
the easiest workaround is to plug in an iPod, although I imagine any 
firewire drive will do.

Attached is the output of diff -r --brief that lists the files that were 
updated at 11:50am on 28 Mar 2004.



Doug White wrote:
> On Fri, 11 Jun 2004, Brandon Fosdick wrote:
> 
> 
>>Doug White wrote:
>>
>>>I'd _love_ to have a machine with broken ATA so I could debug this stuff,
>>>but I either have working ATA or SCSI systems.
>>
>>I'd love for somebody to get this working.
>>
>>Do you need a whole machine or just a motherboard? I'd be happy to send
>>you a board if you're serious about debugging this.
> 
> 
> I don't know if I have the necessary skills to produce a fix, but I'll
> happily confirm the problem for you. :)
> 
> I'd need a full system, since I don't have much in the way of spare parts.
> Definitely no CPUs, memory, or power supplies+chassis.  Lots of SCSI disks
> and SCSI CDROMs tho :)
> 


-- 
Brandon Fosdick
http://www.terranspace.org

-------------- next part --------------
Files 0/src/sys/dev/firewire/firewire.c and 1/src/sys/dev/firewire/firewire.c differ
Files 0/src/sys/dev/firewire/firewire.h and 1/src/sys/dev/firewire/firewire.h differ
Files 0/src/sys/dev/firewire/firewirereg.h and 1/src/sys/dev/firewire/firewirereg.h differ
Files 0/src/sys/dev/firewire/fwcrom.c and 1/src/sys/dev/firewire/fwcrom.c differ
Files 0/src/sys/dev/firewire/fwdev.c and 1/src/sys/dev/firewire/fwdev.c differ
Files 0/src/sys/dev/firewire/fwdma.c and 1/src/sys/dev/firewire/fwdma.c differ
Files 0/src/sys/dev/firewire/fwmem.c and 1/src/sys/dev/firewire/fwmem.c differ
Files 0/src/sys/dev/firewire/fwohci.c and 1/src/sys/dev/firewire/fwohci.c differ
Files 0/src/sys/dev/firewire/fwohci_pci.c and 1/src/sys/dev/firewire/fwohci_pci.c differ
Files 0/src/sys/dev/firewire/fwohcireg.h and 1/src/sys/dev/firewire/fwohcireg.h differ
Files 0/src/sys/dev/firewire/fwohcivar.h and 1/src/sys/dev/firewire/fwohcivar.h differ
Files 0/src/sys/dev/firewire/if_fwe.c and 1/src/sys/dev/firewire/if_fwe.c differ
Files 0/src/sys/dev/firewire/sbp.c and 1/src/sys/dev/firewire/sbp.c differ
Files 0/src/sys/dev/firewire/sbp_targ.c and 1/src/sys/dev/firewire/sbp_targ.c differ


More information about the freebsd-stable mailing list