Changes to kern.geom.debugflags?

Chris Ross cross+freebsd at distal.com
Mon Feb 11 03:16:23 UTC 2013


On Jan 17, 2013, at 19:49 , Chris Ross <cross+freebsd at distal.com> wrote:
> On Jan 17, 2013, at 11:35 , Andriy Gapon <avg at freebsd.org> wrote:
>> Chris,
>> 
>> thank you for triaging and analyzing this problem.  And sorry for the long delay
>> (caused by the New Year craziness you mentioned earlier).
>> 
>> The problem is that arch_zfs_probe methods are expected only to probe for ZFS
>> disks/partitions, but they are not allowed to execute any other ZFS operations.
>> I assumed this to be true and forgot to check sparc64_zfs_probe.  Mea culpa.
>> 
>> Could you please test the following patch?
> 
>  Thank you, Andriy.  Much as you'd expect, that patch solves the problem.  I get some
> of the printf()s that I'd put into zfs_fmtdev(), and the system loads successfully.
> 
>  Please commit that patch, and if you could, change the comment just below the last
> portion of it that is now not quite accurate (since you moved mentioned code).
> 
>  Thanks again!  How long will this take to get to stable/9?  Being new to FreeBSD,
> I'm not too familiar with the process of HEAD/stable/etc.  (In NetBSD, it would be a
> commit followed by a pull request.)

  Sad to say that after hand-testing that patch, I waited for it to appear on stable-9,
(by manual inspection of the relevant code), and tried again.  This time, I get a
slightly different failure:

Rebooting with command: boot                                          
Boot device: disk1  File and args: 
 
>> FreeBSD/sparc64 ZFS boot block
    Boot path:   /pci at 1c,600000/scsi at 2/disk at 1,0:a
ERROR: Last Trap: Memory Address not Aligned

{1} ok

  This is with a zfsloader built from stable-9 as of Feb 2.  I'm updating and rebuilding
now, just to check, but I wanted to send out a note incase anyone else on the
sparc64 list has also seen this.

               - Chris



More information about the freebsd-sparc64 mailing list