On Wed, 29 Nov 2006 16:28:01 +0000
George Harvey <fr30@dial.pipex.com> wrote:

> On Tue, 28 Nov 2006 00:33:26 -0500
> "Stephen M. Rumble" <stephen.rumble@utoronto.ca> wrote:
> 
> > Quoting George Harvey <fr30@dial.pipex.com>:
> > 
[ snip ]

> Tried a local build and got a seg fault after about 2hrs building, no
> hang though. Since I had no problems on an O2 running 3.1, I think
> I'll re-install the Challenge S with 3.1 userland and try the local
> build again with that.

I had a hunt in my disk collection and found a Seagate drive that would
allow the 3.1 install kernel to boot on my Challenge S without locking
up on the SCSI probe. Instead, it panics while creating mfs. The last
lines on the console are included below. I've been trying various
kernels on this machine and that's the first time I've seen an
amap_wipeout. Now that I've found a disk that allows 3.1 to boot, I
think I'll try attaching it to my O2, do the 3.1 install on that, and
then try it back in the Challenge S.

What is the 'amap' and what might cause it to be corrupt?

Regards,
George

Console output from R5000/180 Challenge S booting 3.1 INSTALL:
...
sd0 at scsibus0 target 5 lun 0: <SEAGATE, ST34371N, 0484> disk fixed
sd0: 4148 MB, 5172 cyl, 10 head, 164 sec, 512 bytes/sect x 8496884
sectors
sd0: sync (200.00ns offset 12), 8-bit (5.000MB/s) transfers,tagged
queueing
boot device: <unknown>
root on md0a dumps on md0b
root file system type: ffs
WARNING: clock gained 30 days -- CHECK AND RESET THE DATE!
warning: no /dev/console
init: Creating mfs /dev (407 blocks, 1024 inodes)
panic: amap_wipeout: corrupt amap
Stopped in pid 10.1 (mknod) at  0x882caf94:     jr      ra
                bdslot: nop
db>