boot2 -- Round 2
Jeremy Chadwick
freebsd at jdc.parodius.com
Sun Aug 1 09:20:57 PDT 2004
Small follow-up (useful for any others who are having issues):
1. /boot/boot between "Box A" and "Box B" differ. cmp -x shows
what looks almost like an endian-related problem, but only for a
small portion of boot2. boot0 and boot1 are identical. "medusa"
is "Box B":
medusa# cmp -x /boot/boot2 box_a_boot/boot2
00000910 0c 08
00000914 08 0c
00000920 03 01
00000922 01 03
00000923 43 41
00000924 41 43
00001bc9 01 0c
2. Comparing /boot/boot from 5.2.1-RELEASE and the latest cvs HEAD
shows an immense amount of differences. This is probably justified.
3. Using /boot/boot from 5.2.1-RELEASE on "Box B" works flawlessly.
0:ad(0,a) is chosen as expected, and everything is beautiful.
4. #3 works fine using cvs HEAD's 'disklabel' binary, so the utility
itself is likely fine.
Just wanted to throw all this out there for documentation/discussion
pruposes.
--
| Jeremy Chadwick jdc at parodius.com |
| Parodius Networking http://www.parodius.com/ |
| UNIX Systems Administrator Mountain View, CA, USA |
| Making life hard for others since 1977. |
More information about the freebsd-current
mailing list