this week work on KBOOT

Andy Green andy at
Sat May 31 17:02:09 CEST 2008

Hash: SHA1

Somebody in the thread at some point said:
| Dear Xiangfu,
| sounds good!
| I wouldn't worry about being slow at the beginning. Openmoko is big and
| once you spend some time to understand the system, you will become a lot
| faster in the future.
| Andy, are you saying we should not use Kexec/Kboot? Just a reduced u-boot?
| Unfortunately Werner is on vacation so we have to guess what he would say.
| Can you settle on something and give Xiangfu clear direction? Then
| hopefully once Werner is back he will agree with that...

I explained over a month ago that kexec / kboot is not critical for us
compared to bootloader simplification.

Clear direction:

1) reduce the bootloader to a minimum and use the fixed worst-case
partitioning offset method instead of dynparts,

2) implement backup kernel and rootfs,

3) figure out if kboot / kexec buys us ANYTHING over what we already
have at that point so we should even bother thinking about it.

- -Andy
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora -


More information about the openmoko-kernel mailing list