[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]


I've getting real close to working PReP boot images now.  It seems as
though a lot of things are broken right now in this package.  The biggest
issue right now is that dbootstrap can't identify my architecture.  In
fact, parsefile() doesn't even seem to properly parse /proc/cpuinfo.  How
much has this been tested...I can't believe I'm the first one to use this?!
Anyway, the routine tokenizes all non-white space but it will never match
the "machine:" since it only grabs "machine".  If I hardcode in "PReP" I'm
able to get a debinstaller image that boots and starts the install
process so that's a start.  What's the history on this code?

Next, on my serial console machine I can't use dumpkeys since it attempts 
to open /dev/tty, /dev/tty0, and /dev/console in the order looking for
what it considers to be a "real" console device.  For now, I built
keymaps.tgz on another box but I think the key translation code from
dumpkeys/loadkeys should be liberated into a boot floppies utility so that
any box without a keyboard can build boot-floppies as well.

Finally, I had to remove the language translation code to get up and
running. The scheme to move the translated files to tmp*.c results in
loads of redefinitions when both the originals and translated files are
all linked together.

Any suggestions are welcome,
Matt Porter
This is Linux Country. On a quiet night, you can hear Windows reboot.

Reply to: