Roger Merchberger wrote:Anyway, I'm starting to work on modding the profile (included on the LiveCD) now, should I submit that as a patch, or should I learn SVN & friends...
Well, we could use all the help we could get. Just a few suggestions to get you started.
For Subversion, take a look at http://svnbook.red-bean.com/
I'll start looking into it. My idea of version control never got past "Save As before editing the document." ;-)
That book will help a great deal in understanding the repository and version-control
As far as helping with the livecd, you'd learn a great deal about how it works, and how it is built if you look at our build scripts, which are all done with a series of Makefiles.
Eeek. Makefiles! Having never gotten deep into C, makefiles were always a bit'o'greek to me. The "basics" don't seem difficult from what I've seen so far, but it seems they can get pretty intricate pretty quickly. Ah well, I guess I couldn't avoid it forever... ;-)
=-=-=-=-=
I have a few new questions I need advice and/or help on:
1) Is the included kernel able to support SMP? If not, could I put in a "feature request" for whomever decides what goes in the LiveCD kernel and what doesn't to add that in?
2) I noticed that the "LFS way" to do things in the profiles is to define an environment variable (say: $PARALLEL), issue it a value, then add the parameters in the individual .xml files with something like this:
<make>
<param>-j $PARALLEL</param>
</make>This is all fine and dandy, except: If something goes wonky and $PARALLEL becomes unset (or is never set) for any reason, specifying a -j with a 0 or null value means "infinite." [[ Yes, that seems like it's not exactly "fail-safe" behaviour, but... ;-) ]]
Anyway, it doesn't take long for a computer to run out of all forms of memory whilst compiling gcc or any other good sized package.
The safest way (to me) that I've found is to specify the entire parameter in an xml variable in general.ent, like:
<!ENTITY make_parallel "-j 2">
and use that xml variable in each segment, like:
<make>
<param>&make_parallel;</param>
</make>=-=-=
For "fail-safe" operation, the second works better; for if the variable is not set, deleted, or whatever, nALFS will spew a bunch of errors, but it still seems to proceed, albeit serially instead of insanely. ;-) However, it didn't seem to be the "normal convention" or am I just reading too much into it?
Thanks Roger "Merch" Merchberger
-- Roger "Merch" Merchberger | Anarchy doesn't scale well. -- Me [EMAIL PROTECTED] | SysAdmin, Iceberg Computers
-- http://linuxfromscratch.org/mailman/listinfo/livecd FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
