Update to current packages or PLFS first?

Jeremy Herbison herbie at autobotcity.net
Tue Apr 29 23:37:05 PDT 2003


> As far as I can see we have two options to get development started
again:
> 
> 1) Begin with updating to all the latest packages, or
> 2) Integrating the pure-lfs.txt hint, or
> 
> pure-lfs is based on newer packages for some of the instructions, so
if we
> went the pure-lfs route, it would mean we update only partly to the
latest
> packages (at first) followed by the rest later.
> 
> But if we go the update packages route first, we get into some issues,
> like
> glibc-2.3.2. It'll have a temporary cludge until pure-lfs is applied
after
> which the cludge would be removed (touching a missing header file the
> issue
> was if I recall correctly).
> 
> On a totally different note, Greg let me know he'll be away for a week
or
> so
> and I thought it would be only fair if he were here when the
change-over
> happened. Maybe he doesn't care for that, but it only seems right
since he
> is
> a large part of the pure-lfs effort (yes, you are too Ryan ;)
> 
> The week and a halve that Greg's going to be away would give us all
the
> time
> we need to update to all the latest packages and test the new patches
that
> come with some of the packages. Then there'll be less to test with the
> move
> to pure-lfs itself which is a large undertaking all by itself without
> having
> to update and test new packages alongside of it.
> 
> So that's my preference. Packages first, pure-lfs immediately
afterwards.
> If
> anybody disagrees, speak up (or forever hold your peace).
> 

Like Ryan said, new versions aren't a requirement of the pLFS method. We
could just go to the new method now (with the current packages), and
update packages later. This will ensure lots of testing on the pLFS
method across a number of package versions, and also help to avoid any
headaches when upgrading stuff like glibc. Lets fix the build method
BEFORE we go changing packages!

Jeremy

-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list