Ch6 Glibc Compile Issues
soks86 at gmail.com
Tue Feb 7 18:47:33 PST 2006
Well I was using LiveCD 6.2-Pre2 and it actually does use Perl-5.8.7
along with a patch to remove the preset glibc search path... sooo...
any other ideas? Anyone? Please...
On 2/4/06, paul.rogers at juno.com <paul.rogers at juno.com> wrote:
> Subject: Re: Localedef not working after final Glibc-2.3.6 install
> >If I look through the errors generated by my 'make check' of glibc the
> >first error (and the rest with different binaries) reports that it
> >cannot find the 'mtrace' it compiled to test it out, in fact 'mtrace'
> >does not exist (although I didn't have any make errors that stopped
> >the make process...). Right off the bat that means there is something
> >definately wrong... unless this is normal behaviour? anyone know about
> I had the same problem because the LiveCD I made in mid-December was for LFS-6.1(.0) and provided perl-5.8.6, but I downloaded the LFS book v6.1.1, which had me install Ch5 as though it was 5.8.7. Internal files had directory pointers to 5.8.6, which didn't exist. All I had to do was move the /lib/perl/perl5.8.7 to /lib/perl/perl5.8.6 and installation proceeded. Check that out, maybe you're the third guy with this same problem in the past few weeks.
> >Aside from that... I'm running out of guesses and it's looking like
> >I'm going to have to start over. Although all of my /tools toolchain
> >programs seemed fine... so I can't even guess at what the heck is
> >wrong with this.
> Don't do that. perl-5.8.6 will work for Ch5, just get 5.8.7 for Ch6 install.
> Subject: Re: A way to keep track of packages while installing LFS
> >Yes they are getting slower because basically they search every
> >directories for newer files, which means 'find' have to read every
> >single directories and look for files newer than a specific date, then
> >report these files. The more files/directories the more time.
> Strictly speaking, if you're sure you know where an installation is going to put things then you can have it watch only those directories.
> >Now I have a question (maybe a dumb one ?) Would it be possible to
> >look for the target install (when available) of the makefile before
> >running it ? (Maybe one already did it but I haven't heard of it) That
> >would guarantee a short exec time for the script each time we run it
> >(and we would run it often with a lfs ;-) ).
> A "destdir=" could help. But seriously, I'm running with a P1-233 and I don't find the time that onerous.
> Paul Rogers (paul.rogers at juno.com)
> Rogers' Second Law: "Everything you do communicates."
> (I do not personally endorse any additions after this line. TANSTAAFL :-)
> FAQ: http://www.linuxfromscratch.org/lfs/faq.html
> Unsubscribe: See the above information page
More information about the lfs-support