On 1/6/06, Ken Moffat <[EMAIL PROTECTED]> wrote: > Dan, by adding bison to /tools [ in what is otherwise svn-20060103 with > a bash-documentation cleanup per bug 1679 ] I get a clean build of bison > in chapter 6. That is, it no longer differs when rebuilt in place. > > This means bison depends on itself to build cleanly, but NOT on flex, > IMHO.
Interesting. I'll try that next time. > >>> (v) perl - I'll try creating /etc/hosts before the configure, but I > >>> think this is probably cosmetic. > > Still looks cosmetic. Creating /etc/hosts before configuring perl > means Config_heavy.pl no longer changes when perl is rebuilt in place, > but I can't detect any changes in any other part of perl, either against > rebuilding in place, or against the original build without /etc/hosts > present at configure time. I guess I don't understand how Config_heavy.pl is used. The way that I interpret Bryan's email earlier is that with the perl installed, any perl script that references hostcat will get an empty command. I'm clueless with perl, though. > The prime remaining issues are the toolchain impurity, and (perhaps) > e2fsprogs (I'm hoping that fixing the toolchain might also fix e2fsck). I'm about to add a couple DIYisms into my scripts and see what happens. > We've now got the current utf8 changes in svn, so I'm minded to build > my next cycle against svn plus bison in chapter 5, then compare that to > some sort of toolchain cleanup. Yeah, I'm molding the utf8 changes into my scripts. I'll probably kick one off in the next few days, but I want to see what happens with the new toolchain adjustments first. -- Dan -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
