Date: Wed, 19 Feb 1997 19:57:26 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: thomas@ghpc8.ihf.rwth-aachen.de (Thomas Gellekum) Cc: freebsd-bugs@freebsd.org Subject: Re: RELENG_2_2 fails to build Message-ID: <Mutt.19970219195726.j@uriah.heep.sax.de> In-Reply-To: <199702190730.IAA26332@ghpc6.ihf.rwth-aachen.de>; from Thomas Gellekum on Feb 19, 1997 08:30:38 %2B0100 References: <Mutt.19970219011804.j@uriah.heep.sax.de> <199702190730.IAA26332@ghpc6.ihf.rwth-aachen.de>
next in thread | previous in thread | raw e-mail | index | archive | help
As Thomas Gellekum wrote: > > Hmm, maybe not. texinfo is already needed very early in the build > > process. Anyway, merging the fix from current wouldn't hurt either. > > Didn't my mail get through? I got the same symptoms on my machine > where the makeinfo from teTeX (newer version than in 2.2) comes before > /usr/bin/makeinfo in my $PATH. Hmm Hmm. No, i don't think that's my problem. That's since `make release' bootstraps its own chrooted area (very obviously without any traces from /usr/local), and recompiles itself there. Bootstrapping is done by running `make install' with a DESTDIR set to the chroot area. So, when compiling a RELENG_2_2 release inside a 3.0-current machine, the binaries there are 3.0-current at first, and those are used to recompile the 2.2 tree. Hence, it's ``effectively downgrading''. gnu/lib/libregex's texinfo file isn't compilable with the 3.0-current `makeinfo' command without a special hack. The special hack is in HEAD but not yet in RELENG_2_2. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Mutt.19970219195726.j>