lkml.org 
[lkml]   [1997]   [Mar]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: glibc-2.0.2

I figured out the problem that I was having with __bb_init_func.
It turns out that I was able to build glibc-2.1 without any problem, so it
was merely a matter of looking to see what was different.

The key seems to be that for some reason the no-whole-archive
flag was incorrectly detected by autoconfig for some reason. It looks
like something went wrong the first time attempting to configure
glibc, and as a result the test for no_whole_archive failed, and this
result was cached in config.cache. Any time after this when I ran
configure, the same value was retrieved and not regenerated.

My guess as to *why* this was misconfigured is that I
*probably* had a bogus option in CFLAGS (i.e. something like
CFLAGS=/tmp/, missing the -B), and as a result the little test didn't
compile properly. A test with a bogus CFLAGS gives me the result with
the wrong setting of no-whole-archive, anyways. I am thinking the
configure script may need to be enhanced to test the default CFLAGS to
make sure they are sane - it probably wouldn't be too swift to force
it to ignore CFLAGS, as there may be a good reason for setting things
in there.

-Eric

\
 
 \ /
  Last update: 2005-03-22 13:39    [W:0.028 / U:0.784 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site