[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: molehill ----> mountain
From: |
Gary V . Vaughan |
Subject: |
Re: molehill ----> mountain |
Date: |
Thu, 1 Apr 2004 01:05:26 +0100 |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 1 Apr 2004, at 00:38, Bob Friesenhahn wrote:
On Thu, 1 Apr 2004, Gary V.Vaughan wrote:
Now that we don't RTLD_GLOBAL anymore there is no reason for internal
apis
to leak into libltdl clients... :-)
You are missing the distinction between libltdl and the modules it
loads. Global symbols from libltdl will always permeate dependent
applications. :-)
It's late :-()
Pop quiz: libltdl is a client of xmalloc from libltdl/xalloc.c, and
libm4
is a client of gnulib/xmalloc.c. Both define xmalloc(), with different
semantics, but the same prototype. m4 itself links both libraries. m4.h
declares gnulib/xmalloc.c's xmalloc, but ltdl.h doesn't declare anything
from libltdl/xalloc.c. What happens if m4's main() calls xmalloc()?
I guess you're saying that execution could end up in either :-( This is
exactly what weak symbols are for isn't it? I imagine they are not
portable
enough to be used in libltdl though. Ack.
Cheers,
Gary.
- --
Gary V. Vaughan ())_. address@hidden,gnu.org}
Research Scientist ( '/ http://tkd.kicks-ass.net
GNU Hacker / )= http://www.gnu.org/software/libtool
Technical Author `(_~)_ http://sources.redhat.com/autobook
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (Darwin)
iD8DBQFAa1zKFRMICSmD1gYRAvu7AJ9IQvHGkw7vRS6ZRyoy09GkJcGVdgCdHVYZ
b3YVePITULYm7MoPYoQkNw4=
=xwHo
-----END PGP SIGNATURE-----
Re: 110-gary-rationalise-memory-management.patch, Bob Friesenhahn, 2004/03/31