Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 13 May 2003 08:43:50 -0400
From:      Garance A Drosihn <drosih@rpi.edu>
To:        "M. Warner Losh" <imp@bsdimp.com>, freebsd-arch@freebsd.org
Subject:   Re: `Hiding' libc symbols
Message-ID:  <p05210613bae697537a59@[128.113.24.47]>
In-Reply-To: <20030513.002249.56047366.imp@bsdimp.com>
References:  <20030509154304.GC61844@dragon.nuxi.com> <20030511.103412.15266142.imp@bsdimp.com> <20030511232857.GB66670@dragon.nuxi.com> <20030513.002249.56047366.imp@bsdimp.com>

next in thread | previous in thread | raw e-mail | index | archive | help
At 12:22 AM -0600 5/13/03, M. Warner Losh wrote:
>I have been reading the thread.  Near as I can tell, only you
>and ache were opposing things, but I must admit that I didn't
>count the people that I didn't think understood the issues in
>this list.

I tried to be polite by only adding one response to this thread,
but I still don't see why every symbol needs to be hid.  I can
see having an internal version of _routineB if there is a
routineA which depends on the *internal* implementation of
routineB.  Otherwise, if a package provides a broken routineB,
that is the packages problem.

People have said "but we already hide some symbols!" as if that
is proof that we need to hide all symbols.  Well, we also do
*not* hide a lot of symbols, and it wasn't until this strlcpy
example popped up that those not-hidden symbols were a big issue.
This is not a problem which has been biting us over and over
again, month after month, routine after routine.

If we were to vote, I'd vote to stick with the status quo, but
it is true that I do not feel too strongly about the issue.

-- 
Garance Alistair Drosehn            =   gad@gilead.netel.rpi.edu
Senior Systems Programmer           or  gad@freebsd.org
Rensselaer Polytechnic Institute    or  drosih@rpi.edu



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?p05210613bae697537a59>