[16965] in Kerberos_V5_Development
Re: RFC: libverto nearing release
daemon@ATHENA.MIT.EDU (Nico Williams)
Tue Jul 5 19:27:08 2011
MIME-Version: 1.0
In-Reply-To: <ldvk4bwmi1n.fsf@cathode-dark-space.mit.edu>
Date: Tue, 5 Jul 2011 18:26:40 -0500
Message-ID: <CAK3OfOj1zn-V-O2F172yjAk9OpKJ=4pN_HtAGYM86UEXjipfzw@mail.gmail.com>
From: Nico Williams <nico@cryptonector.com>
To: Tom Yu <tlyu@mit.edu>
Cc: "krbdev@mit.edu" <krbdev@mit.edu>, Simo Sorce <simo@redhat.com>
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Errors-To: krbdev-bounces@mit.edu
On Tue, Jul 5, 2011 at 6:20 PM, Tom Yu <tlyu@mit.edu> wrote:
> Nico Williams <nico@cryptonector.com> writes:
>> The issue is interposers on the global link map.
>
> Wouldn't you want to pick up the symbols from the interposers anyway,
> for the sake of consistency?
In this case yes (though when I responded I hadn't yet looked at the
implementation, so I wasn't sure).
Now that I've glanced at the implementation I don't see a problem with
what it's doing since in no case does libverto export a symbol name
that could conflict with one of the existing event loop libraries'
symbols.
>> What you want is to use RTLD_GROUP (Solaris) or RTLD_DEEPBIND (Linux)
>> if either is available.
>
> Do you mean something lik dlopen(NULL, RTLD_GROUP)?
I meant dlopen(shared_object_name_or_path, RTLD_LOCAL | RTLD_GROUP).
Nico
--
_______________________________________________
krbdev mailing list krbdev@mit.edu
https://mailman.mit.edu/mailman/listinfo/krbdev
