[libvirt] version script and non-existent symbols

[Daniel P. Berrange]

On Mon, Dec 15, 2008 at 09:51:16PM +0000, John Levon wrote:
> On Mon, Dec 15, 2008 at 09:32:03PM +0000, Daniel P. Berrange wrote:
> 
> > > It's actually not clear to me why some of these are even in the version
> > > script. For example, why is brSetEnableSTP() there?
> > 
> > It is used from the network_driver.c file, and that is linked to the
> > daemon
> 
> Hmm I'd missed that. A weird source code setup?

Yeah, its a little wierd, but then moving source files around in CVS
is not too pleasant either.  Ultimately the dlopen() option makes is
neccessary regardless
 
> > reason. Then, if we build drivers as dlopen()'able modules, even more
> 
> Oh, I forgot about that. OK. What is the purpose of the dlopen()
> version? Are people to be expected to deliver modules separately? Is it
> just a minimisation thing?

No, it doesn't allow for 3rd party modules, because we fix the list of
modules we probe for, and also guarentee ABI incompatability in the
internal module interface in every release.

It was basically to allow smaller footprint deployments, and allow 
sysadmins to guarentee that undesired drivers will never be activated
on a host even if the distro maintainer wants to compile all of them 
by default. For example, if you've compiled it in, the UserModeLinux 
driver will always activate itself, but many people won't want it around.
That said, we're not actively using dlopen() currently...

Daniel
-- 
|: Red Hat, Engineering, London   -o-   http://people.redhat.com/berrange/ :|
|: http://libvirt.org  -o-  http://virt-manager.org  -o-  http://ovirt.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: GnuPG: 7D3B9505  -o-  F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|