Re: lt_dlopen and default search paths

[Bob Friesenhahn]

On Tue, 4 Feb 2003, Blake Matheny wrote:

> Thanks for the reply. Yes, I realize that lt_dlopen() has its own search
> algorithm. Regardless, it is still a _wrapper_ for the appropriate function on
> your platform. On my platform, dlopen() is that appropriate target function
> and as far as I know, you can't force dlopen() to _not_ search in its default
> locations. So what is going on so that dlopen() is not searching where it
> should?

You can force dlopen() to not search the default locations by
providing an explicit fully qualified path to the library to load.
That is what libltdl is doing.  If libltdl did things differently then
it wouldn't achieve its objective of working similarly for all
platforms (even Windows!).

Bob

> Whatchu talkin' 'bout, Willis?
> > I believe that lt_dlopen() actually uses its own (documented) search
> > algorithm.  Please refer to the documentation for lt_dlopen() where
> > the algorithm is detailed.
> >
> > Bob
> >
> > On Tue, 4 Feb 2003, Blake Matheny wrote:
> >
> > > Hello All,
> > > I hope I'm addressing the right crowd. I'm doing some development 
> > > utilizing
> > > libltdl, and my question is regarding the lt_dlopen() function.
> > >
> > > Depending on your platform, lt_dlopen() should just make a call to 
> > > dlopen()
> > > (as is the case on my GNU/Linux system). On my GNU/Linux system, the 
> > > dlopen()
> > > function is documented as searching:
> > >  The list of libraries cached in /etc/ld.so.cache
> > >  /usr/lib, followed by /lib
> > >  LD_LIBRARY_PATH
> > >
> > > Using lt_dlopen() and after installing my libraries in /usr/lib, lt_dlopen
> > > fails to find them which indicates to me that dlopen() is either not being
> > > called or is no longer searching those paths. Can someone tell me what 
> > > might
> > > be going on? Thanks.
> > >
> > > -Blake
> > >
> > > I am not on the list, please CC me.
> > >
> > > --
> > > Blake Matheny           "... one of the main causes of the fall of the
> > > address@hidden      Roman Empire was that, lacking zero, they had
> > > http://www.dbaseiv.net   no way to indicate successful termination of
> > > http://ovmj.org/GNUnet/  their C programs." --Robert Firth
> > >
> > >
> > > _______________________________________________
> > > Libtool mailing list
> > > address@hidden
> > > http://mail.gnu.org/mailman/listinfo/libtool
> > >
> >
> > ======================================
> > Bob Friesenhahn
> > address@hidden
> > http://www.simplesystems.org/users/bfriesen
> >
>
> --
> Blake Matheny           "... one of the main causes of the fall of the
> address@hidden      Roman Empire was that, lacking zero, they had
> http://www.dbaseiv.net   no way to indicate successful termination of
> http://ovmj.org/GNUnet/  their C programs." --Robert Firth
>

======================================
Bob Friesenhahn
address@hidden
http://www.simplesystems.org/users/bfriesen