[kde-solaris] Re: question on compiling

[Stefan Teleman]

I actually seriously considered writing something like this myself
back
at the end of October (i found myself in a situation where i had
finished
porting the 80+ required libraries, only to find out that the
directory
tree i was using at the time could no longer be used, so i had to redo
everything again from scratch). Grrrr. Nuff said.

But then i also thought, if i go down this path, what's next. This can
only work if strlen (target RPATH) <= strlen (source RPATH), otherwise
no dice.

I think this would be interesting to do as a side project, for fun,
but
i would not recommend this as a practical solution.

Just my 2c. :-)

--Stefan

----- 

Aaron Williams wrote:
> 
> Look for a tool called setrpath.  This works as long as the new rpath is
> of shorter or equal length.  A Copy of it can be found at
> http://www.martnet.com/~johnny/exploits/linux/change-LD_RUN_PATH
> 
> -Aaron
> 
> Stefan Teleman wrote:
> 
> >This is very useful info. I will try it for 3.2
> >(too late for 3.1.4:-).
> >
> >--Stefan
> >
> >-----
> >
> >Torsten Kasch wrote:
> >
> >
> >>Hi,
> >>
> >>On Monday 08 December 2003 20:44, Stefan Teleman wrote:
> >>[...]
> >>
> >>
> >>>Unfortunately i know of no way of resetting the RPATH in a shared lib
> >>>without
> >>>relinking it.
> >>>
> >>>
> >>There is a solution for creating relocatable shared objects on Solaris. It's
> >>described in the section "Introducing $ORIGIN" in
> >>
> >>        http://docs.sun.com/db/doc/806-0641/6j9vuqukj?a=view
> >>
> >>I had plans to try to get this working with libtool myself but unfortunately I
> >>didn't find some spare time. Maybe someone else wants to give it a try?
> >>
> >>cheers,
> >>        Torsten
> >>
> >>--
> >>   Torsten Kasch                                  tk at CeBiTec.Uni-Bielefeld.DE
> >>      Center for Biotechnology (CeBiTec) / BRF    Office: V6-151
> >>      Bielefeld University                        Voice:  +49 521 106-4828
> >>      D-33594 Bielefeld                           Fax:    +49 521 106-5626
> >>
> >>
> >
> >
> >

-- 
NOTICE: If received in error, please destroy and notify sender. 
Sender does not waive confidentiality or privilege, and use is
prohibited.