On Tue, 2010-01-12 at 16:52 +0000, Simon Marlow wrote:
> On 12/01/10 10:52, Duncan Coutts wrote:
> 
> > This would also be relevant for the ghc build process. Currently for ghc
> > we do not use rpath at all for the .so files for the core packages. I
> > did it that way initially just to get it to work. Ideally we should move
> > to setting the rpath for the final $prefix (which should actually be
> > prefix independent using rpath $ORIGIN) and then use LD_LIBRARY_PATH to
> > override things to use inplace in the build tree.
> 
> Just to summarise my understanding.  If we want a $prefix-independent 
> binary distribution that contains dynamically-linked executables and 
> libraries, then we can either
> 
>   - not use rpath for installed libs/executables, and install
>     the .so files in a system-standard location (or add a new
>     subdir of /usr/lib to the standard set at install time).
> 
>   - or use $ORIGIN rpaths

Right. Though it's slightly hard to call it prefix independent if you
have to install the libs in /usr/lib. But yes, it's location independent
if you adjust the ld.so system-wide search path or set LD_LIBRARY_PATH.

> I've seen various pros/cons of these two go past, perhaps someone could 
> summarise?

I'm not sure I can. Apparently I just don't get what other people think
is wrong with rpath.

Duncan

_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc

Reply via email to