This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH][Ada] Add option to gnatlink to fix gnat.dg on some platforms


On Thu, 7 May 2009, Dave Korn wrote:

> > PS: for reference I've always been opposed to in-tree testing. I wonder
> > how many hundred of lines of dubious stuff would go away if we just
> > tested what we want to test: an installed tree instead of something
> > the user will never see.
> 
>   Speaking for myself, I would always want to have the option to test stuff
> before I install it.

You can install in a different directory using either "make install 
prefix=whatever" or "make install DESTDIR=whatever", and test (taking 
advantage of relocatability) before installing in the final location; 
packaging systems routinely install somewhere other than the final 
location in order to build a package.

(libjava has relocatability bugs - see 
<http://gcc.gnu.org/ml/gcc/2006-12/msg00363.html>.  These can be worked 
around by having the prefix on the "make install" line end with the 
configured prefix (configure --prefix=/some/where, make install 
prefix=/other/some/where), but it would be good for libtool to be fixed to 
work with arbitrary prefix specification at install time, as per the GNU 
Coding Standards.  In addition, some binaries linked with libjava have 
absolute RPATHs hardcoded by default, again breaking relocatability, and 
some scripts installed by libjava don't allow for relocatability either.  
Also, installed testing may not work with libjava at present, even aside 
from these relocatability bugs.)

-- 
Joseph S. Myers
joseph@codesourcery.com


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]