building xorg

[Dan Nicholson]

On Tue, Jun 17, 2008 at 11:15 AM, Chuck Robey <chuckr at telenix.org> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Dan Nicholson wrote:
>>
>> gl.pc comes from mesa, and you really need the mesa version to build
>> GLX and DRI in the server. It needs headers in $includedir/GL like
>> GL/internal/dri_interface.h. Nvidia's gl doesn't provide the DRI
>> interface.
>
> Yes but (famous last words??) don't I need to use the nvidia version, no matter
> how it might be warped, because it's the only one that'll work for my card?
>
> I would be overjoyed to find myself wrong.

At runtime, you'd want Nvidia's libGL to run any GL applications. But
at build time, you need mesa's GL/DRI headers for the server. So, just
make sure mesa's gl.pc is in PKG_CONFIG_PATH at build time and then
make sure the dynamic linker picks up Nvidia's libGL.so first at
runtime (LD_LIBRARY_PATH or /etc/ld.so.conf or just move mesa's libGL
out of the way). At least, that's the idea. I've never actually played
the Nvidia libGL game.

--
Dan