[Mesa-dev] [PATCH 1/2] dri/common: embed drirc into driver binaries

[Kenneth Graunke]

On Wednesday, September 02, 2015 04:47:13 PM Ilia Mirkin wrote:
> On Wed, Sep 2, 2015 at 4:43 PM, Kenneth Graunke <kenneth at whitecape.org> wrote:
> > On Wednesday, September 02, 2015 02:26:55 AM Marek Olšák wrote:
> >> From: Marek Olšák <marek.olsak at amd.com>
> >>
> >> People are having issues with apps because drirc wasn't installed
> >> into /etc. I've lost patience.
> >> ---
> >>  src/mesa/drivers/dri/common/Makefile.am      |   4 +-
> >>  src/mesa/drivers/dri/common/Makefile.sources |   3 +-
> >>  src/mesa/drivers/dri/common/drirc            |  84 --------------------
> >>  src/mesa/drivers/dri/common/drirc_built_in.h | 111 +++++++++++++++++++++++++++
> >>  src/mesa/drivers/dri/common/xmlconfig.c      |  30 +++++++-
> >>  5 files changed, 140 insertions(+), 92 deletions(-)
> >>  delete mode 100644 src/mesa/drivers/dri/common/drirc
> >>  create mode 100644 src/mesa/drivers/dri/common/drirc_built_in.h
> >
> > I really dislike this idea, but I understand your frustration.  Distros
> > often seem to ship bunk /etc/drirc files.
> >
> > It seems that "make install" doesn't actually install /etc/drirc today.
> > Fixing that would probably go a long way in helping this problem...
> 
> Making it load from $sysconfdir seems like the right solution to that
> problem. Won't help the crazies using LIBGL_DRIVERS_PATH, but for the
> normal folk who just do "make install" to a custom dir and use
> LD_LIBRARY_PATH to load from it, should work just fine.

sysconfdir is definitely the right place.

> Perhaps we could install it to the dri directory (where foo_dri.so get
> loaded from) and then just load it from the lib's directory? Might be
> too complicated.
> 
>   -ilia
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.freedesktop.org/archives/mesa-dev/attachments/20150902/542106c2/attachment.sig>