more virtual packages for providing libcuda.so.1 and libOpenCL.so.1

[Graham Inggs]

On 28 June 2013 12:43, Alberto Milone <alberto.milone at canonical.com> wrote:

> So, all it would take is to provide, conflict and replace libcuda5 and
> libopencl1, right?
>

I thought we had decided on libcuda-5.0-1 and libopencl-1.2-1.
However, I did subsequently found a mail in my spam folder from Andreas
(which I cannot find now) saying Debian already has a libopencl1 virtual
package, so maybe provides, conflicts and replaces libcuda-5.0-1 and
libopencl1 is the way to go.


> How did Debian handle it? Did they use libcuda-5.0-1 instead of libcuda5?
>

They have a real package named libcuda1, and a real package named
nvidia-libcopencl1 which provides the libopencl1 virtual package.


>  That would require changes in Mesa and in all the binary drivers. It
> seems a bit excessive to me. Maybe let the package create its own
> symlink (to i386-linux-gnu_GL.conf)  in /etc/ld.so.conf.d so as to work
> around the problem?
>

Thanks, I'll look into that.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-nvidia-devel/attachments/20130628/4cb8c35c/attachment.html>