On 12.06.2012 14:27, Marc-André Lureau wrote: > On Mon, Jun 4, 2012 at 11:33 AM, Christophe Fergeau <cfergeau@xxxxxxxxxx> wrote: >> Makes sense to me though I've only looked quickly through it. Since it >> works for you, I'm in favour of committing it if noone disagrees. > > I also don't like these changes to end up upstream, for the same > reasons Alex mentioned. It makes me sad that Spice will end up with > debian without Celt. So let's decide. Without spice, which means that qemu/kvm, which is a primary user of spice, does not support spice in debian, and which means there's no spice clients in debian (which is less of a problem). Or without celt. (Mind you, it wasn't me who decided that celt is somehow bad to have in Debian, and I myself does not understand the issues 100%. One of the issues, iirc, is that celt does not compile/work on anything but x86). I already replied to email by Christophe, asking why he thinks raw audio is that bad. The same question pops again. For the upstream, and I already mentioned this, the patch might help to identify all places which should be touched if a need to add a new codec emerges. > They should disable audio completely if it's the > case. I would be more in favour of a patch --disable-audio instead. It was one of alternatives. For the reasons I already mentioned (raw pcm isn't bad), making celt optional has been choosen, as it appears to be less restrictive for the user. Thanks, /mjt _______________________________________________ Spice-devel mailing list Spice-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/spice-devel