On Thu, May 04, 2006 at 03:43:29PM +0200, Michel D?nzer wrote: > On Thu, 2006-04-05 at 13:20 +0200, David Mart??nez Moreno wrote: > > El jueves, 4 de mayo de 2006 08:34, Michel D??nzer escribi??: > > [...] > > > This usually happens because there's not enough video RAM available to > > > store the XVideo data. If increasing the amount of RAM reserved for > > > graphics via the VideoRam directive doesn't help, please post full X > > > config and log files. > > > > Which makes me remember that, IIRC, we have lost reportbug scripts in > > the > > transition. I was thinking of: > > > > a) make an external resource 'a la' xsfbs.mk in every tree and ship it with > > every driver. > > b) add it to a common package (say xserver-xorg) and link (symbolically) in > > every driver from /usr/share/bug/blabla/ to our master reporbug script. > > I suggested something like b) before, but it would have to be > xserver-xorg-core, as xserver-xorg is not mandatory (and indeed not > installed here). > > Also, I think that when reporting a bug against xserver-xorg(-core), the > user should be warned to also check for duplicates in other packages, in > particular those of the drivers she's using. This might also apply the > other way around.
The second option sounds infinitely better because then I don't have to update every single package when we change the script. This works out ok for xsfbs more or less (but not great) since this is purely build-related and fairly mature code anyway. What we need to do is figure out exactly what we need out of the scripts, possibly having separate ones for the libs, drivers, and server. We can deploy them in x11-common, since that way we know they're available to us no matter what. I'm still trying to wade through all the RC bugs I've created, and also get us as ready for 7.1 as I can, so I don't have the necessary time to do this. If someone wants to move/update/write the scripts though, I'd *love* to have them ready to deploy. - David Nusinow -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
