On Mon, 2007-06-11 at 11:47 -0700, Keith Packard wrote: > On Mon, 2007-06-11 at 11:14 -0700, Ian Romanick wrote: > > > The problem is that a few of the GLX extensions (e.g., > > GLX_SGI_video_sync and GLX_OML_sync_control) allow applications to query > > the vblank counter directly. I don't know of other hardware that > > maintains an actual counter. I know that MGA doesn't, and I'm pretty > > sure that Via doesn't either. > > Radeon and Intel do keep an internal counter. I posted an outline for > keeping the counter reasonably smoothly monotonic as seen by > applications which should make it work across mode switches too. > > There is a huge power savings advantage to using the chip-resident > counters as we don't have to wake up the CPU every frame to bump a > counter. For chips which do not provide a counter, the driver is welcome > to synthesize one in software. I was thinking that this could even be > done by estimating based on wall clock time, which should be sufficient > in most cases by enabling interrupts periodically to resync to the frame > counter edge. Going from 16ms wakeups to 2000ms wakeups is a huge > potential power savings.
Also, for the time being they can just leave the vblank interrupt(s) enabled and keep more or less the same code they have now. -- Earthling Michel Dänzer | http://tungstengraphics.com Libre software enthusiast | Debian, X and DRI developer ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ -- _______________________________________________ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
