very slow performance of glxgears (68 fps)

[Bryce Harrington]

On Sun, Feb 01, 2009 at 11:55:56AM -0800, Ian Romanick wrote:
> On Fri, 2009-01-30 at 18:11 -0800, Bryce Harrington wrote:
> > On Fri, Jan 30, 2009 at 01:29:49PM -0800, Eric Anholt wrote:
> > > > $ glxgears
> > > > Failed to initialize TTM buffer manager.  Falling back to classic.
> > > > 300 frames in 5.0 seconds = 59.884 FPS
> > > > 299 frames in 5.0 seconds = 59.621 FPS
> > > > 300 frames in 5.0 seconds = 59.818 FPS
> > > 
> > > glxgears is not a benchmark.
> > > 
> > > We sync to vblank because running glxgears at 1000fps is dumb.
> > 
> > I am going to go out on a limb and guess we're going to see a crapload
> > of reports of "performance regression" due to reduced glxgears frame
> > rates.
> > 
> > Mayhaps this could be headed off by changing glxgears output a bit?
> > Something like:
> > 
> >  $ glxgears
> >  Failed to initialize TTM buffer manager.  Falling back to classic.
> >  300 frames in 5.0 seconds = 100.0% synchronized
> >  299 frames in 5.0 seconds = 99.93% synchronized
> >  300 frames in 5.0 seconds = 100.0% synchronized
> 
> I just committed something similar to the Mesa repository.  If it
> detects that vsync is happening, it logs a message to that effect.  The
> detection is not 100% reliable, but it's as good as I can get it.
> 
> [idr at localhost xdemos]$ glxgears
> Mesa: Initializing x86-64 optimizations
> Mesa: Mesa 7.5-devel DEBUG build Jan 30 2009 16:45:54
> Mesa warning: couldn't open libtxc_dxtn.so, software DXTn
> compression/decompression unavailable
> Running synchronized to the vertical refresh.  The framerate should be
> approximately the same as the montior refresh rate.
> 295 frames in 5.0 seconds = 58.948 FPS
> 300 frames in 5.0 seconds = 59.914 FPS

Hi Ian,

Cool, that should help, although won't people ignore the introductory
text and fixate on the fps number?

Bryce