[cairo] Speedup _cairo_matrix_to_pixman_matrix

[Xan Lopez]

So...
I've not yet setup cairo-perf on the hardware, but using torturer and jorn's
timetext the results look like this:

Cairo 1.3.2: Drawn label 2708 times. Average time spent drawing (in
seconds): 0.002021
Cairo 1.3.2 + 3 patches: Drawn label 2716 times. Average time spent drawing
(in seconds): 0.002015
Cairo 1.3.2 + matrix indentity patch: Drawn label 2696 times. Average time
spent drawing (in seconds): 0.002135
Cairo 1.3.2 + cairo_lround patch: Drawn label 2703 times. Average time spent
drawing (in seconds): 0.002085
Cairo 1.3.2 + glyph gstate patch: Drawn label 2693 times. Average time spent
drawing (in seconds): 0.002097

As you can see, the results are so similar that the numbers are not very
meaningfull and any possible win or lose is lost in the noise (I've repeated
all the tests several times, without changes). gtk-theme-torturer seems to
confirm this, as the results are almost completely the same.

I don't think I'll manage to get the cairo-perf numbers today, so if someone
can provide them before tomorrow that would be great, I guess they'll be
much more meaningfull.

Cheers, Xan

On 11/21/06, Xan Lopez <xan.lopez at gmail.com> wrote:
>
>
> On 11/21/06, Daniel Amelang <daniel.amelang at gmail.com> wrote:
> >
> > On 11/21/06, Xan Lopez <xan.lopez at gmail.com> wrote:
> > > Hi Daniel,
> > > In no particular order:
> > > 1) You rock
> > > 2) If no one beats me to it I'll send some numbers to the list
> > tomorrow with
> > > the patches tested on ARM hardware
> > > 3) If you don't get a 770 for Christmas there is no justice in this
> > world :)
> >
> > Thanks, man. Hey, along with adding that torturer-textview patch I
> > sent to this list last week, could you also look into running the
> > timetext.c benchmark that Jorn has here:
> >
> > http://folks.o-hand.com/jorn/pango-benchmarks/
> >
> > and maybe even the built-in cairo-perf test in git? Sorry to nag, it's
> > just that I've been optimizing for all three benchmarks, and it would
> > be really useful to see the effect on all three. The torturer is
> > actually the least helpful for me, as there is a lot of noise from
> > pango/gtk etc, and I'm concentrating on cairo for the moment.
>
>
>
> cairo-perf was on my plans already, but I'll add jorn's benchmarks too.
>
>
> Thanks again,
> >
> > Dan Amelang
> >
>
>
> Cheers, Xan
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freedesktop.org/archives/cairo/attachments/20061122/73e5eda5/attachment.htm