[Pixman] Resurrecting: Added fast path for "pad" type repeats

[Pekka Paalanen]

On Mon, 7 Apr 2014 12:22:15 +0300
Pekka Paalanen <ppaalanen at gmail.com> wrote:

> Hi,
> 
> this thread started from
> http://lists.freedesktop.org/archives/pixman/2013-February/002619.html
> and continued in
> http://lists.freedesktop.org/archives/pixman/2013-March/002677.html
> 
> I'd like to hear what the thoughts of it are now, more questions below.
> 
> 
> On Wed, 13 Feb 2013 08:37:06 +0100
> sandmann at cs.au.dk (Søren Sandmann) wrote:
> 
> > I pushed the first patch to master. For this one, did you try comparing
> > Chris' patch to your on ARMv6? Also, did we ever find out whether it was
> > a bug in Firefox. I'm still somewhat skeptical that it's intended for a
> > PAD image to be accessed so far out of bounds.
> 
> What is the "Chris' patch", has it been merged?
> 
> Do you still think there is a bug in Firefox, and the (trimmed) Cairo
> traces are therefore invalid for PAD type repeats?

...

> In his original email, Ben showed a speedup of "3.86x" in
> t-firefox-chalkboard trimmed Cairo trace on ARMv6.
> 
> However, it didn't come clear to me, whether this "pad" kind of
> optimization would ever be accepted upstream, or what exactly an
> acceptable approach would be. Could you shed some light on that?
> 
> Apparently it should go on the "new infrastructure" way instead?
> 
> I'm trying to gather some understanding on how much work it would be to
> get the "3.86x" improvement upstream, and whether I as a Pixman-novice
> would be able to pull it off in reasonable time.

No idea, anyone?

Also, what about the "[PATCH v5 0/4] Two ARMv6 optimizations"
patch series I posted on April 9th?


Thanks,
pq