[Pixman] [PATCH 0/4] Meet the FPU-based implementation of the core pixel pipeline

Dmitri Vorobiev dmitri.vorobiev at movial.com
Tue Sep 21 05:17:44 PDT 2010


Recently, there was a debate about using floating-point hardware
for a Pixman implementation instead of the fixed point code. In
our company, we have developed an FPU-based implementation of the
core pixel pipeline, including support for all pixel formats and
combiners, and I am now working on the resulted code to make the
latter upstreamable. The patch series that follows is the first
outcome of this work.

We are currently working on further improving performance
characteristics of this implementation, especially in the areas
where GCC is not very good in emitting good enough code for our
primary architecture, namely ARM.

Of course, the new implementation is being introduced in the most
non-intrusive way, and is by default disabled. This way, we have
yet another alternative, yet retain all current characteristics
of the library.


More information about the Pixman mailing list