[FFmpeg-devel] ARM NEON optimisations

[Måns Rullgård]

"Ramiro Polla" <ramiro.polla at gmail.com> writes:

> On Mon, Aug 25, 2008 at 12:06 AM, Mans Rullgard <mans at mansr.com> wrote:
>>
>> NEON is the name of the SIMD unit present in ARMv7 processors such as
>> the Cortex-A8.  Its register file is viewed as 16 128-bit registers or
>> 32 64-bit registers, each a vector of 8/16/32-bit integers or 32-bit
>> floats.
>>
>> This patch series adds NEON optimisations for a the most heavily used
>> FFmpeg dsputil functions.
>
> Nice. I'll tell the guys who have my Beagle Board to test when it hits
> SVN.

They can play with my git tree right now.

> Do you have any kind of benchmarks?

Does "off the scale" count as a benchmark?  I haven't done accurate
benchmarks of everything.  A quick test with an H.264 file shows
roughly 66% speedup with these patches.  That should give you some
idea.

-- 
M?ns Rullg?rd
mans at mansr.com