[FFmpeg-devel] [PATCH] Move MLP's dot product to DSPContext

[Ramiro Polla]

On Wed, Apr 22, 2009 at 4:01 AM, Jason Garrett-Glaser
<darkshikari at gmail.com> wrote:
> On Mon, Apr 20, 2009 at 8:32 PM, Ramiro Polla <ramiro.polla at gmail.com> wrote:
>> On Tue, Apr 21, 2009 at 12:29 AM, Jason Garrett-Glaser
>> <darkshikari at gmail.com> wrote:
>>> 2009/4/20 Ramiro Polla <ramiro.polla at gmail.com>:
[...]
>>>> - on x86_64 (can't run sse4)
>>>> current: ?2070ms
>>>> array of functions in dspcontext:
>>>> c ? ? ?: ?2600ms (badly vectorized)
>>>> c ? ? ?: ?1920ms (not vectorized)
>>>> sse2 ? : ?2450ms
>>>> inlined in mlpdec.c:
>>>> c ? ? ?: ?2800ms (badly vectorized)
>>>> c ? ? ?: ?1980ms (not vectorized)
>>>> sse2 ? : ?2450ms
>>>
>>> Have you tried benching it on a 64-bit system with SSE4?
>>
>> No. I don't have access to any.
>
> I have a strong suspicion that C code on 64-bit will outperform your
> SSE4 loop because of the ability to fit the results in single 64-bit
> registers.

Looking at the sse2 slowdown and seeing that sse4 won't give that much
speedup over it I'd say so too. This should all be under #if !
HAVE_FAST_64BIT

Ramiro Polla