Vitor Sessak <[email protected]> writes:

> On 04/01/2011 09:20 PM, Luca Barbato wrote:
>> On 04/01/2011 07:12 PM, Vitor Sessak wrote:
>>> Hi,
>>>
>>> The following patches add an AVX (an intel x86 extension) FFT
>>> implementation. Since I do not have a Sandybridge myself, I have no idea
>>> of its performance. Benchmarks (for ex., using fft-test -s) are thus
>>> very welcome. Also welcome are suggestions for optimizing it further, in
>>> particular the 8 point FFT (in the T8_AVX macro), which is not much
>>> faster than the SSE version.
>>
>>> ---
>>>   libavcodec/x86/fft.c       |    9 +-
>>>   libavcodec/x86/fft.h       |    1 +
>>>   libavcodec/x86/fft_mmx.asm |  320 
>>> ++++++++++++++++++++++++++++++++++++++++++--
>>>   libavcodec/x86/fft_sse.c   |   12 ++
>>>   libavcodec/x86/x86inc.asm  |   17 +++
>>>   libavutil/mem.c            |   16 +--
>>
>> Might be nicer have an fft_avx.asm file, beside that I don't have an avx
>> system.
>
> It uses several macros from fft_mmx.asm...

Those could be moved to a shared file.

-- 
Måns Rullgård
[email protected]
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to