[FFmpeg-devel] [PATCH] SSE2 version of vf_idet's filter_line()

[James Almer]

On 03/09/14 3:38 AM, Pascal Massimino wrote:
> Hi,
> 
> 
> On Tue, Sep 2, 2014 at 10:26 PM, Reimar Döffinger <Reimar.Doeffinger at gmx.de>
> wrote:
> 
>> On 03.09.2014, at 00:49, Pascal Massimino <pascal.massimino at gmail.com>
>> wrote:
>>> On Tue, Sep 2, 2014 at 9:39 AM, Michael Niedermayer <michaelni at gmx.at>
>>> wrote:
>>>
>>>
>>> [ahem: ffmpeg doesn't feel like using intrinsics, by chance?]
>>
>> I tried that about 5 months back, once more.
>> It still results in code that is slower than the plain C version, even
>> when using SIMD, on trivial NEON audio format conversion (same thing in asm
>> was about 8x faster).
>> So you can get the same effect with less effort by disabling just
>> disabling asm code.
>>
> 
> strange. I exclusively used intrinsics for libwebp (x86, but also
> neon/aarch64) and was pretty
> pleased with the result (say <2% perf loss, but 10x easier maintenance and
> friendliness to non-guru contributors).
> Agreed, coding style is weird, with all these 'const __m128i var = ...',
> but...
> My 2c.
> 
> /skal

Adding to Michael's reasons, there's also the fact you need to compile the files 
with instruction set specific flags, which would be a PITA to handle from the 
build system.
GCC 4.8 and above added some features to work around this, but we support other 
compilers as well as older GCC versions.