2013/4/10 Justin Ruggles <[email protected]>:
> On 04/09/2013 06:57 PM, Christophe Gisquet wrote:
>> No vector tests it.
>
> Do any of the ISO reference vectors test it? It seems almost pointless
> to optimize a function that is literally never used.

Mans once mentioned he did test it, but wasn't able to provide the sample.
fate doesn't test it. I optimized it before noticing it wasn't used.

All I know is it should run when "ac->oc[1].m4ac.ext_sample_rate <
sbr->sample_rate"
(and it seems to be halved). But to find what property exactly it is,
and then match it against official vectors:
ftp://ftp.iis.fhg.de/mpeg4audio-conformance/
is a long run.

>> +    mova       m0, [srcq +  0]
>> +    mova       m1, [srcq + 16]
>> +    shufps     m0, m1, q3131
>> +    shufps     m1, m1, q2020
>
> is pshufd with memory arg any faster here?

Well, I couldn't test it. Probably I should move it towards HEAD of my
branch so as to not export it when running format-patch.

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

Reply via email to