Hi,

On Fri, Apr 29, 2011 at 9:05 AM, Diego Biurrun <[email protected]> wrote:
> On Fri, Apr 29, 2011 at 02:37:35PM +0000, Ronald S. Bultje wrote:
>>
>> On Fri, Apr 29, 2011 at 2:08 PM, Diego Biurrun <[email protected]> wrote:
>> > int/unsigned is the natural memory access type for CPUs, using sized types
>> > for temporary variables, counters and similar just increases code size and
>> > can possibly cause a slowdown.
>> > ---
>> >  libavcodec/alpha/simple_idct_alpha.c |    6 +-
>> >  libavcodec/jfdctfst.c                |   18 ++--
>> >  libavcodec/jfdctint.c                |   18 ++--
>> >  libavcodec/vorbis.c                  |    9 +-
>> >  libavcodec/vorbis.h                  |    2 +-
>> >  libavcodec/vorbisdec.c               |  227 
>> > ++++++++++++++++------------------
>> >  6 files changed, 134 insertions(+), 146 deletions(-)
>>
>> Don't touch the alpha code unless you have access to one.
>
> Mans reviewed that stuff and suggested the change, so I trust it.
> FATE will complain if it fails to compile (I don't see how) and
> then I can revert that part.  It's how I did the last changes to
> Alpha code and for this platform I think it is a valid approach.
>
>> Didn't you just commit such changes to a vorbis decoder already?
>> I'm confused about them re-appearing here.
>
> No, this is a refreshed set of patches that are not pushed yet.
>
>> Regardless, speed tests are
>> missing to confirm no significant slowdown.
>
> On the contrary there is something like ~5% speedup:
>
> time ./ffmpeg -i ~/tmp/FaiF_0x0D_NDAs.ogg -f null -v 0 -

And for the JFDCT?

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

Reply via email to