Diego Biurrun <[email protected]> writes:

> On Fri, Dec 16, 2011 at 08:44:48PM +0000, Måns Rullgård wrote:
>> Diego Biurrun <[email protected]> writes:
>> 
>> > On Fri, Dec 16, 2011 at 08:22:20PM +0000, Mans Rullgard wrote:
>> >> The A32 bitstream reader variant is only used on ARMv5 and for
>> >> Prores due to the larger bit cache this decoder requires.
>> >> 
>> >> In benchmarks on ARMv5 (Marvell Sheeva) with gcc 4.6, the only
>> >> statistically significant difference between ALT and A32 is
>> >> a 4% advantage for ALT in FLAC decoding.  There is thus no (longer)
>> >> any reason to keep the A32 reader from this point of view.
>> >> 
>> >> This patch adds an option to the ALT reader increasing the bit
>> >> cache to 32 bits as required by the Prores decoder.  Benchmarking
>> >> shows no significant change in speed on Intel i7.  Again, the
>> >> A32 reader fails to justify its existence.
>> >> 
>> >> Signed-off-by: Mans Rullgard <[email protected]>
>> >> 
>> >> --- a/libavcodec/get_bits.h
>> >> +++ b/libavcodec/get_bits.h
>> >> @@ -122,8 +102,11 @@ LAST_SKIP_BITS(name, gb, num)
>> >>  
>> >> -#ifdef ALT_BITSTREAM_READER
>> >> +#ifdef LONG_BITSTREAM_READER
>> >> +#   define MIN_CACHE_BITS 32
>> >> +#else
>> >>  #   define MIN_CACHE_BITS 25
>> >> +#endif
>> >
>> > Some decoders have #define ALT_BITSTREAM_READER before header #includes,
>> > I assume these should be deleted now.
>> 
>> Are there any defining plain ALT_BITSTREAM_READER?  I know of a bunch
>> defining ALT_BITSTREAM_READER_LE, and those should of course remain
>> (although we'll rename the macro).
>
> ~/src/libav$ git grep '#define ALT_BITSTREAM_READER$'
> libavcodec/dv.c:#define ALT_BITSTREAM_READER
> libavcodec/imc.c:#define ALT_BITSTREAM_READER

OK, those can go away then.

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

Reply via email to