[FFmpeg-cvslog] r30306 - in trunk/libswscale: rgb2rgb.c rgb2rgb.h rgb2rgb_template.c swscale.c swscale.h swscale_internal.hswscale_template.cx86/yuv2rgb_template.c yuv2rgb.c

[Ramiro Polla]

On Fri, Jan 15, 2010 at 1:50 PM, Zuxy Meng <zuxy.meng at gmail.com> wrote:
> Hi,
>
> "Carl Eugen Hoyos" <cehoyos at ag.or.at> ????
> news:loom.20100115T114848-56 at post.gmane.org...
>>
>> zuxy <subversion <at> mplayerhq.hu> writes:
>>
>>> Log:
>>> Const correctness for src pointer. Remove all constness related warnings
>>> in
>>> libswscale.
>>
>> Is this supposed to change the output of swscale-example?
>> It's only off-by-one, but I'm asking out of curiosity?
>>
>> Carl Eugen
>>
>> --- swscalealt ?2010-01-15 11:43:03.000000000 +0100
>> +++ swscaleneu ?2010-01-15 11:47:15.000000000 +0100
>> @@ -55282,12 +55282,12 @@
>> ?abgr 96x96 -> bgra ? 96x ?64 flags= 8 SSD= ? ?6, ? ?0, ? ?0, ? ?1
>> ?abgr 96x96 -> bgra ? 96x ?64 flags=16 SSD= ? 12, ? ?0, ? ?0, ? 16
>> ?abgr 96x96 -> bgra ? 96x ?64 flags=32 SSD= ? ?9, ? ?1, ? ?2, ? ?1
>> - abgr 96x96 -> bgra ? 96x ?96 flags= 1 SSD= 1745, 1196, ?718, 4526
>> - abgr 96x96 -> bgra ? 96x ?96 flags= 2 SSD= 1745, 1175, ?715, 4526
>> - abgr 96x96 -> bgra ? 96x ?96 flags= 4 SSD= 1745, 1198, ?725, 4526
>> - abgr 96x96 -> bgra ? 96x ?96 flags= 8 SSD= 1745, 1184, ?719, 4526
>> - abgr 96x96 -> bgra ? 96x ?96 flags=16 SSD= 1745, 1230, ?751, 4526
>> - abgr 96x96 -> bgra ? 96x ?96 flags=32 SSD= 1745, 1197, ?724, 4526
>> + abgr 96x96 -> bgra ? 96x ?96 flags= 1 SSD= 1746, 1196, ?718, 4526
>> + abgr 96x96 -> bgra ? 96x ?96 flags= 2 SSD= 1746, 1175, ?716, 4526
>> + abgr 96x96 -> bgra ? 96x ?96 flags= 4 SSD= 1746, 1198, ?725, 4526
>> + abgr 96x96 -> bgra ? 96x ?96 flags= 8 SSD= 1746, 1185, ?719, 4526
>> + abgr 96x96 -> bgra ? 96x ?96 flags=16 SSD= 1746, 1230, ?751, 4526
>> + abgr 96x96 -> bgra ? 96x ?96 flags=32 SSD= 1746, 1197, ?724, 4526
>> ?abgr 96x96 -> bgra ? 96x 128 flags= 1 SSD= ? ?8, ? ?7, ? ?8, ? ?0
>> ?abgr 96x96 -> bgra ? 96x 128 flags= 2 SSD= ? ?8, ? ?1, ? ?1, ? ?0
>> ?abgr 96x96 -> bgra ? 96x 128 flags= 4 SSD= ? ?5, ? ?0, ? ?0, ? ?0
>
> I'll double check.

abgr<->bgra conversions currently read random data (look at the SSD,
they should be 0, the resulting images are wrong).

Also don't worry if swscale-example doesn't match for the flags=1
tests, the fast bilinear code also reads one byte of random data at
the end of each line.