[FFmpeg-devel] [PATCH v28] lavf/movenc: Add palette to video sample description

[Mats Peterson]

On 02/26/2016 11:31 PM, Michael Niedermayer wrote:
> On Fri, Feb 26, 2016 at 11:21:19PM +0100, Mats Peterson wrote:
>> Michael Niedermayer <michael at niedermayer.cc> skrev: (26 februari 2016 22:53:00 CET)
>>> On Fri, Feb 26, 2016 at 06:24:17AM +0100, Mats Peterson wrote:
>>>> On 02/26/2016 05:26 AM, Mats Peterson wrote:
>>>>> On 02/26/2016 05:08 AM, Mats Peterson wrote:
>>>>>> Should hopefully fix the big-endian issue.
>>>>>>
>>>>>
>>>>> Not that I understand why. I thought the palette was already in
>>>>> "big-endian" order, with the A, R, G, B bytes in that order, and
>>> that
>>>>> the previous patch would rather fail instead, so to say. But
>>> alright.
>>>>>
>>>>> Mats
>>>>>
>>>>> _______________________________________________
>>>>> ffmpeg-devel mailing list
>>>>> ffmpeg-devel at ffmpeg.org
>>>>> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>>>
>>>>
>>>> After applying this patch, would you like to try "ffmpeg -i
>>>> 8bpp_129.mov -vcodec rawvideo out.mov" with the file below on that
>>>> mips machine, just to see that nothing is wrong with the palette
>>>> storage on the demuxer side?
>>>
>>> qemu-mips is free software, so is gcc, i think its more efficient
>>> if you test this yourself.
>>> ... Considering the number of patch revissions ;)
>>>
>>> [...]
>>
>> The number of revisions aren't because of not being able to test it on a big-endian machine. I had no idea about this issue until you told me about it. It can't be that hard to try these files on your  mips machine. Or are you perhaps using QEMU yourself? I actually installed it here today for the sole reason of testing with FFmpeg, but it is a slow, cpu-hungry bugger here that makes the fan on my laptop spin constantly.
>
> i use qemu-mips myself
> mips is not fast, building on mips is really slow if one did that
>
> also cross building to mips should be the same speed as building to
> x86 would be
> qemu-mips would only run for a few seconds to do whatever you want
> to test.
> I assume you would test 1 image and not a whole movie
>
> [...]
>
>
>

Ok on your using mips. Nice. But since you already have it set up 
correctly there, and since you were the one who reported this issue 
about wrong colors on a big-endian machine (virtual or not), it wouldn't 
be that hard to try the files I mentioned with v28 of the patch, in 
order to assure that everything's alright. I would be grateful if you 
could do that.

Mats