[FFmpeg-devel] [PATCH] Fix non-rounding up to next 16-bit aligned bug in IFF decoder

[Sebastian Vater]

M?ns Rullg?rd a ?crit :
> Sebastian Vater <cdgs.basty at googlemail.com> writes:
>
>   
>> -    for(; dst < end; dst++) {
>> -        *dst |= get_bits1(&gb) << plane;
>> +    for(; dst < end; dst += 4) {
>> +        const uint32_t v  = AV_RN32A(dst) | lut[get_bits(&gb, 4)];
>> +        AV_WN32A(dst, v);
>>      }
>>     
>
> What happened to a non-mod4 tail?  Is that not allowed?
>
>   
It's not necessary anymore!

Thanks to word boundary alignment required per line, the number of bits
is always divisible by 16.

Since we're reading 4 bits in one loop, this is also always divisible by
16...

-- 

Best regards,
                   :-) Basty/CDGS (-: