[vlc-devel] VLC RBG chroma issues

[Rémi Denis-Courmont]

Le vendredi 6 mars 2009 10:58:43 basos g, vous avez écrit :
> But there are scatered vlc code parts where RBG masks are set with
> kindof arbitrary way for the big endian arch.
> My question is should rgb masks carry the correct endianess
> information ( depending on arch ) or should they have always the big
> endian values and let the compiler do the magic ? For this to be
> ansered i have to understand where mask information are used appart
> from VLC to avutil mappings.

IMHO, it should be the correct value within VLC, i.e. following local 
endianess. Otherwise, we'd get discontinuous bit masks in case of non-byte 
masks (RV15 and RV16 especially), and end call htonl() all over the place.
Besides, the only cases that might not use local endianess is basically raw 
video streaming/recording.

FYI, the non-accelarated X outputs are setting the byte masks. Those are 
provided by the X11 protocol. I don't know if they're correctly implemented 
by filters though.

By the way, has anyone tested RGB2 recently (and why that name?)? I get 
completely wrong colors with both Xlib and XCB outputs, pseudocolor and 
truecolor visuals.

-- 
Rémi Denis-Courmont