Big-Endian problem with Fujitsu CoralPA (Clemens Koller)

[Alex Deucher]

On Nov 15, 2007 4:08 AM, Jarno Manninen <jarno.manninen at tut.fi> wrote:
>
> > I would be glad to do some tests and debugging here, but
> > I am not really happy with the lacking documentation / my lilited
> > knowledge about the details and how it's supposed to be
> > implemented (in an elegant way = without performance
> > penalty / limited functionality).
>
> Hello!
>
> I had the same problem, or still have if you will, with endianess but with
> SMI722. Quick and easy, but still a bit slow solution is to deny all SW
> access to the framebuffer and swap the  bytes correctly in
> UploadTo/DownloadFrom-screen functions. Needless to say this requires EXA to
> work. However this just plain sucks for drawing non accelerated graphics, but
> one could do some imaginative juggling of the fb-layers ROP variables to fix
> that.

the siliconmotion driver in git has EXA support, although it's pretty slow.

>
> Now that I've almost finished renovation/moving and the ISP finally managed to
> connect the cables I can start to figuring out where/how to put my Q&D hacks.

Please post your patches/git tree when it's ready.

Alex