On Tue, 2004-11-16 at 20:04 +0000, Torgeir Veimo wrote:
> On Tue, 2004-11-16 at 20:56 +0100, Olaf Leidinger wrote:
> > Hello !
> >
> > > Yes, you're right - it sais no.
> > >
> > > > Apparently the radeonfb in 2.6.9 doesn't set any acceleration id so
> > > > it's unaccelerated similar to the vesafb.
> > >
> > > So if this bits where set, would it be accelerated?
> > And would directfb be accelerated using vesafb if these bits where set?
>
> I don't think so. A qualified guess without knowing any directfb
> internals would be that it queries the fb driver using an ioctl. You
> might find the sources for fbset and look at what it does to retrieve
> the accelerator id.
Try adding around line 1775 to drivers/video/aty/radeon_base.c;
info->fix.accel = FB_ACCEL_ATI_RADEON;
Don't know if it will make directfb work though, I get (running vdr):
(*) Direct/Thread: Running 'Keyboard Input' (INPUT, 6953)...
(*) DirectFB/Input: Keyboard 0.9 (convergence integrated media GmbH)
(*) DirectFB/Genefx: MMX detected and enabled
(*) DirectFB/Graphics: ATI Radeon 0.2 (Michal D�nzer)
(!) Direct/Modules: Unable to dlopen
`/usr/local/lib/directfb-0.9.21/wm/libdirectfbwm_unique.so'!
--> libuniquewm-0.9.so.21: cannot open shared object file: No such
file or directory
(*) DirectFB/Core/WM: Default 0.2 (Convergence GmbH)
(!) [ 6934: 0.000] --> Caught signal 11 (at (nil), invalid address)
<--
./rundfb: line 6: 6934 Killed vdr -D 2 -s "killall
vdr;/bin/false" -P"softdevice -vo dfb:"
--
Torgeir Veimo <[EMAIL PROTECTED]>
