Hi Guys,

  First thanks to everyone that replied - i'm always grateful for any help.

Antonino A. Daplas wrote:
 >

 > I haven't seen viafb, but the white box is most probably the result of
 > viafb switching to graphics mode when vgacon is still the owner of the
 > console.
 >
 > The proper fix is not to switch to graphics mode at all (fbcon, or any
 > other app will take care of switching to graphics mode) or switch to
 > graphics modes only _after_ the call to register_framebuffer().
 >
 > The call that switches the hardware to graphics mode is typically
 > fb_set_par() in 2.6 kernels.


This could well be the problem. The code doesnt appear to call 
fb_set_par in the pci probe function, but it does have a block of code 
which looks like it does a similar thing to fb_set_par (setmode & 
pandisplay). I've been playing around with this and come to the 
conclusion that fb_set_par doesnt appear to create a valid framebuffer 
console for me.

  Using fbset to swap to a different mode, appears to work, but my 
monitor reports video mode not supported and so it doesnt work (for me 
at any rate). When I took the hardcoded switch to graphic mode out of 
the pci probe function, I get the exact same monitor issue on boot up as 
with fbset. fbset -i remotely from another pc makes all the right noises 
as if its working fine!

  Unless anyone has any ideas, I guess I'll have to see how deep the 
rabbit hole goes...

Liab


_______________________________________________
directfb-users mailing list
[email protected]
http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-users

Reply via email to