Re: [AD] Framebuffer resolution problems

[Milan Mimica <milan.mimica1@xxxxxxxxxx>]

Grzegorz Adam Hankiewicz wrote:
> Well, at http://linux-fbdev.sourceforge.net/ they don't have much
> documentation but you can subscribe to the mailing lists where you
> can ask the fb implementation authors what is happening. Apparently
> fb programming is so easy reading that header file should be
> enough. At least that's the feeling you get from reading thousands
> of micro howtos on how to program it, which are scattered on the
> net and none of them serves as a complete reference.

I thought so.

> If you want to work on this, you first have to get FB support
> compiled and working for your Allegro and try it. Possibly with a
> vesa driver if you don't have supported hardware.

I used vesafb driver and everything was working fine, at least the last time I 
tried it (some latest WIP's). There were some issues, but I think they where 
driver related, like inability to properly change resolution and color depth.

> If Allegro crashes on you on something simple as clear_bitmap(), it
> means the current CVS code doesn't really work for you, and should
> try with the patch I posted, which reverts from 1.29 to 1.27. You
> can see why clear_bitmap() crashes when the height of the bitmap
> has a stupidly high value.

I'm not so sure any more I want to work on this :-/ I just remembered I have a 
problem when switching from X (nvidia drivers) to fb console. The screen is 
messed up.

> Anyway, once you get fb working for you, it would be a matter of
> contacting the affected people (those who proposed changes to the
> fbcon.c file) and ask them to try some code. This code would try
> to printf or log the values used in the variables used in all the
> combinations that the code has gone through.

How come that a fb kernel API is so bad designed? I hope these are just bugs, 
because code using fb API is supposed to work on every hardware:
From Documentation/fb/framebuffer.txt:
"It represents the frame buffer of some video hardware and allows application
software to access the graphics hardware through a well-defined interface, so
the software doesn't need to know anything about the low-level (hardware
register) stuff."


--
Milan Mimica