From: "Petr Vandrovec" <VANDROVE@vc.cvut.cz>
Organization: CC CTU Prague
To: James Simmons <jsimmons@infradead.org>
Date: Tue, 17 Jun 2003 22:29:58 +0200
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Re: FRAMEBUFFER (and console)
Cc: linux-kernel@vger.kernel.org, spyro@f2s.com
Message-ID: <4B38D4B6A6E@vcnet.vc.cvut.cz>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1502
Lines: 31

On 17 Jun 03 at 21:03, James Simmons wrote:
> > My framebuffer (and therefore system console, by definition) come up
> > rather late.
> > 
> > It seems the console doesnt care to check for drivers comming up after a
> > certain time, and thus I get no output despite the driver working.
> 
> The reason for this is because the framebuffers most often depend on the 
> bus being set up. Usually this happening later in the boot process. What 
> are trying to do? Retrieve the earlier printk messages. Do you have 
> DUMMY_CONSOLE set to Y. I believe the data is transfered from dummycon to 
> fbcon after fbcon is initialized. If you having problems with that try 
> increasing the size of dummycon's "screen". See dummycon.c for more 
> details.

Maybe he just enabled vga16 + XXXfb. First vga16 comes up, and start
painting characters. Few microseconds after that XXXfb (for example
matroxfb) comes up, registers itself as /dev/fb1 and reprograms hardware
to non-VGA mode. 

>From that point on vga16fb paints characters to unmapped memory, and
there is only black on screen.
                                                Petr Vandrovec
                                                vandrove@vc.cvut.cz
                                                

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/