Asiliantfb (Frame Buffer) Driver

Asiliantfb (Frame Buffer) Driver

[Dieu Morales]

Greetings, I am working with the latest 2.6.7 kernel
source on a ppc radstone PPC7A board and have included
the CT69000 frame buffer to attempt to get a vga
console.  However, when I set
conswitchp = &dummy_con;
the video initializes (monitor signal LED turns green
from orange) and I get a green screen on the monitor.
And when I set
vgacon_remap_base = 0xfe000000;
conswitchp = &vga_con;
the video initializes (monitor signal LED turns green
from orange)  and 50% of the time I get nothing on the
monitor and the other 50% of the time I get what looks
to be a screen saver that displays colors fading into
one another (red->green->blue...) for about 10 sec
like some kind of a video test. I placed
vgacon_remap_base at the PCI0 I/O base, is this
correct?
I have added console=tty0 to the kernel command line
and have not identified any other configuration items
needed.  The serial console I/O which I am expecting
to see on the monitor shows the driver initialized
without errors.  Am I missing something in terms of
configuration?  I have also noticed that when using
vga_con no graphics card is detected in vgacon.c so
goto no_vga and conswitchp = &dummy_con;
Would this be related to the problem?

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Asiliantfb (Frame Buffer) Driver

[Adrian Cox]

On Sun, 2004-06-20 at 21:40, Dieu Morales wrote:
[snip]
> I have added console=tty0 to the kernel command line
> and have not identified any other configuration items
> needed.  The serial console I/O which I am expecting
> to see on the monitor shows the driver initialized
> without errors.  Am I missing something in terms of
> configuration?  I have also noticed that when using
> vga_con no graphics card is detected in vgacon.c so
> goto no_vga and conswitchp = &dummy_con;
> Would this be related to the problem?

The problem is that vgacon and frame buffer consoles are different
things:

vgacon is a console using the text modes of VGA cards. To stand a chance
of working it requires the boot firmware of your board to execute the
VGA BIOS code on the graphics adapter. It works on PCs and possibly some
non-x86 workstations.

Frame buffer consoles (fbcon) use the graphics modes of the video card,
and chipset specific drivers.  I wrote the original asiliantfb for the
69030 chip. Saito.K & Jeanne replaced the 69030 support with 69000, and
it got into the kernel.

First, try removing vgacon and checking that you have fbcon enabled.
If that fails, try contacting the driver's authors. There's no contact
address for them in the source, though Google finds one other driver by
them with this attribution:
MODULE_AUTHOR("Saito.K & Jeanne <ksaito at interface.co.jp>");

- Adrian Cox
Humboldt Solutions Ltd.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Asiliantfb (Frame Buffer) Driver

[Dieu Morales]

--- Adrian Cox <adrian@humboldt.co.uk> wrote:
> The problem is that vgacon and frame buffer consoles
> are different
> things:
>
> vgacon is a console using the text modes of VGA
> cards. To stand a chance
> of working it requires the boot firmware of your
> board to execute the
> VGA BIOS code on the graphics adapter. It works on
> PCs and possibly some
> non-x86 workstations.
>
> Frame buffer consoles (fbcon) use the graphics modes
> of the video card,
> and chipset specific drivers.  I wrote the original
> asiliantfb for the
> 69030 chip. Saito.K & Jeanne replaced the 69030
> support with 69000, and
> it got into the kernel.
>
> First, try removing vgacon and checking that you
> have fbcon enabled.
> If that fails, try contacting the driver's authors.
> There's no contact
> address for them in the source, though Google finds
> one other driver by
> them with this attribution:
> MODULE_AUTHOR("Saito.K & Jeanne <ksaito at
> interface.co.jp>");
>
> - Adrian Cox
> Humboldt Solutions Ltd.

My firmware does initialize the CT69000 as an
environment option that allows what looks like VGA
mode consisting of a simple black screen with
white text.  As per some suggestions I have removed
vga_con and added fb_con but received different(worse)
results.  Without conswitchp = &fb_con I get an
exception shortly after the asiliant driver has been
initialized. And with conswitchp = &fb_con the system
hangs immediately after ppc.md.setup_arch in setup.c.




Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Asiliantfb (Frame Buffer) Driver

[Adrian Cox]

On Mon, 2004-06-21 at 22:21, Dieu Morales wrote:
> My firmware does initialize the CT69000 as an
> environment option that allows what looks like VGA
> mode consisting of a simple black screen with
> white text.  As per some suggestions I have removed
> vga_con and added fb_con but received different(worse)
> results.  Without conswitchp = &fb_con I get an
> exception shortly after the asiliant driver has been
> initialized. And with conswitchp = &fb_con the system
> hangs immediately after ppc.md.setup_arch in setup.c.

What I do is to have this in my platform setup code:

#ifdef CONFIG_DUMMY_CONSOLE
	conswitchp = &dummy_con;
#endif

Then I select the console from the command line.

I can't vouch for the modified version in the kernel tree. My original
driver for the 69030 had a bug that made it fail if the VGA firmware had
already initialised the card. That wasn't a problem on the target
platform, but it is why I never submitted the driver to the kernel.


- Adrian Cox
Humboldt Solutions Ltd


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Asiliantfb (Frame Buffer) Driver

[Dieu Morales]

Adrian,

Thank you for your suggestion.  The dummy_console with
the fb_con.c driver built in was the combination that
worked.  Seems like I tried every combination except
that one.  Also, the driver works whether or not the
firmware initializes the CT69000, which is good.
Thanks again.

-Michael Morales

--- Adrian Cox <adrian@humboldt.co.uk> wrote:
> On Mon, 2004-06-21 at 22:21, Dieu Morales wrote:
> > My firmware does initialize the CT69000 as an
> > environment option that allows what looks like VGA
> > mode consisting of a simple black screen with
> > white text.  As per some suggestions I have
> removed
> > vga_con and added fb_con but received
> different(worse)
> > results.  Without conswitchp = &fb_con I get an
> > exception shortly after the asiliant driver has
> been
> > initialized. And with conswitchp = &fb_con the
> system
> > hangs immediately after ppc.md.setup_arch in
> setup.c.
>
> What I do is to have this in my platform setup code:
>
> #ifdef CONFIG_DUMMY_CONSOLE
> 	conswitchp = &dummy_con;
> #endif
>
> Then I select the console from the command line.
>
> I can't vouch for the modified version in the kernel
> tree. My original
> driver for the 69030 had a bug that made it fail if
> the VGA firmware had
> already initialised the card. That wasn't a problem
> on the target
> platform, but it is why I never submitted the driver
> to the kernel.
>
>
> - Adrian Cox
> Humboldt Solutions Ltd
>
>
>




Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Asiliantfb (Frame Buffer) Driver

[Geert Uytterhoeven]

On Sun, 20 Jun 2004, Dieu Morales wrote:
> Greetings, I am working with the latest 2.6.7 kernel
> source on a ppc radstone PPC7A board and have included
> the CT69000 frame buffer to attempt to get a vga
> console.  However, when I set
> conswitchp = &dummy_con;
> the video initializes (monitor signal LED turns green
> from orange) and I get a green screen on the monitor.
> And when I set
> vgacon_remap_base = 0xfe000000;
> conswitchp = &vga_con;
> the video initializes (monitor signal LED turns green
> from orange)  and 50% of the time I get nothing on the
> monitor and the other 50% of the time I get what looks
> to be a screen saver that displays colors fading into
> one another (red->green->blue...) for about 10 sec
> like some kind of a video test. I placed
> vgacon_remap_base at the PCI0 I/O base, is this
> correct?
> I have added console=tty0 to the kernel command line
> and have not identified any other configuration items
> needed.  The serial console I/O which I am expecting
> to see on the monitor shows the driver initialized
> without errors.  Am I missing something in terms of
> configuration?  I have also noticed that when using
> vga_con no graphics card is detected in vgacon.c so
> goto no_vga and conswitchp = &dummy_con;
> Would this be related to the problem?

Does you firmware initialize the CT69000?
Asiliantfb may need an initialized CT69000.
Vgacon needs a graphics chip initialized to VGA mode.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/