Linux Frame Buffer Device Development

damien83@... wrote:
> Hello,
> 
> I am trying to implement hardware acceleration to a framebuffer driver, throught
> the functions fb_fillrect, fb_copyarea and fb_imageblit. So I replace the
> cfb_fillrect functions by my own, and add FBINFO_HWACCEL_FILLRECT to
> fb_info->flags. But I try to use both embbeded/QT and Nano-X, and these
> hardware accelerated functions are never called.
> 
> I did a lot of research but probably missed someting... I do not understand why
> these "needed" functions are never called ?

Unfortunately, these accelerated functions are for the console use only.  No
other app can use this functions.

Tony

Florian Echtler wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi, I was trying to have fbcon and nvidiafb as modules, however,
> the console stays plain VGA when I load both. fbset complains 
> about "no such device: /dev/fb0".. (the device itself exists, and
> works when everything is compiled into the kernel).

What kernel version are you using?

Can you post your dmesg after doing a modprobe nvidiafb and modprobe
fbcon?  Note that you can load the modules in any order, it does not
matter.

> 
> Moreover, how can I pass the "video=.." option from the kernel
> commandline to the modules?

modprobe nvidiafb options <name>=<value>

Tony

Forgot to CC.

Florian Echtler wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi, I was trying to have fbcon and nvidiafb as modules, however,
> the console stays plain VGA when I load both. fbset complains 
> about "no such device: /dev/fb0".. (the device itself exists, and
> works when everything is compiled into the kernel).

What kernel version are you using?

Can you post your dmesg after doing a modprobe nvidiafb and modprobe
fbcon?  Note that you can load the modules in any order, it does not
matter.

> 
> Moreover, how can I pass the "video=.." option from the kernel
> commandline to the modules?

modprobe nvidiafb options <name>=<value>

Tony

On Wed, Aug 24, 2005 at 06:05:34AM +0800, Antonino A. Daplas wrote:
> Florian Echtler wrote:
> >-----BEGIN PGP SIGNED MESSAGE-----
> >Hash: SHA1
> >
> >Hi, I was trying to have fbcon and nvidiafb as modules, however,
> >the console stays plain VGA when I load both. fbset complains 
> >about "no such device: /dev/fb0".. (the device itself exists, and
> >works when everything is compiled into the kernel).
> 
> What kernel version are you using?

BTW, would it be possible to use the new initramfs infrastructure, to move the
fbdev drivers out of the kernel and into their own initramfs thingy, and still
be able to start them as early as possible ?

Friendly,

Sven Luther

Sven Luther wrote:
> On Wed, Aug 24, 2005 at 06:05:34AM +0800, Antonino A. Daplas wrote:
>> Florian Echtler wrote:
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>>
>>> Hi, I was trying to have fbcon and nvidiafb as modules, however,
>>> the console stays plain VGA when I load both. fbset complains 
>>> about "no such device: /dev/fb0".. (the device itself exists, and
>>> works when everything is compiled into the kernel).
>> What kernel version are you using?
> 
> BTW, would it be possible to use the new initramfs infrastructure, to move the
> fbdev drivers out of the kernel and into their own initramfs thingy, and still
> be able to start them as early as possible ?

I don't see why not. You can compile fbcon statically, and the drivers as modules.
Or both fbcon and drivers as modules.

Tony

Sven Luther wrote:
> On Fri, Aug 26, 2005 at 10:06:53PM +0800, Antonino A. Daplas wrote:
>> Sven Luther wrote:
>>> On Wed, Aug 24, 2005 at 06:05:34AM +0800, Antonino A. Daplas wrote:
>>>> Florian Echtler wrote:
>>>>> -----BEGIN PGP SIGNED MESSAGE-----
>>>>> Hash: SHA1
>>>>>
>>>>> Hi, I was trying to have fbcon and nvidiafb as modules, however,
>>>>> the console stays plain VGA when I load both. fbset complains 
>>>>> about "no such device: /dev/fb0".. (the device itself exists, and
>>>>> works when everything is compiled into the kernel).
>>>> What kernel version are you using?
>>> BTW, would it be possible to use the new initramfs infrastructure, to move 
>>> the
>>> fbdev drivers out of the kernel and into their own initramfs thingy, and 
>>> still
>>> be able to start them as early as possible ?
>> I don't see why not. You can compile fbcon statically, and the drivers as 
>> modules.
>> Or both fbcon and drivers as modules.
> 
> Yeah, well, the constraint is that this is on powerpc, and there is no vgacon
> or whatever, and we still like to have debug output as soon as possible.
> 

The framebuffer can load as early as you can load initrd. So you'll basically
miss bugs before the filesystem gets loaded. The screen is buffered, so the
contents are there. So if you are fortunate enough to load the framebuffer system
even if the bug happened a bit earlier, you might still be able to see the
message.

Note that the framebuffer system loads a little bit late compared to 2.4. It gets
loaded after PCI, PARISC and rapidio.

Tony

On Sat, Aug 27, 2005 at 08:20:11AM +0800, Antonino A. Daplas wrote:
> Sven Luther wrote:
> >On Fri, Aug 26, 2005 at 10:06:53PM +0800, Antonino A. Daplas wrote:
> >>Sven Luther wrote:
> >>>On Wed, Aug 24, 2005 at 06:05:34AM +0800, Antonino A. Daplas wrote:
> >>>>Florian Echtler wrote:
> >>>>>-----BEGIN PGP SIGNED MESSAGE-----
> >>>>>Hash: SHA1
> >>>>>
> >>>>>Hi, I was trying to have fbcon and nvidiafb as modules, however,
> >>>>>the console stays plain VGA when I load both. fbset complains 
> >>>>>about "no such device: /dev/fb0".. (the device itself exists, and
> >>>>>works when everything is compiled into the kernel).
> >>>>What kernel version are you using?
> >>>BTW, would it be possible to use the new initramfs infrastructure, to 
> >>>move the
> >>>fbdev drivers out of the kernel and into their own initramfs thingy, and 
> >>>still
> >>>be able to start them as early as possible ?
> >>I don't see why not. You can compile fbcon statically, and the drivers as 
> >>modules.
> >>Or both fbcon and drivers as modules.
> >
> >Yeah, well, the constraint is that this is on powerpc, and there is no 
> >vgacon
> >or whatever, and we still like to have debug output as soon as possible.
> >
> 
> The framebuffer can load as early as you can load initrd. So you'll 
> basically
> miss bugs before the filesystem gets loaded. The screen is buffered, so the
> contents are there. So if you are fortunate enough to load the framebuffer 
> system
> even if the bug happened a bit earlier, you might still be able to see the
> message.
> 
> Note that the framebuffer system loads a little bit late compared to 2.4. 
> It gets
> loaded after PCI, PARISC and rapidio.

I guess we have always the serial console if earlier debugging is needed.

Friendly,

Sven Luther

On Fri, Aug 26, 2005 at 10:06:53PM +0800, Antonino A. Daplas wrote:
> Sven Luther wrote:
> >On Wed, Aug 24, 2005 at 06:05:34AM +0800, Antonino A. Daplas wrote:
> >>Florian Echtler wrote:
> >>>-----BEGIN PGP SIGNED MESSAGE-----
> >>>Hash: SHA1
> >>>
> >>>Hi, I was trying to have fbcon and nvidiafb as modules, however,
> >>>the console stays plain VGA when I load both. fbset complains 
> >>>about "no such device: /dev/fb0".. (the device itself exists, and
> >>>works when everything is compiled into the kernel).
> >>What kernel version are you using?
> >
> >BTW, would it be possible to use the new initramfs infrastructure, to move 
> >the
> >fbdev drivers out of the kernel and into their own initramfs thingy, and 
> >still
> >be able to start them as early as possible ?
> 
> I don't see why not. You can compile fbcon statically, and the drivers as 
> modules.
> Or both fbcon and drivers as modules.

Yeah, well, the constraint is that this is on powerpc, and there is no vgacon
or whatever, and we still like to have debug output as soon as possible.

Friendly,

Sven Luther

On Sat, 27 Aug 2005, Sven Luther wrote:
> On Sat, Aug 27, 2005 at 08:20:11AM +0800, Antonino A. Daplas wrote:
> > Sven Luther wrote:
> > >On Fri, Aug 26, 2005 at 10:06:53PM +0800, Antonino A. Daplas wrote:
> > >>Sven Luther wrote:
> > >>>On Wed, Aug 24, 2005 at 06:05:34AM +0800, Antonino A. Daplas wrote:
> > >>>>Florian Echtler wrote:
> > >>>>>-----BEGIN PGP SIGNED MESSAGE-----
> > >>>>>Hash: SHA1
> > >>>>>
> > >>>>>Hi, I was trying to have fbcon and nvidiafb as modules, however,
> > >>>>>the console stays plain VGA when I load both. fbset complains 
> > >>>>>about "no such device: /dev/fb0".. (the device itself exists, and
> > >>>>>works when everything is compiled into the kernel).
> > >>>>What kernel version are you using?
> > >>>BTW, would it be possible to use the new initramfs infrastructure, to 
> > >>>move the
> > >>>fbdev drivers out of the kernel and into their own initramfs thingy, and 
> > >>>still
> > >>>be able to start them as early as possible ?
> > >>I don't see why not. You can compile fbcon statically, and the drivers as 
> > >>modules.
> > >>Or both fbcon and drivers as modules.
> > >
> > >Yeah, well, the constraint is that this is on powerpc, and there is no 
> > >vgacon
> > >or whatever, and we still like to have debug output as soon as possible.
> > >
> > 
> > The framebuffer can load as early as you can load initrd. So you'll 
> > basically
> > miss bugs before the filesystem gets loaded. The screen is buffered, so the
> > contents are there. So if you are fortunate enough to load the framebuffer 
> > system
> > even if the bug happened a bit earlier, you might still be able to see the
> > message.
> > 
> > Note that the framebuffer system loads a little bit late compared to 2.4. 
> > It gets
> > loaded after PCI, PARISC and rapidio.
> 
> I guess we have always the serial console if earlier debugging is needed.

Don't we have promcon, on some PPCs?

Gr{oetje,eeting}s,

						Geert

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

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

On Sun, Aug 28, 2005 at 12:16:54PM +0200, Geert Uytterhoeven wrote:
> On Sat, 27 Aug 2005, Sven Luther wrote:
> > On Sat, Aug 27, 2005 at 08:20:11AM +0800, Antonino A. Daplas wrote:
> > > Sven Luther wrote:
> > > >On Fri, Aug 26, 2005 at 10:06:53PM +0800, Antonino A. Daplas wrote:
> > > >>Sven Luther wrote:
> > > >>>On Wed, Aug 24, 2005 at 06:05:34AM +0800, Antonino A. Daplas wrote:
> > > >>>>Florian Echtler wrote:
> > > >>>>>-----BEGIN PGP SIGNED MESSAGE-----
> > > >>>>>Hash: SHA1
> > > >>>>>
> > > >>>>>Hi, I was trying to have fbcon and nvidiafb as modules, however,
> > > >>>>>the console stays plain VGA when I load both. fbset complains 
> > > >>>>>about "no such device: /dev/fb0".. (the device itself exists, and
> > > >>>>>works when everything is compiled into the kernel).
> > > >>>>What kernel version are you using?
> > > >>>BTW, would it be possible to use the new initramfs infrastructure, to 
> > > >>>move the
> > > >>>fbdev drivers out of the kernel and into their own initramfs thingy, and 
> > > >>>still
> > > >>>be able to start them as early as possible ?
> > > >>I don't see why not. You can compile fbcon statically, and the drivers as 
> > > >>modules.
> > > >>Or both fbcon and drivers as modules.
> > > >
> > > >Yeah, well, the constraint is that this is on powerpc, and there is no 
> > > >vgacon
> > > >or whatever, and we still like to have debug output as soon as possible.
> > > >
> > > 
> > > The framebuffer can load as early as you can load initrd. So you'll 
> > > basically
> > > miss bugs before the filesystem gets loaded. The screen is buffered, so the
> > > contents are there. So if you are fortunate enough to load the framebuffer 
> > > system
> > > even if the bug happened a bit earlier, you might still be able to see the
> > > message.
> > > 
> > > Note that the framebuffer system loads a little bit late compared to 2.4. 
> > > It gets
> > > loaded after PCI, PARISC and rapidio.
> > 
> > I guess we have always the serial console if earlier debugging is needed.
> 
> Don't we have promcon, on some PPCs?

Some, but not all. My main problem is that i build a generic powerpc kernel
for debian, and right now we need many frambuffers builtin for the different
machines. My idea is to build all those as modules, and put them in initramfs,
and still call them early enough. Since initramfs, from what i am told, is
accessed earlier than the initrd was, it will be more usable this way.

Friendly,

Sven Luther

On Sun, 28 Aug 2005, Sven Luther wrote:
> On Sun, Aug 28, 2005 at 12:16:54PM +0200, Geert Uytterhoeven wrote:
> > On Sat, 27 Aug 2005, Sven Luther wrote:
> > > On Sat, Aug 27, 2005 at 08:20:11AM +0800, Antonino A. Daplas wrote:
> > > > Sven Luther wrote:
> > > > >Yeah, well, the constraint is that this is on powerpc, and there is no 
> > > > >vgacon
> > > > >or whatever, and we still like to have debug output as soon as possible.
> > > > 
> > > > The framebuffer can load as early as you can load initrd. So you'll 
> > > > basically
> > > > miss bugs before the filesystem gets loaded. The screen is buffered, so the
> > > > contents are there. So if you are fortunate enough to load the framebuffer 
> > > > system
> > > > even if the bug happened a bit earlier, you might still be able to see the
> > > > message.
> > > > 
> > > > Note that the framebuffer system loads a little bit late compared to 2.4. 
> > > > It gets
> > > > loaded after PCI, PARISC and rapidio.
> > > 
> > > I guess we have always the serial console if earlier debugging is needed.
> > 
> > Don't we have promcon, on some PPCs?
> 
> Some, but not all. My main problem is that i build a generic powerpc kernel
> for debian, and right now we need many frambuffers builtin for the different
> machines. My idea is to build all those as modules, and put them in initramfs,
> and still call them early enough. Since initramfs, from what i am told, is
> accessed earlier than the initrd was, it will be more usable this way.

Another option would be to use offb, and find a way to `overtake' offb later by
a chipset-specific frame buffer device. A bit like vgacon is overtaken by fbdev
on other systems.

Gr{oetje,eeting}s,

						Geert

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

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

On Sun, Aug 28, 2005 at 08:55:16PM +0200, Geert Uytterhoeven wrote:
> On Sun, 28 Aug 2005, Sven Luther wrote:
> > On Sun, Aug 28, 2005 at 12:16:54PM +0200, Geert Uytterhoeven wrote:
> > > On Sat, 27 Aug 2005, Sven Luther wrote:
> > > > On Sat, Aug 27, 2005 at 08:20:11AM +0800, Antonino A. Daplas wrote:
> > > > > Sven Luther wrote:
> > > > > >Yeah, well, the constraint is that this is on powerpc, and there is no 
> > > > > >vgacon
> > > > > >or whatever, and we still like to have debug output as soon as possible.
> > > > > 
> > > > > The framebuffer can load as early as you can load initrd. So you'll 
> > > > > basically
> > > > > miss bugs before the filesystem gets loaded. The screen is buffered, so the
> > > > > contents are there. So if you are fortunate enough to load the framebuffer 
> > > > > system
> > > > > even if the bug happened a bit earlier, you might still be able to see the
> > > > > message.
> > > > > 
> > > > > Note that the framebuffer system loads a little bit late compared to 2.4. 
> > > > > It gets
> > > > > loaded after PCI, PARISC and rapidio.
> > > > 
> > > > I guess we have always the serial console if earlier debugging is needed.
> > > 
> > > Don't we have promcon, on some PPCs?
> > 
> > Some, but not all. My main problem is that i build a generic powerpc kernel
> > for debian, and right now we need many frambuffers builtin for the different
> > machines. My idea is to build all those as modules, and put them in initramfs,
> > and still call them early enough. Since initramfs, from what i am told, is
> > accessed earlier than the initrd was, it will be more usable this way.
> 
> Another option would be to use offb, and find a way to `overtake' offb later by
> a chipset-specific frame buffer device. A bit like vgacon is overtaken by fbdev
> on other systems.

Yep, but less clean. How early is initrramfs compared to when fbdev needs his
stuff. I mean, you could access those modules pretty early.

Friendly,

Sven Luther

This one has a simple bitmapped font that can be printed in different =
sizes and directions in the frame buffer.
http://www.akrobiz.com/ezfb/

~James.   :o)
  ----- Original Message -----=20
  From: William Estrada=20
  To: linux-fbdev-users@...
  Sent: Saturday, August 27, 2005 2:35 PM
  Subject: [Linux-fbdev-users] printing in a framebuffer


  Group,

    I'm just starting programming with framebuffer stuff and I have a =
simple
  question for the group.

    Are there APIs that can be used to 'print' text in a framebuffer?


  --=20
    William Estrada    --  MrUmunhum at popdial dot com  --   =
408-997-0743
    Ymessenger ID: MrUmunhum
    HTTP://Mt-Umunhum-Wireless.net

If you don't see Tux, the penguin, on boot up, you didn't get a working =
frame buffer (usually).

What is your video card?

There are a number of individual choices you need to make in the kernel =
config to make it all work.

Some video chip sets work better than others.

So far, I've had good luck with anything Matrox.

http://www.akrobiz.com/ezfb/

~James.   :o)


  ----- Original Message -----=20
  From: William Estrada=20
  To: linux-fbdev-users@...
  Sent: Wednesday, August 17, 2005 1:34 PM
  Subject: [Linux-fbdev-users] Getting started


  Hi guys,

    Would someone please point me to a 'Getting started' URL for =
framebuffer
  support?

    I am running Fedora Core 4 and I have rebuilt the kernel to make =
sure
  FB support is in the kernel.  I have a simple FB program that I have=20
  compiled but everytime I try to run it, I get:

    Error: cannot open framebuffer device /dev/fb0. - No such device =
rc=3D-1

    The open code looks like this:

        // Open the file for reading and writing
        char*  FB   =3D "/dev/fb0";
        int    fbfd =3D 0;
        fbfd =3D open( FB, O_RDWR);
        if (fbfd < 0) {
          printf("Error: cannot open framebuffer device %s. - %s =
rc=3D%d\n",
                 FB, strerror(errno), fbfd);
            exit(1);
        }
        printf("The framebuffer device was opened successfully. - %d\n", =
fbfd );

    I have done the mkdir:

    [wre00@... FB]$ ls -lrt /dev/fb*
    crwxrwxr-x  1 wre00 root 29, 0 Aug 17 10:05 /dev/fb0

    Any ideas/help??

  --=20
    William Estrada   --  MrUmunhum at popdial dot com  --  408-997-0743
    Ymessenger ID: MrUmunhum
    HTTP://Mt-Umunhum-Wireless.net

Tiancheng,

  My typo!

    $ sudo mknod /dev/fb0 c 29 0
    $ cd /src/FB
    $ sudo ./a.out
    Error: cannot open framebuffer device /dev/fb0. - No such device rc=-1
    $ ls -lrt /dev/fb*
    crw-r--r--  1 root root 29, 0 Aug 18 12:21 /dev/fb0
    $

  Also, the nod disappears after a boot.

  William Estrada 
<http://www.Mt-Umunhum-Wireless.net/resume/william_estrada.html>   --  
MrUmunhum at popdial dot com <mailto:MrUmunhum@...>  --  
408-997-0743
  Ymessenger ID: MrUmunhum
  HTTP://Mt-Umunhum-Wireless.net



Tiancheng Song wrote:

> try use "mknod (root) instead of "mkdir".
>  
>
>

Tiancheng,

  I got it to work.  It had to do with my kernel boot command line in 
grub.conf.  I had to add:

     vga=0x318 video=rivafb

  I used the standard kernel that came with Core 4 and all is well.  
Like magic the /dev/fb and /dev/fb0
were defined by the system (??).  There is still a lot I don't 
understand but it is a start.  Now I need to
know how to make more Frame Buffer devices!

  Thanks for your time.

ps. The HOWTO.html was not any help for my NVidia video card problem.  I 
found it by googling
around for a time.  Bits and pieces!

  William Estrada 
<http://www.Mt-Umunhum-Wireless.net/resume/william_estrada.html>   --  
MrUmunhum at popdial dot com <mailto:MrUmunhum@...>  --  
408-997-0743
  Ymessenger ID: MrUmunhum
  HTTP://Mt-Umunhum-Wireless.net



Tiancheng Song wrote:

> I wonder your kernel is not configured properly. go to 
> http://www.faqs.org/docs/Linux-HOWTO/Framebuffer-HOWTO.html for more 
> details.
>  
> As for the disapearing of node, put "mknod /dev/fb0 c 29 0" in your 
> booting up sript will create the node when boots up.
> Good luck!
>  
> Tiancheng Song
> Software Engineering
> IDERS Incorporated
> Email: tsong@... <mailto:tsong@...>
> Ph: (204) 779-5400 x245
> http://www.iders.ca
>  
> IDERs Incorporated Confidential
> Note: This message is intended solely for the use of the designated 
> recipient(s) and their appointed delegates, and may contain 
> confidential information.  Any unauthorized disclosure, copying or 
> distribution of its contents is strictly prohibited.  If you have 
> received this message in error, please destroy it and advise the 
> sender immediately by phone, Email or facsimile.
>  
>
>     ----- Original Message -----
>     *From:* William Estrada <mailto:mrumunhum@...>
>     *To:* Tiancheng Song <mailto:tsong@...>
>     *Cc:* linux-fbdev-users@...
>     <mailto:linux-fbdev-users@...>
>     *Sent:* Thursday, August 18, 2005 2:25 PM
>     *Subject:* Re: [Linux-fbdev-users] Getting started
>
>     Tiancheng,
>
>       My typo!
>
>         $ sudo mknod /dev/fb0 c 29 0
>         $ cd /src/FB
>         $ sudo ./a.out
>         Error: cannot open framebuffer device /dev/fb0. - No such
>         device rc=-1
>         $ ls -lrt /dev/fb*
>         crw-r--r--  1 root root 29, 0 Aug 18 12:21 /dev/fb0
>         $
>
>       Also, the nod disappears after a boot.
>
>       William Estrada
>     <http://www.Mt-Umunhum-Wireless.net/resume/william_estrada.html>  
>     --  MrUmunhum at popdial dot com <mailto:MrUmunhum@...> 
>     --  408-997-0743
>       Ymessenger ID: MrUmunhum
>       HTTP://Mt-Umunhum-Wireless.net <http://Mt-Umunhum-Wireless.net>
>
>
>
>     Tiancheng Song wrote:
>
>>     try use "mknod (root) instead of "mkdir".
>>      
>>
>>

On 8/19/05, William Estrada <mrumunhum@...> wrote:
> Like magic the /dev/fb and /dev/fb0
>  were defined by the system (??).

This bit of magic is "udev". It gets messages from the kernel when
drivers are loaded, and it creates device nodes for you. When using
udev, the /dev filesystem is actually created in a ramdisk, which
explains why the device nodes disappeared when you rebooted.

> Now I need to know how to make more Frame Buffer devices!

Typically, you get one frame buffer device node per video card.
(/dev/fb is probably a link to /dev/fb0)

--=20
- Charles Lepple

Charles,

  Thanks for the pointer.  That's was good information.

  I will be adding another video card to my system soon.  Do you know
of a URL that could help with this task?  I don't know if I will be
using a NVidia card or not.  Will the system auto-magically configure
it for this setup?


  William Estrada 
<http://www.Mt-Umunhum-Wireless.net/resume/william_estrada.html>   --  
MrUmunhum at popdial dot com <mailto:MrUmunhum@...>  --  
408-997-0743
  Ymessenger ID: MrUmunhum
  HTTP://Mt-Umunhum-Wireless.net



Charles Lepple wrote:

>On 8/19/05, William Estrada <mrumunhum@...> wrote:
>  
>
>>Like magic the /dev/fb and /dev/fb0
>> were defined by the system (??).
>>    
>>
>
>This bit of magic is "udev". It gets messages from the kernel when
>drivers are loaded, and it creates device nodes for you. When using
>udev, the /dev filesystem is actually created in a ramdisk, which
>explains why the device nodes disappeared when you rebooted.
>
>  
>
>>Now I need to know how to make more Frame Buffer devices!
>>    
>>
>
>Typically, you get one frame buffer device node per video card.
>(/dev/fb is probably a link to /dev/fb0)
>
>  
>

If you want to use more than one video card then that implies that you =
need at least one that is PCI. I have had success using 5 Matrox PCI =
cards (all slots full).

EZFB has built in features for multiple video cards and an example =
program that will initialize and put a picture on every working frame =
buffer in your system.

http://www.akrobiz.com/ezfb/

James.   :o)


  ----- Original Message -----=20
  From: William Estrada=20
  To: Charles Lepple=20
  Cc: linux-fbdev-users@...
  Sent: Monday, August 22, 2005 1:23 PM
  Subject: Re: [Linux-fbdev-users] Re: Getting started


  Charles,

    Thanks for the pointer.  That's was good information.

    I will be adding another video card to my system soon.  Do you know
  of a URL that could help with this task?  I don't know if I will be
  using a NVidia card or not.  Will the system auto-magically configure
  it for this setup?



    William Estrada   --  MrUmunhum at popdial dot com  --  408-997-0743
    Ymessenger ID: MrUmunhum
    HTTP://Mt-Umunhum-Wireless.net




  Charles Lepple wrote:=20
On 8/19/05, William Estrada <mrumunhum@...> wrote:
  Like magic the /dev/fb and /dev/fb0
 were defined by the system (??).
   =20
This bit of magic is "udev". It gets messages from the kernel when
drivers are loaded, and it creates device nodes for you. When using
udev, the /dev filesystem is actually created in a ramdisk, which
explains why the device nodes disappeared when you rebooted.

  Now I need to know how to make more Frame Buffer devices!
   =20
Typically, you get one frame buffer device node per video card.
(/dev/fb is probably a link to /dev/fb0)