Bug#401685: Bug#394615: More testing needed

[Attilio Fiandrotti <attilio.fiandrotti@gmail.com>]

Fabrice LORRAIN wrote:

<snip/>

> Follow-up from #394615 sent to #401684.
>
> using iso [1] from yesterday :
>
> cat /proc/fb
> 0 VGA16 VGA
>
> ls -la /dev/fb*
>
> lrwxrwxrwx    1 root     root            4 Dec  6 20:11 /dev/fb0 -> fb/0
>
> /dev/fb:
> crw-rw----    1 root     root      29,   0 Dec  6 20:11 0
>
> grep -i fb syslog
> Dec  6 20:11:48 kernel:  BIOS-e820: 00000000ffb00000 - 0000000100000000 (reserved)
> Dec  6 20:11:50 kernel: vga16fb: initializing
> Dec  6 20:11:50 kernel: vga16fb: mapped to 0xc00a0000
> Dec  6 20:11:50 kernel: fb0: VGA16 VGA frame buffer device
>
> Attaching syslog.gz
>
> I did a test with [2] but I got the following :
> ...DirectFB/FBDev: Error opening '/dev/fb'!
>    --> Is a directory
>
> Hmmm, cat /proc/fb
> 0 VGA16 VGA
>
> No vesa here.

Fabrice, fisrt let me thank you for testing, but i believe the bad news 
is that you'll never be able to run graphical d-i on your Optiplex box.
Googling around i first found many posts from linux and *bsd users 
failing at starting vesafb on their 82810, then i found out [1] that 
vesafb on 82810 simply doesn't work.
Infact, looking at the syslog you provided it looks like vesafb wasn't 
started at all.
We have to handle cases where the vesafb could not be started by falling 
back to newt frontend.

cheers

Attilio

[1] http://www.intel.com/support/graphics/intel810/sb/cs-004014.htm