> Latest working kernel version:2.6.25.4
> Earliest failing kernel version:2.6.25 and up

That can't be right.

Which was the most recent correctly-working kernel version?

Excuse me, I was sealled up.
Latest working kernel version testing me:2.6.24.4

Reply-To: akpm@linux-foundation.org


(switched to email.  Please respond via emailed reply-to-all, not via the
bugzilla web interface).

On Tue, 20 May 2008 22:55:43 -0700 (PDT) bugme-daemon@bugzilla.kernel.org wrote:

> http://bugzilla.kernel.org/show_bug.cgi?id=10766
> 
>            Summary: ceased to work framebuffer with videocard NVIDIA GeForce
>                     MX 440
>            Product: Drivers
>            Version: 2.5
>      KernelVersion: 2.6.25
>           Platform: All
>         OS/Version: Linux
>               Tree: Mainline
>             Status: NEW
>           Severity: normal
>           Priority: P1
>          Component: Console/Framebuffers
>         AssignedTo: jsimmons@infradead.org
>         ReportedBy: ashirapov.denis@mail.ru
> 
> 
> Latest working kernel version:2.6.25.4
> Earliest failing kernel version:2.6.25 and up
> Distribution: Slackware GNU/Linux 12.0 
> Hardware Environment:
>  Processors: AMD Athlon XP 1800(Thunderbird)
>  Video card: Nvidia GeForce MX 440 agp 8x
>  Mother board: Biostar M7NCD (nForce2)
> Software Environment:
> Problem Description: 
>  In kernels >= 2.6.25 loading kernel with append patameters in boot
> loaders(grub) as "vga=317" etc. can't enable requred video mode.
> 
> Steps to reproduce: 
> Compile ketnel 2.6.25 with config:
> http://slackware.perespim.ru/pub/slackware-current/kernels/huge.s/config
> 

Can anyone think of anything we might have done between 2.6.24 and
2.6.25 to break the nvidia driver's boot-command-line handling?

Thanks.

The Night to contribute little clarity in question.
Bag reveals itself in kernel 2.6.25 and above only.
If compiled "vanila" kernel with config including standard "vesafb" in kernel and "nvidiafb module, that does not be fixed videomode vga=0x317 (and other too).
But, sometimes after several reboot and shutdown "vesafb" is work. After like boot in next time again does not work... With other testing me videocard as GeForce2 MX 400 all work correctly. 
Change the files of source kernel:
arch/x86/boot/video-bios.c  
arch/x86/boot/video-vesa.c 
arch/x86/boot/video-vga.c  
arch/x86/boot/video.c  
arch/x86/boot/video.h
version from kernel 2.6.24 problem at all does not solve.

If compiled "vanila" kernel with config including "nvidiafb" in kernel and give to the kernel addition parameters "video=nvidiafb:1024x768-16@75" then framebuffer is work.

Reply-To: akpm@linux-foundation.org

On Wed, 28 May 2008 18:50:06 -0700 (PDT) bugme-daemon@bugzilla.kernel.org wrote:

> http://bugzilla.kernel.org/show_bug.cgi?id=10766
> 
> 
> 
> 
> 
> ------- Comment #4 from ashirapov.denis@mail.ru  2008-05-28 18:50 -------
> The Night to contribute little clarity in question.
> Bag reveals itself in kernel 2.6.25 and above only.
> If compiled "vanila" kernel with config including standard "vesafb" in kernel
> and "nvidiafb module, that does not be fixed videomode vga=0x317 (and other
> too).
> But, sometimes after several reboot and shutdown "vesafb" is work. After like
> boot in next time again does not work... With other testing me videocard as
> GeForce2 MX 400 all work correctly. 
> Change the files of source kernel:
> arch/x86/boot/video-bios.c  
> arch/x86/boot/video-vesa.c 
> arch/x86/boot/video-vga.c  
> arch/x86/boot/video.c  
> arch/x86/boot/video.h
> version from kernel 2.6.24 problem at all does not solve.
> 
> If compiled "vanila" kernel with config including "nvidiafb" in kernel and
> give
> to the kernel addition parameters "video=nvidiafb:1024x768-16@75" then
> framebuffer is work.

OK, thanks but please...

: (switched to email.  Please respond via emailed reply-to-all, not via the
: bugzilla web interface).

Let's handle this bug via email from now on so the linux-fbdev-devel
people get to see it.

Reply-To: krzysztof.h1@poczta.fm

On Wed, 28 May 2008 19:15:26 -0700
Andrew Morton <akpm@linux-foundation.org> wrote:

> On Wed, 28 May 2008 18:50:06 -0700 (PDT) bugme-daemon@bugzilla.kernel.org
> wrote:
> 
> > http://bugzilla.kernel.org/show_bug.cgi?id=10766
> > 
> > 
> > 
> > 
> > 
> > ------- Comment #4 from ashirapov.denis@mail.ru  2008-05-28 18:50 -------
> > The Night to contribute little clarity in question.
> > Bag reveals itself in kernel 2.6.25 and above only.
> > If compiled "vanila" kernel with config including standard "vesafb" in
> kernel
> > and "nvidiafb module, that does not be fixed videomode vga=0x317 (and other
> > too).
> > But, sometimes after several reboot and shutdown "vesafb" is work. After
> like
> > boot in next time again does not work... With other testing me videocard as
> > GeForce2 MX 400 all work correctly. 
> > Change the files of source kernel:
> > arch/x86/boot/video-bios.c  
> > arch/x86/boot/video-vesa.c 
> > arch/x86/boot/video-vga.c  
> > arch/x86/boot/video.c  
> > arch/x86/boot/video.h
> > version from kernel 2.6.24 problem at all does not solve.
> > 
> > If compiled "vanila" kernel with config including "nvidiafb" in kernel and
> give
> > to the kernel addition parameters "video=nvidiafb:1024x768-16@75" then
> > framebuffer is work.
> 
> OK, thanks but please...
> 

I have handled this bug with Denis, so his additional findings are results from this.
I have tried to reproduce it but I do not have the hardware he has.
It is a bug in vesafb or VESA mode setting. 
I am not able to reproduce this bug on Trident card, 3dfx card, Permedia 2,
ATI on-board card (all using vesafb driver).

The current status is that I will try to find someone with 
the NVidia GeForce 4 MX 4x0 card and borrow it. I hope that
Denis' card has some kind of reference board BIOS and not 
BIOS modified by the card's vendor.

Regards,
Krzysztof

----------------------------------------------------------------------
Tanie rozmowy!
Sprawdz >>> http://link.interia.pl/f1e13  

May be I'am can else help with this bug?
Full name videocard: "GeForce 4 MX 440 agp  8x 64Mb ".
The manufacturer by "Palit" if i not mistake.
BIOS  I shall look the version in the evening.
Can I shall be able to conduct the row a test if you will help me...

Reply-To: krzysztof.h1@poczta.fm

On Tue, 10 Jun 2008 07:27:30 +0500
"Аширапов Д." <ashirapov.denis@mail.ru> wrote:

> Hello Krzysztof Helt, 
> 
> 
> 
> >Can you email me the .config file from the 2.6.25 kernel you use to trigger
> the bug?
> 
> Config of the 2.6.25 kernel in attached file. 
> 
> 

Hi Denis,

I have not good news.

I have finally borrowed the card designed as GF4MX440-8X (Geforce MX 440  AGP 8X), but I could not reproduce the bug.
I used the config file you sent me.
I tried also on the Nvidia TNT card and S3 Savage4 card (Number9 SR9). The vesafb worked for both.

The dmesg from my computer with Geforce MX 440 card is attached. The interesting part is:

pci 0000:01:00.0: Boot video device
pci 0000:02:02.0: Firmware left e100 interrupts enabled; disabling
pci_hotplug: PCI Hot Plug PCI Core version: 0.5
vesafb: framebuffer at 0xf0000000, mapped to 0xe0a80000, using 3072k, total 65536k
vesafb: mode is 1024x768x16, linelength=2048, pages=1
vesafb: protected mode interface info at c000:ef40
vesafb: pmi: set display start = c00cef76, set palette = c00cefe0
vesafb: pmi: ports = 3b4 3b5 3ba 3c0 3c1 3c4 3c5 3c6 3c7 3c8 3c9 3cc 3ce 3cf 3d0 3d1 3d2 3d3 3d4 3d5 3da
vesafb: scrolling: redraw
vesafb: Truecolor: size=0:5:6:5, shift=0:11:5:0
Console: switching to colour frame buffer device 128x48
fb0: VESA VGA frame buffer device

As I am not able to reproduce the bug, please test if you can connect to the computer with black screen (e.g. from another one through ssh). I used the 791 constant.
Can you dump the dmesg with disabled screen and to me?

Regards,
Krzysztof


----------------------------------------------------------------------
Tania telefonia internetowa!
Sprawdz >>>  http://link.interia.pl/f1e2e

Is this bug a issue still? Have you tried the nouveau driver?

Bug is still alive and appears only in vesafb with a connected USB flash device only on my motherboard.

This bug report title looks wrong as I have this videocard but vesafb boots and works correctly (except that it conflicts fatally with nVidia binary driver on PAT enabled kernels). Does this problem be limited with a specified resolution only, right?

I can't reproduce this bug using vga=0x317 because isn't recognized as a valid entry. I can reproduce a blank screen if my system boots with vga=317 (0x13D) though, another not listed value...