Debian Bug report logs - #535026
vga= not working with packed color modes on some cards

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Jos van Wolput <wolput@onsneteindhoven.nl>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: KMS not working after booting with grub-pc 1.96+20090628-1 (unstable)

Date: Mon, 29 Jun 2009 11:09:33 +0800

Package: grub-pc
Version: 1.96+20090628-1 (unstable)
Severity: important
System: Debian Sid
Kernel: Linux 2.6.30-rc8-686 (kernel-archive.buildserver.net/debian-kernel)
Processor: Pentium M, 852GM/855GM

I am using KMS with module i915 modeset=1.
When booting with the latest version  of grub-pc (20090628), after loading module i915 modeset=1 the booting process
is no longer shown on the screen until gdm is loaded.
VT screens 1-6 are all showing the gdm login screen although without mouse pointer and no keyboard input.
Only VT screen 7 works normally.

When booting with version grub-pc_1.96+20090611-1, the screen is set to high resolution and KMS works perfectly.

Regards,
Jos van Wolput

Message #8 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: wolput@onsneteindhoven.nl, 535026@bugs.debian.org

Subject: Re: Bug#535026: KMS not working after booting with grub-pc 1.96+20090628-1 (unstable)

Date: Tue, 30 Jun 2009 00:54:52 +0200

forcemerge 534944 535026
thanks

On Mon, Jun 29, 2009 at 11:09:33AM +0800, Jos van Wolput wrote:
> Package: grub-pc
> Version: 1.96+20090628-1 (unstable)
> Severity: important
> System: Debian Sid
> Kernel: Linux 2.6.30-rc8-686 (kernel-archive.buildserver.net/debian-kernel)
> Processor: Pentium M, 852GM/855GM
>
> I am using KMS with module i915 modeset=1.
> When booting with the latest version  of grub-pc (20090628), after loading module i915 modeset=1 the booting process
> is no longer shown on the screen until gdm is loaded.
> VT screens 1-6 are all showing the gdm login screen although without mouse pointer and no keyboard input.
> Only VT screen 7 works normally.
>
> When booting with version grub-pc_1.96+20090611-1, the screen is set to high resolution and KMS works perfectly.

Hi,

Same thing I said in #534944:

I'm uploading 1.96+20090629-1, which includes some cleanup in the Linux
loader.  It's possible your problem might disappear with this version,
please report.

Note: Try it in text mode first (run "terminal_output console" in GRUB
prompt).  If that _does_ work, try from gfxterm then.

Note2: If you can't wait, the patch you want is:
  svn diff -r 2372:2375 svn://svn.savannah.gnu.org/grub/trunk/grub2

Thanks!

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #15 received at 535026-done@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: 535026-done@bugs.debian.org

Cc: wolput@onsneteindhoven.nl, Luca Tettamanti <kronos.it@gmail.com>, Arthur Marsh <arthur.marsh@internode.on.net>

Subject: fixed

Date: Tue, 30 Jun 2009 02:30:39 +0200

Version: 1.96+20090629-1

Arthur reported on IRC that my latest fixes (now in 1.96+20090629-1) resolved
this problem.  Please confirm if you can, thanks.

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #25 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Jos van Wolput <wolput@sohu.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org

Subject: Re: Bug#535026: KMS not working after booting with grub-pc v.1.96+20090628-1 (unstable)

Date: Tue, 30 Jun 2009 15:00:09 +0800

Robert Millan wrote:
> forcemerge 534944 535026
> thanks
>
> On Mon, Jun 29, 2009 at 11:09:33AM +0800, Jos van Wolput wrote:
>   
>> Package: grub-pc
>> Version: 1.96+20090628-1 (unstable)
>> Severity: important
>> System: Debian Sid
>> Kernel: Linux 2.6.30-rc8-686 (kernel-archive.buildserver.net/debian-kernel)
>> Processor: Pentium M, 852GM/855GM
>>
>> I am using KMS with module i915 modeset=1.
>> When booting with the latest version  of grub-pc (20090628), after loading module i915 modeset=1 the booting process
>> is no longer shown on the screen until gdm is loaded.
>> VT screens 1-6 are all showing the gdm login screen although without mouse pointer and no keyboard input.
>> Only VT screen 7 works normally.
>>
>> When booting with version grub-pc_1.96+20090611-1, the screen is set to high resolution and KMS works perfectly.
>>     
>
> Hi,
>
> Same thing I said in #534944:
>
> I'm uploading 1.96+20090629-1, which includes some cleanup in the Linux
> loader.  It's possible your problem might disappear with this version,
> please report.
>
> Note: Try it in text mode first (run "terminal_output console" in GRUB
> prompt).  If that _does_ work, try from gfxterm then.
>
> Note2: If you can't wait, the patch you want is:
>   svn diff -r 2372:2375 svn://svn.savannah.gnu.org/grub/trunk/grub2
>
> Thanks!
>
>   
I just installed v. 1.96+20090629-1. Unfortunately it doesn't work. The 
bug is still the same as I
previously mentioned. I have to revert to v. 1.96+20090611-1.

Jos v. Wolput

-- 
Jos VAN WOLPUT & JIN Yuehua 金月花
Tongelresestraat 204 A, 5613 DS Eindhoven, The Netherlands
P.R.China, 310052 Hangzhou, Binjiang Qu, Jianghan Lu, Zhongxing Huayuan 49-1-1003
中国310052杭州市滨江去江汉路中兴花园49-1-1003
--
phone: 00 31 40 2449399 / 00 86 571 81713287 - 85054977
e-mail: wolput@onsneteindhoven.nl / wolput@sohu.com
website: http://homepages.onsneteindhoven.nl/~wolput
-----
Debian GNU/Linux: the Universal Operating System

Message #28 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Jos van Wolput <wolput@onsneteindhoven.nl>

Cc: 535026@bugs.debian.org

Subject: Re: Bug#535026: KMS not working after booting with grub-pc v.1.96+20090628-1 (unstable)

Date: Tue, 30 Jun 2009 15:49:41 +0200

On Tue, Jun 30, 2009 at 03:11:29PM +0800, Jos van Wolput wrote:
>>   
> I just installed v. 1.96+20090629-1. Unfortunately it doesn't work. The
> bug is still the same as I
> previously mentioned. I have to revert to v. 1.96+20090611-1.

Did you run grub-install after the update?

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #31 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Jos van Wolput <wolput@onsneteindhoven.nl>

Cc: 535026@bugs.debian.org

Subject: Re: Bug#535026: KMS not working after booting with grub-pc v.1.96+20090628-1 (unstable)

Date: Tue, 30 Jun 2009 15:51:59 +0200

On Tue, Jun 30, 2009 at 03:49:41PM +0200, Robert Millan wrote:
> On Tue, Jun 30, 2009 at 03:11:29PM +0800, Jos van Wolput wrote:
> >>   
> > I just installed v. 1.96+20090629-1. Unfortunately it doesn't work. The
> > bug is still the same as I
> > previously mentioned. I have to revert to v. 1.96+20090611-1.
> 
> Did you run grub-install after the update?

Btw, were you using text mode when you tested this?
("terminal_output console" in GRUB command-line)

Does the problem disappear if you disable KMS?

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #38 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Luca Tettamanti <kronos.it@gmail.com>

Cc: 535026@bugs.debian.org

Subject: Re: Bug#534944: grub-pc: text mode virtual terminals blank

Date: Tue, 30 Jun 2009 18:40:13 +0200

On Tue, Jun 30, 2009 at 04:33:37PM +0200, Luca Tettamanti wrote:
> On Mon, Jun 29, 2009 at 7:28 PM, Robert Millan<rmh@aybabtu.com> wrote:
> > Hi again,
> >
> > I'm uploading 1.96+20090629-1, which includes some cleanup in the Linux
> > loader.  It's possible your problem might disappear with this version,
> > please report.
> 
> No joy.

Hi,

Then your bug is #535026.  #534944 was confirmed to be fixed by Arthur.  I'm
changing CC to 535026@bugs.debian.org.

> I see the message that "vga" parameter is deprecated. The mode I'm
> using is 773 (1024x768x256).

Ok.  That's 0x305 aka 1024x768x8, which is supposed to be detected & handled
already.

> This is the output of a patched kernel, loaded with linux16:
> 
> orig_x = 0x0, orig_y =  0x0
> ext_mem_k = 0xffff
> orig_video_page = 0x0
> orig_video_mode = 0x3
> orig_video_cols = 0x50
> orig_video_ega_bx = 0x3
> orig_video_lines = 0x19
> orig_video_isVGA = 0x1
> orig_video_points = 0x10
> lfb_width / lfb_height / lfb_depth = 0x0x0
> lfb_base = 0x0, lfb_size = 0x0
> cl_magic = 0x0, cl_offset = 0x0
> lfb_linelength = 0x0\n
> red_size = 0x0, red_pos = 0x0
> green_size = 0x0, green_pos = 0x0
> blue_size = 0x0, blue_pos = 0x0
> rsvd_size = 0x0, rsvd_pos = 0x0
> vesapm = 0x0:0x0
> pages = 0x0
> vesa_attributes = 0x0
> capabilities = 0x0

This looks like text mode.  I assume you didn't pass "vga=773" to linux16?
I need you to include that parameter in your test.

Thanks!

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #43 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org

Subject: Re: Bug#534944: grub-pc: text mode virtual terminals blank

Date: Tue, 30 Jun 2009 21:36:16 +0200

On Tue, Jun 30, 2009 at 6:40 PM, Robert Millan<rmh@aybabtu.com> wrote:
> On Tue, Jun 30, 2009 at 04:33:37PM +0200, Luca Tettamanti wrote:
>> On Mon, Jun 29, 2009 at 7:28 PM, Robert Millan<rmh@aybabtu.com> wrote:
>> > Hi again,
>> >
>> > I'm uploading 1.96+20090629-1, which includes some cleanup in the Linux
>> > loader.  It's possible your problem might disappear with this version,
>> > please report.
>>
>> No joy.
>
> Hi,
>
> Then your bug is #535026.  #534944 was confirmed to be fixed by Arthur.  I'm
> changing CC to 535026@bugs.debian.org.

I have radeon mobility (M76) card and I'm not using KMS. I also don't
get stale content on text VTs (just black screen).

>> I see the message that "vga" parameter is deprecated. The mode I'm
>> using is 773 (1024x768x256).
>
> Ok.  That's 0x305 aka 1024x768x8, which is supposed to be detected & handled
> already.
>
>> This is the output of a patched kernel, loaded with linux16:
>>
>> orig_x = 0x0, orig_y =  0x0
>> ext_mem_k = 0xffff
>> orig_video_page = 0x0
>> orig_video_mode = 0x3
>> orig_video_cols = 0x50
>> orig_video_ega_bx = 0x3
>> orig_video_lines = 0x19
>> orig_video_isVGA = 0x1
>> orig_video_points = 0x10
>> lfb_width / lfb_height / lfb_depth = 0x0x0
>> lfb_base = 0x0, lfb_size = 0x0
>> cl_magic = 0x0, cl_offset = 0x0
>> lfb_linelength = 0x0\n
>> red_size = 0x0, red_pos = 0x0
>> green_size = 0x0, green_pos = 0x0
>> blue_size = 0x0, blue_pos = 0x0
>> rsvd_size = 0x0, rsvd_pos = 0x0
>> vesapm = 0x0:0x0
>> pages = 0x0
>> vesa_attributes = 0x0
>> capabilities = 0x0
>
> This looks like text mode.  I assume you didn't pass "vga=773" to linux16?
> I need you to include that parameter in your test.

Ops, right:

orig_x = 0x0, orig_y =  0x1
ext_mem_k = 0xffff
orig_video_page = 0x0
orig_video_mode = 0x3
orig_video_cols = 0x50
orig_video_ega_bx = 0x3
orig_video_lines = 0x19
orig_video_isVGA = 0x23
orig_video_points = 0x10
lfb_width / lfb_height / lfb_depth = 1024x768x8
lfb_base = 0xe0000000, lfb_size = 0x100
cl_magic = 0x0, cl_offset = 0x0
lfb_linelength = 0x400
red_size = 0x8, red_pos = 0x0
green_size = 0x8, green_pos = 0x0
blue_size = 0x8, blue_pos = 0x0
rsvd_size = 0x8, rsvd_pos = 0x0
vesapm = 0xc000:0x9eda
pages = 0x12
vesa_attributes = 0xbb
capabilities = 0x0

vesafb is not picking up the right options:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc20010100000, using
768k, total 768k
vesafb: mode is 1024x768x8, linelength=1024, pages=0
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=0:0:0:0, shift=0:0:0:0

both pages and Pseudocolor values look wrong. With older grub I get:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc20010100000, using
1536k, total 16384k
vesafb: mode is 1024x768x8, linelength=1024, pages=18
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=8:8:8:8, shift=0:0:0:0

Luca

Message #46 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: 535026@bugs.debian.org

Cc: Jos van Wolput <wolput@onsneteindhoven.nl>, Luca Tettamanti <kronos.it@gmail.com>

Subject: patch, please test

Date: Wed, 1 Jul 2009 03:54:21 +0200

[Message part 1 (text/plain, inline)]

Hi,

Please could you test if this patch helps?

Thanks

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

[lfb_size.diff (text/x-diff, attachment)]

Message #53 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org, Jos van Wolput <wolput@onsneteindhoven.nl>

Subject: Re: patch, please test

Date: Wed, 1 Jul 2009 23:00:59 +0200

On Wed, Jul 1, 2009 at 3:54 AM, Robert Millan<rmh@aybabtu.com> wrote:
> Please could you test if this patch helps?

It doesn't work. The output of your patched kernel is unchanged while
vesafb picks a different memory size:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc20010100000, using
1536k, total 16384k
vesafb: mode is 1024x768x8, linelength=1024, pages=0
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=0:0:0:0, shift=0:0:0:0

The color data is still wrong though. screen_info is copied from boot
params, which your patched kernel dumps correctly... wtf?

Luca

Message #60 received at 535026-close@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh.debian@aybabtu.com>

To: 535026-close@bugs.debian.org

Subject: Bug#535026: fixed in grub2 1.96+20090702-1

Date: Thu, 02 Jul 2009 12:47:08 +0000

Source: grub2
Source-Version: 1.96+20090702-1

We believe that the bug you reported is fixed in the latest version of
grub2, which is due to be installed in the Debian FTP archive:

grub-common_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-common_1.96+20090702-1_amd64.deb
grub-coreboot_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-coreboot_1.96+20090702-1_amd64.deb
grub-efi-amd64_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-efi-amd64_1.96+20090702-1_amd64.deb
grub-efi-ia32_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-efi-ia32_1.96+20090702-1_amd64.deb
grub-efi_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-efi_1.96+20090702-1_amd64.deb
grub-firmware-qemu_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-firmware-qemu_1.96+20090702-1_amd64.deb
grub-ieee1275_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-ieee1275_1.96+20090702-1_amd64.deb
grub-linuxbios_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-linuxbios_1.96+20090702-1_amd64.deb
grub-pc_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-pc_1.96+20090702-1_amd64.deb
grub-rescue-pc_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub-rescue-pc_1.96+20090702-1_amd64.deb
grub2_1.96+20090702-1.diff.gz
  to pool/main/g/grub2/grub2_1.96+20090702-1.diff.gz
grub2_1.96+20090702-1.dsc
  to pool/main/g/grub2/grub2_1.96+20090702-1.dsc
grub2_1.96+20090702-1_amd64.deb
  to pool/main/g/grub2/grub2_1.96+20090702-1_amd64.deb
grub2_1.96+20090702.orig.tar.gz
  to pool/main/g/grub2/grub2_1.96+20090702.orig.tar.gz



A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 535026@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Robert Millan <rmh.debian@aybabtu.com> (supplier of updated grub2 package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@debian.org)


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.8
Date: Thu, 02 Jul 2009 13:23:51 +0200
Source: grub2
Binary: grub2 grub-linuxbios grub-efi grub-common grub-pc grub-rescue-pc grub-coreboot grub-efi-ia32 grub-efi-amd64 grub-ieee1275 grub-firmware-qemu
Architecture: source amd64
Version: 1.96+20090702-1
Distribution: unstable
Urgency: low
Maintainer: GRUB Maintainers <pkg-grub-devel@lists.alioth.debian.org>
Changed-By: Robert Millan <rmh.debian@aybabtu.com>
Description: 
 grub-common - GRand Unified Bootloader, version 2 (common files)
 grub-coreboot - GRand Unified Bootloader, version 2 (Coreboot version)
 grub-efi   - GRand Unified Bootloader, version 2 (dummy package)
 grub-efi-amd64 - GRand Unified Bootloader, version 2 (EFI-AMD64 version)
 grub-efi-ia32 - GRand Unified Bootloader, version 2 (EFI-IA32 version)
 grub-firmware-qemu - GRUB firmware image for QEMU
 grub-ieee1275 - GRand Unified Bootloader, version 2 (Open Firmware version)
 grub-linuxbios - GRand Unified Bootloader, version 2 (dummy package)
 grub-pc    - GRand Unified Bootloader, version 2 (PC/BIOS version)
 grub-rescue-pc - GRUB bootable rescue images, version 2 (PC/BIOS version)
 grub2      - GRand Unified Bootloader, version 2 (dummy package)
Closes: 535026
Changes: 
 grub2 (1.96+20090702-1) unstable; urgency=low
 .
   * New SVN snapshot.
   * rules: Remove duplicated files in sparc64-ieee1275 port.
   * rules: Comment out -DGRUB_ASSUME_LINUX_HAS_FB_SUPPORT=1 setting.  We'll
     re-evaluate using it when it's more mature.  (Closes: #535026).
Checksums-Sha1: 
 b0af650aaea96628fde0c2b568112a066da490d3 1789 grub2_1.96+20090702-1.dsc
 d32a3a4ed5e023969553ccdaaca7c6f0dfd22ffe 1156404 grub2_1.96+20090702.orig.tar.gz
 d5129dd016f675a32d496584859bc7df86e01117 55034 grub2_1.96+20090702-1.diff.gz
 5c233b8b27a53bd593790a7d0dd488afaa658c1c 176308 grub2_1.96+20090702-1_amd64.deb
 cdb169dcacb0a5c3295273f76f9a11cfb98686f7 175204 grub-linuxbios_1.96+20090702-1_amd64.deb
 96c096aee83a425ea583d38e8e38e43c98c95401 175202 grub-efi_1.96+20090702-1_amd64.deb
 dec8662d29383e4a2c39fe2b05dd4768950d2951 1199746 grub-common_1.96+20090702-1_amd64.deb
 d7e2fbafdede5c1f4dfdcc0d645b282f38d2bb73 630274 grub-pc_1.96+20090702-1_amd64.deb
 e24409efadf93ec5888766733a7756e6df64ac11 759736 grub-rescue-pc_1.96+20090702-1_amd64.deb
 41c072e4a9e7ae8da75586c33d06407702ed2b60 424254 grub-coreboot_1.96+20090702-1_amd64.deb
 0b5456db5491b41733de42948a1f85190a51c07b 450608 grub-efi-ia32_1.96+20090702-1_amd64.deb
 d0a2b666803646dac809ec19bfde659c83cf01ff 505818 grub-efi-amd64_1.96+20090702-1_amd64.deb
 b92dd2116271bb5654bcd24f331fcc6db40cc60f 420344 grub-ieee1275_1.96+20090702-1_amd64.deb
 676188e2b579ba8e8e7b8aabe2c4bcf035a3e259 399922 grub-firmware-qemu_1.96+20090702-1_amd64.deb
Checksums-Sha256: 
 d3c1dbf7a139af354c2d014067d2bcb7351e44171da0e039edf48c9880e1fb40 1789 grub2_1.96+20090702-1.dsc
 9a4f45055a76ba8dac8748d3ad604907311df7a5a3e357c43c0ffcfdb15fd1be 1156404 grub2_1.96+20090702.orig.tar.gz
 63104758ffe60353d2b63a5f2ae615bd62d6e838eeff5238a4cefcfd29ca6bf7 55034 grub2_1.96+20090702-1.diff.gz
 0b7b785ae3c7ca1859e9e930f95f4a361d43dde0c7af83da9248521bf6251876 176308 grub2_1.96+20090702-1_amd64.deb
 71dfc1a0acbadbdf24227f25ab89ab4ee34384b9a19aefe8742511ff56912f19 175204 grub-linuxbios_1.96+20090702-1_amd64.deb
 e347222a85a6ba48c3265b0226f22a2ea0a32f7857188fcca64a692165316326 175202 grub-efi_1.96+20090702-1_amd64.deb
 f388c782485d1a708a55cf901559555fd5303d146f2191752cb5bfc2be66115f 1199746 grub-common_1.96+20090702-1_amd64.deb
 80f839df5a75f81057225e0d99ac38ad625ac836cc96986b509dfd21fdc9bfaf 630274 grub-pc_1.96+20090702-1_amd64.deb
 8b62c2f2ecbac9f54c3a6aa83f8e65520df206671f83a2cb7da093222d787331 759736 grub-rescue-pc_1.96+20090702-1_amd64.deb
 d13d7f6ada0c2097c2c4b46ab26b263ed3a252412bd5bb0541038bd4910698b5 424254 grub-coreboot_1.96+20090702-1_amd64.deb
 2a0f88805733d5d23b9d2c5d4f4f1dfa7d4dce750d0e3f2d1d37e883b6d1aabf 450608 grub-efi-ia32_1.96+20090702-1_amd64.deb
 129a793fc768a12696cdab2dac8f7e91a11d7fa6ba87bd8f2ce2de87ecdc7d1d 505818 grub-efi-amd64_1.96+20090702-1_amd64.deb
 bb4d97458981b1663cbbd55bce4d3e4647b6938d95d2b75bba79ef846c3e9aeb 420344 grub-ieee1275_1.96+20090702-1_amd64.deb
 c7653a6e102eac1264ae005c1891a997e03d092308ef70e96aae90b7f8d62e59 399922 grub-firmware-qemu_1.96+20090702-1_amd64.deb
Files: 
 23a2b86053eab649ecbee0e2bddd7acc 1789 admin extra grub2_1.96+20090702-1.dsc
 a8255a739bf8837a5400c604ff3d7e4d 1156404 admin extra grub2_1.96+20090702.orig.tar.gz
 cbd9b923ea36e26c64851e8e23127bfb 55034 admin extra grub2_1.96+20090702-1.diff.gz
 0cec07840460726fec924befd0389eb9 176308 admin extra grub2_1.96+20090702-1_amd64.deb
 9dc8a224c841beac0abaffd42d27d6e3 175204 admin extra grub-linuxbios_1.96+20090702-1_amd64.deb
 d09295adda4215a6ef5282d72b55b5ce 175202 admin extra grub-efi_1.96+20090702-1_amd64.deb
 16a464deb1c2d5f5786a2bd1a07e01d2 1199746 admin extra grub-common_1.96+20090702-1_amd64.deb
 31e76facbbe92fd2141f978046a4b26e 630274 admin extra grub-pc_1.96+20090702-1_amd64.deb
 6cd74fd775c3c427eb35f01b85c1b45d 759736 admin extra grub-rescue-pc_1.96+20090702-1_amd64.deb
 2cf67e8a8019880a8b81d4fc1ff7b0f2 424254 admin extra grub-coreboot_1.96+20090702-1_amd64.deb
 bcf3248b0fc793ea6e3ba70795a196ee 450608 admin extra grub-efi-ia32_1.96+20090702-1_amd64.deb
 41ce485a0aa4d524576e4f174d143a5e 505818 admin extra grub-efi-amd64_1.96+20090702-1_amd64.deb
 ba17a6c0f8a3455838c7165bb22be114 420344 admin extra grub-ieee1275_1.96+20090702-1_amd64.deb
 b86ac5da363530053e0be64088585488 399922 admin extra grub-firmware-qemu_1.96+20090702-1_amd64.deb

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkpMqbsACgkQC19io6rUCv9uAwCePyxtpgBxkMigfFgEm65+R26z
KVQAnRcTP+GXqwXUwOrXx3ySt/sp33lq
=Yt1t
-----END PGP SIGNATURE-----

Message #65 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: 535026@bugs.debian.org

Cc: Robert Millan <rmh@aybabtu.com>

Subject: Re: patch, please test

Date: Fri, 3 Jul 2009 18:10:01 +0200

The last update to grub didn't fix the bug for me, I still have a blank FB.
vesafb still reports bogus information:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
768k, total 768k
vesafb: mode is 1024x768x8, linelength=1024, pages=0
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=0:0:0:0, shift=0:0:0:0

Luca

Message #70 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Luca Tettamanti <kronos.it@gmail.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Sat, 4 Jul 2009 16:16:53 +0200

[Message part 1 (text/plain, inline)]

On Fri, Jul 03, 2009 at 06:10:01PM +0200, Luca Tettamanti wrote:
> The last update to grub didn't fix the bug for me, I still have a blank FB.
> vesafb still reports bogus information:
> 
> vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
> 768k, total 768k
> vesafb: mode is 1024x768x8, linelength=1024, pages=0
> vesafb: scrolling: redraw
> vesafb: Pseudocolor: size=0:0:0:0, shift=0:0:0:0

Please tell me the output of:

  A) A *patched* Linux being loaded by "linux16" command, using the
     patch in:
     http://lists.gnu.org/archive/html/grub-devel/2009-06/txteL9T65YdVL.txt

     This could be what you already provided in
     http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=535026#43
     except your followup comment "vesafb is not picking up the right options"
     contradicts that (linux16 *is* the old loader, and it is supposed to
     work.  If linux16 doesn't work, we have an entirely different problem).

  B) Linux (patched or not), being loaded by "linux" command, using the
     GRUB patch in:
     http://lists.gnu.org/archive/html/grub-devel/2009-06/txtFlQanGk4Pv.txt
     *plus* the patch I'm attaching.

In both cases, I need both the screen_info dump and (if available) vesafb
output.

Btw, you can also contact me on IRC (nyu @ irc.freenode.net / #grub).

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

[capabilities.diff (text/x-diff, attachment)]

Message #77 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Jos van Wolput <wolput@onsneteindhoven.nl>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org, Luca Tettamanti <kronos.it@gmail.com>

Subject: Re: patch, please test

Date: Tue, 07 Jul 2009 22:58:42 +0800

Robert Millan wrote:
> Hi,
>
> Please could you test if this patch helps?
>
> Thanks
>   
I am sorry I can no longer test while I got a disk crash!

Jos v.W. 

Message #82 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Julien Langer <julien.langer@googlemail.com>

To: 535026@bugs.debian.org

Subject: Re: vga= not working on some cards

Date: Fri, 17 Jul 2009 19:14:50 +0200

I'm also not getting any VGA output when booting with "linux"
"linux16" works fine though.
I'm using vesafb on an nvidia geforce 8800GT

-Julien

Message #87 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Julien Langer <julien.langer@googlemail.com>

To: 535026@bugs.debian.org

Subject: Re: vga= not working on some cards

Date: Fri, 17 Jul 2009 19:17:07 +0200

[Message part 1 (text/plain, inline)]

2009/7/17 Julien Langer <julien.langer@googlemail.com>

> I'm also not getting any VGA output when booting with "linux"
> "linux16" works fine though.
> I'm using vesafb on an nvidia geforce 8800GT


Sorry, forgot to mention that this happens with grub-pc 1.96+20090709-1


>
> -Julien
>
>

[Message part 2 (text/html, inline)]

Message #90 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Julien Langer <julien.langer@googlemail.com>, 535026@bugs.debian.org

Subject: Re: Bug#535026: vga= not working on some cards

Date: Sat, 18 Jul 2009 01:03:54 +0200

On Fri, Jul 17, 2009 at 07:14:50PM +0200, Julien Langer wrote:
> I'm also not getting any VGA output when booting with "linux"
> "linux16" works fine though.
> I'm using vesafb on an nvidia geforce 8800GT

Hi,

Unfortunately I'm unable to reproduce this bug.  I need some feedback
before I can fix it.  Please see:

  http://people.debian.org/~rmh/grub/vga_bug/README

for instructions on how you can help.

Thanks in advance

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #95 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Tue, 21 Jul 2009 22:07:05 +0200

On Sat, Jul 4, 2009 at 4:16 PM, Robert Millan<rmh@aybabtu.com> wrote:
> On Fri, Jul 03, 2009 at 06:10:01PM +0200, Luca Tettamanti wrote:
>> The last update to grub didn't fix the bug for me, I still have a blank FB.
>> vesafb still reports bogus information:
>>
>> vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
>> 768k, total 768k
>> vesafb: mode is 1024x768x8, linelength=1024, pages=0
>> vesafb: scrolling: redraw
>> vesafb: Pseudocolor: size=0:0:0:0, shift=0:0:0:0
>
> Please tell me the output of:
>
>  A) A *patched* Linux being loaded by "linux16" command, using the
>     patch in:
>     http://lists.gnu.org/archive/html/grub-devel/2009-06/txteL9T65YdVL.txt
>
>     This could be what you already provided in
>     http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=535026#43
>     except your followup comment "vesafb is not picking up the right options"
>     contradicts that (linux16 *is* the old loader, and it is supposed to
>     work.  If linux16 doesn't work, we have an entirely different problem).

Sorry for the confusion, linux16 is working fine, linux does not. A
patched kernel with linux16 reports:

screen_info dump:
orig_x = 0x0, orig_y =  0x4
ext_mem_k = 0xffff
orig_video_page = 0x0
orig_video_mode = 0x3
orig_video_cols = 0x50
orig_video_ega_bx = 0x3
orig_video_lines = 0x19
orig_video_isVGA = 0x23
orig_video_points = 0x10
lfb_width / lfb_height / lfb_depth = 1024x768x8
lfb_base = 0xe0000000, lfb_size = 0x100
cl_magic = 0x0, cl_offset = 0x0
lfb_linelength = 0x400
red_size = 0x8, red_pos = 0x0
green_size = 0x8, green_pos = 0x0
blue_size = 0x8, blue_pos = 0x0
rsvd_size = 0x8, rsvd_pos = 0x0
vesapm = 0x0:0x9eda
pages = 0x0
vesa_attributes = 0xbb
capabilities = 0x0

vesafb is happy:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
1536k, total 16384k
vesafb: mode is 1024x768x8, linelength=1024, pages=18
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=8:8:8:8, shift=0:0:0:0

>  B) Linux (patched or not), being loaded by "linux" command, using the
>     GRUB patch in:
>     http://lists.gnu.org/archive/html/grub-devel/2009-06/txtFlQanGk4Pv.txt
>     *plus* the patch I'm attaching.

I'm running 1.96+20090709-1 which already contains the patch to the
capability field. Grub reports:

screen_info dump:
orig_x = 0x0, orig_y =  0x0
ext_mem_k = 0x8000
orig_video_page = 0x0
orig_video_mode = 0x3
orig_video_cols = 0x0
orig_video_ega_bx = 0x0
orig_video_lines = 0x0
orig_video_isVGA = 0x23
orig_video_points = 0x10
lfb_width / lfb_height / lfb_depth = 1024x768x8
lfb_base = 0xe0000000, lfb_size = 0xc
cl_magic = 0xa33f, cl_offset = 0x1000
lfb_linelength = 0x400
red_size = 0x0, red_pos = 0x0
green_size = 0x0, green_pos = 0x0
blue_size = 0x0, blue_pos = 0x0
rsvd_size = 0x0, rsvd_pos = 0x0
vesapm = 0x0:0x0
pages = 0x0
vesa_attributes = 0x0
capabilities = 0x0

the information itself is consistent with what is picked up by vesafb
(which the patch suggested for A), which later shows:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
768k, total 768k
vesafb: mode is 1024x768x8, linelength=1024, pages=0
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=0:0:0:0, shift=0:0:0:0

In this case screen stays blank.

HTH,
Luca

Message #98 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Luca Tettamanti <kronos.it@gmail.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Tue, 21 Jul 2009 22:56:09 +0200

[Message part 1 (text/plain, inline)]

On Tue, Jul 21, 2009 at 10:07:05PM +0200, Luca Tettamanti wrote:
> 
> Sorry for the confusion, linux16 is working fine, linux does not. A
> patched kernel with linux16 reports:
> 
> [...]
> 
> I'm running 1.96+20090709-1 which already contains the patch to the
> capability field. Grub reports:
> 
> screen_info dump:
> orig_x = 0x0, orig_y =  0x0
> ext_mem_k = 0x8000
> orig_video_page = 0x0
> orig_video_mode = 0x3
> orig_video_cols = 0x0
> orig_video_ega_bx = 0x0
> orig_video_lines = 0x0
> orig_video_isVGA = 0x23
> orig_video_points = 0x10
> lfb_width / lfb_height / lfb_depth = 1024x768x8
> lfb_base = 0xe0000000, lfb_size = 0xc
> cl_magic = 0xa33f, cl_offset = 0x1000
> lfb_linelength = 0x400
> red_size = 0x0, red_pos = 0x0
> green_size = 0x0, green_pos = 0x0
> blue_size = 0x0, blue_pos = 0x0
> rsvd_size = 0x0, rsvd_pos = 0x0
> vesapm = 0x0:0x0
> pages = 0x0
> vesa_attributes = 0x0
> capabilities = 0x0

Thanks Luca.  I suspect lfb_size, ${color}_size or maybe vesa_attributes to
be the culprit.

For lfb_size I already made a patch [1], please could you test and see if it
solves the problem?

But I got a hunch that "blank screen" has something to do those color
variables.  For a quick hack see attachment.

[1] http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=46;filename=lfb_size.diff;att=1;bug=535026

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

[color_hack.diff (text/x-diff, attachment)]

Message #105 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Wed, 22 Jul 2009 21:31:50 +0200

On Tue, Jul 21, 2009 at 10:56 PM, Robert Millan<rmh@aybabtu.com> wrote:
> On Tue, Jul 21, 2009 at 10:07:05PM +0200, Luca Tettamanti wrote:
>>
>> Sorry for the confusion, linux16 is working fine, linux does not. A
>> patched kernel with linux16 reports:
>>
>> [...]
>>
>> I'm running 1.96+20090709-1 which already contains the patch to the
>> capability field. Grub reports:
>>
>> screen_info dump:
>> orig_x = 0x0, orig_y =  0x0
>> ext_mem_k = 0x8000
>> orig_video_page = 0x0
>> orig_video_mode = 0x3
>> orig_video_cols = 0x0
>> orig_video_ega_bx = 0x0
>> orig_video_lines = 0x0
>> orig_video_isVGA = 0x23
>> orig_video_points = 0x10
>> lfb_width / lfb_height / lfb_depth = 1024x768x8
>> lfb_base = 0xe0000000, lfb_size = 0xc
>> cl_magic = 0xa33f, cl_offset = 0x1000
>> lfb_linelength = 0x400
>> red_size = 0x0, red_pos = 0x0
>> green_size = 0x0, green_pos = 0x0
>> blue_size = 0x0, blue_pos = 0x0
>> rsvd_size = 0x0, rsvd_pos = 0x0
>> vesapm = 0x0:0x0
>> pages = 0x0
>> vesa_attributes = 0x0
>> capabilities = 0x0
>
> Thanks Luca.  I suspect lfb_size, ${color}_size or maybe vesa_attributes to
> be the culprit.
>
> For lfb_size I already made a patch [1], please could you test and see if it
> solves the problem?
>
> But I got a hunch that "blank screen" has something to do those color
> variables.  For a quick hack see attachment.

The patch for the size of FB is not sufficient; applying the color
hack makes vesafb work:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
1536k, total 16384k
vesafb: mode is 1024x768x8, linelength=1024, pages=0
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=8:8:8:8, shift=0:0:0:0

Luca

Message #108 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Luca Tettamanti <kronos.it@gmail.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Thu, 23 Jul 2009 02:10:16 +0200

[Message part 1 (text/plain, inline)]

On Wed, Jul 22, 2009 at 09:31:50PM +0200, Luca Tettamanti wrote:
> 
> The patch for the size of FB is not sufficient; applying the color
> hack makes vesafb work:

Great.  We're almost finished.  Please could you test:

  - Whether the color hack alone is sufficient, or the patch for FB size
    is also needed.

  - Attached patch to (hopefully) fix color properly instead of the hack.

Many thanks!

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

[dac_palette_width.diff (text/x-diff, attachment)]

Message #113 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Thu, 23 Jul 2009 18:07:52 +0200

On Thu, Jul 23, 2009 at 2:10 AM, Robert Millan<rmh@aybabtu.com> wrote:
> On Wed, Jul 22, 2009 at 09:31:50PM +0200, Luca Tettamanti wrote:
>>
>> The patch for the size of FB is not sufficient; applying the color
>> hack makes vesafb work:
>
> Great.  We're almost finished.  Please could you test:
>
>  - Whether the color hack alone is sufficient, or the patch for FB size
>    is also needed.

The color hack is sufficient.

>  - Attached patch to (hopefully) fix color properly instead of the hack.

It doesn't work. This is what vesafb picks up:

vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
768k, total 768k
vesafb: mode is 1024x768x8, linelength=1024, pages=0
vesafb: scrolling: redraw
vesafb: Pseudocolor: size=0:1:1:1, shift=0:0:0:0

screen_info dump:
orig_x = 0x0, orig_y =  0x0
ext_mem_k = 0x8000
orig_video_page = 0x0
orig_video_mode = 0x3
orig_video_cols = 0x0
orig_video_ega_bx = 0x0
orig_video_lines = 0x0
orig_video_isVGA = 0x23
orig_video_points = 0x10
lfb_width / lfb_height / lfb_depth = 1024x768x8
lfb_base = 0xe0000000, lfb_size = 0xc
cl_magic = 0xa33f, cl_offset = 0x1000
lfb_linelength = 0x400
red_size = 0x1, red_pos = 0x0
green_size = 0x1, green_pos = 0x0
blue_size = 0x1, blue_pos = 0x0
rsvd_size = 0x0, rsvd_pos = 0x0
vesapm = 0x0:0x0
pages = 0x0
vesa_attributes = 0x0
capabilities = 0x0

Luca

Message #116 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Luca Tettamanti <kronos.it@gmail.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Thu, 23 Jul 2009 20:04:04 +0200

On Thu, Jul 23, 2009 at 06:07:52PM +0200, Luca Tettamanti wrote:
> On Thu, Jul 23, 2009 at 2:10 AM, Robert Millan<rmh@aybabtu.com> wrote:
> > On Wed, Jul 22, 2009 at 09:31:50PM +0200, Luca Tettamanti wrote:
> >>
> >> The patch for the size of FB is not sufficient; applying the color
> >> hack makes vesafb work:
> >
> > Great.  We're almost finished.  Please could you test:
> >
> >  - Whether the color hack alone is sufficient, or the patch for FB size
> >    is also needed.
> 
> The color hack is sufficient.
> 
> >  - Attached patch to (hopefully) fix color properly instead of the hack.
> 
> It doesn't work. This is what vesafb picks up:
> 
> vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
> 768k, total 768k
> vesafb: mode is 1024x768x8, linelength=1024, pages=0
> vesafb: scrolling: redraw
> vesafb: Pseudocolor: size=0:1:1:1, shift=0:0:0:0

What vga= paramter are you using?  0x305 ?  What does "vbeinfo" GRUB command
(with unpatched GRUB) print for this mode (0x105, as you need to substract
0x200)?   What does "linux16 /linux.img vga=ask" say about this mode?

Do other modes work?

Thanks!

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #121 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org

Subject: Re: patch, please test

Date: Thu, 23 Jul 2009 21:34:12 +0200

On Thu, Jul 23, 2009 at 8:04 PM, Robert Millan<rmh@aybabtu.com> wrote:
> On Thu, Jul 23, 2009 at 06:07:52PM +0200, Luca Tettamanti wrote:
>> On Thu, Jul 23, 2009 at 2:10 AM, Robert Millan<rmh@aybabtu.com> wrote:
>> > On Wed, Jul 22, 2009 at 09:31:50PM +0200, Luca Tettamanti wrote:
>> >>
>> >> The patch for the size of FB is not sufficient; applying the color
>> >> hack makes vesafb work:
>> >
>> > Great.  We're almost finished.  Please could you test:
>> >
>> >  - Whether the color hack alone is sufficient, or the patch for FB size
>> >    is also needed.
>>
>> The color hack is sufficient.
>>
>> >  - Attached patch to (hopefully) fix color properly instead of the hack.
>>
>> It doesn't work. This is what vesafb picks up:
>>
>> vesafb: framebuffer at 0xe0000000, mapped to 0xffffc90010100000, using
>> 768k, total 768k
>> vesafb: mode is 1024x768x8, linelength=1024, pages=0
>> vesafb: scrolling: redraw
>> vesafb: Pseudocolor: size=0:1:1:1, shift=0:0:0:0
>
> What vga= paramter are you using?  0x305 ?

Yes.

> What does "vbeinfo" GRUB command
> (with unpatched GRUB) print for this mode (0x105, as you need to substract
> 0x200)?

Ah. Packed color...

> What does "linux16 /linux.img vga=ask" say about this mode?

Well it just list the modes (and - yes - 0x305 is present)

> Do other modes work?

All modes work except for the ones using packed color format (8 bits).

Luca

Message #124 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Luca Tettamanti <kronos.it@gmail.com>, 535026@bugs.debian.org

Subject: Re: Bug#535026: patch, please test

Date: Fri, 24 Jul 2009 17:33:27 +0200

[Message part 1 (text/plain, inline)]

Please try this one.

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

[dac_palette_width.diff (text/x-diff, attachment)]

Message #131 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Luca Tettamanti <kronos.it@gmail.com>

To: Robert Millan <rmh@aybabtu.com>

Cc: 535026@bugs.debian.org

Subject: Re: Bug#535026: patch, please test

Date: Fri, 24 Jul 2009 23:51:46 +0200

On Fri, Jul 24, 2009 at 5:33 PM, Robert Millan<rmh@aybabtu.com> wrote:
> Please try this one.

Ok, this patch works fine with 8bit color mode.

thanks,
Luca

Message #134 received at 535026@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>

To: Luca Tettamanti <kronos.it@gmail.com>

Cc: 535026@bugs.debian.org

Subject: Re: Bug#535026: patch, please test

Date: Sat, 25 Jul 2009 01:36:06 +0200

On Fri, Jul 24, 2009 at 11:51:46PM +0200, Luca Tettamanti wrote:
> On Fri, Jul 24, 2009 at 5:33 PM, Robert Millan<rmh@aybabtu.com> wrote:
> > Please try this one.
> 
> Ok, this patch works fine with 8bit color mode.

Finally!

I'll now have to discuss a pair of implementation details with upstream, but
rest assured that a fixed snapshot will be uploaded soon.

Thank you very much.

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."

Message #141 received at 535026-close@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh.debian@aybabtu.com>

To: 535026-close@bugs.debian.org

Subject: Bug#535026: fixed in grub2 1.96+20090725-1

Date: Sat, 25 Jul 2009 19:32:09 +0000

Source: grub2
Source-Version: 1.96+20090725-1

We believe that the bug you reported is fixed in the latest version of
grub2, which is due to be installed in the Debian FTP archive:

grub-common_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-common_1.96+20090725-1_amd64.deb
grub-coreboot_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-coreboot_1.96+20090725-1_amd64.deb
grub-efi-amd64_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-efi-amd64_1.96+20090725-1_amd64.deb
grub-efi-ia32_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-efi-ia32_1.96+20090725-1_amd64.deb
grub-efi_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-efi_1.96+20090725-1_amd64.deb
grub-firmware-qemu_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-firmware-qemu_1.96+20090725-1_amd64.deb
grub-ieee1275_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-ieee1275_1.96+20090725-1_amd64.deb
grub-linuxbios_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-linuxbios_1.96+20090725-1_amd64.deb
grub-pc_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-pc_1.96+20090725-1_amd64.deb
grub-rescue-pc_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub-rescue-pc_1.96+20090725-1_amd64.deb
grub2_1.96+20090725-1.diff.gz
  to pool/main/g/grub2/grub2_1.96+20090725-1.diff.gz
grub2_1.96+20090725-1.dsc
  to pool/main/g/grub2/grub2_1.96+20090725-1.dsc
grub2_1.96+20090725-1_amd64.deb
  to pool/main/g/grub2/grub2_1.96+20090725-1_amd64.deb
grub2_1.96+20090725.orig.tar.gz
  to pool/main/g/grub2/grub2_1.96+20090725.orig.tar.gz



A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 535026@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Robert Millan <rmh.debian@aybabtu.com> (supplier of updated grub2 package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@debian.org)


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.8
Date: Sat, 25 Jul 2009 19:00:53 +0200
Source: grub2
Binary: grub2 grub-linuxbios grub-efi grub-common grub-pc grub-rescue-pc grub-coreboot grub-efi-ia32 grub-efi-amd64 grub-ieee1275 grub-firmware-qemu
Architecture: source amd64
Version: 1.96+20090725-1
Distribution: unstable
Urgency: high
Maintainer: GRUB Maintainers <pkg-grub-devel@lists.alioth.debian.org>
Changed-By: Robert Millan <rmh.debian@aybabtu.com>
Description: 
 grub-common - GRand Unified Bootloader, version 2 (common files)
 grub-coreboot - GRand Unified Bootloader, version 2 (Coreboot version)
 grub-efi   - GRand Unified Bootloader, version 2 (dummy package)
 grub-efi-amd64 - GRand Unified Bootloader, version 2 (EFI-AMD64 version)
 grub-efi-ia32 - GRand Unified Bootloader, version 2 (EFI-IA32 version)
 grub-firmware-qemu - GRUB firmware image for QEMU
 grub-ieee1275 - GRand Unified Bootloader, version 2 (Open Firmware version)
 grub-linuxbios - GRand Unified Bootloader, version 2 (dummy package)
 grub-pc    - GRand Unified Bootloader, version 2 (PC/BIOS version)
 grub-rescue-pc - GRUB bootable rescue images, version 2 (PC/BIOS version)
 grub2      - GRand Unified Bootloader, version 2 (dummy package)
Closes: 535026 538177
Changes: 
 grub2 (1.96+20090725-1) unstable; urgency=high
 .
   * New SVN snapshot.
     - Don't add drivemap call with Windows Vista/7. It breaks Win 7.
       (LP: #402154)
 .
   [ Felix Zielcke ]
   * Don't build grub-efi-amd64 on hurd-i386.
   * Change DEB_BUILD_ARCH to DEB_HOST_ARCH in the check for sparc.
   * Don't add the lintian override for kernel.img for sparc and grub-pc.
   * Add a lintian override for binary-from-other-architecture for
     grub-efi-amd64 and grub-pc on i386.
   * Use wildcards in the lintian overrides.
   * Add a Conflicts/Replaces for all packages except grub-common.
     (Closes: #538177)
 .
   [ Robert Millan ]
   * 008_dac_palette_width.diff: New patch.  Fix blank screen when booting
     Linux with vga= parameter set to a packed color mode (<= 8-bit).
     (Closes: #535026)
   * Set urgency=high because #535026 affects 1.96+20090709-1 which is in
     testing now.
   * patches/907_terminal_output_workaround.diff: Work around recent regression
     with terminal_output command (not critical, just breaks gfxterm).
Checksums-Sha1: 
 040476724380bc6c5a0db6d51cff6280bbee6d5c 1814 grub2_1.96+20090725-1.dsc
 26d14dcf3607f3b86ab7a8c72a0a300a32bacb35 1161862 grub2_1.96+20090725.orig.tar.gz
 28d22c91784277936bed9fb107127b9c050c0056 104940 grub2_1.96+20090725-1.diff.gz
 45efe27decac2d799f36faa51e85d266bd2d1bc7 180504 grub2_1.96+20090725-1_amd64.deb
 5d11a3deea5adbc2b2d4e6cde959811e6a0259bd 179400 grub-linuxbios_1.96+20090725-1_amd64.deb
 d95c0275cbbf75f1ae3793c156c39afc80138c71 179398 grub-efi_1.96+20090725-1_amd64.deb
 768bbad63a1de02b8af9127630b2c355c6fa203a 1206402 grub-common_1.96+20090725-1_amd64.deb
 cdc752656c55d61be7b47ebdad1ec78814257054 660210 grub-pc_1.96+20090725-1_amd64.deb
 230450203ac716545319cd246fb289e9e263dc70 809884 grub-rescue-pc_1.96+20090725-1_amd64.deb
 b520c90aa69591cb62b57a10601d27e30553fc12 432404 grub-coreboot_1.96+20090725-1_amd64.deb
 ef07ba51a91831be9a9c8f11739675e6ff627d42 462814 grub-efi-ia32_1.96+20090725-1_amd64.deb
 270b11a11b56995d22e6bb3782b624c1fc6c468d 519402 grub-efi-amd64_1.96+20090725-1_amd64.deb
 fd82d77125625f9d34c1c6d3769f2a6d467366c0 427678 grub-ieee1275_1.96+20090725-1_amd64.deb
 aca40f33c2195ab85983a54ab8af6ce545f970d7 411172 grub-firmware-qemu_1.96+20090725-1_amd64.deb
Checksums-Sha256: 
 8a5e9dfc1e055b068319d33378ef4e5ac5b30e70f8275812684ade47b11a6347 1814 grub2_1.96+20090725-1.dsc
 f75194a24675a71d9da8ccee4368534dc6bb3e377aad3545227283a9bbeadb1b 1161862 grub2_1.96+20090725.orig.tar.gz
 60c7879f7008846ffa77c5cccce63c5fd746f772f0d453b3336e071633cda2bf 104940 grub2_1.96+20090725-1.diff.gz
 d78443f8ea5cf116b9afbf859d7fe0a06162947afbc3fc450f5cd478ed23cb24 180504 grub2_1.96+20090725-1_amd64.deb
 daf10d895a15d1c0472538e33d8af44dca74261f104aaeba9ca1c825de2d8267 179400 grub-linuxbios_1.96+20090725-1_amd64.deb
 705fe99e4b472489317724d340d4cfc4e5063f9ff313d3a674a02d4c477bac26 179398 grub-efi_1.96+20090725-1_amd64.deb
 c392ee954b7b5398ff703a2a714da26d1e7e40713331d98a4117ba1dcaac05a4 1206402 grub-common_1.96+20090725-1_amd64.deb
 398d0513ebcdc9c0b8fd7b84808abddae6ca85484fc166ea72590b7f48099437 660210 grub-pc_1.96+20090725-1_amd64.deb
 fd9e6c3058cac3b233618e418bf0b1cdf2827f3b366f89740d7e2f2e59a80c5a 809884 grub-rescue-pc_1.96+20090725-1_amd64.deb
 44b0ba8d2d1a74ca9ee5df9495fdebe68d017f3c2e6738179972a215b7dd0fdd 432404 grub-coreboot_1.96+20090725-1_amd64.deb
 4a70c55ae6042699670d5f9b79010b719e6aca82f574ba84c4deafbcba0e4de4 462814 grub-efi-ia32_1.96+20090725-1_amd64.deb
 50873931007e7af330e4ad3cd183a7d2bc143028a5cfaac58992833451a64e25 519402 grub-efi-amd64_1.96+20090725-1_amd64.deb
 56b98c97c4da0b317973534fc367314618b3bb959ee647a07f5a07a5ed461788 427678 grub-ieee1275_1.96+20090725-1_amd64.deb
 31b145f4ab001ea21c7868678472cc03fe22fb1555b3ae6bcd41cb099ca707f6 411172 grub-firmware-qemu_1.96+20090725-1_amd64.deb
Files: 
 91b475bd55bf1760fe3a0edcf6bdb7b3 1814 admin extra grub2_1.96+20090725-1.dsc
 c540297b23407b242bc35b45858d4863 1161862 admin extra grub2_1.96+20090725.orig.tar.gz
 549d72483dc3df87a6cfd6f5f0018b7f 104940 admin extra grub2_1.96+20090725-1.diff.gz
 356a4a233cc1b71ead8df057cf8166c5 180504 admin extra grub2_1.96+20090725-1_amd64.deb
 e380df16b7e3e649a2d60973d2a9bf2a 179400 admin extra grub-linuxbios_1.96+20090725-1_amd64.deb
 95b8056be700082cb5f7b28f2ff7e70e 179398 admin extra grub-efi_1.96+20090725-1_amd64.deb
 e617aeb038111ae6648fa8439e2b73a0 1206402 admin extra grub-common_1.96+20090725-1_amd64.deb
 0109ce250beeadcf882c7a2914b33703 660210 admin extra grub-pc_1.96+20090725-1_amd64.deb
 dab924dff269acaa2590cb1c47355c5a 809884 admin extra grub-rescue-pc_1.96+20090725-1_amd64.deb
 48033ccad60c5fddc82aed65c088f01a 432404 admin extra grub-coreboot_1.96+20090725-1_amd64.deb
 21f12d552b32bb1a55bc54d4b52f587b 462814 admin extra grub-efi-ia32_1.96+20090725-1_amd64.deb
 76d5b3fe3fba85d6810b458cc9fe0518 519402 admin extra grub-efi-amd64_1.96+20090725-1_amd64.deb
 8ed97c003e8883de82610ed1fdc4b503 427678 admin extra grub-ieee1275_1.96+20090725-1_amd64.deb
 cb8291ef4ae211b53327e86dea0c5955 411172 admin extra grub-firmware-qemu_1.96+20090725-1_amd64.deb

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkprWWEACgkQC19io6rUCv9RVwCdEm7gEQe1unglMDq2oE44Z9GC
XQUAmgNnq9NSUHAbKJlQXuzmKy5fD9OB
=pPGY
-----END PGP SIGNATURE-----

Send a report that this bug log contains spam.