Skip to contentSkip to conversation

Bug in SLES8 kernel 2.4.x freezing HP DL740/760

DISCUSSION Oliver Antwerpen 9 participants 13 messages Feb 16, 2005 – Feb 19, 2005

lore
Share & tools
lore
#0 lore raw

Hi out there,

If there is anybody out there using SLES8 on HP ProLiant DL740/760:
BEWARE!

SuSE has patched UNICON into the kernel which will cause these servers to hang when booted with vga=normal. The system will run fine in fb-mode, but not in plain text.

I cannot see, where this UNICON-patch comes from, it seems that is has been developed by some turbolinux-coders.

HP and SuSE have not been able to either fix this problem or at least warn someone about this bug, so I will do it now. The bug is known since Nov 13th 2004.

If there should be anybody who can help, please contact me.

I hope this information will help someone to not run into deep trouble.

Oliver Antwerpen

#1 lore raw
Quoted 16 lines · Oliver Antwerpen

Hi out there,

If there is anybody out there using SLES8 on HP ProLiant DL740/760:
BEWARE!

SuSE has patched UNICON into the kernel which will cause these servers to hang when booted with vga=normal. The system will run fine in fb-mode, but not in plain text.

I cannot see, where this UNICON-patch comes from, it seems that is has been developed by some turbolinux-coders.

HP and SuSE have not been able to either fix this problem or at least warn someone about this bug, so I will do it now. The bug is known since Nov 13th 2004.

If there should be anybody who can help, please contact me.

I hope this information will help someone to not run into deep trouble.

Quoted 9 lines

Oliver Antwerpen

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Well if you don't need unicon, then remove the patch from the .spec file and rebuild the kernel (from the source rpm). Or report it their bug tracking system.

Matthias-Christian Ott

#2 lore raw

Hi,

Quoted 5 lines · Matthias-Christian Ott
Quoted 1 line · Oliver Antwerpen

SuSE has patched UNICON into the kernel which will cause these servers to hang when booted with vga=normal. The system will run fine in fb-mode, but not in plain text.

Well if you don't need unicon, then remove the patch from the .spec file and rebuild the kernel (from the source rpm). Or report it their bug tracking system.

My problem ist, that when I change .config, then I lose my support. So SuSE or HP have to tell me to do so.

Oliver

#3 lore raw
Quoted 6 lines · Oliver Antwerpen
Quoted 5 lines · Matthias-Christian Ott
Quoted 3 lines · Oliver Antwerpen

SuSE has patched UNICON into the kernel which will cause these servers to hang when booted with vga=normal. The system will run fine in fb-mode, but not in plain text.

Quoted 6 lines
Quoted 3 lines

Well if you don't need unicon, then remove the patch from the .spec file and rebuild the kernel (from the source rpm). Or report it their bug tracking system.

My problem ist, that when I change .config, then I lose my support. So SuSE or HP have to tell me to do so.

And they listen here to you?

Bernd

Signature · 3 lines
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services
#4 lore raw

AMD 64 and Kernel AGPart support

In reply to Oliver Antwerpen

Hello,

I have an AMD 64 (gentoo compiled for amd64) and I would like to succeed in my video driver installation (ATI Radeon 9250 :-/). I would need the agpgart support for the Sis Chipset, but all the entry for agpgart are grayed, I can't change anything (Kernel 2.6.9, 2.6.10 ...)

So is it normal or a bug ?? , or am I making a mistake.

NB: one of my friends made the test, without AMD64 and exactly the same kernel he can check these options within agpgart...

Thanks,
Marc

#5 lore raw

Re: AMD 64 and Kernel AGPart support

Quoted 6 lines · On Thu, 17 Feb 2005 15:08:32 +0100 Marc Cramdal <[email protected]> wrote:

I have an AMD 64 (gentoo compiled for amd64) and I would like to succeed in my video driver installation (ATI Radeon 9250 :-/). I would need the agpgart support for the Sis Chipset, but all the entry for agpgart are grayed, I can't change anything (Kernel 2.6.9, 2.6.10 ...)

So is it normal or a bug ?? , or am I making a mistake.

I have an AMD Athlon64 with Radeon 9200SE on VIA K8T800 Chipset... and AGP works fine with only CONFIG_AGP_AMD64 enabled.

The fact is that these AMD processors have an "on-CPU northbridge" for AGP, and the support for the various external AGP bridges (VIA, SiS, Nvidia) is provided directly in "drivers/char/agp/amd64-agp.c" driver (selected by CONFIG_AGP_AMD64).

Reading "drivers/char/agp/amd64-agp.c" I can see that "SIS 755" chipset is supported.

If you have a chipset that isn't supported you can always try with the "agp_try_unsupported=1" option (as stated in the HELP of CONFIG_AGP_AMD64).

Quoted 3 lines

NB: one of my friends made the test, without AMD64 and exactly the same kernel he can check these options within agpgart...

You don't have to be on a i386 to see the options avaiable for i386... just do a "make menuconfig ARCH=i386" and you are done!

But if you are on x86_64, why do you want to enable drivers for other architectures?

Signature · 2 lines
	Paolo Ornati
	Gentoo Linux (kernel 2.6.10-gentoo-r7)
#6 lore raw

Re: ________________: Re: AMD 64 and Kernel AGPart support

Quoted 2 lines · On Thu, 17 Feb 2005 15:42:50 +0100 [email protected] wrote:

ÄúºÃ£º
ÎÒÒÑ_­ÊÕµ_ÄúµÄÀ_ÐÅ

and... what does this means?

Signature · 2 lines
	Paolo Ornati
	Gentoo Linux (kernel 2.6.10-gentoo-r7)
#7 lore raw

Re: ________________: Re: AMD 64 and Kernel AGPart support

Quoted 4 lines · Paolo Ornati
Quoted 2 lines

ÄúºÃ£º
    ÎÒÒÑ_­ÊÕµ_ÄúµÄÀ_ÐÅ

and... what does this means?

SPAM. This looks to me like a new way of spamming though, replying to valid mailing list messages. (I too received couple of these in reply to my messages.)

Parag

#8 lore raw

Re: ________________: Re: AMD 64 and Kernel AGPart support

Quoted 3 lines · On Thu, 17 Feb 2005 10:26:55 -0500 Parag Warudkar <[email protected]>…

SPAM. This looks to me like a new way of spamming though, replying to valid mailing list messages. (I too received couple of these in reply to my messages.)

agrrr.. and I've done a mistake replying!

Signature · 2 lines
	Paolo Ornati
	Gentoo Linux (kernel 2.6.10-gentoo-r7)
#9 lore raw

spam mails with the same Message-ID

In reply to Parag Warudkar
Quoted 8 lines · Parag Warudkar
Quoted 4 lines · Paolo Ornati
Quoted 2 lines

ÄúºÃ£º
    ÎÒÒÑ_­ÊÕµ_ÄúµÄÀ_ÐÅ

and... what does this means?

SPAM. This looks to me like a new way of spamming though, replying to valid mailing list messages. (I too received couple of these in reply to my messages.)

The most interesting fact seems to be that these spam messages have the same message ID as the original Mails.

If you run a program that automatically discards duplicate mails and the spam message reaches you faster than the original email through linux-kernel (which seems to often happen with these mails), the original email will be discarded.

I don't know whether these are known attacks, but the automatic discarding of duplicated emails offers attackers nice opportunities if they know a message ID (as with these emails) or can guess the message ID (since many MUAs have predictable message IDs, an attacker C could use this to suppress a message from person A to person B by sending an email with the message ID to person B bevor person B gets the email from person A).

Quoted 1 line

Parag

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
#10 lore raw

Re: spam mails with the same Message-ID

Quoted 27 lines · Adrian Bunk
Quoted 13 lines · Parag Warudkar
Quoted 7 lines · Paolo Ornati
Quoted 4 lines

ÄúºÃ£º
ÎÒÒÑ_­ÊÕµ_ÄúµÄÀ_ÐÅ

and... what does this means?

SPAM. This looks to me like a new way of spamming though, replying to valid mailing list messages. (I too received couple of these in reply to my messages.)

The most interesting fact seems to be that these spam messages have the same message ID as the original Mails.

If you run a program that automatically discards duplicate mails and the spam message reaches you faster than the original email through linux-kernel (which seems to often happen with these mails), the original email will be discarded.

I don't know whether these are known attacks, but the automatic discarding of duplicated emails offers attackers nice opportunities if they know a message ID (as with these emails) or can guess the message ID (since many MUAs have predictable message IDs, an attacker C
could use this to suppress a message from person A to person B by sending an email with the message ID to person B bevor person B gets the email from person A).

Quoted 9 lines
Quoted 3 lines

Parag

cu
Adrian

The spamers become always cleverer :-) .

Matthias-Christian Ott

#11 lore raw

Sis760 chipset support

In reply to Oliver Antwerpen

Hello,

I can't make agpgart working (even when trying the agp_try_unsupported) option. I have an AMD64 3000+ with a Sis760 chipset and agp doesn't seem to be supported : I only get this with dmesg : "Linux agpgart interface v0.100
(c) Dave Jones". That's all...

So, is Sis760 chipset supported for agpgart under linux kernel ? if not, is there plan to be, tweaks to do (I even tried the Sis Chipset driver for !x86_64 by removing this entry in KConfig ... ) ?

Marc

#12 lore raw

Re: Sis760 chipset support

Quoted 11 lines · Marc Cramdal

Hello,

I can't make agpgart working (even when trying the agp_try_unsupported) option. I have an AMD64 3000+ with a Sis760 chipset and agp doesn't seem to be supported : I only get this with dmesg : "Linux agpgart interface v0.100
(c) Dave Jones". That's all...

So, is Sis760 chipset supported for agpgart under linux kernel ? if not, is there plan to be, tweaks to do (I even tried the Sis Chipset driver for !x86_64 by removing this entry in KConfig ... ) ?

amd64 should be using the amd64-agp driver, as your agpgart is on-cpu on that platform.

Dave