Xen devel - Mar 2009 - Fw: Re: Problems with load the most recent 2.6.29-rc6 & Xen unstable on ASUS P5K Premium

>I probably shouldn''t have been operating machinery yesterday.
>The crash is probably because I committed the e820 memory rearranging
>stuff into the wrong branch, but something may have broken us from
>tip/master as well.
>   J
Kernel was already broken at 2/28/09

--- On Sat, 2/28/09, Marc - A. Dahlhaus <mad@wol.de> wrote:
From: Marc - A. Dahlhaus <mad@wol.de>
Subject: Re: [Xen-devel] Problems with load the most recent 2.6.29-rc6 & Xen
unstable on ASUS P5K Premium
To: bderzhavets@yahoo.com
Cc: "Jeremy Fitzhardinge" <jeremy@goop.org>,
xen-devel@lists.xensource.com
Date: Saturday, February 28, 2009, 10:11 AM

Here we go...

Grub config used to boot:

title XEN pv_ops dom0 with serial console
root (hd0,1)
kernel /boot/xen-3.4-unstable.gz dom0_mem=256M loglvl=all guest_loglvl=all
com1=9600,8n1 console=com1
module /boot/bzImage-2.6.29-rc6-tip ro root=LABEL=ROOT console=hvc0
earlyprintk=xen 4
module /boot/init.rd-2.6.29-rc6-tip

Did a boot test on 3.4-unstable tip and 3.3.1 with dom0-loader-bzImage patch.
Logs attached.

Xen and dom0 kernel are freshly pulled and build from a clean tree. Kernel
config attached. Xen was build with this command after hg clone: make
KERNELS='''' dist

After xen hands over the console to dom0 the screen stays blank...

I''ll do another tests and boot dom0 with console=ttyS0,9600n8 maybe we
get
outputs from the kernel itself then.

Marc

Marc - A. Dahlhaus schrieb:
> Hello Boris,
> hello Jeremy,
> 
> i''ve the same problems here (but i also didn''t got Dom0
up and
running beyond that point before on my Boxes here...).
> I''ll got logs over serial console yesterday (Boris: i hope your
box
actually have a serial port as i had to get me a cheap Asus Barebone for that
;o) and watched them with the screen utility and there are quite a few repeating
backtraces before the whole box stops responding. I''ll do another run
(after
pulling and building Jeremys remaining changes in) and catch the backtraces into
log files over the the next few hours to post them here.
> 
> Marc
> 
> Boris Derzhavets schrieb:
>> The most recent 2.6.29-rc6 with pvops Dom0 support been compiled
>> on my old box (ASUS P5K Premium)  behaves similar to situation
described bellow for ASUS P5Q-E :-
>> 
>> Screen output interrupts immediately after Xen prints scrubbing for
free
>> ram  .  .  .  .   .   .
>> Just black screen for several second followed by Xen host reboot (
system
>> hangs on box built up on ASUS P5Q-E)
>> At mean time i have no any idea how to get boot logs.
>> 
>> --- On *Fri, 2/27/09, Jeremy Fitzhardinge /<jeremy@goop.org>/*
wrote:
>> 
>>     From: Jeremy Fitzhardinge <jeremy@goop.org>
>>     Subject: Re: [Xen-devel] Problems with load the most recent
>>     2.6.29-rc6 & Xen unstable on ASUS P5Q-E
>>     To: bderzhavets@yahoo.com
>>     Cc: xen-devel@lists.xensource.com
>>     Date: Friday, February 27, 2009, 12:59 PM
>> 
>>     Boris Derzhavets wrote:
>>     > Screen output interrupts immediately after Xen prints
scrambling for free
>>     > ram  .  .  .  .   .   .
>>     > Just black screen for ever.
>>     > Novell''s 2.6.27 loads fine under same Xen Unstable.
>>     > 2.6.29-rc3 ( pvops ) also loads fine under same Xen Unstable.
>>     > Box with rc4,rc5 is temporary unavailable.
>>     >
>>     >
>> 
>>     Do you have any diagnostics?  Even a screenshot?
>> 
>>         J
>> 
>>     _______________________________________________
>>     Xen-devel mailing list
>>     Xen-devel@lists.xensource.com
>>     http://lists.xensource.com/xen-devel
>>              
>>
------------------------------------------------------------------------
>> 
>> _______________________________________________
>> Xen-devel mailing list
>> Xen-devel@lists.xensource.com
>> http://lists.xensource.com/xen-devel
>>   
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel



      




_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Boris Derzhavets wrote:
> >I probably shouldn''t have been operating machinery yesterday.
> >The crash is probably because I committed the e820 memory rearranging 
> >stuff into the wrong branch, but something may have broken us from 
> >tip/master as well.
> >   J
>
> Kernel was already broken at *2/28/09
> *
>
Thanks, that''s a useful reference.  I''m in the middle of a
bisect at the
moment, so I should nail down what went wrong fairly soon.

    J

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Jeremy Fitzhardinge schrieb:
> Boris Derzhavets wrote:
>> >I probably shouldn''t have been operating machinery
yesterday.
>> >The crash is probably because I committed the e820 memory 
>> rearranging >stuff into the wrong branch, but something may have 
>> broken us from >tip/master as well.
>> >   J
>>
>> Kernel was already broken at *2/28/09
>> *
>>
> Thanks, that''s a useful reference.  I''m in the middle of
a bisect at
> the moment, so I should nail down what went wrong fairly soon.
>
>    J
The second test i did with console=ttyS0,9600n8 kernel param didn''t
show
output from the kernel.

Marc

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Tue, 3 Mar 2009, Jeremy Fitzhardinge wrote:
>> Kernel was already broken at *2/28/09
>> *
>> 
> Thanks, that''s a useful reference.  I''m in the middle of
a bisect at the
> moment, so I should nail down what went wrong fairly soon.
If you need another reference point a kernel I build from xen/dom0/hackery 
as of 25th Feb works.

 	Michael Young

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Tue, 3 Mar 2009, M A Young wrote:
> If you need another reference point a kernel I build from xen/dom0/hackery
as
> of 25th Feb works.
Or more precisely up to and including changeset
55d8085671863fe4ee6a17b7814bd38180a44e1d
http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=commit;h=55d8085671863fe4ee6a17b7814bd38180a44e1d

 	Michael Young

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

M A Young wrote:
> On Tue, 3 Mar 2009, M A Young wrote:
>
>> If you need another reference point a kernel I build from 
>> xen/dom0/hackery as of 25th Feb works.
>
> Or more precisely up to and including changeset 
> 55d8085671863fe4ee6a17b7814bd38180a44e1d
>
http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=commit;h=55d8085671863fe4ee6a17b7814bd38180a44e1d
>
Thanks.  I make the breakage a bit more recent than that.  The current 
pushed xen/dom0/hackery boots for me now.

    J

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Jeremy Fitzhardinge schrieb:
> M A Young wrote:
>> On Tue, 3 Mar 2009, M A Young wrote:
>>
>>> If you need another reference point a kernel I build from 
>>> xen/dom0/hackery as of 25th Feb works.
>>
>> Or more precisely up to and including changeset 
>> 55d8085671863fe4ee6a17b7814bd38180a44e1d
>>
http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=commit;h=55d8085671863fe4ee6a17b7814bd38180a44e1d
>>
>
> Thanks.  I make the breakage a bit more recent than that.  The current 
> pushed xen/dom0/hackery boots for me now.
>
>    J
Crashes for me on x86_32 with the same symptoms as in my last tests. 
even the backtraces are quite the same as the ones posted before.
I''ll try an x86_64 build tomorrow...

Marc


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Marc - A. Dahlhaus wrote:
> Crashes for me on x86_32 with the same symptoms as in my last tests. 
> even the backtraces are quite the same as the ones posted before.
> I''ll try an x86_64 build tomorrow...
OK,  I have to admit I didn''t look to closely at your results until
I''d
unbroken the kernel for myself.  If you''re seeing stack-protector-like 
crashes, it suggests something wrong with the kernel gs, which is likely 
the result of the new virtually mapped percpu areas.

    J

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

x86_64 build works for me on E8400 box and fails to load on Q9550 box :-
Black screen after Xen scrubbing for free RAM.
I guess a kind of RS232 terminal is required to be connected to COM1 to setup
serial console. I am not sure may any converting cable to be used for connection
to  Samsung 721N.
Please, suggest any appropriate technical specifications or point to my
misunderstanding of this procedure.

--- On Tue, 3/3/09, Marc - A. Dahlhaus <mad@wol.de> wrote:
From: Marc - A. Dahlhaus <mad@wol.de>
Subject: Re: Fw: Re: [Xen-devel] Problems with load the most recent 2.6.29-rc6
& Xen unstable on ASUS P5K Premium
To: "Jeremy Fitzhardinge" <jeremy@goop.org>
Cc: bderzhavets@yahoo.com, "Xen-devel"
<xen-devel@lists.xensource.com>, "Ian Campbell"
<Ian.Campbell@citrix.com>, "M A Young"
<m.a.young@durham.ac.uk>
Date: Tuesday, March 3, 2009, 7:11 PM

Jeremy Fitzhardinge schrieb:
> M A Young wrote:
>> On Tue, 3 Mar 2009, M A Young wrote:
>> 
>>> If you need another reference point a kernel I build from
xen/dom0/hackery as of 25th Feb works.
>> 
>> Or more precisely up to and including changeset
55d8085671863fe4ee6a17b7814bd38180a44e1d
>>
http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=commit;h=55d8085671863fe4ee6a17b7814bd38180a44e1d
> 
> Thanks.  I make the breakage a bit more recent than that.  The current
pushed xen/dom0/hackery boots for me now.
> 
>    J
Crashes for me on x86_32 with the same symptoms as in my last tests. even the
backtraces are quite the same as the ones posted before.
I''ll try an x86_64 build tomorrow...

Marc


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel



      

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hello Boris,


i use a simple serial null modem cable to connect two serial ports on my
test machines with each other and did a simple "cat /dev/ttyS0 | tee
file.log" on the listening box to generate the logs.

Marc

Am Mittwoch, den 04.03.2009, 03:37 -0800 schrieb Boris
Derzhavets:
> x86_64 build works for me on E8400 box and fails to load on Q9550
> box :-
> Black screen after Xen scrubbing for free RAM.
> I guess a kind of RS232 terminal is required to be connected to COM1
> to setup
> serial console. I am not sure may any converting cable to be used for
> connection
> to  Samsung 721N.
> Please, suggest any appropriate technical specifications or point to
> my misunderstanding of this procedure.
> 
> --- On Tue, 3/3/09, Marc - A. Dahlhaus <mad@wol.de> wrote:
>         From: Marc - A. Dahlhaus <mad@wol.de>
>         Subject: Re: Fw: Re: [Xen-devel] Problems with load the most
>         recent 2.6.29-rc6 & Xen unstable on ASUS P5K Premium
>         To: "Jeremy Fitzhardinge" <jeremy@goop.org>
>         Cc: bderzhavets@yahoo.com, "Xen-devel"
>         <xen-devel@lists.xensource.com>, "Ian Campbell"
>         <Ian.Campbell@citrix.com>, "M A Young"
>         <m.a.young@durham.ac.uk>
>         Date: Tuesday, March 3, 2009, 7:11 PM
>         
>         Jeremy Fitzhardinge schrieb:
>         > M A Young wrote:
>         >> On Tue, 3 Mar 2009, M A Young wrote:
>         >> 
>         >>> If you need another reference point a kernel I build
from
>         xen/dom0/hackery as of 25th Feb works.
>         >> 
>         >> Or more precisely up to and including changeset
>         55d8085671863fe4ee6a17b7814bd38180a44e1d
>         >>
>        
http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=commit;h=55d8085671863fe4ee6a17b7814bd38180a44e1d
>         
>         > 
>         > Thanks.  I make the breakage a bit more recent than that.  The
current
>         pushed xen/dom0/hackery boots for me now.
>         > 
>         >    J
>         Crashes for me on x86_32 with the same symptoms as in my last
tests. even the
>         backtraces are quite the same as the ones posted before.
>         I''ll try
>          an x86_64 build tomorrow...
>         
>         Marc
>         
>         
>         _______________________________________________
>         Xen-devel mailing list
>         Xen-devel@lists.xensource.com
>         http://lists.xensource.com/xen-devel


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Thank you , Mark.
I got it.

--- On Wed, 3/4/09, Marc - A. Dahlhaus [ Administration | Westermann GmbH ]
<mad@wol.de> wrote:
From: Marc - A. Dahlhaus [ Administration | Westermann GmbH ] <mad@wol.de>
Subject: Re: Fw: Re: [Xen-devel] Problems with load the most recent 2.6.29-rc6
& Xen unstable on ASUS P5K Premium
To: bderzhavets@yahoo.com
Cc: "Jeremy Fitzhardinge" <jeremy@goop.org>,
"Xen-devel" <xen-devel@lists.xensource.com>, "Ian
Campbell" <Ian.Campbell@citrix.com>, "M A Young"
<m.a.young@durham.ac.uk>
Date: Wednesday, March 4, 2009, 7:51 AM

Hello Boris,


i use a simple serial null modem cable to connect two serial ports on my
test machines with each other and did a simple "cat /dev/ttyS0 | tee
file.log" on the listening box to generate the logs.

Marc

Am Mittwoch, den 04.03.2009, 03:37 -0800 schrieb Boris
Derzhavets:
> x86_64 build works for me on E8400 box and fails to load on Q9550
> box :-
> Black screen after Xen scrubbing for free RAM.
> I guess a kind of RS232 terminal is required to be connected to COM1
> to setup
> serial console. I am not sure may any converting cable to be used for
> connection
> to  Samsung 721N.
> Please, suggest any appropriate technical specifications or point to
> my misunderstanding of this procedure.
> 
> --- On Tue, 3/3/09, Marc - A. Dahlhaus <mad@wol.de> wrote:
>         From: Marc - A. Dahlhaus <mad@wol.de>
>         Subject: Re: Fw: Re: [Xen-devel] Problems with load the most
>         recent 2.6.29-rc6 & Xen unstable on ASUS P5K Premium
>         To: "Jeremy Fitzhardinge" <jeremy@goop.org>
>         Cc: bderzhavets@yahoo.com, "Xen-devel"
>         <xen-devel@lists.xensource.com>, "Ian Campbell"
>         <Ian.Campbell@citrix.com>, "M A Young"
>         <m.a.young@durham.ac.uk>
>         Date: Tuesday, March 3, 2009, 7:11 PM
>         
>         Jeremy Fitzhardinge schrieb:
>         > M A Young wrote:
>         >> On Tue, 3 Mar 2009, M A Young wrote:
>         >> 
>         >>> If you need another reference point a kernel I build
from
>         xen/dom0/hackery as of 25th Feb works.
>         >> 
>         >> Or more precisely up to and including changeset
>         55d8085671863fe4ee6a17b7814bd38180a44e1d
>         >>
>        
http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=commit;h=55d8085671863fe4ee6a17b7814bd38180a44e1d
>         
>         > 
>         > Thanks.  I make the breakage a bit more recent than that. 
The current
>         pushed xen/dom0/hackery boots for me now.
>         > 
>         >    J
>         Crashes for me on x86_32 with the same symptoms as in my last
tests. even the
>         backtraces are quite the same as the ones posted before.
>         I''ll try
>          an x86_64 build tomorrow...
>         
>         Marc
>         
>         
>         _______________________________________________
>         Xen-devel mailing list
>         Xen-devel@lists.xensource.com
>         http://lists.xensource.com/xen-devel


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel



      

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

M A Young wrote:
> On Tue, 3 Mar 2009, M A Young wrote:
>
>> If you need another reference point a kernel I build from 
>> xen/dom0/hackery as of 25th Feb works.
>
> Or more precisely up to and including changeset 
> 55d8085671863fe4ee6a17b7814bd38180a44e1d
>
http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=commit;h=55d8085671863fe4ee6a17b7814bd38180a44e1d
>
I committed a few fixes which gets 32-bit dom0 booting for me.

    J

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel