RE: Re: PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

RE: Re: PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[James Dingwall]

> I was experiencing the same immediate crash with a null pointer
> dereference (log below) on boot with 2.6.35.5.  Reverting
> fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
> me.

Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.

--- drivers/xen/events.c.orig   2010-09-23 16:10:34.257124313 +0100
+++ drivers/xen/events.c        2010-09-23 16:10:44.326194462 +0100
@@ -935,7 +935,7 @@
 	.retrigger		= retrigger_dynirq,
 };

-static struct irq_chip en_percpu_chip __read_mostly = {
+static struct irq_chip xen_percpu_chip __read_mostly = {
 	.name			= "xen-percpu",

 	.disable		= disable_dynirq,

James
This message and the information contained herein is proprietary and confidential and subject to the Amdocs policy statement,
you may review at http://www.amdocs.com/email_disclaimer.asp

Re: PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Paweł Zuzelski]

On Thu, 23 Sep 2010, James Dingwall wrote:

> > I was experiencing the same immediate crash with a null pointer
> > dereference (log below) on boot with 2.6.35.5.  Reverting
> > fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
> > me.
>
> Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.

I confirm. This patch works for me. Thanks.

-- 
Regards,
Paweł Zuzelski

Re: PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Jeremy Fitzhardinge]

 On 09/23/2010 08:23 AM, James Dingwall wrote:
>> I was experiencing the same immediate crash with a null pointer
>> dereference (log below) on boot with 2.6.35.5.  Reverting
>> fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
>> me.
> Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.

Yes, that would do it.  Upstream is OK, so it looks like this crept in
when being cherry-picked into stable-2.6.32/master and stable-2.6.35/master.

    J

> --- drivers/xen/events.c.orig   2010-09-23 16:10:34.257124313 +0100
> +++ drivers/xen/events.c        2010-09-23 16:10:44.326194462 +0100
> @@ -935,7 +935,7 @@
>  	.retrigger		= retrigger_dynirq,
>  };
>
> -static struct irq_chip en_percpu_chip __read_mostly = {
> +static struct irq_chip xen_percpu_chip __read_mostly = {
>  	.name			= "xen-percpu",
>
>  	.disable		= disable_dynirq,
>
> James
> This message and the information contained herein is proprietary and confidential and subject to the Amdocs policy statement,
> you may review at http://www.amdocs.com/email_disclaimer.asp
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: Re: PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Kyle McMartin]

On Thu, Sep 23, 2010 at 04:23:55PM +0100, James Dingwall wrote:
> > I was experiencing the same immediate crash with a null pointer
> > dereference (log below) on boot with 2.6.35.5.  Reverting
> > fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
> > me.
> 
> Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.
> 

Greg, a spelling mistake cropped into this patch during 2.6.35.5 that
James here points out. Adding you to CC since it seems to have missed
stable 2.6.35.6...

Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.

--Kyle

> --- drivers/xen/events.c.orig   2010-09-23 16:10:34.257124313 +0100
> +++ drivers/xen/events.c        2010-09-23 16:10:44.326194462 +0100
> @@ -935,7 +935,7 @@
>  	.retrigger		= retrigger_dynirq,
>  };
> 
> -static struct irq_chip en_percpu_chip __read_mostly = {
> +static struct irq_chip xen_percpu_chip __read_mostly = {
>  	.name			= "xen-percpu",
> 
>  	.disable		= disable_dynirq,
> 
> James
> This message and the information contained herein is proprietary and confidential and subject to the Amdocs policy statement,
> you may review at http://www.amdocs.com/email_disclaimer.asp
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Greg KH]

On Sun, Sep 26, 2010 at 10:40:44PM -0400, Kyle McMartin wrote:
> On Thu, Sep 23, 2010 at 04:23:55PM +0100, James Dingwall wrote:
> > > I was experiencing the same immediate crash with a null pointer
> > > dereference (log below) on boot with 2.6.35.5.  Reverting
> > > fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
> > > me.
> > 
> > Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.
> > 
> 
> Greg, a spelling mistake cropped into this patch during 2.6.35.5 that
> James here points out. Adding you to CC since it seems to have missed
> stable 2.6.35.6...
> 
> Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.

I don't understand, where did the error come from?  Was it in a patch I
applied, or did I mess up some patch?  Is the same problem upstream, and
if not, why not?

And finally, can you send this in a format that I can apply it?

thanks,

greg k-h

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Kyle McMartin]

On Sun, Sep 26, 2010 at 11:13:36PM -0700, Greg KH wrote:
> > Greg, a spelling mistake cropped into this patch during 2.6.35.5 that
> > James here points out. Adding you to CC since it seems to have missed
> > stable 2.6.35.6...
> > 
> > Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.
> 
> I don't understand, where did the error come from?  Was it in a patch I
> applied, or did I mess up some patch?  Is the same problem upstream, and
> if not, why not?

It's correct upstream, and wrong in the patches you added to
stable-queue.git for .32 and .35, no idea how it got introduced. :/

> 
> And finally, can you send this in a format that I can apply it?
> 

I'm just the messenger... James, can you send a patch to Greg with your
signed-off-by and such?

--Kyle

> thanks,
> 
> greg k-h

RE: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot [PATCH]

[James Dingwall]

> On Sun, Sep 26, 2010 at 11:13:36PM -0700, Greg KH wrote:
> > > Greg, a spelling mistake cropped into this patch during 2.6.35.5
> that
> > > James here points out. Adding you to CC since it seems to have
> missed
> > > stable 2.6.35.6...
> > >
> > > Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.
> >
> > I don't understand, where did the error come from?  Was it in a patch
> I
> > applied, or did I mess up some patch?  Is the same problem upstream,
> and
> > if not, why not?
> 
> It's correct upstream, and wrong in the patches you added to
> stable-queue.git for .32 and .35, no idea how it got introduced. :/
> 
> >
> > And finally, can you send this in a format that I can apply it?
> >
> 
> I'm just the messenger... James, can you send a patch to Greg with your
> signed-off-by and such?
> 
> --Kyle
> 
> > thanks,
> >
> > greg k-h

Hi,

My first contribution so I hope I didn't miss any important steps in formatting the patch and that Outlook keeps it intact...  If it needs any fixing up let me know.

James

Correctly name the irq_chip structure to fix an immediate failure when booting as a xen pv_ops guest with a NULL pointer exception. The missing 'x' was introduced in commit [fb412a178502dc498430723b082a932f797e4763] applied to 2.6.3[25]-stable trees.  The commit to mainline was [aaca49642b92c8a57d3ca5029a5a94019c7af69f] which did not have the problem.

Signed-off-by: James Dingwall <james@dingwall.me.uk>
Reported-by: Paweł Zuzelski <pawelz@pld-linux.org>
Tested-by: Paweł Zuzelski <pawelz@pld-linux.org>

--- drivers/xen/events.c.orig	2010-09-27 09:10:09.245652829 +0100
+++ drivers/xen/events.c	2010-09-27 09:10:27.522304658 +0100
@@ -935,7 +935,7 @@ static struct irq_chip xen_dynamic_chip 
 	.retrigger	= retrigger_dynirq,
 };
 
-static struct irq_chip en_percpu_chip __read_mostly = {
+static struct irq_chip xen_percpu_chip __read_mostly = {
 	.name		= "xen-percpu",
 
 	.disable	= disable_dynirq,

This message and the information contained herein is proprietary and confidential and subject to the Amdocs policy statement,
you may review at http://www.amdocs.com/email_disclaimer.asp

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Jeremy Fitzhardinge]

 On 09/26/2010 11:13 PM, Greg KH wrote:
> On Sun, Sep 26, 2010 at 10:40:44PM -0400, Kyle McMartin wrote:
>> On Thu, Sep 23, 2010 at 04:23:55PM +0100, James Dingwall wrote:
>>>> I was experiencing the same immediate crash with a null pointer
>>>> dereference (log below) on boot with 2.6.35.5.  Reverting
>>>> fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
>>>> me.
>>> Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.
>>>
>> Greg, a spelling mistake cropped into this patch during 2.6.35.5 that
>> James here points out. Adding you to CC since it seems to have missed
>> stable 2.6.35.6...
>>
>> Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.
> I don't understand, where did the error come from?  Was it in a patch I
> applied, or did I mess up some patch?  Is the same problem upstream, and
> if not, why not?

I don't know.  It looks like an 'x' was deleted due to an editor
fat-finger or something.  The original upstream patches were cc:d to
stable, so I think you would have got them that way (ie, I don't think I
sent them to you specifically).  Is there an opportunity in your
workflow where this error could have crept in?  The patches should have
been a pretty clean 'git cherry-pick' with no conflicts.

Thanks,
    J

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Greg KH]

On Mon, Sep 27, 2010 at 11:08:43AM -0700, Jeremy Fitzhardinge wrote:
>  On 09/26/2010 11:13 PM, Greg KH wrote:
> > On Sun, Sep 26, 2010 at 10:40:44PM -0400, Kyle McMartin wrote:
> >> On Thu, Sep 23, 2010 at 04:23:55PM +0100, James Dingwall wrote:
> >>>> I was experiencing the same immediate crash with a null pointer
> >>>> dereference (log below) on boot with 2.6.35.5.  Reverting
> >>>> fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
> >>>> me.
> >>> Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.
> >>>
> >> Greg, a spelling mistake cropped into this patch during 2.6.35.5 that
> >> James here points out. Adding you to CC since it seems to have missed
> >> stable 2.6.35.6...
> >>
> >> Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.
> > I don't understand, where did the error come from?  Was it in a patch I
> > applied, or did I mess up some patch?  Is the same problem upstream, and
> > if not, why not?
> 
> I don't know.  It looks like an 'x' was deleted due to an editor
> fat-finger or something.  The original upstream patches were cc:d to
> stable, so I think you would have got them that way (ie, I don't think I
> sent them to you specifically).  Is there an opportunity in your
> workflow where this error could have crept in?  The patches should have
> been a pretty clean 'git cherry-pick' with no conflicts.

Sorry, I don't know what happened, using my scripts, it does drop to an
editor so I can add the "signed-off-by" line, so perhaps I accidentally
did this in there.

It's now in the -stable queue and I'll push out new releases in a few
hours.

thanks,

greg k-h

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot [PATCH]

[Greg KH]

On Mon, Sep 27, 2010 at 09:37:17AM +0100, James Dingwall wrote:
> My first contribution so I hope I didn't miss any important steps in
> formatting the patch and that Outlook keeps it intact...  If it needs
> any fixing up let me know.

Looks good, I've now queued it up, thanks so much for this.

greg k-h

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Jeremy Fitzhardinge]

 On 09/28/2010 11:29 AM, Greg KH wrote:
> On Mon, Sep 27, 2010 at 11:08:43AM -0700, Jeremy Fitzhardinge wrote:
>>  On 09/26/2010 11:13 PM, Greg KH wrote:
>>> On Sun, Sep 26, 2010 at 10:40:44PM -0400, Kyle McMartin wrote:
>>>> On Thu, Sep 23, 2010 at 04:23:55PM +0100, James Dingwall wrote:
>>>>>> I was experiencing the same immediate crash with a null pointer
>>>>>> dereference (log below) on boot with 2.6.35.5.  Reverting
>>>>>> fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
>>>>>> me.
>>>>> Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.
>>>>>
>>>> Greg, a spelling mistake cropped into this patch during 2.6.35.5 that
>>>> James here points out. Adding you to CC since it seems to have missed
>>>> stable 2.6.35.6...
>>>>
>>>> Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.
>>> I don't understand, where did the error come from?  Was it in a patch I
>>> applied, or did I mess up some patch?  Is the same problem upstream, and
>>> if not, why not?
>> I don't know.  It looks like an 'x' was deleted due to an editor
>> fat-finger or something.  The original upstream patches were cc:d to
>> stable, so I think you would have got them that way (ie, I don't think I
>> sent them to you specifically).  Is there an opportunity in your
>> workflow where this error could have crept in?  The patches should have
>> been a pretty clean 'git cherry-pick' with no conflicts.
> Sorry, I don't know what happened, using my scripts, it does drop to an
> editor so I can add the "signed-off-by" line, so perhaps I accidentally
> did this in there.
>
> It's now in the -stable queue and I'll push out new releases in a few
> hours.
>

I should have pointed out it is the same in the stable-2.6.32 tree as well.

    J

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Greg KH]

On Wed, Sep 29, 2010 at 08:48:07AM -0700, Jeremy Fitzhardinge wrote:
>  On 09/28/2010 11:29 AM, Greg KH wrote:
> > On Mon, Sep 27, 2010 at 11:08:43AM -0700, Jeremy Fitzhardinge wrote:
> >>  On 09/26/2010 11:13 PM, Greg KH wrote:
> >>> On Sun, Sep 26, 2010 at 10:40:44PM -0400, Kyle McMartin wrote:
> >>>> On Thu, Sep 23, 2010 at 04:23:55PM +0100, James Dingwall wrote:
> >>>>>> I was experiencing the same immediate crash with a null pointer
> >>>>>> dereference (log below) on boot with 2.6.35.5.  Reverting
> >>>>>> fb412a178502dc498430723b082a932f797e4763 also resolved the problem for
> >>>>>> me.
> >>>>> Checking the diff from this commit it looks like there is a typo in the patch.  With the following (probably mangled by my client) I get 2.6.35.5 to boot.  Suggest it should be a candidate for .32 and .35 stable trees.
> >>>>>
> >>>> Greg, a spelling mistake cropped into this patch during 2.6.35.5 that
> >>>> James here points out. Adding you to CC since it seems to have missed
> >>>> stable 2.6.35.6...
> >>>>
> >>>> Was in xen-use-percpu-interrupts-for-ipis-and-virqs.patch.
> >>> I don't understand, where did the error come from?  Was it in a patch I
> >>> applied, or did I mess up some patch?  Is the same problem upstream, and
> >>> if not, why not?
> >> I don't know.  It looks like an 'x' was deleted due to an editor
> >> fat-finger or something.  The original upstream patches were cc:d to
> >> stable, so I think you would have got them that way (ie, I don't think I
> >> sent them to you specifically).  Is there an opportunity in your
> >> workflow where this error could have crept in?  The patches should have
> >> been a pretty clean 'git cherry-pick' with no conflicts.
> > Sorry, I don't know what happened, using my scripts, it does drop to an
> > editor so I can add the "signed-off-by" line, so perhaps I accidentally
> > did this in there.
> >
> > It's now in the -stable queue and I'll push out new releases in a few
> > hours.
> >
> 
> I should have pointed out it is the same in the stable-2.6.32 tree as well.

Yes, sorry, I know, I only had the chance to get the .35 release out
yesterday, I'll do the .32 one now.

thanks,

greg k-h

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Jeremy Fitzhardinge]

 On 09/29/2010 05:06 PM, Greg KH wrote:
> On Wed, Sep 29, 2010 at 08:48:07AM -0700, Jeremy Fitzhardinge wrote:
>>
>> I should have pointed out it is the same in the stable-2.6.32 tree as well.
> Yes, sorry, I know, I only had the chance to get the .35 release out
> yesterday, I'll do the .32 one now.
>

Ping?  Looks like the fix hasn't hit stable-2.6.32/master yet.

Thanks,
    J

Re: [stable] PROBLEM: [BISECTED] 2.6.35.5 xen domU panics just after the boot

[Greg KH]

On Fri, Oct 01, 2010 at 09:46:56AM -0700, Jeremy Fitzhardinge wrote:
>  On 09/29/2010 05:06 PM, Greg KH wrote:
> > On Wed, Sep 29, 2010 at 08:48:07AM -0700, Jeremy Fitzhardinge wrote:
> >>
> >> I should have pointed out it is the same in the stable-2.6.32 tree as well.
> > Yes, sorry, I know, I only had the chance to get the .35 release out
> > yesterday, I'll do the .32 one now.
> >
> 
> Ping?  Looks like the fix hasn't hit stable-2.6.32/master yet.

No it hasn't, sorry.  Turns out the Embedded Linux kernel "summit" was
too interesting to get a stable kernel out during it, and I've been in a
train/plane/train/ferry/taxi/bed since then.

Should now be out in a few hours...

thanks,

greg k-h