linux-next: manual merge of the xen tree with the tip tree

linux-next: manual merge of the xen tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 493 bytes --]

Hi all,

Today's linux-next merge of the xen tree got a conflict in
arch/x86/include/asm/spinlock.h between a series of commits  from the tip
tree and a smaller series of similar commits from the xen tree.

I see that Linus is commenting on these patches at the moment, and its
not easy to resolve the conflicts, so I will just use the xen tree from
next-20110824 for today.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

Re: linux-next: manual merge of the xen tree with the tip tree

[Jeremy Fitzhardinge]

On 08/24/2011 09:24 PM, Stephen Rothwell wrote:
> Hi all,
>
> Today's linux-next merge of the xen tree got a conflict in
> arch/x86/include/asm/spinlock.h between a series of commits from the tip
> tree and a smaller series of similar commits from the xen tree.
>
> I see that Linus is commenting on these patches at the moment, and its
> not easy to resolve the conflicts, so I will just use the xen tree from
> next-20110824 for today.
>

Thanks Stephen; the xen tree ones are more current, and I want to make
sure I didn't screw up any of the cmpxchg/xadd changes in a wider test env.

    J

Re: linux-next: manual merge of the xen tree with the tip tree

[H. Peter Anvin]

On 08/25/2011 11:13 AM, Jeremy Fitzhardinge wrote:
> On 08/24/2011 09:24 PM, Stephen Rothwell wrote:
>> Hi all,
>>
>> Today's linux-next merge of the xen tree got a conflict in
>> arch/x86/include/asm/spinlock.h between a series of commits from the tip
>> tree and a smaller series of similar commits from the xen tree.
>>
>> I see that Linus is commenting on these patches at the moment, and its
>> not easy to resolve the conflicts, so I will just use the xen tree from
>> next-20110824 for today.
>>
> 
> Thanks Stephen; the xen tree ones are more current, and I want to make
> sure I didn't screw up any of the cmpxchg/xadd changes in a wider test env.
> 

Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
should be dropped.

	-hpa

Kernel Warning

[Carsten Schiers]

Is that a Xen thing that anybody's interested in?

BR,
Carsten.

[    0.011657] ------------[ cut here ]------------
[    0.011734] WARNING: at arch/x86/xen/enlighten.c:729 
init_hw_perf_events+0x337/0x3d7()
[    0.011834] Hardware name: M56S-S3
[    0.011908] Modules linked in:
[    0.012000] Pid: 0, comm: swapper Not tainted 2.6.32.44-xen-amd64 #6
[    0.012000] Call Trace:
[    0.012000]  [<ffffffff8104d70f>] ? warn_slowpath_common+0x76/0x8c
[    0.012000]  [<ffffffff814e216e>] ? init_hw_perf_events+0x337/0x3d7
[    0.012000]  [<ffffffff814e1c7b>] ? identify_boot_cpu+0x15/0x3d
[    0.012000]  [<ffffffff814e1e12>] ? check_bugs+0x9/0x2e
[    0.012000]  [<ffffffff814dace3>] ? start_kernel+0x3cb/0x3e5
[    0.012000]  [<ffffffff814dcba9>] ? xen_start_kernel+0x5e1/0x5e7
[    0.012000] ---[ end trace a7919e7f17c0a725 ]---

Re: Kernel Warning

[Ian Campbell]

On Thu, 2011-08-25 at 20:31 +0100, Carsten Schiers wrote:
> Is that a Xen thing that anybody's interested in?

It's a harmless warning but it gets reported a lot and it taints the
kernel which causes some confusion is there is subsequently another
error. I proposed toning it down a bit a couple of days ago:
http://marc.info/?l=xen-devel&m=131400621622691

Ian.

> 
> BR,
> Carsten.
> 
> [    0.011657] ------------[ cut here ]------------
> [    0.011734] WARNING: at arch/x86/xen/enlighten.c:729 
> init_hw_perf_events+0x337/0x3d7()
> [    0.011834] Hardware name: M56S-S3
> [    0.011908] Modules linked in:
> [    0.012000] Pid: 0, comm: swapper Not tainted 2.6.32.44-xen-amd64 #6
> [    0.012000] Call Trace:
> [    0.012000]  [<ffffffff8104d70f>] ? warn_slowpath_common+0x76/0x8c
> [    0.012000]  [<ffffffff814e216e>] ? init_hw_perf_events+0x337/0x3d7
> [    0.012000]  [<ffffffff814e1c7b>] ? identify_boot_cpu+0x15/0x3d
> [    0.012000]  [<ffffffff814e1e12>] ? check_bugs+0x9/0x2e
> [    0.012000]  [<ffffffff814dace3>] ? start_kernel+0x3cb/0x3e5
> [    0.012000]  [<ffffffff814dcba9>] ? xen_start_kernel+0x5e1/0x5e7
> [    0.012000] ---[ end trace a7919e7f17c0a725 ]---
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel

Re: Kernel Warning

[Jeremy Fitzhardinge]

On 08/25/2011 12:39 PM, Ian Campbell wrote:
> On Thu, 2011-08-25 at 20:31 +0100, Carsten Schiers wrote:
>> Is that a Xen thing that anybody's interested in?
> It's a harmless warning but it gets reported a lot and it taints the
> kernel which causes some confusion is there is subsequently another
> error. I proposed toning it down a bit a couple of days ago:
> http://marc.info/?l=xen-devel&m=131400621622691

I'd tend towards just silently eating APIC writes, unless we want to go
through and try to silence each one at the source.

    J

Re: linux-next: manual merge of the xen tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1227 bytes --]

Hi,

On Thu, 25 Aug 2011 11:26:37 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
>
> On 08/25/2011 11:13 AM, Jeremy Fitzhardinge wrote:
> > On 08/24/2011 09:24 PM, Stephen Rothwell wrote:
> >>
> >> Today's linux-next merge of the xen tree got a conflict in
> >> arch/x86/include/asm/spinlock.h between a series of commits from the tip
> >> tree and a smaller series of similar commits from the xen tree.
> >>
> >> I see that Linus is commenting on these patches at the moment, and its
> >> not easy to resolve the conflicts, so I will just use the xen tree from
> >> next-20110824 for today.
> >>
> > 
> > Thanks Stephen; the xen tree ones are more current, and I want to make
> > sure I didn't screw up any of the cmpxchg/xadd changes in a wider test env.
> > 
> 
> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
> should be dropped.

That's a bit tricky as I get a rolled up tip tree.  The best I could do
is revert the commit that merges the x86/spinlocks branch into
auto-latest ...  I'll do that for today (unless something happens to the
tip tree in the next hour).

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

Re: linux-next: manual merge of the xen tree with the tip tree

[H. Peter Anvin]

On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
>>
>> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
>> should be dropped.
> 
> That's a bit tricky as I get a rolled up tip tree.  The best I could do
> is revert the commit that merges the x86/spinlocks branch into
> auto-latest ...  I'll do that for today (unless something happens to the
> tip tree in the next hour).
> 

OK, let me bother Ingo about it.

	-hpa

Re: linux-next: manual merge of the xen tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 735 bytes --]

On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
>
> On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
> >>
> >> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
> >> should be dropped.
> > 
> > That's a bit tricky as I get a rolled up tip tree.  The best I could do
> > is revert the commit that merges the x86/spinlocks branch into
> > auto-latest ...  I'll do that for today (unless something happens to the
> > tip tree in the next hour).
> > 
> 
> OK, let me bother Ingo about it.

For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
tip tree.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

Re: linux-next: manual merge of the xen tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 999 bytes --]

Hi,

On Fri, 26 Aug 2011 12:54:41 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
> >
> > On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
> > >>
> > >> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
> > >> should be dropped.
> > > 
> > > That's a bit tricky as I get a rolled up tip tree.  The best I could do
> > > is revert the commit that merges the x86/spinlocks branch into
> > > auto-latest ...  I'll do that for today (unless something happens to the
> > > tip tree in the next hour).
> > > 
> > 
> > OK, let me bother Ingo about it.
> 
> For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
> tip tree.

I am still doing this in each linux-next, and it doesn't appear to have
been fixed up the the tree on tesla.tglx.de, yet, I think.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

Re: linux-next: manual merge of the xen tree with the tip tree

[Thomas Gleixner]

On Tue, 13 Sep 2011, Stephen Rothwell wrote:
> Hi,
> 
> On Fri, 26 Aug 2011 12:54:41 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
> > >
> > > On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
> > > >>
> > > >> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
> > > >> should be dropped.
> > > > 
> > > > That's a bit tricky as I get a rolled up tip tree.  The best I could do
> > > > is revert the commit that merges the x86/spinlocks branch into
> > > > auto-latest ...  I'll do that for today (unless something happens to the
> > > > tip tree in the next hour).
> > > > 
> > > 
> > > OK, let me bother Ingo about it.
> > 
> > For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
> > tip tree.
> 
> I am still doing this in each linux-next, and it doesn't appear to have
> been fixed up the the tree on tesla.tglx.de, yet, I think.

We'll take it out. Thanks,

      tglx

Re: linux-next: manual merge of the xen tree with the tip tree

[Jeremy Fitzhardinge]

On 09/13/2011 08:07 AM, Thomas Gleixner wrote:
> On Tue, 13 Sep 2011, Stephen Rothwell wrote:
>> Hi,
>>
>> On Fri, 26 Aug 2011 12:54:41 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>> On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
>>>> On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
>>>>>> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
>>>>>> should be dropped.
>>>>> That's a bit tricky as I get a rolled up tip tree.  The best I could do
>>>>> is revert the commit that merges the x86/spinlocks branch into
>>>>> auto-latest ...  I'll do that for today (unless something happens to the
>>>>> tip tree in the next hour).
>>>>>
>>>> OK, let me bother Ingo about it.
>>> For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
>>> tip tree.
>> I am still doing this in each linux-next, and it doesn't appear to have
>> been fixed up the the tree on tesla.tglx.de, yet, I think.
> We'll take it out. 

Actually, the tip x86/spinlocks was the most up-to-date version of those
patches (since hpa had rebased them to a more recent version of mainline).

But never mind.  Stephen, could you use

    git://github.com/jsgf/linux-xen.git upstream/xen

for linux-next instead of the kernel.org xen.git, and I've re-added the
up-to-date spinlock changes there.

Thanks,
    J

Re: linux-next: manual merge of the xen tree with the tip tree

[Thomas Gleixner]

On Tue, 13 Sep 2011, Jeremy Fitzhardinge wrote:

> On 09/13/2011 08:07 AM, Thomas Gleixner wrote:
> > On Tue, 13 Sep 2011, Stephen Rothwell wrote:
> >> Hi,
> >>
> >> On Fri, 26 Aug 2011 12:54:41 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >>> On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
> >>>> On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
> >>>>>> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
> >>>>>> should be dropped.
> >>>>> That's a bit tricky as I get a rolled up tip tree.  The best I could do
> >>>>> is revert the commit that merges the x86/spinlocks branch into
> >>>>> auto-latest ...  I'll do that for today (unless something happens to the
> >>>>> tip tree in the next hour).
> >>>>>
> >>>> OK, let me bother Ingo about it.
> >>> For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
> >>> tip tree.
> >> I am still doing this in each linux-next, and it doesn't appear to have
> >> been fixed up the the tree on tesla.tglx.de, yet, I think.
> > We'll take it out. 
> 
> Actually, the tip x86/spinlocks was the most up-to-date version of those
> patches (since hpa had rebased them to a more recent version of mainline).

Mooo. You tell that after we did a nasty rebase from hell :(
 
Thanks,

	tglx

Re: linux-next: manual merge of the xen tree with the tip tree

[Jeremy Fitzhardinge]

On 09/13/2011 01:56 PM, Thomas Gleixner wrote:
> On Tue, 13 Sep 2011, Jeremy Fitzhardinge wrote:
>
>> On 09/13/2011 08:07 AM, Thomas Gleixner wrote:
>>> On Tue, 13 Sep 2011, Stephen Rothwell wrote:
>>>> Hi,
>>>>
>>>> On Fri, 26 Aug 2011 12:54:41 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>>>> On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
>>>>>> On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
>>>>>>>> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
>>>>>>>> should be dropped.
>>>>>>> That's a bit tricky as I get a rolled up tip tree.  The best I could do
>>>>>>> is revert the commit that merges the x86/spinlocks branch into
>>>>>>> auto-latest ...  I'll do that for today (unless something happens to the
>>>>>>> tip tree in the next hour).
>>>>>>>
>>>>>> OK, let me bother Ingo about it.
>>>>> For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
>>>>> tip tree.
>>>> I am still doing this in each linux-next, and it doesn't appear to have
>>>> been fixed up the the tree on tesla.tglx.de, yet, I think.
>>> We'll take it out. 
>> Actually, the tip x86/spinlocks was the most up-to-date version of those
>> patches (since hpa had rebased them to a more recent version of mainline).
> Mooo. You tell that after we did a nasty rebase from hell :(

I'd been meaning to take it out of my tree to solve Stephen's problem,
but, well, kernel.org.

    J

Re: linux-next: manual merge of the xen tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1761 bytes --]

Hi Jeremy,

On Tue, 13 Sep 2011 13:53:39 -0700 Jeremy Fitzhardinge <jeremy@goop.org> wrote:
>
> On 09/13/2011 08:07 AM, Thomas Gleixner wrote:
> > On Tue, 13 Sep 2011, Stephen Rothwell wrote:
> >>
> >> On Fri, 26 Aug 2011 12:54:41 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >>> On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
> >>>> On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
> >>>>>> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
> >>>>>> should be dropped.
> >>>>> That's a bit tricky as I get a rolled up tip tree.  The best I could do
> >>>>> is revert the commit that merges the x86/spinlocks branch into
> >>>>> auto-latest ...  I'll do that for today (unless something happens to the
> >>>>> tip tree in the next hour).
> >>>>>
> >>>> OK, let me bother Ingo about it.
> >>> For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
> >>> tip tree.
> >> I am still doing this in each linux-next, and it doesn't appear to have
> >> been fixed up the the tree on tesla.tglx.de, yet, I think.
> > We'll take it out. 
> 
> Actually, the tip x86/spinlocks was the most up-to-date version of those
> patches (since hpa had rebased them to a more recent version of mainline).
> 
> But never mind.  Stephen, could you use
> 
>     git://github.com/jsgf/linux-xen.git upstream/xen
> 
> for linux-next instead of the kernel.org xen.git, and I've re-added the
> up-to-date spinlock changes there.

OK, I have switched to this from today.

My understanding is this:  I do *not* need to revert the spinlock changes
from tip anymore, correct?

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

Re: linux-next: manual merge of the xen tree with the tip tree

[Jeremy Fitzhardinge]

On 09/13/2011 05:32 PM, Stephen Rothwell wrote:
> Hi Jeremy,
>
> On Tue, 13 Sep 2011 13:53:39 -0700 Jeremy Fitzhardinge
<jeremy@goop.org> wrote:
>>
>> On 09/13/2011 08:07 AM, Thomas Gleixner wrote:
>>> On Tue, 13 Sep 2011, Stephen Rothwell wrote:
>>>>
>>>> On Fri, 26 Aug 2011 12:54:41 +1000 Stephen Rothwell
<sfr@canb.auug.org.au> wrote:
>>>>> On Thu, 25 Aug 2011 16:12:33 -0700 "H. Peter Anvin" <hpa@zytor.com>
wrote:
>>>>>> On 08/25/2011 04:06 PM, Stephen Rothwell wrote:
>>>>>>>> Stephen: the x86/spinlocks branch in the -tip tree is obsolete and
>>>>>>>> should be dropped.
>>>>>>> That's a bit tricky as I get a rolled up tip tree. The best I
could do
>>>>>>> is revert the commit that merges the x86/spinlocks branch into
>>>>>>> auto-latest ... I'll do that for today (unless something happens
to the
>>>>>>> tip tree in the next hour).
>>>>>>>
>>>>>> OK, let me bother Ingo about it.
>>>>> For today, I have done "git revert -m 1 6f8fa39c81f1" after merging the
>>>>> tip tree.
>>>> I am still doing this in each linux-next, and it doesn't appear to have
>>>> been fixed up the the tree on tesla.tglx.de, yet, I think.
>>> We'll take it out.
>>
>> Actually, the tip x86/spinlocks was the most up-to-date version of those
>> patches (since hpa had rebased them to a more recent version of mainline).
>>
>> But never mind. Stephen, could you use
>>
>> git://github.com/jsgf/linux-xen.git upstream/xen
>>
>> for linux-next instead of the kernel.org xen.git, and I've re-added the
>> up-to-date spinlock changes there.
>
> OK, I have switched to this from today.
>
> My understanding is this: I do *not* need to revert the spinlock changes
> from tip anymore, correct?

Right.  tglx has removed these changes from tip.git (even though they
were OK), and I've reinstated the proper versions in my tree.  The
xen.git on git.kernel.org still has old versions, but I'll sort that out
once it's back online.  I expect it will ultimately go upstream via
tip.git, but we can work that out later too.

Thanks,
    J

Re: linux-next: manual merge of the xen tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 386 bytes --]

Hi all,

On Wed, 14 Sep 2011 10:32:34 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> My understanding is this:  I do *not* need to revert the spinlock changes
> from tip anymore, correct?

Right, having looked at the tip tree, they are not in there any more.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]