preparing for 4.2.3 and 4.1.6

preparing for 4.2.3 and 4.1.6

[Jan Beulich]

Aiming at a release some time in August, I'd like to cut RC1-s by
Thursday next week (as I'll be on a two week vacation starting
Friday).

Please indicate any bug fixes that so far may have been missed
in the backports already done. A few fixes are currently stuck in
master's staging branch (or pending to make it there) - these
don't need to be named explicitly, I'm already planning to pull over
the hypervisor ones as soon as they get out of staging. To my
knowledge there hasn't been a run through tools side backports
yet - whether that's a sign of there not being much to backport
(apart from security fixes which are already there) I can't tell.

Please note that according to the current plans and with 4.3
having got released, 4.1.6 is expected to be the last
Xenproject.org release of the 4.1 series.

Jan

Re: preparing for 4.2.3 and 4.1.6

[Ian Murray]

----- Original Message -----
> From: Jan Beulich <JBeulich@suse.com>
> To: xen-devel <xen-devel@lists.xenproject.org>
> Cc: Ian Jackson <Ian.Jackson@eu.citrix.com>; Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> Sent: Friday, 12 July 2013, 9:47
> Subject: [Xen-devel] preparing for 4.2.3 and 4.1.6
> 
> Aiming at a release some time in August, I'd like to cut RC1-s by
> Thursday next week (as I'll be on a two week vacation starting
> Friday).
> 
> Please indicate any bug fixes that so far may have been missed
> in the backports already done. A few fixes are currently stuck in
> master's staging branch (or pending to make it there) - these
> don't need to be named explicitly, I'm already planning to pull over
> the hypervisor ones as soon as they get out of staging. To my
> knowledge there hasn't been a run through tools side backports
> yet - whether that's a sign of there not being much to backport
> (apart from security fixes which are already there) I can't tell.
> 
> Please note that according to the current plans and with 4.3
> having got released, 4.1.6 is expected to be the last
> Xenproject.org release of the 4.1 series.
> 

I prepared a partial backport patch for a bug that I came across. Ian Campbell isolated the hunk from a bigger change, which I tested. Ian subsequently asked me to prepare a backport patch. The issue did not occur for me in 4.1.x (can't check exact version as this machine is down at the moment with a hw fault).

[PATCH] [Backport 4.2.x] Fix issue with 'xl list -l' showing domids as -1 when using SXP

http://www.gossamer-threads.com/lists/xen/devel/289351


> Jan
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
> 

Re: preparing for 4.2.3 and 4.1.6

[Jan Beulich]

>>> On 12.07.13 at 11:17, Ian Murray <murrayie@yahoo.co.uk> wrote:
> I prepared a partial backport patch for a bug that I came across. Ian 
> Campbell isolated the hunk from a bigger change, which I tested. Ian 
> subsequently asked me to prepare a backport patch. The issue did not occur 
> for me in 4.1.x (can't check exact version as this machine is down at the 
> moment with a hw fault).
> 
> [PATCH] [Backport 4.2.x] Fix issue with 'xl list -l' showing domids as -1 when 
> using SXP
> 
> http://www.gossamer-threads.com/lists/xen/devel/289351 

As Andrew Cooper stated in a reply to this, IanJ is going to have
to decide whether to take this.

Jan

Re: preparing for 4.2.3 and 4.1.6

[Ian Murray]

----- Original Message -----
> From: Jan Beulich <JBeulich@suse.com>
> To: Ian Murray <murrayie@yahoo.co.uk>
> Cc: Ian Jackson <Ian.Jackson@eu.citrix.com>; Stefano Stabellini <stefano.stabellini@eu.citrix.com>; xen-devel <xen-devel@lists.xenproject.org>
> Sent: Friday, 12 July 2013, 10:22
> Subject: Re: [Xen-devel] preparing for 4.2.3 and 4.1.6
> 
>>>>  On 12.07.13 at 11:17, Ian Murray <murrayie@yahoo.co.uk> 
> wrote:
>>  I prepared a partial backport patch for a bug that I came across. Ian 
>>  Campbell isolated the hunk from a bigger change, which I tested. Ian 
>>  subsequently asked me to prepare a backport patch. The issue did not occur 
>>  for me in 4.1.x (can't check exact version as this machine is down at 
> the 
>>  moment with a hw fault).
>> 
>>  [PATCH] [Backport 4.2.x] Fix issue with 'xl list -l' showing domids 
> as -1 when 
>>  using SXP
>> 
>>  http://www.gossamer-threads.com/lists/xen/devel/289351 
> 
> As Andrew Cooper stated in a reply to this, IanJ is going to have
> to decide whether to take this.
> 
> Jan
>

Understood, but it was a while ago, I presumed it was in the 'lost' category, rather than 'to be decided/rejected/accepted'.

Re: preparing for 4.2.3 and 4.1.6

[Lars Kurth]

Jan,
just asking the question whether you want to build test Days into the 
4.2.3 and 4.1.6 cycles.
Lars

Re: preparing for 4.2.3 and 4.1.6

[Jan Beulich]

>>> On 12.07.13 at 12:05, Lars Kurth <lars.kurth@xen.org> wrote:
> just asking the question whether you want to build test Days into the 
> 4.2.3 and 4.1.6 cycles.

Certainly desirable, if it's not going to need to be me to spend
meaningful amounts of time driving these (which I don't mean to
extend to the need to deal with eventual fallout - that of course
I'm absolutely ready to do even if time is otherwise tight).

Jan

Re: preparing for 4.2.3 and 4.1.6

[Lars Kurth]

Jan,
I am confident that we can manage driving the test days in between 
Dario, Russell and me. It does require a reliable schedule for RC's as 
well as a list of bug fixes that we need to highlight as "please test 
XYZ for RCx" in the Test Day Description. So there would be some impact 
on you in helping pull that info together: but I'd assume that you would 
be on top of the delta between say 4.2.2 and 4.2.3, etc. anyway.
Lars

On 12/07/2013 11:22, Jan Beulich wrote:
>>>> On 12.07.13 at 12:05, Lars Kurth <lars.kurth@xen.org> wrote:
>> just asking the question whether you want to build test Days into the
>> 4.2.3 and 4.1.6 cycles.
> Certainly desirable, if it's not going to need to be me to spend
> meaningful amounts of time driving these (which I don't mean to
> extend to the need to deal with eventual fallout - that of course
> I'm absolutely ready to do even if time is otherwise tight).
>
> Jan
>

Re: preparing for 4.2.3 and 4.1.6

[Ian Murray]

----- Original Message -----
> From: Jan Beulich <JBeulich@suse.com>
> To: Ian Murray <murrayie@yahoo.co.uk>
> Cc: xen-devel <xen-devel@lists.xenproject.org>; Ian Jackson <Ian.Jackson@eu.citrix.com>; Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> Sent: Friday, 12 July 2013, 10:22
> Subject: Re: [Xen-devel] preparing for 4.2.3 and 4.1.6
> 
>>>>  On 12.07.13 at 11:17, Ian Murray <murrayie@yahoo.co.uk> 
> wrote:
>>  I prepared a partial backport patch for a bug that I came across. Ian 
>>  Campbell isolated the hunk from a bigger change, which I tested. Ian 
>>  subsequently asked me to prepare a backport patch. The issue did not occur 
>>  for me in 4.1.x (can't check exact version as this machine is down at 
> the 
>>  moment with a hw fault).
>> 
>>  [PATCH] [Backport 4.2.x] Fix issue with 'xl list -l' showing domids 
> as -1 when 
>>  using SXP
>> 
>>  http://www.gossamer-threads.com/lists/xen/devel/289351 
> 
> As Andrew Cooper stated in a reply to this, IanJ is going to have
> to decide whether to take this.
> 

Ian J, could you please offer comment as to whether you wish to include this or not?

Thanks,

Ian.


> Jan
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
> 

Re: preparing for 4.2.3 and 4.1.6

[Don Slutz]

On 07/12/13 04:47, Jan Beulich wrote:
> Aiming at a release some time in August, I'd like to cut RC1-s by
> Thursday next week (as I'll be on a two week vacation starting
> Friday).
>
> Please indicate any bug fixes that so far may have been missed
> in the backports already done. A few fixes are currently stuck in
> master's staging branch (or pending to make it there) - these
> don't need to be named explicitly, I'm already planning to pull over
> the hypervisor ones as soon as they get out of staging. To my
> knowledge there hasn't been a run through tools side backports
> yet - whether that's a sign of there not being much to backport
> (apart from security fixes which are already there) I can't tell.
>
> Please note that according to the current plans and with 4.3
> having got released, 4.1.6 is expected to be the last
> Xenproject.org release of the 4.1 series.
>
> Jan
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel

I think it is important to get the bug fix:

*[PATCH] libxl_json: Fix backport of JSON_BOOL to 4.2.2

*http://www.gossamer-threads.com/lists/xen/devel/289420 into 4.2.3

Into 4.2.3.  The code does not exist at 4.1
    -Don Slutz

Re: preparing for 4.2.3 and 4.1.6

[Ian Jackson]

Don Slutz writes ("Re: [Xen-devel] preparing for 4.2.3 and 4.1.6"):
> I think it is important to get the bug fix:
> 
> *[PATCH] libxl_json: Fix backport of JSON_BOOL to 4.2.2
> 
> *http://www.gossamer-threads.com/lists/xen/devel/289420 into 4.2.3
> 
> Into 4.2.3.  The code does not exist at 4.1

Done, thanks.

Ian.

Re: preparing for 4.2.3 and 4.1.6

[Konrad Rzeszutek Wilk]

On Fri, Jul 12, 2013 at 09:47:34AM +0100, Jan Beulich wrote:
> Aiming at a release some time in August, I'd like to cut RC1-s by
> Thursday next week (as I'll be on a two week vacation starting
> Friday).
> 
> Please indicate any bug fixes that so far may have been missed
> in the backports already done. A few fixes are currently stuck in
> master's staging branch (or pending to make it there) - these
> don't need to be named explicitly, I'm already planning to pull over
> the hypervisor ones as soon as they get out of staging. To my
> knowledge there hasn't been a run through tools side backports
> yet - whether that's a sign of there not being much to backport
> (apart from security fixes which are already there) I can't tell.
> 
> Please note that according to the current plans and with 4.3
> having got released, 4.1.6 is expected to be the last
> Xenproject.org release of the 4.1 series.

Please backport for tools:
 d513814db6af2b298b8776d7ffc5fb1261e176f4 pygrub/GrubConf: fix boot problem for fedora 19 grub.cfg (2nd attempt)

And for QEMU:
 2133c3847bd2d3e0954e85dadc2ad2606c513cf8 piix4acpi, xen, vcpu hotplug: Split the notification from the changes.
 e28e06225c4ee8793f935b0f72fe2d944a2d9b50 piix4acpi, xen: Clarify that the qemu_set_irq calls just do an IRQ pulse.
 f62079cd7de6ec37f48dfc80fb5906f49fecd6f6 piix4acpi, xen, hotplug: Fix race with ACPI AML code and hotplug.

Thank you!

Re: preparing for 4.2.3 and 4.1.6

[Stefano Stabellini]

On Fri, 12 Jul 2013, Konrad Rzeszutek Wilk wrote:
> On Fri, Jul 12, 2013 at 09:47:34AM +0100, Jan Beulich wrote:
> > Aiming at a release some time in August, I'd like to cut RC1-s by
> > Thursday next week (as I'll be on a two week vacation starting
> > Friday).
> > 
> > Please indicate any bug fixes that so far may have been missed
> > in the backports already done. A few fixes are currently stuck in
> > master's staging branch (or pending to make it there) - these
> > don't need to be named explicitly, I'm already planning to pull over
> > the hypervisor ones as soon as they get out of staging. To my
> > knowledge there hasn't been a run through tools side backports
> > yet - whether that's a sign of there not being much to backport
> > (apart from security fixes which are already there) I can't tell.
> > 
> > Please note that according to the current plans and with 4.3
> > having got released, 4.1.6 is expected to be the last
> > Xenproject.org release of the 4.1 series.
> 
> Please backport for tools:
>  d513814db6af2b298b8776d7ffc5fb1261e176f4 pygrub/GrubConf: fix boot problem for fedora 19 grub.cfg (2nd attempt)
> 
> And for QEMU:
>  2133c3847bd2d3e0954e85dadc2ad2606c513cf8 piix4acpi, xen, vcpu hotplug: Split the notification from the changes.
>  e28e06225c4ee8793f935b0f72fe2d944a2d9b50 piix4acpi, xen: Clarify that the qemu_set_irq calls just do an IRQ pulse.
>  f62079cd7de6ec37f48dfc80fb5906f49fecd6f6 piix4acpi, xen, hotplug: Fix race with ACPI AML code and hotplug.

These qemu-xen-traditional commits seem good backport candidates.

Re: preparing for 4.2.3 and 4.1.6

[Ian Jackson]

Stefano Stabellini writes ("Re: [Xen-devel] preparing for 4.2.3 and 4.1.6"):
> On Fri, 12 Jul 2013, Konrad Rzeszutek Wilk wrote:
> > Please backport for tools:
> >  d513814db6af2b298b8776d7ffc5fb1261e176f4 pygrub/GrubConf: fix boot problem for fedora 19 grub.cfg (2nd attempt)

Done, to 4.1 and 4.2.

> > And for QEMU:
> >  2133c3847bd2d3e0954e85dadc2ad2606c513cf8 piix4acpi, xen, vcpu hotplug: Split the notification from the changes.
> >  e28e06225c4ee8793f935b0f72fe2d944a2d9b50 piix4acpi, xen: Clarify that the qemu_set_irq calls just do an IRQ pulse.
> >  f62079cd7de6ec37f48dfc80fb5906f49fecd6f6 piix4acpi, xen, hotplug: Fix race with ACPI AML code and hotplug.
> 
> These qemu-xen-traditional commits seem good backport candidates.

Also done, to 4.1 and 4.2.

Ian.

Re: preparing for 4.2.3 and 4.1.6

[Daniel Kiper]

On Fri, Jul 12, 2013 at 09:47:34AM +0100, Jan Beulich wrote:
> Aiming at a release some time in August, I'd like to cut RC1-s by
> Thursday next week (as I'll be on a two week vacation starting
> Friday).
>
> Please indicate any bug fixes that so far may have been missed
> in the backports already done. A few fixes are currently stuck in
> master's staging branch (or pending to make it there) - these
> don't need to be named explicitly, I'm already planning to pull over
> the hypervisor ones as soon as they get out of staging. To my
> knowledge there hasn't been a run through tools side backports
> yet - whether that's a sign of there not being much to backport
> (apart from security fixes which are already there) I can't tell.

I think that it will be nice to have:
  - 667d8a84b244d02e9c6a2d02d6a02fc90c2efb4e
    (tools/libfsimage: Fix clean and distclean make targets),
  - 38aabce5424710badc080decca23628442d80673
    (Silently ignore rm and include errors during make distclean),
  - e17295d050110bdbbe0ef19c6e977c8fef7557db
    (docs: Remove tmp files during make clean),
  - 3eb7adb2bad48de7bba4e213c9858de0cbba2fe8
    (stubdom/vtpm: Silently ignore rm errors during make clean),
  - c2eea87c43d1617b2c15c57fce9a64a436679fca
    (stubdom: Do not create dangling links),
  - 43be6972f977660ba9aaf15f78e39c025c9550d8
    (stubdom: Remove xenstore directory during make crossclean),
  - 2bf87074e17e77ca761e3490e5f054175e082e5b
    (stubdom: Clean extras/mini-os directory during make clean),
  - 03b90b005939416463c79a45d91729e8a00742fa
    (tools/xenstat/libxenstat: Remove src/libxenstat.a file during make clean),
  - ea5e515ba19c423e15ca33023cd3c9d2c9aa807f
    (tools/xenmon: Fix typo in Makefile),
  - 38bdfb9197b93262248ff489eed336d80db52b54
    (tools/debugger/kdd: Remove dependencies files during make clean).

Those are build system bug fixes.

Daniel

Re: preparing for 4.2.3 and 4.1.6

[Jan Beulich]

>>> On 15.07.13 at 14:35, Daniel Kiper <dkiper@net-space.pl> wrote:
> I think that it will be nice to have:
>   - 667d8a84b244d02e9c6a2d02d6a02fc90c2efb4e
>     (tools/libfsimage: Fix clean and distclean make targets),
>   - 38aabce5424710badc080decca23628442d80673
>     (Silently ignore rm and include errors during make distclean),
>   - e17295d050110bdbbe0ef19c6e977c8fef7557db
>     (docs: Remove tmp files during make clean),
>   - 3eb7adb2bad48de7bba4e213c9858de0cbba2fe8
>     (stubdom/vtpm: Silently ignore rm errors during make clean),
>   - c2eea87c43d1617b2c15c57fce9a64a436679fca
>     (stubdom: Do not create dangling links),
>   - 43be6972f977660ba9aaf15f78e39c025c9550d8
>     (stubdom: Remove xenstore directory during make crossclean),
>   - 2bf87074e17e77ca761e3490e5f054175e082e5b
>     (stubdom: Clean extras/mini-os directory during make clean),
>   - 03b90b005939416463c79a45d91729e8a00742fa
>     (tools/xenstat/libxenstat: Remove src/libxenstat.a file during make clean),
>   - ea5e515ba19c423e15ca33023cd3c9d2c9aa807f
>     (tools/xenmon: Fix typo in Makefile),
>   - 38bdfb9197b93262248ff489eed336d80db52b54
>     (tools/debugger/kdd: Remove dependencies files during make clean).
> 
> Those are build system bug fixes.

Hmm, I'm not really convinced these are the sorts of changes
we need to backport. But it almost exclusively tools side stuff
anyway, so I'll wanto to hear Ian's opinion on this anyway.

Jan

Re: preparing for 4.2.3 and 4.1.6

[Ian Jackson]

Jan Beulich writes ("Re: [Xen-devel] preparing for 4.2.3 and 4.1.6"):
> On 15.07.13 at 14:35, Daniel Kiper <dkiper@net-space.pl> wrote:
> > I think that it will be nice to have:
...
> Hmm, I'm not really convinced these are the sorts of changes
> we need to backport. But it almost exclusively tools side stuff
> anyway, so I'll wanto to hear Ian's opinion on this anyway.

I think these are bug fixes and so deserving of backports, where they
are low risk.

Ian.

Re: preparing for 4.2.3 and 4.1.6

[Ian Jackson]

Daniel Kiper writes ("Re: [Xen-devel] preparing for 4.2.3 and 4.1.6"):
> I think that it will be nice to have:

Thanks for these suggestions.

Applied to 4.1 and 4.2:

>   - 667d8a84b244d02e9c6a2d02d6a02fc90c2efb4e
>     (tools/libfsimage: Fix clean and distclean make targets),
>   - 38bdfb9197b93262248ff489eed336d80db52b54
>     (tools/debugger/kdd: Remove dependencies files during make clean).

These applied to 4.2 but not to 4.1:

>   - e17295d050110bdbbe0ef19c6e977c8fef7557db
>     (docs: Remove tmp files during make clean),
>   - c2eea87c43d1617b2c15c57fce9a64a436679fca
>     (stubdom: Do not create dangling links),
>   - 03b90b005939416463c79a45d91729e8a00742fa
>     (tools/xenstat/libxenstat: Remove src/libxenstat.a file during make clean),
>   - ea5e515ba19c423e15ca33023cd3c9d2c9aa807f
>     (tools/xenmon: Fix typo in Makefile),

These did not apply to 4.2:

>   - 38aabce5424710badc080decca23628442d80673
>     (Silently ignore rm and include errors during make distclean),
>   - 3eb7adb2bad48de7bba4e213c9858de0cbba2fe8
>     (stubdom/vtpm: Silently ignore rm errors during make clean),
>   - 43be6972f977660ba9aaf15f78e39c025c9550d8
>     (stubdom: Remove xenstore directory during make crossclean),
>   - 2bf87074e17e77ca761e3490e5f054175e082e5b
>     (stubdom: Clean extras/mini-os directory during make clean),

For the patches that didn't apply I haven't investigated why.  For
ones where the patch is applicable, I would welcome a by-hand fixup.

Thanks,
Ian.