[ANNOUNCEMENT] Schedule for Xen 3.0.4

[ANNOUNCEMENT] Schedule for Xen 3.0.4

[Keir Fraser]

Folks,

The xen-unstable tree has seen significant progress since 3.0.3 was forked,
particularly in HVM cleanup and bug fixing. SMP and ACPI HVM guests now work
reliably, as does the Windows Vista installer. Many of these improvements
are the result of large patches and interface cleanups that are unsuitable
for backporting to 3.0.3. Since we'd like to make these improvements
available to Xen users as soon as possible, now seems a good time to call a
feature freeze in preparation for 3.0.4.

I propose the following schedule: At the end of November we declare the
feature freeze. We'd like to get in some more patchsets before this cutoff
-- in particular the kexec/kdump patches and the paravirtual framebuffer
support. Following this, and depending on good testing results, we tag
release candidates for 3.0.4 once or twice a week until we have a tree
suitable for final release (hopefully by the middle of December). Meanwhile
we focus on any lurking performance and correctness bugs that have been
introduced during the recent spike in check-in activity.

These plans are part of a more general move to more frequent Xen releases.
By shortening the development phase of our release cycle we can keep tighter
tabs on the status of the Xen tree, resulting in better prediction of
release dates and faster release of new features and enhancements to the Xen
user community!

 Cheers,
 Keir (and the Xen team)

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[John Levon]

On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:

> release candidates for 3.0.4 once or twice a week until we have a tree
> suitable for final release (hopefully by the middle of December). Meanwhile

Does this mean we have less than a month until the Xen API work is
frozen, as was previously the plan for 3.0.4? We're still discussing
aspects of it and AFAIK there's still a lot of work to do in just
finishing defining it, never mind coding it or any "can I actually
program against this" stuff.

regards,
john

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Ewan Mellor]

On Fri, Nov 24, 2006 at 10:42:53AM +0000, John Levon wrote:

> On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> 
> > release candidates for 3.0.4 once or twice a week until we have a tree
> > suitable for final release (hopefully by the middle of December). Meanwhile
> 
> Does this mean we have less than a month until the Xen API work is
> frozen, as was previously the plan for 3.0.4? We're still discussing
> aspects of it and AFAIK there's still a lot of work to do in just
> finishing defining it, never mind coding it or any "can I actually
> program against this" stuff.

No (sorry for the confusion).  Xen 3.0.4 will contain a "preview" of the
Xen-API (i.e. what we've got now) but this will not be the final version of
the protocol or data-model, and will not be declared stable or supported.  We
will ensure that the 3.0.3 protocols are supported in 3.0.4, and will take the
time during this freeze to make sure that backwards compatibility to
3.0.3-based clients is all in place.

Xen 3.0.4 _will_ contain a working version of "the lifecycle patches", that
is, the xm new, start, shutdown, delete, suspend, and resume functionality.

The official release of the Xen-API has now been bumped to Xen 3.0.5,
hopefully February or March 2007.

Ewan.

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Daniel P. Berrange]

On Fri, Nov 24, 2006 at 10:58:12AM +0000, Ewan Mellor wrote:
> On Fri, Nov 24, 2006 at 10:42:53AM +0000, John Levon wrote:
> 
> > On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> > 
> > > release candidates for 3.0.4 once or twice a week until we have a tree
> > > suitable for final release (hopefully by the middle of December). Meanwhile
> > 
> > Does this mean we have less than a month until the Xen API work is
> > frozen, as was previously the plan for 3.0.4? We're still discussing
> > aspects of it and AFAIK there's still a lot of work to do in just
> > finishing defining it, never mind coding it or any "can I actually
> > program against this" stuff.
> 
> No (sorry for the confusion).  Xen 3.0.4 will contain a "preview" of the
> Xen-API (i.e. what we've got now) but this will not be the final version of
> the protocol or data-model, and will not be declared stable or supported.  We
> will ensure that the 3.0.3 protocols are supported in 3.0.4, and will take the
> time during this freeze to make sure that backwards compatibility to
> 3.0.3-based clients is all in place.
> 
> Xen 3.0.4 _will_ contain a working version of "the lifecycle patches", that
> is, the xm new, start, shutdown, delete, suspend, and resume functionality.

Before 3.0.4 comes out, there are a couple of fixes we previously discussed
on xen-api to ensure the lifecycle patches don't break semantics of the
existing SEXPR / XMLRPC APIs.

http://lists.xensource.com/archives/html/xen-api/2006-10/msg00009.html

It has also previously been mentioned that XenSource were planning some 
work on xenstored to address the terrible performance of its transaction
code. It'd be nice to see that in 3.0.4 if its ready ?

Dan.
-- 
|=- Red Hat, Engineering, Emerging Technologies, Boston.  +1 978 392 2496 -=|
|=-           Perl modules: http://search.cpan.org/~danberr/              -=|
|=-               Projects: http://freshmeat.net/~danielpb/               -=|
|=-  GnuPG: 7D3B9505   F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505  -=| 

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Ewan Mellor]

On Fri, Nov 24, 2006 at 03:54:56PM +0000, Daniel P. Berrange wrote:

> On Fri, Nov 24, 2006 at 10:58:12AM +0000, Ewan Mellor wrote:
> > On Fri, Nov 24, 2006 at 10:42:53AM +0000, John Levon wrote:
> > 
> > > On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> > > 
> > > > release candidates for 3.0.4 once or twice a week until we have a tree
> > > > suitable for final release (hopefully by the middle of December). Meanwhile
> > > 
> > > Does this mean we have less than a month until the Xen API work is
> > > frozen, as was previously the plan for 3.0.4? We're still discussing
> > > aspects of it and AFAIK there's still a lot of work to do in just
> > > finishing defining it, never mind coding it or any "can I actually
> > > program against this" stuff.
> > 
> > No (sorry for the confusion).  Xen 3.0.4 will contain a "preview" of the
> > Xen-API (i.e. what we've got now) but this will not be the final version of
> > the protocol or data-model, and will not be declared stable or supported.  We
> > will ensure that the 3.0.3 protocols are supported in 3.0.4, and will take the
> > time during this freeze to make sure that backwards compatibility to
> > 3.0.3-based clients is all in place.
> > 
> > Xen 3.0.4 _will_ contain a working version of "the lifecycle patches", that
> > is, the xm new, start, shutdown, delete, suspend, and resume functionality.
> 
> Before 3.0.4 comes out, there are a couple of fixes we previously discussed
> on xen-api to ensure the lifecycle patches don't break semantics of the
> existing SEXPR / XMLRPC APIs.
> 
> http://lists.xensource.com/archives/html/xen-api/2006-10/msg00009.html

Yes, we'll get those fixed for 3.0.4.

> It has also previously been mentioned that XenSource were planning some 
> work on xenstored to address the terrible performance of its transaction
> code. It'd be nice to see that in 3.0.4 if its ready ?

There are two different fixes for performance problems that you've been
suffering.  One is a general improvement to xenstore to improve transaction
performance -- I'm afraid that that one isn't going to be ready for 3.0.4.
The other fix, which is in unstable now, is a fix to xm list so that it does
not hit the store unnecessarily.  xm list now explicitly sets a flag to
indicate whether it is doing a "short" or a "long" list, and the short one
does not hit the store at all.  This improves the performance of that call by
a very large factor.  You should consider setting that flag yourself in new
applications.

Ewan.

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Daniel P. Berrange]

On Fri, Nov 24, 2006 at 04:34:47PM +0000, Ewan Mellor wrote:
> On Fri, Nov 24, 2006 at 03:54:56PM +0000, Daniel P. Berrange wrote:
> 
> > On Fri, Nov 24, 2006 at 10:58:12AM +0000, Ewan Mellor wrote:
> > > On Fri, Nov 24, 2006 at 10:42:53AM +0000, John Levon wrote:
> > > 
> > > > On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> > > > 
> > > > > release candidates for 3.0.4 once or twice a week until we have a tree
> > > > > suitable for final release (hopefully by the middle of December). Meanwhile
> > > > 
> > > > Does this mean we have less than a month until the Xen API work is
> > > > frozen, as was previously the plan for 3.0.4? We're still discussing
> > > > aspects of it and AFAIK there's still a lot of work to do in just
> > > > finishing defining it, never mind coding it or any "can I actually
> > > > program against this" stuff.
> > > 
> > > No (sorry for the confusion).  Xen 3.0.4 will contain a "preview" of the
> > > Xen-API (i.e. what we've got now) but this will not be the final version of
> > > the protocol or data-model, and will not be declared stable or supported.  We
> > > will ensure that the 3.0.3 protocols are supported in 3.0.4, and will take the
> > > time during this freeze to make sure that backwards compatibility to
> > > 3.0.3-based clients is all in place.
> > > 
> > > Xen 3.0.4 _will_ contain a working version of "the lifecycle patches", that
> > > is, the xm new, start, shutdown, delete, suspend, and resume functionality.
> > 
> > Before 3.0.4 comes out, there are a couple of fixes we previously discussed
> > on xen-api to ensure the lifecycle patches don't break semantics of the
> > existing SEXPR / XMLRPC APIs.
> > 
> > http://lists.xensource.com/archives/html/xen-api/2006-10/msg00009.html
> 
> Yes, we'll get those fixed for 3.0.4.

Excellant, thanks.

> > It has also previously been mentioned that XenSource were planning some 
> > work on xenstored to address the terrible performance of its transaction
> > code. It'd be nice to see that in 3.0.4 if its ready ?
> 
> There are two different fixes for performance problems that you've been
> suffering.  One is a general improvement to xenstore to improve transaction
> performance -- I'm afraid that that one isn't going to be ready for 3.0.4.
> The other fix, which is in unstable now, is a fix to xm list so that it does
> not hit the store unnecessarily.  xm list now explicitly sets a flag to
> indicate whether it is doing a "short" or a "long" list, and the short one
> does not hit the store at all.  This improves the performance of that call by
> a very large factor.  You should consider setting that flag yourself in new
> applications.

Ok, I'll try out that approach and see how much diference it makes - it
does sound like it ought to solve the common case problem I see for listing
domains.

Dan.
-- 
|=- Red Hat, Engineering, Emerging Technologies, Boston.  +1 978 392 2496 -=|
|=-           Perl modules: http://search.cpan.org/~danberr/              -=|
|=-               Projects: http://freshmeat.net/~danielpb/               -=|
|=-  GnuPG: 7D3B9505   F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505  -=| 

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Ewan Mellor]

On Fri, Nov 24, 2006 at 04:34:47PM +0000, Ewan Mellor wrote:

> On Fri, Nov 24, 2006 at 03:54:56PM +0000, Daniel P. Berrange wrote:
> 
> > On Fri, Nov 24, 2006 at 10:58:12AM +0000, Ewan Mellor wrote:
> > > On Fri, Nov 24, 2006 at 10:42:53AM +0000, John Levon wrote:
> > > 
> > > > On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> > > > 
> > > > > release candidates for 3.0.4 once or twice a week until we have a tree
> > > > > suitable for final release (hopefully by the middle of December). Meanwhile
> > > > 
> > > > Does this mean we have less than a month until the Xen API work is
> > > > frozen, as was previously the plan for 3.0.4? We're still discussing
> > > > aspects of it and AFAIK there's still a lot of work to do in just
> > > > finishing defining it, never mind coding it or any "can I actually
> > > > program against this" stuff.
> > > 
> > > No (sorry for the confusion).  Xen 3.0.4 will contain a "preview" of the
> > > Xen-API (i.e. what we've got now) but this will not be the final version of
> > > the protocol or data-model, and will not be declared stable or supported.  We
> > > will ensure that the 3.0.3 protocols are supported in 3.0.4, and will take the
> > > time during this freeze to make sure that backwards compatibility to
> > > 3.0.3-based clients is all in place.
> > > 
> > > Xen 3.0.4 _will_ contain a working version of "the lifecycle patches", that
> > > is, the xm new, start, shutdown, delete, suspend, and resume functionality.
> > 
> > Before 3.0.4 comes out, there are a couple of fixes we previously discussed
> > on xen-api to ensure the lifecycle patches don't break semantics of the
> > existing SEXPR / XMLRPC APIs.
> > 
> > http://lists.xensource.com/archives/html/xen-api/2006-10/msg00009.html
> 
> Yes, we'll get those fixed for 3.0.4.

These are fixed now, and on their way through testing.  Let me know if you
have any more incompatibility issues with libvirt, and I'll get them fixed in
time for the 3.0.4 freeze.

Ewan.

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Daniel P. Berrange]

On Tue, Nov 28, 2006 at 02:26:23PM +0000, Ewan Mellor wrote:
> On Fri, Nov 24, 2006 at 04:34:47PM +0000, Ewan Mellor wrote:
> 
> > On Fri, Nov 24, 2006 at 03:54:56PM +0000, Daniel P. Berrange wrote:
> > 
> > > On Fri, Nov 24, 2006 at 10:58:12AM +0000, Ewan Mellor wrote:
> > > > On Fri, Nov 24, 2006 at 10:42:53AM +0000, John Levon wrote:
> > > > 
> > > > > On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> > > > > 
> > > > > > release candidates for 3.0.4 once or twice a week until we have a tree
> > > > > > suitable for final release (hopefully by the middle of December). Meanwhile
> > > > > 
> > > > > Does this mean we have less than a month until the Xen API work is
> > > > > frozen, as was previously the plan for 3.0.4? We're still discussing
> > > > > aspects of it and AFAIK there's still a lot of work to do in just
> > > > > finishing defining it, never mind coding it or any "can I actually
> > > > > program against this" stuff.
> > > > 
> > > > No (sorry for the confusion).  Xen 3.0.4 will contain a "preview" of the
> > > > Xen-API (i.e. what we've got now) but this will not be the final version of
> > > > the protocol or data-model, and will not be declared stable or supported.  We
> > > > will ensure that the 3.0.3 protocols are supported in 3.0.4, and will take the
> > > > time during this freeze to make sure that backwards compatibility to
> > > > 3.0.3-based clients is all in place.
> > > > 
> > > > Xen 3.0.4 _will_ contain a working version of "the lifecycle patches", that
> > > > is, the xm new, start, shutdown, delete, suspend, and resume functionality.
> > > 
> > > Before 3.0.4 comes out, there are a couple of fixes we previously discussed
> > > on xen-api to ensure the lifecycle patches don't break semantics of the
> > > existing SEXPR / XMLRPC APIs.
> > > 
> > > http://lists.xensource.com/archives/html/xen-api/2006-10/msg00009.html
> > 
> > Yes, we'll get those fixed for 3.0.4.
> 
> These are fixed now, and on their way through testing.  Let me know if you
> have any more incompatibility issues with libvirt, and I'll get them fixed in
> time for the 3.0.4 freeze.

Thanks for getting those in so quickly - I'll do some testing with it today
/ tomorrow to make sure everything's now working as expected.

Regards,
Dan.
-- 
|=- Red Hat, Engineering, Emerging Technologies, Boston.  +1 978 392 2496 -=|
|=-           Perl modules: http://search.cpan.org/~danberr/              -=|
|=-               Projects: http://freshmeat.net/~danielpb/               -=|
|=-  GnuPG: 7D3B9505   F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505  -=| 

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Daniel P. Berrange]

On Tue, Nov 28, 2006 at 02:26:23PM +0000, Ewan Mellor wrote:
> On Fri, Nov 24, 2006 at 04:34:47PM +0000, Ewan Mellor wrote:
> 
> > On Fri, Nov 24, 2006 at 03:54:56PM +0000, Daniel P. Berrange wrote:
> > 
> > > On Fri, Nov 24, 2006 at 10:58:12AM +0000, Ewan Mellor wrote:
> > > > On Fri, Nov 24, 2006 at 10:42:53AM +0000, John Levon wrote:
> > > > 
> > > > > On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> > > > > 
> > > > > > release candidates for 3.0.4 once or twice a week until we have a tree
> > > > > > suitable for final release (hopefully by the middle of December). Meanwhile
> > > > > 
> > > > > Does this mean we have less than a month until the Xen API work is
> > > > > frozen, as was previously the plan for 3.0.4? We're still discussing
> > > > > aspects of it and AFAIK there's still a lot of work to do in just
> > > > > finishing defining it, never mind coding it or any "can I actually
> > > > > program against this" stuff.
> > > > 
> > > > No (sorry for the confusion).  Xen 3.0.4 will contain a "preview" of the
> > > > Xen-API (i.e. what we've got now) but this will not be the final version of
> > > > the protocol or data-model, and will not be declared stable or supported.  We
> > > > will ensure that the 3.0.3 protocols are supported in 3.0.4, and will take the
> > > > time during this freeze to make sure that backwards compatibility to
> > > > 3.0.3-based clients is all in place.
> > > > 
> > > > Xen 3.0.4 _will_ contain a working version of "the lifecycle patches", that
> > > > is, the xm new, start, shutdown, delete, suspend, and resume functionality.
> > > 
> > > Before 3.0.4 comes out, there are a couple of fixes we previously discussed
> > > on xen-api to ensure the lifecycle patches don't break semantics of the
> > > existing SEXPR / XMLRPC APIs.
> > > 
> > > http://lists.xensource.com/archives/html/xen-api/2006-10/msg00009.html
> > 
> > Yes, we'll get those fixed for 3.0.4.
> 
> These are fixed now, and on their way through testing.  Let me know if you
> have any more incompatibility issues with libvirt, and I'll get them fixed in
> time for the 3.0.4 freeze.

Just tested a build of the latest xen-unstable with these patches and
libvirt looks like it is operating correctly with the lifecycle patches
now.

Regards,
Dan.
-- 
|=- Red Hat, Engineering, Emerging Technologies, Boston.  +1 978 392 2496 -=|
|=-           Perl modules: http://search.cpan.org/~danberr/              -=|
|=-               Projects: http://freshmeat.net/~danielpb/               -=|
|=-  GnuPG: 7D3B9505   F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505  -=| 

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Jan Beulich]

Does this imply 32-on-64 will not be in 3.0.4? Jan

>>> Keir Fraser <keir@xensource.com> 24.11.06 11:28 >>>
Folks,

The xen-unstable tree has seen significant progress since 3.0.3 was forked,
particularly in HVM cleanup and bug fixing. SMP and ACPI HVM guests now work
reliably, as does the Windows Vista installer. Many of these improvements
are the result of large patches and interface cleanups that are unsuitable
for backporting to 3.0.3. Since we'd like to make these improvements
available to Xen users as soon as possible, now seems a good time to call a
feature freeze in preparation for 3.0.4.

I propose the following schedule: At the end of November we declare the
feature freeze. We'd like to get in some more patchsets before this cutoff
-- in particular the kexec/kdump patches and the paravirtual framebuffer
support. Following this, and depending on good testing results, we tag
release candidates for 3.0.4 once or twice a week until we have a tree
suitable for final release (hopefully by the middle of December). Meanwhile
we focus on any lurking performance and correctness bugs that have been
introduced during the recent spike in check-in activity.

These plans are part of a more general move to more frequent Xen releases.
By shortening the development phase of our release cycle we can keep tighter
tabs on the status of the Xen tree, resulting in better prediction of
release dates and faster release of new features and enhancements to the Xen
user community!

 Cheers,
 Keir (and the Xen team)


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

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Chuck Short]

What about using paravirt-ops with xen 3.0.4?

Regards
chuck

On 11/24/06, Keir Fraser <keir@xensource.com> wrote:
> Folks,
>
> The xen-unstable tree has seen significant progress since 3.0.3 was forked,
> particularly in HVM cleanup and bug fixing. SMP and ACPI HVM guests now work
> reliably, as does the Windows Vista installer. Many of these improvements
> are the result of large patches and interface cleanups that are unsuitable
> for backporting to 3.0.3. Since we'd like to make these improvements
> available to Xen users as soon as possible, now seems a good time to call a
> feature freeze in preparation for 3.0.4.
>
> I propose the following schedule: At the end of November we declare the
> feature freeze. We'd like to get in some more patchsets before this cutoff
> -- in particular the kexec/kdump patches and the paravirtual framebuffer
> support. Following this, and depending on good testing results, we tag
> release candidates for 3.0.4 once or twice a week until we have a tree
> suitable for final release (hopefully by the middle of December). Meanwhile
> we focus on any lurking performance and correctness bugs that have been
> introduced during the recent spike in check-in activity.
>
> These plans are part of a more general move to more frequent Xen releases.
> By shortening the development phase of our release cycle we can keep tighter
> tabs on the status of the Xen tree, resulting in better prediction of
> release dates and faster release of new features and enhancements to the Xen
> user community!
>
>  Cheers,
>  Keir (and the Xen team)
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Sean Dague]

[-- Attachment #1.1: Type: text/plain, Size: 1162 bytes --]

On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> Folks,
> 
> The xen-unstable tree has seen significant progress since 3.0.3 was forked,
> particularly in HVM cleanup and bug fixing. SMP and ACPI HVM guests now work
> reliably, as does the Windows Vista installer. Many of these improvements
> are the result of large patches and interface cleanups that are unsuitable
> for backporting to 3.0.3. Since we'd like to make these improvements
> available to Xen users as soon as possible, now seems a good time to call a
> feature freeze in preparation for 3.0.4.

What about the possibility of reving the kernel to something more modern? 
For instance, the nforce5 drivers for Linux didn't make it until 2.6.17, and
didn't get truly stable until 2.6.18.  

I think it would be advantageous if xen-unstable was using a kernel closer
to those being shipped in modern distros, so that xen-unstable testing was
more applicable to current distros, and visa versa.

	-Sean

-- 
Sean Dague
IBM Linux Technology Center                     email: japh@us.ibm.com
Open Hypervisor Team                           alt: sldague@us.ibm.com

[-- Attachment #1.2: Type: application/pgp-signature, Size: 191 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

RE: Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Ian Pratt]

> > The xen-unstable tree has seen significant progress since 3.0.3 was
> forked,
> > particularly in HVM cleanup and bug fixing. SMP and ACPI HVM guests
now
> work
> > reliably, as does the Windows Vista installer. Many of these
improvements
> > are the result of large patches and interface cleanups that are
> unsuitable
> > for backporting to 3.0.3. Since we'd like to make these improvements
> > available to Xen users as soon as possible, now seems a good time to
call
> > feature freeze in preparation for 3.0.4.
> 
> What about the possibility of reving the kernel to something more
modern?
> For instance, the nforce5 drivers for Linux didn't make it until
2.6.17,
> and didn't get truly stable until 2.6.18.
> 
> I think it would be advantageous if xen-unstable was using a kernel
closer
> to those being shipped in modern distros, so that xen-unstable testing
was
> more applicable to current distros, and visa versa.

We really need to get away from having a linux tree in xen-unstable
anyhow. It would be cool if by the time that 3.0.4 goes out we have the
out-of-tree linux 2.6.18/19 level up to the same spec as our 2.6.16. We
can update the top-level Makefiles to make it very easy to build
out-of-tree kernels. We can also make 2.6.18 RPMs and tar balls
available for download. None of this actually has to be directly coupled
to the 3.0.4 release.  

Ian

Re: Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Keir Fraser]

On 29/11/06 21:36, "Ian Pratt" <m+Ian.Pratt@cl.cam.ac.uk> wrote:

> We really need to get away from having a linux tree in xen-unstable
> anyhow. It would be cool if by the time that 3.0.4 goes out we have the
> out-of-tree linux 2.6.18/19 level up to the same spec as our 2.6.16. We
> can update the top-level Makefiles to make it very easy to build
> out-of-tree kernels. We can also make 2.6.18 RPMs and tar balls
> available for download. None of this actually has to be directly coupled
> to the 3.0.4 release.

Yes, we really should decouple Xen releases from guest-OS development.

 -- Keir

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Pasi Kärkkäinen]

Hi!

What's the status of (live) migration of fully virtualized guests? Will it
be in 3.0.4 ?

Thanks.

-- Pasi

On Fri, Nov 24, 2006 at 10:28:43AM +0000, Keir Fraser wrote:
> Folks,
> 
> The xen-unstable tree has seen significant progress since 3.0.3 was forked,
> particularly in HVM cleanup and bug fixing. SMP and ACPI HVM guests now work
> reliably, as does the Windows Vista installer. Many of these improvements
> are the result of large patches and interface cleanups that are unsuitable
> for backporting to 3.0.3. Since we'd like to make these improvements
> available to Xen users as soon as possible, now seems a good time to call a
> feature freeze in preparation for 3.0.4.
> 
> I propose the following schedule: At the end of November we declare the
> feature freeze. We'd like to get in some more patchsets before this cutoff
> -- in particular the kexec/kdump patches and the paravirtual framebuffer
> support. Following this, and depending on good testing results, we tag
> release candidates for 3.0.4 once or twice a week until we have a tree
> suitable for final release (hopefully by the middle of December). Meanwhile
> we focus on any lurking performance and correctness bugs that have been
> introduced during the recent spike in check-in activity.
> 
> These plans are part of a more general move to more frequent Xen releases.
> By shortening the development phase of our release cycle we can keep tighter
> tabs on the status of the Xen tree, resulting in better prediction of
> release dates and faster release of new features and enhancements to the Xen
> user community!
> 
>  Cheers,
>  Keir (and the Xen team)
> 

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Keir Fraser]

On 30/11/06 14:59, "Pasi Kärkkäinen" <pasik@iki.fi> wrote:

> Hi!
> 
> What's the status of (live) migration of fully virtualized guests? Will it
> be in 3.0.4 ?

There will be no HVM save/restore or migration in 3.0.4.

 -- Keir

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Pasi Kärkkäinen]

On Thu, Nov 30, 2006 at 03:02:07PM +0000, Keir Fraser wrote:
> 
> 
> 
> On 30/11/06 14:59, "Pasi Kärkkäinen" <pasik@iki.fi> wrote:
> 
> > Hi!
> > 
> > What's the status of (live) migration of fully virtualized guests? Will it
> > be in 3.0.4 ?
> 
> There will be no HVM save/restore or migration in 3.0.4.
> 
>  -- Keir
> 

OK. Thanks for the reply. 

Is it planned for 3.0.5 ?

-- Pasi

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Keir Fraser]

On 30/11/06 15:08, "Pasi Kärkkäinen" <pasik@iki.fi> wrote:

>> There will be no HVM save/restore or migration in 3.0.4.
>> 
>>  -- Keir
>> 
> 
> OK. Thanks for the reply.
> 
> Is it planned for 3.0.5 ?

Hvm save/restore certainly is. Hopefully we'll have migration at that point
too.

 -- Keir

RE: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Ian Pratt]

> > There will be no HVM save/restore or migration in 3.0.4.
> >
> OK. Thanks for the reply.
> 
> Is it planned for 3.0.5 ?

Absolutely. The patches weren't quite ready, and 3.0.4 was a very opportunistic release based on the tree being in a good state currently.

Thanks,
Ian 

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Keir Fraser]

On 24/11/06 11:38, "Gerd Hoffmann" <kraxel@suse.de> wrote:

>   Hi,
> 
>> I propose the following schedule: At the end of November we declare the
>> feature freeze. We'd like to get in some more patchsets before this cutoff
>> -- in particular the kexec/kdump patches and the paravirtual framebuffer
>> support.
> 
> kexec/kdump means the host kexec?

Yes.

> What about the new domain builder, domU kexec and 32-on-64?

I don't think domU kexec or 32-on-64 will make it. Maybe some of the domain
builder patches could: Emmanuel's been looking those so perhaps he has an
opinion on it, but really I don't see us taking the wholesale rewrite for
3.0.4. After all, the primary benefit is when it's coupled with the 32-on-64
patches which are very unlikely to be in 3.0.4.

However, the advantage of more regular releases is that 3.0.5 won't actually
be that far out. We hope to release that by early March.

 -- Keir

> cheers,
>   Gerd

Re: [ANNOUNCEMENT] Schedule for Xen 3.0.4

[Gerd Hoffmann]

  Hi,

>> What about the new domain builder, domU kexec and 32-on-64?
> 
> I don't think domU kexec or 32-on-64 will make it.

domU kexec most likely probably needs some adjustments anyway after dom0
kexec merge ...

> Maybe some of the domain
> builder patches could: Emmanuel's been looking those so perhaps he has an
> opinion on it, but really I don't see us taking the wholesale rewrite for
> 3.0.4.

Well, *merging* the new domain builder shouldn't be a problem.  No "flag
day" switch required, it can happily live side-by-side with the old one.
 Then we can switch over users one-by-one to the new code as they get
ready, most likely not all of them for 3.0.4.

> After all, the primary benefit is when it's coupled with the 32-on-64
> patches which are very unlikely to be in 3.0.4.

Well, I've seen some ia64 guys want to play with bigendian domains, the
new domain builder (especially the rewritten ELF parser) can handle that
too ...

One of the reasons I want to have it merged soon (preferably before
3.0.4) is to make it easier to have the non-x86 (and other) people look
at it, so we maybe can have everyone using the new code by 3.0.5 and can
drop the old builder code at that point in time.

Merging it also makes it easier to test domU kexec as the userspace tool
 will be able to work with the 3.0.4 libraries.

cheers,
  Gerd

-- 
Gerd Hoffmann <kraxel@suse.de>
http://www.suse.de/~kraxel/julika-dora.jpeg