IA64 KVM

IA64 KVM

[Caleb Moore]

Hi,

We are interested in an IA64 port of KVM, either experimenting with
someone else's or writing our own. I've tried getting in contact with Yu
Fenghua of Intel but it seems that we will be unable to get hold of
their source code until it is in the main tree.

Is such a merge likely in the near future or should we proceed in
creating our own port?

Thank you,
-- 
Caleb Moore
Research Assistant
School of Computer Science and Engineering
University of NSW, Sydney Australia
<cmoore-YbfuJp6tym7X/JP9YwkgDA@public.gmane.org>


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Zhang, Xiantao]

Caleb Moore wrote:

> Is such a merge likely in the near future or should we proceed in
> creating our own port?

Hi, Caleb 
     We have completed kvm IA64 port until the end of  Aug, and showed a
demo in kvm forum of Aug. Then, we could successfully boot windows, and
Linux guest on open guest firmware, and also got a reasonable
performance. Now, we are pushing the code into the upstream. But, the
whole community  have to be working together to re-frame current kvm
source to accomendate new archs first, since current code is not frendly
to new archs support.  So, the code should be pushing out in very near
future.  :-)
Thanks   --Xiantao 

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Alex Williamson]

On Thu, 2007-11-08 at 18:21 +0800, Zhang, Xiantao wrote:
> Caleb Moore wrote:
> 
> > Is such a merge likely in the near future or should we proceed in
> > creating our own port?
> 
> Hi, Caleb 
>      We have completed kvm IA64 port until the end of  Aug, and showed a
> demo in kvm forum of Aug. Then, we could successfully boot windows, and
> Linux guest on open guest firmware, and also got a reasonable
> performance. Now, we are pushing the code into the upstream. But, the
> whole community  have to be working together to re-frame current kvm
> source to accomendate new archs first, since current code is not frendly
> to new archs support.  So, the code should be pushing out in very near
> future.  :-)

Hi Xiantao,

   Perhaps more people would be able to help restructure the tree to
support more architectures if a snapshot of the kvm/ia64 code was made
available.  We might want to think about a kvm-ia64-devel mailing list
too (such as ppc has already).  Thanks,

	Alex

-- 
Alex Williamson                             HP Open Source & Linux Org.


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Akio Takebe]

Hi, Alex and Xiantao

>   Perhaps more people would be able to help restructure the tree to
>support more architectures if a snapshot of the kvm/ia64 code was made
>available.  We might want to think about a kvm-ia64-devel mailing list
>too (such as ppc has already).  Thanks,
>
I also think so. :)

Best Regards,

Akio Takebe


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Zhang, Xiantao]

Akio Takebe wrote:
> Hi, Alex and Xiantao
> 
>>   Perhaps more people would be able to help restructure the tree to
>> support more architectures if a snapshot of the kvm/ia64 code was
>> made available.  We might want to think about a kvm-ia64-devel
>> mailing list too (such as ppc has already).  Thanks,
>> 
> I also think so. :)
H, Alex/Akio 
   Thanks for your suggestions. It is a surely good idea to create a
IA64 mailing list to talk about the IA64-KVM.  But now we have to rebase
them to latest commits before sending out, since the kvm source layout
changed much recently.  Once the final source layout comes out, we will
send out the source code soon.  I believe it should happen in very near
future.   :-)
Best wishes
Xiantao

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Caleb Moore]

On Thu, 2007-11-08 at 09:00 -0700, Alex Williamson wrote:
> Hi Xiantao,
> 
>    Perhaps more people would be able to help restructure the tree to
> support more architectures if a snapshot of the kvm/ia64 code was made
> available.  We might want to think about a kvm-ia64-devel mailing list
> too (such as ppc has already).  Thanks,
> 
> 	Alex

I agree.

My branch of the Gelato Federation (http://www.gelato.unsw.edu.au/)
wants to invest time and effort into IA64 KVM which could mean helping
with a merge, streamlining code, benchmarking, bug hunting or something
else along those lines. We've already got an open source Itanium
virtualization solution called Linux on Linux
(http://www.gelato.unsw.edu.au/IA64wiki/LinuxOnLinux) and we were hoping
to apply what knowledge we gained from that project into something
different. Currently we can't really do anything to help since there is
supposed to be a working port but we can't get to it.

Anyway, I'm currently being payed to improve Itanium virtualization
under Linux, is there anything useful I can do here or should I just
stick with working on LoL? During the last few days I've been attempting
to take out the x86 specific stuff from kvm_main.c but I see that
Xiantao already submitted that change yesterday. Do you Intel guys have
it all under control now?



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Alex Williamson]

On Fri, 2007-11-09 at 09:29 +0800, Zhang, Xiantao wrote:
> H, Alex/Akio 
>    Thanks for your suggestions. It is a surely good idea to create a
> IA64 mailing list to talk about the IA64-KVM.  But now we have to rebase
> them to latest commits before sending out, since the kvm source layout
> changed much recently.  Once the final source layout comes out, we will
> send out the source code soon.  I believe it should happen in very near
> future.   :-)

Hi Xiantao,

   Thanks for your work on KVM/ia64 and efforts to get it into the tree,
we definitely appreciate it.  If it's not too much work, I think many of
us would appreciate looking at any kind of snapshot of your tree, even
if it's out of date from what's currently upstream.  We're eager to
start understanding how it works ;)  Thanks,

	Alex

-- 
Alex Williamson                             HP Open Source & Linux Org.


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Dong, Eddie]

> 
> I agree.
> 
> My branch of the Gelato Federation (http://www.gelato.unsw.edu.au/)
> wants to invest time and effort into IA64 KVM which could mean helping
> with a merge, streamlining code, benchmarking, bug hunting or
> something else along those lines. We've already got an open source
> Itanium virtualization solution called Linux on Linux
> (http://www.gelato.unsw.edu.au/IA64wiki/LinuxOnLinux) and we
> were hoping
> to apply what knowledge we gained from that project into something
> different. Currently we can't really do anything to help since there
> is supposed to be a working port but we can't get to it.
> 
> Anyway, I'm currently being payed to improve Itanium virtualization
> under Linux, is there anything useful I can do here or should I just
> stick with working on LoL? During the last few days I've been

I think paravirt_ops support for IA64 is still blank :-( Nobody are 
interested in?

> attempting to take out the x86 specific stuff from kvm_main.c but I
> see that Xiantao already submitted that change yesterday. Do you
> Intel guys have it all under control now? 
> 
> 

Eddie

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Zhang, Xiantao]

Caleb Moore wrote:
> On Thu, 2007-11-08 at 09:00 -0700, Alex Williamson wrote:
>> Hi Xiantao,
>> 
>>    Perhaps more people would be able to help restructure the tree to
>> support more architectures if a snapshot of the kvm/ia64 code was
>> made available.  We might want to think about a kvm-ia64-devel
>> mailing list too (such as ppc has already).  Thanks,
>> 
>> 	Alex
> 
> I agree.
> 
> My branch of the Gelato Federation (http://www.gelato.unsw.edu.au/)



> Anyway, I'm currently being payed to improve Itanium virtualization
> under Linux, is there anything useful I can do here or should I just
> stick with working on LoL? During the last few days I've been
> attempting to take out the x86 specific stuff from kvm_main.c but I
> see that Xiantao already submitted that change yesterday. Do you
> Intel guys have it all under control now?

Community has begun the split work for two months~. Most code and logic
has been taken out from the kvm_main.c. Currently, we try to finalize
the work in next few days. If the split work done,  we don't see the big
hurdles which blocking IA64 code out. Now, I am trying to refine the
code per new framework, and will send it out soon.
Thanks 
Xiantao

> 
>
------------------------------------------------------------------------
-
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a
> browser. Download your FREE copy of Splunk now >>
> http://get.splunk.com/ _______________________________________________
> kvm-devel mailing list
> kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/kvm-devel

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Carsten Otte]

Zhang, Xiantao wrote:
>    Thanks for your suggestions. It is a surely good idea to create a
> IA64 mailing list to talk about the IA64-KVM.  But now we have to rebase
> them to latest commits before sending out, since the kvm source layout
> changed much recently.  Once the final source layout comes out, we will
> send out the source code soon.  I believe it should happen in very near
> future.   :-)
I think this makes sense: we're shuffling the code around too much at 
the moment. It would be a big burden to have to keep the port 
up-to-date while doing that. We'll soon be done with our portability 
split, and I am sure Xiantao will be one of the first to present a port.
Nevertheless, Akio, Alex and Caleb you're highly welcome to 
participate in the review process of our joined protability work to 
make sure we get everything right for ia64.

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: IA64 KVM

[Isaku Yamahata]

On Fri, Nov 09, 2007 at 11:27:51AM +0800, Dong, Eddie wrote:
> > My branch of the Gelato Federation (http://www.gelato.unsw.edu.au/)
> > wants to invest time and effort into IA64 KVM which could mean helping
> > with a merge, streamlining code, benchmarking, bug hunting or
> > something else along those lines. We've already got an open source
> > Itanium virtualization solution called Linux on Linux
> > (http://www.gelato.unsw.edu.au/IA64wiki/LinuxOnLinux) and we
> > were hoping
> > to apply what knowledge we gained from that project into something
> > different. Currently we can't really do anything to help since there
> > is supposed to be a working port but we can't get to it.
> > 
> > Anyway, I'm currently being payed to improve Itanium virtualization
> > under Linux, is there anything useful I can do here or should I just
> > stick with working on LoL? During the last few days I've been
> 
> I think paravirt_ops support for IA64 is still blank :-( Nobody are 
> interested in?

I have a patch to binary-patch the kernel image to replace some
instructions at boot time. (i.e. paravirt_alt)
Although it was made for Xen/IA64, the patch is quite generic, I belive.

BTW, I'm also very interested in KVM ia64 code too!.
thanks,
-- 
yamahata

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/