VMX status daily report Nov. 3

VMX status daily report Nov. 3

[Shi, Kuiliang]

 
 
We have test latest xen (ChangeSet 7608) on VT platform with Intel 915
chipset and Here is the test summary:
 
IA32: 
    Build&Boot SMP xen0 without problem
    Can startup XenU with different configuration
    
    IA32 issues contain:
    VMX can not be created
    
 
IA32E: 
    Build&Boot SMP xen0 without problem
    Can startup XenU with different configuration
 
   IA32E issues contain:
    VMX can not be created.
 

Thanks and regards
 
Kuiliang shi(kuiliang.shi@intel.com)
Open Source Technology Center
Intel Corp.

Re: VMX status daily report Nov. 3

[Khoa Huynh]

"Shi, Kuiliang" <kuiliang.shi@intel.com> wrote on 11/03/2005 03:14:46 AM:

>
>
> We have test latest xen (ChangeSet 7608) on VT platform with Intel 915
> chipset and Here is the test summary:
>
> IA32:
>     Build&Boot SMP xen0 without problem
>     Can startup XenU with different configuration
>
>     IA32 issues contain:
>     VMX can not be created
>
>
> IA32E:
>     Build&Boot SMP xen0 without problem
>     Can startup XenU with different configuration
>
>    IA32E issues contain:
>     VMX can not be created.
>

I think you need this patch (also from Intel) to create
a VMX domain:

diff -r ae2e13795c63 tools/python/xen/xend/server/DevController.py
--- a/tools/python/xen/xend/server/DevController.py     Tue Nov  1
19:13:06 2005 +0100
+++ b/tools/python/xen/xend/server/DevController.py     Wed Nov  2
14:38:31 2005 +0800
@@ -57,6 +57,8 @@
          @return The ID for the newly created device.
          """
          (devid, back, front) = self.getDeviceDetails(config)
+        if devid == 0:
+            return devid

          self.writeDetails(config, devid, back, front)

I have not seen this patch applied to the latest xen-unstable tree
yet.  I hope it will be applied soon as this works for me.  Thanks.

Regards,

Khoa

Re: VMX status daily report Nov. 3

[Ewan Mellor]

On Thu, Nov 03, 2005 at 09:35:37AM -0600, Khoa Huynh wrote:

> 
> "Shi, Kuiliang" <kuiliang.shi@intel.com> wrote on 11/03/2005 03:14:46 AM:
> 
> >
> >
> > We have test latest xen (ChangeSet 7608) on VT platform with Intel 915
> > chipset and Here is the test summary:
> >
> > IA32:
> >     Build&Boot SMP xen0 without problem
> >     Can startup XenU with different configuration
> >
> >     IA32 issues contain:
> >     VMX can not be created
> >
> >
> > IA32E:
> >     Build&Boot SMP xen0 without problem
> >     Can startup XenU with different configuration
> >
> >    IA32E issues contain:
> >     VMX can not be created.
> >
> 
> I think you need this patch (also from Intel) to create
> a VMX domain:
> 
> diff -r ae2e13795c63 tools/python/xen/xend/server/DevController.py
> --- a/tools/python/xen/xend/server/DevController.py     Tue Nov  1
> 19:13:06 2005 +0100
> +++ b/tools/python/xen/xend/server/DevController.py     Wed Nov  2
> 14:38:31 2005 +0800
> @@ -57,6 +57,8 @@
>           @return The ID for the newly created device.
>           """
>           (devid, back, front) = self.getDeviceDetails(config)
> +        if devid == 0:
> +            return devid
> 
>           self.writeDetails(config, devid, back, front)
> 
> I have not seen this patch applied to the latest xen-unstable tree
> yet.  I hope it will be applied soon as this works for me.  Thanks.

This patch breaks vifs for everyone else (vif IDs are allocated from 0).  I
have a replacement ready and waiting to go.

Ewan.

RE: VMX status daily report Nov. 3

[Shi, Kuiliang]

 
The last workable changset for VMX is r7582. 7583 have the bug:
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=382


-----Original Message-----
From: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of Shi, Kuiliang
Sent: 2005年11月3日 17:15
To: xen-devel@lists.xensource.com
Subject: [Xen-devel] VMX status daily report Nov. 3

 
 
We have test latest xen (ChangeSet 7608) on VT platform with Intel 915
chipset and Here is the test summary:
 
IA32: 
    Build&Boot SMP xen0 without problem
    Can startup XenU with different configuration
    
    IA32 issues contain:
    VMX can not be created
    
 
IA32E: 
    Build&Boot SMP xen0 without problem
    Can startup XenU with different configuration
 
   IA32E issues contain:
    VMX can not be created.
 

Thanks and regards
 
Kuiliang shi(kuiliang.shi@intel.com)
Open Source Technology Center
Intel Corp.


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

Re: VMX status daily report Nov. 3

[Khoa Huynh]

Ewan Mellor <ewan@xensource.com> wrote on 11/03/2005 09:44:31 AM:

> On Thu, Nov 03, 2005 at 09:35:37AM -0600, Khoa Huynh wrote:
>
> >
> > "Shi, Kuiliang" <kuiliang.shi@intel.com> wrote on 11/03/2005 03:14:46
AM:
> >
> > >
> > >
> > > We have test latest xen (ChangeSet 7608) on VT platform with Intel
915
> > > chipset and Here is the test summary:
> > >
> > > IA32:
> > >     Build&Boot SMP xen0 without problem
> > >     Can startup XenU with different configuration
> > >
> > >     IA32 issues contain:
> > >     VMX can not be created
> > >
> > >
> > > IA32E:
> > >     Build&Boot SMP xen0 without problem
> > >     Can startup XenU with different configuration
> > >
> > >    IA32E issues contain:
> > >     VMX can not be created.
> > >
> >
> > I think you need this patch (also from Intel) to create
> > a VMX domain:
> >
> > diff -r ae2e13795c63 tools/python/xen/xend/server/DevController.py
> > --- a/tools/python/xen/xend/server/DevController.py     Tue Nov  1
> > 19:13:06 2005 +0100
> > +++ b/tools/python/xen/xend/server/DevController.py     Wed Nov  2
> > 14:38:31 2005 +0800
> > @@ -57,6 +57,8 @@
> >           @return The ID for the newly created device.
> >           """
> >           (devid, back, front) = self.getDeviceDetails(config)
> > +        if devid == 0:
> > +            return devid
> >
> >           self.writeDetails(config, devid, back, front)
> >
> > I have not seen this patch applied to the latest xen-unstable tree
> > yet.  I hope it will be applied soon as this works for me.  Thanks.
>
> This patch breaks vifs for everyone else (vif IDs are allocated from 0).
I
> have a replacement ready and waiting to go.
>

Ewan, has your patch been picked up in xen-unstable.hg yet ?  If not, can
you post it if it's ready (just so we can apply and use it while waiting
for it to make its way into xen-unstable).  Thanks in advance.

Regards,

Khoa

Re: VMX status daily report Nov. 3

[Ling, Xiaofeng]

[-- Attachment #1: Type: text/plain, Size: 563 bytes --]

Ewan Mellor wrote:
> On Thu, Nov 03, 2005 at 09:35:37AM -0600, Khoa Huynh wrote:
> 
> 
>>"Shi, Kuiliang" <kuiliang.shi@intel.com> wrote on 11/03/2005 03:14:46 AM:
>>
>>
>>>
>>>We have test latest xen (ChangeSet 7608) on VT platform with Intel 915
>>>chipset and Here is the test summary:
>>>
> 
> This patch breaks vifs for everyone else (vif IDs are allocated from 0).  I
> have a replacement ready and waiting to go.
> 
Ewan, how is your replacement patch going on?
Attach is just a simple fix to use "-1" to indicate a ioemu device.
Will that be ok with vif?


[-- Attachment #2: devid-fix.patch --]
[-- Type: text/x-patch, Size: 1257 bytes --]

# HG changeset patch
# User Xiaofeng Ling <xiaofeng.ling@intel.com>
# Node ID ff100cd8f74fd53ad652e5760be956a6631a96d3
# Parent  a969d918674e6deab2a8c61977e4a4016a680ca7
ioemu device for vmx guest doesn't need backend/frontend info in xenstore

Signed-off-by: Xiaofeng Ling <xiaofeng.ling>
diff -r a969d918674e -r ff100cd8f74f tools/python/xen/xend/server/DevController.py
--- a/tools/python/xen/xend/server/DevController.py	Fri Nov  4 16:01:49 2005 +0100
+++ b/tools/python/xen/xend/server/DevController.py	Mon Nov  7 09:46:02 2005 +0800
@@ -58,6 +58,9 @@
         """
         (devid, back, front) = self.getDeviceDetails(config)
 
+        if devid == -1:
+            return devid
+        
         self.writeDetails(config, devid, back, front)
 
         status, fn_ret = self.waitForBackend(devid)
diff -r a969d918674e -r ff100cd8f74f tools/python/xen/xend/server/blkif.py
--- a/tools/python/xen/xend/server/blkif.py	Fri Nov  4 16:01:49 2005 +0100
+++ b/tools/python/xen/xend/server/blkif.py	Mon Nov  7 09:46:02 2005 +0800
@@ -43,7 +43,7 @@
         
         dev = sxp.child_value(config, 'dev')
         if re.match('^ioemu:', dev):
-            return (0,{},{})
+            return (-1,{},{})
 
         devid = blkif.blkdev_name_to_number(dev)
 

[-- Attachment #3: Type: text/plain, Size: 138 bytes --]

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

Re: VMX status daily report Nov. 3

[Ewan Mellor]

[-- Attachment #1: Type: text/plain, Size: 387 bytes --]

On Mon, Nov 07, 2005 at 09:52:52AM +0800, Ling, Xiaofeng wrote:

> Ewan, how is your replacement patch going on?
> Attach is just a simple fix to use "-1" to indicate a ioemu device.
> Will that be ok with vif?

I used None rather than -1, but yes, that's exactly what I've done (attached).

Sorry for the delay in getting that out to you.  It will hopefully push today.

Cheers,

Ewan.

[-- Attachment #2: ioemu.patch --]
[-- Type: text/plain, Size: 1370 bytes --]

# HG changeset patch
# User emellor@leeni.uk.xensource.com
# Node ID 34932eb3f63b6c971504680c3ee8e0e5db07566a
# Parent  de881276da129733b936c6882cd275aac7580a90
Hack around problem whereby VMX domains (using the ioemu device) now fail
because the new hotplug script error checking kicks in.

ioemu really ought to move out of blkif.py.

Signed-off-by: Ewan Mellor <ewan@xensource.com>

diff -r de881276da12 -r 34932eb3f63b tools/python/xen/xend/server/DevController.py
--- a/tools/python/xen/xend/server/DevController.py	Fri Nov  4 16:39:15 2005
+++ b/tools/python/xen/xend/server/DevController.py	Fri Nov  4 16:42:22 2005
@@ -57,8 +57,8 @@
         @return The ID for the newly created device.
         """
         (devid, back, front) = self.getDeviceDetails(config)
-        if devid == 0:
-            return devid
+        if devid == None:
+            return 0
 
         self.writeDetails(config, devid, back, front)
 
diff -r de881276da12 -r 34932eb3f63b tools/python/xen/xend/server/blkif.py
--- a/tools/python/xen/xend/server/blkif.py	Fri Nov  4 16:39:15 2005
+++ b/tools/python/xen/xend/server/blkif.py	Fri Nov  4 16:42:22 2005
@@ -43,7 +43,7 @@
         
         dev = sxp.child_value(config, 'dev')
         if re.match('^ioemu:', dev):
-            return (0,{},{})
+            return (None,{},{})
 
         devid = blkif.blkdev_name_to_number(dev)
 

[-- Attachment #3: Type: text/plain, Size: 138 bytes --]

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

Re: ioemu patch from Ewan Mellor (was: VMX status report Nov. 2)

[Khoa Huynh]

Ewan Mellor <ewan@xensource.com> wrote on 11/07/2005 03:55:07 AM:
> On Mon, Nov 07, 2005 at 09:52:52AM +0800, Ling, Xiaofeng wrote:
>
> > Ewan, how is your replacement patch going on?
> > Attach is just a simple fix to use "-1" to indicate a ioemu device.
> > Will that be ok with vif?
>
> I used None rather than -1, but yes, that's exactly what I've done
(attached).
>
> Sorry for the delay in getting that out to you.  It will hopefully push
today.

The xen-unstable tree which I pulled today should have this patch, but I
still cannot
create a VMX domain - same problem as before.  Am I missing anything ?
Thanks.

Regards,

Khoa