thr3ads.net - Xen devel - Test report for xen-unstable and qemu-xen [Jun 2013]

If this information is useful, please help other people find it:
Share via:

Fabio Fantoni

2013-Jun-12 14:33 UTC

Test report for xen-unstable and qemu-xen

Dom0:
Wheezy 64 bit with kernel from package linux-image-3.2.0-4-amd64 version 
3.2.41-2+deb7u2, package blktap-dkms and all dependency packages for 
xen, spice and usb redirection.
Seabios 1.7.2-3 and latest spice and usbredir compiled from debian 
unstable sources.
-------------------------
/etc/modules
------------
loop max_loop=64
xenfs
xen-evtchn
blktap
-------------------------
git clone git://xenbits.xen.org/xen.git (in this build commit is 
6859874b61d5ddaf5289e72ed2b2157739b72ca5)
-------------------------
Added some patches:
- tools-configure-qemu-for-spice-and-usbredir
- libxl-Add-spice-vdagent-support-for-upstream-qemu
- tools-firmware-seabios-packaged
- tools: Improve make deb
- vgabios-stdvga.bin custom build from Frediano
- qemu-xen change for support other resolutions:
hw/vga.c: # from Frediano patch
     - if ((val <= VBE_DISPI_MAX_XRES) && ((val & 7) == 0)) {
     + if ((val <= VBE_DISPI_MAX_XRES) && ((val & 1) == 0)) {
- updated qemu upstream
Config-mk:
QEMU_UPSTREAM_REVISION ?= master
-------------------------
./configure --prefix=/usr
-------------------------
make debball

-------------------------
Issues solved from my previous test build report:
-------------
- restore with xl on xendomain not work (now solved upstream)
-------------------------

-------------------------
Old issue:
- Network is not working after restore on Windows domU with qemu-xen and 
gplpv
   Workaround: fixed mac address
-------------
- nested hvm probably bugged: started test it on windows 7 pro 64 domU 
on qemu-xen,
   xpmode crash on install without show error, no error on logs and 
havdetection works
-------------
- Starting xendomain with "xl create" show some debug options:
   Starting auto Xen domains: W7.cfgxc: info: VIRTUAL MEMORY ARRANGEMENT:
   Loader:        0000000000100000->000000000019eb68
   Modules:       0000000000000000->0000000000000000
   TOTAL:         0000000000000000->000000007f000000
   ENTRY ADDRESS: 0000000000100000
xc: info: PHYSICAL MEMORY ALLOCATION:
   4KB PAGES: 0x0000000000000200
   2MB PAGES: 0x00000000000003f7
   1GB PAGES: 0x0000000000000000
[done].
I think it is would be better hiding them here
-------------
- on W7 pro 64 bit domU error in the begin of qemu log, while domU seem 
to work:
xc: error: linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS failed (22 
= Invalid argument): Internal error
xen be: qdisk-832: xc_gnttab_set_max_grants failed: Invalid argument
-------------------------

Fabio Fantoni

2013-Jul-01 13:19 UTC

head link

Test report for xen-unstable and qemu-xen

Dom0:
Wheezy 64 bit with kernel from package linux-image-3.2.0-4-amd64 version 
3.2.41-2+deb7u2, package blktap-dkms and all dependency packages for 
xen, spice and usb redirection.
Seabios 1.7.2-3 and latest spice and usbredir compiled from debian 
unstable sources.
-------------------------
/etc/modules
------------
loop max_loop=64
xenfs
xen-evtchn
blktap
-------------------------
git clone git://xenbits.xen.org/xen.git (in this build commit is 
bcb49301c21ed69b4e8c164fa215efe9db30621a)
-------------------------
Added some patches:
- tools-configure-qemu-for-spice-and-usbredir
- libxl-Add-spice-vdagent-support-for-upstream-qemu
- tools-firmware-seabios-packaged
- tools: Improve make deb
- vgabios-stdvga.bin custom build from Frediano
- qemu-xen change for support other resolutions:
hw/vga.c: # from Frediano patch
     - if ((val <= VBE_DISPI_MAX_XRES) && ((val & 7) == 0)) {
     + if ((val <= VBE_DISPI_MAX_XRES) && ((val & 1) == 0)) {
-------------------------
./configure --prefix=/usr
-------------------------
make debball

-------------------------
Issues solved from my previous test build report:
-------------
- nested hvm was bugged:

   Before: started test it on windows 7 pro 64 domU on qemu-xen,
   xpmode crashed on install without show error, no error on logs and 
havdetection works.

   Now xpmode installs and starts correctly except integration features
   (disks,printers,clipboard sharing) and usb passthrough (of xpmode)
-------------------------

-------------------------
Old issue:
- Network is not working after restore on Windows domU with qemu-xen and 
gplpv
   Workaround: fixed mac address
-------------
- Starting xendomain with "xl create" show some debug options:
   Starting auto Xen domains: W7.cfgxc: info: VIRTUAL MEMORY ARRANGEMENT:
   Loader:        0000000000100000->000000000019eb68
   Modules:       0000000000000000->0000000000000000
   TOTAL:         0000000000000000->000000007f000000
   ENTRY ADDRESS: 0000000000100000
xc: info: PHYSICAL MEMORY ALLOCATION:
   4KB PAGES: 0x0000000000000200
   2MB PAGES: 0x00000000000003f7
   1GB PAGES: 0x0000000000000000
[done].
I think it is would be better hiding them here
-------------
- on W7 pro 64 bit domU error in the begin of qemu log, while domU seem 
to work:
xc: error: linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS failed (22 
= Invalid argument): Internal error
xen be: qdisk-832: xc_gnttab_set_max_grants failed: Invalid argument
-------------------------

Stefano Stabellini

2013-Jul-01 13:50 UTC

head link

Re: Test report for xen-unstable and qemu-xen

On Mon, 1 Jul 2013, Fabio Fantoni wrote:> - on W7 pro 64 bit domU error in the begin of qemu log, while domU seem to
> work:
> xc: error: linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS failed (22
> Invalid argument): Internal error
> xen be: qdisk-832: xc_gnttab_set_max_grants failed: Invalid argument
> -------------------------
This is an harmless message if you are running a pvops kernel, it only
indicates a problem if you are running on a SUSE kernel.

Jan, could we figure out from the return code whether it was a real
problem and only print an error in that case?
I think that users might get confused otherwise.

Jan Beulich

2013-Jul-01 14:07 UTC

head link

Re: Test report for xen-unstable and qemu-xen

>>> On 01.07.13 at 15:50, Stefano Stabellini
<stefano.stabellini@eu.citrix.com> wrote:
> On Mon, 1 Jul 2013, Fabio Fantoni wrote:
>> - on W7 pro 64 bit domU error in the begin of qemu log, while domU seem
to
>> work:
>> xc: error: linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS failed (22
>> Invalid argument): Internal error
>> xen be: qdisk-832: xc_gnttab_set_max_grants failed: Invalid argument
>> -------------------------
> 
> This is an harmless message if you are running a pvops kernel, it only
> indicates a problem if you are running on a SUSE kernel.
> 
> Jan, could we figure out from the return code whether it was a real
> problem and only print an error in that case?
> I think that users might get confused otherwise.
This is supposed to happen already, by special casing the ENOTTY
errno value. Question is why EINVAL is being seen here instead.

Jan

Matthew Daley

2013-Jul-03 08:25 UTC

head link

Re: Test report for xen-unstable and qemu-xen

On Tue, Jul 2, 2013 at 2:07 AM, Jan Beulich <JBeulich@suse.com>
wrote:>>>> On 01.07.13 at 15:50, Stefano Stabellini
<stefano.stabellini@eu.citrix.com> wrote:
>> On Mon, 1 Jul 2013, Fabio Fantoni wrote:
>>> - on W7 pro 64 bit domU error in the begin of qemu log, while domU
seem to
>>> work:
>>> xc: error: linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS failed
(22 >>> Invalid argument): Internal error
>>> xen be: qdisk-832: xc_gnttab_set_max_grants failed: Invalid
argument
>>> -------------------------
>>
>> This is an harmless message if you are running a pvops kernel, it only
>> indicates a problem if you are running on a SUSE kernel.
>>
>> Jan, could we figure out from the return code whether it was a real
>> problem and only print an error in that case?
>> I think that users might get confused otherwise.
>
> This is supposed to happen already, by special casing the ENOTTY
> errno value. Question is why EINVAL is being seen here instead.
Probably because
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/fs/ioctl.c?id=07d106d0a33d6063d2061305903deb02489eba20
only went in time for Linux 3.3 (it''s not in my Debian unstable, for
example).

I personally use the patch below to keep a clean log (I see this
message too), but I''m sure it would be NAK''d for continuing
the
ENOTTY/EINVAL confusion and making legitmate EINVALs disappear.
Perhaps instead the message could be softened from an error to a
warning?

- Matthew


From 8a47313339e4778e0cb282ad5b60da1449c8da34 Mon Sep 17 00:00:00 2001
From: Matthew Daley <mattjd@gmail.com>
Date: Thu, 23 May 2013 18:26:36 +1200
Subject: [PATCH] libxc: actually ignore missing ioctl in
 linux_gnttab_set_max_grants on Linux <= 3.2

Linux 3.2 and earlier (incorrectly) convert the ENOIOCTLCMD returned by
gntdev''s ioctl handler to EINVAL instead of ENOTTY.

Handle this in linux_gnttab_set_max_grants, otherwise false negatives
can lead to misleading error messages, like one logged by qemu when a
qemu-backed xen virtual disk is created.

Signed-off-by: Matthew Daley <mattjd@gmail.com>
---
 tools/libxc/xc_linux_osdep.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index 73860a2..3f85054 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -573,8 +573,10 @@ static int linux_gnttab_set_max_grants(xc_gnttab
*xch, xc_osdep_handle h,
         /*
          * Newer (e.g. pv-ops) kernels don''t implement this IOCTL,
          * so ignore the resulting specific failure.
+         * gntdev in <= 3.2 kernels returns EINVAL instead of ENOTTY on
+         * unknown IOCTLs.
          */
-        if (errno == ENOTTY)
+        if (errno == ENOTTY || errno == EINVAL)
             rc = 0;
         else
             PERROR("linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS
failed");
--
1.7.10.4

Jan Beulich

2013-Jul-03 10:16 UTC

head link

Re: Test report for xen-unstable and qemu-xen

>>> On 03.07.13 at 10:25, Matthew Daley <mattjd@gmail.com> wrote:
> On Tue, Jul 2, 2013 at 2:07 AM, Jan Beulich <JBeulich@suse.com>
wrote:
>>>>> On 01.07.13 at 15:50, Stefano Stabellini
<stefano.stabellini@eu.citrix.com>
> wrote:
>>> On Mon, 1 Jul 2013, Fabio Fantoni wrote:
>>>> - on W7 pro 64 bit domU error in the begin of qemu log, while
domU seem to
>>>> work:
>>>> xc: error: linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS
failed (22 >>>> Invalid argument): Internal error
>>>> xen be: qdisk-832: xc_gnttab_set_max_grants failed: Invalid
argument
>>>> -------------------------
>>>
>>> This is an harmless message if you are running a pvops kernel, it
only
>>> indicates a problem if you are running on a SUSE kernel.
>>>
>>> Jan, could we figure out from the return code whether it was a real
>>> problem and only print an error in that case?
>>> I think that users might get confused otherwise.
>>
>> This is supposed to happen already, by special casing the ENOTTY
>> errno value. Question is why EINVAL is being seen here instead.
> 
> Probably because
>
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/fs/io
> ctl.c?id=07d106d0a33d6063d2061305903deb02489eba20
> only went in time for Linux 3.3 (it''s not in my Debian unstable,
for
> example).
Very much so, and kernels should be backporting this if upstream
didn''t.
> I personally use the patch below to keep a clean log (I see this
> message too), but I''m sure it would be NAK''d for
continuing the
> ENOTTY/EINVAL confusion and making legitmate EINVALs disappear.
> Perhaps instead the message could be softened from an error to a
> warning?
Making this a warning may be acceptable, but I''m not eager to work
around kernel bugs like this (and in effect risk hiding _real_ errors,
which ought to not be logged as just a warning).

Jan

Stefano Stabellini

2013-Jul-03 10:58 UTC

head link

Re: Test report for xen-unstable and qemu-xen

On Wed, 3 Jul 2013, Jan Beulich wrote:> >>> On 03.07.13 at 10:25, Matthew Daley <mattjd@gmail.com>
wrote:
> > On Tue, Jul 2, 2013 at 2:07 AM, Jan Beulich <JBeulich@suse.com>
wrote:
> >>>>> On 01.07.13 at 15:50, Stefano Stabellini
<stefano.stabellini@eu.citrix.com>
> > wrote:
> >>> On Mon, 1 Jul 2013, Fabio Fantoni wrote:
> >>>> - on W7 pro 64 bit domU error in the begin of qemu log,
while domU seem to
> >>>> work:
> >>>> xc: error: linux_gnttab_set_max_grants: ioctl
SET_MAX_GRANTS failed (22 > >>>> Invalid argument): Internal
error
> >>>> xen be: qdisk-832: xc_gnttab_set_max_grants failed:
Invalid argument
> >>>> -------------------------
> >>>
> >>> This is an harmless message if you are running a pvops kernel,
it only
> >>> indicates a problem if you are running on a SUSE kernel.
> >>>
> >>> Jan, could we figure out from the return code whether it was a
real
> >>> problem and only print an error in that case?
> >>> I think that users might get confused otherwise.
> >>
> >> This is supposed to happen already, by special casing the ENOTTY
> >> errno value. Question is why EINVAL is being seen here instead.
> > 
> > Probably because
> >
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/fs/io
> > ctl.c?id=07d106d0a33d6063d2061305903deb02489eba20
> > only went in time for Linux 3.3 (it''s not in my Debian
unstable, for
> > example).
> 
> Very much so, and kernels should be backporting this if upstream
> didn''t.
Did anybody ask for a backport? It doesn''t look like the patch was sent
to stable.

Ren, Yongjie

2013-Jul-03 15:12 UTC

head link

Re: Test report for xen-unstable and qemu-xen

> -----Original Message-----
> From: xen-devel-bounces@lists.xen.org
> [mailto:xen-devel-bounces@lists.xen.org] On Behalf Of Fabio Fantoni
> Sent: Monday, July 01, 2013 9:20 PM
> To: xen-devel; Stefano Stabellini; Ian Campbell; Ian Jackson; Xu, Dongxiao;
> Jan Beulich
> Subject: [Xen-devel] Test report for xen-unstable and qemu-xen
> 
> Dom0:
> Wheezy 64 bit with kernel from package linux-image-3.2.0-4-amd64
> version
> 3.2.41-2+deb7u2, package blktap-dkms and all dependency packages for
> xen, spice and usb redirection.
> Seabios 1.7.2-3 and latest spice and usbredir compiled from debian
> unstable sources.
> -------------------------
> /etc/modules
> ------------
> loop max_loop=64
> xenfs
> xen-evtchn
> blktap
> -------------------------
> git clone git://xenbits.xen.org/xen.git (in this build commit is
> bcb49301c21ed69b4e8c164fa215efe9db30621a)
> -------------------------
> Added some patches:
> - tools-configure-qemu-for-spice-and-usbredir
> - libxl-Add-spice-vdagent-support-for-upstream-qemu
> - tools-firmware-seabios-packaged
> - tools: Improve make deb
> - vgabios-stdvga.bin custom build from Frediano
> - qemu-xen change for support other resolutions:
> hw/vga.c: # from Frediano patch
>      - if ((val <= VBE_DISPI_MAX_XRES) && ((val & 7) == 0))
{
>      + if ((val <= VBE_DISPI_MAX_XRES) && ((val & 1) == 0))
{
> -------------------------
> ./configure --prefix=/usr
> -------------------------
> make debball
> 
> -------------------------
> Issues solved from my previous test build report:
> -------------
> - nested hvm was bugged:
> 
>    Before: started test it on windows 7 pro 64 domU on qemu-xen,
>    xpmode crashed on install without show error, no error on logs and
> havdetection works.
> 
>    Now xpmode installs and starts correctly except integration features
>    (disks,printers,clipboard sharing) and usb passthrough (of xpmode)
> -------------------------
> Hi Fabio,
I''m also testing Win7 XP mode in nested VMX environment, but I met some
problems. In the Windows 7 HVM guest, a tool from Microsoft showed
hardware-assisted virtualization is supported, but XP mode showed
"hardware-
assisted virtualization is disabled" and it refused to boot XP.
(see the attached 2 pictures to get the two conflicting messages..)

I followed the MS'' below doc to set up XP mode.
http://windows.microsoft.com/en-us/windows7/install-and-use-windows-xp-mode-in-windows-7
And, used the tool http://go.microsoft.com/fwlink/p/?LinkId=163321 to check
whether hardware-assisted virtualization is enabled or not.

Test environment:
hardware: Sandy Bridge - EP server
L0: latest Xen 4.3, Dom0 3.10.0, qemu-xen (not traditional)
L1: Windows 7 Enterprise
L2: XP mode

my Win7 guest config file is below; can you give me any hints?
----------------------
builder = "hvm"
name = "jay-hvm-1"
memory = 2048
vcpus = 2
vif = [ '''' ]
disk = [ ''/home/jay/win7-1.qcow2,qcow2,xvda,w'' ]
vnc = 1
stdvga = 1
usb=1
usbdevice=''tablet''
nestedhvm=1
----------------------



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

Fabio Fantoni

2013-Jul-04 07:55 UTC

head link

Re: Test report for xen-unstable and qemu-xen

Il 03/07/2013 17:12, Ren, Yongjie ha scritto:>> -----Original Message-----
>> From: xen-devel-bounces@lists.xen.org
>> [mailto:xen-devel-bounces@lists.xen.org] On Behalf Of Fabio Fantoni
>> Sent: Monday, July 01, 2013 9:20 PM
>> To: xen-devel; Stefano Stabellini; Ian Campbell; Ian Jackson; Xu,
Dongxiao;
>> Jan Beulich
>> Subject: [Xen-devel] Test report for xen-unstable and qemu-xen
>>
>> Dom0:
>> Wheezy 64 bit with kernel from package linux-image-3.2.0-4-amd64
>> version
>> 3.2.41-2+deb7u2, package blktap-dkms and all dependency packages for
>> xen, spice and usb redirection.
>> Seabios 1.7.2-3 and latest spice and usbredir compiled from debian
>> unstable sources.
>> -------------------------
>> /etc/modules
>> ------------
>> loop max_loop=64
>> xenfs
>> xen-evtchn
>> blktap
>> -------------------------
>> git clone git://xenbits.xen.org/xen.git (in this build commit is
>> bcb49301c21ed69b4e8c164fa215efe9db30621a)
>> -------------------------
>> Added some patches:
>> - tools-configure-qemu-for-spice-and-usbredir
>> - libxl-Add-spice-vdagent-support-for-upstream-qemu
>> - tools-firmware-seabios-packaged
>> - tools: Improve make deb
>> - vgabios-stdvga.bin custom build from Frediano
>> - qemu-xen change for support other resolutions:
>> hw/vga.c: # from Frediano patch
>>       - if ((val <= VBE_DISPI_MAX_XRES) && ((val & 7) ==
0)) {
>>       + if ((val <= VBE_DISPI_MAX_XRES) && ((val & 1) ==
0)) {
>> -------------------------
>> ./configure --prefix=/usr
>> -------------------------
>> make debball
>>
>> -------------------------
>> Issues solved from my previous test build report:
>> -------------
>> - nested hvm was bugged:
>>
>>     Before: started test it on windows 7 pro 64 domU on qemu-xen,
>>     xpmode crashed on install without show error, no error on logs and
>> havdetection works.
>>
>>     Now xpmode installs and starts correctly except integration
features
>>     (disks,printers,clipboard sharing) and usb passthrough (of xpmode)
>> -------------------------
>>
> Hi Fabio,
> I''m also testing Win7 XP mode in nested VMX environment, but I met
some
> problems. In the Windows 7 HVM guest, a tool from Microsoft showed
> hardware-assisted virtualization is supported, but XP mode showed
"hardware-
> assisted virtualization is disabled" and it refused to boot XP.
> (see the attached 2 pictures to get the two conflicting messages..)
>
> I followed the MS'' below doc to set up XP mode.
>
http://windows.microsoft.com/en-us/windows7/install-and-use-windows-xp-mode-in-windows-7
> And, used the tool http://go.microsoft.com/fwlink/p/?LinkId=163321 to check
> whether hardware-assisted virtualization is enabled or not.
>
> Test environment:
> hardware: Sandy Bridge - EP server
> L0: latest Xen 4.3, Dom0 3.10.0, qemu-xen (not traditional)
> L1: Windows 7 Enterprise
> L2: XP mode
>
> my Win7 guest config file is below; can you give me any hints?
> ----------------------
> builder = "hvm"
> name = "jay-hvm-1"
> memory = 2048
> vcpus = 2
> vif = [ '''' ]
> disk = [ ''/home/jay/win7-1.qcow2,qcow2,xvda,w'' ]
> vnc = 1
> stdvga = 1
> usb=1
> usbdevice=''tablet''
> nestedhvm=1
> ----------------------Have you tried with same xen commit? 
(bcb49301c21ed69b4e8c164fa215efe9db30621a) or atleast to this that 
probably have solved my problem: nested vmx: Fix the booting of L2 PAE 
guest - 49b55619e4923aa1a43b7c027dd3e352575e4e52

Ren, Yongjie

2013-Jul-04 14:55 UTC

head link

Re: Test report for xen-unstable and qemu-xen

> -----Original Message-----
> From: Fabio Fantoni [mailto:fabio.fantoni@m2r.biz]
> Sent: Thursday, July 04, 2013 3:55 PM
> To: Ren, Yongjie
> Cc: xen-devel; Stefano Stabellini; Ian Campbell; Ian Jackson; Xu, Dongxiao;
> Jan Beulich
> Subject: Re: [Xen-devel] Test report for xen-unstable and qemu-xen
> 
> Il 03/07/2013 17:12, Ren, Yongjie ha scritto:
> >> -----Original Message-----
> >> From: xen-devel-bounces@lists.xen.org
> >> [mailto:xen-devel-bounces@lists.xen.org] On Behalf Of Fabio
Fantoni
> >> Sent: Monday, July 01, 2013 9:20 PM
> >> To: xen-devel; Stefano Stabellini; Ian Campbell; Ian Jackson; Xu,
> Dongxiao;
> >> Jan Beulich
> >> Subject: [Xen-devel] Test report for xen-unstable and qemu-xen
> >>
> >> Dom0:
> >> Wheezy 64 bit with kernel from package linux-image-3.2.0-4-amd64
> >> version
> >> 3.2.41-2+deb7u2, package blktap-dkms and all dependency packages
> for
> >> xen, spice and usb redirection.
> >> Seabios 1.7.2-3 and latest spice and usbredir compiled from debian
> >> unstable sources.
> >> -------------------------
> >> /etc/modules
> >> ------------
> >> loop max_loop=64
> >> xenfs
> >> xen-evtchn
> >> blktap
> >> -------------------------
> >> git clone git://xenbits.xen.org/xen.git (in this build commit is
> >> bcb49301c21ed69b4e8c164fa215efe9db30621a)
> >> -------------------------
> >> Added some patches:
> >> - tools-configure-qemu-for-spice-and-usbredir
> >> - libxl-Add-spice-vdagent-support-for-upstream-qemu
> >> - tools-firmware-seabios-packaged
> >> - tools: Improve make deb
> >> - vgabios-stdvga.bin custom build from Frediano
> >> - qemu-xen change for support other resolutions:
> >> hw/vga.c: # from Frediano patch
> >>       - if ((val <= VBE_DISPI_MAX_XRES) && ((val &
7) == 0)) {
> >>       + if ((val <= VBE_DISPI_MAX_XRES) && ((val &
1) == 0)) {
> >> -------------------------
> >> ./configure --prefix=/usr
> >> -------------------------
> >> make debball
> >>
> >> -------------------------
> >> Issues solved from my previous test build report:
> >> -------------
> >> - nested hvm was bugged:
> >>
> >>     Before: started test it on windows 7 pro 64 domU on qemu-xen,
> >>     xpmode crashed on install without show error, no error on logs
> and
> >> havdetection works.
> >>
> >>     Now xpmode installs and starts correctly except integration
> features
> >>     (disks,printers,clipboard sharing) and usb passthrough (of
> xpmode)
> >> -------------------------
> >>
> > Hi Fabio,
> > I''m also testing Win7 XP mode in nested VMX environment, but
I met
> some
> > problems. In the Windows 7 HVM guest, a tool from Microsoft showed
> > hardware-assisted virtualization is supported, but XP mode showed
> "hardware-
> > assisted virtualization is disabled" and it refused to boot XP.
> > (see the attached 2 pictures to get the two conflicting messages..)
> >
> > I followed the MS'' below doc to set up XP mode.
> >
> http://windows.microsoft.com/en-us/windows7/install-and-use-windows-x
> p-mode-in-windows-7
> > And, used the tool http://go.microsoft.com/fwlink/p/?LinkId=163321 to
> check
> > whether hardware-assisted virtualization is enabled or not.
> >
> > Test environment:
> > hardware: Sandy Bridge - EP server
> > L0: latest Xen 4.3, Dom0 3.10.0, qemu-xen (not traditional)
> > L1: Windows 7 Enterprise
> > L2: XP mode
> >
> > my Win7 guest config file is below; can you give me any hints?
> > ----------------------
> > builder = "hvm"
> > name = "jay-hvm-1"
> > memory = 2048
> > vcpus = 2
> > vif = [ '''' ]
> > disk = [ ''/home/jay/win7-1.qcow2,qcow2,xvda,w'' ]
> > vnc = 1
> > stdvga = 1
> > usb=1
> > usbdevice=''tablet''
> > nestedhvm=1
> > ----------------------
> Have you tried with same xen commit?
> (bcb49301c21ed69b4e8c164fa215efe9db30621a) or atleast to this that
> probably have solved my problem: nested vmx: Fix the booting of L2 PAE
> guest - 49b55619e4923aa1a43b7c027dd3e352575e4e52Yes, I tried that commit you mentioned. I also think Dongxiao''s fix 
"nested vmx: Fix the booting of L2 PAE" can fix some issue as you
tested.

But in my testing, with or without that fix, I met the same issue. I
can''t
even use ''XP mode'' in Win7 guest. ''XP Moe''
showed hardware VT is disabled;
while MS''s another tool to check VT showed hardware VT is supported. 
I was confused and can''t even start XP mode in Win7 HVM guest. 
(Before that fix, you can see your XP mode crash; but I can''t start XP
mode at all.)

thanks!  - Jay

Maybe Matching Threads

Search for more maybe matching threads

Xen devel - Jun 2013 - Test report for xen-unstable and qemu-xen

Test report for xen-unstable and qemu-xen

Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Re: Test report for xen-unstable and qemu-xen

Maybe Matching Threads