Xen devel - Feb 2010 - tapdisk2 segfaults with xen-unstable+linux-2.6-pvops

Hi,

I''m trying to run guest pv vm on my machine running xen-unstable
(xen-4.0.0-rc3.pre + linux-2.6-pvops). However, I get blktap error:

"
sudo xm create -c xmexample
Using config file "./xmexample".
Error: Failed to create device.
    stdout: unrecognized child response
    stderr:
Check that target "/home/ashish/xen/hd.img" exists and that blktap2
driver installed in dom0.
"

Dom0 dmesg shows:

"
[ 7112.648271] blktap_sysfs_create: adding attributes for dev eb5506c0
[ 7112.648287] blktap_control_allocate_tap: allocated tap de87e000
[ 7112.651564] tapdisk2[7325]: segfault at 10 ip b763a540 sp bffd548c
error 4 in libpthread-2.7.so[b7633000+14000]
"

tapdisk2 -n aio:/home/ashish/xen/hd.img output shows the same above segfault msg

Any ideas what could be wrong?

config, xmexample, and dmesg attached.

Thanks,
Ashish




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

Signed-off-by: Daniel Stodden <daniel.stodden@citrix.com>




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

Hi,

Disable memshr in tools/blktap2/Makefile. Find a patch next to this
mail, unless Keir pulled it in by the time you read this.

Disable CONFIG_HIGHPTE, the config you posted has it enabled.

The pvops kernel wants a bunch of patches which haven''t made it into
Jeremy''s tree yet. The latest version was posted here.

http://lists.xensource.com/archives/html/xen-devel/2010-02/msg00331.html

Let me know if this work for you. Especially if it doesn''t, but I
haven''t heard about more open issues lately.

Thanks,
Daniel

On Wed, 2010-02-10 at 20:45 -0500, Ashish Bijlani wrote:
> Hi,
> 
> I''m trying to run guest pv vm on my machine running xen-unstable
> (xen-4.0.0-rc3.pre + linux-2.6-pvops). However, I get blktap error:
> 
> "
> sudo xm create -c xmexample
> Using config file "./xmexample".
> Error: Failed to create device.
>     stdout: unrecognized child response
>     stderr:
> Check that target "/home/ashish/xen/hd.img" exists and that
blktap2
> driver installed in dom0.
> "
> 
> Dom0 dmesg shows:
> 
> "
> [ 7112.648271] blktap_sysfs_create: adding attributes for dev eb5506c0
> [ 7112.648287] blktap_control_allocate_tap: allocated tap de87e000
> [ 7112.651564] tapdisk2[7325]: segfault at 10 ip b763a540 sp bffd548c
> error 4 in libpthread-2.7.so[b7633000+14000]
> "
> 
> tapdisk2 -n aio:/home/ashish/xen/hd.img output shows the same above
segfault msg
> 
> Any ideas what could be wrong?
> 
> config, xmexample, and dmesg attached.
> 
> Thanks,
> Ashish


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

Hey Daniel,

Disabling msmshr worked for me. Thanks !!

- Ashish

On Thu, Feb 11, 2010 at 2:56 AM, Daniel Stodden
<daniel.stodden@citrix.com> wrote:
>
> Hi,
>
> Disable memshr in tools/blktap2/Makefile. Find a patch next to this
> mail, unless Keir pulled it in by the time you read this.
>
> Disable CONFIG_HIGHPTE, the config you posted has it enabled.
>
> The pvops kernel wants a bunch of patches which haven''t made it
into
> Jeremy''s tree yet. The latest version was posted here.
>
> http://lists.xensource.com/archives/html/xen-devel/2010-02/msg00331.html
>
> Let me know if this work for you. Especially if it doesn''t, but I
> haven''t heard about more open issues lately.
>
> Thanks,
> Daniel
>
> On Wed, 2010-02-10 at 20:45 -0500, Ashish Bijlani wrote:
>> Hi,
>>
>> I''m trying to run guest pv vm on my machine running
xen-unstable
>> (xen-4.0.0-rc3.pre + linux-2.6-pvops). However, I get blktap error:
>>
>> "
>> sudo xm create -c xmexample
>> Using config file "./xmexample".
>> Error: Failed to create device.
>>     stdout: unrecognized child response
>>     stderr:
>> Check that target "/home/ashish/xen/hd.img" exists and that
blktap2
>> driver installed in dom0.
>> "
>>
>> Dom0 dmesg shows:
>>
>> "
>> [ 7112.648271] blktap_sysfs_create: adding attributes for dev eb5506c0
>> [ 7112.648287] blktap_control_allocate_tap: allocated tap de87e000
>> [ 7112.651564] tapdisk2[7325]: segfault at 10 ip b763a540 sp bffd548c
>> error 4 in libpthread-2.7.so[b7633000+14000]
>> "
>>
>> tapdisk2 -n aio:/home/ashish/xen/hd.img output shows the same above
segfault msg
>>
>> Any ideas what could be wrong?
>>
>> config, xmexample, and dmesg attached.
>>
>> Thanks,
>> Ashish
>
>
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hello Daniel,

Patches applied, and it seems to be working so far !
Thx.

Hope that Jeremy hasn''t ignored the patches because of the name of the
diffs.

--
Sander

Thursday, February 11, 2010, 12:22:33 PM, you wrote:
> On Thu, 2010-02-11 at 06:18 -0500, Sander Eikelenboom wrote:
>> Hello Daniel,
>> 
>> Ah these patches are actually for pvops (since the attached diffs are
all called linux-2.6.18-xen .....)
>> Will give them a try then :-)
> Sorry if that was confusing. I thought the subject line makes it clear.
> Most of them were ported forward from 2.6.18 and the naming just keeps
> track of which is which.
> Daniel



-- 
Best regards,
 Sander                            mailto:linux@eikelenboom.it


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

Hello Daniel,

Done some benchmarking with bonnie++ and it seems blktap2 at present performs
slightly better than loopback with raw image files, didn''t see crashes
or other problems so far.
Will the disabling of shared mem for blktap2 give a big performance hit ?

--

Sander

Thursday, February 11, 2010, 10:03:02 PM, you wrote:
> On Thu, 2010-02-11 at 09:17 -0500, Sander Eikelenboom wrote: 
>> Hello Daniel,
>> 
>> Patches applied, and it seems to be working so far !
>> Thx.
>> 
>> Hope that Jeremy hasn''t ignored the patches because of the
name of the diffs.
> Valid point.
> Hey Jeremy. Is what''s the situation wrt getting those into git?
> http://lists.xensource.com/archives/html/xen-devel/2010-02/msg00331.html
> Daniel




-- 
Best regards,
 Sander                            mailto:linux@eikelenboom.it


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

On Fri, 2010-02-12 at 08:21 -0500, Sander Eikelenboom
wrote:
> Hello Daniel,
> 
> Done some benchmarking with bonnie++ and it seems blktap2 at present
performs slightly better than loopback with raw image files, didn''t see
crashes or other problems so far.
Good to hear. Thanks for the feedback and keep me cc''d if you can.
> Will the disabling of shared mem for blktap2 give a big performance hit ?
Memshr is mainly a h/v extension, only driven by tapdisk.

I didn''t catch up on the implementation yet. But the idea is to
establish cow mappings on clean buffers read from shared disk images.
Provided the mappings are shadowed in one way or the other. Think
OS-level pagecache.

It''s a really good idea. But unless you''re actually running a
significant number of clones from a shared master image it will be
generating nothing but overhead.

I think any non-manual decision on whether to turn it on or off should
reside in the hands of a host agent overseeing the potential benefit.

Which is beyond tapdisk, and certainly beyond a hardcoded Makefile
macro.

Whether crashing or not, the medium term update should turn it into a
proper block driver layer.

Cheers,
Daniel



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