Xen devel - Nov 2006 - VMX status report 12281:8d0e06c38c0c, block issue detected

We have tested the latest xen on VT platform with Intel 915/E8500
chipset.  Here is the test summary:

Bugs have been fixed:
1>.Create VMX domain cause IA32E xen0 reboot bug has been fixed.
2> unmodified drivers build issue is fixed and current it could be built
successfully

[Block issue]:
1).i386/pae/x86_64: Create VMX domain twice causes xen0 reboot while
xenU is Okay. See attachments for console logs.  

This issue blocks our test for VMX.


Best Regards
Ping
 





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

On 8/11/06 7:19 am, "Yu, Ping Y" <ping.y.yu@intel.com> wrote:
> [Block issue]:
> 1).i386/pae/x86_64: Create VMX domain twice causes xen0 reboot while
> xenU is Okay. See attachments for console logs.
Somehow I''ve started corrupting memory. :-(

At least it looks easy to reproduce!

 -- Keir



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

On 8/11/06 08:35, "Keir Fraser" <Keir.Fraser@cl.cam.ac.uk>
wrote:
>> [Block issue]:
>> 1).i386/pae/x86_64: Create VMX domain twice causes xen0 reboot while
>> xenU is Okay. See attachments for console logs.
> 
> Somehow I''ve started corrupting memory. :-(
> 
> At least it looks easy to reproduce!
FYI, the bug is the final xfree(vlapic) in vlapic_destroy(). If that line is
removed then HVM guest destruction works again.

The patch is in the staging tree, c/s 12296:1fcdb3a35db3. It should get
pushed to the public tree fairly soon.

 -- Keir



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