Lu Baolu wrote:
Hi Folks,

Has anybody tried to install OpenSolaris as a HVM guest on xvm3.3
built from source in gate?


Hmm, no. we haven't seen this... Could you sync up to the
bits which were pushed today and see if you still see the
problem?

If you see it with the current bits, we'll need to look into it further.


Thanks,


MRJ


There are lots of message lines as below out from serial port of domain 0.

... ...
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2941000) failed, pte=0x401000002f6b8667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2942000) failed, pte=0x401000002f6b1667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2943000) failed, pte=0x401000002f6b0667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2944000) failed, pte=0x401000002f69d667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2945000) failed, pte=0x401000002f69c667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2946000) failed, pte=0x401000002f925667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2947000) failed, pte=0x401000002f924667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c295e000) failed, pte=0x401000002f91f667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c295f000) failed, pte=0x401000002f91e667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2960000) failed, pte=0x401000002f919667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2961000) failed, pte=0x401000002f918667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2962000) failed, pte=0x401000002f915667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2932000) failed, pte=0x401000002f914667
Apr  7 14:29:56 blu-nhm unix: WARNING:
xen_fix_foreign(va=0xffffff01c2933000) failed, pte=0x401000002f913667
... ...

This kind of warning messages starts when a OpenSolaris (both
OpenSolaris200811 and Nevada builds) HVM guest starts and will not
stop until domain 0 restart.

The qemu-dm log repeats a same failure.

[al...@blu-nhm:~]tail -f /var/log/xen/qemu-dm-0811.log
... ...
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
track_dirty_vram(f0000000, 26) failed (-1, 22)
... ...

No such messages log when OpenSolaris HVM guest is shutdown.

There are not any other significant misbehaviors expect that HVM guest
is a little slow. Other HVM guest (Windows, Linux) don't cause this
warning. Is this a bug in list?

Thanks,
Allen
_______________________________________________
xen-discuss mailing list
[email protected]
_______________________________________________
xen-discuss mailing list
[email protected]

Reply via email to