Hi,

OK I believe I found the root course of the panics people are getting
from the Intel IOMMU code.
This was though to have been due to buggy BIOS's that have not cough
up with the times of VT. However I don't believe this is the case and
is in fact a Solaris kernel bug.

BugID: http://bugs.opensolaris.org/view_bug.do?bug_id=6811593

Our panics seem to be from recent BIOS's initializing the USB ports
for things like 'USB HDD boot support' or 'USB CDROM boot support',
turning this off solved the hang/panic. What I believe Solaris is
doing is trying to reinitialize the hardware and registers and
returning unexpected values thus we are in a panic. Linux does not
seem to suffer from the same problem. What I believe is that Solaris
is missing the check to see if the hardware is already initialized or
not _before_ it tired too do such. There is still some funy stuff in
my dmesg after disabling this in the BIOS. I would be happy to provide
more info if required. If someone from Sun and/or Intel could contact
me who is working on this bug (there are others just like this one) I
would be happy to work with them so this is fixed for the OSOL2009.x
release as I know your thinking of disabling IOMMU by default if you
can't fix it.

I have tested this theory on 3 boxes and it seems to be consistent
although I have not had the time to review the code in detail yet.

Best Regards,
Edward O'Callaghan.

-- 
All Documents adhered to the ISO/IEC 26300  standard file format for
electronic office documents, such as spreadsheets, charts,
presentations and word processing documents from this email address.
The author does not take responsibility of the recipients inability to
read international standards and who use proprietary products such as
MS Office.
See: http://www.openoffice.org/
Edward O'Callaghan.
--
http://fedoraproject.org/wiki/EdwardOcallaghan
http://moonshine.opn4.org/
http://www.pcbsd.org/

Reply via email to