Le samedi 05 mai 2012 à 19:40 +0200, Jaromír Červenka a écrit :
today, one of my servers frozen out of nowhere. After hard reset, I
found this in /var/log/messages
Linux onca 3.1.9-1.4-default #1 SMP Fri Jan 27 08:55:10 UTC 2012
(efb5ff4) x86_64 x86_64 x86_64 GNU/Linux
Could it be a bug of kernel module or HW error?
That would be a hardware error. What would be interesting to know is
what happened right before the SMBus transaction timeout. No hint in the
I can help you more if you provide additional information on your
system. In particular :
* Does the w83795 driver return sane values usually?
* Does your board support IPMI, and do you have any ipmi driver loaded?
* Is there a BMC connected to the machine?
One frequent issue with the w83795 driver issue is when the monitoring
device is accessed both by the w83795 driver and by IPMI (either from
the host or from a BMC). As there is no synchronisation between them,
it's only a matter of time before a collision happens and everybody gets
Another possibility is that you have another chip on the SMBus which
went wrong and this blocked the SMBus completely, preventing other chips
on the bus from being accessed. It may be possible to improve the
i2c-i801 driver to recover from such situations. The problem is that
it's not easy to differentiate between the bus being genuinely busy
because another master is driving a transfer, and the bus looking busy
because a slave is wrongly holding the lines low. Also, if a slave
really went mad, there is nothing i2c-i801 can do, it's up to the
slave's driver to reset the chip if it has a way to do so (but the
hardware may be such that there is no such way other than a reboot.)
To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-kernel+owner(a)opensuse.org