Spinlock and cache coherency

[Bloch, Jack]

I am running a SuSE 2.4.19 Kernel. We have a quad CPU system (actually two
physical CPU's with hyperthreading. We have encountered a problem where
three of four CPU's are locked in the NET_RX_ACTION waiting for the
BR_NETPROTO_LOCK spinlock. The fourth is spiinning on another internal
driver spinlock (spinlock_irqsave). The author of this driver points out to
us the cache coherency issues related to spinlock specifically with the
Intel errata G40 "Potential Loss of Data Coherency During MP Data Ownership
Transfer". Anyone seen any of these issues?


Please CC me directly on any responses.


Regards,


Jack
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/