> In my opinion, relying on apparent ordering instead of enforcing it may > lead to subtil race conditions. We may, for example, be victimized by some > hardware erratum due to brain-deaded hardware optimizations, if we try to > be clever when we should avoid to do so. >
I agree, but
- Ingo molnar noticed a 4 % speed-up with lmbench - IMHO this means that we
should try hard to use the optimization.
- it's a CPU semaphore, ie a bad PCI card cannot cause any problems.
- both CPU's are very similar [Intel says that both should have the same
stepping, and the list of supported combinations with different steppings is
short]
--
Manfred
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/