Well, maybe you like under testing things, but personally I don't. What
do you want a microoptimization? Get real, you don't accept
microoptimizations to networking code without numbers from the bigger
picture. So that means running real tests on real applications like
Oracle or RHDB. I think you'll find that the cross cpu case will be
quite happy with a range flush. Heck, it's the kind of IPI that can
even legally be asynchronous. Why don't you go do something useful
like implement that instead of making me spend my time to correct other
people's obvious mistakes?
-ben
-- Fish. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/