On Wed, Aug 17, 2011 at 10:08:04AM -0700, Paolo Bonzini wrote: > On 08/17/2011 07:31 AM, Mathieu Desnoyers wrote: > >So I guess that in that sense, synchronize_rcu() of MB and QSBR could be > >merged, is that what you are trying to get at ? (sorry for being slow on > >the uptake);-) > > Yes, and perhaps MEMBARRIER too---leaving just SIGNAL as the oddball > one. That's also why I started doing the futex optimization in > QSBR. > > Did you measure SIGNAL to have better performance in practice than MB?
Yep. See http://www.rdrop.com/users/paulmck/preprints/, click on "main paper" and see figures 6 and later. Note that the "Signal-based RCU" line is significantly higher than the "General-purpose RCU" line, and keep in mind that these plots use logscale y axes. Of course, your mileage may vary. Thanx, Paul _______________________________________________ ltt-dev mailing list [email protected] http://lists.casi.polymtl.ca/cgi-bin/mailman/listinfo/ltt-dev
