Re: RCU fanout leaf balancing

2014-10-26 Thread Daniel J Blueman
On 25/10/2014 21:48, Paul E. McKenney wrote: On Sat, Oct 25, 2014 at 02:59:24PM +0800, Daniel J Blueman wrote: Hi Paul, Finding earlier reference to increasing RCU fanout leaf for the purpose of "decrease[ing] cache-miss overhead for large systems", would your suggestion be to increase the valu

RCU fanout leaf balancing

2014-10-25 Thread Daniel J Blueman
Hi Paul, Finding earlier reference to increasing RCU fanout leaf for the purpose of "decrease[ing] cache-miss overhead for large systems", would your suggestion be to increase the value to the next hierarchy core-count above 16? If we have say 32 interconnected 48-core servers; 3 sockets of

Re: RCU fanout leaf balancing

2014-10-25 Thread Paul E. McKenney
On Sat, Oct 25, 2014 at 02:59:24PM +0800, Daniel J Blueman wrote: > Hi Paul, > > Finding earlier reference to increasing RCU fanout leaf for the > purpose of "decrease[ing] cache-miss overhead for large systems", > would your suggestion be to increase the value to the next hierarchy > core-count a