From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1422755AbXCWJkR (ORCPT ); Fri, 23 Mar 2007 05:40:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1422757AbXCWJkQ (ORCPT ); Fri, 23 Mar 2007 05:40:16 -0400 Received: from pfx2.jmh.fr ([194.153.89.55]:37014 "EHLO pfx2.jmh.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1422755AbXCWJkP (ORCPT ); Fri, 23 Mar 2007 05:40:15 -0400 Date: Fri, 23 Mar 2007 10:40:17 +0100 From: Eric Dumazet To: Nick Piggin Cc: Linux Kernel Mailing List , Ravikiran G Thirumalai , Ingo Molnar Subject: Re: [rfc][patch] queued spinlocks (i386) Message-Id: <20070323104017.cc1ea4fe.dada1@cosmosbay.com> In-Reply-To: <20070323085910.GA11577@wotan.suse.de> References: <20070323085910.GA11577@wotan.suse.de> X-Mailer: Sylpheed 2.3.1 (GTK+ 2.10.6; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 23 Mar 2007 09:59:11 +0100 Nick Piggin wrote: > > Implement queued spinlocks for i386. This shouldn't increase the size of > the spinlock structure, while still able to handle 2^16 CPUs. > > Not completely implemented with assembly yet, to make the algorithm a bit > clearer. > > The queued spinlock has 2 fields, a head and a tail, which are indexes > into a FIFO of waiting CPUs. To take a spinlock, a CPU performs an > "atomic_inc_return" on the head index, and keeps the returned value as > a ticket. The CPU then spins until the tail index is equal to that > ticket. > > To unlock a spinlock, the tail index is incremented (this can be non > atomic, because only the lock owner will modify tail). > > Implementation inefficiencies aside, this change should have little > effect on performance for uncontended locks, but will have quite a > large cost for highly contended locks [O(N) cacheline transfers vs > O(1) per lock aquisition, where N is the number of CPUs contending]. > The benefit is is that contended locks will not cause any starvation. > > Just an idea. Big NUMA hardware seems to have fairness logic that > prevents starvation for the regular spinlock logic. But it might be > interesting for -rt kernel or systems with starvation issues. It's a very nice idea Nick. You also have for free the number or cpus that are before you. On big SMP/NUMA, we could use this information to call a special lock_cpu_relax() function to avoid cacheline transferts. asm volatile(LOCK_PREFIX "xaddw %0, %1\n\t" : "+r" (pos), "+m" (lock->qhead) : : "memory"); for (;;) { unsigned short nwait = pos - lock->qtail; if (likely(nwait == 0)) break; lock_cpu_relax(lock, nwait); } lock_cpu_relax(raw_spinlock_t *lock, unsigned int nwait) { unsigned int cycles = nwait * lock->min_cycles_per_round; busy_loop(cycles); }