From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sat, 21 Apr 2007 02:55:17 -0700 From: Andrew Morton Subject: Re: [PATCH 04/10] lib: percpu_counter_mod64 Message-Id: <20070421025517.d9f9bc14.akpm@linux-foundation.org> In-Reply-To: <20070420155502.787144532@chello.nl> References: <20070420155154.898600123@chello.nl> <20070420155502.787144532@chello.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Peter Zijlstra Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, miklos@szeredi.hu, neilb@suse.de, dgc@sgi.com, tomoki.sekiyama.qu@hitachi.com, nikita@clusterfs.com, trond.myklebust@fys.uio.no, yingchao.zhou@gmail.com List-ID: On Fri, 20 Apr 2007 17:51:58 +0200 Peter Zijlstra wrote: > Add percpu_counter_mod64() to allow large modifications. > > Signed-off-by: Peter Zijlstra > --- > include/linux/percpu_counter.h | 9 +++++++++ > lib/percpu_counter.c | 28 ++++++++++++++++++++++++++++ > 2 files changed, 37 insertions(+) > > Index: linux-2.6/include/linux/percpu_counter.h > =================================================================== > --- linux-2.6.orig/include/linux/percpu_counter.h 2007-04-12 13:54:55.000000000 +0200 > +++ linux-2.6/include/linux/percpu_counter.h 2007-04-12 14:00:21.000000000 +0200 > @@ -36,6 +36,7 @@ static inline void percpu_counter_destro > } > > void percpu_counter_mod(struct percpu_counter *fbc, s32 amount); > +void percpu_counter_mod64(struct percpu_counter *fbc, s64 amount); > s64 percpu_counter_sum(struct percpu_counter *fbc); > > static inline s64 percpu_counter_read(struct percpu_counter *fbc) > @@ -81,6 +82,14 @@ percpu_counter_mod(struct percpu_counter > preempt_enable(); > } > > +static inline void > +percpu_counter_mod64(struct percpu_counter *fbc, s64 amount) > +{ > + preempt_disable(); > + fbc->count += amount; > + preempt_enable(); > +} > + > static inline s64 percpu_counter_read(struct percpu_counter *fbc) > { > return fbc->count; > Index: linux-2.6/lib/percpu_counter.c > =================================================================== > --- linux-2.6.orig/lib/percpu_counter.c 2006-07-31 13:07:38.000000000 +0200 > +++ linux-2.6/lib/percpu_counter.c 2007-04-12 14:17:12.000000000 +0200 > @@ -25,6 +25,34 @@ void percpu_counter_mod(struct percpu_co > } > EXPORT_SYMBOL(percpu_counter_mod); > > +void percpu_counter_mod64(struct percpu_counter *fbc, s64 amount) > +{ > + long count; > + s32 *pcount; > + int cpu; > + > + if (amount >= FBC_BATCH || amount <= -FBC_BATCH) { > + spin_lock(&fbc->lock); > + fbc->count += amount; > + spin_unlock(&fbc->lock); > + return; > + } This is wrong, a little. If the counter was at -FBC_BATCH/2 and the caller passed in FBC_BATCH, we could just set the cpu-local counter to FBC_BATCH/2 instead of going for the lock. Probably doesn't matter though. > + cpu = get_cpu(); > + pcount = per_cpu_ptr(fbc->counters, cpu); > + count = *pcount + amount; > + if (count >= FBC_BATCH || count <= -FBC_BATCH) { > + spin_lock(&fbc->lock); > + fbc->count += count; > + *pcount = 0; > + spin_unlock(&fbc->lock); > + } else { > + *pcount = count; > + } > + put_cpu(); > +} > +EXPORT_SYMBOL(percpu_counter_mod64); Bloaty. Surely we won't be needing this on 32-bit kernels? Even monster PAE has only 64,000,000 pages and won't be using deltas of more than 4 gigapages? -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org