From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754247Ab0CRBFF (ORCPT ); Wed, 17 Mar 2010 21:05:05 -0400 Received: from mga11.intel.com ([192.55.52.93]:31738 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751790Ab0CRBFD (ORCPT ); Wed, 17 Mar 2010 21:05:03 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.51,262,1267430400"; d="scan'208";a="781767589" Date: Thu, 18 Mar 2010 09:05:00 +0800 From: Shaohua Li To: Corrado Zoccolo Cc: Jens Axboe , "linux-kernel@vger.kernel.org" , "vgoyal@redhat.com" , "jmoyer@redhat.com" , "guijianfeng@cn.fujitsu.com" , "Shi, Alex" Subject: Re: [RFC]cfq-iosched: fix a kbuild regression Message-ID: <20100318010459.GA26716@sli10-desk.sh.intel.com> References: <20100316025656.GA15390@sli10-desk.sh.intel.com> <20100317133019.GV5768@kernel.dk> <4e5e476b1003171124y791a25a8n6c31368f182b52a9@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4e5e476b1003171124y791a25a8n6c31368f182b52a9@mail.gmail.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Mar 18, 2010 at 02:24:10AM +0800, Corrado Zoccolo wrote: > On Wed, Mar 17, 2010 at 2:30 PM, Jens Axboe wrote: > > On Tue, Mar 16 2010, Shaohua Li wrote: > >> Alex Shi reported a kbuild regression which is about 10% performance lost. > >> He bisected to this commit: 3dde36ddea3e07dd025c4c1ba47edec91606fec0. > >> The reason is cfqq_close() can't find close cooperator. If we store the seek > >> distance to the value before the commit like below, the regression fully goes > >> away. If this is too invasive, just changing the cfq_rq_close() for the > >> !for_preempt is ok too. > > > > Corrado, any objections to widening the seek threshold? > > The previous seek threshold value was meant to be compared with the > average seek distance, so it was large to account for variations. > Since we handle the variations differently, we should have a smaller > value now (the 100 * 8 was the result of a benchmark run on several > disks). Our test doesn't find any issue with the seek threshold so far, so it proberly is ok. > I agreee, though, that for merging queues, it is now too small, so we > should have two thresholds, the current one used to determine if a > request causes a seek, and an other to be used inside cfq_close (with > the older value, regardless of for_preempt). That makes sense. Updated patch. Alex Shi reported a kbuild regression which is about 10% performance lost. He bisected to this commit: 3dde36ddea3e07dd025c4c1ba47edec91606fec0. The reason is cfqq_close() can't find close cooperator. Restoring cfq_rq_close()'s threshold to original value makes the regression go away. Since for_preempt parameter isn't used anymore, this patch deletes it. Reported-by: Alex Shi Signed-off-by: Shaohua Li diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index dee9d93..8d5a2f2 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -47,6 +47,7 @@ static const int cfq_hist_divisor = 4; #define CFQ_SERVICE_SHIFT 12 #define CFQQ_SEEK_THR (sector_t)(8 * 100) +#define CFQQ_CLOSE_THR (sector_t)(8 * 1024) #define CFQQ_SECT_THR_NONROT (sector_t)(2 * 32) #define CFQQ_SEEKY(cfqq) (hweight32(cfqq->seek_history) > 32/8) @@ -1660,9 +1661,9 @@ static inline sector_t cfq_dist_from_last(struct cfq_data *cfqd, } static inline int cfq_rq_close(struct cfq_data *cfqd, struct cfq_queue *cfqq, - struct request *rq, bool for_preempt) + struct request *rq) { - return cfq_dist_from_last(cfqd, rq) <= CFQQ_SEEK_THR; + return cfq_dist_from_last(cfqd, rq) <= CFQQ_CLOSE_THR; } static struct cfq_queue *cfqq_close(struct cfq_data *cfqd, @@ -1689,7 +1690,7 @@ static struct cfq_queue *cfqq_close(struct cfq_data *cfqd, * will contain the closest sector. */ __cfqq = rb_entry(parent, struct cfq_queue, p_node); - if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq, false)) + if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq)) return __cfqq; if (blk_rq_pos(__cfqq->next_rq) < sector) @@ -1700,7 +1701,7 @@ static struct cfq_queue *cfqq_close(struct cfq_data *cfqd, return NULL; __cfqq = rb_entry(node, struct cfq_queue, p_node); - if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq, false)) + if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq)) return __cfqq; return NULL; @@ -3103,7 +3104,7 @@ cfq_should_preempt(struct cfq_data *cfqd, struct cfq_queue *new_cfqq, * if this request is as-good as one we would expect from the * current cfqq, let it preempt */ - if (cfq_rq_close(cfqd, cfqq, rq, true)) + if (cfq_rq_close(cfqd, cfqq, rq)) return true; return false;