Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758098Ab1COP4W (ORCPT ); Tue, 15 Mar 2011 11:56:22 -0400 Received: from mx1.redhat.com ([209.132.183.28]:27713 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758076Ab1COP4U (ORCPT ); Tue, 15 Mar 2011 11:56:20 -0400 Date: Tue, 15 Mar 2011 11:56:13 -0400 From: Vivek Goyal To: Shaohua Li Cc: Jens Axboe , lkml , Jeff Moyer , Corrado Zoccolo , Gui Jianfeng Subject: Re: [PATCH]cfq-iosched: removing unnecessary think time checking Message-ID: <20110315155613.GA2587@redhat.com> References: <1300157851.2337.78.camel@sli10-conroe> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1300157851.2337.78.camel@sli10-conroe> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2490 Lines: 65 On Tue, Mar 15, 2011 at 10:57:31AM +0800, Shaohua Li wrote: > Removing think time checking. A high thinktime queue might means the queue > dispatches several requests and then do away. Limitting such queue seems > meaningless. And also this can simplify code. This is suggested by Vivek. > > Signed-off-by: Shaohua Li > Looks good to me. Acked-by: Vivek Goyal I really can't find a relation between why should we allow unlimited dispatch only for low thinktime queues and not for high thinktime queue. concern here is that due to preemtion it can happen that async IO is starved. I think high think time queue automatically mitigates that up to some extent that there will be time interval when async IO will get time to dispatch. If we run into issues of sync starving async, then it should probably controlled by not giving unlimited queue depth to sync but by limiting it to some higher number. Thanks Vivek > --- > block/cfq-iosched.c | 13 ++++--------- > 1 file changed, 4 insertions(+), 9 deletions(-) > > Index: linux/block/cfq-iosched.c > =================================================================== > --- linux.orig/block/cfq-iosched.c 2011-03-10 13:52:13.000000000 +0800 > +++ linux/block/cfq-iosched.c 2011-03-10 13:53:08.000000000 +0800 > @@ -2418,19 +2418,14 @@ static bool cfq_may_dispatch(struct cfq_ > return false; > > /* > - * If there is only one sync queue, and its think time is > - * small, we can ignore async queue here and give the sync > + * If there is only one sync queue > + * we can ignore async queue here and give the sync > * queue no dispatch limit. The reason is a sync queue can > * preempt async queue, limiting the sync queue doesn't make > * sense. This is useful for aiostress test. > */ > - if (cfq_cfqq_sync(cfqq) && cfqd->busy_sync_queues == 1) { > - struct cfq_io_context *cic = RQ_CIC(cfqq->next_rq); > - > - if (sample_valid(cic->ttime_samples) && > - cic->ttime_mean < cfqd->cfq_slice_idle) > - promote_sync = true; > - } > + if (cfq_cfqq_sync(cfqq) && cfqd->busy_sync_queues == 1) > + promote_sync = true; > > /* > * We have other queues, don't allow more IO from this one > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/