All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: Jeff Moyer <jmoyer@redhat.com>
Cc: linux-kernel@vger.kernel.org, jens.axboe@oracle.com,
	nauman@google.com, dpshah@google.com, lizf@cn.fujitsu.com,
	ryov@valinux.co.jp, fernando@oss.ntt.co.jp,
	s-uchida@ap.jp.nec.com, taka@valinux.co.jp,
	guijianfeng@cn.fujitsu.com, balbir@linux.vnet.ibm.com,
	righi.andrea@gmail.com, m-ikeda@ds.jp.nec.com,
	akpm@linux-foundation.org, riel@redhat.com,
	kamezawa.hiroyu@jp.fujitsu.com
Subject: Re: [PATCH 15/20] blkio: Take care of preemptions across groups
Date: Wed, 4 Nov 2009 14:27:25 -0500	[thread overview]
Message-ID: <20091104192725.GM2870@redhat.com> (raw)
In-Reply-To: <x49d43yt8ry.fsf@segfault.boston.devel.redhat.com>

On Wed, Nov 04, 2009 at 02:00:33PM -0500, Jeff Moyer wrote:
> Vivek Goyal <vgoyal@redhat.com> writes:
> 
> > +static bool cfq_should_preempt_group(struct cfq_data *cfqd,
> > +		struct cfq_queue *cfqq, struct cfq_queue *new_cfqq)
> > +{
> > +	struct cfq_entity *cfqe = &cfqq->entity;
> > +	struct cfq_entity *new_cfqe = &new_cfqq->entity;
> > +
> > +	if (cfqq_to_cfqg(cfqq) != &cfqd->root_group)
> > +		cfqe = parent_entity(&cfqq->entity);
> > +
> > +	if (cfqq_to_cfqg(new_cfqq) != &cfqd->root_group)
> > +		new_cfqe = parent_entity(&new_cfqq->entity);
> > +
> > +	/*
> > +	 * Allow an RT request to pre-empt an ongoing non-RT cfqq timeslice.
> > +	 */
> > +
> > +	if (new_cfqe->ioprio_class == IOPRIO_CLASS_RT
> > +	    && cfqe->ioprio_class != IOPRIO_CLASS_RT)
> > +		return true;
> > +	/*
> > +	 * Allow an BE request to pre-empt an ongoing IDLE clas timeslice.
> > +	 */
> > +
> > +	if (new_cfqe->ioprio_class == IOPRIO_CLASS_BE
> > +	    && cfqe->ioprio_class == IOPRIO_CLASS_IDLE)
> > +		return true;
> > +
> > +	return false;
> > +}
> 
> What was the motivation for this?  It seems like this would really break
> isolation.  What if one group has all RT priority tasks, will it starve
> out the other groups?
> 

It will not as we traverse up the hierarchy and look for the ioprio class
of the group entity.

So if you got following configuration where G1 and G2 are two groups. G1
is prio class RT and G2 is prio class BE, then any queue in G1 will
preempt any queue in G2 as at highest level, G1 and G2 are different class
altogether.

			root
			/  \
		       G1  G2

Normal cfqq preemption checks will not catch this. So if G2 has some BE
cfqq running, and some BE queue gets backlogged in G1, this new queue wil
not preempt the queue in G2 and it should have.

That's why preemption checks at group level.

Secondly if G1 and G2 are of ioprioclass BE and all the jobs in G1 are of
RT nature, they will not preempt the queues in G2, hence providing
isolation.

Thanks
Vivek

  reply	other threads:[~2009-11-04 19:28 UTC|newest]

Thread overview: 88+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-11-03 23:43 [RFC] Block IO Controller V1 Vivek Goyal
2009-11-03 23:43 ` [PATCH 01/20] blkio: Documentation Vivek Goyal
2009-11-04 13:37   ` Jeff Moyer
2009-11-04 17:21   ` Balbir Singh
2009-11-04 17:52     ` Vivek Goyal
2009-11-04 23:36       ` Balbir Singh
2009-11-03 23:43 ` [PATCH 02/20] blkio: Change CFQ to use CFS like queue time stamps Vivek Goyal
2009-11-04 14:30   ` Jeff Moyer
2009-11-04 16:37     ` Vivek Goyal
2009-11-04 17:59       ` Corrado Zoccolo
2009-11-04 18:54         ` Vivek Goyal
2009-11-05  2:44       ` Divyesh Shah
2009-11-05 14:39         ` Vivek Goyal
2009-11-04 21:18   ` Corrado Zoccolo
2009-11-04 22:25     ` Vivek Goyal
2009-11-05  8:36       ` Corrado Zoccolo
2009-11-04 23:22     ` Vivek Goyal
2009-11-05  8:27       ` Corrado Zoccolo
2009-11-05  0:05     ` Vivek Goyal
2009-11-06 22:22     ` [RFC] Workload type Vs Groups (Was: Re: [PATCH 02/20] blkio: Change CFQ to use CFS like queue time stamps) Vivek Goyal
2009-11-09 17:33       ` Nauman Rafique
2009-11-09 21:47       ` Corrado Zoccolo
2009-11-09 23:12         ` Vivek Goyal
2009-11-10 11:29           ` Corrado Zoccolo
2009-11-10 13:31             ` Vivek Goyal
2009-11-10 14:12               ` Vivek Goyal
2009-11-10 18:05                 ` Corrado Zoccolo
2009-11-10 19:15                   ` Vivek Goyal
2009-11-12  8:53                     ` Corrado Zoccolo
2009-11-11  0:48   ` [PATCH 02/20] blkio: Change CFQ to use CFS like queue time stamps Gui Jianfeng
2009-11-12 23:07     ` Vivek Goyal
2009-11-13  0:59       ` Gui Jianfeng
2009-11-13  1:24         ` Vivek Goyal
2009-11-13  2:05           ` Gui Jianfeng
2009-11-03 23:43 ` [PATCH 03/20] blkio: Introduce the notion of weights Vivek Goyal
2009-11-04 15:06   ` Jeff Moyer
2009-11-04 15:41     ` Vivek Goyal
2009-11-04 17:07       ` Divyesh Shah
2009-11-04 19:00         ` Vivek Goyal
2009-11-04 19:15       ` Jeff Moyer
2009-11-03 23:43 ` [PATCH 04/20] blkio: Introduce the notion of cfq entity Vivek Goyal
2009-11-03 23:43 ` [PATCH 05/20] blkio: Introduce the notion of cfq groups Vivek Goyal
2009-11-03 23:43 ` [PATCH 06/20] blkio: Introduce cgroup interface Vivek Goyal
2009-11-04 15:23   ` Jeff Moyer
2009-11-04 16:47     ` Vivek Goyal
2009-11-03 23:43 ` [PATCH 07/20] blkio: Provide capablity to enqueue/dequeue group entities Vivek Goyal
2009-11-04 15:34   ` Jeff Moyer
2009-11-04 16:54     ` Vivek Goyal
2009-11-03 23:43 ` [PATCH 08/20] blkio: Add support for dynamic creation of cfq_groups Vivek Goyal
2009-11-04 16:01   ` Jeff Moyer
2009-11-03 23:43 ` [PATCH 09/20] blkio: Porpogate blkio cgroup weight or ioprio class updation to cfq groups Vivek Goyal
2009-11-05  5:35   ` Gui Jianfeng
2009-11-05 14:42     ` Vivek Goyal
2009-11-03 23:43 ` [PATCH 10/20] blkio: Implement cfq group deletion and reference counting support Vivek Goyal
2009-11-04 18:44   ` Jeff Moyer
2009-11-04 19:00     ` Vivek Goyal
2009-11-03 23:43 ` [PATCH 11/20] blkio: Some CFQ debugging Aid Vivek Goyal
2009-11-04 18:52   ` Jeff Moyer
2009-11-04 19:12     ` Vivek Goyal
2009-11-04 19:25       ` Jeff Moyer
2009-11-05  3:10   ` Divyesh Shah
2009-11-05 14:42     ` Vivek Goyal
2009-11-06  0:56       ` Divyesh Shah
2009-11-03 23:43 ` [PATCH 12/20] blkio: Export disk time and sectors dispatched from cgroup interface Vivek Goyal
2009-11-03 23:43 ` [PATCH 13/20] blkio: Add a group dequeue interface in cgroup for debugging Vivek Goyal
2009-11-03 23:43 ` [PATCH 14/20] blkio: Do not allow request merging across cfq groups Vivek Goyal
2009-11-03 23:43 ` [PATCH 15/20] blkio: Take care of preemptions across groups Vivek Goyal
2009-11-04 19:00   ` Jeff Moyer
2009-11-04 19:27     ` Vivek Goyal [this message]
2009-11-04 19:30       ` Jeff Moyer
2009-11-06  7:55   ` Gui Jianfeng
2009-11-06 22:10     ` Vivek Goyal
2009-11-09  7:41       ` Gui Jianfeng
2009-11-03 23:43 ` [PATCH 16/20] blkio: do not select co-operating queues from different cfq groups Vivek Goyal
2009-11-03 23:43 ` [PATCH 17/20] blkio: Wait for queue to get backlogged before it expires Vivek Goyal
2009-11-03 23:43 ` [PATCH 18/20] blkio: arm idle timer even if think time is great then time slice left Vivek Goyal
2009-11-04 19:04   ` Jeff Moyer
2009-11-04 19:17     ` Vivek Goyal
2009-11-03 23:43 ` [PATCH 19/20] blkio: Arm slice timer even if there are requests in driver Vivek Goyal
2009-11-03 23:43 ` [PATCH 20/20] blkio: Drop the reference to queue once the task changes cgroup Vivek Goyal
2009-11-04 19:09   ` Jeff Moyer
2009-11-04 19:18     ` Vivek Goyal
2009-11-04  7:43 ` [RFC] Block IO Controller V1 Jens Axboe
2009-11-04 13:39   ` Vivek Goyal
2009-11-04 19:12 ` Jeff Moyer
2009-11-04 19:19   ` Vivek Goyal
2009-11-04 19:27     ` Jeff Moyer
2009-11-04 19:38       ` Vivek Goyal

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20091104192725.GM2870@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=balbir@linux.vnet.ibm.com \
    --cc=dpshah@google.com \
    --cc=fernando@oss.ntt.co.jp \
    --cc=guijianfeng@cn.fujitsu.com \
    --cc=jens.axboe@oracle.com \
    --cc=jmoyer@redhat.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lizf@cn.fujitsu.com \
    --cc=m-ikeda@ds.jp.nec.com \
    --cc=nauman@google.com \
    --cc=riel@redhat.com \
    --cc=righi.andrea@gmail.com \
    --cc=ryov@valinux.co.jp \
    --cc=s-uchida@ap.jp.nec.com \
    --cc=taka@valinux.co.jp \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.