From: Tejun Heo <tj@kernel.org>
To: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: linux-kernel@vger.kernel.org, Lai Jiangshan <laijs@cn.fujitsu.com>
Subject: Re: [PATCHSET] workqueue: remove gcwq and make worker_pool the only backend abstraction
Date: Tue, 22 Jan 2013 17:07:58 -0800 [thread overview]
Message-ID: <20130123010758.GH5359@htj.dyndns.org> (raw)
In-Reply-To: <20130122053702.GA10675@lge.com>
Hey.
On Tue, Jan 22, 2013 at 02:37:02PM +0900, Joonsoo Kim wrote:
> > It seems like we'll need to support worker pools with custom
> > attributes, which is planned to be implemented as extra worker_pools
> > for the unbound CPU. Removing gcwq and having worker_pool as the top
> > level abstraction makes things much simpler for such designs. Also,
> > there's scalability benefit to not sharing locking and busy hash among
> > different worker pools as worker pools w/ custom attributes are likely
> > to have widely different memory / cpu locality characteristics.
>
> Could you tell me why extra worker_pools with custom attributes are needed?
> Or could you give a reference link for this?
Currently, there are two expected users - writeback and crypto. The
former currently implements its own worker pool and the latter is
using per-cpu workqueue but not particularly happy with it. Being
bound to the issuing CPU seems a bit too limiting.
Thanks.
--
tejun
next prev parent reply other threads:[~2013-01-23 1:08 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-17 1:42 Tejun Heo
2013-01-17 1:42 ` [PATCH 01/17] workqueue: unexport work_cpu() Tejun Heo
2013-01-17 1:42 ` [PATCH 02/17] workqueue: use std_ prefix for the standard per-cpu pools Tejun Heo
2013-01-17 1:42 ` [PATCH 03/17] workqueue: make GCWQ_DISASSOCIATED a pool flag Tejun Heo
2013-01-17 1:42 ` [PATCH 04/17] workqueue: make GCWQ_FREEZING " Tejun Heo
2013-01-17 1:42 ` [PATCH 05/17] workqueue: introduce WORK_OFFQ_CPU_NONE Tejun Heo
2013-01-17 1:42 ` [PATCH 06/17] workqueue: add worker_pool->id Tejun Heo
2013-01-17 1:42 ` [PATCH 07/17] workqueue: record pool ID instead of CPU in work->data when off-queue Tejun Heo
2013-01-17 1:42 ` [PATCH 08/17] workqueue: move busy_hash from global_cwq to worker_pool Tejun Heo
2013-01-17 1:42 ` [PATCH 09/17] workqueue: move global_cwq->cpu " Tejun Heo
2013-01-17 1:42 ` [PATCH 10/17] workqueue: move global_cwq->lock " Tejun Heo
2013-01-17 1:42 ` [PATCH 11/17] workqueue: make hotplug processing per-pool Tejun Heo
2013-01-17 1:42 ` [PATCH 12/17] workqueue: make freezing/thawing per-pool Tejun Heo
2013-01-17 1:42 ` [PATCH 13/17] workqueue: replace for_each_worker_pool() with for_each_std_worker_pool() Tejun Heo
2013-01-17 1:42 ` [PATCH 14/17] workqueue: remove worker_pool->gcwq Tejun Heo
2013-01-17 1:42 ` [PATCH 15/17] workqueue: remove global_cwq Tejun Heo
2013-01-22 6:50 ` Joonsoo Kim
2013-01-23 1:09 ` Tejun Heo
2013-01-23 18:09 ` [PATCH v2 " Tejun Heo
2013-01-24 9:29 ` Joonsoo Kim
2013-01-24 18:44 ` Tejun Heo
2013-01-17 1:42 ` [PATCH 16/17] workqueue: rename nr_running variables Tejun Heo
2013-01-17 1:42 ` [PATCH 17/17] workqueue: post global_cwq removal cleanups Tejun Heo
2013-01-17 1:48 ` [PATCHSET] workqueue: remove gcwq and make worker_pool the only backend abstraction Tejun Heo
2013-01-17 3:25 ` Wanlong Gao
2013-01-17 19:11 ` Tejun Heo
2013-01-22 5:37 ` Joonsoo Kim
2013-01-23 1:07 ` Tejun Heo [this message]
2013-01-24 13:36 ` Lai Jiangshan
2013-01-24 18:51 ` Tejun Heo
2013-01-24 19:03 ` Tejun Heo
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=20130123010758.GH5359@htj.dyndns.org \
--to=tj@kernel.org \
--cc=iamjoonsoo.kim@lge.com \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®