Date: Fri, 17 Jul 2015 17:35:09 +0200
From: Frederic Weisbecker <fweisbec@gmail.com>
To: Tejun Heo <tj@kernel.org>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Lai Jiangshan <jiangshanlai@gmail.com>, Rik van Riel <riel@redhat.com>,
        "Luis Claudio R. Goncalves" <lclaudio@uudg.org>
Subject: Re: [RFC] workqueue: avoiding unbounded wq on isolated CPUs by
 default
Message-ID: <20150717153507.GB20260@lerouge>
References: <9e53de7c91c885ee255e16ee25f401d9eedf08d9.1437067317.git.bristot@redhat.com>
 <20150716192448.GY15934@mtj.duckdns.org>
 <1437107190.3438.23.camel@gmail.com>
 <20150717152720.GD15934@mtj.duckdns.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150717152720.GD15934@mtj.duckdns.org>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1593
Lines: 40

On Fri, Jul 17, 2015 at 11:27:20AM -0400, Tejun Heo wrote:
> Hello, Mike.
> 
> On Fri, Jul 17, 2015 at 06:26:30AM +0200, Mike Galbraith wrote:
> > Hm, I thought the plan was that after the Lai's unbound series landed,
> > his ordered wq patch would follow, but perhaps not.
> 
> Yes, that still is the plan but this is kinda unrelated to that
> change.  This just initializes wq cpumask according to cpu isolation.
> I'm just curious whether there was any specific reason we didn't do
> this before (ISTR people discussing it back then too).

Initializing wq unbound cpumask to housekeeping_mask is still the
plan. I just remember we didn't do it in Lai's series because it
was slightly unrelated. When a patchset is complicated, like Lai's,
it's better to keep it focus to a single purpose.

Anyway that patch is welcome.

> 
> > I'm referring to the somewhat aged patch below. (freshly wedged into
> > master, and maybe not properly, but it should at least look familiar).
> 
> Yeah, I think I asked Lai to try a different approach where we
> regulate it from queueing path rather than playing with pwqs.  I think
> that'd end up quite a bit simpler.

Ordered workqueues aren't handled currently? I tried setting the unbound
cpumask and it also applied to khelper which is a singlethread (and thus
ordered) workqueue.

> Thanks.
> 
> -- 
> tejun
--
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/