Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Tejun Heo <htejun@gmail.com>
Date:   Mon, 9 Nov 2020 11:10:07 -0500
From:   "tj@kernel.org" <tj@kernel.org>
To:     Trond Myklebust <trondmy@hammerspace.com>
Cc:     "jiangshanlai@gmail.com" <jiangshanlai@gmail.com>,
        "mhocko@suse.com" <mhocko@suse.com>,
        "peterz@infradead.org" <peterz@infradead.org>,
        "juri.lelli@redhat.com" <juri.lelli@redhat.com>,
        "mingo@redhat.com" <mingo@redhat.com>,
        "vincent.guittot@linaro.org" <vincent.guittot@linaro.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "neilb@suse.de" <neilb@suse.de>
Subject: Re: [PATCH rfc] workqueue: honour cond_resched() more effectively.
Message-ID: <20201109161007.GF7496@mtj.duckdns.org>
References: <87v9efp7cs.fsf@notabene.neil.brown.name>
 <20201109080038.GY2594@hirez.programming.kicks-ass.net>
 <aec65c71c09e803285688d5974193a98b4422428.camel@hammerspace.com>
 <20201109140141.GE7496@mtj.duckdns.org>
 <d2c79d91e29134ef6184138de5fc856ca530d2a5.camel@hammerspace.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <d2c79d91e29134ef6184138de5fc856ca530d2a5.camel@hammerspace.com>
Precedence: bulk

Hello,

On Mon, Nov 09, 2020 at 02:11:42PM +0000, Trond Myklebust wrote:
> That means changing all filesystem code to use cpu-intensive queues. As
> far as I can tell, they all use workqueues (most of them using the
> standard system queue) for fput(), dput() and/or iput() calls.

I suppose the assumption was that those operations couldn't possiby be
expensive enough to warrant other options, which doesn't seem to be the case
unfortunately. Switching the users to system_unbound_wq, which should be
pretty trivial, seems to be the straight forward solution.

I can definitely see benefits in making workqueue smarter about
concurrency-managed work items taking a long time. Given that nothing on
these types of workqueues can be latency sensitive and the problem being
reported is on the scale of tens of seconds, I think a more palatable
approach could be through watchdog mechanism rather than hooking into
cond_resched(). Something like:

* Run watchdog timer more frequently - e.g. 1/4 of threshold.

* If a work item is occupying the local concurrency for too long, set
  WORKER_CPU_INTENSIVE for the worker and, probably, generate a warning.

I still think this should generate a warning and thus can't replace
switching to unbound wq. The reason is that the concurrency limit isn't the
only problem. A kthread needing to run on one particular CPU for tens of
seconds just isn't great.

Thanks.

-- 
tejun