All of lore.kernel.org
 help / color / mirror / Atom feed
From: Waiman Long <waiman.long@hpe.com>
To: Davidlohr Bueso <dave@stgolabs.net>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Ding Tianhong <dingtianhong@huawei.com>,
	Jason Low <jason.low2@hp.com>, Ingo Molnar <mingo@redhat.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	"Paul E. McKenney" <paulmck@us.ibm.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Will Deacon <Will.Deacon@arm.com>,
	Tim Chen <tim.c.chen@linux.intel.com>,
	Waiman Long <Waiman.Long@hp.com>
Subject: Re: [PATCH] locking/mutex: Avoid spinner vs waiter starvation
Date: Wed, 03 Feb 2016 17:07:51 -0500	[thread overview]
Message-ID: <56B27A37.6070105@hpe.com> (raw)
In-Reply-To: <20160202211906.GD16147@linux-uzut.site>

On 02/02/2016 04:19 PM, Davidlohr Bueso wrote:
> On Mon, 01 Feb 2016, Peter Zijlstra wrote:
>
>> Subject: locking/mutex: Avoid spinner vs waiter starvation
>> From: Peter Zijlstra <peterz@infradead.org>
>> Date: Fri, 22 Jan 2016 12:06:53 +0100
>>
>> Ding Tianhong reported that under his load the optimistic spinners
>> would totally starve a task that ended up on the wait list.
>>
>> Fix this by ensuring the top waiter also partakes in the optimistic
>> spin queue.
>>
>> There are a few subtle differences between the assumed state of
>> regular optimistic spinners and those already on the wait list, which
>> result in the @acquired complication of the acquire path.
>>
>> Most notable are:
>>
>> - waiters are on the wait list and need to be taken off
>> - mutex_optimistic_spin() sets the lock->count to 0 on acquire
>>   even though there might be more tasks on the wait list.
>
> Right, the main impact I see with these complications are that the
> window of when a waiter takes the lock via spinning and then acquires
> the wait_lock to remove itself from the list, will allow an unlock
> thread to set the lock as available in the fastpath which could in
> turn allow a third thread the steal the lock. With high contention,
> this window will be come obviously larger as we contend for the
> wait_lock.
>
> CPU-0                                  CPU-1            CPU-3
> __mutex_lock_common              mutex_optimistic_spin
>   (->count now 0)
>             __mutex_fastpath_unlock
>             (->count now 1)                 __mutex_fastpath_lock
>                                       (stolen)
>
> spin_lock_mutex(&lock->wait_lock, flags);
>
> But we've always been bad when it comes to counter and waiters.
>
> Thanks,
> Davidlohr

I don't quite get how that can happen. CPU0 cannot change the count to 0 
unless CPU1, the lock holder, does the unlock first. Once CPU0 sees a 
count of 1 and change it to 0, it is the lock holder and there can be no 
other CPU that can do the unlock.

Cheers,
Longman

  parent reply	other threads:[~2016-02-03 22:07 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-21  9:29 [PATCH RFC] locking/mutexes: don't spin on owner when wait list is not NULL Ding Tianhong
2016-01-21 21:23 ` Tim Chen
2016-01-22  2:41   ` Paul E. McKenney
2016-01-22  2:48     ` Davidlohr Bueso
2016-01-22  3:13       ` Paul E. McKenney
2016-01-21 23:02 ` Waiman Long
2016-01-22  6:09   ` Davidlohr Bueso
2016-01-22 13:38     ` Waiman Long
2016-01-22 16:46       ` Davidlohr Bueso
2016-01-25  2:23         ` [PATCH] locking/mutex: Allow next waiter lockless wakeup Davidlohr Bueso
2016-01-25 23:02           ` Waiman Long
2016-02-29 11:21           ` [tip:locking/core] " tip-bot for Davidlohr Bueso
2016-01-22  8:54   ` [PATCH RFC] locking/mutexes: don't spin on owner when wait list is not NULL Peter Zijlstra
2016-01-22 10:20     ` Jason Low
2016-01-22 10:53       ` Peter Zijlstra
2016-01-22 10:56         ` Peter Zijlstra
2016-01-22 11:06           ` Peter Zijlstra
2016-01-22 13:59             ` Waiman Long
2016-01-24  8:03               ` Ding Tianhong
2016-01-29  9:53                 ` Peter Zijlstra
2016-01-30  1:18                   ` Ding Tianhong
2016-02-01  3:29                     ` huang ying
2016-02-01  3:35                       ` Huang, Ying
2016-02-01 10:08                     ` [PATCH] locking/mutex: Avoid spinner vs waiter starvation Peter Zijlstra
2016-02-02 21:19                       ` Davidlohr Bueso
2016-02-03  7:10                         ` Ding Tianhong
2016-02-03 19:24                           ` Davidlohr Bueso
2016-02-04  1:20                             ` Ding Tianhong
2016-02-12 18:33                               ` Waiman Long
2016-02-03 22:07                         ` Waiman Long [this message]
2016-02-04  1:35                       ` Jason Low
2016-02-04  8:55                         ` huang ying
2016-02-04 22:49                           ` Jason Low
2016-01-22 13:41     ` [PATCH RFC] locking/mutexes: don't spin on owner when wait list is not NULL Waiman Long

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=56B27A37.6070105@hpe.com \
    --to=waiman.long@hpe.com \
    --cc=Waiman.Long@hp.com \
    --cc=Will.Deacon@arm.com \
    --cc=dave@stgolabs.net \
    --cc=dingtianhong@huawei.com \
    --cc=jason.low2@hp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=paulmck@us.ibm.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=tim.c.chen@linux.intel.com \
    --cc=torvalds@linux-foundation.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 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.