Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1754777AbbHCTJ4 (ORCPT <rfc822;w@1wt.eu>);
	Mon, 3 Aug 2015 15:09:56 -0400
Received: from mx2.suse.de ([195.135.220.15]:48357 "EHLO mx2.suse.de"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1754650AbbHCTJz (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Mon, 3 Aug 2015 15:09:55 -0400
Message-ID: <1438628982.17146.20.camel@stgolabs.net>
Subject: Re: [PATCH v4 1/7] locking/pvqspinlock: Unconditional PV kick with
 _Q_SLOW_VAL
From: Davidlohr Bueso <dave@stgolabs.net>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Waiman Long <Waiman.Long@hp.com>, Ingo Molnar <mingo@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>, "H. Peter Anvin" <hpa@zytor.com>,
        x86@kernel.org, linux-kernel@vger.kernel.org,
        Scott J Norton <scott.norton@hp.com>,
        Douglas Hatch <doug.hatch@hp.com>
Date: Mon, 03 Aug 2015 12:09:42 -0700
In-Reply-To: <20150803183745.GY25159@twins.programming.kicks-ass.net>
References: <1438395724-25910-1-git-send-email-Waiman.Long@hp.com>
	 <1438395724-25910-2-git-send-email-Waiman.Long@hp.com>
	 <20150801222903.GC25159@twins.programming.kicks-ass.net>
	 <1438626129.17146.9.camel@stgolabs.net>
	 <20150803183745.GY25159@twins.programming.kicks-ass.net>
Content-Type: text/plain; charset="UTF-8"
X-Mailer: Evolution 3.12.11 
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 1115
Lines: 25

On Mon, 2015-08-03 at 20:37 +0200, Peter Zijlstra wrote:
> OK, so there's no 'fix'? The patch claims we can loose a wakeup and I
> just don't see how that is true.

Taking another look, I think you could hit something like this:

CPU0 (lock):					CPU1 (unlock):
  pv_wait_head					  __pv_queued_spin_unlock
						    <load ->state> [bogus ->state != halted]
    <spin>					    smp_store_release(&l->locked, 0);
						    
    WRITE_ONCE(pn->state, vcpu_halted);             
    pv_wait(&l->locked, _Q_SLOW_VAL);		    if (->state == vcpu_halted)
					              pv_kick(node->cpu); <-- missing wakeup, never called

So basically you can miss a wakeup if node->state load is done while the
locking thread is spinning and hasn't gotten a chance to update the
state to halted. That would also imply that it occurs right when the
threshold limit is about to be reached.

--
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/