Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 11 Sep 2018 10:17:55 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Davidlohr Bueso <dave@stgolabs.net>
Cc:     Waiman Long <longman@redhat.com>, Ingo Molnar <mingo@redhat.com>,
        Will Deacon <will.deacon@arm.com>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] locking/rwsem: Make owner store task pointer of last
 owning reader
Message-ID: <20180911081755.GO24106@hirez.programming.kicks-ass.net>
References: <1536265114-10842-1-git-send-email-longman@redhat.com>
 <20180910093156.GS24082@hirez.programming.kicks-ass.net>
 <83b333ea-8ef2-84ea-6810-4ab1a67eef88@redhat.com>
 <20180910171550.GA3902@linux-r8p5>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180910171550.GA3902@linux-r8p5>
User-Agent: Mutt/1.10.0 (2018-05-17)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Sep 10, 2018 at 10:15:50AM -0700, Davidlohr Bueso wrote:
> On Mon, 10 Sep 2018, Waiman Long wrote:
> 
> > One major issue with a combined count/owner is that we may have to use
> > cmpxchg for reader lock which will certainly impact reader-heavy
> > workloads. I have also thought about ways to compress the task pointer
> > address so that it can use fewer bits and leave the rests for reader
> > count. It is probably doable on 64-bit systems, but likely not on 32-bit
> > system given that there are less bits to play around.
> 
> Yeah we've discussed this before. As a cleanup it would obviously be good,
> but I fear about raw performance loss when using cmpxchg instead of xadd.

Does it really matter though? Last time I looked at something similar
(refcount_t) the "LOCK INCL" vs "LOCK CMPXCHG" was something like 15 vs
23 cycles (and that was with the cmpxchg loop actually doing a lot
more).

Do we really care about the down_read() path _that_ much? I thought that
with the main pain point, pagefaults, the problem was mostly the line
bouncing, not a few extra cycles.