Message-ID: <1435348637.4110.83.camel@edumazet-glaptop2.roam.corp.google.com>
Subject: Re: [GIT PULL] Locking changes for v4.2
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Waiman Long <waiman.long@hp.com>
Cc: Ingo Molnar <mingo@kernel.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        linux-kernel@vger.kernel.org, Peter Zijlstra <a.p.zijlstra@chello.nl>,
        Thomas Gleixner <tglx@linutronix.de>,
        "Paul E. McKenney" <paulmck@us.ibm.com>,
        Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 26 Jun 2015 21:57:17 +0200
In-Reply-To: <558D8B66.3060108@hp.com>
References: <20150622073015.GA1986@gmail.com>
	 <1435309910.4110.74.camel@edumazet-glaptop2.roam.corp.google.com>
	 <558D8B66.3060108@hp.com>
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 997
Lines: 32

On Fri, 2015-06-26 at 13:27 -0400, Waiman Long wrote:

> I am glad that the qspinlock patch helps. May I know what kind of 
> machine (cpu type, # of sockets, etc) that you are running the test on? 
> Usually the more sockets the machine has, the bigger the improvement you 
> will see.

This particular host had two Intel(R) Xeon(R) CPU E5-2696 v2 @ 2.50GHz

(12 cores per socket, total of 48 hyper threads)

Note that if pipebench uses 48 threads (instead of 16 as my previous
mail did), there is almost no degradation.

lpaa23:~# ./pipebench -n 16
threads:16 sleep:30 pinned:1
total operations: 20228142, ops/sec 674271
20228142

lpaa23:~# ./pipebench -n 48
threads:48 sleep:30 pinned:1
total operations: 19763616, ops/sec 658787
19763616


Thanks

--
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/