From mboxrd@z Thu Jan 1 00:00:00 1970 From: M Kelly Subject: Re: =?utf-8?b?c3Bpbl9sb2Nr?= cause ? Date: Wed, 1 Jun 2016 19:25:20 +0000 (UTC) Message-ID: References: <574F15E4.5070008@lexisnexis.com> <4716588.WReR5L1Wkn@milian-kdab2> <20160601181153.GU2563@kernel.org> <574F2F20.2090007@hpe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-path: Received: from plane.gmane.org ([80.91.229.3]:49739 "EHLO plane.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750744AbcFATZc (ORCPT ); Wed, 1 Jun 2016 15:25:32 -0400 Received: from list by plane.gmane.org with local (Exim 4.69) (envelope-from ) id 1b8Blh-0003iG-JM for linux-perf-users@vger.kernel.org; Wed, 01 Jun 2016 21:25:30 +0200 Received: from 209.243.55.99 ([209.243.55.99]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 01 Jun 2016 21:25:29 +0200 Received: from mark.kelly by 209.243.55.99 with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 01 Jun 2016 21:25:29 +0200 Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: linux-perf-users@vger.kernel.org Rick Jones hpe.com> writes: > > On 06/01/2016 11:49 AM, M Kelly wrote: > > So that is helpful. > > I do not understand the system_call_fastpath->futex->_spin_lock part > > yet, > > but that is a separate issue! > > Just a guess, but if your application is using threads, contention on > the synchonrization primitives you use to keep things straight can > end-up punting to the kernel. > > rick jones > Hi, I think I recall you from hp, years ago when I was there :-) Thanks for the info. I suspect we are too heavy on pthread_mutex_lock()/unlock() which could probably produce these call graphs: --- _spin_lock | |--49.11%-- futex_wake | do_futex | sys_futex | system_call_fastpath | __lll_unlock_wake | |--44.31%-- futex_wait_setup | futex_wait | do_futex | sys_futex | system_call_fastpath | __lll_lock_wait thanks again, mark