From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: Question to perf spending a large amount of time monitoring a java process Date: Tue, 5 Dec 2017 18:52:58 +0100 Message-ID: <20171205175258.GN3165@worktop.lehotels.local> References: <20d3fa12-295c-84d3-2eae-aa32b5757d19@huawei.com> <87bmjdjcpa.fsf@linux.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from merlin.infradead.org ([205.233.59.134]:39020 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752673AbdLERxH (ORCPT ); Tue, 5 Dec 2017 12:53:07 -0500 Content-Disposition: inline In-Reply-To: <87bmjdjcpa.fsf@linux.intel.com> Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: Andi Kleen Cc: zhangmengting , linux-perf-users@vger.kernel.org, acme@redhat.com, namhyung@kernel.org, jolsa@redhat.com, huawei.libin@huawei.com, cj.chengjian@huawei.com, alexey.budankov@intel.com On Tue, Dec 05, 2017 at 09:40:01AM -0800, Andi Kleen wrote: > zhangmengting writes: > > > Hi all, > > > > I found that perf spends a large amount of time attaching and monitoring > > a java process with lock, although the execution time of the java process > > is below 1 minute. > > You could check if this patchkit helps > > https://lkml.org/lkml/2017/9/8/118 > > Not sure why it's not moved forward. Peter? Because its absolutely broken and I don't have had the time to work on it much. There's a pile of patches here: git://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git perf/core that fix the worst of the fallout.