From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752089AbdJXNqc (ORCPT ); Tue, 24 Oct 2017 09:46:32 -0400 Received: from mail-wm0-f65.google.com ([74.125.82.65]:56537 "EHLO mail-wm0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932895AbdJXM7s (ORCPT ); Tue, 24 Oct 2017 08:59:48 -0400 X-Google-Smtp-Source: ABhQp+RTUnFDkmEjAVWsz0hgA6xDAMfplRt6d8eBLv/+kpgVxXHpMJr0kggKmpPfSwuroTILcjZGQg== Date: Tue, 24 Oct 2017 14:59:44 +0200 From: Ingo Molnar To: Jiri Olsa Cc: "Liang, Kan" , "acme@kernel.org" , "mingo@redhat.com" , "linux-kernel@vger.kernel.org" , "peterz@infradead.org" , "jolsa@kernel.org" , "wangnan0@huawei.com" , "hekuang@huawei.com" , "namhyung@kernel.org" , "alexander.shishkin@linux.intel.com" , "Hunter, Adrian" , "ak@linux.intel.com" Subject: Re: [PATCH V3 0/6] event synthesization multithreading for perf record Message-ID: <20171024125944.uswroptykcqrgjox@gmail.com> References: <1508529934-369393-1-git-send-email-kan.liang@intel.com> <20171023114822.ijbixdkhysinlwqv@gmail.com> <37D7C6CF3E00A74B8858931C1DB2F077537D874E@SHSMSX103.ccr.corp.intel.com> <20171024092200.wef6b66ecmhrvaja@gmail.com> <20171024114755.GA2716@krava> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20171024114755.GA2716@krava> User-Agent: NeoMutt/20170609 (1.8.3) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Jiri Olsa wrote: > I recently made some changes on threaded record, which are based > on Namhyungs time* API, which is needed to read/sort the data afterwards > > but I wasn't able to get any substantial and constant reduce of LOST events > and then I got sidetracked and did not finish, but it's in here: So, in the context of system-wide profiling, the way that would work best I think is the following: thread #0 binds itself to CPU#0 (via sched_setaffinity) and creates a per-CPU event on CPU#0 thread #1 binds itself to CPU#1 (via sched_setaffinity) and creates a per-CPU event on CPU#1 thread #2 binds itself to CPU#2 (via sched_setaffinity) and creates a per-CPU event on CPU#2 etc. Is this how you implemented it? If the threads in the thread pool are just free-running then the scheduler might not migrate it to the 'right' CPU that is streaming the perf events and there will be a lot of cross-talking between CPUs. Inherited events (default 'perf record') is tougher. Thanks, Ingo