From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756144AbaCNVMd (ORCPT ); Fri, 14 Mar 2014 17:12:33 -0400 Received: from one.firstfloor.org ([193.170.194.197]:34491 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755180AbaCNVMc (ORCPT ); Fri, 14 Mar 2014 17:12:32 -0400 Date: Fri, 14 Mar 2014 22:12:30 +0100 From: Andi Kleen To: Jiri Olsa Cc: Andi Kleen , acme@infradead.org, mingo@kernel.org, linux-kernel@vger.kernel.org, peterz@infradead.org, eranian@google.com, namhyung@kernel.org, Andi Kleen Subject: Re: [PATCH 8/8] perf, tools, test: Add test case for alias and JSON parsing Message-ID: <20140314211230.GA22728@two.firstfloor.org> References: <1394048978-15909-1-git-send-email-andi@firstfloor.org> <1394048978-15909-9-git-send-email-andi@firstfloor.org> <20140311143149.GA31431@krava.redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140311143149.GA31431@krava.redhat.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > I think the problem is in parse_events_fixup, which assumes > the 'cpu' pmu globaly for all events I investigated this now. The problem is that the event is listed, but cannot be set up. I saw it also with 'data_writes' on another system. That's a generic problem, the test just exposes it. It's a side effect of the fragile kernel->user interface in perf and likely hard to fix. The only thing I can do is to ignore failures here and only print messages. -Andi