From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758592AbcCVKbd (ORCPT ); Tue, 22 Mar 2016 06:31:33 -0400 Received: from outbound-smtp12.blacknight.com ([46.22.139.17]:55726 "EHLO outbound-smtp12.blacknight.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752822AbcCVKbZ (ORCPT ); Tue, 22 Mar 2016 06:31:25 -0400 Date: Tue, 22 Mar 2016 10:31:21 +0000 From: Mel Gorman To: Sukadev Bhattiprolu Cc: Peter Zijlstra , LKML Subject: Translating perf events to raw event codes via sysfs Message-ID: <20160322103121.GH31585@techsingularity.net> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Sukadev, I'm curious about commit e0728b50d480 ("powerpc/perf: Export Power8 generic and cache events to sysfs"). This commit translates perf high-level event names into raw codes but only on one architecture for one CPU. While I can see how that is useful, I'm wondering why this was implemented as being specific to a single processor family. Should this not be done generically for all architectures? I accept that translating event names to raw codes for perf can be difficult but there are also tools like evt2raw whose output can then be used with perf. -- Mel Gorman SUSE Labs