From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752730Ab3FZQ7p (ORCPT ); Wed, 26 Jun 2013 12:59:45 -0400 Received: from one.firstfloor.org ([193.170.194.197]:37787 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751475Ab3FZQ7o (ORCPT ); Wed, 26 Jun 2013 12:59:44 -0400 Date: Wed, 26 Jun 2013 18:59:42 +0200 From: Andi Kleen To: Peter Zijlstra Cc: "Yan, Zheng" , linux-kernel@vger.kernel.org, mingo@kernel.org, eranian@google.com, andi@firstfloor.org Subject: Re: [PATCH 0/7] perf, x86: Haswell LBR call stack support Message-ID: <20130626165942.GF6123@two.firstfloor.org> References: <1372150039-15151-1-git-send-email-zheng.z.yan@intel.com> <20130626120749.GH28407@twins.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130626120749.GH28407@twins.programming.kicks-ass.net> 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 > These patches are also done wrong; the first patches should add HSW > support for the existing features. Only then should you do patches for > new fail^wfeatures. The basic Haswell LBR support is already in tip. It's pretty much the same as Sandy Bridge, except the the sign extension need to extend two additional bits. The only other part not yet in are the exposed filters for INTX/ABORT, plus the workaround for the duplicated ABORT record. These are orthogonal to callstack LBR. -Andi