From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755899Ab0AVQWq (ORCPT ); Fri, 22 Jan 2010 11:22:46 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755414Ab0AVQWp (ORCPT ); Fri, 22 Jan 2010 11:22:45 -0500 Received: from casper.infradead.org ([85.118.1.10]:36071 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755378Ab0AVQWn (ORCPT ); Fri, 22 Jan 2010 11:22:43 -0500 Subject: Re: [PATCH 08/10] perf_event: Optimize the constraint searching bits From: Peter Zijlstra To: Stephane Eranian Cc: fweisbec@gmail.com, paulus@samba.org, mingo@elte.hu, davem@davemloft.net, robert.richter@amd.com, linux-kernel@vger.kernel.org In-Reply-To: References: <20100122155044.391857484@chello.nl> <20100122155535.887853503@chello.nl> Content-Type: text/plain; charset="UTF-8" Date: Fri, 22 Jan 2010 17:22:33 +0100 Message-ID: <1264177353.4283.1541.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2010-01-22 at 17:08 +0100, Stephane Eranian wrote: > On Fri, Jan 22, 2010 at 4:50 PM, Peter Zijlstra wrote: > > Instead of copying bitmasks around, pass pointers to the constraint structure. > > > The reason I changed that is because of the AMD situation where constraints > are dynamically generated. So we need to get some storage from somwhere. > See my AMD NB patch, Can't we still do that by embedding 4 event_constraint structures in struct amd_nb and rewriting those as we need?