Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S268947AbUINFHG (ORCPT ); Tue, 14 Sep 2004 01:07:06 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S268972AbUINFHF (ORCPT ); Tue, 14 Sep 2004 01:07:05 -0400 Received: from adsl-63-197-226-105.dsl.snfc21.pacbell.net ([63.197.226.105]:30634 "EHLO cheetah.davemloft.net") by vger.kernel.org with ESMTP id S268947AbUINFHC (ORCPT ); Tue, 14 Sep 2004 01:07:02 -0400 Date: Mon, 13 Sep 2004 22:05:07 -0700 From: "David S. Miller" To: William Lee Irwin III Cc: akpm@osdl.org, raybry@sgi.com, jbarnes@engr.sgi.com, linux-kernel@vger.kernel.org Subject: Re: [profile] amortize atomic hit count increments Message-Id: <20040913220507.1a269816.davem@davemloft.net> In-Reply-To: <20040914044748.GZ9106@holomorphy.com> References: <20040913015003.5406abae.akpm@osdl.org> <20040914044748.GZ9106@holomorphy.com> X-Mailer: Sylpheed version 0.9.12 (GTK+ 1.2.10; sparc-unknown-linux-gnu) X-Face: "_;p5u5aPsO,_Vsx"^v-pEq09'CU4&Dc1$fQExov$62l60cgCc%FnIwD=.UF^a>?5'9Kn[;433QFVV9M..2eN.@4ZWPGbdi<=?[:T>y?SD(R*-3It"Vj:)"dP Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 520 Lines: 14 William, any reason not to fully per-cpu the profile buffer and then only traverse the array when the user attempts to capture the counters? Then we can undo the atomics altogether, as well as the cacheline traffic, for the extremely common case. Are there space concerns? - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/