From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: cat /proc/net/tcp takes 0.5 seconds on x86_64 Date: Thu, 28 Aug 2008 00:54:08 +0200 Message-ID: <87k5e2qdbj.fsf@basil.nowhere.org> References: <200808261549.m7QFnVUN032543@bz-web1.app.phx.redhat.com> <20080826163719.GA25066@redhat.com> <87zlmyr5nz.fsf@basil.nowhere.org> <20080827214700.GB26610@one.firstfloor.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Dave Jones , netdev@vger.kernel.org, j.w.r.degoede@hhs.nl To: Trent Piepho Return-path: Received: from one.firstfloor.org ([213.235.205.2]:52171 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751747AbYH0WyP (ORCPT ); Wed, 27 Aug 2008 18:54:15 -0400 In-Reply-To: <20080827214700.GB26610@one.firstfloor.org> (Andi Kleen's message of "Wed, 27 Aug 2008 23:47:00 +0200") Sender: netdev-owner@vger.kernel.org List-ID: >> >> Umm, 0.4 to 0.8 seconds is on average _slower_ than 0.44 seconds. Did you >> mean .04 to .08? > > Yes noticed that too -- i think it must have been .04-.08 and the zeros > got lost somewhere. I remember that it was significantly faster. It's really > a benchmark how fast an read lock aquire/release is, but even on > the best systems it's quite slow compared to lockless code. FWIW I redid the benchmarks now and it's 0.046s (without) to 0.005s (with patch) on a different 2GB Core2 system with nearly empty hash table. -Andi -- ak@linux.intel.com