Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754097AbYC3QSX (ORCPT ); Sun, 30 Mar 2008 12:18:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753007AbYC3QSP (ORCPT ); Sun, 30 Mar 2008 12:18:15 -0400 Received: from mail.gmx.net ([213.165.64.20]:59905 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1752971AbYC3QSO (ORCPT ); Sun, 30 Mar 2008 12:18:14 -0400 X-Authenticated: #14349625 X-Provags-ID: V01U2FsdGVkX185xA0ZPiw5fLxDWzISy53zh43hKSa33xweuJcX// JDvzHGMpkBcbLH Subject: Re: Scalability requirements for sysv ipc (+namespaces broken with SEM_UNDO) From: Mike Galbraith To: Manfred Spraul Cc: Linux Kernel Mailing List , paulmck@linux.vnet.ibm.com, Nadia Derbey , Andrew Morton , Peter Zijlstra , Pavel Emelianov In-Reply-To: <47EF9FC2.3050901@colorfullife.com> References: <47E382DB.70503@colorfullife.com> <47E3ADE3.4030304@bull.net> <47E3B924.3000304@colorfullife.com> <20080321141339.GB9618@linux.vnet.ibm.com> <47E3DD8C.4000904@colorfullife.com> <1206164608.3659.11.camel@marge.simson.net> <47E4DB06.3020003@colorfullife.com> <1206186832.4545.1.camel@marge.simson.net> <47E51637.8060102@colorfullife.com> <47E55945.10109@colorfullife.com> <1206460206.4414.21.camel@marge.simson.net> <47EF9FC2.3050901@colorfullife.com> Content-Type: text/plain Date: Sun, 30 Mar 2008 19:18:08 +0200 Message-Id: <1206897488.4593.11.camel@marge.simson.net> Mime-Version: 1.0 X-Mailer: Evolution 2.12.0 Content-Transfer-Encoding: 7bit X-Y-GMX-Trusted: 0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1242 Lines: 36 On Sun, 2008-03-30 at 16:12 +0200, Manfred Spraul wrote: > Mike Galbraith wrote: > > On Sat, 2008-03-22 at 20:08 +0100, Manfred Spraul wrote: > > > > > >> just the normal performance of 2.6.25-rc3 is abyssimal, 55 to 60% slower > >> than 2.6.18.8: > >> > > > > After manually reverting 3e148c79938aa39035669c1cfa3ff60722134535, > > 2.6.25.git scaled linearly > We can't just revert that patch: with IDR, a global lock is mandatory :-( > We must either revert the whole idea of using IDR or live with the > reduced scalability. Yeah, I looked at the problem, but didn't know what the heck to do about it, so just grabbed my axe to verify/quantify. > Actually, there are further bugs: the undo structures are not > namespace-aware, thus semop with SEM_UNDO, unshare, create new array > with same id, but more semaphores, another semop with SEM_UNDO will > corrupt kernel memory :-( > I'll try to clean up the bugs first, then I'll look at the scalability > again. Great! -Mike -- 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/