Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 26 Jul 2002 09:29:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 26 Jul 2002 09:29:23 -0400 Received: from pc2-cwma1-5-cust12.swa.cable.ntl.com ([80.5.121.12]:45045 "EHLO irongate.swansea.linux.org.uk") by vger.kernel.org with ESMTP id ; Fri, 26 Jul 2002 09:29:23 -0400 Subject: Re: [PATCH] 2.5-rmap: VM strict overcommit From: Alan Cox To: Pavel Machek Cc: Robert Love , akpm@zip.com.au, riel@conectiva.com.br, linux-kernel@vger.kernel.org, Linus Torvalds In-Reply-To: <20020726103104.GA279@elf.ucw.cz> References: <1026928763.1116.11.camel@sinai> <20020726103104.GA279@elf.ucw.cz> Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Ximian Evolution 1.0.3 (1.0.3-6) Date: 26 Jul 2002 15:46:43 +0100 Message-Id: <1027694803.13428.43.camel@irongate.swansea.linux.org.uk> Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 947 Lines: 23 On Fri, 2002-07-26 at 11:31, Pavel Machek wrote: > In what scenario can "strict overcommit" kill? When the kernel grabs over 50% of RAM. Remember that includes page tables. I've seen the kernel taking 35% of RAM. > > > +4 - (NEW) paranoid overcommit. The total address space commit > > + for the system is not permitted to exceed swap. The machine > > + will never kill a process accessing pages it has mapped > > + except due to a bug (ie report it!). > > ...and why is that scenario impossible on "paranoid overcommit"? You are guaranteed user pages are either discardable or all fit in swap. If you run out of memory the kernel itself runs out of allocation space for any kernel owned page. - 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/