From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.12] helo=sc8-sf-mx2.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1Bp6Ip-00020u-9E for user-mode-linux-devel@lists.sourceforge.net; Mon, 26 Jul 2004 07:18:07 -0700 Received: from smtp002.mail.ukl.yahoo.com ([217.12.11.33]) by sc8-sf-mx2.sourceforge.net with smtp (Exim 4.34) id 1Bp6Io-0005Up-Mb for user-mode-linux-devel@lists.sourceforge.net; Mon, 26 Jul 2004 07:18:07 -0700 From: BlaisorBlade Subject: Re: [uml-devel] I ported skas3 to fc2 kernel References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline Message-Id: <200407261620.23840.blaisorblade_spam@yahoo.it> Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Mon, 26 Jul 2004 16:20:23 +0200 Content-Transfer-Encoding: quoted-printable To: Jinnah Dylan Hosein , user-mode-linux-devel@lists.sourceforge.net Alle 01:13, venerd=EC 23 luglio 2004, Jinnah Dylan Hosein ha scritto: > I ported the host-skas3-2.6.7-v1.patch to the Fedora Core 2 kernel. > > I used the latest fc2 2.6.6-1.435.2.3 kernel rpm. > > I've tested it and I was able to run a 2.4.19 uml kernel with a redhat > 7.2 rootfs. I wasn't able to make it leak memory by running the "while > true" bash line, but I don't understand exactly what triggers the leak so > it's likely I didn't properly test. Well, it's a bit difficult to trigger; to the best of my understanding, it = can=20 be triggered if UML is a child or grandchild of a multi-threaded process=20 using the TLS, i.e. Thread-Local Storage (as KDE Konsole) running either wi= th=20 LinuxThreads support (i.e. either glibc is not NPTL-aware or is running ont= o=20 a 2.4 host kernel). More technically, if any process modifies its LDT (by using the modify_ldt(= )=20 syscall), any of its child processes will have the LDT set, and if UML has = the LDT set it will then leak memory. That said, I don't have time to look at it well but I don't see anything ba= d,=20 at a quick look; the port is not a simple rediff so there could be some=20 little changes needed, but if it runs Ok, I wouldn't worry. For the SKAS + UML patch: it's easy to rebuild it from the splitout version= =20 (you must browse the folders to find it, should be something like=20 patches/2.6.7-v1), but you can get by without it, so I'll build it only if = I=20 get free time. I've not even had the time to join the SKAS - leak fix with = the rest of the patch! --=20 Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_idG21&alloc_id=10040&op=CCk _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel