From owner-freebsd-net@FreeBSD.ORG Wed Dec 25 13:33:59 2013 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id BC3F04E5; Wed, 25 Dec 2013 13:33:59 +0000 (UTC) Received: from cell.glebius.int.ru (glebius.int.ru [81.19.69.10]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id 3F4E716A3; Wed, 25 Dec 2013 13:33:58 +0000 (UTC) Received: from cell.glebius.int.ru (localhost [127.0.0.1]) by cell.glebius.int.ru (8.14.7/8.14.7) with ESMTP id rBPDXvWE008291 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO); Wed, 25 Dec 2013 17:33:57 +0400 (MSK) (envelope-from glebius@FreeBSD.org) Received: (from glebius@localhost) by cell.glebius.int.ru (8.14.7/8.14.7/Submit) id rBPDXuhb008290; Wed, 25 Dec 2013 17:33:56 +0400 (MSK) (envelope-from glebius@FreeBSD.org) X-Authentication-Warning: cell.glebius.int.ru: glebius set sender to glebius@FreeBSD.org using -f Date: Wed, 25 Dec 2013 17:33:56 +0400 From: Gleb Smirnoff To: Mark Felder Subject: Re: 10.0-RC1: bad mbuf leak? Message-ID: <20131225133356.GL71033@FreeBSD.org> References: <1387204500.12061.60192349.19EAE1B4@webmail.messagingengine.com> <3A115E20-3ADB-49BA-885D-16189B97842B@FreeBSD.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3A115E20-3ADB-49BA-885D-16189B97842B@FreeBSD.org> User-Agent: Mutt/1.5.22 (2013-10-16) Cc: FreeBSD Net , FreeBSD Stable Mailing List X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 25 Dec 2013 13:33:59 -0000 Mark, On Tue, Dec 24, 2013 at 10:54:33AM -0600, Mark Felder wrote: M> > finally found some free time today to try to look into this. I was digging into the SVN changelogs of sys/dev/e1000 and couldn't see any obvious changes that I should revert. Instead I went a different route and jumped to HEAD/CURRENT. I'm not seeing the mbufs leaking yet. I'll need another 24 hours to confirm. Hopefully this is a worthwhile clue. I'm a bit surprised nobody else has reported this type of behavior... maybe 10 isn't getting the amount of testing we expect? ...or maybe it's just my lonely, haunted hardware :( M> M> Ok, I feel safe confirming that 10.0-RCs are not stable on my hardware. The mbuf problem went away completely when I jumped to head/current. M> M> Can someone please suggest what patch I can attempt to back out to fix this? I'd like to try to assist in fixing this before 10.0-RELEASE happens or we're going to have some very angry users. Is it possible for you to bisect head from the stable/10 branchpoint up to the current date and narrow down the revisions that introduced (and later fixed?) the leak? -- Totus tuus, Glebius.