From owner-freebsd-current@FreeBSD.ORG Thu Sep 17 22:59:08 2009 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2611F106566B for ; Thu, 17 Sep 2009 22:59:08 +0000 (UTC) (envelope-from patfbsd@davenulle.org) Received: from smtp.lamaiziere.net (net.lamaiziere.net [91.121.44.19]) by mx1.freebsd.org (Postfix) with ESMTP id DF70B8FC14 for ; Thu, 17 Sep 2009 22:59:07 +0000 (UTC) Received: from baby-jane.lamaiziere.net (106.10.87-79.rev.gaoland.net [79.87.10.106]) by smtp.lamaiziere.net (Postfix) with ESMTPA id D543F633322 for ; Fri, 18 Sep 2009 00:59:06 +0200 (CEST) Received: from baby-jane.lamaiziere.net (localhost [127.0.0.1]) by baby-jane.lamaiziere.net (Postfix) with ESMTP id 2AA95C8F7 for ; Fri, 18 Sep 2009 00:59:44 +0200 (CEST) Date: Fri, 18 Sep 2009 00:59:42 +0200 From: Patrick Lamaiziere To: freebsd-current@freebsd.org Message-ID: <20090918005942.0a4d5ca8@baby-jane.lamaiziere.net> In-Reply-To: <20090917132127.2a0ee973@baby-jane.lamaiziere.net> References: <20090917132127.2a0ee973@baby-jane.lamaiziere.net> X-Mailer: Claws Mail 3.7.2 (GTK+ 2.16.6; i386-portbld-freebsd8.0) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Subject: Re: [ZFS] panic vm_fault : fault on no fault entry (when superpages are enabled) X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Sep 2009 22:59:08 -0000 Le Thu, 17 Sep 2009 13:21:27 +0200, Patrick Lamaiziere a =E9crit : Hello, > [8.0 BETA-4/i386] >=20 > I've got this panic when copying /usr on a new pool. > It's 100% reproductible here with=20 > tar cf - -C /usr . |=A0tar xpvf - -C /pool >=20 > The panic occurs always with free() in the ZFS code. ... > /boot/loader.conf > # ZFS > vm.kmem_size=3D"512M" > vm.kmem_size_max=3D"512M" > vfs.zfs.arc_max=3D"40M" > vfs.zfs.vdev.cache.size=3D"5M" >=20 > and options KVA_PAGES=3D512 in my kernel I think I've found the culprit: superpages. With vm.pmap.pg_ps_enabled=3D"1", my box will panic in the next five minutes when doing a simple tar on ZFS. Looks good without the superpages. Is it a known problem? Thanks, regards.