From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751925Ab0L3WwM (ORCPT ); Thu, 30 Dec 2010 17:52:12 -0500 Received: from ogre.sisk.pl ([217.79.144.158]:45396 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750967Ab0L3WwL (ORCPT ); Thu, 30 Dec 2010 17:52:11 -0500 From: "Rafael J. Wysocki" To: carlos palma Subject: Re: [Bug #25402] kernel (2.6.37-8-generic_amd64) panic on boot (with message "map_single: bounce buffer is not DMA'ble) - possible regression !!! Date: Thu, 30 Dec 2010 23:51:48 +0100 User-Agent: KMail/1.13.5 (Linux/2.6.37-rc8+; KDE/4.4.4; x86_64; ; ) Cc: Linux Kernel Mailing List , Kernel Testers List , Maciej Rutecki , Florian Mickler References: <96DQe4a_2tH.A.4RE.497GNB@chimera> <4D1CFA99.1070701@ono.com> In-Reply-To: <4D1CFA99.1070701@ono.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201012302351.48156.rjw@sisk.pl> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thursday, December 30, 2010, carlos palma wrote: > I can confirm that this "regression" still exists with the kernel > 2.6.37-rc8. > The Call Trace is basically the same: > > ... > Kernel panic - not syncing: map_single ... > Pid 935, comm: NetworkManager Tainted > call trace > panic+oxbf/0x200 > ? default_spin_lock_flags+0x9/0x10 > ?swiotlb_tlb_map_single+0x1ce/0x250 > ?map_single+0x46/0x60 > swiotlb_map_page+0xa6/0xf > ... and then the FREEZE ... only a hard reset brings to live (with > 2.6.36-2) Linux. Thanks for the info. Why did you put "regression" into double quotes? Rafael