From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sabrina Dubroca Subject: Re: 3.19-rc7 netconsole locking bug. Date: Thu, 5 Feb 2015 00:41:32 +0100 Message-ID: <20150204234132.GA13621@kria> References: <20150204193902.GB652@codemonkey.org.uk> <20150204195901.GA1682@codemonkey.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Cc: netdev@vger.kernel.org To: Dave Jones Return-path: Received: from smtp4-g21.free.fr ([212.27.42.4]:37902 "EHLO smtp4-g21.free.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756083AbbBDXld (ORCPT ); Wed, 4 Feb 2015 18:41:33 -0500 Content-Disposition: inline In-Reply-To: <20150204195901.GA1682@codemonkey.org.uk> Sender: netdev-owner@vger.kernel.org List-ID: 2015-02-04, 14:59:01 -0500, Dave Jones wrote: > On Wed, Feb 04, 2015 at 02:39:02PM -0500, Dave Jones wrote: > > I set up netconsole this morning, and noticed when I rebooted the machine > > that this happens when it initializes.. > > ... > > > > Despite the positive sounding last two lines, nothing actually gets logged > > over the netconsole. > > After changing the boot process to modprobe netconsole later, I still > see the same spew, but the console now works. > However, shortly afterwards, there's another locking warning.. > > BUG: sleeping function called from invalid context at kernel/irq/manage.c:104 > in_atomic(): 1, irqs_disabled(): 1, pid: 0, name: swapper/1 > 4 locks held by swapper/1/0: > #0: (rcu_read_lock){......}, at: [] __netif_receive_skb_core+0x78/0xc00 > #1: (rcu_read_lock_bh){......}, at: [] ip_finish_output2+0x9f/0xaa0 > #2: (rcu_read_lock_bh){......}, at: [] __dev_queue_xmit+0x63/0xbe0 > #3: (_xmit_ETHER#2){+.-...}, at: [] sch_direct_xmit+0x11f/0x330 > irq event stamp: 136019 > hardirqs last enabled at (136018): [] _raw_spin_unlock_irqrestore+0x5a/0xb0 > hardirqs last disabled at (136019): [] irq_work_interrupt+0x6a/0x80 > softirqs last enabled at (135536): [] _local_bh_enable+0x25/0x80 > softirqs last disabled at (135537): [] irq_exit+0x27d/0x2e0 > CPU: 1 PID: 0 Comm: swapper/1 Not tainted 3.19.0-rc7+ #13 > ffffffffa0d3d363 2097f77ab05a8c13 ffff88007ec03160 ffffffffa09a5efa > 0000000000000000 0000000000000000 ffff88007ec03190 ffffffffa00ca277 > ffff88007ec03190 ffffffffa0d3d363 0000000000000068 0000000000000000 > Call Trace: > [] dump_stack+0x84/0xb9 > [] ___might_sleep+0x2d7/0x390 > [] __might_sleep+0x80/0x100 > [] synchronize_irq+0x67/0x120 > [] ? __disable_irq_nosync+0x69/0xc0 > [] disable_irq+0x34/0x50 > [] e1000_netpoll+0x55/0x1d0 > [] netpoll_poll_dev+0x99/0x300 > [] netpoll_send_skb_on_dev+0x1df/0x470 > [] netpoll_send_udp+0x30c/0x5a0 This looks like the issue I started working on: https://lkml.org/lkml/2014/10/29/523 https://marc.info/?l=linux-netdev&m=141813589016500&w=2 I will continue. Sorry I let this slide. -- Sabrina