From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752681AbbJBNKd (ORCPT ); Fri, 2 Oct 2015 09:10:33 -0400 Received: from mx1.redhat.com ([209.132.183.28]:34732 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752374AbbJBNKb (ORCPT ); Fri, 2 Oct 2015 09:10:31 -0400 Date: Fri, 2 Oct 2015 09:10:27 -0400 From: Neil Horman To: Vinson Lee Cc: "David S. Miller" , Jiri Pirko , Pravin B Shelar , Nikolay Aleksandrov , Sony Chacko , Dept-HSGLinuxNICDev@qlogic.com, Netdev , LKML Subject: Re: netpoll_send_skb_on_dev warning with bnx2 Message-ID: <20151002131027.GC10731@hmsreliant.think-freely.org> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Oct 01, 2015 at 08:25:46PM -0700, Vinson Lee wrote: > Hi. > > I am seeing a netpoll_send_skb_on_dev warning with bnx2. It happens on > Linux 4.1 and I am able to reproduce the warning with Linux 4.3-rc3. > > ------------[ cut here ]------------ > WARNING: CPU: 11 PID: 3110 at net/core/netpoll.c:368 > netpoll_send_skb_on_dev+0x183/0x201() > netpoll_send_skb_on_dev(): eth0 enabled interrupts in poll > (bnx2_start_xmit+0x0/0x5d4 [bnx2]) > Modules linked in: netconsole(+) configfs ipv6 ppdev parport_pc lp > parport tcp_diag inet_diag ipmi_devintf serio_raw iTCO_wdt > iTCO_vendor_support ipmi_si ipmi_msghandler hpilo hpwdt bnx2 coretemp > kvm_intel kvm crc32c_intel ghash_clmulni_intel microcode lpc_ich > mfd_core i7core_edac edac_core pcc_cpufreq acpi_cpufreq shpchp > sch_fq_codel hpsa radeon ttm drm_kms_helper drm fb_sys_fops sysimgblt > sysfillrect syscopyarea i2c_algo_bit i2c_core > CPU: 11 PID: 3110 Comm: modprobe Not tainted 4.3.0-rc3 #1 > 0000000000000170 ffff88060be17a18 ffffffff812bc84c ffffffff818b1a78 > ffff88060be17a68 ffff88060be17a58 ffffffff81064712 000000000be17a68 > ffffffff814ec12e ffff88060bc10900 0000000000000000 ffff8800d941c000 > Call Trace: > [] dump_stack+0x48/0x5c > [] warn_slowpath_common+0xa1/0xbb > [] ? netpoll_send_skb_on_dev+0x183/0x201 > [] warn_slowpath_fmt+0x46/0x48 > [] ? bnx2_run_loopback+0x391/0x391 [bnx2] > [] netpoll_send_skb_on_dev+0x183/0x201 > [] netpoll_send_udp+0x3df/0x3f1 > [] write_msg+0xaf/0xe9 [netconsole] > [] call_console_drivers.clone.2+0xd1/0xe9 > [] console_unlock+0x30e/0x3a5 > [] register_console+0x2af/0x322 > [] init_netconsole+0x1b4/0x224 [netconsole] > [] ? 0xffffffffa0497000 > [] do_one_initcall+0xf7/0x182 > [] ? kmem_cache_alloc_trace+0xb6/0xf0 > [] ? do_init_module+0x31/0x1e4 > [] do_init_module+0x69/0x1e4 > [] load_module+0x1451/0x160b > [] ? mod_kobject_put+0x4d/0x4d > [] ? __vmalloc_node+0x3e/0x40 > [] SyS_init_module+0x14f/0x155 > [] entry_SYSCALL_64_fastpath+0x12/0x6a > ---[ end trace 878215466e581776 ]--- > > Cheers, > Vinson Hmm, that would suggest that someone called local_irq_enable in either the xmit, poll_controller or napi poll paths (or one of its always-disable, bretheren), but I dont see where that might be happening. Are you able to instrument the kernel (either by cloning that WARN_ON_ONCE call or via stap script), so we can further isolate which call path the problem is happening in? Neil