From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Greear Subject: ixgbe issues in 2.6.29.3 Date: Tue, 19 May 2009 08:51:37 -0700 Message-ID: <4A12D589.7040002@candelatech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit To: NetDev Return-path: Received: from mail.candelatech.com ([208.74.158.172]:54037 "EHLO ns3.lanforge.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752108AbZESPvf (ORCPT ); Tue, 19 May 2009 11:51:35 -0400 Received: from [192.168.100.195] (firewall.candelatech.com [70.89.124.249]) (authenticated bits=0) by ns3.lanforge.com (8.14.2/8.14.2) with ESMTP id n4JFpaPm014870 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Tue, 19 May 2009 08:51:37 -0700 Sender: netdev-owner@vger.kernel.org List-ID: I'm using 2.6.29.3 (plus my hacks, but no changes to ixgbe for what that's worth). I just got some new hardware: Dual Intel E5530 processors, 12GB RAM, etc and put in a dual-port 10G NIC with Intel chipset (though made by Silicom). I tried changing MTU from 1500 to 9000 and started pktgen, and I noticed that a process that does ethtool ioctls was using 99% of a CPU and strace showed nothing (and would not ctrl-C to exit). I have modified pktgen, btw. Some other applications hung as well, and I found this in the logs. There is a lot of new stuff in my setup, and this could be symptom instead of cause, but thought I'd post the trace in case it helps. ixgbe: eth1: ixgbe_change_mtu: changing MTU from 1500 to 9000 ------------[ cut here ]------------ WARNING: at /home/greearb/git/linux-2.6.dev.28/net/sched/sch_generic.c:226 dev_watchdog+0xe7/0x148() Hardware name: X8DTU NETDEV WATCHDOG: eth1 (ixgbe): transmit timed out Modules linked in: nf_defrag_ipv4 nfs lockd nfs_acl auth_rpcgss 8021q garp redirdev arc4 michael_mic macvlan wanlink(P) pktgen stp bnep sco l2cap bluetooth sunrpc ipv6 cpufreq_ondemand acpi_cpufreq freq_table loop dm_multipath scsi_dh uinput pcspkr sg i2c_i801 i2c_core iTCO_wdt iTCO_vendor_support e1000e ixgbe igb dca inet_lro button sr_mod cdrom joydev dm_snapshot dm_zero dm_mirror dm_region_hash dm_log dm_mod ata_piix ata_generic pata_acpi libata sd_mod scsi_mod ext3 jbd mbcache uhci_hcd ohci_hcd ssb ehci_hcd [last unloaded: x_tables] Pid: 0, comm: swapper Tainted: P W 2.6.29.3 #4 Call Trace: [] warn_slowpath+0xd3/0xf2 [] ? __enqueue_entity+0x74/0x76 [] ? enqueue_entity+0x1a0/0x1a9 [] ? enqueue_task_fair+0x3e/0x43 [] ? _spin_unlock_irqrestore+0x40/0x44 [] ? task_rq_unlock+0xc/0xe [] ? try_to_wake_up+0x2a7/0x2b9 [] ? autoremove_wake_function+0x11/0x38 [] ? __cycles_2_ns+0x23/0x42 [] ? _spin_lock+0x1e/0x22 [] ? _spin_unlock+0x2d/0x38 [] ? __netif_tx_unlock+0x14/0x16 [] ? netif_tx_lock+0x4d/0x66 [] ? dev_watchdog+0x0/0x148 [] dev_watchdog+0xe7/0x148 [] ? hrtimer_hres_active+0x1a/0x30 [] run_timer_softirq+0x126/0x19c [] __do_softirq+0x66/0xe8 [] call_softirq+0x1c/0x34 [] do_softirq+0x34/0x78 [] irq_exit+0x36/0x8a [] smp_apic_timer_interrupt+0x76/0x83 [] apic_timer_interrupt+0x13/0x20 [] ? acpi_idle_enter_bm+0x236/0x26d [] ? acpi_idle_enter_bm+0x22c/0x26d [] ? cpuidle_idle_call+0x80/0xb8 [] ? cpu_idle+0x4f/0x92 [] ? start_secondary+0x181/0x186 ---[ end trace 8256d1ae7f76313d ]--- Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com