Message-Id: <1226986433.32373.5.camel@stinkpad>
Date:	Mon, 17 Nov 2008 21:33:53 -0800
From:	"Jeffrey W. Baker" <jwbaker@...il.com>
To:	netdev@...r.kernel.org
Subject: e1000e mutex contention log spew on 2.6.27

I have a huge number of these log entries on a ThinkPad X61 (82566MM
rev 03).  My system is generating one every two seconds.  Here's the
first one, which contains a substantial amount of spurious memory
contents:

WARNING: at /build/buildd/linux-2.6.27/drivers/net/e1000e/ich8lan.c:403 e1000_acquire_swflag_ich8lan+0x47/0xd0 [e1000e]()
e1000e mutex contention. Owned by pid -1
Modules linked in: af_packet i915 drm rfcomm sco bridge stp bnep l2cap ppdev ipv6 container sbs sbshc wmi pci_slot acpi_cpufreq cpufreq_powersave cpufreq_stats cpufreq_conservative cpufreq_userspace cpufreq_ondemand freq_table iptable_filter ip_tables x_tables sbp2 parport_pc lp parport btusb bluetooth pcmcia arc4 ecb crypto_blkcipher thinkpad_acpi nvram evdev serio_raw psmouse pcspkr iwlagn iwlcore snd_hda_intel rfkill snd_pcm_oss snd_mixer_oss sdhci_pci battery sdhci led_class yenta_socket rsrc_nonstatic ac mac80211 snd_pcm mmc_core cfg80211 pcmcia_core video output snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device snd button intel_agp iTCO_wdt iTCO_vendor_support soundcore agpgart shpchp pci_hotplug snd_page_alloc ext3 jbd mbcache sd_mod crc_t10dif ata_generic sg pata_acpi ahci ata_piix ohci1394 libata ieee1394 scsi_mod ehci_hcd uhci_hcd usbcore dock e1000e thermal processor fan fbcon tileblit font bitblit soft
 ursor fuse
Pid: 5337, comm: NetworkManager Not tainted 2.6.27-7-generic #1
 [<c0131d65>] warn_slowpath+0x65/0x90
 [<c0123186>] ? set_next_entity+0x126/0x160
 [<c0102df6>] ? __switch_to+0xa6/0x160
 [<c01284cb>] ? finish_task_switch+0x2b/0xe0
 [<c037c999>] ? schedule+0x429/0x790
 [<c01376e6>] ? __do_softirq+0xf6/0x120
 [<c037e511>] ? _spin_lock_irqs22 4.084550] etlink_e+] ]88iespeodod:x0cl]l4>  [link_rc f01?.e ect _0r-lxe12/
 8551>] netl cb
 Rlfpintis>80 c[ s+0x0/.085920018[=/ eeksth.8d00t]6n0/0x]  [<c001at- fbi _ie6s .
 0<kgx30
 [ 28816881.09k19cceet o_[a7cex
 
 .2n]  [<c04b0>] ?n0 ]ncp_meeu50f0i0x 0+0>[ 8880]  [<cre-[eccodci6?0e2fe]d n0x21/0x30.<4>[ 28/xie nuanvu M]]yq08[_<<+f75  [<c02e7sf0neds__1]]?0_eh>/ />k_smove_wact<40pa ipir1f3c_ e/>x>ockca>] ? >h0]bqysi  a1.[r0i7e200? a02edc151o 0eep_o1 23_0<<+xats2c01474b0>]02e66f1>+[/tfs h_[8  r]10?>22[_fy_>[ 28818 (seid_<030</1801eo0]10/02]  [<c6=idovoohm.6c0e c0vid3 <x3a/0x130
 .102821nclcsde10 0x
 o3e38ddmcimer_sta2f
 80nti_af39e e0o+l2cf+00imer_s2f
 806eisac39e ee08.2n/e<rrtimer0x2f
 <ai b_mfs0e
 p8/_20[/. <[ 2881.184f7898 bepuri -b[l0t0p[ ac20t e0
 [ 28861db9484: ewcOc17? 24c]0 [fd[7[_2[8> 63--
 <4 lg_6 te0080 8/b>] 1000eopmm1711e f[.x] syse00e/_pcadl/1c]2 8 < cd in: sfs h[>e>>>6> d--1
 <ccih 08q_rp: c>3_ ek 51 _cpsmousq _08 20 0e0
 _tablcled_ctg1xc8h4a_[8_c]dreq_taci lee4[8s8ieddf8py8e psnd_sed f<_ x[
 r[ diro_rmidi si+k1[0]efk
 o0c snd_miinte 8?f0>0_ivideoic]8127d[5 0 _Ob4_l_agp cx20s1l147 =ertpic sg c]w0e00c<c0_._nC ilebll0n41
 _13<00erpage_allomm: N10c[+7 1 a +4epcnp uhcih0115<._35?>cel_aNetwo55.610_4l0m2itci ata_g5]  9[8[ .22nu2c/qhege56028f 8<8l
 .1candid
 253ea1 _9ee0c 0c8aaepi<c01310x40 >8
 ]c4/21.>aamc_] ?1 frfe]2_] ]>pq[   eeend? co-lai< 80[   [< ee>xlcsqr41x0ke<0 _] ? c]?-codaG7x4ecx0>_e] ? 8lccida 00/00 4tendmsg? cop stmbrc 9u7e_[5_   [<c04eveai  n8ec] t_s] ? c0<eedor[5398110eendmsg+0xef/0x120
 [<c0234775>] ? apparmor_socket_recvmsg+0x15/0x20
 [<c01474b0>] ? autoremove_wake_function+0x0/0x50
 [<c01474b0>] ? autoremove_wake_function+0x0/0x50
 [<c02548ca>] ? copy_from_user+0x3a/0x130
 [<c02edc15>] ? verify_iovec+0x35/0xb0
 [<c02e66f1>] sys_sendmsg+0x111/0x230
 [<c014b474>] ? hrtimer_start+0xc4/0x1c0
 [<c01284cb>] ? finish_task_switch+0x2b/0xe0
 [<c037c999>] ? schedule+0x429/0x790
 [<c02e6edb>] sys_socketcall+0xeb/0x2d0
 [<c0103f7b>] sysenter_do_call+0x12/0x2f
 =======================
---[ end trace 1aa61db9484f7898 ]---

That's pretty sexy.  I hope I didn't lose any hardware here.  Both
LEDs on the NIC are on steady.  Later log entries seem more sane I
guess:

WARNING: at /build/buildd/linux-2.6.27/drivers/net/e1000e/ich8lan.c:403 e1000_acquire_swflag_ich8lan+0x47/0xd0 [e1000e]()
e1000e mutex contention. Owned by pid -1
Modules linked in: af_packet i915 drm rfcomm sco bridge stp bnep l2cap ppdev ipv6 container sbs sbshc wmi pci_slot acpi_cpufreq cpufreq_powersave cpufreq_stats cpufreq_conservative cpufreq_userspace cpufreq_ondemand freq_table iptable_filter ip_tables x_tables sbp2 parport_pc lp parport btusb bluetooth pcmcia arc4 ecb crypto_blkcipher thinkpad_acpi nvram evdev serio_raw psmouse pcspkr iwlagn iwlcore snd_hda_intel rfkill snd_pcm_oss snd_mixer_oss sdhci_pci battery sdhci led_class yenta_socket rsrc_nonstatic ac mac80211 snd_pcm mmc_core cfg80211 pcmcia_core video output snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device snd button intel_agp iTCO_wdt iTCO_vendor_support soundcore agpgart shpchp pci_hotplug snd_page_alloc ext3 jbd mbcache sd_mod crc_t10dif ata_generic sg pata_acpi ahci ata_piix ohci1394 libata ieee1394 scsi_mod ehci_hcd uhci_hcd usbcore dock e1000e(-) thermal processor fan fbcon tileblit font bitblit s
 ftcursor fuse
Pid: 8697, comm: modprobe Tainted: G        W 2.6.27-7-generic #1
 [<c0131d65>] warn_slowpath+0x65/0x90
 [<c037c999>] ? schedule+0x429/0x790
 [<c01376e6>] ? __do_softirq+0xf6/0x120
 [<c0253dde>] ? delay_tsc+0xe/0x70
 [<f8856028>] ? fuse_dev_fasync+0x28/0x40 [fuse]
 [<c0253ea1>] ? __const_udelay+0x31/0x40
 [<f88c243c>] ? e1000_flash_cycle_ich8lan+0x4c/0x70 [e1000e]
 [<c037d3f9>] ? mutex_trylock+0x9/0x30
 [<f88c2c17>] e1000_acquire_swflag_ich8lan+0x47/0xd0 [e1000e]
 [<f88c7502>] e1000e_write_phy_reg_igp+0x22/0x70 [e1000e]
 [<f88c2932>] e1000_phy_hw_reset_ich8lan+0x182/0x1b0 [e1000e]
 [<f88d2c10>] e1000_remove+0xc8/0xca [e1000e]
 [<c02630be>] pci_device_remove+0x1e/0x40
 [<c02c3a19>] __device_release_driver+0x79/0xc0
 [<c02c3aff>] driver_detach+0x9f/0xb0
 [<c02c2d1b>] bus_remove_driver+0x7b/0xb0
 [<c02c4009>] driver_unregister+0x39/0x40
 [<c02632e9>] pci_unregister_driver+0x29/0x80
 [<c0144fb0>] ? wakeme_after_rcu+0x0/0x20
 [<f88d2b37>] e1000_exit_module+0x12/0x23 [e1000e]
 [<c015ad07>] sys_delete_module+0x167/0x230
 [<c0199d7a>] ? do_munmap+0x1da/0x230
 [<c014ba38>] ? up_write+0x8/0x20
 [<c0103f7b>] sysenter_do_call+0x12/0x2f
 [<c0370000>] ? default_device_exit+0x60/0xb0
 =======================
---[ end trace 1aa61db9484f7898 ]---

I removed e1000e and now the system is acting in the usual way again.

-jwb


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists