Red Hat Bugzilla – Bug 711489
atl1c: transmit queue timeout (Acer Aspire One 522)
Last modified: 2011-10-24 15:18:39 EDT
Created attachment 503519 [details]
Description of problem:
Linux seems to crash when I am using wifi on a ASUS 522. I does not happen instantly but after using wifi for a while the whole computer locks up.
Version-Release number of selected component (if applicable):
kernel.i686 22.214.171.124-30.fc15 @updates
Steps to Reproduce:
1. Enable wifi.
2. Use computer for a while.
Computer crashes, no response.
Computer should not crash!
I also get some strange warnings when booting, do you want to see them?
After some more testing I have found out that Linux crashes when there is no ethernet cable present, regardless if there is any wireless or not...
If you remove the ethernet cable, let the computer crash (sound becomes garbled and you can not interact in any way with the computer) and then plug in the ethernet cable again everything seems to work again, except for the connected USB mouse.
Created attachment 503527 [details]
dmesg after a took out and reinserted ethernet cable
Tried kernel-126.96.36.199-31.fc15.i686.rpm, same results but USB mouse worked after I reinserted ethernet cable.
Here is some possibly relevant info from dmesg (dmesg4.txt):
0.284648] pci_root PNP0A08:00: address space collision: host bridge window [mem 0x000cc000-0x000cffff] conflicts with Video ROM [mem 0x000c0000-0x000ce1ff]
After ethernet cable removal
[ 106.116023] atl1c 0000:06:00.0: atl1c: em1 NIC Link is Down
[ 138.275882] psmouse.c: TouchPad at isa0060/serio1/input0 lost synchronization, throwing 2 bytes away.
[ 138.281331] atl1c 0000:06:00.0: atl1c: em1 NIC Link is Up<100 Mbps Full Duplex>
[ 138.803406] psmouse.c: resync failed, issuing reconnect request
[ 143.712312] ------------[ cut here ]------------
[ 143.712335] WARNING: at net/sched/sch_generic.c:256 dev_watchdog+0xe2/0x147()
[ 143.712342] Hardware name: Aspire One 522
[ 143.712348] NETDEV WATCHDOG: em1 (atl1c): transmit queue 0 timed out
[ 143.712353] Modules linked in: fuse cpufreq_ondemand powernow_k8 mperf 8021q garp stp llc snd_hda_codec_conexant snd_hda_codec_hdmi snd_hda_intel snd_hda_codec arc4 snd_hwdep ath9k snd_seq snd_seq_device snd_pcm mac80211 snd_timer ath9k_common sp5100_tco snd soundcore ath9k_hw uvcvideo sparse_keymap i2c_piix4 microcode ath snd_page_alloc cfg80211 rfkill k10temp atl1c videodev wmi joydev ipv6 video radeon ttm drm_kms_helper drm i2c_algo_bit i2c_core [last unloaded: scsi_wait_scan]
[ 143.712442] Pid: 0, comm: swapper Not tainted 188.8.131.52-31.fc15.i686 #1
[ 143.712449] Call Trace:
[ 143.712462] [<c043aad9>] warn_slowpath_common+0x7c/0x91
[ 143.712472] [<c074f92a>] ? dev_watchdog+0xe2/0x147
[ 143.712482] [<c074f92a>] ? dev_watchdog+0xe2/0x147
[ 143.712491] [<c043ab79>] warn_slowpath_fmt+0x33/0x35
[ 143.712500] [<c074f92a>] dev_watchdog+0xe2/0x147
[ 143.712511] [<c07d6483>] ? _raw_spin_unlock_irqrestore+0x13/0x15
[ 143.712521] [<c044da73>] ? __queue_work+0x1fc/0x204
[ 143.712532] [<c0445525>] run_timer_softirq+0x152/0x207
[ 143.712542] [<c048a4d0>] ? __rcu_process_callbacks+0x64/0x1f5
[ 143.712552] [<c074f848>] ? dev_watchdog+0x0/0x147
[ 143.712562] [<c043fdc4>] __do_softirq+0xa9/0x163
[ 143.712571] [<c043fd1b>] ? __do_softirq+0x0/0x163
[ 143.712576] <IRQ> [<c043ff17>] ? irq_exit+0x3c/0x70
[ 143.712593] [<c041a7fd>] ? smp_apic_timer_interrupt+0x6b/0x78
[ 143.712603] [<c07d6d09>] ? apic_timer_interrupt+0x31/0x38
[ 143.712615] [<c06316ca>] ? arch_local_irq_enable+0x7/0xb
[ 143.712625] [<c0632309>] ? acpi_idle_enter_simple+0x105/0x13d
[ 143.712635] [<c0712fb4>] ? cpuidle_idle_call+0xc5/0x136
[ 143.712645] [<c0401fa0>] ? cpu_idle+0x8e/0xa8
[ 143.712655] [<c07ba9d5>] ? rest_init+0x5d/0x5f
[ 143.712667] [<c0a59816>] ? start_kernel+0x357/0x35d
[ 143.712676] [<c0a5920c>] ? unknown_bootoption+0x0/0x19e
[ 143.712686] [<c0a590d6>] ? i386_start_kernel+0xd6/0xdc
[ 143.712693] ---[ end trace 44c3b2c404a5df17 ]---
[ 143.749261] atl1c 0000:06:00.0: irq 42 for MSI/MSI-X
[ 143.749885] atl1c 0000:06:00.0: atl1c: em1 NIC Link is Up<100 Mbps Full Duplex>
Created attachment 503732 [details]
Tried with latest firmware, same results.
Hrm. I see this bug reported against Ubuntu and in the kernel bugzilla, with no useful responses.
Could you try booting with "pci=nomsi" to rule out an interrupt related problem?
Also, could you please attach the output of "lspci -vvnn" as root?
*** Bug 699963 has been marked as a duplicate of this bug. ***
*** Bug 709560 has been marked as a duplicate of this bug. ***
Created attachment 504524 [details]
I am using latest from git (3.0.0-rc2-helber.x86_64-00256-gb99ca60-dirty):
[ 311.712032] ------------[ cut here ]------------
[ 311.712050] WARNING: at net/sched/sch_generic.c:256 dev_watchdog+0x164/0x205()
[ 311.712058] Hardware name: Aspire 4540
[ 311.712065] NETDEV WATCHDOG: p5p1 (atl1c): transmit queue 0 timed out
[ 311.712071] Modules linked in: vboxnetadp vboxnetflt sunrpc vboxdrv powernow_k8 mperf thermal atl1c wmi processor virtio_net virtio virtio_ring [last unloaded: scsi_wait_scan]
[ 311.712109] Pid: 5134, comm: mpegaudioparse0 Not tainted 3.0.0-rc2-helber.x86_64-00256-gb99ca60-dirty #111
[ 311.712116] Call Trace:
[ 311.712121] <IRQ> [<ffffffff8107d2f5>] warn_slowpath_common+0x83/0x9b
[ 311.712146] [<ffffffff8107d3b0>] warn_slowpath_fmt+0x46/0x48
[ 311.712155] [<ffffffff81509c4b>] dev_watchdog+0x164/0x205
[ 311.712164] [<ffffffff8108ac78>] ? mod_timer+0x212/0x274
[ 311.712174] [<ffffffff81089ee5>] run_timer_softirq+0x209/0x354
[ 311.712183] [<ffffffff81509ae7>] ? qdisc_reset+0x41/0x41
[ 311.712192] [<ffffffff8104ee05>] ? lapic_next_event+0x1d/0x21
[ 311.712202] [<ffffffff81083452>] __do_softirq+0xd7/0x1ef
[ 311.712212] [<ffffffff810a68af>] ? tick_program_event+0x1f/0x21
[ 311.712223] [<ffffffff8162d41c>] call_softirq+0x1c/0x30
[ 311.712232] [<ffffffff8103ca98>] do_softirq+0x46/0x86
[ 311.712241] [<ffffffff810837a8>] irq_exit+0x44/0x96
[ 311.712251] [<ffffffff8162dd4d>] smp_apic_timer_interrupt+0x8a/0x98
[ 311.712260] [<ffffffff8162cbd3>] apic_timer_interrupt+0x13/0x20
[ 311.712265] <EOI>
[ 311.712272] ---[ end trace 44d4634399b23574 ]---
[ 311.725659] atl1c 0000:03:00.0: Unable to allocate MSI interrupt Error: -22
[ 311.725708] atl1c 0000:03:00.0: atl1c: p5p1 NIC Link is Up<100 Mbps Full Duplex>
Created attachment 504550 [details]
boot cmdline: ro root=/dev/sda3 rd_NO_LUKS rd_NO_LVM rd_NO_MD rd_NO_DM LANG=pt_BR.UTF-8 SYSFONT=latarcyrheb-sun16 KEYTABLE=us-acentos quiet pci=nomsi
OS Release: Fedora release 15 (Lovelock)
I simply tried to configure Samba and started downloading http://www.samba.org/samba/docs/Samba-Guide.pdf Downloading freezed and pages stopped downloading. After that error appeared.
Still happens on latest F15 Linux. Is anything going on upstream?
Still happens in latest Linux, but seems to take slightly longer to trigger.
Set as blocker.
Is there any workaround this bug, perhaps by disabling the ethernet chip?
Can you confirm that this affects Fedora 16? "latest Linux" is ambiguous.
We usually don't consider very hardware-specific bugs to be blockers; is there any indication this bug is affecting a particularly wide range of users?
Fedora Bugzappers volunteer triage team
If this only affects a small subset of hardware, I'm -1 blocker, -1 NTH on this.
The system is Acer, not ASUS. =) Helber's system also appears to be an Acer. Nick, what's your hardware?
Discussed in the 2011-07-29 blocker review meeting. Rejected as F16 alpha blocker due to small scope of affected hardware.
If the impact of this bug turns out to be greater after investigation, please re-propose as either blocker or NTH.
Tried 2.6.40-3.fc15, no change. Will try F16 version as soon as liveUSB is available.
For all affected, here is a proper workaround:
in /etc/modprobe.d/blacklist.conf add the follwing line
(This will disable the ethernet chip.)
Tried with 20110906.17 live-cd, same problems.
Proposed nice ot have for F16.
Discussed at 2011-09-30 NTH review meeting, accepted as NTH due to the severity of the impact of the issue: it makes it very difficult to install / use F16 on an affected system.
*** Bug 746125 has been marked as a duplicate of this bug. ***
*** This bug has been marked as a duplicate of bug 717211 ***