2013-01-29 21:10:05

by Richard Hartmann

[permalink] [raw]
Subject: drivers/net/wireless/iwlwifi/dvm/tx.c and arch/x86/kernel/apic/ipi.c: Kernel trace when roaming between different Access Points with the same ESSID

And another one; same circumstances. But this time, I got two stack
traces right behind each other.

As I got two possibly related traces, I am sending to the maintainers
of arch/x86/kernel/apic/ipi.c as well.


[ 5777.806736] ------------[ cut here ]------------
[ 5777.806745] WARNING: at
/build/buildd-linux_3.7.1-1~experimental.1-i386-Au6THQ/linux-3.7.1/arch/x86/kernel/apic/ipi.c:109
default_send_IPI_mask_logical+0x23/0x7a()
[ 5777.806746] Hardware name: 4286CTO
[ 5777.806747] empty IPI mask
[ 5777.806748] Modules linked in: parport_pc ppdev lp parport rfcomm
bnep cpufreq_userspace cpufreq_stats cpufreq_powersave
cpufreq_conservative uinput nfsd auth_rpcgss nfs_acl nfs lockd
dns_resolver fscache sunrpc ext4 jbd2 mbcache loop fuse
snd_hda_codec_hdmi coretemp snd_hda_codec_conexant arc4 iwldvm
mac80211 kvm_intel kvm iwlwifi crc32c_intel snd_hda_intel aesni_intel
snd_hda_codec snd_hwdep snd_pcm snd_page_alloc aes_i586 btusb
thinkpad_acpi snd_seq joydev cfg80211 xts nvram bluetooth
snd_seq_device snd_timer i915 rfkill snd lrw drm_kms_helper drm
i2c_algo_bit ac crc16 battery gf128mul iTCO_wdt iTCO_vendor_support
ablk_helper acpi_cpufreq mperf button video pcspkr mei soundcore
lpc_ich cryptd mfd_core i2c_i801 processor evdev psmouse i2c_core
serio_raw wmi tpm_tis tpm tpm_bios xfs sg sd_mod crc_t10dif mmc_block
xhci_hcd microcode ahci libahci sdhci_pci libata sdhci mmc_core
scsi_mod ehci_hcd usbcore usb_common e1000e thermal thermal_sys
[ 5777.806798] Pid: 4314, comm: Chrome_CacheThr Not tainted
3.7-trunk-686-pae #1 Debian 3.7.1-1~experimental.1
[ 5777.806799] Call Trace:
[ 5777.806802] [<c10310c5>] ? warn_slowpath_common+0x68/0x79
[ 5777.806805] [<c10238ce>] ? default_send_IPI_mask_logical+0x23/0x7a
[ 5777.806808] [<c103113e>] ? warn_slowpath_fmt+0x29/0x2d
[ 5777.806810] [<c10238ce>] ? default_send_IPI_mask_logical+0x23/0x7a
[ 5777.806812] [<c102263a>] ? native_send_call_func_ipi+0x42/0x45
[ 5777.806814] [<c106be06>] ? smp_call_function_many+0x178/0x1a8
[ 5777.806817] [<c102d1a5>] ? do_flush_tlb_all+0x2e/0x2e
[ 5777.806819] [<c102d22a>] ? native_flush_tlb_others+0x22/0x26
[ 5777.806821] [<c102d11f>] ? flush_tlb_others+0x9/0xf
[ 5777.806823] [<c102d43e>] ? flush_tlb_page+0x5b/0x60
[ 5777.806825] [<c102c54c>] ? ptep_set_access_flags+0x44/0x4b
[ 5777.806828] [<c10bb4dc>] ? do_wp_page+0x2ff/0x69d
[ 5777.806831] [<c10e0d65>] ? pipe_read+0x331/0x33f
[ 5777.806832] [<c10bd17c>] ? handle_pte_fault+0x88a/0x903
[ 5777.806834] [<c102936e>] ? arch_flush_lazy_mmu_mode+0x5/0x13
[ 5777.806836] [<c10bd45f>] ? handle_mm_fault+0x1e5/0x1fb
[ 5777.806840] [<c12eb330>] ? __do_page_fault+0x350/0x38a
[ 5777.806842] [<c10dac18>] ? do_sync_write+0x89/0x89
[ 5777.806844] [<c100fc2e>] ? read_tsc+0xa/0x28
[ 5777.806847] [<c1061b8a>] ? timekeeping_get_ns.constprop.10+0x10/0x56
[ 5777.806849] [<c1050bc1>] ? should_resched+0x5/0x1e
[ 5777.806851] [<c12e80e6>] ? _cond_resched+0x5/0x18
[ 5777.806854] [<c117bb1a>] ? copy_to_user+0x2b/0x48
[ 5777.806856] [<c106c032>] ? generic_smp_call_function_interrupt+0xed/0x106
[ 5777.806858] [<c12eb36a>] ? __do_page_fault+0x38a/0x38a
[ 5777.806860] [<c12e90af>] ? error_code+0x67/0x6c
[ 5777.806863] [<c12e0000>] ? set_cpu_sibling_map+0x22c/0x2e5
[ 5777.806867] ---[ end trace ab21f7f1719b3f0e ]---
[ 5824.597874] iwlwifi 0000:03:00.0: fail to flush all tx fifo queues
[ 5826.603337] iwlwifi 0000:03:00.0: fail to flush all tx fifo queues
[ 5826.621539] ------------[ cut here ]------------
[ 5826.621574] WARNING: at
/build/buildd-linux_3.7.1-1~experimental.1-i386-Au6THQ/linux-3.7.1/drivers/net/wireless/iwlwifi/dvm/tx.c:1189
iwlagn_rx_reply_tx+0x288/0x6e2 [iwldvm]()
[ 5826.621578] Hardware name: 4286CTO
[ 5826.621581] Modules linked in: parport_pc ppdev lp parport rfcomm
bnep cpufreq_userspace cpufreq_stats cpufreq_powersave
cpufreq_conservative uinput nfsd auth_rpcgss nfs_acl nfs lockd
dns_resolver fscache sunrpc ext4 jbd2 mbcache loop fuse
snd_hda_codec_hdmi coretemp snd_hda_codec_conexant arc4 iwldvm
mac80211 kvm_intel kvm iwlwifi crc32c_intel snd_hda_intel aesni_intel
snd_hda_codec snd_hwdep snd_pcm snd_page_alloc aes_i586 btusb
thinkpad_acpi snd_seq joydev cfg80211 xts nvram bluetooth
snd_seq_device snd_timer i915 rfkill snd lrw drm_kms_helper drm
i2c_algo_bit ac crc16 battery gf128mul iTCO_wdt iTCO_vendor_support
ablk_helper acpi_cpufreq mperf button video pcspkr mei soundcore
lpc_ich cryptd mfd_core i2c_i801 processor evdev psmouse i2c_core
serio_raw wmi tpm_tis tpm tpm_bios xfs sg sd_mod crc_t10dif mmc_block
xhci_hcd microcode ahci libahci sdhci_pci libata sdhci mmc_core
scsi_mod ehci_hcd usbcore usb_common e1000e thermal thermal_sys
[ 5826.621732] Pid: 0, comm: swapper/0 Tainted: G W
3.7-trunk-686-pae #1 Debian 3.7.1-1~experimental.1
[ 5826.621737] Call Trace:
[ 5826.621750] [<c10310c5>] ? warn_slowpath_common+0x68/0x79
[ 5826.621768] [<f86e81e2>] ? iwlagn_rx_reply_tx+0x288/0x6e2 [iwldvm]
[ 5826.621776] [<c10310e3>] ? warn_slowpath_null+0xd/0x10
[ 5826.621790] [<f86e81e2>] ? iwlagn_rx_reply_tx+0x288/0x6e2 [iwldvm]
[ 5826.621808] [<f86ed344>] ? iwl_rx_dispatch+0x57/0x63 [iwldvm]
[ 5826.621822] [<f864ad14>] ? iwl_irq_tasklet+0x39c/0x698 [iwlwifi]
[ 5826.621833] [<c12e0b0d>] ? setup_APIC_timer+0x33/0x5e
[ 5826.621841] [<c1037457>] ? tasklet_action+0x62/0xa5
[ 5826.621847] [<c103711b>] ? __do_softirq+0x90/0x192
[ 5826.621854] [<c103708b>] ? perf_trace_irq_handler_entry+0xd7/0xd7
[ 5826.621857] <IRQ> [<c10372cb>] ? irq_exit+0x32/0x7d
[ 5826.621872] [<c100d070>] ? do_IRQ+0x65/0x76
[ 5826.621880] [<c12ed2f3>] ? common_interrupt+0x33/0x38
[ 5826.621888] [<c10300d8>] ? copy_process.part.33+0x8a8/0xf0c
[ 5826.621895] [<c11b1c03>] ? intel_idle+0xc4/0xec
[ 5826.621905] [<c122ea2e>] ? cpuidle_enter+0xb/0xe
[ 5826.621912] [<c122edb9>] ? cpuidle_enter_state+0x9/0x33
[ 5826.621920] [<c122ee8e>] ? cpuidle_idle_call+0xab/0x146
[ 5826.621927] [<c10114d7>] ? cpu_idle+0x88/0xac
[ 5826.621935] [<c1464966>] ? start_kernel+0x334/0x339
[ 5826.621939] ---[ end trace ab21f7f1719b3f0f ]---
[ 5826.631527] cfg80211: Calling CRDA to update world regulatory domain
[ 5826.641195] cfg80211: World regulatory domain updated:
[ 5826.641204] cfg80211: (start_freq - end_freq @ bandwidth),
(max_antenna_gain, max_eirp)
[ 5826.641210] cfg80211: (2402000 KHz - 2472000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[ 5826.641215] cfg80211: (2457000 KHz - 2482000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[ 5826.641220] cfg80211: (2474000 KHz - 2494000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[ 5826.641224] cfg80211: (5170000 KHz - 5250000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[ 5826.641229] cfg80211: (5735000 KHz - 5835000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[ 5828.636680] iwlwifi 0000:03:00.0: fail to flush all tx fifo queues


2013-02-09 19:57:57

by Richard Hartmann

[permalink] [raw]
Subject: Re: drivers/net/wireless/iwlwifi/dvm/tx.c and arch/x86/kernel/apic/ipi.c: Kernel trace when roaming between different Access Points with the same ESSID

Yet another one, this time during high USB i/o load.

Could this be a hardware issue as well?

Does it still make sense to send these reports to anyone?


[95553.449219] ------------[ cut here ]------------
[95553.449257] WARNING: at
/build/buildd-linux_3.7.1-1~experimental.1-i386-Au6THQ/linux-3.7.1/drivers/net/wireless/iwlwifi/dvm/tx.c:1189
iwlagn_rx_reply_tx+0x288/0x6e2 [iwldvm]()
[95553.449262] Hardware name: 4286CTO
[95553.449265] Modules linked in: nls_utf8 nls_cp437 vfat fat
mmc_block usb_storage parport_pc ppdev rfcomm lp parport bnep
cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_conservative
uinput nfsd auth_rpcgss nfs_acl nfs lockd dns_resolver fscache sunrpc
ext4 jbd2 mbcache loop fuse snd_hda_codec_hdmi snd_hda_codec_conexant
arc4 coretemp iwldvm mac80211 kvm_intel snd_hda_intel thinkpad_acpi
snd_hda_codec nvram snd_hwdep snd_pcm snd_page_alloc kvm i915
drm_kms_helper crc32c_intel joydev drm i2c_algo_bit snd_seq btusb
bluetooth iTCO_wdt snd_seq_device snd_timer aesni_intel aes_i586 crc16
iwlwifi iTCO_vendor_support xts lrw i2c_i801 pcspkr snd gf128mul
i2c_core ablk_helper cryptd soundcore lpc_ich mfd_core cfg80211
psmouse serio_raw acpi_cpufreq mperf mei rfkill ac wmi battery tpm_tis
tpm evdev processor tpm_bios button video xfs sg sd_mod crc_t10dif
xhci_hcd ahci libahci libata microcode scsi_mod ehci_hcd sdhci_pci
sdhci mmc_core usbcore usb_common e1000e thermal thermal_sys
[95553.449427] Pid: 0, comm: swapper/0 Tainted: G W
3.7-trunk-686-pae #1 Debian 3.7.1-1~experimental.1
[95553.449431] Call Trace:
[95553.449446] [<c10310c5>] ? warn_slowpath_common+0x68/0x79
[95553.449464] [<f87291e2>] ? iwlagn_rx_reply_tx+0x288/0x6e2 [iwldvm]
[95553.449472] [<c10310e3>] ? warn_slowpath_null+0xd/0x10
[95553.449487] [<f87291e2>] ? iwlagn_rx_reply_tx+0x288/0x6e2 [iwldvm]
[95553.449502] [<f859008b>] ? check_event_type_and_length+0x37/0x4b [snd_seq]
[95553.449520] [<f872e344>] ? iwl_rx_dispatch+0x57/0x63 [iwldvm]
[95553.449537] [<f859fd14>] ? iwl_irq_tasklet+0x39c/0x698 [iwlwifi]
[95553.449546] [<c1037457>] ? tasklet_action+0x62/0xa5
[95553.449552] [<c103711b>] ? __do_softirq+0x90/0x192
[95553.449559] [<c103708b>] ? perf_trace_irq_handler_entry+0xd7/0xd7
[95553.449563] <IRQ> [<c10372cb>] ? irq_exit+0x32/0x7d
[95553.449577] [<c100d070>] ? do_IRQ+0x65/0x76
[95553.449586] [<c12ed2f3>] ? common_interrupt+0x33/0x38
[95553.449593] [<c10300d8>] ? copy_process.part.33+0x8a8/0xf0c
[95553.449602] [<c11b1c03>] ? intel_idle+0xc4/0xec
[95553.449612] [<c122ea2e>] ? cpuidle_enter+0xb/0xe
[95553.449619] [<c122edb9>] ? cpuidle_enter_state+0x9/0x33
[95553.449626] [<c122ee8e>] ? cpuidle_idle_call+0xab/0x146
[95553.449634] [<c10114d7>] ? cpu_idle+0x88/0xac
[95553.449641] [<c1464966>] ? start_kernel+0x334/0x339
[95553.449646] ---[ end trace e2ce8294be2c8dca ]---

--
Richard

2013-02-25 16:42:24

by Drew Raines

[permalink] [raw]
Subject: Re: drivers/net/wireless/iwlwifi/dvm/tx.c and arch/x86/kernel/apic/ipi.c: Kernel trace when roaming between different Access Points with the same ESSID

Drew Raines wrote:

> Richard Hartmann wrote:
>
>> Yet another one, this time during high USB i/o load.
>
> I've been suffering this issue as well since some version of the
> experimental Wheezy packages of 3.7.x, but it's also happening with a
> make-kpkg build of 3.8.
>
> If I switch APs in my house, my connection immediately fails and it
> won't work until I power the radio off and back on with the hardware
> switch on the side of the laptop. I can't even `modprobe -r iwlwifi'
> the driver ("FATAL: Module iwlwifi is in use.").

As I was messing around with downgrading my kernel package (where'd
3.5-trunk go??) it occurred to me that I hadn't yet tried a more
recent version of firmware-iwlwifi. I had been on 0.36+wheezy.1.
After installing 0.37 from experimental, this issue seems to have
disappeared. I've been roaming around the house watching my wlan0
reassociate several times with a `ping -i3' running and it hasn't
lost its connection.

Richard, if you haven't updated, you might want to try:

sudo apt-get -t experimental install firmware-iwlwifi

Thanks to the Intel guys for keeping this excellent hardware going on
Linux!

-Drew


2013-02-25 16:34:46

by Drew Raines

[permalink] [raw]
Subject: Re: drivers/net/wireless/iwlwifi/dvm/tx.c and arch/x86/kernel/apic/ipi.c: Kernel trace when roaming between different Access Points with the same ESSID

Richard Hartmann wrote:

> Yet another one, this time during high USB i/o load.

I've been suffering this issue as well since some version of the
experimental Wheezy packages of 3.7.x, but it's also happening with a
make-kpkg build of 3.8.

If I switch APs in my house, my connection immediately fails and it
won't work until I power the radio off and back on with the hardware
switch on the side of the laptop. I can't even `modprobe -r iwlwifi'
the driver ("FATAL: Module iwlwifi is in use.").

I do not have this problem when waking from sleep.

Thanks for any help!

-Drew

Details:

Debian Wheezy
Linux beta 3.8.0 #1 SMP Thu Feb 21 12:30:32 CST 2013 x86_64 GNU/Linux
Intel(R) Centrino(R) Ultimate-N 6300 AGN, REV=0x74
Thinkpad X230

lspci and kernel logs attached.


Attachments:
lspci.txt (6.20 kB)
lspci output
kern.log.txt (53.21 kB)
Debian kern.log since waking from sleep and changing access points
Download all attachments

2013-02-28 16:10:39

by Drew Raines

[permalink] [raw]
Subject: Re: drivers/net/wireless/iwlwifi/dvm/tx.c and arch/x86/kernel/apic/ipi.c: Kernel trace when roaming between different Access Points with the same ESSID

Drew Raines wrote:

> As I was messing around with downgrading my kernel package (where'd
> 3.5-trunk go??) it occurred to me that I hadn't yet tried a more
> recent version of firmware-iwlwifi. I had been on 0.36+wheezy.1.
> After installing 0.37 from experimental, this issue seems to have
> disappeared.

Apparently after a few hibernations and suspends the issue
resurfaces. I move access points and my connection stop functioning
and eventually get this log message:

Feb 28 09:51:26 beta kernel: [105937.778241] iwlwifi 0000:03:00.0: fail to flush all tx fifo queues

The only fix is a reset of the hardware radio switch on the side of
the laptop.

What can I do to diagnose further?

-Drew