Skip to contentSkip to conversation

Kernelpanic on eeePC and MSI Wind (both using rt2800pci) with Linux 3.x

DISCUSSION Michael Basse 3 participants 8 messages Nov 4, 2011 – Nov 23, 2011

lore
Share & tools
lore
#0 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi everyone,
first of all, i hope that this procedure is the correct one for reporting issues with the kernel.

Because of 404 on
https://www.kernel.org/pub/linux/docs/lkml/reporting-bugs.html please forgive me when not everything is as you would suggest.

Me (and many other) ubuntu-users are facing issues with different Version of Kernel 3.x. As it seems all users are using "rt2800pci" for there wifi.

Model: "RaLink RT2860"
Vendor: pci 0x1814 "RaLink"
Device: pci 0x0781 "RT2860"
SubVendor: pci 0x1814 "RaLink"
SubDevice: pci 0x2790
Driver: "rt2800pci"
Driver Modules: "rt2800pci"

As it seems, the panics are happening more often, when on battery.

Also, this bug never happened on 2.6.38-x, Just with 3.x

My initial bug-report can be found here

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/869502

Because i am also facing the kernel-panic with 3.1.0-0301rc9-generic (which is not a specific ubuntu-kernel) i think its an upstream-bug

Here is a summary

3.0.0-11.18 = no panic after (3-4 days of testing)

3.0.0-12.19 = panic (3-5 days of testing)
https://launchpadlibrarian.net/82161301/IMG_20111006_222737.jpg (no rt2800 in the text)
https://launchpadlibrarian.net/82196440/IMG_20111007_103327.jpg (no rt2800 in the text)

3.0.0-12.20 = panic (10-10 days of testing)
https://launchpadlibrarian.net/82558459/IMG_20111011_211226.jpg
(rt2800 in the text)
https://launchpadlibrarian.net/82620705/IMG_20111012_163536.jpg (no rt2800 in the text)

3.0.6-030006-generic = panic (14 days of testing)
https://launchpadlibrarian.net/84309173/IMG_20111103_032854.jpg

3.1.0-0301rc9-generic = panic (14 days of testing)
https://launchpadlibrarian.net/83329981/IMG_20111020_231054.jpg (no rt2800 in the text)

for the last panic i a some dmesg and syslog-infos

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/869502/comments/67

Additional Screenshots from others users can be found in the bug
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/869502

Please let me know if you need further informations from me.

Greetings from germany

Michael Basse
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJOtABuAAoJEMIajL8dFT4GXCAIALGXZ/K8EwTxHeYdrZ0m5ORR 2P3MThbmJF/LrxN/qPFYWTIrck9qLm62pf7o5MeyY5gmIz7ToS+8caj2gcwBvy62 m5X6+jjWz+h3IldkzwxgbWPiu0GO5DXF8lJVfcx3ATnjpRUwe7SNkC2Sex4sqiXY 1thTER7ckeAt1PnImEKHPTwYW0Sc97FoyTYTUfwc4Pky7Y8dZSsCk4uB1cdxyH1t MfRT+mgNbUHuv0+aXwdGY0EiIttfyLq0qq2LmGVpqVcWcPJNuiCdaafIPdVCrqPE uBmtKDnnCxy+jNRjnPVk/FRknZVG0uNTDQ2YulbaOVhcHrGYUlQVJlmUSx5qk0g= =SIF1
-----END PGP SIGNATURE-----

#1 lore raw

Hi everyone,

here is a complete netconsole-output from a kernel-panic (provided by Hauke Jung) regarding to the topic
http://article.gmane.org/gmane.linux.kernel/1211290

[ 3888.374899] BUG: scheduling while atomic: kworker/u:21/5593/0x10010200
[ 3888.377441] Modules linked in: netconsole rt2800pci rt2800lib

crc_ccitt rt2x00pci rt2x00lib mac80211 cfg80211 eeprom_93cx6 r8169 bnep rfcomm bluetooth pci_stub vboxpci vboxnetadp vboxnetflt vboxdrv deflate zlib_deflate ctr twofish_generic parport_pc twofish_i586 twofish_common ppdev camellia serpent blowfish cast5 des_generic xcbc rmd160 sha512_generic crypto_null af_key snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device psmouse serio_raw binfmt_misc snd i915 soundcore drm_kms_helper drm snd_page_alloc i2c_algo_bit wmi video configfs arc4 lp parport ums_realtek usb_storage uas [last unloaded: netconsole]

[ 3888.396165] Pid: 5593, comm: kworker/u:21 Tainted: G        W
3.0.0-13-generic #21-Ubuntu
[ 3888.399461] Call Trace:
[ 3888.402723]  [<c151902f>] ? printk+0x2d/0x2f
[ 3888.406014]  [<c15188c8>] __schedule_bug+0x5e/0x64
[ 3888.409302]  [<c152ab94>] __schedule+0x614/0x620
[ 3888.412562]  [<c12c034b>] ? fb_set_var+0x17b/0x2d0
[ 3888.415801]  [<c12846a5>] ? put_dec+0x85/0x90
[ 3888.419043]  [<c12858fa>] ? number.isra.3+0x30a/0x320
[ 3888.422289]  [<c10412c5>] schedule+0x35/0x50
[ 3888.425528]  [<c152b996>] __mutex_lock_slowpath+0xc6/0x120
[ 3888.428810]  [<c152b644>] mutex_lock+0x24/0x40
[ 3888.432088]  [<f81f1659>] drm_fb_helper_pan_display+0x29/0xb0
[drm_kms_helper]
[ 3888.435437]  [<f81f1630>] ? drm_fb_helper_fill_var+0x1a0/0x1a0
[drm_kms_helper]
[ 3888.438823]  [<c12bfa09>] fb_pan_display+0xb9/0x160
[ 3888.442203]  [<c12ced11>] bit_update_start+0x21/0x50
[ 3888.445571]  [<c12cc1a4>] fbcon_switch+0x3c4/0x570
[ 3888.448930]  [<c1324eab>] redraw_screen+0x13b/0x240
[ 3888.452370]  [<c12cb64d>] fbcon_blank+0x23d/0x300
[ 3888.455690]  [<c1284b41>] ? format_decode+0x311/0x380
[ 3888.459024]  [<c1286a38>] ? vsnprintf+0x2c8/0x390
[ 3888.462332]  [<c1026038>] ? default_spin_lock_flags+0x8/0x10
[ 3888.465653]  [<c152ca3d>] ? _raw_spin_lock_irqsave+0x2d/0x40
[ 3888.468976]  [<c106b6ec>] ? down_trylock+0x2c/0x40
[ 3888.472289]  [<c1026038>] ? default_spin_lock_flags+0x8/0x10
[ 3888.475611]  [<c152ca3d>] ? _raw_spin_lock_irqsave+0x2d/0x40
[ 3888.478965]  [<c105578b>] ? lock_timer_base.isra.30+0x2b/0x50
[ 3888.482334]  [<c1056838>] ? mod_timer+0x118/0x280
[ 3888.485718]  [<c108c62d>] ? crash_kexec+0x7d/0xe0
[ 3888.489100]  [<c1325a1f>] do_unblank_screen.part.20+0x7f/0x130
[ 3888.492508]  [<c1325b10>] do_unblank_screen+0x40/0x70
[ 3888.495835]  [<c1325b4f>] unblank_screen+0xf/0x20
[ 3888.499051]  [<c12894e5>] bust_spinlocks+0x15/0x40
[ 3888.502127]  [<c152de04>] oops_end+0x34/0xd0
[ 3888.505060]  [<c1518142>] no_context+0x13c/0x144
[ 3888.507844]  [<c151826a>] __bad_area_nosemaphore+0x120/0x128
[ 3888.510541]  [<c106a11a>] ? __run_hrtimer+0x9a/0x1b0
[ 3888.513102]  [<c152fd40>] ? vmalloc_fault+0xee/0xee
[ 3888.515566]  [<c12cf557>] ? bit_putcs+0x237/0x410
[ 3888.517979]  [<c12d0999>] ? bitfill_aligned+0x29/0x30
[ 3888.520379]  [<c12d06dc>] ? cfb_fillrect+0x14c/0x2c0
[ 3888.522768]  [<c12cf557>] ? bit_putcs+0x237/0x410
[ 3888.525181]  [<c12d0999>] ? bitfill_aligned+0x29/0x30
[ 3888.527571]  [<c12d06dc>] ? cfb_fillrect+0x14c/0x2c0
[ 3888.529978]  [<c12ca8ee>] ? get_color.isra.17+0x2e/0x140
[ 3888.532394]  [<c12d0970>] ? bitfill_aligned.part.1+0x120/0x120
[ 3888.534811]  [<c12ceb28>] ? bit_clear+0xb8/0xe0
[ 3888.537223]  [<f81e2565>] ? rtl8169_start_xmit+0x175/0x3d0 [r8169]
[ 3888.539659]  [<c142e828>] ? __alloc_skb+0x58/0x1f0
[ 3888.542065]  [<c144d8c3>] ? netpoll_send_udp+0x1c3/0x1e0
[ 3888.544442]  [<f8075311>] ? write_msg+0xc1/0xd0 [netconsole]
[ 3888.546781]  [<f8075250>] ? netconsole_netdev_event+0x1c0/0x1c0
[netconsole]
[ 3888.549193]  [<c1047b0d>] ? __call_console_drivers+0x7d/0x90
[ 3888.551599]  [<c1026038>] ? default_spin_lock_flags+0x8/0x10
[ 3888.554027]  [<c152ca3d>] ? _raw_spin_lock_irqsave+0x2d/0x40
[ 3888.556459]  [<c1047fe2>] ? console_unlock+0xa2/0xf0
[ 3888.558871]  [<c10f6889>] ? __mod_zone_page_state+0x59/0x60
[ 3888.561321]  [<c10e4e11>] ? free_pcppages_bulk+0x2b1/0x340
[ 3888.563752]  [<c10e4ec7>] ? drain_pages+0x27/0x90
[ 3888.566181]  [<c101c2fc>] ? clear_local_APIC+0x13c/0x2c0
[ 3888.568609]  [<c101c4d2>] ? disable_local_APIC+0x52/0x90
[ 3888.571012]  [<c104e4d0>] ? local_bh_enable_ip+0x90/0x90
[ 3888.573418]  [<c100a121>] ? stop_this_cpu+0x41/0x50
[ 3888.575790]  [<c101b3e7>] ? smp_reboot_interrupt+0x27/0x30
[ 3888.578277]  [<c152ce79>] ? reboot_interrupt+0x31/0x38
[ 3888.580679]  [<c104e4d0>] ? local_bh_enable_ip+0x90/0x90
[ 3888.583060]  [<c104007b>] ? proc_sched_show_task+0x1a1b/0x1a70
[ 3888.585488]  [<c104e515>] ? __do_softirq+0x45/0x1a0
[ 3888.587895]  [<c104e4d0>] ? local_bh_enable_ip+0x90/0x90
[ 3888.590327]  <IRQ>  [<c104e896>] ? irq_exit+0x76/0xa0
[ 3888.592785]  [<c1041832>] ? scheduler_ipi+0x32/0x40
[ 3888.595230]  [<c101b417>] ? smp_reschedule_interrupt+0x27/0x30
[ 3888.597712]  [<c152cd9d>] ? reschedule_interrupt+0x31/0x38
[ 3888.600184]  [<c10300d8>] ? tg_schedulable+0x198/0x270
[ 3888.602653]  [<c1037d43>] ? finish_task_switch+0x43/0xc0
[ 3888.605144]  [<c152a871>] ? __schedule+0x2f1/0x620
[ 3888.607623]  [<c10af55d>] ? rcu_irq_exit+0xd/0x10
[ 3888.610115]  [<c104e85c>] ? irq_exit+0x3c/0xa0
[ 3888.612594]  [<c10412c5>] ? schedule+0x35/0x50
[ 3888.615044]  [<c152ccc2>] ? work_resched+0x5/0x20
[ 3888.617516]  [<c1520000>] ? ext4_check_descriptors+0x224/0x34e
[ 3888.622833] ------------[ cut here ]------------
[ 3888.624011] WARNING: at
/build/buildd/linux-3.0.0/arch/x86/kernel/apic/ipi.c:113
default_send_IPI_mask_logical+0xdc/0xf0()
[ 3888.624011] Hardware name: U-100
[ 3888.624011] Modules linked in: netconsole rt2800pci rt2800lib

crc_ccitt rt2x00pci rt2x00lib mac80211 cfg80211 eeprom_93cx6 r8169 bnep rfcomm bluetooth pci_stub vboxpci vboxnetadp vboxnetflt vboxdrv deflate zlib_deflate ctr twofish_generic parport_pc twofish_i586 twofish_common ppdev camellia serpent blowfish cast5 des_generic xcbc rmd160 sha512_generic crypto_null af_key snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device psmouse serio_raw binfmt_misc snd i915 soundcore drm_kms_helper drm snd_page_alloc i2c_algo_bit wmi video configfs arc4 lp parport ums_realtek usb_storage uas [last unloaded: netconsole]

[ 3888.624011] Pid: 559, comm: rsyslogd Tainted: G        W
3.0.0-13-generic #21-Ubuntu
[ 3888.624011] Call Trace:
[ 3888.624011]  [<c151902f>] ? printk+0x2d/0x2f
[ 3888.624011]  [<c1047982>] warn_slowpath_common+0x72/0xa0
[ 3888.624011]  [<c101d1dc>] ? default_send_IPI_mask_logical+0xdc/0xf0
[ 3888.624011]  [<c101d1dc>] ? default_send_IPI_mask_logical+0xdc/0xf0
[ 3888.624011]  [<c10479d2>] warn_slowpath_null+0x22/0x30
[ 3888.624011]  [<c101d1dc>] default_send_IPI_mask_logical+0xdc/0xf0
[ 3888.624011]  [<c10af55d>] ? rcu_irq_exit+0xd/0x10
[ 3888.624011]  [<c104e85c>] ? irq_exit+0x3c/0xa0
[ 3888.624011]  [<c102a97a>] flush_tlb_others_ipi+0xba/0xd0
[ 3888.624011]  [<c102aafd>] native_flush_tlb_others+0xd/0x10
[ 3888.624011]  [<c102ac3d>] flush_tlb_page+0x4d/0x90
[ 3888.624011]  [<c1029e3c>] ptep_clear_flush_young+0x2c/0x40
[ 3888.624011]  [<c1105aa0>] page_referenced_one+0x60/0x1e0
[ 3888.624011]  [<c1105cdf>] page_referenced_file+0xbf/0x110
[ 3888.624011]  [<c11072c5>] page_referenced+0x65/0xf0
[ 3888.624011]  [<c10ebee7>] page_check_references.isra.43+0x27/0x80
[ 3888.624011]  [<c10edff0>] shrink_page_list+0x1c0/0x4c0
[ 3888.624011]  [<c10ee631>] shrink_inactive_list+0x101/0x3e0
[ 3888.624011]  [<c10ecd61>] ? get_scan_count+0x271/0x380
[ 3888.624011]  [<c10ee9db>] shrink_list+0xcb/0x150
[ 3888.624011]  [<c10eeaed>] shrink_zone+0x8d/0x1a0
[ 3888.624011]  [<c10710bb>] ? ktime_get_ts+0xeb/0x120
[ 3888.624011]  [<c10ef375>] shrink_zones+0x55/0xe0
[ 3888.624011]  [<c10ef474>] do_try_to_free_pages+0x74/0x270
[ 3888.624011]  [<c10ef8cb>] try_to_free_pages+0x8b/0x120
[ 3888.624011]  [<c10e595f>] ? get_page_from_freelist+0x1df/0x2a0
[ 3888.624011]  [<c10e5f84>] __alloc_pages_nodemask+0x424/0x6e0
[ 3888.624011]  [<c10fb530>] do_anonymous_page+0x70/0x280
[ 3888.624011]  [<c10feec8>] handle_pte_fault+0x1f8/0x220
[ 3888.624011]  [<c10feff8>] handle_mm_fault+0x108/0x210
[ 3888.624011]  [<c152fd40>] ? vmalloc_fault+0xee/0xee
[ 3888.624011]  [<c152fe9b>] do_page_fault+0x15b/0x4a0
[ 3888.624011]  [<c1070bf4>] ? getnstimeofday+0x54/0x120
[ 3888.624011]  [<c12886c0>] ? copy_to_user+0x40/0x60
[ 3888.624011]  [<c104cfc2>] ? sys_gettimeofday+0x32/0x70
[ 3888.624011]  [<c152fd40>] ? vmalloc_fault+0xee/0xee
[ 3888.624011]  [<c152d30f>] error_code+0x67/0x6c
[ 3888.624011]  [<c1520000>] ? ext4_check_descriptors+0x224/0x34e
[ 3888.624011] ---[ end trace 4d06842269ea82cb ]---

Greetings

Michael Basse

#4 lore raw

From: Michael Basse <[email protected]>
Date: Tue, 22 Nov 2011 21:49:49 +0100

Quoted 7 lines

Hi,

just for the record. This thread is hopefully providing a patch

http://thread.gmane.org/gmane.linux.kernel.wireless.general/80759

already built the kernel and doing some testing now.

That patch is wrong, it returns IRQ_HANDLED when it can't tell if the chip generated the interrupt or not.

Therefore, if any other device shares that interrupt line it's interrupts can be lost when this case triggers.

This has to be fixed another way.

#5 lore raw
Quoted 11 lines · David Miller

From: Michael Basse<[email protected]>
Date: Tue, 22 Nov 2011 21:49:49 +0100

Quoted 7 lines

Hi,

just for the record. This thread is hopefully providing a patch

http://thread.gmane.org/gmane.linux.kernel.wireless.general/80759

already built the kernel and doing some testing now.

Quoted 7 lines

That patch is wrong, it returns IRQ_HANDLED when it can't tell if the chip generated the interrupt or not.

Therefore, if any other device shares that interrupt line it's interrupts can be lost when this case triggers.

This has to be fixed another way.

Huh? IRQ_HANDLED is the "safe" return value when you don't know for sure whether your device generated the interrupt. Returning that doesn't prevent other handlers on the same IRQ from being called.

#6 lore raw

From: Robert Hancock <[email protected]>
Date: Tue, 22 Nov 2011 19:49:33 -0600

Quoted 15 lines · David Miller
Quoted 14 lines

From: Michael Basse<[email protected]>
Date: Tue, 22 Nov 2011 21:49:49 +0100

Quoted 7 lines

Hi,

just for the record. This thread is hopefully providing a patch

http://thread.gmane.org/gmane.linux.kernel.wireless.general/80759

already built the kernel and doing some testing now.

That patch is wrong, it returns IRQ_HANDLED when it can't tell if the chip generated the interrupt or not.

Quoted 8 lines
Quoted 4 lines

Therefore, if any other device shares that interrupt line it's interrupts can be lost when this case triggers.

This has to be fixed another way.

Huh? IRQ_HANDLED is the "safe" return value when you don't know for sure whether your device generated the interrupt. Returning that doesn't prevent other handlers on the same IRQ from being called.

But it prevents being able to notice spurious interrupts from other devices sharing the IRQ.

You absolutely must return an accurate return value here for correct operation, you simply cannot "guess".

#7 lore raw
Quoted 27 lines · David Miller

From: Robert Hancock <[email protected]>
Date: Tue, 22 Nov 2011 19:49:33 -0600

Quoted 15 lines · David Miller
Quoted 14 lines

From: Michael Basse<[email protected]>
Date: Tue, 22 Nov 2011 21:49:49 +0100

Quoted 7 lines

Hi,

just for the record. This thread is hopefully providing a patch

http://thread.gmane.org/gmane.linux.kernel.wireless.general/80759

already built the kernel and doing some testing now.

That patch is wrong, it returns IRQ_HANDLED when it can't tell if the chip generated the interrupt or not.

Quoted 8 lines
Quoted 4 lines

Therefore, if any other device shares that interrupt line it's interrupts can be lost when this case triggers.

This has to be fixed another way.

Huh? IRQ_HANDLED is the "safe" return value when you don't know for sure whether your device generated the interrupt. Returning that doesn't prevent other handlers on the same IRQ from being called.

Quoted 5 lines

But it prevents being able to notice spurious interrupts from other devices sharing the IRQ.

You absolutely must return an accurate return value here for correct operation, you simply cannot "guess".

Certainly it's best to do so when possible, but with some hardware it simply isn't possible to do this. In some cases there's no register that reliably indicates whether or not the device raised an interrupt.