BUG: scheduling while atomic: irq/51-iwlwifi/408/0x7ffffc00

[Cansheng Dong]

I was playing some audio files, the following message appears,
        ALSA lib pcm.c:8424:(snd_pcm_recover) underrun occurred
And the WIFI is disconnected.
I have experienced problems,
(1)The gnome-sound-recorder sometime works sometime doesn't.
(2) Click the LibreOffice Calc, after the empty worksheet loads, when i try to open another document by using "File->Open" on the menu, the application is not responding at all and I have to force it quit

ProblemType: KernelOops
DistroRelease: Ubuntu 18.10
Package: linux-image-4.18.0-8-generic 4.18.0-8.9
ProcVersionSignature: Ubuntu 4.18.0-8.9-generic 4.18.7
Uname: Linux 4.18.0-8-generic x86_64
Annotation: Your system might become unstable now and might need to be restarted.
ApportVersion: 2.20.10-0ubuntu11
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: dcansheng 2110 F.... pulseaudio
 /dev/snd/controlC1: dcansheng 2110 F.... pulseaudio
Date: Fri Oct 5 10:56:26 2018
Failure: oops
HibernationDevice: RESUME=UUID=2bbccbb3-89de-4207-9146-7a829c7e83a8
InstallationDate: Installed on 2018-08-29 (36 days ago)
InstallationMedia: Ubuntu 16.04 LTS "Xenial Xerus" - Release amd64 (20160420.1)
MachineType: LENOVO 20392
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.18.0-8-generic root=UUID=0bcc11bd-c39c-4a17-9eb3-67098d0bca7f ro quiet splash vt.handoff=1
PulseList: Error: command ['pacmd', 'list'] failed with exit code 1: No PulseAudio daemon running, or not running as session daemon.
RelatedPackageVersions: kerneloops-daemon N/A
SourcePackage: linux
Title: BUG: scheduling while atomic: irq/51-iwlwifi/408/0x7ffffc00
UpgradeStatus: Upgraded to cosmic on 2018-08-30 (35 days ago)
dmi.bios.date: 11/21/2014
dmi.bios.vendor: LENOVO
dmi.bios.version: 9DCN29WW(V2.09)
dmi.board.asset.tag: No Asset Tag
dmi.board.name: Lenovo B40-70
dmi.board.vendor: LENOVO
dmi.board.version: 31900058WIN
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: Lenovo B40-70
dmi.modalias: dmi:bvnLENOVO:bvr9DCN29WW(V2.09):bd11/21/2014:svnLENOVO:pn20392:pvrLenovoB40-70:rvnLENOVO:rnLenovoB40-70:rvr31900058WIN:cvnLENOVO:ct10:cvrLenovoB40-70:
dmi.product.family: IDEAPAD
dmi.product.name: 20392
dmi.product.sku: LENOVO_MT_20392_BU_idea_FM_Lenovo B40-70
dmi.product.version: Lenovo B40-70
dmi.sys.vendor: LENOVO

[i (i-linux-kernel-bugs)]

Created attachment 278231
relevant lines from journalctl -b

When connecting to an Eduroam hotspot with wrong credentials, iwlwifi prints out lots of (scary) error messages, followed by a hardware restart.

The distro is Arch Linux, and a relevant log file is attached.

[luca (luca-linux-kernel-bugs)]

Do you have any idea if this was happening before 4.18?

Can you try our core39 release, as explained in our wiki[1] and see if the problem still happens with that?

We had a bunch of changes in the IRQ rx handler recently, so the problem may be solved in our internal tree already.

[1] https://wireless.wiki.kernel.org/en/users/drivers/iwlwifi/core_release

[i (i-linux-kernel-bugs)]

No idea, but I'll give a try with LTS 4.14.72.

I will also try core39 release. The new -41 firmware is also needed, right?

[i (i-linux-kernel-bugs)]

Here's some test result:

1. The issue does not happen (no "BUG" line or stack traces) on 4.14.72 but is reproducible on 4.18.11.

2. The issue also does not happen with the new core39 release and -41 firmware.

3. The "Unhandled alg: 0x707" line appears in all cases.

Test method:

Using 123456/123456 as username/password, and TTLS/MSCHAPv2 as the authentication mechanism for a WPA2-EAP Eduroam hotspot. In my case, the hotspot is provided by some Cisco device.

[i (i-linux-kernel-bugs)]

As a side note, the scripts/uninstall.sh in backport-iwlwifi does not support xz compression for kernel modules.

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This change was made by a bot.

[Cristian Aravena Romero (caravena)]

Did this issue start happening after an update/upgrade? Was there a prior kernel version where you were not having this particular problem?

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v4.19-rc7 kernel[0].

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.19-rc7/

[Cristian Aravena Romero (caravena)]

https://bugzilla.kernel.org/show_bug.cgi?id=201297
--
Cristian Aravena Romero (caravena)

[luca (luca-linux-kernel-bugs)]

Thanks for the initial analysis, we'll try to find if there's any patch that fixes it in core39.

[johannes (johannes-linux-kernel-bugs)]

(In reply to Haochen Tong from comment #4)
> As a side note, the scripts/uninstall.sh in backport-iwlwifi does not
> support xz compression for kernel modules.

Fixed that in upstream backports, it'll take a while to propagate to backport-iwlwifi, but it eventually will.

Thanks for the report!

[luca (luca-linux-kernel-bugs)]

Thanks, Johannes! I'll pull your fix into our core release soon, during my next round of merging back from upstream.

The patch that fixes this "scheduling while atomic" issue is already in net-next.git[1] and should go to 4.20. When it reaches 4.20 it will be automatically backported to the relevant stable kernels. I could have sent it to the 4.19-rc series, but since it's not a regression in 4.19, I preferred to send it to -next instead.

[1] https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git/commit/?id=3baf7528d6f832b28622d1ddadd2e47f6c2b5e08

[luca (luca-linux-kernel-bugs)]

*** Bug 201297 has been marked as a duplicate of this bug. ***

[luca (luca-linux-kernel-bugs)]

*** Bug 201355 has been marked as a duplicate of this bug. ***

[Cristian Aravena Romero (caravena)]

https://bugzilla.kernel.org/show_bug.cgi?id=200991#c7

"Thanks, Johannes! I'll pull your fix into our core release soon, during my next round of merging back from upstream.

The patch that fixes this "scheduling while atomic" issue is already in net-next.git[1] and should go to 4.20. When it reaches 4.20 it will be automatically backported to the relevant stable kernels. I could have sent it to the 4.19-rc series, but since it's not a regression in 4.19, I preferred to send it to -next instead.

[1] https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git/commit/?id=3baf7528d6f832b28622d1ddadd2e47f6c2b5e08"

Best regards,
--
Cristian Aravena Romero (caravena)

[f.kargl (f.kargl-linux-kernel-bugs)]

Created attachment 279255
iwlwifi scheduling while atomic, dmesg output

Just ran into a similar issue while trying to connect to eduroam, although with correct credentials (stored in password manager). I know it's been fixed already, but more data can't hurt.

Debian 4.18.10-2,
iwlwifi-7260-17.ucode,
firmware version: 17.948900127.0