i915: GPU HANG: ecode 9:0:0x00000000, hang on rcs0

[Egbert van der Wal]

I just installed Kubuntu 19.10 on my brand new Dell XPS 15 7590 laptop.

The display freezes quite regularly for a couple of seconds. In that period, the mouse pointer does move, but nothing else happens: I cannot interact, move windows, scroll, type or whatever. Afterwards the system becomes responsive again. In the kernel log I see messages like:

[ 2110.867998] i915 0000:00:02.0: GPU HANG: ecode 9:0:0x00000000, hang on rcs0
[ 2110.869007] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
[ 2120.523573] iwlwifi 0000:3b:00.0: Unhandled alg: 0x71b
[ 2146.443128] iwlwifi 0000:3b:00.0: Unhandled alg: 0x71b
[ 2173.551329] iwlwifi 0000:3b:00.0: Unhandled alg: 0x71b
[ 2204.008717] iwlwifi 0000:3b:00.0: Unhandled alg: 0x71b
[ 4404.960469] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
[ 6020.935285] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0

I found that this problem occurs significantly more frequent when I'm using Jetbrains IntelliJ IDE, but I did experience it in several other occassions as well.

I found similar reports both in UBuntu as other places that seem to be related:

https://bugs.launchpad.net/ubuntu/+source/linux-signed-5.4/+bug/1863087
https://bugs.freedesktop.org/show_bug.cgi?id=111805
https://bugs.archlinux.org/task/64725
https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/1862865
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1861395

The ones on Launchpad are marked as invalid for Eoan and only apply for Focal.

The XPS 7590 has a Intel chip and nvidia chip, I'm experiencing the problem both with and without the nvidia driver loaded. I ran apport when it wasn't loaded to make sure the kernel isn't tainted with it.

Based on the feedback I see that this may be fixed in the most recent releases of kernel 5.4 (>= 5.4.0-17) but I don't see a proper way to install this in Eoan.

ProblemType: Bug
DistroRelease: Ubuntu 19.10
Package: linux-image-5.3.0-47-generic 5.3.0-47.39
ProcVersionSignature: Ubuntu 5.3.0-47.39-generic 5.3.18
Uname: Linux 5.3.0-47-generic x86_64
ApportVersion: 2.20.11-0ubuntu8.8
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: evdwal 1280 F.... pulseaudio
CurrentDesktop: KDE
Date: Thu Apr 9 17:25:04 2020
InstallationDate: Installed on 2020-04-06 (2 days ago)
InstallationMedia: Kubuntu 19.10 "Eoan Ermine" - Release amd64 (20191017)
MachineType: Dell Inc. XPS 15 7590
ProcFB: 0 i915drmfb
ProcKernelCmdLine: \\vmlinuz-5.3.0-47-generic root=/dev/mapper/vgkubuntu-root ro quiet splash net.ifnames=0 biosdevname=0 vt.handoff=7 initrd=\initrd.img-5.3.0-47-generic
RelatedPackageVersions:
 linux-restricted-modules-5.3.0-47-generic N/A
 linux-backports-modules-5.3.0-47-generic N/A
 linux-firmware 1.183.5
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 12/25/2019
dmi.bios.vendor: Dell Inc.
dmi.bios.version: 1.5.0
dmi.board.name: 0VYV0G
dmi.board.vendor: Dell Inc.
dmi.board.version: A00
dmi.chassis.type: 10
dmi.chassis.vendor: Dell Inc.
dmi.modalias: dmi:bvnDellInc.:bvr1.5.0:bd12/25/2019:svnDellInc.:pnXPS157590:pvr:rvnDellInc.:rn0VYV0G:rvrA00:cvnDellInc.:ct10:cvr:
dmi.product.family: XPS
dmi.product.name: XPS 15 7590
dmi.product.sku: 0905
dmi.sys.vendor: Dell Inc.

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

This change was made by a bot.

[Egbert van der Wal (eggie)]

Another remark: I already enabled -propsed for Eoan to see if that includes an updated, therefore I'm now running linux-image-5.3.0-47-generic from eoan-proposed. I had the exact same problems while running the kernel from the main repo linux-image-5.3.0-46-generic.

[CrazyCoder (sbaranov)]

JetBrains team has received a dozen of similar reports in the last couple of days. They are currently linked to the old issue at https://youtrack.jetbrains.com/issue/JBR-2265. See also the related tickets.

While the original issue was probably caused by something different, these new reports indicate a different problem that started to occur after the recent Ubuntu (kernel?) updates.

[Geir Isene (qc-e-9h)]

Rather than adding a report as the dozen+1, I will just chime in the I have the same on my Dell XPS15 after upgrading three days ago.

[Steve Murphy (smurphos)]

Same experience with Mint 19.3 on a bionic base with kernel 5.3.0-46-generic

[Egbert van der Wal (eggie)]

Using the `mainline` utility from https://launchpad.net/~cappelikan/+archive/ubuntu/ppa I now installed a mainline kernel version 5.4.28-050428-generic - the latest currently available in the 5.4 series.

Using this kernel, I don't experience this issue at all - no hangs, not even during the use of IntelliJ. I tested it for a couple of hours in which I normally had dozens of hangs.

If it helps diagnosing this issue, I'm willing to bisect the issue with various mainline versions between 5.3.0-47 and 5.4.28. Let me know if this has any value. For now I'm sticking with this new kernel version.

Not sure if related - I did have an additional issue with the graphics. Directly after boot, but also after suspend, during connecting to wifi, I'd get loads of the `iwlwifi 0000:3b:00.0: Unhandled alg: 0x71b` messages and the display would also freeze for couple of seconds (10 - 20). During this specific freeze, the mouse pointer would also be unresponsive. This would be accompanied by loads of messages in the kernel log like:

nouveau 0000:01:00.0: fifo: SCHED_ERROR 08 []

I managed to fix this by adding nouveau.modeset=0 to the kernel command line.

Like I said, I don't know if these are related. Does the i915 GPU hang also occur on laptops without a nvidia gpu?

[Matthieu Vion (magentix)]

Same experience here. I upgrade from 5.3.0-47-generic to 5.3.0-51-generic the problem is still here :

Apr 30 10:30:45 xxxx kernel: [ 7955.957968] i915 0000:00:02.0: GPU HANG: ecode 9:0:0x00000000, hang on rcs0
Apr 30 10:30:45 xxxx kernel: [ 7955.958978] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0

System freeze randomly for 5 seconds. Ubuntu 18.04 was installed on my laptop in december 2019, with probably 5.3.0-23 or 5.3.0-24 and worked find...

[Grzegorz (g-kuznik)]

I'm facing same issue. I thought that is the JetBrains software update issue, because I updated PHPStorm to 2020.1

I have two different issues. Only PHPStorm freezes or whole system freezes.

*** Log for PHPStorm freeze: ***

Apr 30 11:18:57 XXXX gnome-shell[4130]: Could not create selection source for X11: Format TARGETS not supported
Apr 30 11:20:08 XXXX kernel: [11685.466519] [UFW BLOCK] IN=wlp2s0 OUT= MAC=01:00:5e:00:00:01:70:4f:57:55:f3:38:08:00 SRC=192.168.1.1 DST=224.0.0.1 LEN=36 TOS=0x00 PREC=0x00 TTL=1 ID=0 DF PROTO=2
Apr 30 11:22:12 XXXX kernel: [11809.996191] [UFW BLOCK] IN=wlp2s0 OUT= MAC=01:00:5e:00:00:01:70:4f:57:55:f3:38:08:00 SRC=192.168.1.1 DST=224.0.0.1 LEN=36 TOS=0x00 PREC=0x00 TTL=1 ID=0 DF PROTO=2
Apr 30 11:22:28 XXXX kernel: [11825.857676] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Apr 30 11:18:59 XXXX gnome-shell[4130]: message repeated 7 times: [ Could not create selection source for X11: Format TARGETS not supported]
Apr 30 11:22:28 XXXX jetbrains-phpstorm.desktop[4130]: 2020-04-30 11:22:28,719 [7663939] WARN - .diagnostic.PerformanceWatcher - UI was frozen for 7585ms, details saved to /home/grego/.cache/JetBrains/PhpStorm2020.1/log/threadDumps-freeze-20200430-112226-PS-201.6668.153-XlibWrapper.XQueryPointer-7sec

*** Log for system freeze: ***

Sometimes have two lines:

Apr 30 08:33:53 XXXX kernel: [ 1710.739980] i915 0000:00:02.0: GPU HANG: ecode 9:0:0x00000000, hang on rcs0
Apr 30 08:33:53 XXXX kernel: [ 1710.740998] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0

Sometimes second line appears alone:

Apr 30 13:50:14 XXXX kernel: [20691.862458] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0

[Peeter Rannou (thedotwriter)]

Like Egbert, upgrading to the lastest stable 5.4 kernel version (5.4.41-050441-generic in my case) resolved the issue. I have been using PHPStorm (v 2020.1) for a week and didn't experience any freeze since then.