You are not logged in.
Hello!
I recently installed the system on a new laptop. Sometimes the image on my screen freezes (sound is still playing). After a while, the image unfreezes
The error I see in the journalctl is: i915 0000:00:02.0: [drm] *ERROR* [CRTC:131:pipe B] flip_done timed out.
I don't use any xf86-video driver.
I am using kernel 6.3.7-arch1-1.
journalctl -b -1: http://0x0.st/HcvB.txt
Last edited by stefanovca (2023-06-12 06:02:41)
Offline
Same issue here.
Apparently, it is possible to regain computer control by switching to a console (Ctrl+Alt+F1) and back to the graphical environment (Ctrl+Alt+F7).
Offline
That's right, that's exactly how I am solving the problem now.
Offline
Jun 11 16:10:39 rakis kernel: i915 0000:00:02.0: [drm] *ERROR* [CRTC:131:pipe B] flip_done timed out
Jun 11 16:10:50 rakis kernel: i915 0000:00:02.0: [drm] *ERROR* flip_done timed out
Jun 11 16:10:50 rakis kernel: i915 0000:00:02.0: [drm] *ERROR* [CRTC:131:pipe B] commit wait timed out
Jun 11 16:14:01 rakis kernel: i915 0000:00:02.0: [drm] *ERROR* flip_done timed out
Jun 11 16:14:01 rakis kernel: i915 0000:00:02.0: [drm] *ERROR* [PLANE:82:plane 1B] commit wait timed out
Jun 11 16:14:01 rakis kernel: ------------[ cut here ]------------
Jun 11 16:14:01 rakis kernel: i915 0000:00:02.0: vblank wait timed out on crtc 1
xrandr -q
loginctl session-status # mainly whether and what compositor, though I guess som hw accelerated browser might be at play as well
@alzuin, likewise vblank timeout?
Offline
It happened again:
journalctl -b -0 : http://0x0.st/HcYN.txt
xrandr -q: http://0x0.st/HcYq.txt
loginctl session-status: http://0x0.st/HcYb.txt
Offline
You could try to replicate it
1. w/ a disabled compositor, https://wiki.archlinux.org/title/Xfwm#Composite_manager and
2. w/ only one output attached (ie. w/o the HDMI monitor)
Offline
@seth, yep, the same.
For more context, my laptop is a Lenovo P14s with an Intel i915 and an Nvidia Quadro for 3D acceleration.
The setup I made is explained in https://wiki.archlinux.org/title/NVIDIA_Optimus, and I use https://github.com/Askannz/optimus-manager to control it.
At the moment of the error, optimus-manager was configured to use the Nvidia card only (optimus-manager --switch nvidia). Now I'm trying to force using the i915 (optimus-manager --switch integrated), and I'm waiting to see if the problem happens again so I can send some logs.
Honestly, I don't know if the problem above is related to the above: the laptop is pretty new and I always had PC with a single card; perhaps I'm entirely wrong, and it is not related to this setup, also because stefanovca logs do not mention nvidia anywhere.
Last bit: searching for these errors in the forums, I tried adding some custom flags to grub: `intel_idle.max_cstate=1 i915.enable_psr=0 intremap=off button.lid_init_state=open`, but the problem persists.
Last edited by alzuin (2023-06-12 20:02:19)
Offline
OP doesn' have a hybrid setup (no nvidia PCI ID in the logs either)
Please also post journal, session status and xrandr output.
Offline
You could try to replicate it
1. w/ a disabled compositor, https://wiki.archlinux.org/title/Xfwm#Composite_manager and
2. w/ only one output attached (ie. w/o the HDMI monitor)
Tested for two days w/o the HDMI monitor. I haven't had a problem so far.
Offline
https://wiki.archlinux.org/title/Intel_ … on_(VSYNC)
Do you use the TearFree option w/ the modesetting driver?
Offline
https://wiki.archlinux.org/title/Intel_ … on_(VSYNC)
Do you use the TearFree option w/ the modesetting driver?
No, I don't use that option. I forgot to write that I use HP ProBook 450 G9 with Alder Lake-UP3 GT2 [Iris Xe Graphics].
Offline
Tried to disable vsync (and then using both outputs)?
Offline
Tried to disable vsync (and then using both outputs)?
I disabled vsync and tested for a few days. It sounds strange, but the problem only occurred when I put one monitor on top of the other in the display settings.
Offline
At that moment or under that condition?
The vertical layout means more vertical lines…
Jun 20 22:51:57 rakis kernel: RIP: 0010:drm_wait_one_vblank+0x1f7/0x210
And something is clearly waiting for the vblank signal.
What if you disable PSR, https://wiki.archlinux.org/title/Intel_ … flickering
Offline
having exactly the same issue since some time (some kernel 6.x update iirc) on my
HP ZBook Firefly 14 inch G9 Mobile Workstation PC/8ABB, BIOS U71 Ver. 01.07.00 03/23/2023
CPU: 12th Gen Intel i7-1260P (16) @ 4.700GHz
GPU: Intel Alder Lake-P
not using any xf86-video at all
the issue only shows up when external HDMI monitor is connected (and seems to preferably happen while streaming or
youtube vids, all in firefox). Turing off VSYNC leads to black external monitor (after login; mouse pointer is visible).
I have PSR disabled since install (otherwise the laptop was not usable at all)
journalctl -b: http://0x0.st/HT6k.txt
Offline
Turing off VSYNC leads to black external monitor (after login; mouse pointer is visible)
Please post an xorg log, and the outputs of
xrandr -q
DISPLAY=:0.1 xrandr -q
DISPLAY=:1 xrandr -q
in that situation and check whether you can still move windows onto the "black" output and whether the situation changes when you suspend the compositor (SHIFT+Alt+F12 to toggle it)
Offline
thx for helping!
with disabled VSYNC I get:
Xorg log: http://0x0.st/HT0C.txt
xrandr -q: http://0x0.st/HT0h.txt
for both values for DISPLAY I get "Can't open display..."
moving windows back and forth to the HDMI monitor works flawlessly, just the screen stays "black"
suspending the compositor doesn't change a thing
Offline
[ 21.656] (WW) modeset(0): Present-flip: queue async flip during flip on CRTC 1 failed: Device or resource busy
The only other occurrence known to the oracle of mountain view is https://forum.manjaro.org/t/issues-with … nux/142014 (more typically there's EINVAL, not EBUSY) - does anyone else get this error?
Also try the behavior w/ the LTS kernel.
Offline
Running on 6.1.35-1-lts for 4 days now without any issues...so guess will stick to this for now
Offline
You probably also want to check into the drm-tip kernels to see whether this is meawhile resolved.
https://aur.archlinux.org/packages?O=0&K=drm-tip
Because at some point the current kernel will become the new LTS one…
Offline
I would suggest linux-mainline 6.4 available from miffe and if that still has the issues I can build you either a drm-tip or drm-next kernel if that would help.
Offline
Issue still persists under linux 6.4.1.arch2-1; all good when using 6.1.37-1-lts
Offline
Issue still persists under linux 6.4.1.arch2-1; all good when using 6.1.37-1-lts
What about https://git.kernel.org/pub/scm/linux/ke … d3342ede18 linked below
https://drive.google.com/file/d/16fLZRT … sp=sharing linux-6.4.r12274.g5133c9e51de4-1-x86_64.pkg.tar.zst
https://drive.google.com/file/d/1odBFmr … sp=sharing linux-headers-6.4.r12274.g5133c9e51de4-1-x86_64.pkg.tar.zst
Offline
crashes at start up!
installed via:
paru -U Downloads/linux-headers-6.4.r12274.g5133c9e51de4-1-x86_64.pkg.tar.zst Downloads/linux-6.4.r12274.g5133c9e51de4-1-x86_64.pkg.tar.zst
leads to:
sudo journalctl -b -1: http://0x0.st/H1gi.txt
or am I missing something?
Offline
Thanks for testing can you try blacklisting the iwlwifi module to see if 6.5 introduced a new issue in that module.
Edit:
Possibly this bug https://bugzilla.kernel.org/show_bug.cgi?id=217622
Edit2:
Possible fix for the iwlwifi bug from https://lore.kernel.org/all/0068af47-e4 … inger.net/ applied to kernel linked below
https://drive.google.com/file/d/1aui_CO … sp=sharing linux-6.4.r12274.g5133c9e51de4-1.1-x86_64.pkg.tar.zst
https://drive.google.com/file/d/1XlCcw9 … sp=sharing linux-headers-6.4.r12274.g5133c9e51de4-1.1-x86_64.pkg.tar.zst
Last edited by loqs (2023-07-08 10:01:37)
Offline