You are not logged in.

#26 2020-07-17 08:47:47

Pillgar
Member
Registered: 2017-07-05
Posts: 41

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

I've downgraded to nvidia 440-dkms since my last post, and haven't noticed any lockups.  That's with XFCE, lightdm (though I auto-login), Kernel 5.7.8-1-ck-ivybridge, and a 1070ti card. 

@d_fajardo do you have nvidia-drm.modeset=1 enabled?

Offline

#27 2020-07-18 14:43:06

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

@Pillgar Yes I have nvidia-drm.modeset=1 enabled. However...

I finally decided to do a complete reinstall of archlinux rather than chasing after this fantom of a solution. I made the decision not only of this problem. Well first I didn't really like that sddm crashing in the middle of a session. Secondly I have noticed a spike in my cpu usage even when idle which I was also trying to trace (pulseaudio and xorg were the culprits). Finally I just discovered I have a syslinux directory in my /boot directory. I use refind and never used syslinux so how that directory ended there is a big mystery. So because of those things, I felt something happened with my system.  It also seems to me that, except for a few, not many people are experiencing same problems to call for a bug ,  so instead of wasting more time trying to hunt for a solution, I decided to take that time instead to reinstall taking the opportunity to clear up some clutter in the process.

And with this, the problems are gone. nvidia and KDE are in love again, cpu runs on bare minimum on idle, and a leaner system.

I thank particularly V1del for helping look at the problem. I will close this thread with the solution to this particular problem to this particular system - reinstall.

Offline

#28 2020-07-21 21:09:20

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

I'm afraid I have to reopen this thread again. My reinstall fixed most issues I was having personally but I was mistaken and the issue with KDE and nvidia hasn't gone away. In the beginning, a minimal plasma install seemed OK but I still get freezes on PC running idle (kscreenlocker issue it seems) and the compositor crashing adter wakeup from suspend. The logs look exactly the same as I posted.

There are others experiencing these problems - here, here and here.

Isn't this looking like a bug? And if it is, which one - nvidia or KDE?

Any thoughts? How does one file a bug report?

Offline

#29 2020-07-22 01:05:27

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 21,431

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

If you actually run into issues from waking up from suspend (that wasn't clear to me, session idling and kscreenlock invoking shouldn't matter if it isn't actually set to suspend, if you actually intend to wake up from suspension that is an entirely different ballpark), that is likely an issue with the nvidia driver  (... possibly in combination with the KDE related work they actually did in that driver, e.g. a feature I particularly like is that they've managed to fix a non updating plasma panel in this driver version, it's not too far fetched that something they've worked on here bites itself on restoration of the compositor). It could also be kscreenlock-specific in particular, as kscreenlock uses a sandbox that only allows a certain specific set of OpenGL calls to be made within it's context. It's possible that a new command was added here on the nvidia side during wakeup that bites itself with what the kscreenlock sandbox expects, which could indeed make this something that is fixable on KDE's side of things.

However nvidia has some newer methods for suspension/wakeup that aren't enabled by default that should probably be tested if they aren't yet active: http://us.download.nvidia.com/XFree86/L … ement.html Pay attention to the needed kernel module parameters and the size considerations mentioned there, you don't need to do the install part mentioned and can just enable the relevant services if you use the normal Arch packages.

Regardless of this, if you want to make a bug report about this it would be best to post a thread on their Linux forum including detailed reproduction steps as well as an attachment of the generated tarball from running nvidia-bug-report.sh as root (from what I'm reading you might have to change the extension to something like txt to fool the board software to attach that). If it's in any way possible attach that after reproduction of the issue. (Is it actually "frozen" ? Can you not switch VTs with ctrl+alt+F2 or so?)

Last edited by V1del (2020-07-22 01:10:19)

Offline

#30 2020-07-22 06:45:30

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

Thanks so much for the link V1del. I will definitely have a look at the nvidia and bug tracker sites. And I can switch to a TTY. Does this mean it is a purely graphics problem?

Edit: Another thing to mention as well is I'm getting I believe are tearings in some applications in this new install. Konsole has produces lines and KPatience produces lines when cards are dragged. Search in the internet and some say it has to do with scaling (I'm in 4k so I have to scale things up). Anyone has any thoughts on these as well?

Last edited by d_fajardo (2020-07-22 07:09:07)

Offline

#31 2020-07-24 12:19:06

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

The nvidia suspension/wakeup services didn't really fix the problem. I also had a look at the Linux nvidia forum. According to Phoronix a fix was done for to fix plasma crashing with the compositor disabled for this current driver and perhaps that fix contributed to the current problem. I think I'll just sit it tight for the moment. After recent updates, some things seem better - I don't get the compositor crashing anymore on wakeup from sleep but instead I get the message 'Desktop effects were restarted due to graphics reset'. I can l live with that for the moment. The manual invocation of the screenlock works, but I have to disable auto-screenlock otherwise I get a freeze when KDE tries to trigger it by itself.

Last edited by d_fajardo (2020-07-24 12:28:27)

Offline

#32 2020-07-29 20:51:15

radiobla
Member
Registered: 2020-02-19
Posts: 65

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

I'm having the very same issue with nvidia 450.57 (linux 5.7.10), but although kscreenlock seems to invoke the issue, the process hanging is the kwin_x11.

I was able to switch to TTY4 (for some reason TTY2 and TTY3 were not available) and see kwin_x11 on "top" consuming 99% of CPU. I killed the process (kwin_x11) and restarted display-manager.service through systemctl.

This way I was able to recover the system without needing to do a "cold boot".

You can try a

kwin_x11 --replace

if you see no decorations on windows.

Have you guys found a better workaround?

Offline

#33 2020-07-29 20:53:19

radiobla
Member
Registered: 2020-02-19
Posts: 65

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

One more thing:

The issue doesn't happen consistently every single time that screen lock is invoked. It happens intermittently.

Offline

#34 2020-07-30 18:34:46

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

@rabiola kwin-x11 --replace crashes badly for me but read on.

Anyway, just to update, I have filed bug reports to both NVIDIA and KDE and the good news is both have replied. I received an email from NVIDIA that they have taken the bug report internally and will setup a system to try to reproduce it.
The current state of the KDE bug report is here.

Fingers crossed there will be a favourable outcome to all these.

EDIT:
The auto screen lock function seems to be fixed now after the recent upgrades. I can leave the machine to lock by itself without it freezing.
The problem with suspend continues however and issuing kwin-x11 --replace crashes badly.

Last edited by d_fajardo (2020-07-31 05:57:43)

Offline

#35 2020-08-11 16:55:57

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

Another update in this issue. Initially, nvidia tested it on an Ubuntu setup and couldn't reproduce the problems. Suspecting it might a newer kernel against the 450 driver, I requested that they test on an Archlinux setup and they are now confirming the same issues I'm experiencing.
So it does look like the problem stems from the newer kernel (possible the 5.7) and Nvidia 450.
Nvidia is now investigating the problem and hopefully come up with a fix in the next nvidia update.

Offline

#36 2020-08-14 06:20:34

kokoko3k
Member
Registered: 2008-11-14
Posts: 2,390

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

Not really a solution, but since nvidia now supports render offload, if you have an integrated gpu, you can just use it and use the discrete gpu just when it is needed.

kwin and kde works so much better with my haswell now, power consumption is lower and the (eco)system feels more coherent.
There's a small performance hit (2...5%) when using the nvidia gpu, but i live with that.

Last edited by kokoko3k (2020-08-14 06:21:15)


Help me to improve ssh-rdp !
Retroarch User? Try my koko-aio shader !

Offline

#37 2020-08-14 07:33:28

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

@kokoko3k As you said it's not a solution but I'd like to try that particularly if it solves the Suspend problem. How do you initiate render offload?

Edit: Nevermind. I found it here.

Last edited by d_fajardo (2020-08-14 11:46:01)

Offline

#38 2020-08-23 21:04:58

kokoko3k
Member
Registered: 2008-11-14
Posts: 2,390

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

Did it solved the suspend issue?


Help me to improve ssh-rdp !
Retroarch User? Try my koko-aio shader !

Offline

#39 2020-08-23 21:25:50

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

@kokoko3k No not for me. It didn't solve the suspend issue and I am wondering actually why not since the change made intel as the default desktop gpu. Anyway I reverted back since I didn't like the graphics performance in general.
Aside from the suspend, nvidia graphics work fine. I just have to wait till nvidia gets a fix. They eventually will have to since the other distros will have to upgrade to the newer kernels and they will then be experiencing these problems.

Offline

#40 2020-08-24 21:06:21

radiobla
Member
Registered: 2020-02-19
Posts: 65

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

Neither for me. It's happening less often, but it hasn't happened so far since some updates on nvidia drivers.

pacman -Qi nvidia
Name            : nvidia
Version         : 450.66-14

I found a workaround to get the system back if kscreenlockd and/or kwin_x11 hangs. Go to your TTY (i.e. CTRL+Alt+F2) and kill "kscreenlockd" then get back to your graphical mode (in my case ctrl+alt+f1).

This usually unfreezes the graphical mode. If the problem persists, specially if you are using compositor, try to kill "kwin_x11". You will get out of window borders/decorations, if so run "kwin_x11 --replace" or maybe just "kwin_x11" is gonna work. Do it in a terminal on graphical mode or try "DISPLAY=:0 kwin_x11 --replace" still on TTY (preferably if you're logged in with the same user). This should bring window decorations back.

That workaround is not ideal, but it can save you from losing files in case the system freezes so that you don't need to reboot.

My current kernel is 5.8.3.

Let us know if the workaround works for you on next freeze.

Last edited by radiobla (2020-08-24 21:09:06)

Offline

#41 2020-08-24 21:54:02

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

No, kwin_x11 --replace doesn't work.
A few people has tried that with same result as noted in this bug report.

The problem with kscreenlock seems to have disappeared and freezing doesn't seem to be a problem any longer as well. It's still just the compositor crashing on wakeup from suspend and this does point to a glx problem with the new nvidia driver. And as noted kwin_x11 replace doesn't solve the issue.

Offline

#42 2020-08-24 22:28:21

radiobla
Member
Registered: 2020-02-19
Posts: 65

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

The kscreenlockd issue has just happened here. I just killed the process and got the system back. However, Plasma Compositor was not enabled, if that info can help you. kwin_x11 didn't go to 100% CPU usage, neither hung.

Offline

#43 2020-09-01 18:43:48

Doimhneacht
Member
Registered: 2020-09-01
Posts: 1

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

radiobla wrote:

I found a workaround to get the system back if kscreenlockd and/or kwin_x11 hangs. Go to your TTY (i.e. CTRL+Alt+F2) and kill "kscreenlockd" then get back to your graphical mode (in my case ctrl+alt+f1).

Thank you, radiobla. Your workaround did the trick for me.

I've been experiencing (seemingly) random freezes for a few months now. This thread helped me to realize that it was always because of screen lock. Today it happened to me twice in a row, both times after attempt to lock screen manually (as opposed to autolock).

I have Nvidia 450.57-9 and Linux 5.8.1.

Offline

#44 2020-09-02 01:43:24

radiobla
Member
Registered: 2020-02-19
Posts: 65

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

I had to disable screen locking. It haven't happened so far. It's been a few months since this issue began to happen.

Offline

#45 2020-09-02 07:54:01

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

Strange that I don't have a problem with screenlock now, both auto-invocation and manually.
Has anyone tried it on linux-lts and see if it a fix?

Offline

#46 2020-09-04 18:16:12

Xakep_SDK
Member
Registered: 2017-09-23
Posts: 61

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

There are 2 bugs:
1. DPMS bug - screen sometimes does not wake up after power saving mode, simply switch to TTY and back one or many times
2. kscreenlock_greet bug - for me, it happens always next lock after DPMS bug. Switch to TTY and run this: sudo loginctl unlock-sessions, switch back.
No need to kill any programs, everything works fine, except maybe some compositing bugs with firefox, just restart it.

Offline

#47 2020-10-03 07:20:49

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

Just an update for those having the same issues. According to KDE Bug Report, they have traced the problem to a driver regression.
A fix looks like forthcoming.

Offline

#48 2020-10-05 11:00:37

VxMxPx
Member
From: Slovenia
Registered: 2012-12-22
Posts: 12

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

I have similar issue on Gnome. Last week or so random freezes, music will continue playing for a while everything else stops. No keys responds, no mouse movements. Can't do anything else but hard reboot.

I did try to SSH from another machine, htop shows xorg at 100%.

› pacman -Qi nvidia
Name            : nvidia
Version         : 455.23.04-4

› uname -r
5.8.13-arch1-1

Offline

#49 2020-10-09 05:11:55

Orbital_sFear
Member
Registered: 2014-10-13
Posts: 43

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

I have recently been getting random lockups as well. 980 TI. Normal use, 4 monitors with youtube and writing code.

Here is what my crash looks like

Oct 08 21:53:55 STI kernel: BUG: kernel NULL pointer dereference, address: 0000000000000020
Oct 08 21:53:55 STI kernel: #PF: supervisor read access in kernel mode
Oct 08 21:53:55 STI kernel: #PF: error_code(0x0000) - not-present page
Oct 08 21:53:55 STI kernel: PGD 80000006ee339067 P4D 80000006ee339067 PUD 0 
Oct 08 21:53:55 STI kernel: Oops: 0000 [#1] PREEMPT SMP PTI
Oct 08 21:53:55 STI kernel: CPU: 0 PID: 674 Comm: irq/139-nvidia Tainted: P           OE     5.8.13-arch1-1 #1
Oct 08 21:53:55 STI kernel: Hardware name: System manufacturer System Product Name/Z170-A, BIOS 1302 11/09/2015
Oct 08 21:53:55 STI kernel: RIP: 0010:_nv027467rm+0x9/0x90 [nvidia]
Oct 08 21:53:55 STI kernel: Code: 90 ff e8 ea b0 00 00 31 c0 48 83 c4 08 c3 31 c0 eb bf 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 83 ec 08 48 85 ff 74 57 <48> 8b 17 31 c0 48 85 d2 75 0e eb 2b 0f 1f 00 48 8b>
Oct 08 21:53:55 STI kernel: RSP: 0018:ffff8e8ec0ed7be0 EFLAGS: 00010202
Oct 08 21:53:55 STI kernel: RAX: 0000000000000020 RBX: 0000000000000020 RCX: 0000000000000010
Oct 08 21:53:55 STI kernel: RDX: ffff88d12f12c6c8 RSI: ffffffffffffffff RDI: 0000000000000020
Oct 08 21:53:55 STI kernel: RBP: ffff88d5ed66a940 R08: ffffffffc250bbf0 R09: ffff88d5ed66a920
Oct 08 21:53:55 STI kernel: R10: ffffffffc11e3160 R11: ffff88d60759e008 R12: 0000000000000020
Oct 08 21:53:55 STI kernel: R13: 0000000000000000 R14: ffff88d5ed66aaa8 R15: ffff88d5ed66abb0
Oct 08 21:53:55 STI kernel: FS:  0000000000000000(0000) GS:ffff88d616400000(0000) knlGS:0000000000000000
Oct 08 21:53:55 STI kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Oct 08 21:53:55 STI kernel: CR2: 0000000000000020 CR3: 0000000373508004 CR4: 00000000003606f0
Oct 08 21:53:55 STI kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Oct 08 21:53:55 STI kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Oct 08 21:53:55 STI kernel: Call Trace:
Oct 08 21:53:55 STI kernel:  ? _nv029871rm+0x1b/0x90 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv025414rm+0x18/0x60 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv011658rm+0x13d/0x1c0 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv000083rm+0x12f/0x1a0 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv011586rm+0xff/0x180 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv018400rm+0x1af/0x210 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv018340rm+0xd9a/0xe90 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv018341rm+0xde/0x260 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv018307rm+0x72/0xc0 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv018321rm+0x235/0x2d0 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv026016rm+0x10/0x10 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv018354rm+0xac/0xe0 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv027674rm+0x820/0xdc0 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv007560rm+0x155/0x270 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv027682rm+0x8d/0x180 [nvidia]
Oct 08 21:53:55 STI kernel:  ? _nv000711rm+0xa9/0x200 [nvidia]
Oct 08 21:53:55 STI kernel:  ? disable_irq_nosync+0x10/0x10
Oct 08 21:53:55 STI kernel:  ? rm_isr_bh+0x1c/0x60 [nvidia]
Oct 08 21:53:55 STI kernel:  ? nvidia_isr_kthread_bh+0x1b/0x40 [nvidia]
Oct 08 21:53:55 STI kernel:  ? irq_thread_fn+0x20/0x60
Oct 08 21:53:55 STI kernel:  ? irq_thread+0xf5/0x1a0
Oct 08 21:53:55 STI kernel:  ? irq_finalize_oneshot.part.0+0xe0/0xe0
Oct 08 21:53:55 STI kernel:  ? irq_thread_check_affinity+0xd0/0xd0
Oct 08 21:53:55 STI kernel:  ? kthread+0x142/0x160
Oct 08 21:53:55 STI kernel:  ? __kthread_bind_mask+0x60/0x60
Oct 08 21:53:55 STI kernel:  ? ret_from_fork+0x22/0x30
Oct 08 21:53:55 STI kernel: Modules linked in: fuse nvidia_drm(POE) nvidia_modeset(POE) nls_iso8859_1 nls_cp437 vfat fat nvidia(POE) intel_rapl_msr intel_rapl_common mei_hdcp eeepc_wmi asus_wmi x86_pkg_temp_therm>
Oct 08 21:53:55 STI kernel:  drm_ttm_helper ttm drm_kms_helper cec rc_core syscopyarea sysfillrect sysimgblt fb_sys_fops vboxsf vboxguest sg drm crypto_user agpgart ip_tables x_tables hid_generic usbhid hid btrfs>
Oct 08 21:53:55 STI kernel: CR2: 0000000000000020
Oct 08 21:53:55 STI kernel: ---[ end trace 19808a81dd5bfbab ]---
Oct 08 21:53:55 STI kernel: RIP: 0010:_nv027467rm+0x9/0x90 [nvidia]
Oct 08 21:53:55 STI kernel: Code: 90 ff e8 ea b0 00 00 31 c0 48 83 c4 08 c3 31 c0 eb bf 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 83 ec 08 48 85 ff 74 57 <48> 8b 17 31 c0 48 85 d2 75 0e eb 2b 0f 1f 00 48 8b>
Oct 08 21:53:55 STI kernel: RSP: 0018:ffff8e8ec0ed7be0 EFLAGS: 00010202
Oct 08 21:53:55 STI kernel: RAX: 0000000000000020 RBX: 0000000000000020 RCX: 0000000000000010
Oct 08 21:53:55 STI kernel: RDX: ffff88d12f12c6c8 RSI: ffffffffffffffff RDI: 0000000000000020
Oct 08 21:53:55 STI kernel: RBP: ffff88d5ed66a940 R08: ffffffffc250bbf0 R09: ffff88d5ed66a920
Oct 08 21:53:55 STI kernel: R10: ffffffffc11e3160 R11: ffff88d60759e008 R12: 0000000000000020
Oct 08 21:53:55 STI kernel: R13: 0000000000000000 R14: ffff88d5ed66aaa8 R15: ffff88d5ed66abb0
Oct 08 21:53:55 STI kernel: FS:  0000000000000000(0000) GS:ffff88d616400000(0000) knlGS:0000000000000000
Oct 08 21:53:55 STI kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Oct 08 21:53:55 STI kernel: CR2: 0000000000000020 CR3: 0000000373508004 CR4: 00000000003606f0
Oct 08 21:53:55 STI kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Oct 08 21:53:55 STI kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Oct 08 21:53:55 STI kernel: BUG: kernel NULL pointer dereference, address: 0000000000000027
Oct 08 21:53:55 STI kernel: #PF: supervisor read access in kernel mode
Oct 08 21:53:55 STI kernel: #PF: error_code(0x0000) - not-present page
Oct 08 21:53:55 STI kernel: PGD 80000006ee339067 P4D 80000006ee339067 PUD 0 
Oct 08 21:53:55 STI kernel: Oops: 0000 [#2] PREEMPT SMP PTI
Oct 08 21:53:55 STI kernel: CPU: 0 PID: 674 Comm: irq/139-nvidia Tainted: P      D    OE     5.8.13-arch1-1 #1
Oct 08 21:53:55 STI kernel: Hardware name: System manufacturer System Product Name/Z170-A, BIOS 1302 11/09/2015
Oct 08 21:53:55 STI kernel: RIP: 0010:do_exit+0x371/0xab0
Oct 08 21:53:55 STI kernel: Code: d5 d5 26 00 48 89 df e8 7d 90 28 00 45 84 ed 0f 85 34 04 00 00 48 89 df e8 4c 50 02 00 e8 d7 f7 01 00 48 89 df e8 7f f3 f9 ff <f6> 43 27 02 0f 85 69 05 00 00 48 89 df e8 5d aa 17>
Oct 08 21:53:55 STI kernel: RSP: 0018:ffff8e8ec0ed7eb0 EFLAGS: 00010286
Oct 08 21:53:55 STI kernel: RAX: 0000000080000000 RBX: 0000000000000000 RCX: 0000000000000000
Oct 08 21:53:55 STI kernel: RDX: 0000000000000001 RSI: 0000000000000000 RDI: 00000000ffffffff
Oct 08 21:53:55 STI kernel: RBP: ffff88d5f932dd00 R08: 0000000000000001 R09: 0000000000000000
Oct 08 21:53:55 STI kernel: R10: ffff88d60f5c7c00 R11: ffff8e8ec0ed7801 R12: ffff88d5f932e4ac
Oct 08 21:53:55 STI kernel: R13: 0000000000000001 R14: 0000000000000001 R15: ffff88d5f932dd00
Oct 08 21:53:55 STI kernel: FS:  0000000000000000(0000) GS:ffff88d616400000(0000) knlGS:0000000000000000
Oct 08 21:53:55 STI kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Oct 08 21:53:55 STI kernel: CR2: 0000000000000027 CR3: 0000000373508004 CR4: 00000000003606f0
Oct 08 21:53:55 STI kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Oct 08 21:53:55 STI kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Oct 08 21:53:55 STI kernel: Call Trace:
Oct 08 21:53:55 STI kernel:  ? task_work_run+0x5c/0x90
Oct 08 21:53:55 STI kernel:  ? do_exit+0x369/0xab0
Oct 08 21:53:55 STI kernel:  ? kthread+0x142/0x160
Oct 08 21:53:55 STI kernel:  ? rewind_stack_do_exit+0x17/0x17
Oct 08 21:53:55 STI kernel: Modules linked in: fuse nvidia_drm(POE) nvidia_modeset(POE) nls_iso8859_1 nls_cp437 vfat fat nvidia(POE) intel_rapl_msr intel_rapl_common mei_hdcp eeepc_wmi asus_wmi x86_pkg_temp_therm>
Oct 08 21:53:55 STI kernel:  drm_ttm_helper ttm drm_kms_helper cec rc_core syscopyarea sysfillrect sysimgblt fb_sys_fops vboxsf vboxguest sg drm crypto_user agpgart ip_tables x_tables hid_generic usbhid hid btrfs>
Oct 08 21:53:55 STI kernel: CR2: 0000000000000027
Oct 08 21:53:55 STI kernel: ---[ end trace 19808a81dd5bfbac ]---
Oct 08 21:53:55 STI kernel: RIP: 0010:_nv027467rm+0x9/0x90 [nvidia]
Oct 08 21:53:55 STI kernel: Code: 90 ff e8 ea b0 00 00 31 c0 48 83 c4 08 c3 31 c0 eb bf 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 83 ec 08 48 85 ff 74 57 <48> 8b 17 31 c0 48 85 d2 75 0e eb 2b 0f 1f 00 48 8b>
Oct 08 21:53:55 STI kernel: RSP: 0018:ffff8e8ec0ed7be0 EFLAGS: 00010202
Oct 08 21:53:55 STI kernel: RAX: 0000000000000020 RBX: 0000000000000020 RCX: 0000000000000010
Oct 08 21:53:55 STI kernel: RDX: ffff88d12f12c6c8 RSI: ffffffffffffffff RDI: 0000000000000020
Oct 08 21:53:55 STI kernel: RBP: ffff88d5ed66a940 R08: ffffffffc250bbf0 R09: ffff88d5ed66a920
Oct 08 21:53:55 STI kernel: R10: ffffffffc11e3160 R11: ffff88d60759e008 R12: 0000000000000020
Oct 08 21:53:55 STI kernel: R13: 0000000000000000 R14: ffff88d5ed66aaa8 R15: ffff88d5ed66abb0
Oct 08 21:53:55 STI kernel: FS:  0000000000000000(0000) GS:ffff88d616400000(0000) knlGS:0000000000000000
Oct 08 21:53:55 STI kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Oct 08 21:53:55 STI kernel: CR2: 0000000000000027 CR3: 0000000373508004 CR4: 00000000003606f0
Oct 08 21:53:55 STI kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Oct 08 21:53:55 STI kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Oct 08 21:53:55 STI kernel: Fixing recursive fault but reboot is needed!

Offline

#50 2020-10-09 06:50:27

d_fajardo
Member
Registered: 2017-07-28
Posts: 1,563

Re: [Solved] Random Freezes and Compositor Crashing with nvidia 450.57

The nvidia 455.28 update seems to have fixed the problem. However, it doesn't look like it's been perfectly ironed out - I still get the message 'desktop effects were restarted' when I wake up from suspend but kwin doesn't crash any longer and the problem with the black, blank artifacts in the KDE system is fixed on resume.
I have suppressed the 'desktop effects were restarted' notification for now in KDE and will continue to monitor how the suspend-resume functions behaves.

Offline

Board footer

Powered by FluxBB