ha! Yes I suppose AMD drivers would have very limited impact on an Intel chipset. Having all your RAM being consumed certainly sounds strange, I’m a bit out of my depth but I hope you find a solution to your problem.
I figured out what was going on, and it was totally unrelated. A lot of containers were running at start up and exhausting all of the memory. This was only noticeable when I had to remove one of the memory chips on the laptop because of an issue.
Glad you found the issue! and relieved to know I won’t be randomly running out of memory anytime soon.
I’ll also throw my specs into the ring here. Same issues and mine always seems triggered by quick scrubbing during video playback (specifically jumping time using the arrows). I also usually have 100 tabs open, so I just wrote it off as a memory issue.
FW13 Ryzen 7640U Arch 6.11.9 Gnome 47 on Wayland
Update from the same conf: it happens more and more often, and not only when watching videos. Happy to run tests and gather logs if needed.
1 Like
After you edit the grub file ( sudo nano /etc/default/grub) you need to update grub.
Grub file location:
/etc/default/grub
- Find the line with
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash" - Inside the last double-quote, add your Linux kernel boot parameters and/or Module-level parameters.
- I’m using: GRUB_CMDLINE_LINUX=“rhgb quiet amdgpu.dcdebugmask=0x10”
- Update Grub
Update grub using the below commands:
sudo grub2-mkconfig -o /etc/grub2.cfg
Followed by
sudo grub2-mkconfig -o /etc/grub2-efi.cfg
Reboot
Source:
see also:
6 Likes
Same. It happens too frequently, on my Ubuntu machine, and my partner’s Manjaro/Windows machine. Haven’t caught it happening in Windows yet, and probably won’t spend enough time to, but it happens to my machine about once every ~30 hours of system uptime. Once it happens, it does not recover. This is in gnome.
I’m still experiencing this issue on and off, I cannot find a way to monitor what is causing it to occur. The logs show nothing. The system becomes completely unresponsive while watching Youtube in Firefox. From brief search it looks like LOTS of people are experiencing this problem. Is there any recourse because until further notice my laptop is now classified as unstable, buggy and unperformant. Extremely frustrated with this on-going issue. I cannot state this enough.
I am also experiencing this issue with the only reliable workaround being using linux 6.10.9 with 1:24.2.2-1 (or possibly 1:24.2.7-1, just started testing) versions of mesa, lib32-mesa, vulkan-radeon and lib32-vulkan-radeon (Arch packages, these possibly have different names on Fedora). This has been going on since around August and my frustration has just been deepened, as trying the newest linux 6.12.7 has led to a general feeling of lag throughout the entire system (but mostly with the cursor). One positive thing about the new kernel is that when freezes occurred, the system fully recovers within 10 seconds (unlike the slowdowns, which require hibernation to recover), but they were way more frequent, which I still consider unusable.
I updated kernel to 6.12, and it stopped. I did this on both an ubuntu and a manjaro (think I may have bumped the manjaro to 6.13?) and haven’t suffered anymore.
I’m on 6.12 and still experiencing the issue and efindus is right, I’ve been seeing this problem since about end of August or so. Also experiencing the laggy touchpad cursor issue for a week or so now.
1 Like
Try 6.13? I haven’t had the laggy touchpad cursor, except when the freezing happened, and that was the well-documented problem of watching YouTube videos in Firefox.
I have a Frame.Work laptop 13 AMD 7640U with Debian 12 and the latest kernel from backports (6.11.10+bpo-amd64), and I experience the same issue described, with the system slowing down in stutters at random moments while watching videos on YouTube on Firefox (.deb package, 133.0.3). I assumed it was a rendering issue with VP-9 hardware acceleration, so I forced AV1 acceleration and everything seems to work great!
Before making any system changes, try this approach and let me know if it works for you too.
I’ll wait for 6.13 to enter stable. Thanks Daniel I’ll try AV1 out. But the question remains why would a video codec bring the entire system to a halt?
1 Like
I just got a Framework laptop for the first time a few days ago and I’ve run into this same issue several times today and yesterday. I’ve got a Framework 13, AMD Ryzen 7840, running Ubuntu 24.04 LTS, BIOS version 3.05, kernel version 6.8. Like a few other people in this thread, the issue seems to usually occur when a video is running in Firefox, though each time I’ve had a fair number of programs open so it was hard to narrow it down, and the slowness persists until a restart. Has anyone pinned down the source of this issue yet?
Fedora Kinoite on 6.12.7-200.fc41.x86_64
I’ve had stutters for some time now (seems like an eternity ahaha) every time I watch a video or scroll fast enough on anything gpu rendered it seems. I thought my laptop had an issue, kind of glad it’s a kernel one.
I wish the patch will be merged and published soon. Argh
Follow-up, 5 or 6 days ago I set the amdgpu.dcdebugmask=0x10 kernel parameter (which, if I understand the above discussion correctly, disables PSR) and the issue hasn’t recurred since then, so whatever the root cause is, this fix seems to work.
Is there any significant downside to leaving PSR off? From what I’ve read, it seems to just be a power-saving optimization, but it’s not clear to me whether it makes a significant difference in practice for a laptop’s battery life.
Same except I’m on a 7640U. Still ubuntu 24.04, BIOS v3.5, and kernel version 6.8.0-51-generic.
I was using the amdgpu.dcdebugmask=0x10 workaround for the last few weeks, but I think it could have introduced out of a sudden freezes to the system caused by null pointer exceptions, it took some time for me to figure out that this workaround could make my laptop freeze every ~30 minutes to few hours, something could have triggered it.
Now that I removed it, everything is stable, no freezes.
I just wanted to share, if someone experiences the same.
BUG: kernel NULL pointer dereference, address: 0000000000000010
Mar 06 17:10:28 fedora kernel: #PF: supervisor read access in kernel mode
Mar 06 17:10:28 fedora kernel: #PF: error_code(0x0000) - not-present page
Mar 06 17:10:28 fedora kernel: PGD 0 P4D 0
Mar 06 17:10:28 fedora kernel: Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI
Mar 06 17:10:28 fedora kernel: CPU: 4 UID: 1000 PID: 6203 Comm: code:cs0 Not tainted 6.11.4-301.fc41.x86_64 #1
Mar 06 17:10:28 fedora kernel: Hardware name: Framework Laptop 13 (AMD Ryzen 7040Series)/FRANMDCP05, BIOS 03.05 03/29/2024
Mar 06 17:10:28 fedora kernel: RIP: 0010:amdgpu_vm_bo_update+0x67c/0x760 [amdgpu]
Mar 06 17:10:28 fedora kernel: Code: 00 00 00 49 83 ce 08 48 c7 44 24 10 00 00 00 00 41 89 c7 48 8b 92 b8 f7 ff ff 41 83 e7 01 e9 8a fb ff ff 49 8b 83 c8 01 00 00 <8b> 40 10 83>
Mar 06 17:10:28 fedora kernel: RSP: 0018:ffffa79c04a6f790 EFLAGS: 00010246
Mar 06 17:10:28 fedora kernel: RAX: 0000000000000000 RBX: ffff996c8cd80000 RCX: ffff996e2c5ea540
Mar 06 17:10:28 fedora kernel: RDX: 0000001020000000 RSI: ffff996cd7fba120 RDI: ffff996e2c5ea540
Mar 06 17:10:28 fedora kernel: RBP: ffff996e2c5ea538 R08: ffff996e2c5ea500 R09: 0000000000000000
and another
BUG: kernel NULL pointer dereference, address: 0000000000000010
Mar 06 15:11:34 fedora kernel: #PF: supervisor read access in kernel mode
Mar 06 15:11:34 fedora kernel: #PF: error_code(0x0000) - not-present page
Mar 06 15:11:34 fedora kernel: PGD 19a43c067 P4D 19a43c067 PUD 0
Mar 06 15:11:34 fedora kernel: Oops: Oops: 0000 [#2] PREEMPT SMP NOPTI
Mar 06 15:11:34 fedora kernel: CPU: 4 UID: 1000 PID: 2674 Comm: gnome-shel:cs0 Tainted: G D 6.11.4-301.fc41.x86_64 #1
Mar 06 15:11:34 fedora kernel: Tainted: [D]=DIE
Mar 06 15:11:34 fedora kernel: Hardware name: Framework Laptop 13 (AMD Ryzen 7040Series)/FRANMDCP05, BIOS 03.05 03/29/2024
Mar 06 15:11:34 fedora kernel: RIP: 0010:amdgpu_vm_bo_update+0x67c/0x760 [amdgpu]
Mar 06 15:11:34 fedora kernel: Code: 00 00 00 49 83 ce 08 48 c7 44 24 10 00 00 00 00 41 89 c7 48 8b 92 b8 f7 ff ff 41 83 e7 01 e9 8a fb ff ff 49 8b 83 c8 01 00 00 <8b> 40 10 83>
Mar 06 15:11:34 fedora kernel: RSP: 0018:ffffc11d8803b7c0 EFLAGS: 00010246
Unfortunately I can confirm this is still an issue on kernel 6.13.5. I removed the amdgpu.dcdebugmask=0x10 kernel parameter after upgrading to kernel 6.13.5 about a week ago, and the same issue just recurred today. (I’m now using Secureblue, a bootable container image based on Fedora Silverblue; the choice of distro doesn’t seem to matter much for this issue as I’ve had it happen on Ubuntu, Fedora Workstation, and Secureblue.)
1 Like