Persistent System Crashes and Performance Issues on Fedora 40 with ASUS Vivobook 15

[lucas-winicius]

Hello,

My name is Lucas, and I own an ASUS Vivobook 15, AMD Ryzen 7, M1502IA-EJ252. Recently, I decided to switch from Pop!_OS to Fedora 40, attracted by its latest GNOME version. However, after some time using Fedora, I started encountering two significant issues.

1. Severe Freezing During Heavy Browser Usage: My first issue involves extreme system freezes when running multiple applications that rely on a web browser. For example, if I open around seven tabs in Firefox and simultaneously run apps like ‘Zap Zap,’ VS Code, and Beekeeper, my swap and RAM usage spikes to their maximum, causing my laptop to become unresponsive. While I’ve attempted to mitigate this by drastically reducing the number of open tabs, this has been largely unsuccessful. Even with minimal usage—such as four tabs in Firefox, Lunacy, VS Code, and Warp Terminal—I still experience frequent freezes, particularly while developing web applications.
2. Unexpected Logouts and System Instability: The second, more critical issue I’ve been facing is unexpected system logouts. Over the past two days, my screen has suddenly gone black as if the system has shut down. After less than a second, the login screen reappears, and upon logging in, I find that all my applications have closed. Yesterday, August 11, 2024, this happened twice within a three-minute span. Today, August 12, 2024, while studying with only Firefox open, I suspended my laptop and left. Upon returning, I discovered that my laptop had logged off again, even though it was suspended. Firefox had four tabs open at the time: localhost, a Tailwind color finder site, GitHub, and Lucide. Although I wasn’t working on anything critical, these incidents have made me uneasy about using Fedora, as I constantly worry about losing my work.

This is my first experience with Fedora, and despite these issues, I truly appreciate the system and would prefer not to switch to another distribution. For reference, my system is fully updated (Linux lucas 6.10.3-200.fc40.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Aug 5 14:30:00 UTC 2024 x86_64 GNU/Linux). I appreciate any assistance you can provide.

[catanzaro (Michael Catanzaro)]

The first problem sounds like you’re just running out of memory?

For the second problem, see AMD APU regression (full halt) on kernel 6.10 - how to best report?. I think kernel 6.10 is just bad. Try downgrading to kernel 6.9.

[stephaneklein (Stéphane Klein)]

@lucas-winicius

Unexpected Logouts and System Instability: The second, more critical issue I’ve been facing is unexpected system logouts. Over the past two days, my screen has suddenly gone black as if the system has shut down. After less than a second, the login screen reappears, and upon logging in, I find that all my applications have closed. Yesterday, August 11, 2024, this happened twice within a three-minute span. Today, August 12, 2024, while studying with only Firefox open, I suspended my laptop and left. Upon returning, I discovered that my laptop had logged off again, even though it was suspended.

Same issue on my workstation since 2024-08-05.

[maxmousee (Natan Silva)]

same issue here on Ryzen 7840/Framework 13"
I think it started with Kernel 6.10

[yuntaz (Y.Z.)]

Framework 13 Ryzen 7840 with Kernel 6.10 is likely related to the Mesa issue mentioned above.

@catanzaro

For the second problem, see AMD APU regression (full halt) on kernel 6.10 - how to best report? . I think kernel 6.10 is just bad. Try downgrading to kernel 6.9.

[maxmousee (Natan Silva)]

Indeed, I think that could be the case! Thanks!

[zsarge (Zack Sargent)]

Same issue on Fedora workstation in the same time period.

I detailed my experiences in another form post.

[tanhowzur]

Same-ish crashes to mine. To your points:

1. Oddly, even with browser usage. I’m using Brave myself. Frequent crashes/restarts.
2. My forced restarts are usually un-resolvable. I’ll be logged out, then after logging back in, it will terminate my main monitor, and only display on my laptop screen. That screen though, I’m unable to type anything. So I’ll do a hard restart. Sometimes the restart solves it, other times the loop will continue.

EDIT: to include $ inxi -Fzxi

System:
  Kernel: 6.10.6-200.fc40.x86_64 arch: x86_64 bits: 64 compiler: gcc
    v: 2.41-37.fc40
  Desktop: KDE Plasma v: 6.1.4 Distro: Fedora Linux 40 (Workstation Edition)
Machine:
  Type: Laptop System: HP product: OMEN by HP Laptop 17-ck0xxx v: N/A
    serial: <superuser required>
  Mobo: HP model: 88F7 v: 86.43 serial: <superuser required> UEFI: AMI
    v: F.47 date: 03/28/2024
Battery:
  ID-1: BAT0 charge: 75.6 Wh (100.0%) condition: 75.6/83.0 Wh (91.1%)
    volts: 12.6 min: 11.6 model: HP Primary status: full
CPU:
  Info: 8-core model: 11th Gen Intel Core i7-11800H bits: 64 type: MT MCP
    arch: Tiger Lake rev: 1 cache: L1: 640 KiB L2: 10 MiB L3: 24 MiB
  Speed (MHz): avg: 1031 high: 1400 min/max: 800/4600 cores: 1: 849 2: 1319
    3: 1216 4: 801 5: 800 6: 800 7: 1261 8: 1136 9: 982 10: 1205 11: 800
    12: 1190 13: 997 14: 1400 15: 941 16: 801 bogomips: 73728
  Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx
Graphics:
  Device-1: Intel TigerLake-H GT1 [UHD Graphics] vendor: Hewlett-Packard
    driver: i915 v: kernel arch: Gen-12.1 bus-ID: 0000:00:02.0
  Device-2: NVIDIA GA106M [GeForce RTX 3060 Mobile / Max-Q]
    vendor: Hewlett-Packard driver: nvidia v: 560.35.03 arch: Ampere
    bus-ID: 0000:01:00.0
  Device-3: Luxvisions Innotech HP Wide Vision HD Camera driver: uvcvideo
    type: USB bus-ID: 3-6:3
  Display: wayland server: X.org v: 1.20.14 with: Xwayland v: 24.1.2
    compositor: kwin_wayland driver: X: loaded: modesetting,nvidia dri: iris
    gpu: i915,nvidia resolution: 2560x1440
  API: EGL v: 1.5 drivers: iris,nvidia,swrast platforms:
    active: gbm,wayland,x11,surfaceless,device inactive: device-2
  API: OpenGL v: 4.6.0 compat-v: 4.5 vendor: nvidia mesa v: 560.35.03
    glx-v: 1.4 direct-render: yes renderer: NVIDIA GeForce RTX 3060 Laptop
    GPU/PCIe/SSE2
  API: Vulkan v: 1.3.290 drivers: N/A surfaces: xcb,xlib,wayland devices: 3
Audio:
  Device-1: Intel Tiger Lake-H HD Audio vendor: Hewlett-Packard
    driver: sof-audio-pci-intel-tgl bus-ID: 0000:00:1f.3
  Device-2: NVIDIA GA106 High Definition Audio vendor: Hewlett-Packard
    driver: snd_hda_intel v: kernel bus-ID: 0000:01:00.1
  API: ALSA v: k6.10.6-200.fc40.x86_64 status: kernel-api
  Server-1: JACK v: 1.9.22 status: off
  Server-2: PipeWire v: 1.0.7 status: active
Network:
  Device-1: Realtek RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet
    vendor: Hewlett-Packard driver: r8169 v: kernel port: 3000
    bus-ID: 0000:2d:00.0
  IF: eno1 state: down mac: <filter>
  Device-2: Intel Wi-Fi 6E AX210/AX1675 2x2 [Typhoon Peak] driver: iwlwifi
    v: kernel bus-ID: 0000:2e:00.0
  IF: wlo1 state: up mac: <filter>
  IP v4: <filter> type: dynamic noprefixroute scope: global
  IP v6: <filter> type: noprefixroute scope: link
  WAN IP: <filter>
Bluetooth:
  Device-1: Intel AX210 Bluetooth driver: btusb v: 0.8 type: USB bus-ID: 3-7:4
  Report: btmgmt ID: hci0 rfk-id: 0 state: down bt-service: enabled,running
    rfk-block: hardware: no software: yes address: <filter> bt-v: 5.3 lmp-v: 12
RAID:
  Hardware-1: Intel Volume Management Device NVMe RAID Controller driver: vmd
    v: 0.6 bus-ID: 0000:00:0e.0
Drives:
  Local Storage: total: 1.84 TiB used: 224.3 GiB (11.9%)
  ID-1: /dev/nvme0n1 vendor: Kingston model: SNV2S1000G size: 931.51 GiB
    temp: 49.9 C
  ID-2: /dev/nvme1n1 vendor: Samsung model: MZVL21T0HCLR-00BH1
    size: 953.87 GiB temp: 45.9 C
Partition:
  ID-1: / size: 929.93 GiB used: 223.82 GiB (24.1%) fs: btrfs
    dev: /dev/nvme0n1p3
  ID-2: /boot size: 973.4 MiB used: 471.8 MiB (48.5%) fs: ext4
    dev: /dev/nvme0n1p2
  ID-3: /boot/efi size: 598.8 MiB used: 19 MiB (3.2%) fs: vfat
    dev: /dev/nvme0n1p1
  ID-4: /home size: 929.93 GiB used: 223.82 GiB (24.1%) fs: btrfs
    dev: /dev/nvme0n1p3
Swap:
  ID-1: swap-1 type: zram size: 8 GiB used: 1.43 GiB (17.9%) dev: /dev/zram0
Sensors:
  System Temperatures: cpu: N/A mobo: N/A
  Fan Speeds (rpm): cpu: 2585 fan-2: 2428
Info:
  Memory: total: 16 GiB note: est. available: 15.26 GiB used: 5.86 GiB (38.4%)
  Processes: 491 Uptime: 2h 11m Init: systemd target: graphical (5)
  Packages: 16 Compilers: gcc: 14.2.1 Shell: Bash v: 5.2.26 inxi: 3.3.34

[zsarge (Zack Sargent)]

I downgraded my Kernel to Fedora 6.8.1-300.fc40 using this guide, which has stopped the logouts for the past few days.

I still experience occasional browser freezes in Chrome, but that may be legitimate given my hardware.

[stephaneklein (Stéphane Klein)]

2 months later and I’m still suffering from the same problem .

@litemotiv you wrote here:

For that second issue a kernel driver update is needed that will likely land in kernel 6.11. This particular bug is a regression and does not occur in kernel 6.9, if you have an older kernel still installed you can boot into that to avoid it. Kernel 6.10 is affected, so for a more definitive fix you would have to wait until kernel 6.11.

I’m currently using the kernel 6.11.4-201.fc40.x86_64 and I still have the bug. Do you have any idea why? How has the situation evolved?

[yuntaz (Y.Z.)]

I’m still experiencing this display issue on Framework 13. A mitigating workaround that disables PSR is available, but the workaround likely messes up some KDE multi display functions like overview and workspace. Some errors include Firefox [4053]: Couldn't map window 0x7f491ff54ae0 as subsurface because its parent is not mapped.

[litemotiv (Olivier)]

@stephaneklein

I’m currently using the kernel 6.11.4-201.fc40.x86_64 and I still have the bug. Do you have any idea why? How has the situation evolved?

I’m still on kernel 6.9 myself Stéphane, i’ll see this weekend if i can find some more information about the current situation.

[stephaneklein (Stéphane Klein)]

@yuntaz Are you sure your bug is the same as the one discussed in this thread?

[litemotiv (Olivier)]

@stephaneklein i’ve upgraded to Fedora 41 yesterday which carries kernel 6.11.5-300 and mesa 24.2.4-1 and these versions seem to work well on my system. I have not experienced any issues since upgrading. Hopefully this could also improve your existing situation, please let us know once you have upgraded yourself.

[stephaneklein (Stéphane Klein)]

@litemotiv I’ve just updated to version 41, after 2 hours of use, I’ve just had the bug again , Gnome Shell has quit and returned to the login screen.

I use 2 external monitor + laptop monitor.

[litemotiv (Olivier)]

Ai that’s unfortunate i haven’t had issues anymore on my system.

Can you perhaps trace it back to specific programs or tasks you are doing, or does it seem completely random?

[tqcharm (Mike B.)]

coredumpctl info shows the backtrace of the latest crash. It might be of help troubleshooting.

[stephaneklein (Stéphane Klein)]

coredumpctl info shows the backtrace of the latest crash. It might be of help troubleshooting.

@tqcharm thanks for the tip .

[stephaneklein (Stéphane Klein)]

After analysis, I think I’m the victim of several AMD driver bugs.

In this post, I’ll present just one.

At 13:13:33, when I plugged the USB-C cable from my external monitor into my laptop.

Here are the kernel logs:

13:13:33 t14s kernel: retire_capture_urb: 58 callbacks suppressed

...

13:13:54 t14s kernel: amdgpu 0000:33:00.0: [drm] REG_WAIT timeout 1us * 100 tries - dcn31_program_compbuf_size line:141
13:13:54 t14s kernel: ------------[ cut here ]------------
13:13:54 t14s kernel: WARNING: CPU: 12 PID: 30057 at drivers/gpu/drm/amd/amdgpu/../display/dc/hubbub/dcn31/dcn31_hubbub.c:151 dcn31_program_compbuf_size+0xd1/0x230 [amdgpu]

...

More complete version of these logs: kernel.log · GitHub

I found line 151 in the source code file that generates the warning: drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c#L151

Here’s the bug I found by searching on the keyword dcn31_program_compbuf_size: AMD GPU screen blanking for seconds with a warning

I run Fedora 40 on a ThinkPad T14 Gen3 - comes with AMD Ryzen 7 PRO 6850U with Radeon Graphics. I have my monitor connected via the ThinkPad dock, which is over a USB-C connection.

Yesterday, I updated to F41 with the 6.11.5-300.fc41 kernel. The screen blanking shot up to maybe 30x per minute, with 1-2s blanking each time, effectively giving me an unusable display.

I have the same hardware as this person and I seem to be experiencing the same problem.

However, I seem to have had gnome-shell crashes even without the action of plugging an external monitor into USB-C . I think I’m a victim of several bugs.

4 seconds after this warning, Discord generates a coredump :

$ coredumpctl list --since "2024-11-07"
TIME                           PID  UID  GID SIG     COREFILE EXE                           SIZE
Thu 2024-11-07 10:09:21 CET 352125 1000 1000 SIGTRAP present  /app/main/mattermost-desktop 10.6M
Thu 2024-11-07 13:13:58 CET  54929 1000 1000 SIGABRT present  /usr/lib64/discord/Discord   56.9M

Here are the contents of :

$ coredumpctl info 54929 # /usr/lib64/discord/Discord 

https://gist.github.com/stephane-klein/56b9097e1a22390ef3350757bf3132dc

And that of

$ coredumpctl info 352125 # /app/main/mattermost-desktop

https://gist.github.com/stephane-klein/1b3790d45aef6383f98341f31ea40444

Question: would you advise me to create another thread on Fedora Discussion with the title “AMD GPU screen blanking for seconds with dcn31_program_compbuf_size warning”?