Random second-long keyboard freezes on last batch Framework 16

Framework Laptop 16 Linux
1

Hello!

I received my Framework 16 laptop with integrated GPU earlier in July. I run Arch Linux with Gnome 46. I’m quite happy with it, except a weird keyboard lag that appears now and then. It is not restricted to typing, for example using alt + tab to change windows on Gnome 46 also suffers from the lag. When I type (this message for example) I need to wait unpractical amounts of time. It happens randomly, sometimes I can work for an hour with no noticeable issue, some other time it’s really unhelpful. I have tried on battery and on wired power with no difference. I am not able to try an external keyboard because I don’t have any. When my keyboard and mouse are frozen I have noticed that the time display by Gnome is still being updated second after second. As I type this message I’ve seen several times a Gnome infobox asking me whether I wished to kill Firefox that had become unresponsive. It happens in the Gnome searchbox (after pressing the super key), in Firefox, in Emacs, but much less in Gedit for some reason.

Mouse also is affected, but I guess focusing on the keyboard will already provide enough details.

I have included several log files. I have also included the stdout of me installing amd-ucode, where several warnings have appeared. Hopefully it might help?

Would you have any cue on how to solve this issue?

Best,
X

Maybe related: [TRACKING] Freezing Arch Linux AMD - #35 by Enrico

2

If you happen to run powertop or similar, exclude the keyboard from power saving.

3

open a console and check the kernel ring buffer and the system journal. Both can be seen merged with:

sudo journalctl -f

Note the sudo, as without you will not see everything.
When this happens again, check the console. It should show you something related.

4

Thanks for your very early replies!! I’m much obliged.

It took me some time to type to try and include the logfiles, I ended up putting them in a GitHub Gist. Hopefully it will give some colour to this issue.

I do not use powertop:

$ pacman -Ss powertop 
extra/powertop 2.15-2
    A tool to diagnose issues with power consumption and power management
1 Like
5

Ha, here is the journalctl output executed as root. I can see some random stuff about bijiben an app from Gnome (I don’t use bijiben) but also some issue related to i2c:

Aug 06 12:57:29 C40C04 kernel: PM: suspend entry (s2idle)
Aug 06 12:57:29 C40C04 kernel: Filesystems sync: 0.023 seconds
Aug 06 13:49:21 C40C04 kernel: Freezing user space processes
Aug 06 13:49:21 C40C04 kernel: Freezing user space processes completed (elapsed 0.002 seconds)
Aug 06 13:49:21 C40C04 kernel: OOM killer disabled.
Aug 06 13:49:21 C40C04 kernel: Freezing remaining freezable tasks
Aug 06 13:49:21 C40C04 kernel: Freezing remaining freezable tasks completed (elapsed 0.001 seconds)
Aug 06 13:49:21 C40C04 kernel: printk: Suspending console(s) (use no_console_suspend to debug)
Aug 06 13:49:21 C40C04 kernel: ACPI: EC: interrupt blocked
Aug 06 13:49:21 C40C04 kernel: ACPI: EC: interrupt unblocked
Aug 06 13:49:21 C40C04 kernel: nvme nvme0: 16/0/0 default/read/poll queues
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: MES failed to respond to msg=MISC (WAIT_REG_MEM)
Aug 06 13:49:21 C40C04 kernel: [drm:amdgpu_mes_reg_write_reg_wait [amdgpu]] *ERROR* failed to reg_write_reg_wait
Aug 06 13:49:21 C40C04 kernel: [drm] PCIE GART of 512M enabled (table at 0x000000801FD00000).
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: SMU is resuming...
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: SMU is resumed successfully!
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: [drm:jpeg_v4_0_hw_init [amdgpu]] JPEG decode initialized successfully.
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring jpeg_dec uses VM inv eng 1 on hub 8
Aug 06 13:49:21 C40C04 kernel: amdgpu 0000:c1:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 13 on hub 0
Aug 06 13:49:21 C40C04 kernel: [drm] ring gfx_32772.1.1 was added
Aug 06 13:49:21 C40C04 kernel: [drm] ring compute_32772.2.2 was added
Aug 06 13:49:21 C40C04 kernel: [drm] ring sdma_32772.3.3 was added
Aug 06 13:49:21 C40C04 kernel: [drm] ring gfx_32772.1.1 ib test pass
Aug 06 13:49:21 C40C04 kernel: [drm] ring compute_32772.2.2 ib test pass
Aug 06 13:49:21 C40C04 kernel: [drm] ring sdma_32772.3.3 ib test pass
Aug 06 13:49:21 C40C04 kernel: OOM killer enabled.
Aug 06 13:49:21 C40C04 kernel: Restarting tasks ... done.
Aug 06 13:49:21 C40C04 kernel: random: crng reseeded on system resumption
Aug 06 13:49:21 C40C04 kernel: PM: suspend exit
Aug 06 13:49:24 C40C04 kernel: wlan0: authenticate with 7c:c1:77:99:17:35 (local address=fc:b0:de:18:1d:63)
Aug 06 13:49:24 C40C04 kernel: wlan0: send auth to 7c:c1:77:99:17:35 (try 1/3)
Aug 06 13:49:24 C40C04 kernel: wlan0: authenticated
Aug 06 13:49:24 C40C04 kernel: wlan0: associate with 7c:c1:77:99:17:35 (try 1/3)
Aug 06 13:49:24 C40C04 kernel: wlan0: RX AssocResp from 7c:c1:77:99:17:35 (capab=0x1511 status=0 aid=6)
Aug 06 13:49:24 C40C04 kernel: wlan0: associated
Aug 06 13:49:24 C40C04 kernel: wlan0: Limiting TX power to 26 (26 - 0) dBm as advertised by 7c:c1:77:99:17:35
Aug 06 14:10:52 C40C04 kernel: bijiben-shell-s[58127]: segfault at 79d20d5b6de6 ip 000079d5c2a09f81 sp 00007fffe0d15378 error 4 in libgobject-2.0.so.0.8000.4[39f81,79d5c29dc000+37000] likely on CPU 7 (core 3, socket 0)
Aug 06 14:10:52 C40C04 kernel: Code: 00 00 00 4c 89 ce e9 6e e6 ff ff 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 48 85 ff 74 47 48 8b 07 48 85 c0 74 3f <48> 8b 00 48 3d fc 03 00 00 77 2c 48 8d 15 ad 41 02 00 48 c1 e8 02
Aug 06 14:15:47 C40C04 kernel: i2c_hid_acpi i2c-FRMW0003:00: i2c_hid_get_input: incomplete report (7/65535)
Aug 06 14:33:42 C40C04 kernel: bijiben-shell-s[78815]: segfault at 78e9b2e3abe6 ip 000078ee6d13bf81 sp 00007ffc5f9b0bc8 error 4 in libgobject-2.0.so.0.8000.4[39f81,78ee6d10e000+37000] likely on CPU 8 (core 4, socket 0)
6

Looks like the 2 last lines are the hint. hid → Human input devie. And segfault. Don’t know what the bijiben-shell-s is though

7

I uninstalled bijiben and installed Vulkan. After a restart it seems amdgpu is way less noisy in the dmesg logs and so far the lag issue hasn’t surfaced back yet:

$ sudo dmesg | grep amdgpu
[sudo] password for caocoa: 
[    2.001307] [drm] amdgpu kernel modesetting enabled.
[    2.011158] amdgpu: Virtual CRAT table created for CPU
[    2.011170] amdgpu: Topology: Add CPU node
[    2.011321] amdgpu 0000:c1:00.0: enabling device (0006 -> 0007)
[    2.015937] amdgpu 0000:c1:00.0: amdgpu: Fetched VBIOS from VFCT
[    2.015939] amdgpu: ATOM BIOS: 113-PHXGENERIC-001
[    2.018982] amdgpu 0000:c1:00.0: [drm:jpeg_v4_0_early_init [amdgpu]] JPEG decode is enabled in VM mode
[    2.088406] amdgpu 0000:c1:00.0: vgaarb: deactivate vga console
[    2.088415] amdgpu 0000:c1:00.0: amdgpu: Trusted Memory Zone (TMZ) feature enabled
[    2.088500] amdgpu 0000:c1:00.0: amdgpu: VRAM: 512M 0x0000008000000000 - 0x000000801FFFFFFF (512M used)
[    2.088503] amdgpu 0000:c1:00.0: amdgpu: GART: 512M 0x00007FFF00000000 - 0x00007FFF1FFFFFFF
[    2.088753] [drm] amdgpu: 512M of VRAM memory ready
[    2.088756] [drm] amdgpu: 31742M of GTT memory ready.
[    2.090305] amdgpu 0000:c1:00.0: amdgpu: Will use PSP to load VCN firmware
[    2.115895] amdgpu 0000:c1:00.0: amdgpu: reserve 0x4000000 from 0x8018000000 for PSP TMR
[    2.672698] amdgpu 0000:c1:00.0: amdgpu: RAS: optional ras ta ucode is not available
[    2.681041] amdgpu 0000:c1:00.0: amdgpu: RAP: optional rap ta ucode is not available
[    2.681044] amdgpu 0000:c1:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[    2.712460] amdgpu 0000:c1:00.0: amdgpu: SMU is initialized successfully!
[    2.821645] amdgpu 0000:c1:00.0: [drm:jpeg_v4_0_hw_init [amdgpu]] JPEG decode initialized successfully.
[    2.823811] kfd kfd: amdgpu: Allocated 3969056 bytes on gart
[    2.823826] kfd kfd: amdgpu: Total number of KFD nodes to be created: 1
[    2.823946] amdgpu: Virtual CRAT table created for GPU
[    2.824084] amdgpu: Topology: Add dGPU node [0x15bf:0x1002]
[    2.824086] kfd kfd: amdgpu: added device 1002:15bf
[    2.824097] amdgpu 0000:c1:00.0: amdgpu: SE 1, SH per SE 2, CU per SH 6, active_cu_number 12
[    2.824103] amdgpu 0000:c1:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[    2.824104] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[    2.824106] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[    2.824107] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[    2.824108] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[    2.824109] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[    2.824110] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[    2.824111] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[    2.824112] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[    2.824113] amdgpu 0000:c1:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[    2.824114] amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
[    2.824115] amdgpu 0000:c1:00.0: amdgpu: ring jpeg_dec uses VM inv eng 1 on hub 8
[    2.824116] amdgpu 0000:c1:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 13 on hub 0
[    2.829222] amdgpu 0000:c1:00.0: amdgpu: Runtime PM not available
[    2.830405] [drm] Initialized amdgpu 3.57.0 20150101 for 0000:c1:00.0 on minor 1
[    2.856201] fbcon: amdgpudrmfb (fb0) is primary device
[    2.856212] amdgpu 0000:c1:00.0: [drm] fb0: amdgpudrmfb frame buffer device
[    4.425640] snd_hda_intel 0000:c1:00.1: bound 0000:c1:00.0 (ops amdgpu_dm_audio_component_bind_ops [amdgpu])

I’ll let you know if it happens again. Thanks for your help!

8

Also, i2c is way less suspicious now in dmesg logs:

sudo dmesg | grep i2c
[    4.259896] hid-generic 0018:32AC:001B.0005: hidraw4: I2C HID v1.00 Device [FRMW0003:00 32AC:001B] on i2c-FRMW0003:00
[    4.271921] input: PIXA3854:00 093A:0274 Mouse as /devices/platform/AMDI0010:03/i2c-1/i2c-PIXA3854:00/0018:093A:0274.0006/input/input9
[    4.272028] input: PIXA3854:00 093A:0274 Touchpad as /devices/platform/AMDI0010:03/i2c-1/i2c-PIXA3854:00/0018:093A:0274.0006/input/input10
[    4.272114] hid-generic 0018:093A:0274.0006: input,hidraw5: I2C HID v1.00 Mouse [PIXA3854:00 093A:0274] on i2c-PIXA3854:00
[    4.285572] hid-sensor-hub 0018:32AC:001B.0005: hidraw4: I2C HID v1.00 Device [FRMW0003:00 32AC:001B] on i2c-FRMW0003:00
[    4.432358] input: PIXA3854:00 093A:0274 Mouse as /devices/platform/AMDI0010:03/i2c-1/i2c-PIXA3854:00/0018:093A:0274.0006/input/input14
[    4.432458] input: PIXA3854:00 093A:0274 Touchpad as /devices/platform/AMDI0010:03/i2c-1/i2c-PIXA3854:00/0018:093A:0274.0006/input/input15
[    4.432524] hid-multitouch 0018:093A:0274.0006: input,hidraw5: I2C HID v1.00 Mouse [PIXA3854:00 093A:0274] on i2c-PIXA3854:00

for the sake of completion:

lsmod | grep i2c 
Module                  Size  Used by
i2c_piix4              36864  0
i2c_hid_acpi           12288  0
i2c_hid                40960  1 i2c_hid_acpi
i2c_algo_bit           20480  1 amdgpu
9

However, I fail to see any causation between i2c and GPU stuff, so perhaps the log lines I fixed are completely unrelated to the first bug I described.

10

Hello, unfortunately this bug completely vanished away for a few hours, and now is back. I can see nothing in journalctl -f.

Would you have any idea where else to look at?

11

I feel like I’m getting closer.

$ libinput debug-events --verbose
-event2   DEVICE_ADDED            Video Bus                         seat0 default group1  cap:k
-event0   DEVICE_ADDED            Lid Switch                        seat0 default group2  cap:S
-event1   DEVICE_ADDED            Power Button                      seat0 default group3  cap:k
-event3   DEVICE_ADDED            Framework Laptop 16 Keyboard Module - ISO seat0 default group4  cap:k
-event4   DEVICE_ADDED            Framework Laptop 16 Keyboard Module - ISO System Control seat0 default group4  cap:k
-event5   DEVICE_ADDED            Framework Laptop 16 Keyboard Module - ISO Consumer Control seat0 default group4  cap:kp scroll-nat
-event6   DEVICE_ADDED            Framework Laptop 16 Keyboard Module - ISO Wireless Radio Control seat0 default group4  cap:k
-event7   DEVICE_ADDED            Framework Laptop 16 Keyboard Module - ISO Keyboard seat0 default group4  cap:k
-event12  DEVICE_ADDED            PIXA3854:00 093A:0274 Mouse       seat0 default group5  cap:p left scroll-nat scroll-button
-event13  DEVICE_ADDED            PIXA3854:00 093A:0274 Touchpad    seat0 default group5  cap:pg  size 124x77mm tap(dl off) left scroll-nat scroll-2fg-edge click-buttonareas-clickfinger dwt-on dwtp-on

…
[and then quite correlated with my typing]:
event13 - palm: keyboard timeout
…
event13 - PIXA3854:00 093A:0274 Touchpad: kernel bug: Touch jump detected and discarded.
See https://wayland.freedesktop.org/libinput/doc/1.26.1/touchpad-jumping-cursors.html for details

When I actually place my real palm on the touchpad I can see:

1053: event13 - palm: touch 0 (TOUCH_BEGIN), palm detected (typing)
  ... event13 - button state: touch 0 from BUTTON_STATE_NONE    event BUTTON_EVENT_IN_AREA     to BUTTON_STATE_AREA   
 event7   KEYBOARD_KEY            +72.194s      *** (-1) released
1091: event13 - button state: touch 0 from BUTTON_STATE_AREA    event BUTTON_EVENT_UP          to BUTTON_STATE_NONE   
 event7   KEYBOARD_KEY            +72.291s      *** (-1) released
1099: event13 - palm: keyboard timeout
1100: event13 - button state: touch 0 from BUTTON_STATE_NONE    event BUTTON_EVENT_IN_AREA     to BUTTON_STATE_AREA   
  ... event13 - gesture state GESTURE_STATE_NONE → GESTURE_EVENT_FINGER_DETECTED → GESTURE_STATE_UNKNOWN
1106: event13 - gesture state GESTURE_STATE_UNKNOWN → GESTURE_EVENT_HOLD_TIMEOUT → GESTURE_STATE_HOLD
-event13  GESTURE_HOLD_BEGIN      +74.461s      1
1111: event13 - gesture state GESTURE_STATE_HOLD → GESTURE_EVENT_HOLD_AND_MOTION → GESTURE_STATE_HOLD_AND_MOTION

It happens even I do absolutely not touch the touchpad in any way. It’s like the keyboard would stop responding, but then all the keys pressed at once. It happens even if I type very very lightly on the keyboard, far from the touchpad.

During the freeze time, the mouse disappears from the window, however it can still be seen if moved out of it.

The option « Disabling touchpad while typing » doesn’t change the issue if active or not. Disabling the touchpad itself doesn’t change anything either.

When looking in evtest’s output, you can see I just hit backspace at timestamp 1722968903.198551 and there is nothing special going on.

sudo evtest                  
[sudo] password for caocoa: 
No device specified, trying to scan all of /dev/input/event*
Available devices:
/dev/input/event0:      Lid Switch
/dev/input/event1:      Power Button
/dev/input/event10:     HD-Audio Generic HDMI/DP,pcm=7
/dev/input/event11:     HD-Audio Generic HDMI/DP,pcm=8
/dev/input/event12:     PIXA3854:00 093A:0274 Mouse
/dev/input/event13:     PIXA3854:00 093A:0274 Touchpad
/dev/input/event2:      Video Bus
/dev/input/event3:      Framework Laptop 16 Keyboard Module - ISO
/dev/input/event4:      Framework Laptop 16 Keyboard Module - ISO System Control
/dev/input/event5:      Framework Laptop 16 Keyboard Module - ISO Consumer Control
/dev/input/event6:      Framework Laptop 16 Keyboard Module - ISO Wireless Radio Control
/dev/input/event7:      Framework Laptop 16 Keyboard Module - ISO Keyboard
/dev/input/event8:      PC Speaker
/dev/input/event9:      HD-Audio Generic HDMI/DP,pcm=3
Select the device event number [0-13]: 7
Event: time 1722968903.161876, type 1 (EV_KEY), code 14 (KEY_BACKSPACE), value 2
Event: time 1722968903.161876, -------------- SYN_REPORT ------------
Event: time 1722968903.198551, type 1 (EV_KEY), code 14 (KEY_BACKSPACE), value 2
Event: time 1722968903.198551, -------------- SYN_REPORT ------------
Event: time 1722968903.235229, type 1 (EV_KEY), code 14 (KEY_BACKSPACE), value 2
Event: time 1722968903.235229, -------------- SYN_REPORT ------------

Anyone suffering from this bug? It makes a laptop pretty much unusable.

12

Looks like the touchpad is doing issues.
Can you disable the palm autodetection on the touchpad? (it essentially disables the touchpad when you type).
I’d check the connectors of the touchpad. Also, I’d open a support case with Framework. The techs should know more about this.

13

Thanks a lot for your kind help. In my previous message I noted:

the option « Disabling touchpad while typing » doesn’t change the issue if active or not. Disabling the touchpad itself doesn’t change anything either.

but on the other hand it is true that I don’t remember experiencing this issue when using my laptop with the lid closed with bluetooth keyboard and touchpad.

14

I know they had some issues with the firmware of the FW13 touchpad … reason I’d rather ask support for help here.

15

What sort of delay is it? In seconds, or msec or whatever. Just saying “too long” is not particularly helpful.
Possible causes:

  1. disk / storage activity.
  2. EC bugs, try forcing an EC restart.
  3. powertop or similar tool config.
  4. GPU bugs, firmware
  5. too many interrupts
  6. suspend bugs
16

Thanks James for chiming in.

I gathered more evidences here: 2024-08-07T11:23:51,319896500+02:00-dmesg.txt · GitHub

To reply to your message and your list:

On delay length and behaviour during the freeze

The delay can be anything from half a second to fifteen seconds. They can appear on any key sequence in Emacs, gedit, Firefox textareas or input fields, even Gnome UI itself.

Example 1

For example when editing a message I pressed → and got a delay. I maintained this key pushed down for 30 seconds, the caret didn’t move in the text area. Then I unpressed it: still nothing. After a couple of seconds it seems that all the keypress events were sent very quickly and the caret moved right like crazy. When I used the touchpad several times to reset its position somewhere else in the textarea it kept moving right for more than 30s.

Example 2

When editing this message I got mostly lucky, but I still had to wait several times more than several seconds at random key presses. During one freeze I pressed ctrl+alt+F3 to move to TTY 3, that happened immediately. Then I went back to Gnome UI with ctrl+alt+F2 and the freeze was still happening. Then, after a couple of seconds the keypress event were released and the text appeared.

On your list

  1. disk / storage activity: kind of non-existent
  2. EC bugs, try forcing an EC restart: do you mean embedded controller bugs? I guess they would appear in dmesg or journalctl outputs, I didn’t find any.
  3. powertop or similar tool config: I do not use any.
  4. GPU bugs, firmware: I saw several log lines, for example here:
[ 4375.206190] amdgpu 0000:c1:00.0: amdgpu: MES failed to respond to msg=MISC (WAIT_REG_MEM)
[ 4375.206193] [drm:amdgpu_mes_reg_write_reg_wait [amdgpu]] *ERROR* failed to reg_write_reg_wait
[ 4375.208511] [drm] PCIE GART of 512M enabled (table at 0x000000801FD00000).
[ 4375.208598] amdgpu 0000:c1:00.0: amdgpu: SMU is resuming...
[ 4375.212534] amdgpu 0000:c1:00.0: amdgpu: SMU is resumed successfully!
  1. too many interrupts

That’s a great point. I don’t know what a normal amount of interrupts is. I added two files with timestamps in their names to count interrupts across time:

Between these two timestamps (today 11:47:30 and 11:48:35) several freeze happened, around a second-long each. To read two datapoints:

  • For CPU3 with IRQ 6: 15545722 - 15521859 = 23863 interruptions
  • For CPU5 with IRQ 7: 595639 - 594607 = 1032 interruptions
  1. suspend bugs

I don’t know. I know I have an issue when the laptop hibernates, but I guess it’s not related. Let’s not try to fix everything at once, except if you think it might be related.

17

Thanks for this kind suggestion. I just opened a case, hopefully soon it’ll be an issue of the past.

18

Thanks all for your kind and swift help! The issue was caused by a Gnome misconfiguration. I was able to get rid of it with a mini-reset of Gnome:

$ dconf reset -f /org/gnome/

For future users I’d rather suggest to dump the dconf values first.

2 Likes
19

I tried performing Gnome mini-reset and It seems to solve the issue. But the freezes got back after a few days… It’s a bit annoying if I need to setup Gnome again every few days. Did you experience something similar ?