Urgent help needed for strange system freezing issue

D

Deleted member 143446

Guest
Hi again, I am having very strange system freezing issues. I post similar issue "about help me to find what eats my CPU" titled post. In there, I tought that mem and swap was the problem. After several days of using system profiler I noticed that problem is not related with memory. Since I have sent a sysprof tool's capture log with this post. I tested that:
1. I only opened a geany(a text editor) and a file
2. I save a file
3. I press space and press save file shortcut(CTRL+S) three times and system freezes.
4. It also happens when I changed window from one program to another.
I attached captured sysprof tool log(capture.syscap MD5 (capture.syscap) = a07bd786d172b2b295fdaab5833245d8 ) while doing this. Please, if anybody can help, help me to solve this strange freezing issue.
I used another Linux distro on the same hardware but no symptoms of such thing there. Only in my Debian Bullseye.
My system info :

OS : 5.10.0-8-amd64 #1 SMP Debian 5.10.46-4 (2021-08-03) x86_64 GNU/Linux
CPU : Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz
Memory : 2 GB
Graphics Card : ATI Radeon HD 4350
Thanks.
 


1. What was the other distribution and what kernel was used there?
2. When it freezes are you able to switch to a different tty(by pressing ctrl+alt+f3 and then logging in) and then run dmesg to see what output and errors you get there?
3. Is the firmware-amd-grapics package installed?
Without this package installed, poor 2D/3D performance is commonly experienced. Some GPUs may require firmware to function properly at all.
4. I installed a vm with Debian Bullseye(11), looks like the most recent kernel for Bullseye is a kernel from 2022.
Code:
tux@debian:~$ uname -a
Linux debian 5.10.0-16-amd64 #1 SMP Debian 5.10.127-2 (2022-07-23) x86_64 GNU/Linux
Looks like your kernel is from 2021, try updating your system.
 
Last edited:
1. What was the other distribution and what kernel was used there?
2. When it freezes are you able to switch to a different tty and then run dmesg to see what output and errors you get there?
3. Is the firmware-amd-grapics package installed?

4. I installed a vm with Debian Bullseye(11), looks like the most recent kernel for Bullseye is a kernel from 2022.
Code:
tux@debian:~$ uname -a
Linux debian 5.10.0-16-amd64 #1 SMP Debian 5.10.127-2 (2022-07-23) x86_64 GNU/Linux
Looks like your kernel is from 2021, try updating your system.
1. I used Slackware Linux, so not Debian
2. I tried to change tty1 while high CPU usage started, and find only these lines as errors :
[ 262.201757] perf: interrupt took too long (2510 > 2500), lowering kernel.perf_event_max_sample_rate to 79500
[ 262.224018] perf: interrupt took too long (3138 > 3137), lowering kernel.perf_event_max_sample_rate to 63500
[ 262.278077] perf: interrupt took too long (3951 > 3922), lowering kernel.perf_event_max_sample_rate to 50500
[ 263.098894] perf: interrupt took too long (4984 > 4938), lowering kernel.perf_event_max_sample_rate to 40000
[ 295.858237] perf: interrupt took too long (6274 > 6230), lowering kernel.perf_event_max_sample_rate to 31750
Other error is the same as in my previous post of usb modem error(it became standard error, and even if modem is not plugged and no browser is opened, same systematic freezing occurs).
3. Yes I installed firmware-amd-graphics - the latest version from the repo
4. I do not know how can I do that :) What is kernel file, where it is , is it related with .img file , where should I get them and put it? ... can it be related with kernel update? Thanks.
 
1. 1. I used Slackware Linux, so not Debian
2. I tried to change tty1 while high CPU usage started, and find only these lines as errors :
Other error is the same as in my previous post of usb modem error(it became standard error, and even if modem is not plugged and no browser is opened, same systematic freezing occurs).
4. I do not know how can I do that :) What is kernel file, where it is , is it related with .img file , where should I get them and put it? ... can it be related with kernel update? Thanks.
1. What Kernel was Slackware using just as a comparison?
2. Try doing it again but then run this instead: dmesg | egrep "amdgpu|radeon"
4. Login as root or use sudo and run: sudo apt update && sudo apt upgrade
Also check if one of these packages are installed.
 
This looks nicer :

Code:
$ sudo iostat
Linux 5.10.0-8-amd64 (debian)     14-08-2022     _x86_64_    (1 CPU)

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           4,92    8,15    4,09   25,26    0,00   57,58

And I used your advise and got :
Code:
[    2.559754] [drm] radeon kernel modesetting enabled.
[    2.559850] radeon 0000:01:00.0: vgaarb: deactivate vga console
[    2.561298] radeon 0000:01:00.0: VRAM: 1024M 0x0000000000000000 - 0x000000003FFFFFFF (1024M used)
[    2.561303] radeon 0000:01:00.0: GTT: 1024M 0x0000000040000000 - 0x000000007FFFFFFF
[    2.561390] [drm] radeon: 1024M of VRAM memory ready
[    2.561393] [drm] radeon: 1024M of GTT memory ready.
[    2.561431] radeon 0000:01:00.0: firmware: direct-loading firmware radeon/RV710_pfp.bin
[    2.561457] radeon 0000:01:00.0: firmware: direct-loading firmware radeon/RV710_me.bin
[    2.561476] radeon 0000:01:00.0: firmware: direct-loading firmware radeon/R700_rlc.bin
[    2.561503] radeon 0000:01:00.0: firmware: direct-loading firmware radeon/RV710_smc.bin
[    2.562497] [drm] radeon: dpm initialized
[    2.562563] radeon 0000:01:00.0: firmware: direct-loading firmware radeon/RV710_uvd.bin
[    2.563869] [drm] enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0
[    2.575361] radeon 0000:01:00.0: WB enabled
[    2.575366] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000040000c00
[    2.575370] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000040000c0c
[    2.575581] radeon 0000:01:00.0: fence driver on ring 5 use gpu addr 0x000000000005c598
[    2.575671] radeon 0000:01:00.0: radeon: MSI limited to 32-bit
[    2.575720] radeon 0000:01:00.0: radeon: using MSI.
[    2.575744] [drm] radeon: irq initialized.
[    3.562486] fbcon: radeondrmfb (fb0) is primary device
[    3.588486] radeon 0000:01:00.0: [drm] fb0: radeondrmfb frame buffer device
[    3.588557] [drm] Initialized radeon 2.50.0 20080528 for 0000:01:00.0 on minor 0

I did not use Slackware for kernel comparison I used it as OS. Also test whether the same thing happens there, not. Even usb live version of Slackware works flawlessly. I am suspecting that one service or program does that. I have uploaded "capture.syscap" file but not accepted, there one can see more details. It is hard to sent every moment of capture by images here...
 
One screenshot from sysprof tool's capture :
 

Attachments

  • cpu_freeze.png
    cpu_freeze.png
    113 KB · Views: 221
Problems like this are hard to troubleshoot, it looks like do_syscall_64 ksys_read and _x64_sys_ioctl use the most of the cpu. Maybe another idea is to run top and then make it freeze to see what program is using the most cpu at the time of freezing.
 
Last edited:
I did not use Slackware for kernel comparison I used it as OS. Also test whether the same thing happens there, not. Even usb live version of Slackware works flawlessly. I am suspecting that one service or program does that.
Also It's not really a good comparison running something live vs actually installed. A good comparison would be comparing both running live, what happens when you run Debian Bullseye live?
 
Problems like this are hard to troubleshoot, it looks like do_syscall_64 ksys_read and _x64_sys_ioctl use the most of the cpu. Maybe another idea is to run top and then make it freeze to see what program is using the most cpu at the time of freezing.
How can I do that? I know how to use top but I do not know how to freeze it?
 
Actually I am running Debian Bullseye Live. I installed it into USB and using persistence mode.
 
How can I do that? I know how to use top but I do not know how to freeze it?
Have top open in one Window and then make it freeze as you described in your OP.
 
Actually I am running Debian Bullseye Live. I installed it into USB and using persistence mode.
Why not just install it normally so on a normal hard drive, constantly running updates on a usb drive tends to make them break faster. Also it would be interesting to see if installed normally if you run into the same problem or not because I'm assuming you don't update regurarly when running live with persistent usb storage.
 
Also what DE are you using for you Live setup, trying changing DE and see if it happens there?
 
Code:
    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND       
   1336 root      20   0  579340  52488  20860 R  99,9   2,6   6:19.08 Xorg           
  12503 user      20   0   10100   4032   3264 R  50,0   0,2   0:00.12 top           
      1 root      20   0  165528   8480   5256 S   0,0   0,4   0:07.81 systemd       
      2 root      20   0       0      0      0 S   0,0   0,0   0:00.00 kthreadd       
      3 root       0 -20       0      0      0 I   0,0   0,0   0:00.00 rcu_gp         
      4 root       0 -20       0      0      0 I   0,0   0,0   0:00.00 rcu_par_gp     
      6 root       0 -20       0      0      0 I   0,0   0,0   0:00.00 kworker/0:0H-+
      8 root       0 -20       0      0      0 I   0,0   0,0   0:00.00 mm_percpu_wq   
      9 root      20   0       0      0      0 S   0,0   0,0   0:00.00 rcu_tasks_rud+
     10 root      20   0       0      0      0 S   0,0   0,0   0:00.00 rcu_tasks_tra+
     11 root      20   0       0      0      0 S   0,0   0,0   0:02.98 ksoftirqd/0   
     12 root      20   0       0      0      0 I   0,0   0,0   0:15.59 rcu_sched     
     13 root      rt   0       0      0      0 S   0,0   0,0   0:00.06 migration/0   
     15 root      20   0       0      0      0 S   0,0   0,0   0:00.00 cpuhp/0       
     17 root      20   0       0      0      0 S   0,0   0,0   0:00.00 kdevtmpfs     
     18 root       0 -20       0      0      0 I   0,0   0,0   0:00.00 netns         
     19 root      20   0       0      0      0 S   0,0   0,0   0:00.00 kauditd       
     20 root      20   0       0      0      0 S   0,0   0,0   0:00.01 khungtaskd     
     21 root      20   0       0      0      0 S   0,0   0,0   0:00.00 oom_reaper     
     22 root       0 -20       0      0      0 I   0,0   0,0   0:00.00 writeback
I captured this : %99.9 CPU goes to Xorg! I am using LXDE , but I tried xfce4 as desktop two days long, nothing changed.
The strange thing is that, there was no such problem before. I did not do big changes before this freezes started. I am not even playing a computer games, just a chess ! :) :)
As I suspected, there is a problem with my Xorg ! Is there any alternative to Xorg? I suspect that my situation looks like the one in :
.There OP solved the problem by installing old mesa and Xorg but how? How can I do that? I do not know, in synaptic package manager, there is only the newest ones.

Also does my system crashes when I uninstall Xorg instantly? How can I install the old version without crashing the GUI?
 
Last edited by a moderator:
I mean X window system by GUI in my prev post.
 
Problem continues still... I added as zip the sysprof tool capture file as sysprocCapture.zip file. New screenshot from sysprof tool below.
How can I add amdgpu.runpm=0 to kernel boot option to test whether the problem is due to power management (off/on cycle) of gpu?
 

Attachments

  • highcpu.png
    highcpu.png
    116.5 KB · Views: 189
  • sysprofCapture.zip
    739.1 KB · Views: 233
Last edited by a moderator:

Members online


Top