Wierd System Slowlessness at unknown random boot

Superadmin · 2021-12-17 03:37:08

Hi Everyone !!

Yesterday and today I noticed a very weird thing going around in my laptop. Every single process in my laptop was and is running slow. I used to take around 17 seconds for my laptop to boot up now its getting over 3 minutes. Even opening up browser which used to be like click of a finger is now taking significant time. When I run htop it shows one cpu resources being used at unprecedented rate while other are at 0.3 percent but ram resources is I guess normal. I had an update which 2 days before and yesterday.
Here is the pacman logs : https://pastebin.com/C9sUX9GC

Here is the dmesg report : http://ix.io/3IDq

Here are the journal ctl of my last five boots from yesterday since today.
From today: Boot 0 : http://ix.io/3IDs
Boot 1: http://ix.io/3IDt
Boot 2: http://ix.io/3IDu
Boot 3: http://ix.io/3IDv
Boot 4: http://ix.io/3IDw
Boot 5: http://ix.io/3IDx

I have not installed anything or used any sort of cpu intensive application with casual browsing and file management.
I have no idea how this occurred on the first. But after some hours of usage the sluggishness gets removed by itself and everything is back to normal. This sluggishness symptoms only shows when opening computer after a long time like 12 hours.

Can anyone help me what this is and how can I solve this??

EDIT:
Arch linux went through a major issues lately and lot of user like me experienced such issues.
New python update has a lot of impact on our usage and everything
I found out that auto-cpufreq was broken. I don't know if that was what brought hiccup in my system but I rebuild it again and now its working fine
I for now will leave this post solved and would report if the system sluggishness is seen once again.

Edit2: The problem again came up in my system. No the problem didn't lie in auto-cpu freq it was in something else. Will somebody help me find out what the problem lied in
Here is the journalctl of boot0: http://ix.io/3IOH
There is something I noticed in journal

kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53  hero kernel: ACPI: \_SB_.PLTF.C001: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C002: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C003: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C004: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C005: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C006: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C007: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C008: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C009: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C00A: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C00B: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C00C: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C00D: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C00E: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order
Dec 19 09:28:53 hero kernel: ACPI: \_SB_.PLTF.C00F: Found 3 idle states
Dec 19 09:28:53 hero kernel: ACPI: FW issue: working around C-state latencies out of order

I don't know if this is or not but also I am posting this

Have a great day

Last edited by Superadmin (2021-12-30 13:31:22)

Superadmin · 2021-12-19 04:17:44

Will anyone help me figure out where the problem lied in?
I am just unknown about the ways to figure out where the problem lied in

May the symptoms were not shown when I had not updated my system.
when an update comes I run into troubles.
here is the pacman logs: https://pastebin.com/6W8uzMVc

seth · 2021-12-19 08:44:41

The pacman log excerpts you keep posting are from september, you don't seem to patch the cpu microcode early, your GPUs are both late (no early kms) and the hybrid configuration seems off - especially there're issue w/ nvidia being loaded by some python script and that alone takes ~11 seconds

Dec 19 09:29:17 superadmin python3[568]: [15258] INFO: Loading module nvidia
Dec 19 09:29:28 superadmin kernel: nvidia: module license 'NVIDIA' taints kernel.

Also

When I run htop it shows one cpu resources being used at unprecedented rate

don't paraphrase - it would certainly be helpful to qunatify "unprecedented" and also by which process.
Likewise

This sluggishness symptoms only shows when opening computer after a long time like 12 hours.

What does that actually mean? "open"? Resume from suspend or hibernation or after it's been laying in the sun sauna for 12h straight…
Is there another OS (particularily a hibernating windows) on that machine?

Finally

Every single process in my laptop was and is running slow. I used to take around 17 seconds for my laptop to boot up now its getting over 3 minutes. Even opening up browser which used to be like click of a finger is now taking significant time.
…
But after some hours of usage the sluggishness gets removed by itself and everything is back to normal.

This isn't really isolating the CPU as cause - you could face some long running I/O task (file tracker, database update) that also charges a cpu core (because it looks like the system is encrypted)

time echo "scale=5000; a(1)*4" | bc -l

will charge your CPU a bit w/o relying on IO or (too much) RAM so you could compare it's performance in the slow and fast context to see whether it's actually CPU related.

Superadmin · 2021-12-20 00:36:02

seth wrote:

The pacman log excerpts you keep posting are from september,

My Bad
Here is the one: https://pastebin.com/hGw9ZnJG

seth wrote:

you don't seem to patch the cpu microcode early, your GPUs are both late (no early kms) and the hybrid configuration seems off - especially there're issue w/ nvidia being loaded by some python script and that alone takes ~11 seconds

I use optimus manager to control this hybrid graphics card (AMD and Nvidia). How do I patch this cpu microcode early and make GPU to be loaded early in kms?

seth wrote:

What does that actually mean? "open"? Resume from suspend or hibernation or after it's been laying in the sun sauna for 12h straight…
Is there another OS (particularily a hibernating windows) on that machine?

I always shutdown my laptop every time. I rarely use sleep and hibernation is off grid for me. There is no any other operating system here.

seth wrote:

time echo "scale=5000; a(1)*4" | bc -l

time echo "scale=5000; a(1)*4" | bc -l
3.141592653589793238462643383279502884197169399375105820974944592307\
81640628620899862803482534211706798214808651328230664709384460955058\
22317253594081284811174502841027019385211055596446229489549303819644\
28810975665933446128475648233786783165271201909145648566923460348610\
45432664821339360726024914127372458700660631558817488152092096282925\
40917153643678925903600113305305488204665213841469519415116094330572\
70365759591953092186117381932611793105118548074462379962749567351885\
75272489122793818301194912983367336244065664308602139494639522473719\
07021798609437027705392171762931767523846748184676694051320005681271\
45263560827785771342757789609173637178721468440901224953430146549585\
37105079227968925892354201995611212902196086403441815981362977477130\
99605187072113499999983729780499510597317328160963185950244594553469\
08302642522308253344685035261931188171010003137838752886587533208381\
42061717766914730359825349042875546873115956286388235378759375195778\
18577805321712268066130019278766111959092164201989380952572010654858\
63278865936153381827968230301952035301852968995773622599413891249721\
77528347913151557485724245415069595082953311686172785588907509838175\
46374649393192550604009277016711390098488240128583616035637076601047\
10181942955596198946767837449448255379774726847104047534646208046684\
25906949129331367702898915210475216205696602405803815019351125338243\
00355876402474964732639141992726042699227967823547816360093417216412\
19924586315030286182974555706749838505494588586926995690927210797509\
30295532116534498720275596023648066549911988183479775356636980742654\
25278625518184175746728909777727938000816470600161452491921732172147\
72350141441973568548161361157352552133475741849468438523323907394143\
33454776241686251898356948556209921922218427255025425688767179049460\
16534668049886272327917860857843838279679766814541009538837863609506\
80064225125205117392984896084128488626945604241965285022210661186306\
74427862203919494504712371378696095636437191728746776465757396241389\
08658326459958133904780275900994657640789512694683983525957098258226\
20522489407726719478268482601476990902640136394437455305068203496252\
45174939965143142980919065925093722169646151570985838741059788595977\
29754989301617539284681382686838689427741559918559252459539594310499\
72524680845987273644695848653836736222626099124608051243884390451244\
13654976278079771569143599770012961608944169486855584840635342207222\
58284886481584560285060168427394522674676788952521385225499546667278\
23986456596116354886230577456498035593634568174324112515076069479451\
09659609402522887971089314566913686722874894056010150330861792868092\
08747609178249385890097149096759852613655497818931297848216829989487\
22658804857564014270477555132379641451523746234364542858444795265867\
82105114135473573952311342716610213596953623144295248493718711014576\
54035902799344037420073105785390621983874478084784896833214457138687\
51943506430218453191048481005370614680674919278191197939952061419663\
42875444064374512371819217999839101591956181467514269123974894090718\
64942319615679452080951465502252316038819301420937621378559566389377\
87083039069792077346722182562599661501421503068038447734549202605414\
66592520149744285073251866600213243408819071048633173464965145390579\
62685610055081066587969981635747363840525714591028970641401109712062\
80439039759515677157700420337869936007230558763176359421873125147120\
53292819182618612586732157919841484882916447060957527069572209175671\
16722910981690915280173506712748583222871835209353965725121083579151\
36988209144421006751033467110314126711136990865851639831501970165151\
16851714376576183515565088490998985998238734552833163550764791853589\
32261854896321329330898570642046752590709154814165498594616371802709\
81994309924488957571282890592323326097299712084433573265489382391193\
25974636673058360414281388303203824903758985243744170291327656180937\
73444030707469211201913020330380197621101100449293215160842444859637\
66983895228684783123552658213144957685726243344189303968642624341077\
32269780280731891544110104468232527162010526522721116603966655730925\
47110557853763466820653109896526918620564769312570586356620185581007\
29360659876486117910453348850346113657686753249441668039626579787718\
55608455296541266540853061434443185867697514566140680070023787765913\
44017127494704205622305389945613140711270004078547332699390814546646\
45880797270826683063432858785698305235808933065757406795457163775254\
20211495576158140025012622859413021647155097925923099079654737612551\
76567513575178296664547791745011299614890304639947132962107340437518\
95735961458901938971311179042978285647503203198691514028708085990480\
10941214722131794764777262241425485454033215718530614228813758504306\
33217518297986622371721591607716692547487389866549494501146540628433\
66393790039769265672146385306736096571209180763832716641627488880078\
69256029022847210403172118608204190004229661711963779213375751149595\
01566049631862947265473642523081770367515906735023507283540567040386\
74351362222477158915049530984448933309634087807693259939780541934144\
73774418426312986080998886874132604720

real	0m37.389s
user	0m37.379s
sys	0m0.003s

I don't think I have used any encryption system excepts for the one KDE uses by default to store passwords and others.
I will provide the output of the above code when I see the sluggishness again(to compare the results). I am confused weather or not to mark this thread solved or not.

This sluggishness refers to eternal booting time where you will be able to see single fraction of processes like "Welcome To Grub" for particularly 3 sec long.
 Before I didn't even noticed that even existed and login screen manager takes longer time to be loaded. 
Every application are loaded slowly as if my SSD got replaced with an HDD. Terminal, Browser, Even response time after clicking the wifi bluetooth button are so slow. 
I have touchegg and the gestures are responded only after 3 sec late with transition effect that I have never seen that slow and sluggish.

thank you seth (truly).

Sry for everything

Have a great day

Last edited by Superadmin (2021-12-20 01:18:44)

seth · 2021-12-20 07:55:14

https://wiki.archlinux.org/title/Microcode

3.1415926535897932…

Yeah, I know - https://en.wikipedia.org/wiki/Pi

I don't think I have used any encryption system

your journal wrote:

Dec 17 08:40:59 superadmin kernel: EXT4-fs (dm-0): mounted filesystem with ordered data mode. Opts: (null). Quota mode: none.

lsblk -f

Edit: also please compare

glxinfo -B

in fast and slow case.

Last edited by seth (2021-12-20 07:56:33)

Superadmin · 2021-12-21 02:47:05

It happened again today just now..... this is the boot that showed the same thing again
Here is the journalctl : http://ix.io/3J0c

The symptoms still is persistent at unknown boot. It shows up sometime..... and sometime it does not..... But the time period in which it shows up is 12 hours typically.
( I always shutdown my system ; and rarely use sleep.)

Wow !!
I did (time echo "scale") thing and there is no answer till now.....

$ time echo "scale=5000; a(1)*4" | bc -l
3.141592........
 
real	6m24.848s
user	6m24.419s
sys	0m0.144s

lsblk -f

$ lsblk -f
NAME                  FSTYPE      FSVER    LABEL UUID                                   FSAVAIL FSUSE% MOUNTPOINTS
nvme0n1                                                                                                
├─nvme0n1p1           vfat        FAT32          3A72-B173                                             
└─nvme0n1p2           LVM2_member LVM2 001       cVHLaP-9TGI-reRF-6fJr-Zwfc-OAdY-Wevn4q                
  └─volgroup0-lv_root ext4        1.0            7e645fba-0fac-4fda-a6b4-391204510eed    395.1G    10% /

glxinfo -B

$ glxinfo -B
name of display: :0
display: :0  screen: 0
direct rendering: Yes
Extended renderer info (GLX_MESA_query_renderer):
    Vendor: AMD (0x1002)
    Device: AMD RENOIR (DRM 3.42.0, 5.15.8-arch1-1, LLVM 13.0.0) (0x1636)
    Version: 21.3.1
    Accelerated: yes
    Video memory: 512MB
    Unified memory: no
    Preferred profile: core (0x1)
    Max core profile version: 4.6
    Max compat profile version: 4.6
    Max GLES1 profile version: 1.1
    Max GLES[23] profile version: 3.2
Memory info (GL_ATI_meminfo):
    VBO free memory - total: 60 MB, largest block: 60 MB
    VBO free aux. memory - total: 2943 MB, largest block: 2943 MB
    Texture free memory - total: 60 MB, largest block: 60 MB
    Texture free aux. memory - total: 2943 MB, largest block: 2943 MB
    Renderbuffer free memory - total: 60 MB, largest block: 60 MB
    Renderbuffer free aux. memory - total: 2943 MB, largest block: 2943 MB
Memory info (GL_NVX_gpu_memory_info):
    Dedicated video memory: 512 MB
    Total available memory: 3584 MB
    Currently available dedicated video memory: 60 MB
OpenGL vendor string: AMD
OpenGL renderer string: AMD RENOIR (DRM 3.42.0, 5.15.8-arch1-1, LLVM 13.0.0)
OpenGL core profile version string: 4.6 (Core Profile) Mesa 21.3.1
OpenGL core profile shading language version string: 4.60
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile

OpenGL version string: 4.6 (Compatibility Profile) Mesa 21.3.1
OpenGL shading language version string: 4.60
OpenGL context flags: (none)
OpenGL profile mask: compatibility profile

OpenGL ES profile version string: OpenGL ES 3.2 Mesa 21.3.1
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20

I don't know I had OpenGl since I had installed nvidia-utils but I had to install mesa-utils to get the above results....
May be we need mesa package.....

Edit: I am on the same boot. And suddenly the symptoms goes away like puff
here is the time echo results

$ time echo "scale=5000; a(1)*4" | bc -l
3.1415......

real	0m37.347s
user	0m37.337s
sys	0m0.004s

I am changing the topic title, I know that I am not good in title....

Last edited by Superadmin (2021-12-21 02:59:17)

seth · 2021-12-21 09:13:09

Either there's some massive sideload hogging all cores or https://aur.archlinux.org/packages/auto-cpufreq/ underscales the frequency.
It might also be the (assumed) power supply state (eg. auto-cpufreq thinks you're on battery or you actually are on battery)

Superadmin · 2021-12-21 14:12:51

seth wrote:

Either there's some massive sideload hogging all cores or https://aur.archlinux.org/packages/auto-cpufreq/ underscales the frequency.
It might also be the (assumed) power supply state (eg. auto-cpufreq thinks you're on battery or you actually are on battery)

I have reinstalled auto-cpufreq since I don't want to give up performance for battery
Auto-cpufreq used to down scale the frequency to 1400 on battery and 2900 when on charging.... There had been no issue till date but seems like new python updates got a little epidemic.
During uninstall it showed this message....

:: Do you want to remove these packages? [Y/
(1/4) removing auto-cpufreq                        [----------------------] 100%
rm: cannot remove '/usr/bin/cpufreqctl': No such file or directory

After installation I didn't work as expected

$ sudo auto-cpufreq --install

--------------------- Deploying auto-cpufreq as a daemon ----------------------

* Turn off bluetooth on boot

* Deploy auto-cpufreq install script
Traceback (most recent call last):
  File "/usr/bin/auto-cpufreq", line 198, in <module>
    main()
  File "/usr/lib/python3.10/site-packages/click/core.py", line 1128, in __call__
    return self.main(*args, **kwargs)
  File "/usr/lib/python3.10/site-packages/click/core.py", line 1053, in main
    rv = self.invoke(ctx)
  File "/usr/lib/python3.10/site-packages/click/core.py", line 1395, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/usr/lib/python3.10/site-packages/click/core.py", line 754, in invoke
    return __callback(*args, **kwargs)
  File "/usr/bin/auto-cpufreq", line 175, in main
    deploy_daemon()
  File "/usr/lib/python3.10/site-packages/auto_cpufreq/core.py", line 350, in deploy_daemon
    shutil.copy(
  File "/usr/lib/python3.10/shutil.py", line 417, in copy
    copyfile(src, dst, follow_symlinks=follow_symlinks)
  File "/usr/lib/python3.10/shutil.py", line 254, in copyfile
    with open(src, 'rb') as fsrc:
FileNotFoundError: [Errno 2] No such file or directory: '/usr/share/auto-cpufreq/scripts/auto-cpufreq-install.sh'

It's completely uninstalled now

I will let you know asap about the changes after uninstallation.....

seth wrote:

(eg. auto-cpufreq thinks you're on battery or you actually are on battery)

I actually use laptop while on battery and also while charging so not consistent but I never had auto-cpufreq behave like this before

Lets see how the system acts afterwards

Superadmin · 2021-12-22 02:44:12

Superadmin wrote:

Lets see how the system acts afterwards

Nooooooooo
None of the issue was solved......
Journal again: http://ix.io/3J5w

$ time echo "scale=5000; a(1)*4" | bc -l
3.141592653...

real	6m22.415s
user	6m22.195s
sys	0m0.060s

I am confused what might have brought such issue

seth · 2021-12-22 08:27:42

https://wiki.archlinux.org/title/Microc … ly_loading

Though things seem to rather brak down after

Dec 22 08:11:44 superadmin kernel: acpi_cpufreq: overriding BIOS provided _PSD data

Compare

sudo tail /sys/devices/system/cpu/cpufreq/*/*

for fast and slow environments (doesn't tell about "why", but maybe at least "what")

Superadmin · 2021-12-22 13:33:38

seth wrote:

Compare
sudo tail /sys/devices/system/cpu/cpufreq/*/* 
for fast and slow environments (doesn't tell about "why", but maybe at least "what")

sudo tail /sys/devices/system/cpu/cpufreq/*/* | curl -F 'f:1=<-' ix.io
tail: error reading '/sys/devices/system/cpu/cpufreq/policy0/stats': Is a directory
....
tail: error reading '/sys/devices/system/cpu/cpufreq/policy11/stats': Is a directory
http://ix.io/3J7Y

And even after becoming root user I was denied to read files like this

# sudo tail /sys/devices/system/cpu/cpufreq/*/*/* | curl -F 'f:1=<-' ix.io
tail: cannot open '/sys/devices/system/cpu/cpufreq/policy0/stats/reset' for reading: Permission denied
.....
tail: cannot open '/sys/devices/system/cpu/cpufreq/policy9/stats/reset' for reading: Permission denied
http://ix.io/3J86

Last edited by Superadmin (2021-12-22 13:36:10)

seth · 2021-12-22 13:51:58

That's expected and we don't really care about the stats (nor to reset anything - for the moment)
Is http://ix.io/3J7Y from a fast or slow condition? (It's at the min of 1.4GHz on a max of 2.9 GHz for all but three cores that are on 2.9GHz - this seems expectable and normal)

Superadmin · 2021-12-22 14:05:37

seth wrote:

Is http://ix.io/3J7Y from a fast or slow condition? (It's at the min of 1.4GHz on a max of 2.9 GHz for all but three cores that are on 2.9GHz - this seems expectable and normal)

The output is for fast condition..... This boot had no issue...
I will let you know again.....

Last edited by Superadmin (2021-12-23 03:31:11)

Superadmin · 2021-12-23 02:41:43

Edit: 12 hours again symptoms seen... I am just tired of this now and can't even solve such issues

sudo tail /sys/devices/system/cpu/cpufreq/*/* | curl -F 'f:1=<-' ix.io
http://ix.io/3JbY

This is when my computer is slow

Do I have to install arch back from the beginning ?
I don't know how to make this right
I am learning slowly but haven't reached that level where I can make things correct

sudo journalctl | grep processor | curl -F 'f:1=<-' ix.io
http://ix.io/3Jc3

Processor frequency info from September till date
and If we look into journal it says

Dec 23 08:10:04 superadmin kernel: Unstable clock detected, switching default tracing clock to "global"
                                   If you want to keep using the local clock, then add:
                                     "trace_clock=local"
                                   on the kernel command line

May be a little info on what the cpu is of mine
This is for when everything was back to normal
I will post the cpu info when I detect some differences whilst my computer is slow

# sudo turbostat CPU
turbostat version 21.05.04 - Len Brown <lenb@kernel.org>
CPUID(0): AuthenticAMD 0x10 CPUID levels
CPUID(1): family:model:stepping 0x17:60:1 (23:96:1) microcode 0x0
CPUID(0x80000000): max_extended_levels: 0x80000020
CPUID(1): SSE3 MONITOR - - - TSC MSR - HT -
CPUID(6): APERF, No-TURBO, No-DTS, No-PTM, No-HWP, No-HWPnotify, No-HWPwindow, No-HWPepp, No-HWPpkg, No-EPB
CPUID(7): No-SGX
RAPL: 234 sec. Joule Counter Range, at 280 Watts
/dev/cpu_dma_latency: 2000000000 usec (default)
current_driver: acpi_idle
current_governor: menu
current_governor_ro: menu
cpu0: POLL: CPUIDLE CORE POLL IDLE
cpu0: C1: ACPI FFH MWAIT 0x0
cpu0: C2: ACPI IOPORT 0x414
cpu0: C3: ACPI IOPORT 0x415
cpu0: cpufreq driver: acpi-cpufreq
cpu0: cpufreq governor: schedutil
cpufreq boost: 1
cpu0: MSR_RAPL_PWR_UNIT: 0x000a1003 (0.125000 Watts, 0.000015 Joules, 0.000977 sec.)
turbostat: exec CPU: No such file or directory
0.006510 sec
Core	CPU	Avg_MHz	Busy%	Bzy_MHz	TSC_MHz	IPC	IRQ	POLL	C1	C2	C3	POLL%	C1%	C2%	C3%	CorWatt	PkgWatt
-	-	205	12.95	1582	2956	0.73	61	0	24	1	33	0.00	7.51	0.66	82.32	1.11	3.97
0	0	632	25.54	2476	2899	0.34	10	0	14	1	6	0.00	62.96	10.39	4.44	0.25	3.84
0	1	131	5.18	2536	2886	0.31	3	0	0	0	2	0.00	0.00	0.00	95.27
1	2	82	6.44	1273	2895	0.41	2	0	0	0	1	0.00	0.00	0.00	94.12	0.04
1	3	102	7.87	1298	2892	0.39	5	0	1	0	3	0.00	6.81	0.00	87.47
2	4	72	5.67	1273	2895	0.49	2	0	0	0	1	0.00	0.00	0.00	95.23	0.05
2	5	131	8.09	1616	2919	0.29	4	0	0	0	3	0.00	0.00	0.00	94.13
3	6	140	10.92	1280	2896	0.35	4	0	4	0	2	0.00	24.83	0.00	65.38	0.08
3	7	56	4.43	1273	2895	0.63	3	0	0	0	1	0.00	0.00	0.00	96.36
4	8	95	6.88	1380	2894	0.40	3	0	1	0	1	0.00	0.79	0.00	93.48	0.06
4	9	111	7.97	1397	2895	0.35	4	0	0	0	2	0.00	0.00	0.00	93.77
5	10	310	22.60	1371	2894	1.03	2	0	1	0	1	0.00	1.41	0.00	77.06	0.12
5	11	91	6.57	1388	2894	0.37	4	0	0	0	1	0.00	0.00	0.00	94.33
6	12	430	28.85	1492	2894	1.13	4	0	0	0	4	0.00	0.00	0.00	75.40	0.16
6	13	84	5.75	1465	2894	0.41	2	0	0	0	1	0.00	0.00	0.00	95.01
7	14	136	10.18	1339	2892	0.32	5	0	0	0	3	0.00	0.00	0.00	92.14	0.30
7	15	589	43.47	1355	2895	1.44	4	0	3	0	1	0.00	20.91	0.00	37.54

Last edited by Superadmin (2021-12-23 03:31:24)

seth · 2021-12-23 08:26:29

All cores now have "scaling_cur_freq = 398662" (400 MHz, welcome to the time when the one and only Matrix movie premiered) as opposed to the "good" run where it was between 1.4GHz and 2GHz
And this seems to be because there isn't *any* /sys/devices/system/cpu/cpufreq/*/scaling_governor in the "bad" case - no idea why, though.

1. Have you enabled early µcode patches?
2.

systool -vm cpufreq

3. Can you set a governor in the slow case and does that help anything? https://wiki.archlinux.org/title/CPU_fr … _governors (you're defaulting to "schedutil")

Superadmin · 2021-12-23 13:33:36

seth wrote:

All cores now have "scaling_cur_freq = 398662" (400 MHz, welcome to the time when the one and only Matrix movie premiered) as opposed to the "good" run where it was between 1.4GHz and 2GHz
And this seems to be because there isn't *any* /sys/devices/system/cpu/cpufreq/*/scaling_governor in the "bad" case - no idea why, though.

I don't know why
Is that very bad?
I mean I am starting to get anxiety regarding my laptop...... I don't want this to blow up one day just because I did something

1. Have you enabled early µcode patches?

No, I haven't touched anything regarding like that
I just installed amd-ucode during installation
Then other than auto-cpufreq I didn't touch anything regarding cpu frequency and stuff

$ systool -vm cpufreq
Module = "cpufreq"

  Attributes:
    uevent              = <store method only>

  Parameters:
    default_governor    = "schedutil"
    off                 = "0"

3. Can you set a governor in the slow case and does that help anything? (you're defaulting to "schedutil")

You mean" powersave" to Run the CPU at the minimum frequency. ?
I did

cpupower frequency-set -g powersave

I will let you know asap

Last edited by Superadmin (2021-12-23 13:34:56)

seth · 2021-12-23 13:55:15

The CPU runs at an extremely low frequency - you're probably not taking damages from that.

cpufreq is loaded and w/ default settings, nothing special there.

I just installed amd-ucode during installation

Then you'll now head over to https://wiki.archlinux.org/title/Microcode and make sure to load the µcode patches early.

You mean" powersave" to Run the CPU at the minimum frequency?

No, I mean *any* governor. There's none and that's likely the reason for the low cpu frequency.

Superadmin · 2021-12-23 15:04:48

seth wrote:

Then you'll now head over to https://wiki.archlinux.org/title/Microcode and make sure to load the µcode patches early.

In the Microcode(GRUB) section it has suggested to add

echo 'Loading initial ramdisk'
initrd	/boot/cpu_manufacturer-ucode.img /boot/initramfs-linux.img

- in the /boot/grub/grub.cfg
I don't want to break the grub configuration so I am writing it at /etc/default/grub
But where am I supposed to write the above line in?

The website also says to "" Repeat it for each menu entry. ""
which I don't know what it means?

Can you help me?

seth · 2021-12-23 15:26:01

Unless you "grub-mkconfig" edits to /etc/default/grub are meaningless and the former will write /boot/grub/grub.cfg
The wiki also states that "grub-mkconfig" will actually add the microcode automagically (but for the fallback images)

Superadmin · 2021-12-24 02:10:34

After 12 Hours the problem was again noticed

The frequency was downclocked to 400Hz which was mind blowing and worrysome

$ sudo cpupower monitor
    | Mperf              || Idle_Stats                
 CPU| C0   | Cx   | Freq  || POLL | C1   | C2   | C3    
   0|  3.21| 96.79|   399||  0.00|  0.01|  0.00| 97.35
   1|  2.16| 97.84|   399||  0.00|  0.11|  0.02| 97.58
   2|  3.42| 96.58|   399||  0.00|  0.00|  0.00| 97.01
   3|  4.86| 95.14|   399||  0.00|  0.00|  0.00| 95.31
   4|  1.64| 98.36|   399||  0.00|  0.00|  0.00| 98.61
   5|  4.23| 95.77|   399||  0.00|  0.05|  0.00| 97.28
   6|  3.07| 96.93|   399||  0.00|  0.00|  0.00| 97.53
   7|  0.84| 99.16|   399||  0.00|  0.00|  0.00| 99.28
   8|  0.18| 99.82|   399||  0.00|  0.00|  0.00| 99.85
   9|  0.05| 99.95|   399||  0.00|  0.00|  0.00| 99.92
  10|  0.90| 99.10|   399||  0.00|  0.53|  0.00| 98.73
  11|  0.26| 99.74|   399||  0.00|  0.00|  0.00| 99.75
  12|  0.74| 99.26|   399||  0.00|  0.00|  0.00| 99.37
  13|  0.84| 99.16|   399||  0.00|  0.00|  0.00| 99.28
  14|  0.29| 99.71|   399||  0.00|  0.00|  0.00| 99.76
  15|  0.91| 99.09|   399||  0.00|  0.00|  0.00| 99.17

I had set the value of governer to performance....(I can assure you that it is in performance mode)

Yet the frequency had no any changes to it

I also have a question but I would like to ask that at the end when the issue is solved

Last edited by Superadmin (2021-12-24 04:08:07)

Superadmin · 2021-12-24 02:38:44

seth wrote:

Unless you "grub-mkconfig" edits to /etc/default/grub are meaningless and the former will write /boot/grub/grub.cfg
The wiki also states that "grub-mkconfig" will actually add the microcode automagically (but for the fallback images)

In /boot/grub/grub.cfg
there are if..else condition in the header section
then there are menu entry
like this

if...
....
### BEGIN /etc/grub.d/10_linux ###
menuentry 'Arch Linux' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-simple-7e645fba-0fac-4fda-a6b4-391204510eed' {
        load_video
        set gfxpayload=keep
....
         linux   /boot/vmlinuz-linux root=/dev/mapper/volgroup0-lv_root rw  loglevel=3 quiet acpi_backlight=native
        echo    'Loading initial ramdisk ...'
        initrd  /boot/amd-ucode.img /boot/initramfs-linux.img

....
submenu 'Advanced options for Arch Linux' $menuentry_id_option 'gnulinux-advanced-7e645fba-0fac-4fda-a6b4-391204510eed' {
        menuentry 'Arch Linux, with Linux linux' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-linux-advanced-7e645fba-0fac-4fda-a6b4-391204510eed' {
                load_video
.....
        linux   /boot/vmlinuz-linux root=/dev/mapper/volgroup0-lv_root rw  loglevel=3 quiet acpi_backlight=native
        echo    'Loading initial ramdisk ...'
        initrd  /boot/amd-ucode.img /boot/initramfs-linux.img

....
menuentry 'Arch Linux, with Linux linux (fallback initramfs)' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-linux-fallback-7e645fba-0fac-4fda-a6b4-391204510eed' {
                load_video
....
         linux   /boot/vmlinuz-linux root=/dev/mapper/volgroup0-lv_root rw  loglevel=3 quiet acpi_backlight=native
        echo    'Loading initial ramdisk ...'
        initrd  /boot/amd-ucode.img /boot/initramfs-linux.img
....

There is already the line in each menu entry..... That means I already have early loading setup of ucode

Last edited by Superadmin (2021-12-24 02:40:09)

seth · 2021-12-24 07:53:24

I can assure you that it is in performance mode

By looking at /sys/devices/system/cpu/cpufreq/*/scaling_governor ?
Because the original situation was that there was none and while "I can assure you that it is in performance mode" is nice and fine, being able to point to the contents of that file would be even more assuring.

That means I already have early loading setup of ucode

If you ran grub-mkconfig after installing the µcode package, that's expectable - it should also now be reflected in your system journal.

Can you trigger the slowness when keeping the system busy?
Add

processor.max_cstate=1

to the kernel parameters.

Superadmin · 2021-12-24 14:02:09

seth wrote:

I can assure you that it is in performance mode
By looking at /sys/devices/system/cpu/cpufreq/*/scaling_governor ?

There are policies0, policies1 etc. in that location with scaling_governor all set to powersave.... I shifted to powersave to save some battery

# nano /sys/devices/system/cpu/cpufreq/*/scaling_governor
/sys/devices/system/cpu/cpufreq/policy0/scaling_governor
powersave
......
/sys/devices/system/cpu/cpufreq/policy9/scaling_governor
powersave

Here is the journalctl: http://ix.io/3JlE

seth wrote:

Can you trigger the slowness when keeping the system busy?
Add
processor.max_cstate=1
to the kernel parameters.

Sorry I didn't get you a little bit
Am I supposed to trigger that slowness ?
Do let me know
Have a great day seth !!

Last edited by Superadmin (2021-12-24 14:05:59)

seth · 2021-12-24 14:12:44

There are policies0, policies1 etc. in that location with scaling_governor all set to powersave

While the system is slow and the scaling_cur_freq is 400MHz??

Am I supposed to trigger that slowness ?

The hope is that enforcing a lower c-state prevents the performance issues, so add that kernel parameter and wait and see whether the system still comes up slow.

Have a great day seth !!

https://www.youtube.com/watch?v=7z3kVCiGfhk

Superadmin · 2021-12-24 14:23:03

seth wrote:

There are policies0, policies1 etc. in that location with scaling_governor all set to powersave
While the system is slow and the scaling_cur_freq is 400MHz??
Am I supposed to trigger that slowness ?
The hope is that enforcing a lower c-state prevents the performance issues, so add that kernel parameter and wait and see whether the system still comes up slow.

No, the slowness was not seen.....
I guess it is because I have also set the governer to powersave from boot via cpupower systemd unit

seth wrote:

https://www.youtube.com/watch?v=7z3kVCiGfhk

And btw that link was dead
try this one: https://www.youtube.com/watch?v=Z44ZPjYpcv4

But genuinely
Merry Christmas with nothing in between

Edit: I will mark this as solved since I haven't found the slowness in last 24 hours

Last edited by Superadmin (2021-12-25 16:00:30)

Arch Linux

#1 2021-12-17 03:37:08

Wierd System Slowlessness at unknown random boot

#2 2021-12-19 04:17:44

Re: Wierd System Slowlessness at unknown random boot

#3 2021-12-19 08:44:41

Re: Wierd System Slowlessness at unknown random boot

#4 2021-12-20 00:36:02

Re: Wierd System Slowlessness at unknown random boot

#5 2021-12-20 07:55:14

Re: Wierd System Slowlessness at unknown random boot

#6 2021-12-21 02:47:05

Re: Wierd System Slowlessness at unknown random boot

#7 2021-12-21 09:13:09

Re: Wierd System Slowlessness at unknown random boot

#8 2021-12-21 14:12:51

Re: Wierd System Slowlessness at unknown random boot

#9 2021-12-22 02:44:12

Re: Wierd System Slowlessness at unknown random boot

#10 2021-12-22 08:27:42

Re: Wierd System Slowlessness at unknown random boot

#11 2021-12-22 13:33:38

Re: Wierd System Slowlessness at unknown random boot

#12 2021-12-22 13:51:58

Re: Wierd System Slowlessness at unknown random boot

#13 2021-12-22 14:05:37

Re: Wierd System Slowlessness at unknown random boot

#14 2021-12-23 02:41:43

Re: Wierd System Slowlessness at unknown random boot

#15 2021-12-23 08:26:29

Re: Wierd System Slowlessness at unknown random boot

#16 2021-12-23 13:33:36

Re: Wierd System Slowlessness at unknown random boot

#17 2021-12-23 13:55:15

Re: Wierd System Slowlessness at unknown random boot

#18 2021-12-23 15:04:48

Re: Wierd System Slowlessness at unknown random boot

#19 2021-12-23 15:26:01

Re: Wierd System Slowlessness at unknown random boot

#20 2021-12-24 02:10:34

Re: Wierd System Slowlessness at unknown random boot

#21 2021-12-24 02:38:44

Re: Wierd System Slowlessness at unknown random boot

#22 2021-12-24 07:53:24

Re: Wierd System Slowlessness at unknown random boot

#23 2021-12-24 14:02:09

Re: Wierd System Slowlessness at unknown random boot

#24 2021-12-24 14:12:44

Re: Wierd System Slowlessness at unknown random boot

#25 2021-12-24 14:23:03

Re: Wierd System Slowlessness at unknown random boot

Board footer