CPU frequency stuck at minimum speed
- Login o registrati per inviare commenti
My Trisquel 8 system, with the 4.15.0-72-generic image of Linux (that of the "linux-image-generic-hwe-8.0" package), runs on top of an Intel i5-7400 processor: https://ark.intel.com/content/www/us/en/ark/products/97147/intel-core-i5-7400-processor-6m-cache-up-to-3-50-ghz.html
cpufreq sometimes works. Nevertheless, it sometimes does not, even right after a reboot: all four cores run at 800 MHz, the minimum hardware-defined speed, although I have a process that has been taking 100% of one of the core for several minutes. Even weirder, the command 'for i in 0 1 2 3; do sudo cpufreq-set -c $i -d 3GHz; done', to set a minimum frequency of 3 GHz for all four cores, seems to work, because 'cpufreq-info' then tells "current policy: frequency should be within 3.00 GHz and 3.50 GHz" ("3.00 GHz" has replaced "800 MHz"), but does not because the "current CPU frequency" remains at "800 MHz" (and my CPU-intensive task completes in more than four times the time it sometimes takes), a now "forbidden" frequency ("current policy: frequency should be within 3.00 GHz and 3.50 GHz") that 'cpufreq-info' reports anyway:
$ cpufreq-info
cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to name at domain, please.
analyzing CPU 0:
driver: intel_pstate
CPUs which run at the same hardware frequency: 0
CPUs which need to have their frequency coordinated by software: 0
maximum transition latency: 4294.55 ms.
hardware limits: 800 MHz - 3.50 GHz
available cpufreq governors: performance, powersave
current policy: frequency should be within 3.00 GHz and 3.50 GHz.
The governor "powersave" may decide which speed to use
within this range.
current CPU frequency is 800 MHz.
analyzing CPU 1:
driver: intel_pstate
CPUs which run at the same hardware frequency: 1
CPUs which need to have their frequency coordinated by software: 1
maximum transition latency: 4294.55 ms.
hardware limits: 800 MHz - 3.50 GHz
available cpufreq governors: performance, powersave
current policy: frequency should be within 3.00 GHz and 3.50 GHz.
The governor "powersave" may decide which speed to use
within this range.
current CPU frequency is 800 MHz.
analyzing CPU 2:
driver: intel_pstate
CPUs which run at the same hardware frequency: 2
CPUs which need to have their frequency coordinated by software: 2
maximum transition latency: 4294.55 ms.
hardware limits: 800 MHz - 3.50 GHz
available cpufreq governors: performance, powersave
current policy: frequency should be within 3.00 GHz and 3.50 GHz.
The governor "powersave" may decide which speed to use
within this range.
current CPU frequency is 800 MHz.
analyzing CPU 3:
driver: intel_pstate
CPUs which run at the same hardware frequency: 3
CPUs which need to have their frequency coordinated by software: 3
maximum transition latency: 4294.55 ms.
hardware limits: 800 MHz - 3.50 GHz
available cpufreq governors: performance, powersave
current policy: frequency should be within 3.00 GHz and 3.50 GHz.
The governor "powersave" may decide which speed to use
within this range.
current CPU frequency is 800 MHz.
I have not found any relevant setting in the BIOS. Has somebody here any idea to fix or bypass that problem?
The same problem apparently occurs whatever the version of the kernel: besides version 4.15, I tried versions 4.4 (Trisquel 8's default) and 5.4 (from jxself's APT Repository for Linux-libre). I am on AC-power and the battery is plugged. There is no file /sys/devices/system/cpu/cpu*/cpufreq/bios_limit. I have written 1 in /sys/module/processor/parameters/ignore_ppc: no effect. 100 in /sys/devices/system/cpu/intel_pstate/min_perf_pct: no effect either. According to lm-sensors, the temperature of the cores, about 50°C, is not high, defined as 80°C.
I am running out of ideas...
Magic Banana's 'puter is putzing along in low gear ...
Thinking inside the box: A few of years ago I was working on a project involving the storage
of grapes in a produce warehouse, wherein the grapes were in small crates protected against
mold by sulfur dioxide generated from packets of a sulfate that release the gas by reacting
with moisture in the air.
Sulfur dioxide dissolves in water to form sulfurous acid, which is rather corrosive of copper
alloys.
I was given one of those packets to find out how they work ... whereupon the USB ports on my
laptop started to misbehave.
Oxides of transition metals are semiconductors; and the diode drops across the transition
metal contacts were causing the contacts to introduce electrical noise that greatly slowed
the USB transmissions.
It might be that your 'puter's internal connections are being affected by sulfur dioxide in
urban air pollution, and so re-making the 'puter's internal connections might just work a
cure, at least for a while.
Sulfur dioxide is also formed by the combustion of fossil fuels containing sulfur compounds;
the oderant in methane-containing gas is also a sulfur compound.
I've worked in a chemistry lab. Computer chassis were covered with rust within a few months. But no observable performance drop (before the computer was totally unusable).
If I were able to re-make the internal connections of the Intel processors, I would definitely first re-connect the programming fuses to disable Boot Guard.
(Should this post be redirected to Troll Lounge?)
- Login o registrati per inviare commenti