On 09/01/2015 12:46 PM, Stanislav Baiduzhyi wrote:

On Tue, Sep 1, 2015 at 9:21 PM, Brandon Vincent wrote:

...

On Tue, Sep 1, 2015 at 12:17 PM, Stanislav Baiduzhyi wrote:

...

Can anyone please suggest me some way to investigate what is going on in the system? Who or what is causing that?

I'd be interested to see the output of cpufreq-info(1), if its installed.

The output is below. Also, forgot to mention, it's openSUSE 13.2 x86_64, tried both official kernel and 4.2.0, without any difference.

user@zenbook:~> cpufreq-info cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009 Report errors and bugs to http://bugs.opensuse.org, 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: 0.97 ms. hardware limits: 800 MHz - 3.00 GHz available cpufreq governors: performance, powersave current policy: frequency should be within 800 MHz and 3.00 GHz. The governor "powersave" may decide which speed to use within this range. current CPU frequency is 1.10 GHz. 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: 0.97 ms. hardware limits: 800 MHz - 3.00 GHz available cpufreq governors: performance, powersave current policy: frequency should be within 800 MHz and 3.00 GHz. The governor "powersave" may decide which speed to use within this range. current CPU frequency is 802 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: 0.97 ms. hardware limits: 800 MHz - 3.00 GHz available cpufreq governors: performance, powersave current policy: frequency should be within 800 MHz and 3.00 GHz. The governor "powersave" may decide which speed to use within this range. current CPU frequency is 897 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: 0.97 ms. hardware limits: 800 MHz - 3.00 GHz available cpufreq governors: performance, powersave current policy: frequency should be within 800 MHz and 3.00 GHz. The governor "powersave" may decide which speed to use within this range. current CPU frequency is 998 MHz.

See this thread: https://bbs.archlinux.org/viewtopic.php?id=163253 This is an known issue with inetl_pstate in some situations. See especially post #4, and perhaps give that a try. You might notice that all but one of your CPUs are running close to their lowest frequency. In short your machine is trying to save power. What little work there is to do is being done by your machine without ramping up the CPU frequency. Were it not for the heat issue, this would seem the wise thing to do, but obviously there must be brief jumps to higher frequencies. -- After all is said and done, more is said than done. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 09/01/2015 02:23 PM, Stanislav Baiduzhyi wrote:

Tried that, load average is still very high, but feels like heating dropped a little.

Load average means nothing. The heat is important, pay attention to that, and ignore load average. Look, in an intelligently scaling system, where CPUs are ramped down to save power, the load average will go up, because less cpu power is available to handle the load. (It will approach 1.xx, but seldom more, because all tasks waiting for CPU time are being handled by the available (throtteled) CPU. When it raises above one, an intelligent CPU management system will throttle up the CPU(s). There is no point in running CPUs at full frequency to accomplish a tiny amount of work. A smart system makes just enough CPU cycles available to accomplish the tasks waiting for CPU cycles. So naturally, Load Average (which only counts task waiting for a time slice) will approach 1 on a properly scaling architecture. https://en.wikipedia.org/wiki/Load_(computing)#Unix-style_load_calculation -- After all is said and done, more is said than done. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 09/01/2015 02:43 PM, Xen wrote:

On Tue, 1 Sep 2015, John Andersen wrote:

...

Load average means nothing. The heat is important, pay attention to that, and ignore load average.

So load figures are really quite meaningless then?.

Never knew that... Never knew anything about it, really... ;-). :-/.

Until they get rather high, yes, meaningless. But even ONE task that is constantly needing even small amounts of CPU cycles can bring a single CPU to Load Average of 1. When there are dozens of tasks waiting for processor time, and your load average climes to 8 or 10, the machine is busy. I've seen load averages of 50 in a multiprocessor machine which was still quite responsive. -- After all is said and done, more is said than done. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 09/01/2015 02:55 PM, Xen wrote:

the thing is supposed to have two cores,

while watching top in a terminal, press the number 1 key, and it will break out each cpu % on another line.

From what you posted before, I believe you have 4 cores.

-- After all is said and done, more is said than done. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, 2015-09-01 at 23:43 +0200, Xen wrote:

On Tue, 1 Sep 2015, John Andersen wrote:

...

Load average means nothing. The heat is important, pay attention to that, and ignore load average. So load figures are really quite meaningless then?.

No, they have a very particular meaning. That meaning is only relevant when a system is under load however; 'resting' load average on modern systems can be deceiving. -- Adam Tauno Williams mailto:awilliam@whitemice.org GPG D95ED383 Systems Administrator, Python Developer, LPI / NCLA -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-09-02 07:57, Stanislav Baiduzhyi wrote:

Hm, makes sense. But it is still 800MHz cpu fully loaded while idle, no?

Not at all according to "top". - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlXmwckACgkQja8UbcUWM1xn7wD/dlm7UIcAp5oatmTuld77JFuE 8W1E6YVJt19zvZTOYisBAIaalpF4Xiekx3B3X6ZvAANohTguIuNPcCh2mHxKGw70 =6Fe4 -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, Sep 1, 2015 at 10:06 PM, Carlos E. R. wrote:

On 2015-09-01 21:17, Stanislav Baiduzhyi wrote:

...

I've noticed that my Asus Zenbook is hot all the time, even when idle. Load average never gets lower than 1.02, but mostly is 1.10. Eliminating plasma5, X, network, audio and everything else I got to runlevel 1, with only like 7 or 8 processes running, still load average is above 1.04 and spiking to 1.18 from time to time.

And what is running? You could try "top" in a terminal at runlevel 1.

Is your machine plugged to the mains, or running on battery?

Running plugged all the time. Here is the output of 'top -bn1' right after runlevel1 boot, so the load average is still raising. https://gist.github.com/TheIndifferent/115e33fd3dacb337a697 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

There was a high load, this cannot be related to just battery since there is a software indication that it relates to the system. Right?. On Tue, 1 Sep 2015, Carlos E. R. wrote:

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256

On 2015-09-01 22:32, Stanislav Baiduzhyi wrote:

...

On Tue, Sep 1, 2015 at 10:06 PM, Carlos E. R. <> wrote:

...

...

Is your machine plugged to the mains, or running on battery?

Running plugged all the time.

Then try unplugged. The heat might come from a damaged battery that is overcharging all the time.

- -- Cheers / Saludos,

Carlos E. R.

(from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux)

iF4EAREIAAYFAlXmDFwACgkQja8UbcUWM1y4wAD8C09fA5xdKzP1cQRby7ajXl5U +u6vNyvH5QliLRyr1ocA/15juRzUu1+ONYTJH6uQ0qL+beK7O4GsHqUcjrPPyjxq =1eOk -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, Sep 1, 2015 at 10:36 PM, Carlos E. R. wrote:

On 2015-09-01 22:32, Stanislav Baiduzhyi wrote:

...

On Tue, Sep 1, 2015 at 10:06 PM, Carlos E. R. <> wrote:

...

Is your machine plugged to the mains, or running on battery? Running plugged all the time. Then try unplugged. The heat might come from a damaged battery that is overcharging all the time.

Battery is on the opposite side of the laptop, right under the touchpad, I feel when it's getting warm, but vents from fans are under the display, and are really felt. On Tue, Sep 1, 2015 at 10:55 PM, Carlos E. R. wrote:

On 2015-09-01 22:51, Xen wrote:

...

There was a high load, this cannot be related to just battery since there is a software indication that it relates to the system.

Right?.

try and find out :-)

Your load is similar to mine.

You mean that you too have constant load average higher than 1.0? On another 2 systems that I have (thinkpad and zotac zbox) the load average is normally low, and no overheating when the laptop is idle. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, Sep 1, 2015 at 11:19 PM, Carlos E. R. wrote:

On 2015-09-01 23:08, Stanislav Baiduzhyi wrote:

...

On Tue, Sep 1, 2015 at 10:36 PM, Carlos E. R. <> wrote:

...

...

Your load is similar to mine.

You mean that you too have constant load average higher than 1.0? On another 2 systems that I have (thinkpad and zotac zbox) the load average is normally low, and no overheating when the laptop is idle.

No, the link you posted shows a load the last minute of 0.56, and 0.04 during the last 15 minutes. Mine is 0.4 all the time; but the CPU load as displayed by gkrellm is less than 10% on the two cores.

Uptime is 0 minutes as well. System always starts with lower load, but raises to 1.10 in a minute. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, Sep 1, 2015 at 11:38 PM, Carlos E. R. wrote:

Well, take a paste of top after 10 minutes and post that :-)

And use a "-d 10" switch to get an average load. No need to post a link, just the first dozen lines suffices, pasted in an email.

Here it is: top - 07:47:31 up 43 min, 0 users, load average: 1.04, 1.04, 0.97 Tasks: 128 total, 1 running, 127 sleeping, 0 stopped, 0 zombie %Cpu(s): 0.0 us, 0.0 sy, 0.0 ni, 99.9 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem: 10061420 total, 171268 used, 9890152 free, 5344 buffers KiB Swap: 15728636 total, 0 used, 15728636 free. 78020 cached Mem PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 819 root 20 0 22144 2484 2184 R 6.250 0.025 0:00.01 top 1 root 20 0 36008 5960 3568 S 0.000 0.059 0:02.05 systemd 2 root 20 0 0 0 0 S 0.000 0.000 0:00.00 kthreadd 3 root 20 0 0 0 0 S 0.000 0.000 0:00.00 ksoftirqd/0 5 root 0 -20 0 0 0 S 0.000 0.000 0:00.00 kworker/0:0H 7 root 20 0 0 0 0 S 0.000 0.000 0:00.03 rcu_preempt 8 root 20 0 0 0 0 S 0.000 0.000 0:00.00 rcu_sched 9 root 20 0 0 0 0 S 0.000 0.000 0:00.00 rcu_bh 10 root 20 0 0 0 0 S 0.000 0.000 0:00.03 rcuop/0 11 root 20 0 0 0 0 S 0.000 0.000 0:00.00 rcuos/0 12 root 20 0 0 0 0 S 0.000 0.000 0:00.00 rcuob/0 13 root rt 0 0 0 0 S 0.000 0.000 0:00.00 migration/0 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-09-02 07:49, Stanislav Baiduzhyi wrote:

On Tue, Sep 1, 2015 at 11:38 PM, Carlos E. R. <> wrote:

...

Well, take a paste of top after 10 minutes and post that :-)

And use a "-d 10" switch to get an average load. No need to post a link, just the first dozen lines suffices, pasted in an email.

Here it is:

top - 07:47:31 up 43 min, 0 users, load average: 1.04, 1.04, 0.97 Tasks: 128 total, 1 running, 127 sleeping, 0 stopped, 0 zombie %Cpu(s): 0.0 us, 0.0 sy, 0.0 ni, 99.9 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem: 10061420 total, 171268 used, 9890152 free, 5344 buffers KiB Swap: 15728636 total, 0 used, 15728636 free. 78020 cached Mem

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 819 root 20 0 22144 2484 2184 R 6.250 0.025 0:00.01 top 1 root 20 0 36008 5960 3568 S 0.000 0.059 0:02.05 systemd 2 root 20 0 0 0 0 S 0.000 0.000 0:00.00 kthreadd 3 root 20 0 0 0 0 S 0.000 0.000 0:00.00 ksoftirqd/0

Well, load average is 1, but as has been explained means nothing (a really loaded machine goes up to 50), and it is not an indicator of CPU busy time. That the CPU runs at the lowest frequency indicates that it is doing nothing. And top confirms: one task, "top" precisely, taking 6% cpu (which is a lot: on my machine it takes 0.6%), but total load is 0.1% (which is curious). I don't see why your machine is hot, but it is not the CPU load. I would have a look at the output of the command "sensors" (assuming you have them configured already; if not, run "sensors-detect" first. On 2015-09-02 09:08, Carl Hartung wrote:

On Wed, 2 Sep 2015 07:49:56 +0200 Stanislav Baiduzhyi wrote: 8< - - - - - trimmed - - - - - >8

See that "1 running, 127 sleeping"? You didn't catch any 'resource hog' processes this time around. Since 'top' refreshes every N seconds (default is 3,)

But I told him to use "top -d 10", which would average the last 10 seconds. You can increase it even more, to minutes. - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlXmxL4ACgkQja8UbcUWM1xb6QD+L/m8ARSEwOa0F3pCjVqYduv5 r28rUWpadvmq0zeC0q0A/iy/saEthR1ufiDBF3xxOwCfDrBqE4YWEt9W7bMMAFec =ClYL -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org