[opensuse] Boot message and cpu speed
I am running leap 14.3. I have tried to join the kernel list, but while I did receive a welcome message, nothing else followed. So I am posing the question here. On boot, with printout enabled, I receive numerous messages about temperature above threshold. But this occurs from an off state where the system is cold! From my dmesg log I have the following. The first elements are what I receive on my boot message, but right after it says all are normal. These are consecutive lines, so I do not understand that something happened between the first set of messages and the second. Any comments? Don CPU5: Core temperature above threshold, cpu clock throttled (total events = 1) CPU4: Core temperature above threshold, cpu clock throttled (total events = 1) CPU2: Package temperature above threshold, cpu clock throttled (total events = 2) CPU3: Package temperature above threshold, cpu clock throttled (total events = 2) CPU6: Package temperature above threshold, cpu clock throttled (total events = 2) CPU1: Package temperature above threshold, cpu clock throttled (total events = 2) CPU0: Package temperature above threshold, cpu clock throttled (total events = 2) CPU7: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Package temperature above threshold, cpu clock throttled (total events = 2) CPU5: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Core temperature/speed normal CPU5: Core temperature/speed normal CPU6: Package temperature/speed normal CPU1: Package temperature/speed normal CPU0: Package temperature/speed normal CPU7: Package temperature/speed normal CPU4: Package temperature/speed normal CPU5: Package temperature/speed normal CPU3: Package temperature/speed normal CPU2: Package temperature/speed normal microcode: CPU0 microcode updated early to revision 0x24, date = 2018-01-21 Initializing cgroup subsys cpuset Initializing cgroup subsys cpu Initializing cgroup subsys cpuacct -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 05/21/2018 01:27 PM, don fisher wrote:
I am running leap 14.3. I have tried to join the kernel list, but while I did receive a welcome message, nothing else followed. So I am posing the question here.
On boot, with printout enabled, I receive numerous messages about temperature above threshold. But this occurs from an off state where the system is cold!
From my dmesg log I have the following. The first elements are what I receive on my boot message, but right after it says all are normal. These are consecutive lines, so I do not understand that something happened between the first set of messages and the second.
Any comments?
Don CPU5: Core temperature above threshold, cpu clock throttled (total events = 1) CPU4: Core temperature above threshold, cpu clock throttled (total events = 1) CPU2: Package temperature above threshold, cpu clock throttled (total events = 2) CPU3: Package temperature above threshold, cpu clock throttled (total events = 2) CPU6: Package temperature above threshold, cpu clock throttled (total events = 2) CPU1: Package temperature above threshold, cpu clock throttled (total events = 2) CPU0: Package temperature above threshold, cpu clock throttled (total events = 2) CPU7: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Package temperature above threshold, cpu clock throttled (total events = 2) CPU5: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Core temperature/speed normal CPU5: Core temperature/speed normal CPU6: Package temperature/speed normal CPU1: Package temperature/speed normal CPU0: Package temperature/speed normal CPU7: Package temperature/speed normal CPU4: Package temperature/speed normal CPU5: Package temperature/speed normal CPU3: Package temperature/speed normal CPU2: Package temperature/speed normal microcode: CPU0 microcode updated early to revision 0x24, date = 2018-01-21 Initializing cgroup subsys cpuset Initializing cgroup subsys cpu Initializing cgroup subsys cpuacct
Sorry, when I read the email I realized I had 14.3 rather than 42.3. Don -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 05/21/2018 01:27 PM, don fisher wrote:
CPU5: Core temperature above threshold, cpu clock throttled (total events = 1) CPU4: Core temperature above threshold, cpu clock throttled (total events = 1) CPU2: Package temperature above threshold, cpu clock throttled (total events = 2) CPU3: Package temperature above threshold, cpu clock throttled (total events = 2) CPU6: Package temperature above threshold, cpu clock throttled (total events = 2) CPU1: Package temperature above threshold, cpu clock throttled (total events = 2) CPU0: Package temperature above threshold, cpu clock throttled (total events = 2) CPU7: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Package temperature above threshold, cpu clock throttled (total events = 2) CPU5: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Core temperature/speed normal CPU5: Core temperature/speed normal CPU6: Package temperature/speed normal CPU1: Package temperature/speed normal CPU0: Package temperature/speed normal CPU7: Package temperature/speed normal CPU4: Package temperature/speed normal CPU5: Package temperature/speed normal CPU3: Package temperature/speed normal CPU2: Package temperature/speed normal microcode: CPU0 microcode updated early to revision 0x24, date = 2018-01-21 Initializing cgroup subsys cpuset Initializing cgroup subsys cpu Initializing cgroup subsys cpuacct
Just off hand, I suspect that your boot processes got your machine busy in a hurry and all CPUs were working hard, and temperature raised rapidly. The system throttled the cpus to protect the machine, (reducing overall machine capability), by reducing power consumption. The CPU fan kicked in bringing down the CPU temperature, so the system throttled UP the CPUs to normal speed again, now that the fan caught up and brought the chip temp down. Seems perfectly normal. -- 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 05/21/2018 02:14 PM, John Andersen wrote:
On 05/21/2018 01:27 PM, don fisher wrote:
CPU5: Core temperature above threshold, cpu clock throttled (total events = 1) CPU4: Core temperature above threshold, cpu clock throttled (total events = 1) CPU2: Package temperature above threshold, cpu clock throttled (total events = 2) CPU3: Package temperature above threshold, cpu clock throttled (total events = 2) CPU6: Package temperature above threshold, cpu clock throttled (total events = 2) CPU1: Package temperature above threshold, cpu clock throttled (total events = 2) CPU0: Package temperature above threshold, cpu clock throttled (total events = 2) CPU7: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Package temperature above threshold, cpu clock throttled (total events = 2) CPU5: Package temperature above threshold, cpu clock throttled (total events = 2) CPU4: Core temperature/speed normal CPU5: Core temperature/speed normal CPU6: Package temperature/speed normal CPU1: Package temperature/speed normal CPU0: Package temperature/speed normal CPU7: Package temperature/speed normal CPU4: Package temperature/speed normal CPU5: Package temperature/speed normal CPU3: Package temperature/speed normal CPU2: Package temperature/speed normal microcode: CPU0 microcode updated early to revision 0x24, date = 2018-01-21 Initializing cgroup subsys cpuset Initializing cgroup subsys cpu Initializing cgroup subsys cpuacct
Just off hand, I suspect that your boot processes got your machine busy in a hurry and all CPUs were working hard, and temperature raised rapidly.
The system throttled the cpus to protect the machine, (reducing overall machine capability), by reducing power consumption.
The CPU fan kicked in bringing down the CPU temperature, so the system throttled UP the CPUs to normal speed again, now that the fan caught up and brought the chip temp down.
Seems perfectly normal.
Thanks. I wasn't too worried once I looked at the dmesg output and saw that things were set to normal. I should have mentioned that none of the fans came on, and that the two sections of text I posted were consecutive in the dmesg printout. I assume not separated my more than a millisecond or so. The boot message kept me up last night, until I looked at the total output. Curious as to why that message is printed as a boot message when it is immediately corrected. Thanks again, Don -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-05-21 23:55, don fisher wrote:
On 05/21/2018 02:14 PM, John Andersen wrote:
Thanks. I wasn't too worried once I looked at the dmesg output and saw that things were set to normal. I should have mentioned that none of the fans came on, and that the two sections of text I posted were consecutive in the dmesg printout. I assume not separated my more than a millisecond or so. The boot message kept me up last night, until I looked at the total output. Curious as to why that message is printed as a boot message when it is immediately corrected.
Don't assume timing, you can get the printout with times, relative or absolute. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
On 05/21/2018 03:05 PM, Carlos E. R. wrote:
On 2018-05-21 23:55, don fisher wrote:
On 05/21/2018 02:14 PM, John Andersen wrote:
Thanks. I wasn't too worried once I looked at the dmesg output and saw that things were set to normal. I should have mentioned that none of the fans came on, and that the two sections of text I posted were consecutive in the dmesg printout. I assume not separated my more than a millisecond or so. The boot message kept me up last night, until I looked at the total output. Curious as to why that message is printed as a boot message when it is immediately corrected.
Don't assume timing, you can get the printout with times, relative or absolute.
I set this up this log file a long time ago, and do not remember how to read the deltas. My command is: dmesg -d -e -T -w >> ~/messages & which makes a log of the ring buffer to the file messages. The segment of interest is attached. I can read most of the line, but what is inside the <..> escapes me. I assume it is a delta, but do not know what the units are. The entire sequence is completed 17:37:24, + ? Don
On 2018-05-22 01:56, don fisher wrote:
On 05/21/2018 03:05 PM, Carlos E. R. wrote:
On 2018-05-21 23:55, don fisher wrote:
On 05/21/2018 02:14 PM, John Andersen wrote:
Thanks. I wasn't too worried once I looked at the dmesg output and saw that things were set to normal. I should have mentioned that none of the fans came on, and that the two sections of text I posted were consecutive in the dmesg printout. I assume not separated my more than a millisecond or so. The boot message kept me up last night, until I looked at the total output. Curious as to why that message is printed as a boot message when it is immediately corrected.
Don't assume timing, you can get the printout with times, relative or absolute.
I set this up this log file a long time ago, and do not remember how to read the deltas. My command is:
dmesg -d -e -T -w >> ~/messages &
-d, --show-delta Display the timestamp and the time delta spent between messages. If used together with --notime then only the time delta without the timestamp is printed. -e, --reltime Display the local time and the delta in human-readable format. Be aware that conversion to the local time could be inaccurate (see -T for more details). -T, --ctime Print human-readable timestamps. Be aware that the timestamp could be inaccurate! The time source used for the logs is not updated after system SUSPEND/RESUME. -w, --follow Wait for new messages. This feature is supported only on systems with a readable /dev/kmsg (since kernel 3.5.0). I don't know why use -w, this means that you have this running continuously. You could instead simply query the system log when you need them, no need to create a log of your own. For example: journalctl -b -o short-precise
which makes a log of the ring buffer to the file messages. The segment of interest is attached. I can read most of the line, but what is inside the <..> escapes me. I assume it is a delta, but do not know what the units are. The entire sequence is completed 17:37:24, + ?
Seconds, I think, from the previous line to the current one perhaps. Can't be from boot. Then this is a software problem, the entire thing is less than one second; there is no time for temperature to build up and cool down. Guessing, perhaps the software initially reads wrong the temps, then some tables are loaded, then they are read correctly. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
participants (3)
-
Carlos E. R. -
don fisher -
John Andersen