You are not logged in.
I'm experiencing the fan spinning indefinitely after wake from suspend. Reboot doesn't fix, but a shutdown+boot fixes it.
This wasn't on 4.8 for sure, and it's still not fixed in 4.10. ([testing] enabled.)
One of my friend using X1 Carbon (rev. 4, Skylake) also confirms this issue (after upgrade to 4.9).
Not sure what's causing it, any solutions?
Offline
I can confirm this issue with my ThinkPad X1 Carbon 4th (20FB0043GE).
It doesn't always happen after awake from suspend but most of the time and I haven't found any solution yet.
I would like to rule out that this problem was caused by an UEFI update.
Could you give me your UEFI version ?
# dmidecode | grep 'BIOS Revision'
BIOS Revision: 1.23
Offline
Yeah, I also updated recently (no more rollbacks, lol):
[ 0.000000] DMI: LENOVO 20FRCTO1WW/20FRCTO1WW, BIOS N1FET49W (1.23 ) 02/08/2017
However, I doubt it's related to the problem. It happened before I upgrade to 1.23, and it didn't happen in 4.8. Two possibilities as far as I thought:
1. Sensor register corruption, leading to the thermal breakage.
2. EC register corruption, leading to the fan control breakage.
Please note that your thermal control is broken after waking from suspend. If it's spinning low, it will continue to spin low. I have got several kernel panics probably due to memory corruption with heat.
Offline
Downgraded to 4.8.14, everything is at normal.
Offline
Thanks for your reply!
I was actually not sure if that problem already happened before the UEFI update.
Also thank you for the remark about the completely broken thermal control. I was not aware of that.
Did you already raise a bug at https://bugzilla.kernel.org/?
Offline
Yes, Bug 191181.
Offline
Awesome.
Its title makes it actually hard to find.
Offline
Hi,
I think I have a similar issue with my Thinkpad t460s and Linux 4.10.3 Arch stock: sometimes (but not always) after suspend, when I resume the fan goes at top speed, even with no task and the laptop completely cool. I honestly don't remember what happened with previous kernels, but I guess there was a time a few months back when this issue did not arise, and I suspect it's the same issue that bothers you, although mine is a different model. Also I guess it's not UEFI, and I noticed you opened a bug. Is there something I could check, both in the cases where this problem arises after resume and when it doesn't, like powertop or similar?
Thank you
Offline
@nicolo - downgrade linux -> 4.8.14.
Offline
Thanks, I understand downgrading may solve it, but I'd like to use the most recent kernel in general: has the bug been fixed? or do we understand where the problem was introduced? in my case, I believe it has to do with the sensor
acpitz-virtual-0
Adapter: Virtual device
temp1: +29.0°C (crit = +128.0°C)
randomly giving 10+ degrees above this standard value, which triggers the fan.
Offline
No kernel devs are aware of this bug.
Anyway, this is the "correct" value from 4.8:
High stress:
iwlwifi-virtual-0
Adapter: Virtual device
temp1: +32.0°C
acpitz-virtual-0
Adapter: Virtual device
temp1: +76.0°C (crit = +128.0°C)
coretemp-isa-0000
Adapter: ISA adapter
Physical id 0: +74.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +74.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +73.0°C (high = +100.0°C, crit = +100.0°C)
thinkpad-isa-0000
Adapter: ISA adapter
fan1: 6276 RPM
pch_skylake-virtual-0
Adapter: Virtual device
temp1: +56.5°C
Cooled down:
iwlwifi-virtual-0
Adapter: Virtual device
temp1: +34.0°C
acpitz-virtual-0
Adapter: Virtual device
temp1: +54.0°C (crit = +128.0°C)
coretemp-isa-0000
Adapter: ISA adapter
Physical id 0: +51.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +50.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +49.0°C (high = +100.0°C, crit = +100.0°C)
thinkpad-isa-0000
Adapter: ISA adapter
fan1: 3244 RPM
pch_skylake-virtual-0
Adapter: Virtual device
temp1: +49.5°C
Offline
Do you agree that the problem consists precisely in acpitz-virtual-0 (randomly) sometimes (after suspend) reporting 20+ degrees than actual temperature, thus triggering the fan?
So they must somehow have broken control for this sensor in newer kernels.
Last edited by nicolo (2017-03-22 03:09:40)
Offline
I was suspecting Intel Sensor Hub is to the blame, and it's proven true today.
Blacklisting the module doesn't seem to work. Instead, build the kernel without CONFIG_INTEL_ISH. Everything is running normally on 4.11-rc4. (This will disable the crappy sensors support together)
Offline
Uh, there also seems to be some ISH patches in 4.11 which is not tested by me. Can anyone probably use a prebuilt kernel (which has the ISH support as default) to test it?
Offline
Bug still confirmed with
linux-image-4.11.0-041100rc3-generic_4.11.0-041100rc3.201703192331_amd64.deb
(downloaded from Ubuntu pages)
Last edited by Claudio Sacerdoti Coen (2017-03-28 15:01:46)
Offline
Thanks for your report. I have shouted to the Intel idiots, so let's wait them to figure out.
Offline
Same behaviour for rc4 (using Ubuntu package as well). I also confirm that blacklisting the *ish* modules makes no difference.
Offline
Blacklisting themselves has no effect for some reason, and you must compile without ISH. (I guess you're running a decent machine that suffers from this bug so you can get the kernel compiled in a decent amount of time)
Offline
Small addendum: hibernation does not trigger the issue. Moreover, when the fan is running at full throttle after a resume from suspend, hibernating and resuming fixes the issues... until the next suspend ofc
Offline
That's expected. However, have you tried the kernel built with ISH off in config? That should solve the problem too, but I'm too lazy to build one and currently living comfortably on 4.8.
Offline
I am getting into same problematic situation.
I have thinkpad t460, I did update bios some days ago to 1.43 and running kernel 5.15-6-arch2-1
For last 8 days (current uptime) I did suspend and wake a lot of times and there was not problem. But today it happen again. I did think that I did solve it by updating BIOS. But looks that not.
any idea?
thanks
NOTE:
temperatures from htop or i7z are ok => cca 26C
output from sensors cmd is also saying 26C for CPU but for Adapter: ACPI interface is 48C
and fan1 cca 4800RPM
Last edited by jancici (2021-12-13 13:20:06)
Offline
I'm seeing this (ACPI temperature is 48 celsius, fan blowing) recently on my ThinkPad X1 Carbon 4th gen. No recent firmware updates though.
Some recent reports from kernel bugzilla:
https://bugzilla.kernel.org/show_bug.cgi?id=196129
https://bugzilla.kernel.org/show_bug.cgi?id=214205
Though these reports are quite old already. I wonder why this started happening now.
Offline