[SOLVED] ath10k_pci: What's the critical temperature?

alexander_linux_user · 04-14-2021, 11:48 PM

Hello, friends!

This is my very first question here, so I hope I haven't done anything wrong while posting this.

I'm concerned about whether the PCI adapter on my Dell Inspiron 3582 laptop is overheating or not. Here is what "sensors" show:

Quote:

ath10k_hwmon-pci-0200
Adapter: PCI adapter
temp1: +98.0°C

Here's what "hardinfo" shows:

Quote:

ath10k_pci/temp1 Temperature 98,00°C

It changes from around +90.0°C to +103.0°C, usually, it's around +98.0°C.

Any other adapter doesn't go higher than +60.0°C.

In terms of performance, I don't have complaints everything just works fine, except for when I play the 720p60 video on YouTube long enough, "tlp" automatically shuts pc down. And yes, the video card is integrated, no discrete video card.

The laptop itself is slightly warm and quiet while working.

Other info that might be helpful:

Code:

PCI Devices
-----------

-PCI Devices-
Host bridge		: Intel Corporation Device 31f0 (rev 03)
Signal processing controller		: Intel Corporation Celeron/Pentium Silver Processor Dynamic Platform and Thermal Framework Processor Participant (rev 03)
System peripheral		: Intel Corporation Device 3190 (rev 03)
VGA compatible controller		: Intel Corporation UHD Graphics 605 (rev 03) (prog-if 00 [VGA controller])
Audio device		: Intel Corporation Device 3198 (rev 03)
Communication controller		: Intel Corporation Celeron/Pentium Silver Processor Trusted Execution Engine Interface (rev 03)
SATA controller		: Intel Corporation Device 31e3 (rev 03) (prog-if 01 [AHCI 1.0])
PCI bridge		: Intel Corporation Device 31d8 (rev f3) (prog-if 00 [Normal decode])
PCI bridge		: Intel Corporation Device 31d6 (rev f3) (prog-if 00 [Normal decode])
PCI bridge		: Intel Corporation Device 31d7 (rev f3) (prog-if 00 [Normal decode])
USB controller		: Intel Corporation Device 31a8 (rev 03) (prog-if 30 [XHCI])
Signal processing controller		: Intel Corporation Celeron/Pentium Silver Processor Serial IO I2C Host Controller (rev 03)
Signal processing controller		: Intel Corporation Celeron/Pentium Silver Processor Serial IO I2C Host Controller (rev 03)
Signal processing controller		: Intel Corporation Device 31b0 (rev 03)
Signal processing controller		: Intel Corporation Device 31b2 (rev 03)
Signal processing controller		: Intel Corporation Device 31b4 (rev 03)
Signal processing controller		: Intel Corporation Device 31b6 (rev 03)
Signal processing controller		: Intel Corporation Device 31b8 (rev 03)
Signal processing controller		: Intel Corporation Device 31ba (rev 03)
ISA bridge		: Intel Corporation Device 31e8 (rev 03)
SMBus		: Intel Corporation Celeron/Pentium Silver Processor Gaussian Mixture Model (rev 03)
Non-Volatile memory controller		: Lite-On Technology Corporation Device 9100 (rev 03) (prog-if 02 [NVM Express])
Network controller		: Qualcomm Atheros QCA9377 802.11ac Wireless Network Adapter (rev 31)
[/QUOTE]

Summary
-------

Code:

-Computer-
Processor		: Intel(R) Pentium(R) Silver N5000 CPU @ 1.10GHz
Memory		: 7969MB (3937MB used)
Machine Type		: Notebook
Operating System		: Debian GNU/Linux 10 (buster)
User Name		: alexander (Alexander)
Date/Time		: чт, 15-кві-2021 07:31:22 +0300
-Display-
Resolution		: 0x0 pixels
OpenGL Renderer		: (Unknown)
X11 Vendor		: The X.Org Foundation
-Audio Devices-
Audio Adapter		: HDA-Intel - HDA Intel PCH
-Input Devices-
 AT Translated Set 2 keyboard
 Lid Switch
 Sleep Button
 Power Button
 Video Bus
 ETPS/2 Elantech Touchpad
 Logitech USB Optical Mouse
 Intel HID events
 Intel HID 5 button array
 PC Speaker
 Dell WMI hotkeys
 DELL08D5:00 04F3:30C4 Touchpad
 HDA Digital PCBeep
 HDA Intel PCH Headphone Mic
 HDA Intel PCH HDMI/DP,pcm:3
 HDA Intel PCH HDMI/DP,pcm:7
 HDA Intel PCH HDMI/DP,pcm:8
 HDA Intel PCH HDMI/DP,pcm:9
 HDA Intel PCH HDMI/DP,pcm:10
-Printers-
No printers found

Code:

Operating System
----------------

-Version-
Kernel		: Linux 4.19.0-16-amd64 (x86_64)
Version		: #1 SMP Debian 4.19.181-1 (2021-03-19)
C Library		: GNU C Library / (Debian GLIBC 2.28-10) 2.28
Distribution		: Debian GNU/Linux 10 (buster)
-Current Session-
Computer Name		: alexander-pc
User Name		: alexander (Alexander)
Language		: LC_CTYPE=en_US.UTF-8;LC_NUMERIC=uk_UA.UTF-8;LC_TIME=uk_UA.UTF-8;LC_COLLATE=en_US.UTF-8;LC_MONETARY=uk_UA.UTF-8;LC_MESSAGES=en_US.UTF-8;LC_PAPER=uk_UA.UTF-8;LC_NAME=uk_UA.UTF-8;LC_ADDRESS=uk_UA.UTF-8;LC_TELEPHONE=uk_UA.UTF-8;LC_MEASUREMENT=uk_UA.UTF-8;LC_IDENTIFICATION=uk_UA.UTF-8 (en_US.UTF-8)
Home Directory		: /home/alexander
-Misc-
Uptime		: 2 hours 18 minutes
Load Average		: 0,67, 0,64, 0,62
Available entropy in /dev/random		: 3826 bits (healthy)

While googling for "ath10k_hwmon-pci-0200" the temperature values I've seen were about +78.0°C or less but I've noticed that there were other PCI adapters with ath10k in reports. In other words, I couldn't find any comprehensible information about what's the critical temperature for ath10k or what should I do to find out the answer.

Any suggestions are appreciated

pan64 · 04-15-2021, 09:06 AM

I guess 100°C is not really acceptable. You may need to find the reason and lower it. (probably clean it?)

alexander_linux_user · 04-15-2021, 09:52 AM

Quote:

Originally Posted by pan64

I guess 100°C is not really acceptable. You may need to find the reason and lower it. (probably clean it?)

Could a different kernel help in your opinion?

Quote:

dell_smm-virtual-0
Adapter: Virtual device
CPU: +52.0°C
SODIMM: +41.0°C

acpitz-acpi-0
Adapter: ACPI interface
temp1: +52.0°C (crit = +95.0°C)

ath10k_hwmon-pci-0200
Adapter: PCI adapter
temp1: +100.0°C

coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +52.0°C (high = +105.0°C, crit = +105.0°C)
Core 0: +53.0°C (high = +105.0°C, crit = +105.0°C)
Core 1: +53.0°C (high = +105.0°C, crit = +105.0°C)
Core 2: +53.0°C (high = +105.0°C, crit = +105.0°C)
Core 3: +53.0°C (high = +105.0°C, crit = +105.0°C)

It's interesting that "sensors" tool show high temperatures for other adapters but not for PCI adapter, like it doesn't matter.

computersavvy · 04-16-2021, 10:19 AM

If you are relying on "sensors" to tell you the actual temps and the max limits you might read up on sensors. Most of the sensors detected actually have valid conversion factors in the config file(s) but some don't. I seriously doubt an ethernet interface is reporting 100C as an actual value but would more consider it is an invalid conversion in sensors configs.

Try looking in /etc/sensors3.conf and at the files in /etc/sensors.d to see what you might find out about the settings. Maybe you can adjust the displayed values to a valid number and input the max limits you expect at the same time.

ondoho · 04-17-2021, 02:23 PM

Did you run "sensors-detect" first?
Maybe it's just misconfigured.
ath10k appears to have something to do with a wireless device. I have no idea why that would run at 100 degrees celsius.

Maybe

Code:

lspci -k

could shed some light.

alexander_linux_user · 04-18-2021, 01:07 AM

Quote:

Originally Posted by ondoho

Did you run "sensors-detect" first?
Maybe it's just misconfigured.
ath10k appears to have something to do with a wireless device. I have no idea why that would run at 100 degrees celsius.

Maybe

Code:

lspci -k

could shed some light.

Yes, I did run "sensors-detect" first.

"lspci -k" did shed some light!

Quote:

02:00.0 Network controller: Qualcomm Atheros QCA9377 802.11ac Wireless Network Adapter (rev 31)
Subsystem: Dell QCA9377 802.11ac Wireless Network Adapter
Kernel driver in use: ath10k_pci
Kernel modules: ath10k_pci

Googling "Qualcomm Atheros QCA9377 working temperatures" and "Qualcomm Atheros QCA9377 critical temperature" didn't result in something clear, so it looks like there is nothing to worry about, I guess.

Thanks for your help!

alexander_linux_user · 04-18-2021, 09:16 AM

Quote:

Originally Posted by computersavvy

If you are relying on "sensors" to tell you the actual temps and the max limits you might read up on sensors. Most of the sensors detected actually have valid conversion factors in the config file(s) but some don't. I seriously doubt an ethernet interface is reporting 100C as an actual value but would more consider it is an invalid conversion in sensors configs.

Try looking in /etc/sensors3.conf and at the files in /etc/sensors.d to see what you might find out about the settings. Maybe you can adjust the displayed values to a valid number and input the max limits you expect at the same time.

Here is my sensors3.conf content:

Quote:

# libsensors configuration file
# -----------------------------
#
# This default configuration file only includes statements which do not
# differ from one mainboard to the next. Only label, compute and set
# statements for internal voltage and temperature sensors are included.
#
# In general, local changes should not be added to this file, but rather
# placed in custom configuration files located in /etc/sensors.d. This
# approach makes further updates much easier.
#
# Such custom configuration files for specific mainboards can be found in
# "configs" directory of lm-sensors package.
#
# Please contribute back a configuration of your board so other users with
# the same hardware won't need to recreate it again and again.

chip "lm78-*" "lm79-*" "lm80-*" "lm96080-*"

label temp1 "M/B Temp"

chip "w83792d-*"

label in0 "VcoreA"
label in1 "VcoreB"
label in6 "+5V"
label in7 "5VSB"
label in8 "Vbat"

set in6_min 5.0 * 0.90
set in6_max 5.0 * 1.10
set in7_min 5.0 * 0.90
set in7_max 5.0 * 1.10
set in8_min 3.0 * 0.90
set in8_max 3.0 * 1.10

chip "w83793-*"

label in0 "VcoreA"
label in1 "VcoreB"
label in7 "+5V"
label in8 "5VSB"
label in9 "Vbat"

set in7_min 5.0 * 0.90
set in7_max 5.0 * 1.10
set in8_min 5.0 * 0.90
set in8_max 5.0 * 1.10
set in9_min 3.0 * 0.90
set in9_max 3.0 * 1.10

chip "w83795g-*" "w83795adg-*"

label in12 "+3.3V"
label in13 "3VSB"
label in14 "Vbat"

set in12_min 3.3 * 0.90
set in12_max 3.3 * 1.10
set in13_min 3.3 * 0.90
set in13_max 3.3 * 1.10
set in14_min 3.0 * 0.90
set in14_max 3.3 * 1.10

chip "via686a-*"

label in0 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in4_min 12.0 * 0.90
set in4_max 12.0 * 1.10

chip "adm1025-*" "ne1619-*"

label in1 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"
label in5 "VCC"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in5_min 3.3 * 0.90
set in5_max 3.3 * 1.10
# Depending on how your chip is hardwired, you may or may not have
# +12V readings.
# set in4_min 12.0 * 0.90
# set in4_max 12.0 * 1.10

label temp1 "CPU Temp"
label temp2 "M/B Temp"

chip "lm87-*" "adm1024-*"

label in1 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in4_min 12.0 * 0.90
set in4_max 12.0 * 1.10

label temp1 "M/B Temp"
label temp2 "CPU Temp"

chip "it87-*" "it8712-*" "it8716-*" "it8718-*" "it8720-*"

label in8 "Vbat"

chip "fscpos-*" "fscher-*"
#FSC "Hermes"

label in0 "+12V"
label in1 "+5V"
label in2 "Vbat"

label temp1 "CPU Temp"
label temp2 "M/B Temp"
label temp3 "Aux Temp"

chip "fscscy-*"
#FSC "Scylla"

label in0 "+12V"
label in1 "+5V"
label in2 "+3.3V"

label temp1 "CPU0 Temp"
label temp2 "CPU1 Temp"
label temp3 "M/B Temp"
label temp4 "Aux Temp"

chip "fschds-*"
# Fujitsu Technology Solutions, "Hades"-Chip

# Temperatures
label temp1 "CPU Temp"
label temp2 "Super I/O Temp"
label temp3 "System Temp"

# Fans
label fan1 "PSU Fan"
label fan2 "CPU Fan"
label fan3 "System FAN2"
label fan4 "System FAN3"
label fan5 "System FAN4"

# Voltages
label in0 "+12V"
label in1 "+5V"
label in2 "Vbat"

chip "fscsyl-*"
# Fujitsu Technology Solutions, "Syleus"-Chip

# Temperatures
label temp1 "CPU Temp"
label temp4 "Super I/O Temp"
label temp5 "Northbridge Temp"

# Fans
label fan1 "CPU Fan"
label fan2 "System FAN2"
label fan3 "System FAN3"
label fan4 "System FAN4"
label fan7 "PSU Fan"

# Voltages
label in0 "+12V"
label in1 "+5V"
label in2 "Vbat"
label in3 "+3.3V"
label in5 "+3.3V-Aux"

chip "vt1211-*"

label in5 "+3.3V"

label temp2 "SIO Temp"

chip "vt8231-*"

label in5 "+3.3V"

chip "smsc47m192-*"

label in1 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"
label in5 "VCC"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in4_min 12.0 * 0.90
set in4_max 12.0 * 1.10
set in5_min 3.3 * 0.90
set in5_max 3.3 * 1.10

label temp1 "SIO Temp"

chip "lm85-*" "lm85b-*" "lm85c-*" "adm1027-*" "adt7463-*" "adt7468-*" \
"emc6d100-*" "emc6d102-*" "emc6d103-*" "emc6d103s-*"

label in1 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
# Depending on how your chip is hardwired, you may or may not have
# +12V readings.
# set in4_min 12.0 * 0.90
# set in4_max 12.0 * 1.10

label temp2 "M/B Temp"

chip "emc6w201-*"

label in2 "+3.3V"
label in3 "+5V"

label temp6 "M/B Temp"

chip "pc87365-*" "pc87366-*"

# Voltage inputs

label in7 "3VSB"
label in8 "VDD"
label in9 "Vbat"
label in10 "AVDD"

compute in7 @*2, @/2
compute in8 @*2, @/2
compute in10 @*2, @/2

# These are the operating conditions as recommended by National
# Semiconductor
set in7_min 3.0
set in7_max 3.6
set in8_min 3.0
set in8_max 3.6
set in10_min 3.0
set in10_max 3.6
# Depending on the hardware setup, the battery voltage may or may not
# be monitored.
# set in9_min 2.4
# set in9_max 3.6

label temp3 "SIO Temp"

set temp3_min 0
set temp3_max 70
set temp3_crit 85

chip "adm1030-*" "adm1031-*"

label temp1 "M/B Temp"

chip "w83627thf-*"

label in3 "+5V"
label in7 "5VSB"
label in8 "Vbat"

# Internal resistors
compute in3 @ * (1 + 34/51), @ / (1 + 34/51)
compute in7 @ * (1 + 34/51), @ / (1 + 34/51)

set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in7_min 5.0 * 0.90
set in7_max 5.0 * 1.10
# The battery voltage may or may not be monitored.
# set in8_min 3.0 * 0.90
# set in8_max 3.0 * 1.10

chip "w83627ehf-*" "w83627dhg-*" "w83667hg-*" "nct6775-*" "nct6776-*" \
"nct6779-*" "nct6791-*" "nct6795-*" "nct6796-*"

label in0 "Vcore"
label in2 "AVCC"
label in3 "+3.3V"
label in7 "3VSB"
label in8 "Vbat"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 3.3 * 0.90
set in3_max 3.3 * 1.10
set in7_min 3.3 * 0.90
set in7_max 3.3 * 1.10
set in8_min 3.0 * 0.90
set in8_max 3.3 * 1.10

chip "w83627uhg-*"

label in2 "AVCC"
label in3 "+5V"
label in7 "5VSB"
label in8 "Vbat"

set in2_min 5.0 * 0.90
set in2_max 5.0 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in7_min 5.0 * 0.90
set in7_max 5.0 * 1.10
set in8_min 3.0 * 0.90
set in8_max 3.3 * 1.10

chip "f71805f-*"

label in0 "+3.3V"

set in0_min 3.3 * 0.90
set in0_max 3.3 * 1.10

chip "f71872f-*"

label in0 "+3.3V"
label in9 "Vbat"
label in10 "3VSB"

set in0_min 3.3 * 0.90
set in0_max 3.3 * 1.10
set in9_min 3.0 * 0.90
set in9_max 3.0 * 1.10
set in10_min 3.3 * 0.90
set in10_max 3.3 * 1.10

chip "k8temp-*"

label temp1 "Core0 Temp"
label temp2 "Core0 Temp"
label temp3 "Core1 Temp"
label temp4 "Core1 Temp"

chip "dme1737-*"

label in0 "5VSB"
label in1 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"
label in5 "3VSB"
label in6 "Vbat"

label temp2 "SIO Temp"

set in0_min 5.0 * 0.90
set in0_max 5.0 * 1.10
set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in4_min 12.0 * 0.90
set in4_max 12.0 * 1.10
set in5_min 3.3 * 0.90
set in5_max 3.3 * 1.10
set in6_min 3.0 * 0.90
set in6_max 3.0 * 1.10

chip "sch311x-*"

label in1 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"
label in5 "3VSB"
label in6 "Vbat"

label temp2 "SIO Temp"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
set in4_min 12.0 * 0.90
set in4_max 12.0 * 1.10
set in5_min 3.3 * 0.90
set in5_max 3.3 * 1.10
set in6_min 3.0 * 0.90
set in6_max 3.0 * 1.10

chip "sch5027-*"

label in0 "5VSB"
label in1 "Vcore"
label in2 "+3.3V"
label in5 "3VSB"
label in6 "Vbat"

label temp2 "SIO Temp"

set in0_min 5.0 * 0.90
set in0_max 5.0 * 1.10
set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in5_min 3.3 * 0.90
set in5_max 3.3 * 1.10
set in6_min 3.0 * 0.90
set in6_max 3.0 * 1.10

chip "sch5127-*"

label in2 "+3.3V"
label in5 "3VSB"
label in6 "Vbat"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in5_min 3.3 * 0.90
set in5_max 3.3 * 1.10
set in6_min 3.0 * 0.90
set in6_max 3.0 * 1.10

chip "f71808e-*" "f71808a-*" "f71862fg-*" "f71869-*" "f71869a-*" "f71882fg-*" \
"f71889fg-*" "f71889ed-*" "f71889a-*"

label in0 "+3.3V"
label in7 "3VSB"
label in8 "Vbat"

compute in0 @*2, @/2
compute in7 @*2, @/2
compute in8 @*2, @/2

chip "f71858fg-*" "f8000-*"

label in0 "+3.3V"
label in1 "3VSB"
label in2 "Vbat"

compute in0 @*2, @/2
compute in1 @*2, @/2
compute in2 @*2, @/2

chip "f71868a-*"

label in0 "+3.3V"
label in7 "3VSB"
label in8 "Vbat"
label in9 "5VSB"

compute in0 @*2, @/2
compute in7 @*2, @/2
compute in8 @*2, @/2
compute in9 @*3, @/3

chip "f81865f-*"

label in0 "+3.3V"
label in5 "3VSB"
label in6 "Vbat"

compute in0 @*2, @/2
compute in5 @*2, @/2
compute in6 @*2, @/2

chip "adt7473-*" "adt7475-*"

label in2 "+3.3V"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10

label temp2 "Board Temp"

chip "adt7476-*" "adt7490-*"

label in1 "Vcore"
label in2 "+3.3V"
label in3 "+5V"
label in4 "+12V"

set in2_min 3.3 * 0.90
set in2_max 3.3 * 1.10
set in3_min 5.0 * 0.90
set in3_max 5.0 * 1.10
# Depending on how your ADT7476 is hardwired, you may or may not have
# +12V readings.
# set in4_min 12.0 * 0.90
# set in4_max 12.0 * 1.10

label temp2 "M/B Temp"

No mentions of "ath10k" or Qualcomm Atheros QCA9377, as I see it. "sensors.d" has one hidden file called ".placeholder" which is empty. Any thoughts?

Thanks for participation!

obobskivich · 04-19-2021, 02:59 AM

This is probably not quite on the same 'thread' as where you've been going (and I'll leave it to others to give their feedback), but in my experience some hardware just has 'bum' sensors that report insane values (I have an SSD that, for example, always reports itself at 110* C and I've seen CPUs report absurd values like 950* C or -70* C), in some cases this is due to a software adjustment/conversion being wrong, but in some cases its just a bad read from the device. On a desktop I would suggest you try taking its temperature in some way (non-contact thermometer is a good/safe idea - using your hand is a free/maybe not safe idea) - in my SSD example it runs at room temperature in actuality, it just reports insane values (and the value never changes which is also 'wrong' - it should at minimum change as room temperature changes). But with a laptop that's a bit harder (and 100* C is unfortunately believable on modern laptops). The only possible idea I can think of (from a hardware angle): can you easily remove that card from the laptop and try another one in its place? (I'm assuming this is a mini PCI or mini PCIe card - but I have no idea if this laptop is easy to open or not - some let you get at those devices pretty nicely, some require you to basically tear the entire machine apart and hope you don't break anything).

I found the developer manual for the 9377 itself (https://developer.qualcomm.com/qfile...cification.pdf), and it indicates a maximum temperature of 115* C in operation (Tcase) on page 17, but lists a lower value for '9377-3 variant' at 85* C.

I found this YouTube video that (I think) shows your laptop being opened: https://www.youtube.com/watch?v=nBhOkkMvdds - it looks like the WiFi card is near the DRAM towards the hinge and is removable - depending on how much one of these modules costs you may try replacing it just to see if it fixes things. If you're very brave, you could in theory try running the laptop with the cover off and taking the temperature of that device while it runs (I'd recommend a non-contact or at least non-human probe until you know its not actually running at >90* C).

business_kid · 04-22-2021, 06:56 AM

The ath10k is a kernel module for a wifi chip, right? I have a very similar Qualcomm chip using the ath9k module and it's not warm to the touch.

Wifi chips consume in the region of 100mA, with more perhaps on 802.11n and perhaps if they are being driven extremely fast by the pcie bus. 78ºC is a temperature I'd expect from an APU, which consumes amps. Your ath10k chip is overheating, imho, or your sensor feedback is crazy. 78ºF (=25.5ºC) is closer to what I'd expect.

Does your wifi work? Does it have a heatsink?

A good yardstick from my electronic repair days was to touch a finger on any low current suspect chip after 2 minutes from cold. If a board was consuming current, I'd let it cool, cover it with a towel or pullover, and turn on. Then I'd pull the towel and feel the chips after 1-2 minutes. The low current chips that continued to get hot were duds. Any hot to the touch were probably duds, although current varies a bit and I'd take that into account.

Increasingly, there's p channel fet(to +V) & n channel fet(to 0V) at most points inside chips, and one should always be off at all times. When they are partially on, leakage occurs and the chip gets hot. When many are leaking and getting hot, you'll get a high temperature. The chip might actually work for a bit, but not long.

alexander_linux_user · 04-22-2021, 07:18 AM

Quote:

Originally Posted by business_kid

The ath10k is a kernel module for a wifi chip, right? I have a very similar Qualcomm chip using the ath9k module and it's not warm to the touch.

Wifi chips consume in the region of 100mA, with more perhaps on 802.11n and perhaps if they are being driven extremely fast by the pcie bus. 78ºC is a temperature I'd expect from an APU, which consumes amps. Your ath10k chip is overheating, imho, or your sensor feedback is crazy. 78ºF (=25.5ºC) is closer to what I'd expect.

Does your wifi work? Does it have a heatsink?

A good yardstick from my electronic repair days was to touch a finger on any low current suspect chip after 2 minutes from cold. If a board was consuming current, I'd let it cool, cover it with a towel or pullover, and turn on. Then I'd pull the towel and feel the chips after 1-2 minutes. The low current chips that continued to get hot were duds. Any hot to the touch were probably duds, although current varies a bit and I'd take that into account.

Increasingly, there's p channel fet(to +V) & n channel fet(to 0V) at most points inside chips, and one should always be off at all times. When they are partially on, leakage occurs and the chip gets hot. When many are leaking and getting hot, you'll get a high temperature. The chip might actually work for a bit, but not long.

WiFi is alright, performance is alright, compared to my previous laptop the current one is cool, that's for sure. It was Fujitsu I used for about 10 years, and still, it worked fine, although I did clean it from time to time. Maybe I'll clean my Dell too.

I suppose my sensor feedback is crazy, it doesn't even know what the critical temperature is like if it's not a big deal if it will be +150ºC or more. What's a critical temperature for ath9k?

Right now it's showing +92ºC, while a few days ago it was showing +100ºC and I did absolutely nothing to make it lower.

business_kid · 04-22-2021, 08:31 AM

At that temperature wet the surface of the chip and it should disappear in a few seconds.

Perform my experiment as laid out in my last post. Does it really get that hot? If so, use it until it dies. Max junction temperature is usually 150ºC but there's layers of thermal insulation between outside and wafer, and 92ºC on the case is dangerous territory indeed for something that isn't expecting heat.

Also when you switch on, monitor the temperature every 30 seconds or so for 5 minutes. It may initially rise but should taper off . See can you match the temperature to case temperature.

It gets very uncomfortable on the case much above 40ºC. Figure out what is going on and tell us.

Emerson · 04-22-2021, 09:12 AM

Try some good well-polished distro (liveCD), it may have sensors properly configured, then you at least can see if this reading is bogus (it probably is).

alexander_linux_user · 04-22-2021, 09:57 AM

Quote:

Originally Posted by Emerson

Try some good well-polished distro (liveCD), it may have sensors properly configured, then you at least can see if this reading is bogus (it probably is).

Is there a more well-polished distro than Debian?

Emerson · 04-22-2021, 10:12 AM

Debian stable has ridiculously old kernels (and other packages) and they backport support for newer hardware, this approach may work for video cards and such. I wouldn't bet they have backported support for sensors. I'd never run Debian stable in a modern desktop or laptop. My 2¢.

alexander_linux_user · 04-22-2021, 10:19 AM

Quote:

Originally Posted by Emerson

Debian stable has ridiculously old kernels (and other packages) and they backport support for newer hardware, this approach may work for video cards and such. I wouldn't bet they have backported support for sensors. I'd never run Debian stable in a modern desktop or laptop. My 2¢.

It uses 4.19 LTS kernel. I don't want Ubuntu, Fedora didn't run well, Linux Mint is not interesting anymore, Manjaro or Arch maybe. Are you on Arch? What would you recommend?

What do you use instead of pulseaudio?