Re: 3.13.?: Strange / dangerous fan policy...

From: Manuel Krause
Date: Tue Mar 11 2014 - 17:59:54 EST

Next message: Michel Lespinasse: "Re: mm: mmap_sem lock assertion failure in __mlock_vma_pages_range"
Previous message: Josh Triplett: "[PATCHv2 3/3] x86: Support compiling out userspace I/O (iopl and ioperm)"
In reply to: Manuel Krause: "Re: 3.13.?: Strange / dangerous fan policy..."
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On 2014-03-10 02:49, Manuel Krause wrote:

On 2014-03-09 18:58, Rafael J. Wysocki wrote:

On Sunday, March 09, 2014 01:10:25 AM Manuel Krause wrote:

On 2014-03-08 16:59, Guenter Roeck wrote:

On 03/08/2014 03:08 AM, Jean Delvare wrote:

On Fri, 7 Mar 2014 14:52:30 -0800, Guenter Roeck wrote:

On Fri, Mar 07, 2014 at 11:04:29PM +0100, Manuel Krause wrote:

Hi, and thanks for the quick response!
No special fancy "fan control policy". 'fancontrol' isn't
up or
running.
Vanilla kernels 3.11.* and 3.12.* had been working on here
without
any extra work.
--
# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1: +71.0°C (crit = +256.0°C)
temp2: +69.0°C (crit = +110.0°C)
temp3: +52.0°C (crit = +105.0°C)
temp4: +25.0°C (crit = +110.0°C)
temp5: +58.0°C (crit = +110.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0: +62.0°C (high = +105.0°C, crit = +105.0°C)
Core 1: +60.0°C (high = +105.0°C, crit = +105.0°C)
--
My notebook (HP/Compaq 6730b) does not have a seperate fan
sensor.
This is with 3.12.13 with my normal workload.

Please, trust my above mentionned values of 94 °C vs. 74°C
as I
don't like to boot 3.13.6 anymore, to avoid harm to the
notebook's
casing.

Understood. Unfortunately, we'll need to get information
from the new kernel to be able to track down the problem.

Indeed. Not only the run-time temperatures, but also the high
and crit
limits.

But I'd do to test any improvement-patch.

So far I have no idea what is going on. I don't see anything
in the
drivers providing above data that would explain the behavior,
but I might be missing something.

Looks like a regression in the acpi subsystem or in power
management,
not hwmon. Hwmon is merely reporting the temperatures, it's not
responsible for the actual temperatures.

I would agree. I don't think we have enough information to be
sure,
though. There might be some unintended interaction or
interference.

gpu is a good hint ... for example, look at commit b9ed919f1c8
(drm/nouveau/drm/pm: remove everything except the hwmon
interfaces
to THERM). nouveau does export pwm and fan control information,
so any change in that code may have unintended side effects.
Similar, I don't know how ec39f64bba (drm/radeon/dpm: Convert to
use devm_hwmon_register_with_groups) could have the observed
impact,
as it is purely passive, but I prefer to be rather safe than
sorry.

This problem has now been submitted into bugzilla as
https://bugzilla.kernel.org/show_bug.cgi?id=71711.

Guenter

Sorry, for beeing late, had to search for/accumulate much info
for you...
I hope, you like me to put it into one answer to you all CCing
you.

My GFX is a GM45 Intel (mobile), shared memory, running the
opensource Mesa drivers/extensions.
kernel-module: i915

According to the output of 'cpupower': I have
CPUidle driver: acpi_idle
CPUidle governor: menu

CPUfreq:
driver: acpi-cpufreq
available cpufreq governors: ondemand, performance
-
And "ondemand" is running.
--

# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1: +41.0°C (crit = +256.0°C)
temp2: +92.0°C (crit = +110.0°C)
temp3: +71.0°C (crit = +105.0°C)
temp4: +26.5°C (crit = +110.0°C)
temp5: +25.0°C (crit = +110.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0: +86.0°C (high = +105.0°C, crit = +105.0°C)
Core 1: +84.0°C (high = +105.0°C, crit = +105.0°C)

FROM a critical "smelly" situation today, kernel-compilation, fan
@100%.
--

Additional findings:

Identification from bootup ACPI initialisation vs. sensors:
temp1 = DTSZ
temp2 = CPUZ --> triggering Cooling in 3.12.13 if > 74°C
temp3 = SKNZ
temp4 = BATZ "Battery Zone" always calm ~ +6°C of ambient T
temp5 = FDTZ --- in 3.12.13 a representation of the cooling-fan
(25 - 45 - 58 - max?)
Core 0 & Core 1 are the internal CPU T sensors.

With the 3.13.x (.5+) kernels the first gatherered cooling
settings from bootup do stay forever. Means, rebooting a hot
system will get a FDTZ @45°C+ and won't make any problems, as it
does cool enough (even for kernel compiling on here). If it gets
25°C @bootup the system goes into emergency cooling somewhen.
Same is with a suspend/resume.

Kernel 3.12.13 adjusts the cooling on it's own, but
appropriately.

This almost certainly is an ACPI regression, but I'm not sure
whether
thermal management or CPU power management is broken on your
system.

Can you compare the contents of /sys/class/thermal/ from
working and
not working kernels, please?

Rafael

Hi again,
unfortunately you didn't specify how deeply I should dig into
/sys/class/thermal. So you get the lines from # BOF # to # EOF #
below. I hope they're readable without more comments.

The most remarkable changes, in my eyes, had happened within
"thermal_zone1".

Best regards,
Manuel Krause

# BOF #
Following ones are all from /sys/class/thermal/ which are links
to -> ../../devices/virtual/thermal/

I've listed the directories in sections of cooling_devices and
thermal_zones separately for each bad/good kernel. For Emailing
purposes only. You can merge them into a spreadsheet for your
evaluation on your own. I've left out reporting some subdirs and
subdir's values that _really_ didn't seem to need attention.

Also, I've had collected the #sensors output for each readout,
having reproduced nearly the same workload, represented by the
"Fan speed" (thermal_zone4==FDTZ).

And I've done my very best to not produce typos or c&p errors.

3.13.5 -- 20140309 -- 20:52 -- bad
=============================
dir |-
/type /cur_state /max_state
cooling_device0 Processor 0 10
cooling_device1 Processor 0 10
cooling_device2 Fan 0 1
cooling_device3 Fan 1 1
cooling_device4 Fan 0 1
cooling_device5 Fan 0 1
cooling_device6 Fan 0 1
cooling_device7 LCD 0 24

3.12.13 -- 20140310 -- 00:26 -- good
==============================
dir |-
/type /cur_state /max_state
cooling_device0 Processor 0 10
cooling_device1 Processor 0 10
cooling_device2 Fan 0 1
cooling_device3 Fan 1 1
cooling_device4 Fan 1 1
cooling_device5 Fan 1 1
cooling_device6 Fan 1 1
cooling_device7 LCD 0 24

3.13.5 -- 20140309 -- 20:52 -- bad
=============================
dir |-
/passive /temp |- /cdev?_ /trip_ /trip_
trip_ point_ point_
point ?_temp ?_type
thermal_zone0 0 68000 ?=0 n.a. 256000 critical
thermal_zone1 n.a. 70000 |-
?=0 6 110000 critical
?=1 5 107000 passive
?=2 4 90000 active
?=3 3 75000 active
?=4 2 55000 active
?=5 1 45000 active
?=6 1 30000 active
thermal_zone2 n.a. 54000 |-
?=0 1 105000 critical
?=1 1 95000 passive
thermal_zone3 n.a. 25800 |-
?=0 1 110000 critical
?=1 1 60000 passive
thermal_zone4 0 58000 ?=0 n.a. 110000 critical

3.12.13 -- 20140310 -- 00:26 -- good
==============================
dir |-
/passive /temp |- /cdev?_ /trip_ /trip_
trip_ point_ point_
point ?_temp ?_type
thermal_zone0 0 50000 ?=0 n.a. 256000 critical
thermal_zone1 n.a. 70000 |-
?=0 1 110000 critical
?=1 1 107000 passive
?=2 2 90000 active
?=3 3 67000 active
?=4 4 55000 active
?=5 5 45000 active
?=6 6 30000 active
thermal_zone2 n.a. 53000 |-
?=0 1 105000 critical
?=1 1 95000 passive
thermal_zone3 n.a. 25600 |-
?=0 1 110000 critical
?=1 1 60000 passive
thermal_zone4 0 58000 ?=0 n.a. 110000 critical

---
Legend here:
/type is always acpitz
/mode enabled
/policy step_wise

- from kernel ACPI initialisation: thermal_zone0==DTSZ,
thermal_zone1==CPUZ, thermal_zone2==SKNZ,
thermal_zone3==BATZ, thermal_zone4==FDTZ
- n.a. means file or value is not available
___
Legend in general:
/power/control is always auto
/power/runtime_status unsupported
/uevent ''==empty

----------------------------------------------------------------

3.13.5 -- 20140309 -- 20:52 -- bad
=============================
# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1: +68.0°C (crit = +256.0°C)
temp2: +70.0°C (crit = +110.0°C)
temp3: +54.0°C (crit = +105.0°C)
temp4: +25.8°C (crit = +110.0°C)
temp5: +58.0°C (crit = +110.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0: +66.0°C (high = +105.0°C, crit = +105.0°C)
Core 1: +63.0°C (high = +105.0°C, crit = +105.0°C)

3.12.13 -- 20140310 -- 00:26 -- good
==============================
# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1: +50.0°C (crit = +256.0°C)
temp2: +70.0°C (crit = +110.0°C)
temp3: +53.0°C (crit = +105.0°C)
temp4: +25.6°C (crit = +110.0°C)
temp5: +58.0°C (crit = +110.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0: +65.0°C (high = +105.0°C, crit = +105.0°C)
Core 1: +61.0°C (high = +105.0°C, crit = +105.0°C)

# EOF #

Hi, and thank you for your attention ^^

at the bottom of this email you'd get the actual values for the new 3.12.14 kernel for two different levels of usage and ambient temperature.
You'd read, in kernel 3.12.14 the /cdev?_trip_point enumeration has changed to the way of 3.13.? and also one /trip_point_?_temp did. But 3.12.14 is working as well as 3.12.13. (So my first eyecatcher didn't lead to useful things.)
I'm not capaple of finding or understanding the related code, but, please, let me present an idea of what MAY be going on:

In 3.12.13+, on my system, the effective cooling fan speed seems to be an accumulation, maybe bitwise, of cooling_device[2-6]/cur_state, that each get activated (=1) by a certain other temperature value or level; each of the cooling_device[2-6]/cur_state stays @1 as long as their ref. temp. does not undershoot. For my system this ref. temp. would most likely be triggered by temp2 == thermal_zone1/temp [CPUZ].

In 3.13.? there seems to get only one of cooling_device[2-6]/cur_state be set to 1, the others left and/or rewritten with 0. And the fan speed algorithm then accumulates only one 1 without seeing the [_LEVEL_] number of cooling_device[2-6]... or re-requesting the related trigger temperature.

I hope this leads you developers nearer to a conclusion on how to fix it,
best regards, Manuel Krause

_____________________________
3.12.14 -- 20140311 -- 19:07 -- changed, not broken -- normal use
=============================
/sys/class/thermal/* which
are links to -> ../../devices/virtual/thermal/*

dir |-
/type /cur_state /max_state Maybe
trigger
/PWM
...
cooling_device2 Fan 0 1 not yet
observed
cooling_device3 Fan 0 1 FDTZ==58°C
cooling_device4 Fan 1 1 FDTZ==45°C
cooling_device5 Fan 1 1 FDTZ==34°C
cooling_device6 Fan 1 1 FDTZ==25°C
...

dir |-
/passive /temp |- /cdev?_ /trip_ /trip_
trip_ point_ point_
point ?_temp ?_type
...
thermal_zone1 n.a. 73000 |- (CPUZ)
?=0 6 110000 critical
?=1 5 107000 passive
?=2 4 90000 active
?=3 3 75000 active
?=4 2 55000 active
?=5 1 45000 active
?=6 1 30000 active
...
thermal_zone4 n.a. 45000 ?=0 n.a. 110000 critical (FDTZ)
...

# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1: +46.0°C (crit = +256.0°C)
temp2: +73.0°C (crit = +110.0°C)
temp3: +57.0°C (crit = +105.0°C)
temp4: +26.3°C (crit = +110.0°C)
temp5: +45.0°C (crit = +110.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0: +68.0°C (high = +105.0°C, crit = +105.0°C)
Core 1: +66.0°C (high = +105.0°C, crit = +105.0°C)

_____________________________
3.12.14 -- 20140311 -- 21:09 -- changed, not broken -- idle state
=============================

dir |-
/type /cur_state /max_state Maybe
trigger
/PWM
...
cooling_device2 Fan 0 1 not yet
observed
cooling_device3 Fan 0 1 FDTZ==58°C
cooling_device4 Fan 0 1 FDTZ==45°C
cooling_device5 Fan 0 1 FDTZ==34°C
cooling_device6 Fan 1 1 FDTZ==25°C
...

dir |-
/passive /temp
thermal_zone1 n.a. 46000 ... (CPUZ)
...
thermal_zone4 n.a. 25000 ... (FDTZ)
...

# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1: +50.0°C (crit = +256.0°C)
temp2: +46.0°C (crit = +110.0°C)
temp3: +44.0°C (crit = +105.0°C)
temp4: +25.7°C (crit = +110.0°C)
temp5: +25.0°C (crit = +110.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0: +41.0°C (high = +105.0°C, crit = +105.0°C)
Core 1: +41.0°C (high = +105.0°C, crit = +105.0°C)
_____________________________

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Next message: Michel Lespinasse: "Re: mm: mmap_sem lock assertion failure in __mlock_vma_pages_range"
Previous message: Josh Triplett: "[PATCHv2 3/3] x86: Support compiling out userspace I/O (iopl and ioperm)"
In reply to: Manuel Krause: "Re: 3.13.?: Strange / dangerous fan policy..."
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]