Re: [lm-sensors] Missing Code v4.5-rc1 [Patch v3 25/25] IT8628E

From: Guenter Roeck <linux@roeck-us.net>
To: lm-sensors@vger.kernel.org
Subject: Re: [lm-sensors] Missing Code v4.5-rc1 [Patch v3 25/25] IT8628E
Date: Sun, 28 Feb 2016 17:37:37 +0000	[thread overview]
Message-ID: <56D33061.9010704@roeck-us.net> (raw)
In-Reply-To: <56D1B0EF.9050603@gmx.de>

On 02/28/2016 08:59 AM, Sardaukai wrote:
> Small Update
>> ...
>>> In addition i searched drivers/hwmon/it87.c in order to find the patched code from kernel 4.5-rc5.
>>> Unfortunately i found no code changes from Patch v3 (i tested about 5 patches).
>>> Should it not be patched within 4.5-rc1 or do i miss here something?
>>>
>> Does it say anywhere that the patches were reviewed and/or accepted
>> for upstream integration ?
>>
>> Problem is that there is hardly anyone willing to review hwmon kernel
>> patches nowadays, meaning they have to go in unreviewed or not at all.
>> Given the complexity and scope of the series, I don't want to just push
>> the patches into the upstream kernel without review. With that in mind,
>> it may take a while for the code to be upstream.
>>
> I see. You wrote that it should go up with rc-1, so i thought it is already there, my mistake.
>>> Second question
>>> Is it possible to take your code from github to patch/replace it in another kernel, like 4.3 / 4.4 ?
>>>
>> Yes, that should be possible. I have been running it on earlier kernels
>> myself.
>
> With GitHub module code for it87 chip's and a boot parameter "acpi_enforce_resources=lax"
> i was able to load it.
>
> it8628-isa-0a40
> Adapter: ISA adapter
> in0:          +0.43 V  (min =  +0.00 V, max =  +3.06 V)     <--- seems to be VCore, tested with cpuBurn
> in1:          +2.04 V  (min =  +0.00 V, max =  +3.06 V)
> in2:          +2.00 V  (min =  +0.00 V, max =  +3.06 V)
> in3:          +2.04 V  (min =  +0.00 V, max =  +3.06 V)
> in4:          +0.01 V  (min =  +0.00 V, max =  +3.06 V)
> in5:          +1.07 V  (min =  +0.00 V, max =  +3.06 V)
> in6:          +1.20 V  (min =  +0.00 V, max =  +3.06 V)     <--- could be RAM Voltage
> 3VSB:         +3.36 V  (min =  +0.00 V, max =  +6.12 V)
> Vbat:         +3.17 V              <--- Battery Voltage from EFI, really?!
> fan1:           0 RPM  (min =    0 RPM)      <--- CPU_FAN not connected
> fan2:        2280 RPM  (min =    0 RPM) <--- CPU_OPT (my H80i pump)
> fan3:        1078 RPM  (min =    0 RPM) <--- SYS_FAN3 or 2 (radiator fan)
> fan4:           0 RPM  (min =    0 RPM)      <--- SYS_FAN1 not connected
> fan5:        1819 RPM  (min =    0 RPM) <--- SYS_FAN2 or 3 (radiator fan #2)
> temp1:        +43.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermistor
> temp2:         -8.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermistor
> temp3:        +22.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = Intel PECI
> intrusion0:  OK               <--- Nice, i have to check where and how to use that
>
If it is connected. On most motherboards it isn't.

> Kernel 4.3.5 with Debian Testing (Stretch)
> sensors version 3.4.0 with libsensors version 3.4.0
> fancontrol 3.4.0-2
>
> I'll try to get that names step by step. Further i tried fancontrol, i got funny stuff with fan5.
> It seems to be that PWM values from fan5 are inverted.
>
> Following fancontrol setting stops all fans.
>
> # Configuration file generated by pwmconfig, changes will be lost
> INTERVAL=10
> DEVPATH=hwmon2=devices/platform/it87.2624
> DEVNAME=hwmon2=it8628
> FCTEMPS=hwmon2/pwm5=hwmon2/temp3_input hwmon2/pwm3=hwmon2/temp3_input
> FCFANS=hwmon2/pwm5=hwmon2/fan5_input hwmon2/pwm3=hwmon2/fan3_input
> MINTEMP=hwmon2/pwm5=20 hwmon2/pwm3=20
> MAXTEMP=hwmon2/pwm5=60 hwmon2/pwm3=60
> MINSTOP=hwmon2/pwm3=1 hwmon2/pwm5=254
> MINSTART=hwmon2/pwm3=1 hwmon2/pwm5=255
> MINPWM=hwmon2/pwm3=0 hwmon2/pwm5=254
> MAXPWM=hwmon2/pwm3=2 hwmon2/pwm5=255
>
> I tried some different EFI settings with system fans (Silent, Normal, PWM per degree) without
> any changes. Fan3 works out of the box. I will check all other fans soon.
> Any idea how to fix that?
>

Sorry, no.

I prefer to stick with automatic fan control; I consider it to be less risky,
since it keeps working even if the user space code fails. the Super-IO chips usually
to a pretty good job.

Guenter

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors