[lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Matt Sheumack]

[-- Attachment #1.1: Type: text/plain, Size: 3866 bytes --]

Hi All,

Been scanning the recent activity around how 2.6.31 has broken fan control
on many m/boards, and am hoping to make a contribution so this problem can
be fixed. Given that my coding skills are limited to Fortran 77, best I can
do is help with data and test. I've not (yet) enabled the "lax" workaround,
so I don't know if it works. Happy to follow instructions from here to
provide H/W input.

Regards

Matt

- i7-920 + ASUS P6T.

What worked:
- Under Ubuntu 9.0.4 with 2.6.14 through 2.6.16 (canonical distributions).
- Manual modprobe w83627ehf added to rc script
- lm-sensors 3.1.1 (tarball from sensors web-site). Fan control found 3 fans
(1 CPU + 2 x chassis), with 2 x PWM devices (1 CPU + 1 chassis). The two m/b
connected fans were sucessfully controlled using:

      Settings for hwmon8/device/pwm2:
         Depends on hwmon8/device/temp2_input
         Controls hwmon8/device/fan4_input
<lm-sensors@lm-sensors.org>
What doesn't work:
- Under Ubuntu 9.10 with 2.6.31 (Canonical distribution, online upgrade from
9.0.4)
- sensors version 3.1.1 with libsensors version 3.1.1 (canonical
distributions)
- sensors sees CPU and core temps + 3 fans.
- There are no pwm-capable sensor modules installed devices found (hence no
fan-control)

# sensors
atk0110-acpi-0
Adapter: ACPI interface
Vcore Voltage:      +1.05 V  (min =  +0.80 V, max =  +1.60 V)
 +3.3 Voltage:      +3.30 V  (min =  +2.97 V, max =  +3.63 V)
 +5 Voltage:        +4.97 V  (min =  +4.50 V, max =  +5.50 V)
 +12 Voltage:      +11.97 V  (min = +10.20 V, max = +13.80 V)
CPU FAN Speed:     2481 RPM  (min =  600 RPM)
CHASSIS1 FAN Speed:3590 RPM  (min =  600 RPM)
CHASSIS2 FAN Speed:3245 RPM  (min =  600 RPM)
POWER FAN Speed:      0 RPM  (min =    0 RPM)
CPU Temperature:    +45.0°C  (high = +60.0°C, crit = +75.0°C)
MB Temperature:     +40.0°C  (high = +45.0°C, crit = +75.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0:      +46.0°C  (high = +80.0°C, crit = +100.0°C)
...
coretemp-isa-0007
Adapter: ISA adapter
Core 7:      +43.0°C  (high = +80.0°C, crit = +100.0°C)

dmesg:
[   13.045551] Adding 19044984k swap on /dev/md2.  Priority:-1 extents:1
across:19044984k
[   13.178591] nvidia: module license 'NVIDIA' taints kernel.
[   13.178594] Disabling lock debugging due to kernel taint
[   13.203853] Bluetooth: Generic Bluetooth USB driver ver 0.5
[   13.203949] usbcore: registered new interface driver btusb
[   13.222574] lp: driver loaded but no devices found
[   13.431026]   alloc irq_desc for 24 on node 0
[   13.431028]   alloc kstat_irqs on node 0
[   13.431035] nvidia 0000:02:00.0: PCI INT A -> GSI 24 (level, low) -> IRQ
24
[   13.431040] nvidia 0000:02:00.0: setting latency timer to 64
[   13.431147] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  173.14.20
Thu Jun 25 18:15:32 PDT 2009
[   13.511995] coretemp coretemp.0: Using relative temperature scale!
[   13.512023] coretemp coretemp.1: Using relative temperature scale!
[   13.512044] coretemp coretemp.2: Using relative temperature scale!
[   13.512065] coretemp coretemp.3: Using relative temperature scale!
[   13.512089] coretemp coretemp.4: Using relative temperature scale!
[   13.512111] coretemp coretemp.5: Using relative temperature scale!
[   13.512136] coretemp coretemp.6: Using relative temperature scale!
[   13.512161] coretemp coretemp.7: Using relative temperature scale!
[   13.570109] cfg80211: Calling CRDA to update world regulatory domain
[   13.797234] ip_tables: (C) 2000-2006 Netfilter Core Team
[   14.204574] w83627ehf: Found W83627EHG chip at 0x290
[   14.204594] ACPI: I/O resource w83627ehf [0x295-0x296] conflicts with
ACPI region HWRE [0x290-0x299]
[   14.204596] ACPI: If an ACPI driver is available for this device, you
should use it instead of the native driver

-- 
Matt Sheumack

[-- Attachment #1.2: Type: text/html, Size: 4111 bytes --]

[-- Attachment #2: Type: text/plain, Size: 153 bytes --]

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Reinhard Nissl]

Hi,

Am 14.11.2009 20:38, schrieb Matt Sheumack:

> Been scanning the recent activity around how 2.6.31 has broken fan
> control on many m/boards, and am hoping to make a contribution so this
> problem can be fixed. Given that my coding skills are limited to Fortran
> 77, best I can do is help with data and test. I've not (yet) enabled the
> "lax" workaround, so I don't know if it works. Happy to follow
> instructions from here to provide H/W input.

I recently upgraded to openSUSE 11.2 which uses kernel 2.6.31.5
looking forward to using the new hwmon driver ATK0110. But
gkrellm doesn't show the sensors, though sensors does, just like
in your report.

I wonder which sysfs layout is correct as for example coretemp
works with gkrellm. coretemp provides its temperatures below
/sys/class/hwmon/hwmon1/device/ while ATK0110 provides it below
/sys/class/hwmon/hwmon0/.

I would be glad if this could be fixed in the 2.6.31 kernel tree.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:rnissl@gmx.de

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Jean Delvare]

On Sun, 15 Nov 2009 21:23:10 +0100, Reinhard Nissl wrote:
> Hi,
> 
> Am 14.11.2009 20:38, schrieb Matt Sheumack:
> 
> > Been scanning the recent activity around how 2.6.31 has broken fan
> > control on many m/boards, and am hoping to make a contribution so this
> > problem can be fixed. Given that my coding skills are limited to Fortran
> > 77, best I can do is help with data and test. I've not (yet) enabled the
> > "lax" workaround, so I don't know if it works. Happy to follow
> > instructions from here to provide H/W input.
> 
> I recently upgraded to openSUSE 11.2 which uses kernel 2.6.31.5
> looking forward to using the new hwmon driver ATK0110. But
> gkrellm doesn't show the sensors, though sensors does, just like
> in your report.
> 
> I wonder which sysfs layout is correct as for example coretemp
> works with gkrellm. coretemp provides its temperatures below
> /sys/class/hwmon/hwmon1/device/ while ATK0110 provides it below
> /sys/class/hwmon/hwmon0/.

Both are correct, libsensors supports both.

> I would be glad if this could be fixed in the 2.6.31 kernel tree.

If "sensors" shows things correctly then you can be certain that
everything is right on the kernel side.

-- 
Jean Delvare

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Reinhard Nissl]

Hi,

Am 15.11.2009 22:26, schrieb Jean Delvare:

>> I wonder which sysfs layout is correct as for example coretemp
>> works with gkrellm. coretemp provides its temperatures below
>> /sys/class/hwmon/hwmon1/device/ while ATK0110 provides it below
>> /sys/class/hwmon/hwmon0/.
> 
> Both are correct, libsensors supports both.
> 
>> I would be glad if this could be fixed in the 2.6.31 kernel tree.
> 
> If "sensors" shows things correctly then you can be certain that
> everything is right on the kernel side.

What's the idea behind providing it "here" or(/and) "there"?

Looks like gkrellm doesn't use libsensors so the location matters
until it supports both locations too.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:rnissl@gmx.de

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Jean Delvare]

On Sun, 15 Nov 2009 22:33:59 +0100, Reinhard Nissl wrote:
> Hi,
> 
> Am 15.11.2009 22:26, schrieb Jean Delvare:
> 
> >> I wonder which sysfs layout is correct as for example coretemp
> >> works with gkrellm. coretemp provides its temperatures below
> >> /sys/class/hwmon/hwmon1/device/ while ATK0110 provides it below
> >> /sys/class/hwmon/hwmon0/.
> > 
> > Both are correct, libsensors supports both.
> > 
> >> I would be glad if this could be fixed in the 2.6.31 kernel tree.
> > 
> > If "sensors" shows things correctly then you can be certain that
> > everything is right on the kernel side.
> 
> What's the idea behind providing it "here" or(/and) "there"?
> 
> Looks like gkrellm doesn't use libsensors so the location matters
> until it supports both locations too.

Recent versions of gkrellm (2.3.1 and later) do support libsensors, if
built properly.

-- 
Jean Delvare

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Reinhard Nissl]

Hi,

Am 15.11.2009 22:53, schrieb Jean Delvare:

>> Looks like gkrellm doesn't use libsensors so the location matters
>> until it supports both locations too.
> 
> Recent versions of gkrellm (2.3.1 and later) do support libsensors, if
> built properly.

openSUSE 11.2 provides 2.3.1. Changelog says, they upgraded to
2.3.1 about 2 years ago. But ldd /usr/gkrellm doesn't mention
libsensors, although openSUSE 11.2 provides both v3 and v4.

So, looks like poor me is using a gkrellm without libsensors support.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:rnissl@gmx.de

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Jean Delvare]

On Sun, 15 Nov 2009 23:08:01 +0100, Reinhard Nissl wrote:
> Hi,
> 
> Am 15.11.2009 22:53, schrieb Jean Delvare:
> 
> >> Looks like gkrellm doesn't use libsensors so the location matters
> >> until it supports both locations too.
> > 
> > Recent versions of gkrellm (2.3.1 and later) do support libsensors, if
> > built properly.
> 
> openSUSE 11.2 provides 2.3.1. Changelog says, they upgraded to
> 2.3.1 about 2 years ago. But ldd /usr/gkrellm doesn't mention
> libsensors, although openSUSE 11.2 provides both v3 and v4.
> 
> So, looks like poor me is using a gkrellm without libsensors support.

Indeed. I've just tested with openSUSE 11.1 and I see the same, ldd
doesn't list libsensors and the sensor labels suggest libsensors isn't
used. It is odd that nobody noticed so far.

Would you please report this bug on bugzilla.novell.com?

-- 
Jean Delvare

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Luca Tettamanti]

On Sat, Nov 14, 2009 at 8:38 PM, Matt Sheumack <matt@sheumack.com> wrote:
> Hi All,
>
> Been scanning the recent activity around how 2.6.31 has broken fan control
> on many m/boards, and am hoping to make a contribution so this problem can
> be fixed. Given that my coding skills are limited to Fortran 77, best I can
> do is help with data and test. I've not (yet) enabled the "lax" workaround,
> so I don't know if it works. Happy to follow instructions from here to
> provide H/W input.
[...]
> # sensors
> atk0110-acpi-0
> Adapter: ACPI interface
> Vcore Voltage:      +1.05 V  (min =  +0.80 V, max =  +1.60 V)
>  +3.3 Voltage:      +3.30 V  (min =  +2.97 V, max =  +3.63 V)
>  +5 Voltage:        +4.97 V  (min =  +4.50 V, max =  +5.50 V)
>  +12 Voltage:      +11.97 V  (min = +10.20 V, max = +13.80 V)
> CPU FAN Speed:     2481 RPM  (min =  600 RPM)
> CHASSIS1 FAN Speed:3590 RPM  (min =  600 RPM)
> CHASSIS2 FAN Speed:3245 RPM  (min =  600 RPM)
> POWER FAN Speed:      0 RPM  (min =    0 RPM)
> CPU Temperature:    +45.0°C  (high = +60.0°C, crit = +75.0°C)
> MB Temperature:     +40.0°C  (high = +45.0°C, crit = +75.0°C)

atk0110 does not support fan control. If you want it then you have to:
- blacklist asus_atk0110 module
- pass "acpi_enforce_resources=lax" on kernel command list
- load w83627ehf

Be sure to read the FAQ before trying it:
http://www.lm-sensors.org/wiki/FAQ/Chapter3#Mysensorshavestoppedworkinginkernel2.6.31

Luca

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

Re: [lm-sensors] ASUS P6T, 2.6.31-14, no pwm devices

[Jean Delvare]

Hallo Reinhard,

On Sun, 15 Nov 2009 23:08:01 +0100, Reinhard Nissl wrote:
> Am 15.11.2009 22:53, schrieb Jean Delvare:
> 
> >> Looks like gkrellm doesn't use libsensors so the location matters
> >> until it supports both locations too.
> > 
> > Recent versions of gkrellm (2.3.1 and later) do support libsensors, if
> > built properly.
> 
> openSUSE 11.2 provides 2.3.1. Changelog says, they upgraded to
> 2.3.1 about 2 years ago. But ldd /usr/gkrellm doesn't mention

Do you mean "ldd /usr/bin/gkrellm"?

> libsensors, although openSUSE 11.2 provides both v3 and v4.
> 
> So, looks like poor me is using a gkrellm without libsensors support.

I have double-checked with the relevant package maintainer and openSUSE
11.2 has gkrellm 2.3.2, not 2.3.1. And it definitely has libsensors
support.

So either you are running an older version of openSUSE, or you somehow
messed up your ldd command when checking.

-- 
Jean Delvare
http://khali.linux-fr.org/wishlist.html

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