* Re: [lm-sensors] ASRock 970 Extreme3
2012-05-21 13:45 [lm-sensors] ASRock 970 Extreme3 James
@ 2012-05-21 18:11 ` Guenter Roeck
2012-05-21 20:07 ` James
2012-05-21 20:15 ` Guenter Roeck
2 siblings, 0 replies; 4+ messages in thread
From: Guenter Roeck @ 2012-05-21 18:11 UTC (permalink / raw)
To: lm-sensors
On Mon, 2012-05-21 at 09:45 -0400, James wrote:
> I use the nct6776-isa-0290
> Adapter: ISA adapter
>
> I have 3 fans:
> 1. The CPU fan (about 3300 RPM)
> 2. The rear case fan (about 1050 RPM)
> 3. The front case fan (about 1350 RPM)
>
> Do the fans in Power Supplies typically get monitored?
>
> I have 6 fan connectors:
> 1. Chassis fan connector (#12 on the motherboard)
> 2. Chassis fan connector (#15 on the motherboard)
> 3. Chassis fan connector (#2 on the motherboard)
> 1. CPU fan connector (#6 on the motherboard)
> 2. CPU fan connector (#5 on the motherboard)
> 1. Power fan connector (#10 on the motherboard)
>
> Maybe someone can see a pattern.
>
> 1. I plugged the rear case fan in #12
> fan1 was 1060 RPM (the rear case fan probably)
> fan2 was 3300 RPM (the CPU fan probably)
> other fans reported 0 RPMs
>
> 2. I plugged the rear case fan in #2
> all fans reported 0 RPM
> other fans reported 0 RPMs
>
> 3. I plugged the rear case fan in #12 and the front case fan in #15.
> fan1 was 1053 RPM (the rear case fan probably)
> fan2 was 3316 (the CPU fan probably)
> other fans reported 0 RPMs
>
> 4. I plugged the rear case fan in #10 and the front case fan in #12.
> fan1 was 1384 (the front case fan probably)
> other fans reported 0 RPMs
Problem is that ASRock uses a GPIO based multiplexer to connect the
fans. So there is no 1:1 relationship between fans connectors and fan
attributes. The GPIO pins used vary from board to board.
There are some possibilities to support this, such as to modify the
fancontrol script to also support setting gpio pins (if there is a gpio
driver for the given HW), or to somehow implement the necessary support
in the kernel. But those are all not really good solutions, and there
are some questions if it can be made to work reliably in the first
place. The AIDA64 folks tried to implement it but gave up (see
http://forums.aida64.com/index.php?/topic/386-asrocks-fan-controller-sensor-trouble/), which is not a good sign.
Guenter
_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [lm-sensors] ASRock 970 Extreme3
2012-05-21 13:45 [lm-sensors] ASRock 970 Extreme3 James
2012-05-21 18:11 ` Guenter Roeck
@ 2012-05-21 20:07 ` James
2012-05-21 20:15 ` Guenter Roeck
2 siblings, 0 replies; 4+ messages in thread
From: James @ 2012-05-21 20:07 UTC (permalink / raw)
To: lm-sensors
On 05/21/12 14:11, Guenter Roeck wrote:
> On Mon, 2012-05-21 at 09:45 -0400, James wrote:
>> I use the nct6776-isa-0290
>> Adapter: ISA adapter
>>
>> I have 3 fans:
>> 1. The CPU fan (about 3300 RPM)
>> 2. The rear case fan (about 1050 RPM)
>> 3. The front case fan (about 1350 RPM)
>>
>> Do the fans in Power Supplies typically get monitored?
>>
>> I have 6 fan connectors:
>> 1. Chassis fan connector (#12 on the motherboard)
>> 2. Chassis fan connector (#15 on the motherboard)
>> 3. Chassis fan connector (#2 on the motherboard)
>> 1. CPU fan connector (#6 on the motherboard)
>> 2. CPU fan connector (#5 on the motherboard)
>> 1. Power fan connector (#10 on the motherboard)
>>
>> Maybe someone can see a pattern.
>>
>> 1. I plugged the rear case fan in #12
>> fan1 was 1060 RPM (the rear case fan probably)
>> fan2 was 3300 RPM (the CPU fan probably)
>> other fans reported 0 RPMs
>>
>> 2. I plugged the rear case fan in #2
>> all fans reported 0 RPM
>> other fans reported 0 RPMs
>>
>> 3. I plugged the rear case fan in #12 and the front case fan in #15.
>> fan1 was 1053 RPM (the rear case fan probably)
>> fan2 was 3316 (the CPU fan probably)
>> other fans reported 0 RPMs
>>
>> 4. I plugged the rear case fan in #10 and the front case fan in #12.
>> fan1 was 1384 (the front case fan probably)
>> other fans reported 0 RPMs
>
> Problem is that ASRock uses a GPIO based multiplexer to connect the
> fans. So there is no 1:1 relationship between fans connectors and fan
> attributes. The GPIO pins used vary from board to board.
>
> There are some possibilities to support this, such as to modify the
> fancontrol script to also support setting gpio pins (if there is a gpio
> driver for the given HW), or to somehow implement the necessary support
> in the kernel. But those are all not really good solutions, and there
> are some questions if it can be made to work reliably in the first
> place. The AIDA64 folks tried to implement it but gave up (see
> http://forums.aida64.com/index.php?/topic/386-asrocks-fan-controller-sensor-trouble/), which is not a good sign.
>
> Guenter
>
>
>
It looks like I can get two fans.
So can I assume that there are two sensors for the 6 connectors and 3 are multiplexed for each sensor?
Is it correct that there are GPIO pins to switch which fan the sensor reads from?
_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [lm-sensors] ASRock 970 Extreme3
2012-05-21 13:45 [lm-sensors] ASRock 970 Extreme3 James
2012-05-21 18:11 ` Guenter Roeck
2012-05-21 20:07 ` James
@ 2012-05-21 20:15 ` Guenter Roeck
2 siblings, 0 replies; 4+ messages in thread
From: Guenter Roeck @ 2012-05-21 20:15 UTC (permalink / raw)
To: lm-sensors
On Mon, 2012-05-21 at 16:07 -0400, James wrote:
> On 05/21/12 14:11, Guenter Roeck wrote:
> > On Mon, 2012-05-21 at 09:45 -0400, James wrote:
> >> I use the nct6776-isa-0290
> >> Adapter: ISA adapter
> >>
> >> I have 3 fans:
> >> 1. The CPU fan (about 3300 RPM)
> >> 2. The rear case fan (about 1050 RPM)
> >> 3. The front case fan (about 1350 RPM)
> >>
> >> Do the fans in Power Supplies typically get monitored?
> >>
> >> I have 6 fan connectors:
> >> 1. Chassis fan connector (#12 on the motherboard)
> >> 2. Chassis fan connector (#15 on the motherboard)
> >> 3. Chassis fan connector (#2 on the motherboard)
> >> 1. CPU fan connector (#6 on the motherboard)
> >> 2. CPU fan connector (#5 on the motherboard)
> >> 1. Power fan connector (#10 on the motherboard)
> >>
> >> Maybe someone can see a pattern.
> >>
> >> 1. I plugged the rear case fan in #12
> >> fan1 was 1060 RPM (the rear case fan probably)
> >> fan2 was 3300 RPM (the CPU fan probably)
> >> other fans reported 0 RPMs
> >>
> >> 2. I plugged the rear case fan in #2
> >> all fans reported 0 RPM
> >> other fans reported 0 RPMs
> >>
> >> 3. I plugged the rear case fan in #12 and the front case fan in #15.
> >> fan1 was 1053 RPM (the rear case fan probably)
> >> fan2 was 3316 (the CPU fan probably)
> >> other fans reported 0 RPMs
> >>
> >> 4. I plugged the rear case fan in #10 and the front case fan in #12.
> >> fan1 was 1384 (the front case fan probably)
> >> other fans reported 0 RPMs
> >
> > Problem is that ASRock uses a GPIO based multiplexer to connect the
> > fans. So there is no 1:1 relationship between fans connectors and fan
> > attributes. The GPIO pins used vary from board to board.
> >
> > There are some possibilities to support this, such as to modify the
> > fancontrol script to also support setting gpio pins (if there is a gpio
> > driver for the given HW), or to somehow implement the necessary support
> > in the kernel. But those are all not really good solutions, and there
> > are some questions if it can be made to work reliably in the first
> > place. The AIDA64 folks tried to implement it but gave up (see
> > http://forums.aida64.com/index.php?/topic/386-asrocks-fan-controller-sensor-trouble/), which is not a good sign.
> >
> > Guenter
> >
> >
> >
> It looks like I can get two fans.
> So can I assume that there are two sensors for the 6 connectors and 3 are multiplexed for each sensor?
> Is it correct that there are GPIO pins to switch which fan the sensor reads from?
I really don't know details. Supposedly there is a .xml file somewhere
in the ASRock software describing which connector is connected to which
fan, and which fans are multiplexed. The NCT6776F has up to 5 fan
inputs, so theoretically you could get up to 5 fans working. But then I
have no idea which header connects to which fan input, and which headers
are multiplexed.
Guenter
_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
^ permalink raw reply [flat|nested] 4+ messages in thread