Problems with PowerMac 2.7GHz cooling with 2.6.15-1.1824_FC4

[Brian D. Carlstrom]

Following up on my own post with more information. The therm_pm72
problem I reported against 2.6.15-1.1823_FC4 and 2.6.15-1.1824_FC4 still
seems present after updating to the released 2.6.15-1.1830_FC4. 

I ended up applying the "powerpc: g5 thermal overtemp bug" patch from
2.6.15 to 2.6.14-1.1656_FC4 to fix the problem.

/var/log/dmesg seems to give a hint of what is going wrong. here is
output from five different configurations on 3 different kinds of
hardware.

The first is 2.6.15-1.1830_FC4 on dual 2.0GHz machine that do not have
the liquid cooling pumps. As expected, no PUMP controls appear:

    PowerMac G5 Thermal control driver 1.2b2
    Detected fan controls:
      0: PWM fan, id 1, location: BACKSIDE,SYS CTRLR FAN
      1: RPM fan, id 2, location: DRIVE BAY
      2: PWM fan, id 2, location: SLOT,PCI FAN
      3: RPM fan, id 3, location: CPU A INTAKE
      4: RPM fan, id 4, location: CPU A EXHAUST
p      5: RPM fan, id 5, location: CPU B INTAKE
      6: RPM fan, id 6, location: CPU B EXHAUST

The second output is 2.6.15-1.1830_FC4 on dual 2.5Ghz machine. These
show the detection of the PUMP controls. These do not have any fan
issues:

    PowerMac G5 Thermal control driver 1.2b2
    Detected fan controls:
      0: PWM fan, id 1, location: BACKSIDE,SYS CTRLR FAN
      1: RPM fan, id 2, location: DRIVE BAY
      2: PWM fan, id 2, location: SLOT,PCI FAN
      3: RPM fan, id 3, location: CPU A INTAKE
      4: RPM fan, id 4, location: CPU A EXHAUST
      5: RPM fan, id 5, location: CPU B INTAKE
      6: RPM fan, id 6, location: CPU B EXHAUST
      7: RPM fan, id 1, location: CPU A PUMP
      8: RPM fan, id 0, location: CPU B PUMP

The third output is 2.6.15-1.1830_FC4 on dual 2.7Ghz machine. These
show the detection of the PUMP controls. However, the fans are blaring
full speed:

    PowerMac G5 Thermal control driver 1.2b2
    Detected fan controls:
      0: PWM fan, id 1, location: BACKSIDE,SYS CTRLR FAN
      1: RPM fan, id 2, location: DRIVE BAY
      2: PWM fan, id 2, location: SLOT,PCI FAN
      3: RPM fan, id 3, location: CPU A INTAKE
      4: RPM fan, id 4, location: CPU A EXHAUST
      5: RPM fan, id 5, location: CPU B INTAKE
      6: RPM fan, id 6, location: CPU B EXHAUST
      7: RPM fan, id 1, location: CPU A PUMP
      8: RPM fan, id 0, location: CPU B PUMP

The fourth output is running the old 2.6.14-1.1656_FC4 on dual 2.7GHz
machine. This was causing overheating on two machines because it lacks
the patch. However, it does not have blaring fans. Note in addition to
the PUMP lines, there is an extra message about "Liquid cooling pumps
detected":

    PowerMac G5 Thermal control driver 1.2b2
    Detected fan controls:
      0: PWM fan, id 1, location: BACKSIDE,SYS CTRLR FAN
      1: RPM fan, id 2, location: DRIVE BAY
      2: PWM fan, id 2, location: SLOT,PCI FAN
      3: RPM fan, id 3, location: CPU A INTAKE
      4: RPM fan, id 4, location: CPU A EXHAUST
      5: RPM fan, id 5, location: CPU B INTAKE
      6: RPM fan, id 6, location: CPU B EXHAUST
      7: RPM fan, id 1, location: CPU A PUMP
      8: RPM fan, id 0, location: CPU B PUMP
    Liquid cooling pumps detected, using new algorithm !

The fifth output is running the old 2.6.14-1.1656_FC4.bdc with the
therm_pm72 patch on dual 2.5Ghz machine. this does not overheat or have
fan issues. Note it has the "Liquid cooling pumps detected, using new
algorithm !" that is missing from the 1830 build:

    PowerMac G5 Thermal control driver 1.2b2
    Detected fan controls:
      0: PWM fan, id 1, location: BACKSIDE,SYS CTRLR FAN
      1: RPM fan, id 2, location: DRIVE BAY
      2: PWM fan, id 2, location: SLOT,PCI FAN
      3: RPM fan, id 3, location: CPU A INTAKE
      4: RPM fan, id 4, location: CPU A EXHAUST
      5: RPM fan, id 5, location: CPU B INTAKE
      6: RPM fan, id 6, location: CPU B EXHAUST
      7: RPM fan, id 1, location: CPU A PUMP
      8: RPM fan, id 0, location: CPU B PUMP
    Liquid cooling pumps detected, using new algorithm !

So I build a version of therm_pm72.c with #define DEBUG to see what I
could learn. I also put in some DBG message of my own. 

The problem seems to be that therm_pm72_attach isn't get called with
the "u3 0" adapter, since I see no "found U3-0" in the dmesg output,
but only "Found K2".

For now I'm going back to my own patched 2.6.14 kernel.

-bri