tech-userlevel archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: dbcool, envsys, powerd shutting down my machine


> The above is not necessarily correct -- many of the chips supported by
> the lm(4) driver, on NetBSD, OpenBSD and DragonFly BSD, certainly do,
> in fact, implement a whole bunch of fan-controlling features; see
> for a sample BSD implementation of the
> fan controlling logic.

Ah - I looked at the LM manuals, as I had the references handy, and
assumed the others were similar.  They are almost different enough to
have separate drivers.  However, I think that it makes sense to add
the changes in the patch to our driver.

> Winbond W83627HF is supported by both,n,d/wbsio.4 and
>,o,d/lm.4, and is one of the chips supported by the
> fanctl patch above, too.

Looks like wbsio handles the attachment glue (rather than lm at iic).

> However, the above manual from Tyan doesn't appear to specify how
> exactly are the fans actually wired up (e.g., to which pins on which
> chip).

That seems fairly standard.  If the firmware can display the sensor
values, it's possible to do some correlation, but otherwise I haven't
found a good way, especially with the range of temperatures and voltage
sources that can be measured.

As an example where the firmware can show the values, I can translate
sensors on a Sun fire V240:

               F0.RS:      5720                                       RPM
        MB.P0.T_CORE:    47.000                                      degC
       MB.V_GBE_+2V5:     2.508                                         V
            MB.T_ENC:    11.000                                      degC

which is nice.



PS.  Approximate translations:

  F0.RS		fan 0 rotational speed
  MB.P0.T_CORE	main board processor 0 temperature (core)
  MB.V_GBE_+2V5 main board gigabit ethernet +2.5V
  MB.T_ENC	main board temperature (enclosure)

   My other computer runs NetBSD too   -

Home | Main Index | Thread Index | Old Index