Subject: Re: lm(4) driver sync with OpenBSD, supporting new chips.
To: Juan RP <juan@xtrarom.org>
From: Nicolas Joly <njoly@pasteur.fr>
List: tech-kern
Date: 03/07/2007 18:42:51
On Wed, Mar 07, 2007 at 04:12:16PM +0100, Juan RP wrote:
> 
> Hi, I just updated our lm(4) driver with the OpenBSD code (which is IMHO
> much cleaner and simplified) and that gives us much more accurate
> reported values and support for new chips.
> 
> If you are using the lm(4) driver, could you please test the patch?

Works for me on my amd64 workstation; it does not report bogus temp
values for unconnected temp3 sensor anymore. For 2 other erroneous
values (+12V and -5V), real sensors are located on a ADT7463 chipset.

lm0 at isa0 port 0x290-0x297: Winbond W83627HF Hardware monitor

before:
VCORE A:      1.552 V
VCORE B:      1.568 V
  +3.3V:      3.328 V
    +5V:      5.047 V
   +12V:      5.046 V
   -12V:     -1.405 V
    -5V:     -3.544 V
  +5VSB:      5.042 V
   VBAT:      0.000 V
 Temp 1:     43.000 degC
 Temp 2:     37.500 degC
 Temp 3:    208.000 degC
  Fan 1:          0 RPM
  Fan 2:          0 RPM
  Fan 3:          0 RPM

after:
VCore A:      1.552 V
VCore B:      1.568 V
  +3.3V:      3.328 V
    +5V:      5.053 V
   +12V:      4.986 V
   -12V:     -1.337 V
    -5V:     -3.540 V
   5VSB:      5.042 V
   VBAT:      0.000 V
  Temp0:     43.000 degC
  Temp1:     37.000 degC

-- 
Nicolas Joly

Biological Software and Databanks.
Institut Pasteur, Paris.