lm_sensors 2.6.4 on Tyan Thunder K7 (S2462) (w83781d driver)
Mark D. Studebaker
mds at paradyne.com
Wed Aug 14 04:13:55 CEST 2002
see the many tickets on this topic.
links on our 'old news' page.
Andrew Bray wrote:
>
> -----BEGIN PGP SIGNED MESSAGE-----
>
> Folks,
>
> This motherboard is claimed in its manual of having a Winbond W83627HF
> Super I/O ASIC and as a build option a Winbond W83627D hardware
> monitoring.
>
> The Winbond W83627D hardware monitoring chip is claimed to include:
> 3-pin fan monitoring headers
> 2-pin chassis intrusion header
> CPU temperature and voltage monitoring
>
> The Winbond web site doesn't have data on the W83627D chip.
>
> lm_sensors 2.6.4 detects a w83627hf at SMBUS address 0x2c, and a w83782d
> at address 0x2d.
>
> There are only two subclients detected at addresses 0x48 and 0x49.
>
> I have a suspidin that a W83627D is actually a W83782D with no subclients.
>
> Anyway, I cannot have both installed at once, because of a lack of
> subclient addresses.
>
> I have created a shell script that dumps all the files in
> /proc/sys/dev/sensors/w*/*.
>
> Here is the output of my script and 'sensors' for the two chips (I used
> init=0 to retain the BIOS initialisation):
>
> > insmod w83781d ignore=0,0x2d init=0
> >
> > w83627hf-i2c-0-2c:
> > alarms=4096
> > beep=0 0
> > fan1=750 0
> > fan2=750 0
> > fan3=750 0
> > fan_div=8 8 8
> > in0=4.08 4.08 1.74
> > in1=4.08 4.08 1.72
> > in2=2.97 3.63 3.29
> > in3=2.68 3.26 2.94
> > in4=2.84 3.45 2.91
> > in5=0.33 0.78 0.59
> > in6=0.70 1.02 0.72
> > in7=2.68 3.26 3.21
> > in8=2.70 3.29 0.00
> > pwm1=255
> > pwm2=255
> > sensor1=2
> > sensor2=2
> > sensor3=2
> > temp1=60.0 127.0 56.0
> > temp2=72.5 32.0 40.5
> > temp3=72.5 32.0 45.0
> > vid=1.850
> > vrm=8.2
> >
> > w83627hf-i2c-0-2c
> > Adapter: SMBus AMD766 adapter at 80e0
> > Algorithm: Non-I2C SMBus adapter
> > VCore 1: +1.74 V (min = +4.08 V, max = +4.08 V)
> > VCore 2: +1.74 V (min = +4.08 V, max = +4.08 V)
> > +3.3V: +3.29 V (min = +2.97 V, max = +3.63 V)
> > +5V: +4.94 V (min = +4.50 V, max = +5.48 V)
> > +12V: +11.06 V (min = +10.79 V, max = +13.11 V)
> > -12V: -11.88 V (min = -13.21 V, max = -10.90 V)
> > -5V: -5.45 V (min = -5.51 V, max = -4.51 V) ALARM
> > V5SB: +5.39 V (min = +4.50 V, max = +5.48 V)
> > VBat: +0.00 V (min = +2.70 V, max = +3.29 V)
>
> The BIOS hardware monitor page lists core voltages, plus +6 and +/-12V
> foltages, which closely match these here.
>
> > fan1: 0 RPM (min = 750 RPM, div = 8)
> > fan2: 0 RPM (min = 750 RPM, div = 8)
> > fan3: 0 RPM (min = 750 RPM, div = 8)
>
> The BIOS lists 2 CPU fans, and 4 chassis fans. The chassis fans were all
> reported at 0 RPM.
>
> > temp1: +55°C (limit = +60°C) sensor = 3904 transistor
> > temp2: +41.5°C (limit = +72°C, hysteresis = +32°C) sensor = 3904 transistor
> > temp3: +45.0°C (limit = +72°C, hysteresis = +32°C) sensor = 3904 transistor
>
> The BIOS only lists 2 temperature for the two CPUs, which wwere both 43 degC
>
> > vid: +1.850 V
> > alarms: Chassis intrusion detection ALARM
>
> My case does not make any connection with the chassis intrusion connector
> on the mootherboard.
>
> > beep_enable:
> > Sound alarm disabled
> >
> > insmod w83781d ignore=0,0x2c init=0
> > w83782d-i2c-0-2d:
> > alarms=198656
> > beep=1 0
> > fan1=3000 5314
> > fan2=3000 5152
> > fan3=1500 0
> > fan_div=2 2 4
> > in0=4.08 4.08 1.76
> > in1=4.08 4.08 2.84
> > in2=2.97 3.63 1.52
> > in3=2.68 3.26 2.86
> > in4=2.84 3.45 1.23
> > in5=0.33 0.78 2.48
> > in6=0.70 1.02 3.29
> > in7=2.68 3.26 2.51
> > in8=2.70 3.29 0.00
> > pwm1=255
> > pwm2=255
> > pwm3=255
> > pwm4=255
> > sensor1=2
> > sensor2=2
> > sensor3=2
> > temp1=100.0 127.0 46.0
> > temp2=72.5 32.0 41.5
> > temp3=72.5 32.0 45.0
> > vid=1.850
> > vrm=8.2
> >
> > w83782d-i2c-0-2d
> > Adapter: SMBus AMD766 adapter at 80e0
> > Algorithm: Non-I2C SMBus adapter
> > VCore 1: +1.76 V (min = +4.08 V, max = +4.08 V)
> > VCore 2: +2.84 V (min = +4.08 V, max = +4.08 V)
> > +3.3V: +1.53 V (min = +2.97 V, max = +3.63 V)
> > +5V: +4.80 V (min = +4.50 V, max = +5.48 V)
> > +12V: +4.56 V (min = +10.79 V, max = +13.11 V)
> > -12V: -2.16 V (min = -13.21 V, max = -10.90 V)
> > -5V: +2.62 V (min = -5.51 V, max = -4.51 V)
> > V5SB: +4.22 V (min = +4.50 V, max = +5.48 V)
> > VBat: +0.00 V (min = +2.70 V, max = +3.29 V)
>
> These voltages don't seem to corrrespond with anything reported by th
> BIOS.
>
> > fan1: 5314 RPM (min = 3000 RPM, div = 2)
> > fan2: 5113 RPM (min = 3000 RPM, div = 2)
>
> These are very close to the CPU fan speeds reported by the BIOS.
>
> > fan3: 0 RPM (min = 1500 RPM, div = 4)
> > temp1: +46°C (limit = +100°C) sensor = 3904 transistor
> > temp2: +42.0°C (limit = +72°C, hysteresis = +32°C) sensor = 3904 transistor
> > temp3: +32.0°C (limit = +72°C, hysteresis = +32°C) sensor = 3904 transistor
>
> These traces were taken only a minute or two apart, as both tests were
> using the same subcliient addresses.
>
> I would have expected the same temperature readings in both tests.
>
> > vid: +1.850 V
> > alarms:
> > beep_enable:
> > Sound alarm enabled
>
> To conclude:
>
> 1. Does it look like I have analysed things correctly?
>
> 2. Is there a way I can use force_subclients to operate one of the two
> Winbond chips without subclients, and have both installed at once?
>
> Regards,
>
> Andy
>
> - --
> - -----------------------------------------------------------------------------
> Andrew Bray, PWMS, MA, | preferred: mailto:andy at chaos.org.uk
> London, England | or: mailto:andy at madhouse.demon.co.uk
> PGP id/fingerprint: D811F5C9/26 B5 42 C6 F4 00 B2 71 BA EA 9B 81 6C 65 59 07
>
> -----BEGIN PGP SIGNATURE-----
> Version: 2.6.2i
>
> iQCVAwUBPVkbl0jMoRfYEfXJAQHS8AP/egFwnjH/EsJfHZSNGq85PPsjZbH8ldNa
> FxEblvX/MJXorQQQDnZB2PMf097qERo3dEhqc6f527q1081udxkATnjB6KnWYFZY
> 5ZKq+yR2vxx4lwPPWGYTZi01Kz41IAlKHg45ztPXrAmjLBwVMzey/mASKuVpO93O
> BrGCE2eEdUU=
> =HhjT
> -----END PGP SIGNATURE-----
More information about the lm-sensors
mailing list