[lm-sensors] Urgent: need some more info about the Hadess IC

[lm-sensors] Urgent: need some more info about the Hadess IC

[Hans de Goede]

Hi,

Some time ago we mailed about the Hadess management IC. You
said it was 100% identical to the Heimdall except for the
voltage registers, which should be 0x20 - 0x22.

This cannot be right, as this would give us the following
register definitions for the Hadess:

#define FSCHMD_REG_WDOG_PRESET		0x28
#define FSCHMD_REG_WDOG_STATE		0x23
#define FSCHMD_REG_WDOG_CONTROL		0x21

static const u8 FSCHMD_REG_VOLT[3] = { 0x20, 0x21, 0x22 };

Notice that register 0x21 now is seen as both a
voltage register and the watchdog control register, which
cannot be right. Can you please check if the watchdog registers
perhaps have changed with the Hadess too ?

And if possible double check nothing else has changed?

I put urgent in the subject, because I would like to
get the Hadess support in to kernel-2.6.30, and the
merge window for that has just opened (and doesn't stay open
all that long).

Thanks & Regards,

Hans


p.s.

2.6.30 will come with DMI based detection (so auto loading)
of the Hermes and Syleus drivers, and with Syleus support,
leading to 100% plug and play hwmon support for FSC systems
with those hwmon IC's :)

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Urgent: need some more info about the Hadess IC

[Cestonaro, Thilo]

Hey Hans,

thats correct. The watchdog stuff has changed too. Sorry for that mistake.

0x20 => 5V Actual
0x21 => 12V Actual
0x22 => Bat Actual

0x28 => WD Control
0x29 => WD Status
0x2A => WD PreSet

As syleus has it (except the new voltages).
I've double checked the rest against the fschmd.c I got from you. The rest is correct!


DMI in3-in5:
I'm checking the stuff and I think you'll get new informations on Monday.
Anyway, what BIOS Ver. do you have? Perhaps the voltages are missing there.

Ciao Thilo

-----Original Message-----
From: Hans de Goede [mailto:hdegoede@redhat.com] 
Sent: Friday, March 27, 2009 1:19 PM
To: Cestonaro, Thilo
Cc: Jean Delvare; LM Sensors
Subject: Urgent: need some more info about the Hadess IC

Hi,

Some time ago we mailed about the Hadess management IC. You said it was 100% identical to the Heimdall except for the voltage registers, which should be 0x20 - 0x22.

This cannot be right, as this would give us the following register definitions for the Hadess:

#define FSCHMD_REG_WDOG_PRESET		0x28
#define FSCHMD_REG_WDOG_STATE		0x23
#define FSCHMD_REG_WDOG_CONTROL		0x21

static const u8 FSCHMD_REG_VOLT[3] = { 0x20, 0x21, 0x22 };

Notice that register 0x21 now is seen as both a voltage register and the watchdog control register, which cannot be right. Can you please check if the watchdog registers perhaps have changed with the Hadess too ?

And if possible double check nothing else has changed?

I put urgent in the subject, because I would like to get the Hadess support in to kernel-2.6.30, and the merge window for that has just opened (and doesn't stay open all that long).

Thanks & Regards,

Hans


p.s.

2.6.30 will come with DMI based detection (so auto loading) of the Hermes and Syleus drivers, and with Syleus support, leading to 100% plug and play hwmon support for FSC systems with those hwmon IC's :)

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Urgent: need some more info about the Hadess IC

[Hans de Goede]

Thanks,

That was very helpful, I've just send a patch adding Hades support to Jean.

And I'm looking forward to some more info about the DMI tables for in3-5

Regards,

Hans


On 03/27/2009 03:03 PM, Cestonaro, Thilo wrote:
> Hey Hans,
>
> thats correct. The watchdog stuff has changed too. Sorry for that mistake.
>
> 0x20 =>  5V Actual
> 0x21 =>  12V Actual
> 0x22 =>  Bat Actual
>
> 0x28 =>  WD Control
> 0x29 =>  WD Status
> 0x2A =>  WD PreSet
>
> As syleus has it (except the new voltages).
> I've double checked the rest against the fschmd.c I got from you. The rest is correct!
>
>
> DMI in3-in5:
> I'm checking the stuff and I think you'll get new informations on Monday.
> Anyway, what BIOS Ver. do you have? Perhaps the voltages are missing there.
>
> Ciao Thilo
>
> -----Original Message-----
> From: Hans de Goede [mailto:hdegoede@redhat.com]
> Sent: Friday, March 27, 2009 1:19 PM
> To: Cestonaro, Thilo
> Cc: Jean Delvare; LM Sensors
> Subject: Urgent: need some more info about the Hadess IC
>
> Hi,
>
> Some time ago we mailed about the Hadess management IC. You said it was 100% identical to the Heimdall except for the voltage registers, which should be 0x20 - 0x22.
>
> This cannot be right, as this would give us the following register definitions for the Hadess:
>
> #define FSCHMD_REG_WDOG_PRESET		0x28
> #define FSCHMD_REG_WDOG_STATE		0x23
> #define FSCHMD_REG_WDOG_CONTROL		0x21
>
> static const u8 FSCHMD_REG_VOLT[3] = { 0x20, 0x21, 0x22 };
>
> Notice that register 0x21 now is seen as both a voltage register and the watchdog control register, which cannot be right. Can you please check if the watchdog registers perhaps have changed with the Hadess too ?
>
> And if possible double check nothing else has changed?
>
> I put urgent in the subject, because I would like to get the Hadess support in to kernel-2.6.30, and the merge window for that has just opened (and doesn't stay open all that long).
>
> Thanks&  Regards,
>
> Hans
>
>
> p.s.
>
> 2.6.30 will come with DMI based detection (so auto loading) of the Hermes and Syleus drivers, and with Syleus support, leading to 100% plug and play hwmon support for FSC systems with those hwmon IC's :)

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors