CONFIG_SENSORS_OCC_P8_I2C

CONFIG_SENSORS_OCC_P8_I2C

[Borislav Petkov]

Hi,

make oldconfig asks me here on a x86 machine about this below and it
looks like this should depend on some P8 config item so that it doesn't
show on other architectures.

Right?

Thx.

CONFIG_SENSORS_OCC_P8_I2C:

This option enables support for monitoring sensors provided by the
On-Chip Controller (OCC) on a POWER8 processor. However, this driver
can only run on a baseboard management controller (BMC) connected to
the P8, not the POWER processor itself. Communications with the OCC are
established through I2C bus.

This driver can also be built as a module. If so, the module will be
called occ-p8-hwmon.

Symbol: SENSORS_OCC_P8_I2C [=n]
Type  : tristate
Defined at drivers/hwmon/occ/Kconfig:6
  Prompt: POWER8 OCC through I2C
  Depends on: HWMON [=y] && I2C [=y]
  Location:
    -> Device Drivers
      -> Hardware Monitoring support (HWMON [=y])
        -> POWER8 OCC through I2C (SENSORS_OCC_P8_I2C [=n])
Selects: SENSORS_OCC [=n]

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: CONFIG_SENSORS_OCC_P8_I2C

[Guenter Roeck]

On 12/4/22 09:55, Borislav Petkov wrote:
> Hi,
> 
> make oldconfig asks me here on a x86 machine about this below and it
> looks like this should depend on some P8 config item so that it doesn't
> show on other architectures.
> 
> Right?
> 

I don't immediately see why this should depend on a P8 configuration.
The driver runs on a BMC, after all. My understanding is that a BMC can be
of any architecture. Is there a guarantee that there are and never will be
any x86 BMCs, and/or that BMCs are limited to a certain subset of
architectures ?

Thanks,
Guenter

> Thx.
> 
> CONFIG_SENSORS_OCC_P8_I2C:
> 
> This option enables support for monitoring sensors provided by the
> On-Chip Controller (OCC) on a POWER8 processor. However, this driver
> can only run on a baseboard management controller (BMC) connected to
> the P8, not the POWER processor itself. Communications with the OCC are
> established through I2C bus.
> 
> This driver can also be built as a module. If so, the module will be
> called occ-p8-hwmon.
> 
> Symbol: SENSORS_OCC_P8_I2C [=n]
> Type  : tristate
> Defined at drivers/hwmon/occ/Kconfig:6
>    Prompt: POWER8 OCC through I2C
>    Depends on: HWMON [=y] && I2C [=y]
>    Location:
>      -> Device Drivers
>        -> Hardware Monitoring support (HWMON [=y])
>          -> POWER8 OCC through I2C (SENSORS_OCC_P8_I2C [=n])
> Selects: SENSORS_OCC [=n]
> 

Re: CONFIG_SENSORS_OCC_P8_I2C

[Borislav Petkov]

On Sun, Dec 04, 2022 at 10:37:09AM -0800, Guenter Roeck wrote:
> I don't immediately see why this should depend on a P8 configuration.
> The driver runs on a BMC, after all. My understanding is that a BMC can be
> of any architecture. Is there a guarantee that there are and never will be
> any x86 BMCs, and/or that BMCs are limited to a certain subset of
> architectures ?

Well, then it should depend on the architectures which are currently
running BMCs for P8 machines. ARM maybe?

At least on x86 the majority of BMCs are ARM-based AFAIK.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: CONFIG_SENSORS_OCC_P8_I2C

[Guenter Roeck]

On 12/4/22 11:13, Borislav Petkov wrote:
> On Sun, Dec 04, 2022 at 10:37:09AM -0800, Guenter Roeck wrote:
>> I don't immediately see why this should depend on a P8 configuration.
>> The driver runs on a BMC, after all. My understanding is that a BMC can be
>> of any architecture. Is there a guarantee that there are and never will be
>> any x86 BMCs, and/or that BMCs are limited to a certain subset of
>> architectures ?
> 
> Well, then it should depend on the architectures which are currently
> running BMCs for P8 machines. ARM maybe?
> 
> At least on x86 the majority of BMCs are ARM-based AFAIK.
> 

We are not typically that restrictive in the hwmon subsystem. That isn't done
for many of the other drivers which are typically only used on x86 either.
Why is that suddenly a problem ? Just don't enable it if you don't need it.

Anyway,

commit 58c0399db8bf6bf88c7580b742b0cb45e02f55d5
Author:     Jeremy Kerr <jk@codeconstruct.com.au>
AuthorDate: Mon Oct 24 16:15:27 2022 +0800
Commit:     Guenter Roeck <linux@roeck-us.net>
CommitDate: Thu Dec 1 09:21:06 2022 -0800

     hwmon: (occ) OCC sensors aren't arch-specific

     Commit c112d75840fb ("hwmon: OCC drivers are ARM-only") made the OCC
     sensor drivers not selectable on powerpc64:

       These drivers are for a BMC inside PowerPC servers. The BMC runs on
       ARM hardware, so only propose the drivers on this architecture, unless
       build-testing.

     ... but we now have a powerpc64 BMC (still for a powerpc64 host), so
     drop the `depends on` that excludes building for this platform.

     Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
     Acked-by: Joel Stanley <joel@jms.id.au>
     Link: https://lore.kernel.org/r/20221024081527.3842565-1-jk@codeconstruct.com.au
     Signed-off-by: Guenter Roeck <linux@roeck-us.net>

I really don't want to end up in a situation where I constantly have
to deal with changing architecture support, so I would strongly prefer
to keep it that way unless it is a real problem.

Guenter

Re: CONFIG_SENSORS_OCC_P8_I2C

[Borislav Petkov]

On Sun, Dec 04, 2022 at 01:32:50PM -0800, Guenter Roeck wrote:
> We are not typically that restrictive in the hwmon subsystem. That isn't done
> for many of the other drivers which are typically only used on x86 either.
> Why is that suddenly a problem ? Just don't enable it if you don't need it.

Yeah, ok, fair enough. I've switched my build script to use
olddefconfig.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette