[PATCH] THERMAL must not select HWMON

[PATCH] THERMAL must not select HWMON

[Bastian Blank]

[-- Attachment #1: Type: text/plain, Size: 862 bytes --]

On Fri, Mar 14, 2008 at 11:21:12AM -0700, Randy Dunlap wrote:

[ select forces HWMON to yes even if deps are never applicable ]

> Yes, that's a known & longstanding problem (that select does not
> follow/honor dependencies).

Okay.

So THERMAL can't select HWMON, instead it needs to do a proper depends.

Bastian

Signed-off-by: Bastian Blank <waldi@debian.org>
Cc: Zhang Rui <rui.zhang@intel.com>

diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index 3ab313e..394912c 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -4,7 +4,7 @@
 
 menuconfig THERMAL
 	bool "Generic Thermal sysfs driver"
-	select HWMON
+	depends on HWMON
 	default y
 	help
 	  Generic Thermal Sysfs driver offers a generic mechanism for
-- 
You canna change the laws of physics, Captain; I've got to have thirty minutes!

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [PATCH] THERMAL must not select HWMON

[Andrew Morton]

On Fri, 14 Mar 2008 19:37:58 +0100
Bastian Blank <bastian@waldi.eu.org> wrote:

> On Fri, Mar 14, 2008 at 11:21:12AM -0700, Randy Dunlap wrote:
> 
> [ select forces HWMON to yes even if deps are never applicable ]
> 
> > Yes, that's a known & longstanding problem (that select does not
> > follow/honor dependencies).
> 
> Okay.
> 
> So THERMAL can't select HWMON, instead it needs to do a proper depends.
> 
> Bastian
> 
> Signed-off-by: Bastian Blank <waldi@debian.org>
> Cc: Zhang Rui <rui.zhang@intel.com>
> 
> diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
> index 3ab313e..394912c 100644
> --- a/drivers/thermal/Kconfig
> +++ b/drivers/thermal/Kconfig
> @@ -4,7 +4,7 @@
>  
>  menuconfig THERMAL
>  	bool "Generic Thermal sysfs driver"
> -	select HWMON
> +	depends on HWMON
>  	default y
>  	help
>  	  Generic Thermal Sysfs driver offers a generic mechanism for

Sorry, but we have insufficient information here.

Presumably it fixes some build error, but the changelog should tell us (at
least) what that error was, why it occurs and how the patch fixes it.

Often (for bugfixes) I'll go off and find this stuff out from the mailing
list and put together a proper description.  I also need to do this so I
can work out whether the patch is needed in 2.6.25 or -stable.  But I can't
even find any reference to the original problem in the lkml archives.

Re: [PATCH] THERMAL must not select HWMON

[Randy Dunlap]

On Fri, 14 Mar 2008 13:43:01 -0700 Andrew Morton wrote:

> On Fri, 14 Mar 2008 19:37:58 +0100
> Bastian Blank <bastian@waldi.eu.org> wrote:
> 
> > On Fri, Mar 14, 2008 at 11:21:12AM -0700, Randy Dunlap wrote:
> > 
> > [ select forces HWMON to yes even if deps are never applicable ]
> > 
> > > Yes, that's a known & longstanding problem (that select does not
> > > follow/honor dependencies).
> > 
> > Okay.
> > 
> > So THERMAL can't select HWMON, instead it needs to do a proper depends.
> > 
> > Bastian
> > 
> > Signed-off-by: Bastian Blank <waldi@debian.org>
> > Cc: Zhang Rui <rui.zhang@intel.com>
> > 
> > diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
> > index 3ab313e..394912c 100644
> > --- a/drivers/thermal/Kconfig
> > +++ b/drivers/thermal/Kconfig
> > @@ -4,7 +4,7 @@
> >  
> >  menuconfig THERMAL
> >  	bool "Generic Thermal sysfs driver"
> > -	select HWMON
> > +	depends on HWMON
> >  	default y
> >  	help
> >  	  Generic Thermal Sysfs driver offers a generic mechanism for
> 
> Sorry, but we have insufficient information here.
> 
> Presumably it fixes some build error, but the changelog should tell us (at
> least) what that error was, why it occurs and how the patch fixes it.

Bastian can report if he actually had a build error or not.

What he reported on the linux-kbuild mailing list is that
THERMAL selects HWMON.  However, HWMON depends on HAS_IOMEM, but
arch/s390 does not enable/support HAS_IOMEM.  Bang.

> Often (for bugfixes) I'll go off and find this stuff out from the mailing
> list and put together a proper description.  I also need to do this so I
> can work out whether the patch is needed in 2.6.25 or -stable.  But I can't
> even find any reference to the original problem in the lkml archives.


---
~Randy