Skip to contentSkip to conversation

FW_LOADER should select HOTPLUG

PATCH New Adrian Bunk 2 participants 6 messages Dec 2, 2006 – Dec 3, 2006

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 FW_LOADER should select HOTPLUG

Since FW_LOADER is an option that is always select'ed by the code using it, it mustn't depend on HOTPLUG.

It's only relevant in the EMBEDDED=y case, but this might have resulted in illegal FW_LOADER=, HOTPLUG=n configurations.

drivers/base/Kconfig renamed from drivers/base/Kconfig.old +2 -2
@@ -19,8 +19,8 @@
 	  If unsure say Y here.
 
 config FW_LOADER
-	tristate "Userspace firmware loading support"
-	depends on HOTPLUG
+	tristate
+	select HOTPLUG
 	---help---
 	  This option is provided for the case where no in-kernel-tree modules
 	  require userspace firmware loading support, but a module built outside
#1 lore raw

PATCH 1/1 FW_LOADER should select HOTPLUG

[ fixed patch below ]

Since FW_LOADER is an option that is always select'ed by the code using it, it mustn't depend on HOTPLUG.

It's only relevant in the EMBEDDED=y case, but this might have resulted in illegal FW_LOADER=, HOTPLUG=n configurations.

drivers/base/Kconfig renamed from drivers/base/Kconfig.old +2 -2
@@ -19,8 +19,8 @@
 	  If unsure say Y here.
 
 config FW_LOADER
-	tristate "Userspace firmware loading support"
-	depends on HOTPLUG
+	tristate
+	select HOTPLUG
 	---help---
 	  This option is provided for the case where no in-kernel-tree modules
 	  require userspace firmware loading support, but a module built outside
drivers/pcmcia/Kconfig renamed from drivers/pcmcia/Kconfig.old +1 -1
@@ -6,7 +6,7 @@
 
 config PCCARD
 	tristate "PCCard (PCMCIA/CardBus) support"
-	depends on HOTPLUG
+	select HOTPLUG
 	---help---
 	  Say Y here if you want to attach PCMCIA- or PC-cards to your Linux
 	  computer.  These are credit-card size devices such as network cards,
#2 lore raw
Quoted 11 lines · On Sun, 3 Dec 2006 08:16:37 +0100 Adrian Bunk <[email protected]> wrote:

Since FW_LOADER is an option that is always select'ed by the code using it, it mustn't depend on HOTPLUG.

It's only relevant in the EMBEDDED=y case, but this might have resulted in illegal FW_LOADER=, HOTPLUG=n configurations.

drivers/base/Kconfig line 23

Quoted 7 lines
drivers/base/Kconfig renamed from drivers/base/Kconfig.old +2 -2
@@ -19,8 +19,8 @@
 	  If unsure say Y here.
 
 config FW_LOADER
-	tristate "Userspace firmware loading support"
-	depends on HOTPLUG
+	tristate
+	select HOTPLUG

It would be a retrograde step to start selecting HOTPLUG - we've managed to avoid it thus far.

It'd be better to make those drivers which select FW_LOADER dependent upon HOTPLUG.

#3 lore raw
Quoted 16 lines · Andrew Morton
Quoted 13 lines · On Sun, 3 Dec 2006 08:16:37 +0100 Adrian Bunk <[email protected]> wrote:

Since FW_LOADER is an option that is always select'ed by the code using it, it mustn't depend on HOTPLUG.

It's only relevant in the EMBEDDED=y case, but this might have resulted in illegal FW_LOADER=, HOTPLUG=n configurations.

drivers/base/Kconfig renamed from drivers/base/Kconfig.old +0 -0
@@ -19,8 +19,8 @@
 	  If unsure say Y here.
 
Quoted 8 lines

drivers/base/Kconfig line 23

Quoted 5 lines
drivers/base/Kconfig renamed from drivers/base/Kconfig.old +2 -2
 config FW_LOADER
-	tristate "Userspace firmware loading support"
-	depends on HOTPLUG
+	tristate
+	select HOTPLUG

It would be a retrograde step to start selecting HOTPLUG - we've managed to avoid it thus far.

$ grep -r "select HOTPLUG" * | wc -l
4
$ 
Quoted 2 lines

It'd be better to make those drivers which select FW_LOADER dependent upon HOTPLUG.

$ grep -r "select FW_LOADER" * | wc -l
71
$ 

And since the only case where depends<->select makes a difference is CONFIG_EMBEDDED=y, people will always continue to forget the dependency on HOTPLUG when select'ing FW_LOADER.

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
#4 lore raw
Quoted 30 lines · On Sun, 3 Dec 2006 09:15:52 +0100 Adrian Bunk <[email protected]> wrote:
Quoted 16 lines · Andrew Morton
Quoted 13 lines · On Sun, 3 Dec 2006 08:16:37 +0100 Adrian Bunk <[email protected]> wrote:

Since FW_LOADER is an option that is always select'ed by the code using it, it mustn't depend on HOTPLUG.

It's only relevant in the EMBEDDED=y case, but this might have resulted in illegal FW_LOADER=, HOTPLUG=n configurations.

drivers/base/Kconfig renamed from drivers/base/Kconfig.old +0 -0
@@ -19,8 +19,8 @@
 	  If unsure say Y here.
 
Quoted 8 lines

drivers/base/Kconfig line 23

Quoted 5 lines
drivers/base/Kconfig renamed from drivers/base/Kconfig.old +2 -2
 config FW_LOADER
-	tristate "Userspace firmware loading support"
-	depends on HOTPLUG
+	tristate
+	select HOTPLUG

It would be a retrograde step to start selecting HOTPLUG - we've managed to avoid it thus far.

$ grep -r "select HOTPLUG" * | wc -l
4
$ 
Quoted 5 lines
Quoted 2 lines

It'd be better to make those drivers which select FW_LOADER dependent upon HOTPLUG.

$ grep -r "select FW_LOADER" * | wc -l
71

46 actually. And it's immaterial.

Quoted 4 lines

And since the only case where depends<->select makes a difference is CONFIG_EMBEDDED=y, people will always continue to forget the dependency on HOTPLUG when select'ing FW_LOADER.

I realise that. But having HOTPLUG magically and mysteriously turn itself back on again is a pita.

#5 lore raw
Quoted 33 lines · Andrew Morton
Quoted 30 lines · On Sun, 3 Dec 2006 09:15:52 +0100 Adrian Bunk <[email protected]> wrote:
Quoted 16 lines · Andrew Morton
Quoted 13 lines · On Sun, 3 Dec 2006 08:16:37 +0100 Adrian Bunk <[email protected]> wrote:

Since FW_LOADER is an option that is always select'ed by the code using it, it mustn't depend on HOTPLUG.

It's only relevant in the EMBEDDED=y case, but this might have resulted in illegal FW_LOADER=, HOTPLUG=n configurations.

drivers/base/Kconfig renamed from drivers/base/Kconfig.old +0 -0
@@ -19,8 +19,8 @@
 	  If unsure say Y here.
 
Quoted 8 lines

drivers/base/Kconfig line 23

Quoted 5 lines
drivers/base/Kconfig renamed from drivers/base/Kconfig.old +2 -2
 config FW_LOADER
-	tristate "Userspace firmware loading support"
-	depends on HOTPLUG
+	tristate
+	select HOTPLUG

It would be a retrograde step to start selecting HOTPLUG - we've managed to avoid it thus far.

$ grep -r "select HOTPLUG" * | wc -l
4
$ 
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

It'd be better to make those drivers which select FW_LOADER dependent upon HOTPLUG.

$ grep -r "select FW_LOADER" * | wc -l
71

46 actually. And it's immaterial.

Sorry, I had some .old and .orig files in my tree...

Quoted 6 lines
Quoted 3 lines

And since the only case where depends<->select makes a difference is CONFIG_EMBEDDED=y, people will always continue to forget the dependency on HOTPLUG when select'ing FW_LOADER.

I realise that. But having HOTPLUG magically and mysteriously turn itself back on again is a pita.

In practice, it's already always turned on (except for CONFIG_EMBEDDED=n).

Unless Kconfig gets changed to automatically inherit dependencies of select'ed symbols, FW_LOADER not select'ing HOTPLUG is simply a constant source of bugs.

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed