Skip to contentSkip to conversation

cris: don't use module_init for non-modular core eeprom.c code

PATCH In review Paul Gortmaker 2 participants 3 messages Jun 14, 2015 – Jun 15, 2015

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 cris: don't use module_init for non-modular core eeprom.c code

The eeprom.c code is compiled based on the Kconfig setting ETRAX_I2C_EEPROM, which is bool. So the code is either built in or absent. It will never be modular, so using module_init as an alias for __initcall is rather misleading.

Fix this up now, so that we can relocate module_init from init.h into module.h in the future. If we don't do this, we'd have to add module.h to obviously non-modular code, and that would be a worse thing.

Direct use of __initcall is discouraged, vs prioritized ones. Use of device_initcall is consistent with what __initcall maps onto, and hence does not change the init order, making the
impact of this change zero. Should someone with real hardware
for boot testing want to change it later to arch_initcall or something different, they can do that at a later date.

---

[To be appended to the content originally sent as:
"Replace module_init with device_initcall in non modules"
https://lkml.kernel.org/r/[email protected]
]

 arch/cris/arch-v10/drivers/eeprom.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
arch/cris/arch-v10/drivers/eeprom.c +1 -2
index 5047a33043bd..f679a19dfeb8 100644
@@ -848,5 +848,4 @@ static void eeprom_disable_write_protect(void)
     /* Write protect disabled */
   }
 }
-
-module_init(eeprom_init);
+device_initcall(eeprom_init);
Signature · 1 line
2.2.1
#1 lore raw
Quoted 10 lines · Paul Gortmaker

The eeprom.c code is compiled based on the Kconfig setting ETRAX_I2C_EEPROM, which is bool. So the code is either built in or absent. It will never be modular, so using module_init as an alias for __initcall is rather misleading.

Fix this up now, so that we can relocate module_init from init.h into module.h in the future. If we don't do this, we'd have to add module.h to obviously non-modular code, and that would be a worse thing.

Quoted 8 lines

Direct use of __initcall is discouraged, vs prioritized ones. Use of device_initcall is consistent with what __initcall maps onto, and hence does not change the init order, making the
impact of this change zero. Should someone with real hardware
for boot testing want to change it later to arch_initcall or something different, they can do that at a later date.

Cc Mikael Starvik <[email protected]>

Looks good, I'll take it in the CRIS-tree.

Quoted 2 lines

/^JN - Jesper Nilsson

Signature · 1 line
               Jesper Nilsson -- [email protected]
#2 lore raw
Quoted 18 lines · Jesper Nilsson
Quoted 17 lines · Paul Gortmaker

The eeprom.c code is compiled based on the Kconfig setting ETRAX_I2C_EEPROM, which is bool. So the code is either built in or absent. It will never be modular, so using module_init as an alias for __initcall is rather misleading.

Fix this up now, so that we can relocate module_init from init.h into module.h in the future. If we don't do this, we'd have to add module.h to obviously non-modular code, and that would be a worse thing.

Direct use of __initcall is discouraged, vs prioritized ones. Use of device_initcall is consistent with what __initcall maps onto, and hence does not change the init order, making the
impact of this change zero. Should someone with real hardware
for boot testing want to change it later to arch_initcall or something different, they can do that at a later date.

Quoted 3 lines
Quoted 1 line
Cc Mikael Starvik <[email protected]>

Looks good, I'll take it in the CRIS-tree.

That is fine if you want, but I need to keep a copy here in this series (as per the part of the msg you snipped) ; otherwise I'll introduce a bisect fail into the commit history if my tree gets merged ahead of the cris tree.

Thanks,
Paul.
--

Quoted 6 lines
Quoted 3 lines

/^JN - Jesper Nilsson