hwmon should subsys_init
PATCH 1/1 hwmon should subsys_init
Subsystem infrastructure should normally register with "subsys_initcall", so that it's available to drivers that may need to initialize early.
This patch updates "hwmon" to do so. It's common for embedded systems to have multifunction chips with hardware monitoring interfaces, and to have those chips be used during system bringup ... before a normal "module_init" would kick, or maybe just linked so they'd init before hwmon.
Index: at91/drivers/hwmon/hwmon.c
=================================--- at91.orig/drivers/hwmon/hwmon.c 2006-04-06 18:30:25.000000000 -0700
+++ at91/drivers/hwmon/hwmon.c 2006-12-21 03:16:13.000000000 -0800
@@ -101,7 +101,7 @@ static void __exit hwmon_exit(void)
class_destroy(hwmon_class);
}
-module_init(hwmon_init);
+subsys_initcall(hwmon_init);
module_exit(hwmon_exit);
EXPORT_SYMBOL_GPL(hwmon_device_register);
Jean Delvare
AppliedPATCH 1/1 hwmon should subsys_init
Quoted 14 lines · David Brownell
Subsystem infrastructure should normally register with "
subsys_initcall", so that it's available to drivers that may need to initialize early.This patch updates "hwmon" to do so. It's common for embedded systems to have multifunction chips with hardware monitoring interfaces, and to have those chips be used during system bringup ... before a normal "
module_init" would kick, or maybe just linked so they'd init before hwmon.Signed-off-by David Brownell <dbrownell at users.sourceforge.net>Index:
at91/drivers/hwmon/hwmon.c
=================================> ---at91.orig/drivers/hwmon/hwmon.c2006-04-06 18:30:25.000000000 -0700
+++at91/drivers/hwmon/hwmon.c2006-12-21 03:16:13.000000000 -0800
Review static void __exit hwmon_exit(void) · line 107
Quoted 8 lines
(quoted diff) +1
-1@@ -101,7 +101,7 @@ static void __exit hwmon_exit(void) class_destroy(hwmon_class); } -module_init(hwmon_init); +subsys_initcall(hwmon_init); module_exit(hwmon_exit); EXPORT_SYMBOL_GPL(hwmon_device_register);
Applied, thanks.
Signature · 1 line
Jean Delvare