thermal: core: Drop trips_disabled bitmask
PATCH 1/1 thermal: core: Drop trips_disabled bitmask
From: Rafael J. Wysocki <[email protected]>
After recent changes, thermal_zone_get_trip() cannot fail, as invoked from thermal_zone_device_register_with_trips(), so the only role of the trips_disabled bitmask is struct thermal_zone_device is to make handle_thermal_trip() skip trip points whose temperature was initially zero. However, since the unit of temperature in the thermal core is millicelsius, zero may very well be a valid temperature value at least in some usage scenarios and the trip temperature may as well change later. Thus there is no reason to permanently disable trip points with initial temperature equal to zero.
Accordingly, drop the trips_disabled bitmask along with the code related to it.
--- drivers/thermal/thermal_core.c | 13 ------------- include/linux/thermal.h | 2 -- 2 files changed, 15 deletions(-)
drivers/thermal/thermal_core.c +0 -13
@@ -347,10 +347,6 @@ static void handle_thermal_trip(struct t {struct thermal_trip trip; -/* Ignore disabled trip points */ -if (test_bit (trip_id, &tz->trips_disabled)) -return ; -__thermal_zone_get_trip (tz, trip_id, &trip);if (trip.temperature ==THERMAL_TEMP_INVALID )
@@ -1231,7 +1227,6 @@ thermal_zone_device_register_with_trips(struct thermal_zone_device *tz;int id;int result; -int count;struct thermal_governor *governor;if (!type ||strlen (type) ==0 ) {
@@ -1328,14 +1323,6 @@ thermal_zone_device_register_with_trips(if (result) goto release_device; -for (count =0 ; count < num_trips; count++) { -struct thermal_trip trip; - - result =thermal_zone_get_trip (tz, count, &trip); -if (result || !trip.temperature) -set_bit (count, &tz->trips_disabled); - } -/* Update 'this' zone's governor information */ mutex_lock (&thermal_governor_lock);
include/linux/thermal.h +0 -2
@@ -122,7 +122,6 @@ struct thermal_cooling_device {* @devdata: private pointer for device private data * @trips: an array of struct thermal_trip * @num_trips: number of trip points the thermal zone supports -* @trips_disabled; bitmap for disabled trips * @passive_delay_jiffies: number of jiffies to wait between polls when * performing passive cooling. * @polling_delay_jiffies: number of jiffies to wait between polls when
@@ -163,7 +162,6 @@ struct thermal_zone_device {void *devdata;struct thermal_trip *trips;int num_trips; -unsigned long trips_disabled;/* bitmap for disabled trips */ unsigned long passive_delay_jiffies;unsigned long polling_delay_jiffies;int temperature;
Quoted 12 lines · Rafael J. Wysocki
From: Rafael J. Wysocki <[email protected]>
After recent changes,
thermal_zone_get_trip()cannot fail, as invoked fromthermal_zone_device_register_with_trips(), so the only role of thetrips_disabledbitmask is structthermal_zone_deviceis to makehandle_thermal_trip()skip trip points whose temperature was initially zero. However, since the unit of temperature in the thermal core is millicelsius, zero may very well be a valid temperature value at least in some usage scenarios and the trip temperature may as well change later. Thus there is no reason to permanently disable trip points with initial temperature equal to zero.
Quoted 4 lines
Accordingly, drop the
trips_disabledbitmask along with the code related to it.Signed-off-by Rafael J. Wysocki <[email protected]>
I guess I was copied because of commit f1b80a3878b2 ("thermal: core:
Restore behavior regarding invalid trip points"). Since then we stopped relying on this behavior with commit 5601ef91fba8 ("mlxsw: core_thermal:
Use static trip points for transceiver modules").
Tested your patch and didn't see any regressions:
Thanks
Rafael J. Wysocki
maintainerQuoted 14 lines · Ido Schimmel
Quoted 12 lines · Rafael J. Wysocki
From: Rafael J. Wysocki <[email protected]>
After recent changes,
thermal_zone_get_trip()cannot fail, as invoked fromthermal_zone_device_register_with_trips(), so the only role of thetrips_disabledbitmask is structthermal_zone_deviceis to makehandle_thermal_trip()skip trip points whose temperature was initially zero. However, since the unit of temperature in the thermal core is millicelsius, zero may very well be a valid temperature value at least in some usage scenarios and the trip temperature may as well change later. Thus there is no reason to permanently disable trip points with initial temperature equal to zero.
Quoted 7 lines
Quoted 4 lines
Accordingly, drop the
trips_disabledbitmask along with the code related to it.Signed-off-by Rafael J. Wysocki <[email protected]>I guess I was copied because of commit
f1b80a3878b2("thermal: core:
Restore behavior regarding invalid trip points").
That's correct.
Quoted 7 lines
Since then we stopped
relying on this behavior with commit5601ef91fba8("mlxsw:core_thermal:
Use static trip points for transceiver modules").Tested your patch and didn't see any regressions:
Tested-by Ido Schimmel <[email protected]>
Thank you!
Quoted 12 lines · Rafael J. Wysocki
From: Rafael J. Wysocki <[email protected]>
After recent changes,
thermal_zone_get_trip()cannot fail, as invoked fromthermal_zone_device_register_with_trips(), so the only role of thetrips_disabledbitmask is structthermal_zone_deviceis to makehandle_thermal_trip()skip trip points whose temperature was initially zero. However, since the unit of temperature in the thermal core is millicelsius, zero may very well be a valid temperature value at least in some usage scenarios and the trip temperature may as well change later. Thus there is no reason to permanently disable trip points with initial temperature equal to zero.
Quoted 4 lines
Accordingly, drop the
trips_disabledbitmask along with the code related to it.Signed-off-by Rafael J. Wysocki <[email protected]>
The logical change after changing how the trip point are handled ;)
Signature · 5 lines
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook | <http://twitter.com/#!/linaroorg> Twitter | <http://www.linaro.org/linaro-blog/> Blog