Skip to contentSkip to conversation

thermal: introduce by-name softlink

PATCH In review Wei Wang 6 participants 22 messages Dec 5, 2019 – Dec 16, 2019 Thermal

lore
Share & tools
lore

Wei Wang

author
#0 lore raw

v3 cover thermal: introduce by-name softlink

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

Changes since v1 [1]:
* Split cooling device registration into a seperate patch
Changes since v2 [2]:
* Split improve error message in thermal zone registration

[1]: v1: https://lore.kernel.org/patchwork/patch/1139450/
[2]: v2: https://lkml.org/lkml/2019/12/4/1291

Wei Wang (3):
thermal: prevent cooling device with no type to be registered
thermal: improve error message in thermal zone registration
thermal: create softlink by name for thermal_zone and cooling_device

 drivers/thermal/thermal_core.c | 55 +++++++++++++++++++++++++++-------
 1 file changed, 44 insertions(+), 11 deletions(-)
Signature · 1 line
2.24.0.393.g34dc348eaf-goog
author
#1 lore raw

PATCH v3 1/3 thermal: prevent cooling device with no type to be registered

commit 54fa38cc2eda ("thermal: core: prevent zones with no types to be registered") added logic to prevent thermal zone with empty type to be registered. Similarly, there are APIs that rely on cdev->type. This patch prevents cooling device without valid type to be registered.

---
 drivers/thermal/thermal_core.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/thermal/thermal_core.c +13 -3
index d4481cc8958f..974e2d91c30b 100644
@@ -954,12 +954,22 @@ __thermal_cooling_device_register(struct device_node *np,
 	struct thermal_zone_device *pos = NULL;
 	int result;
 
-	if (type && strlen(type) >= THERMAL_NAME_LENGTH)
+	if (!type || !type[0]) {
+		pr_err("Error: No cooling device type defined\n");
 		return ERR_PTR(-EINVAL);
+	}
+
+	if (strlen(type) >= THERMAL_NAME_LENGTH) {
+		pr_err("Error: Cooling device name over %d chars: %s\n",
+			THERMAL_NAME_LENGTH, type);
+		return ERR_PTR(-EINVAL);
+	}
 
 	if (!ops || !ops->get_max_state || !ops->get_cur_state ||
-	    !ops->set_cur_state)
+	    !ops->set_cur_state) {
+		pr_err("Error: Cooling device missing callbacks: %s\n", type);
 		return ERR_PTR(-EINVAL);
+	}
 
 	cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
 	if (!cdev)
@@ -972,7 +982,7 @@ __thermal_cooling_device_register(struct device_node *np,
 	}
 
 	cdev->id = result;
-	strlcpy(cdev->type, type ? : "", sizeof(cdev->type));
+	strlcpy(cdev->type, type, sizeof(cdev->type));
 	mutex_init(&cdev->lock);
 	INIT_LIST_HEAD(&cdev->thermal_instances);
 	cdev->np = np;
Signature · 1 line
2.24.0.393.g34dc348eaf-goog
Reviewed-by Tested-by on v3 1/3
#2 lore raw

Re: [PATCH v3 1/3] thermal: prevent cooling device with no type to be registered

Quoted 7 lines · Wei Wang

commit 54fa38cc2eda ("thermal: core: prevent zones with no types to be registered") added logic to prevent thermal zone with empty type to be registered. Similarly, there are APIs that rely on cdev->type. This patch prevents cooling device without valid type to be registered.

Looks better now. Thanks.

Quoted 45 lines
---
 drivers/thermal/thermal_core.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/thermal/thermal_core.c +13 -4
index d4481cc8958f..974e2d91c30b 100644
@@ -954,12 +954,22 @@ __thermal_cooling_device_register(struct device_node *np,
        struct thermal_zone_device *pos = NULL;
        int result;
 
-       if (type && strlen(type) >= THERMAL_NAME_LENGTH)
+       if (!type || !type[0]) {
+               pr_err("Error: No cooling device type defined\n");
                return ERR_PTR(-EINVAL);
+       }
+
+       if (strlen(type) >= THERMAL_NAME_LENGTH) {
+               pr_err("Error: Cooling device name over %d chars: %s\n",
+                       THERMAL_NAME_LENGTH, type);
+               return ERR_PTR(-EINVAL);
+       }
 
        if (!ops || !ops->get_max_state || !ops->get_cur_state ||
-           !ops->set_cur_state)
+           !ops->set_cur_state) {
+               pr_err("Error: Cooling device missing callbacks: %s\n", type);
                return ERR_PTR(-EINVAL);
+       }
 
        cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
        if (!cdev)
@@ -972,7 +982,7 @@ __thermal_cooling_device_register(struct device_node *np,
        }
 
        cdev->id = result;
-       strlcpy(cdev->type, type ? : "", sizeof(cdev->type));
+       strlcpy(cdev->type, type, sizeof(cdev->type));
        mutex_init(&cdev->lock);
        INIT_LIST_HEAD(&cdev->thermal_instances);
        cdev->np = np;
--

2.24.0.393.g34dc348eaf-goog

author
#3 lore raw

Re: [PATCH v3 1/3] thermal: prevent cooling device with no type to be registered

Quoted 12 lines · Amit Kucheria
Quoted 7 lines · Wei Wang

commit 54fa38cc2eda ("thermal: core: prevent zones with no types to be registered") added logic to prevent thermal zone with empty type to be registered. Similarly, there are APIs that rely on cdev->type. This patch prevents cooling device without valid type to be registered.

Looks better now. Thanks.

Thanks Amit for the review and testing.

Quoted 48 lines
Quoted 45 lines
---
 drivers/thermal/thermal_core.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/thermal/thermal_core.c +13 -4
index d4481cc8958f..974e2d91c30b 100644
@@ -954,12 +954,22 @@ __thermal_cooling_device_register(struct device_node *np,
        struct thermal_zone_device *pos = NULL;
        int result;
 
-       if (type && strlen(type) >= THERMAL_NAME_LENGTH)
+       if (!type || !type[0]) {
+               pr_err("Error: No cooling device type defined\n");
                return ERR_PTR(-EINVAL);
+       }
+
+       if (strlen(type) >= THERMAL_NAME_LENGTH) {
+               pr_err("Error: Cooling device name over %d chars: %s\n",
+                       THERMAL_NAME_LENGTH, type);
+               return ERR_PTR(-EINVAL);
+       }
 
        if (!ops || !ops->get_max_state || !ops->get_cur_state ||
-           !ops->set_cur_state)
+           !ops->set_cur_state) {
+               pr_err("Error: Cooling device missing callbacks: %s\n", type);
                return ERR_PTR(-EINVAL);
+       }
 
        cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
        if (!cdev)
@@ -972,7 +982,7 @@ __thermal_cooling_device_register(struct device_node *np,
        }
 
        cdev->id = result;
-       strlcpy(cdev->type, type ? : "", sizeof(cdev->type));
+       strlcpy(cdev->type, type, sizeof(cdev->type));
        mutex_init(&cdev->lock);
        INIT_LIST_HEAD(&cdev->thermal_instances);
        cdev->np = np;
--

2.24.0.393.g34dc348eaf-goog

#4 lore raw

Re: [PATCH v3 1/3] thermal: prevent cooling device with no type to be registered

In reply to Wei Wang
Quoted 15 lines · Wei Wang

commit 54fa38cc2eda ("thermal: core: prevent zones with no types to be registered") added logic to prevent thermal zone with empty type to be registered. Similarly, there are APIs that rely on cdev->type. This patch prevents cooling device without valid type to be registered.

---
 drivers/thermal/thermal_core.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/thermal/thermal_core.c +0 -0
index d4481cc8958f..974e2d91c30b 100644

drivers/thermal/thermal_core.c __thermal_cooling_device_register(struct device_node *np, · line 957

Quoted 5 lines
drivers/thermal/thermal_core.c +1 -1
@@ -954,12 +954,22 @@ __thermal_cooling_device_register(struct device_node *np,
 	struct thermal_zone_device *pos = NULL;
 	int result;
 
-	if (type && strlen(type) >= THERMAL_NAME_LENGTH)
+	if (!type || !type[0]) {

Why not use strlen(type) == 0 ?

Quoted 29 lines
drivers/thermal/thermal_core.c +12 -2
+		pr_err("Error: No cooling device type defined\n");
 		return ERR_PTR(-EINVAL);
+	}
+
+	if (strlen(type) >= THERMAL_NAME_LENGTH) {
+		pr_err("Error: Cooling device name over %d chars: %s\n",
+			THERMAL_NAME_LENGTH, type);
+		return ERR_PTR(-EINVAL);
+	}
 
 	if (!ops || !ops->get_max_state || !ops->get_cur_state ||
-	    !ops->set_cur_state)
+	    !ops->set_cur_state) {
+		pr_err("Error: Cooling device missing callbacks: %s\n", type);
 		return ERR_PTR(-EINVAL);
+	}
 
 	cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
 	if (!cdev)
@@ -972,7 +982,7 @@ __thermal_cooling_device_register(struct device_node *np,
 	}
 
 	cdev->id = result;
-	strlcpy(cdev->type, type ? : "", sizeof(cdev->type));
+	strlcpy(cdev->type, type, sizeof(cdev->type));
 	mutex_init(&cdev->lock);
 	INIT_LIST_HEAD(&cdev->thermal_instances);
 	cdev->np = np;
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
author
#5 lore raw

Re: [PATCH v3 1/3] thermal: prevent cooling device with no type to be registered

Quoted 20 lines · Daniel Lezcano
Quoted 18 lines · Wei Wang

commit 54fa38cc2eda ("thermal: core: prevent zones with no types to be registered") added logic to prevent thermal zone with empty type to be registered. Similarly, there are APIs that rely on cdev->type. This patch prevents cooling device without valid type to be registered.

---
 drivers/thermal/thermal_core.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/thermal/thermal_core.c +0 -0
index d4481cc8958f..974e2d91c30b 100644
@@ -954,12 +954,22 @@ __thermal_cooling_device_register(struct device_node *np,
      struct thermal_zone_device *pos = NULL;
      int result;
 
Quoted 4 lines

drivers/thermal/thermal_core.c __thermal_cooling_device_register(struct device_node *np, · line 957

Quoted 2 lines
drivers/thermal/thermal_core.c +1 -1
-     if (type && strlen(type) >= THERMAL_NAME_LENGTH)
+     if (!type || !type[0]) {

Why not use strlen(type) == 0 ?

Checking empty is faster than getting length and this is already a pattern used in this file:
https://github.com/torvalds/linux/blob/v5.4/drivers/thermal/thermal_core.c#L63

Quoted 39 lines
Quoted 22 lines
drivers/thermal/thermal_core.c +11 -1
+             pr_err("Error: No cooling device type defined\n");
              return ERR_PTR(-EINVAL);
+     }
+
+     if (strlen(type) >= THERMAL_NAME_LENGTH) {
+             pr_err("Error: Cooling device name over %d chars: %s\n",
+                     THERMAL_NAME_LENGTH, type);
+             return ERR_PTR(-EINVAL);
+     }
 
      if (!ops || !ops->get_max_state || !ops->get_cur_state ||
-         !ops->set_cur_state)
+         !ops->set_cur_state) {
+             pr_err("Error: Cooling device missing callbacks: %s\n", type);
              return ERR_PTR(-EINVAL);
+     }
 
      cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
      if (!cdev)

drivers/thermal/thermal_core.c __thermal_cooling_device_register(struct device_node *np, · line 988

Quoted 8 lines
drivers/thermal/thermal_core.c +1 -1
@@ -972,7 +982,7 @@ __thermal_cooling_device_register(struct device_node *np,
      }
 
      cdev->id = result;
-     strlcpy(cdev->type, type ? : "", sizeof(cdev->type));
+     strlcpy(cdev->type, type, sizeof(cdev->type));
      mutex_init(&cdev->lock);
      INIT_LIST_HEAD(&cdev->thermal_instances);
      cdev->np = np;

--
<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

author
#6 lore raw

PATCH v3 2/3 thermal: improve error message in thermal zone registration

In reply to Wei Wang

Follow up with commit 67eed44b8a8a ("thermal: Add some error messages") to clean up checkpatch warning on line length and also add more message for developers.

---
 drivers/thermal/thermal_core.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
drivers/thermal/thermal_core.c +11 -7
index 974e2d91c30b..9db7f72e70f8 100644
@@ -1255,29 +1255,33 @@ thermal_zone_device_register(const char *type, int trips, int mask,
 	int count;
 	struct thermal_governor *governor;
 
-	if (!type || strlen(type) == 0) {
+	if (!type || !type[0]) {
 		pr_err("Error: No thermal zone type defined\n");
 		return ERR_PTR(-EINVAL);
 	}
 
-	if (type && strlen(type) >= THERMAL_NAME_LENGTH) {
-		pr_err("Error: Thermal zone name (%s) too long, should be under %d chars\n",
-		       type, THERMAL_NAME_LENGTH);
+	if (strlen(type) >= THERMAL_NAME_LENGTH) {
+		pr_err("Error: Thermal zone name over %d chars: %s\n",
+			THERMAL_NAME_LENGTH, type);
 		return ERR_PTR(-EINVAL);
 	}
 
 	if (trips > THERMAL_MAX_TRIPS || trips < 0 || mask >> trips) {
-		pr_err("Error: Incorrect number of thermal trips\n");
+		pr_err("Error: Incorrect number of thermal trips: %s\n", type);
 		return ERR_PTR(-EINVAL);
 	}
 
 	if (!ops) {
-		pr_err("Error: Thermal zone device ops not defined\n");
+		pr_err("Error: Thermal zone device ops not defined: %s\n",
+			type);
 		return ERR_PTR(-EINVAL);
 	}
 
-	if (trips > 0 && (!ops->get_trip_type || !ops->get_trip_temp))
+	if (trips > 0 && (!ops->get_trip_type || !ops->get_trip_temp)) {
+		pr_err("Error: Thermal zone device missing callback: %s\n",
+			type);
 		return ERR_PTR(-EINVAL);
+	}
 
 	tz = kzalloc(sizeof(*tz), GFP_KERNEL);
 	if (!tz)
Signature · 1 line
2.24.0.393.g34dc348eaf-goog
Reviewed-by on v3 2/3
#7 lore raw

Re: [PATCH v3 2/3] thermal: improve error message in thermal zone registration

Quoted 6 lines · Wei Wang

Follow up with commit 67eed44b8a8a ("thermal: Add some error messages") to clean up checkpatch warning on line length and also add more message for developers.

Quoted 52 lines
---
 drivers/thermal/thermal_core.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
drivers/thermal/thermal_core.c +11 -8
index 974e2d91c30b..9db7f72e70f8 100644
@@ -1255,29 +1255,33 @@ thermal_zone_device_register(const char *type, int trips, int mask,
        int count;
        struct thermal_governor *governor;
 
-       if (!type || strlen(type) == 0) {
+       if (!type || !type[0]) {
                pr_err("Error: No thermal zone type defined\n");
                return ERR_PTR(-EINVAL);
        }
 
-       if (type && strlen(type) >= THERMAL_NAME_LENGTH) {
-               pr_err("Error: Thermal zone name (%s) too long, should be under %d chars\n",
-                      type, THERMAL_NAME_LENGTH);
+       if (strlen(type) >= THERMAL_NAME_LENGTH) {
+               pr_err("Error: Thermal zone name over %d chars: %s\n",
+                       THERMAL_NAME_LENGTH, type);
                return ERR_PTR(-EINVAL);
        }
 
        if (trips > THERMAL_MAX_TRIPS || trips < 0 || mask >> trips) {
-               pr_err("Error: Incorrect number of thermal trips\n");
+               pr_err("Error: Incorrect number of thermal trips: %s\n", type);
                return ERR_PTR(-EINVAL);
        }
 
        if (!ops) {
-               pr_err("Error: Thermal zone device ops not defined\n");
+               pr_err("Error: Thermal zone device ops not defined: %s\n",
+                       type);
                return ERR_PTR(-EINVAL);
        }
 
-       if (trips > 0 && (!ops->get_trip_type || !ops->get_trip_temp))
+       if (trips > 0 && (!ops->get_trip_type || !ops->get_trip_temp)) {
+               pr_err("Error: Thermal zone device missing callback: %s\n",
+                       type);
                return ERR_PTR(-EINVAL);
+       }
 
        tz = kzalloc(sizeof(*tz), GFP_KERNEL);
        if (!tz)
--

2.24.0.393.g34dc348eaf-goog

author
#8 lore raw

PATCH v3 3/3 thermal: create softlink by name for thermal_zone and cooling_device

In reply to Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

---
 drivers/thermal/thermal_core.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)
drivers/thermal/thermal_core.c +20 -1
index 9db7f72e70f8..7872bd527f3f 100644
@@ -22,6 +22,7 @@
 #include <net/netlink.h>
 #include <net/genetlink.h>
 #include <linux/suspend.h>
+#include <linux/kobject.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/thermal.h>
@@ -46,6 +47,8 @@ static DEFINE_MUTEX(poweroff_lock);
 
 static atomic_t in_suspend;
 static bool power_off_triggered;
+static struct kobject *cdev_link_kobj;
+static struct kobject *tz_link_kobj;
 
 static struct thermal_governor *def_governor;
 
@@ -999,9 +1002,15 @@ __thermal_cooling_device_register(struct device_node *np,
 		return ERR_PTR(result);
 	}
 
-	/* Add 'this' new cdev to the global cdev list */
+	/* Add 'this' new cdev to the global cdev list and create link*/
 	mutex_lock(&thermal_list_lock);
 	list_add(&cdev->node, &thermal_cdev_list);
+	if (!cdev_link_kobj)
+		cdev_link_kobj = kobject_create_and_add("cdev-by-name",
+						cdev->device.kobj.parent);
+	if (!cdev_link_kobj || sysfs_create_link(cdev_link_kobj,
+						&cdev->device.kobj, cdev->type))
+		dev_err(&cdev->device, "Failed to create cdev-by-name link\n");
 	mutex_unlock(&thermal_list_lock);
 
 	/* Update binding information for 'this' new cdev */
@@ -1167,6 +1176,8 @@ void thermal_cooling_device_unregister(struct thermal_cooling_device *cdev)
 			}
 		}
 	}
+	if (cdev_link_kobj)
+		sysfs_remove_link(cdev_link_kobj, cdev->type);
 
 	mutex_unlock(&thermal_list_lock);
 
@@ -1354,6 +1365,12 @@ thermal_zone_device_register(const char *type, int trips, int mask,
 
 	mutex_lock(&thermal_list_lock);
 	list_add_tail(&tz->node, &thermal_tz_list);
+	if (!tz_link_kobj)
+		tz_link_kobj = kobject_create_and_add("tz-by-name",
+						tz->device.kobj.parent);
+	if (!tz_link_kobj || sysfs_create_link(tz_link_kobj,
+						&tz->device.kobj, tz->type))
+		dev_err(&tz->device, "Failed to create tz-by-name link\n");
 	mutex_unlock(&thermal_list_lock);
 
 	/* Bind cooling devices for this zone */
@@ -1425,6 +1442,8 @@ void thermal_zone_device_unregister(struct thermal_zone_device *tz)
 			}
 		}
 	}
+	if (tz_link_kobj)
+		sysfs_remove_link(tz_link_kobj, tz->type);
 
 	mutex_unlock(&thermal_list_lock);
 
Signature · 1 line
2.24.0.393.g34dc348eaf-goog
Reviewed-by Tested-by on v3 3/3
#9 lore raw

Re: [PATCH v3 3/3] thermal: create softlink by name for thermal_zone and cooling_device

Quoted 5 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

This is useful, especially for systems with plenty of thermal zones :)

Quoted 67 lines
---
 drivers/thermal/thermal_core.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)
drivers/thermal/thermal_core.c +18 -1
index 9db7f72e70f8..7872bd527f3f 100644
@@ -22,6 +22,7 @@
 #include <net/netlink.h>
 #include <net/genetlink.h>
 #include <linux/suspend.h>
+#include <linux/kobject.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/thermal.h>
@@ -46,6 +47,8 @@ static DEFINE_MUTEX(poweroff_lock);
 
 static atomic_t in_suspend;
 static bool power_off_triggered;
+static struct kobject *cdev_link_kobj;
+static struct kobject *tz_link_kobj;
 
 static struct thermal_governor *def_governor;
 
@@ -999,9 +1002,15 @@ __thermal_cooling_device_register(struct device_node *np,
 		return ERR_PTR(result);
 	}
 
-	/* Add 'this' new cdev to the global cdev list */
+	/* Add 'this' new cdev to the global cdev list and create link*/
 	mutex_lock(&thermal_list_lock);
 	list_add(&cdev->node, &thermal_cdev_list);
+	if (!cdev_link_kobj)
+		cdev_link_kobj = kobject_create_and_add("cdev-by-name",
+						cdev->device.kobj.parent);
+	if (!cdev_link_kobj || sysfs_create_link(cdev_link_kobj,
+						&cdev->device.kobj, cdev->type))
+		dev_err(&cdev->device, "Failed to create cdev-by-name link\n");
 	mutex_unlock(&thermal_list_lock);
 
 	/* Update binding information for 'this' new cdev */
@@ -1167,6 +1176,8 @@ void thermal_cooling_device_unregister(struct thermal_cooling_device *cdev)
 			}
 		}
 	}
+	if (cdev_link_kobj)
+		sysfs_remove_link(cdev_link_kobj, cdev->type);
 
 	mutex_unlock(&thermal_list_lock);
 
@@ -1354,6 +1365,12 @@ thermal_zone_device_register(const char *type, int trips, int mask,
 
 	mutex_lock(&thermal_list_lock);
 	list_add_tail(&tz->node, &thermal_tz_list);
+	if (!tz_link_kobj)
+		tz_link_kobj = kobject_create_and_add("tz-by-name",
+						tz->device.kobj.parent);
+	if (!tz_link_kobj || sysfs_create_link(tz_link_kobj,
+						&tz->device.kobj, tz->type))
+		dev_err(&tz->device, "Failed to create tz-by-name link\n");
 	mutex_unlock(&thermal_list_lock);
 
 	/* Bind cooling devices for this zone */

drivers/thermal/thermal_core.c void thermal_zone_device_unregister(struct thermal_zone_device *tz) · line 1449

Quoted 8 lines
drivers/thermal/thermal_core.c +2 -0
@@ -1425,6 +1442,8 @@ void thermal_zone_device_unregister(struct thermal_zone_device *tz)
 			}
 		}
 	}
+	if (tz_link_kobj)
+		sysfs_remove_link(tz_link_kobj, tz->type);
 
 	mutex_unlock(&thermal_list_lock);
 
Tested-by on v3 3/3
#10 lore raw

Re: [PATCH v3 3/3] thermal: create softlink by name for thermal_zone and cooling_device

In reply to Wei Wang
Quoted 9 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

Quoted 76 lines
---
 drivers/thermal/thermal_core.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)
drivers/thermal/thermal_core.c +20 -2
index 9db7f72e70f8..7872bd527f3f 100644
@@ -22,6 +22,7 @@
 #include <net/netlink.h>
 #include <net/genetlink.h>
 #include <linux/suspend.h>
+#include <linux/kobject.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/thermal.h>
@@ -46,6 +47,8 @@ static DEFINE_MUTEX(poweroff_lock);
 
 static atomic_t in_suspend;
 static bool power_off_triggered;
+static struct kobject *cdev_link_kobj;
+static struct kobject *tz_link_kobj;
 
 static struct thermal_governor *def_governor;
 
@@ -999,9 +1002,15 @@ __thermal_cooling_device_register(struct device_node *np,
                return ERR_PTR(result);
        }
 
-       /* Add 'this' new cdev to the global cdev list */
+       /* Add 'this' new cdev to the global cdev list and create link*/
        mutex_lock(&thermal_list_lock);
        list_add(&cdev->node, &thermal_cdev_list);
+       if (!cdev_link_kobj)
+               cdev_link_kobj = kobject_create_and_add("cdev-by-name",
+                                               cdev->device.kobj.parent);
+       if (!cdev_link_kobj || sysfs_create_link(cdev_link_kobj,
+                                               &cdev->device.kobj, cdev->type))
+               dev_err(&cdev->device, "Failed to create cdev-by-name link\n");
        mutex_unlock(&thermal_list_lock);
 
        /* Update binding information for 'this' new cdev */
@@ -1167,6 +1176,8 @@ void thermal_cooling_device_unregister(struct thermal_cooling_device *cdev)
                        }
                }
        }
+       if (cdev_link_kobj)
+               sysfs_remove_link(cdev_link_kobj, cdev->type);
 
        mutex_unlock(&thermal_list_lock);
 
@@ -1354,6 +1365,12 @@ thermal_zone_device_register(const char *type, int trips, int mask,
 
        mutex_lock(&thermal_list_lock);
        list_add_tail(&tz->node, &thermal_tz_list);
+       if (!tz_link_kobj)
+               tz_link_kobj = kobject_create_and_add("tz-by-name",
+                                               tz->device.kobj.parent);
+       if (!tz_link_kobj || sysfs_create_link(tz_link_kobj,
+                                               &tz->device.kobj, tz->type))
+               dev_err(&tz->device, "Failed to create tz-by-name link\n");
        mutex_unlock(&thermal_list_lock);
 
        /* Bind cooling devices for this zone */
@@ -1425,6 +1442,8 @@ void thermal_zone_device_unregister(struct thermal_zone_device *tz)
                        }
                }
        }
+       if (tz_link_kobj)
+               sysfs_remove_link(tz_link_kobj, tz->type);
 
        mutex_unlock(&thermal_list_lock);
 
--

2.24.0.393.g34dc348eaf-goog

#11 lore raw
In reply to Wei Wang
Quoted 5 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

Instead of adding another ABI, I suggest we put the current one deprecated with a warning in the dmesg, update the documentation and change the name the next version.

Quoted 16 lines

Changes since v1 [1]:
* Split cooling device registration into a seperate patch
Changes since v2 [2]:
* Split improve error message in thermal zone registration

[1]: v1: https://lore.kernel.org/patchwork/patch/1139450/
[2]: v2: https://lkml.org/lkml/2019/12/4/1291

Wei Wang (3):
thermal: prevent cooling device with no type to be registered
thermal: improve error message in thermal zone registration
thermal: create softlink by name for thermal_zone and cooling_device

 drivers/thermal/thermal_core.c | 55 +++++++++++++++++++++++++++-------
 1 file changed, 44 insertions(+), 11 deletions(-)
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
author
#12 lore raw
Quoted 8 lines · Daniel Lezcano
Quoted 5 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

Quoted 5 lines

Instead of adding another ABI, I suggest we put the current one deprecated with a warning in the dmesg, update the documentation and change the name the next version.

IMHO, we should keep the existing path which is a common pattern for sysfs interface. There are reasons we need couple thermal zone and cooling device in one class, but might be worth considering split as the latter might be used for other purposes e.g. battery current limit for preventive vdrop prevention. By nature, thermal zone are sensors, and cooling devices are usually components with potential high power use.

Quoted 25 lines
Quoted 8 lines

Changes since v1 [1]:
* Split cooling device registration into a seperate patch
Changes since v2 [2]:
* Split improve error message in thermal zone registration

[1]: v1: https://lore.kernel.org/patchwork/patch/1139450/
[2]: v2: https://lkml.org/lkml/2019/12/4/1291

Quoted 8 lines

Wei Wang (3):
thermal: prevent cooling device with no type to be registered
thermal: improve error message in thermal zone registration
thermal: create softlink by name for thermal_zone and cooling_device

 drivers/thermal/thermal_core.c | 55 +++++++++++++++++++++++++++-------
 1 file changed, 44 insertions(+), 11 deletions(-)

--
<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

#13 lore raw
Quoted 16 lines · Wei Wang
Quoted 8 lines · Daniel Lezcano
Quoted 5 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

Quoted 5 lines

Instead of adding another ABI, I suggest we put the current one deprecated with a warning in the dmesg, update the documentation and change the name the next version.

Quoted 7 lines

IMHO, we should keep the existing path which is a common pattern for sysfs interface. There are reasons we need couple thermal zone and cooling device in one class, but might be worth considering split as the latter might be used for other purposes e.g. battery current limit for preventive vdrop prevention. By nature, thermal zone are sensors, and cooling devices are usually components with potential high power use.

[Added Greghk and Rafael in Cc]

I understand but I would like to have Greg's and Rafael's opinion on that.

The result is:

ls -al /sys/devices/virtual/thermal/
total 0
drwxr-xr-x 22 root root 0 Dec 10 12:34 .
drwxr-xr-x 15 root root 0 Dec 10 12:34 ..

drwxr-xr-x  2 root root 0 Dec 10 13:18 cdev-by-name
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device0
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device1
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device10
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device11
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device12
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device13
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device14
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device15
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device2
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device3
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device4
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device5
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device6
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device7
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device8
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device9
drwxr-xr-x  3 root root 0 Dec 10 12:34 thermal_zone0
drwxr-xr-x  3 root root 0 Dec 10 12:34 thermal_zone1
drwxr-xr-x  2 root root 0 Dec 10 13:18 tz-by-name

ls -al /sys/devices/virtual/thermal/cdev-by-name/ total 0
drwxr-xr-x 2 root root 0 Dec 10 13:18 . drwxr-xr-x 22 root root 0 Dec 10 12:34 .. lrwxrwxrwx 1 root root 0 Dec 10 13:18 thermal-cpufreq-0 -> ../cooling_device0
lrwxrwxrwx 1 root root 0 Dec 10 13:18 thermal-cpufreq-1 -> ../cooling_device1

lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-0 -> ../cooling_device2
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-1 -> ../cooling_device3
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-10 ->

../cooling_device12
lrwxrwxrwx 1 root root 0 Dec 10 13:18 thermal-idle-11 -> ../cooling_device13
lrwxrwxrwx 1 root root 0 Dec 10 13:18 thermal-idle-12 -> ../cooling_device14
lrwxrwxrwx 1 root root 0 Dec 10 13:18 thermal-idle-13 -> ../cooling_device15

lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-2 -> ../cooling_device4
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-3 -> ../cooling_device5
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-4 -> ../cooling_device6
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-5 -> ../cooling_device7
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-6 -> ../cooling_device8
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-7 -> ../cooling_device9
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-8 -> ../cooling_device10
lrwxrwxrwx  1 root root 0 Dec 10 13:18 thermal-idle-9 -> ../cooling_device11

ls -al /sys/devices/virtual/thermal/tz-by-name/ total 0
drwxr-xr-x 2 root root 0 Dec 10 13:18 .
drwxr-xr-x 22 root root 0 Dec 10 12:34 ..
lrwxrwxrwx 1 root root 0 Dec 10 20:53 cpu -> ../thermal_zone0 lrwxrwxrwx 1 root root 0 Dec 10 20:53 gpu -> ../thermal_zone1

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
#14 lore raw
Quoted 29 lines · Daniel Lezcano
Quoted 16 lines · Wei Wang
Quoted 8 lines · Daniel Lezcano
Quoted 5 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.
The patch creates tz-by-name' and cdev-by-name' for thermal zone and cooling_device respectively.

Quoted 5 lines

Instead of adding another ABI, I suggest we put the current one deprecated with a warning in the dmesg, update the documentation and change the name the next version.

Quoted 7 lines

IMHO, we should keep the existing path which is a common pattern for sysfs interface. There are reasons we need couple thermal zone and cooling device in one class, but might be worth considering split as the latter might be used for other purposes e.g. battery current limit for preventive vdrop prevention. By nature, thermal zone are sensors, and cooling devices are usually components with potential high power use.

[Added Greghk and Rafael in Cc]

I understand but I would like to have Greg's and Rafael's opinion on that.

Quoted 7 lines

The result is:

ls -al /sys/devices/virtual/thermal/
total 0
drwxr-xr-x 22 root root 0 Dec 10 12:34 . drwxr-xr-x 15 root root 0 Dec 10 12:34 .. drwxr-xr-x 2 root root 0 Dec 10 13:18 cdev-by-name

Ick!

Quoted 20 lines
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device0
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device1
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device10
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device11
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device12
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device13
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device14
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device15
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device2
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device3
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device4
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device5
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device6
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device7
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device8
drwxr-xr-x  3 root root 0 Dec 10 12:34 cooling_device9
drwxr-xr-x  3 root root 0 Dec 10 12:34 thermal_zone0
drwxr-xr-x  3 root root 0 Dec 10 12:34 thermal_zone1
drwxr-xr-x  2 root root 0 Dec 10 13:18 tz-by-name
Quoted 6 lines

ls -al /sys/devices/virtual/thermal/cdev-by-name/ total 0
drwxr-xr-x 2 root root 0 Dec 10 13:18 . drwxr-xr-x 22 root root 0 Dec 10 12:34 .. lrwxrwxrwx 1 root root 0 Dec 10 13:18 thermal-cpufreq-0 -> ../cooling_device0

Ick ick ick!

What is this all for?

Where is the Documentation/ABI/ entry trying to explain this mess? Please provide that and we can go from there as I have no idea what this is trying to "help with".

thanks,

greg k-h

#15 lore raw
In reply to Daniel Lezcano
Quoted 8 lines · Daniel Lezcano
Quoted 7 lines · Wei Wang
Quoted 5 lines · Daniel Lezcano
Quoted 3 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.

tree traversal is supposed to be done in userspace code :)

But what userspace code needs to do this, and for what?

thanks,

greg k-h

author
#16 lore raw
Quoted 6 lines · Greg Kroah-Hartman
Quoted 4 lines · Daniel Lezcano
Quoted 3 lines · Wei Wang
Quoted 1 line · Daniel Lezcano
Quoted 7 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines · Wei Wang
Quoted 3 lines

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.

tree traversal is supposed to be done in userspace code :)

Yes, that can be done in userspace, but given the amount of thermal zones we have in some mobile devices, this will bring a lot of convenience.

e.g. this is on Pixel 4 XL:
coral:/ # ls /sys/devices/virtual/thermal/
cdev-by-name cooling_device15 cooling_device22 cooling_device3
cooling_device9 thermal_zone15 thermal_zone22 thermal_zone3

thermal_zone37  thermal_zone44  thermal_zone51  thermal_zone59
thermal_zone66  thermal_zone73  thermal_zone80  thermal_zone88
cooling_device0   cooling_device16  cooling_device23  cooling_device30
 thermal_zone0    thermal_zone16  thermal_zone23  thermal_zone30
thermal_zone38  thermal_zone45  thermal_zone52  thermal_zone6
thermal_zone67  thermal_zone74  thermal_zone81  thermal_zone9
cooling_device1   cooling_device17  cooling_device24  cooling_device31
 thermal_zone1    thermal_zone17  thermal_zone24  thermal_zone31
thermal_zone39  thermal_zone46  thermal_zone53  thermal_zone60
thermal_zone68  thermal_zone75  thermal_zone82  tz-by-name
cooling_device10  cooling_device18  cooling_device25  cooling_device4
 thermal_zone10   thermal_zone18  thermal_zone25  thermal_zone32
thermal_zone4   thermal_zone47  thermal_zone54  thermal_zone61
thermal_zone69  thermal_zone76  thermal_zone83
cooling_device11  cooling_device19  cooling_device26  cooling_device5
 thermal_zone11   thermal_zone19  thermal_zone26  thermal_zone33
thermal_zone40  thermal_zone48  thermal_zone55  thermal_zone62
thermal_zone7   thermal_zone77  thermal_zone84
cooling_device12  cooling_device2   cooling_device27  cooling_device6
 thermal_zone12   thermal_zone2   thermal_zone27  thermal_zone34
thermal_zone41  thermal_zone49  thermal_zone56  thermal_zone63
thermal_zone70  thermal_zone78  thermal_zone85
cooling_device13  cooling_device20  cooling_device28  cooling_device7
 thermal_zone13   thermal_zone20  thermal_zone28  thermal_zone35
thermal_zone42  thermal_zone5   thermal_zone57  thermal_zone64
thermal_zone71  thermal_zone79  thermal_zone86
cooling_device14  cooling_device21  cooling_device29  cooling_device8
 thermal_zone14   thermal_zone21  thermal_zone29  thermal_zone36
thermal_zone43  thermal_zone50  thermal_zone58  thermal_zone65
thermal_zone72  thermal_zone8   thermal_zone87
Quoted 1 line

But what userspace code needs to do this, and for what?

In Android, thermal daemon and thermal HAL as well as some init.rc script would use those thermal paths for managing and monitoring thermal. The daemon/HAL could have logic pipled in, however Android's init.rc script would be really tricky.
On a related note, we also create /dev/block/by-name links from userspace.

Thanks!
-Wei

Quoted 4 lines

thanks,

greg k-h

#17 lore raw
Quoted 13 lines · Wei Wang
Quoted 11 lines · Greg Kroah-Hartman
Quoted 8 lines · Daniel Lezcano
Quoted 7 lines · Wei Wang
Quoted 5 lines · Daniel Lezcano
Quoted 3 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.

Quoted 5 lines
Quoted 2 lines

tree traversal is supposed to be done in userspace code :)

Yes, that can be done in userspace, but given the amount of thermal zones we have in some mobile devices, this will bring a lot of convenience.

But usually all these thermal zones are fixed and building the name<->tz association is just a question of a few lines of code from userspace, no? What would be the benefit of adding more ABI?

If there is a thermal zone change, then a notification can be used, so the userspace code rebuild the name<->tz, no?

Quoted 49 lines

e.g. this is on Pixel 4 XL:
coral:/ # ls /sys/devices/virtual/thermal/
cdev-by-name cooling_device15 cooling_device22 cooling_device3
cooling_device9 thermal_zone15 thermal_zone22 thermal_zone3

thermal_zone37  thermal_zone44  thermal_zone51  thermal_zone59
thermal_zone66  thermal_zone73  thermal_zone80  thermal_zone88
cooling_device0   cooling_device16  cooling_device23  cooling_device30
 thermal_zone0    thermal_zone16  thermal_zone23  thermal_zone30
thermal_zone38  thermal_zone45  thermal_zone52  thermal_zone6
thermal_zone67  thermal_zone74  thermal_zone81  thermal_zone9
cooling_device1   cooling_device17  cooling_device24  cooling_device31
 thermal_zone1    thermal_zone17  thermal_zone24  thermal_zone31
thermal_zone39  thermal_zone46  thermal_zone53  thermal_zone60
thermal_zone68  thermal_zone75  thermal_zone82  tz-by-name
cooling_device10  cooling_device18  cooling_device25  cooling_device4
 thermal_zone10   thermal_zone18  thermal_zone25  thermal_zone32
thermal_zone4   thermal_zone47  thermal_zone54  thermal_zone61
thermal_zone69  thermal_zone76  thermal_zone83
cooling_device11  cooling_device19  cooling_device26  cooling_device5
 thermal_zone11   thermal_zone19  thermal_zone26  thermal_zone33
thermal_zone40  thermal_zone48  thermal_zone55  thermal_zone62
thermal_zone7   thermal_zone77  thermal_zone84
cooling_device12  cooling_device2   cooling_device27  cooling_device6
 thermal_zone12   thermal_zone2   thermal_zone27  thermal_zone34
thermal_zone41  thermal_zone49  thermal_zone56  thermal_zone63
thermal_zone70  thermal_zone78  thermal_zone85
cooling_device13  cooling_device20  cooling_device28  cooling_device7
 thermal_zone13   thermal_zone20  thermal_zone28  thermal_zone35
thermal_zone42  thermal_zone5   thermal_zone57  thermal_zone64
thermal_zone71  thermal_zone79  thermal_zone86
cooling_device14  cooling_device21  cooling_device29  cooling_device8
 thermal_zone14   thermal_zone21  thermal_zone29  thermal_zone36
thermal_zone43  thermal_zone50  thermal_zone58  thermal_zone65
thermal_zone72  thermal_zone8   thermal_zone87
Quoted 1 line

But what userspace code needs to do this, and for what?

In Android, thermal daemon and thermal HAL as well as some init.rc script would use those thermal paths for managing and monitoring thermal. The daemon/HAL could have logic pipled in, however Android's init.rc script would be really tricky.
On a related note, we also create /dev/block/by-name links from userspace.

Thanks!
-Wei

Quoted 4 lines

thanks,

greg k-h

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
author
#18 lore raw
Quoted 17 lines · Daniel Lezcano
Quoted 15 lines · Wei Wang
Quoted 13 lines · Greg Kroah-Hartman
Quoted 8 lines · Daniel Lezcano
Quoted 7 lines · Wei Wang
Quoted 5 lines · Daniel Lezcano
Quoted 3 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.

tree traversal is supposed to be done in userspace code :)

Quoted 7 lines
Quoted 3 lines

Yes, that can be done in userspace, but given the amount of thermal zones we have in some mobile devices, this will bring a lot of convenience.

But usually all these thermal zones are fixed and building the name<->tz association is just a question of a few lines of code from userspace, no? What would be the benefit of adding more ABI?

code reuse as tz mapping is per device tree setup and we try to use the same _type_ in thermal zone in shared client code for Android thermal API.

Quoted 3 lines

If there is a thermal zone change, then a notification can be used, so the userspace code rebuild the name<->tz, no?

Assuming you meant for type<->tz, this mapping won't change in runtime in our case (not unloading things), but it could be changed due to device tree change, e.g. adding/removing thermal zones.

Quoted 59 lines
Quoted 38 lines

e.g. this is on Pixel 4 XL:
coral:/ # ls /sys/devices/virtual/thermal/
cdev-by-name cooling_device15 cooling_device22 cooling_device3
cooling_device9 thermal_zone15 thermal_zone22 thermal_zone3

thermal_zone37  thermal_zone44  thermal_zone51  thermal_zone59
thermal_zone66  thermal_zone73  thermal_zone80  thermal_zone88
cooling_device0   cooling_device16  cooling_device23  cooling_device30
 thermal_zone0    thermal_zone16  thermal_zone23  thermal_zone30
thermal_zone38  thermal_zone45  thermal_zone52  thermal_zone6
thermal_zone67  thermal_zone74  thermal_zone81  thermal_zone9
cooling_device1   cooling_device17  cooling_device24  cooling_device31
 thermal_zone1    thermal_zone17  thermal_zone24  thermal_zone31
thermal_zone39  thermal_zone46  thermal_zone53  thermal_zone60
thermal_zone68  thermal_zone75  thermal_zone82  tz-by-name
cooling_device10  cooling_device18  cooling_device25  cooling_device4
 thermal_zone10   thermal_zone18  thermal_zone25  thermal_zone32
thermal_zone4   thermal_zone47  thermal_zone54  thermal_zone61
thermal_zone69  thermal_zone76  thermal_zone83
cooling_device11  cooling_device19  cooling_device26  cooling_device5
 thermal_zone11   thermal_zone19  thermal_zone26  thermal_zone33
thermal_zone40  thermal_zone48  thermal_zone55  thermal_zone62
thermal_zone7   thermal_zone77  thermal_zone84
cooling_device12  cooling_device2   cooling_device27  cooling_device6
 thermal_zone12   thermal_zone2   thermal_zone27  thermal_zone34
thermal_zone41  thermal_zone49  thermal_zone56  thermal_zone63
thermal_zone70  thermal_zone78  thermal_zone85
cooling_device13  cooling_device20  cooling_device28  cooling_device7
 thermal_zone13   thermal_zone20  thermal_zone28  thermal_zone35
thermal_zone42  thermal_zone5   thermal_zone57  thermal_zone64
thermal_zone71  thermal_zone79  thermal_zone86
cooling_device14  cooling_device21  cooling_device29  cooling_device8
 thermal_zone14   thermal_zone21  thermal_zone29  thermal_zone36
thermal_zone43  thermal_zone50  thermal_zone58  thermal_zone65
thermal_zone72  thermal_zone8   thermal_zone87
Quoted 1 line

But what userspace code needs to do this, and for what?

Quoted 12 lines

In Android, thermal daemon and thermal HAL as well as some init.rc script would use those thermal paths for managing and monitoring thermal. The daemon/HAL could have logic pipled in, however Android's init.rc script would be really tricky.
On a related note, we also create /dev/block/by-name links from userspace.

Thanks!
-Wei

Quoted 4 lines

thanks,

greg k-h

--
<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

#19 lore raw

Hi Wei,

(resending because I accidentally replied only to you. Also adding everyone else back in the thread, not sure what happened before).

Quoted 23 lines · Wei Wang
Quoted 21 lines · Daniel Lezcano
Quoted 18 lines · Wei Wang
Quoted 6 lines · Greg Kroah-Hartman
Quoted 4 lines · Daniel Lezcano
Quoted 3 lines · Wei Wang
Quoted 1 line · Daniel Lezcano
Quoted 7 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines · Wei Wang
Quoted 3 lines

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.

tree traversal is supposed to be done in userspace code :)

Yes, that can be done in userspace, but given the amount of thermal zones we have in some mobile devices, this will bring a lot of convenience.

Quoted 8 lines
Quoted 3 lines

But usually all these thermal zones are fixed and building the name<->tz association is just a question of a few lines of code from userspace, no? What would be the benefit of adding more ABI?

code reuse as tz mapping is per device tree setup and we try to use the same _type_ in thermal zone in shared client code for Android thermal API.

CMIIW, but the tz mapping today can still be done by parsing through the directory the other way? (i.e. tz->type) instead of type->tz?

Also, I understand the resistance to add a new ABI since it will have to be kept around forever. What would be worth knowing is if these mappings are static (built once) or you have to go through and parse the directories every time? AFAIU, the parsing (map creation) seems to only happen once during boot, so we (Android) should be ok?

Quoted 9 lines
Quoted 4 lines

If there is a thermal zone change, then a notification can be used, so the userspace code rebuild the name<->tz, no?

Assuming you meant for type<->tz, this mapping won't change in runtime in our case (not unloading things), but it could be changed due to device tree change, e.g. adding/removing thermal zones.

I actually think we might need this after all. This assumes all tz devices are registered when userspace (thermal hal/daemon) is coming up. However, that may soon not be the case and we have to adjust for devices being added in parallel with userspace?

So, may be we need to start monitoring uevents like we do for power supplies too? In that case, the mapping can also be done dynamically.

Lastly, I think we can make a case further if there are benefits we can share here other than convenience. Like startup time, handing hardware variations etc? (Also point to the current userspace code that uses this)

- ssp

Quoted 61 lines
Quoted 61 lines
Quoted 40 lines

e.g. this is on Pixel 4 XL:
coral:/ # ls /sys/devices/virtual/thermal/
cdev-by-name cooling_device15 cooling_device22 cooling_device3
cooling_device9 thermal_zone15 thermal_zone22 thermal_zone3

thermal_zone37  thermal_zone44  thermal_zone51  thermal_zone59
thermal_zone66  thermal_zone73  thermal_zone80  thermal_zone88
cooling_device0   cooling_device16  cooling_device23  cooling_device30
 thermal_zone0    thermal_zone16  thermal_zone23  thermal_zone30
thermal_zone38  thermal_zone45  thermal_zone52  thermal_zone6
thermal_zone67  thermal_zone74  thermal_zone81  thermal_zone9
cooling_device1   cooling_device17  cooling_device24  cooling_device31
 thermal_zone1    thermal_zone17  thermal_zone24  thermal_zone31
thermal_zone39  thermal_zone46  thermal_zone53  thermal_zone60
thermal_zone68  thermal_zone75  thermal_zone82  tz-by-name
cooling_device10  cooling_device18  cooling_device25  cooling_device4
 thermal_zone10   thermal_zone18  thermal_zone25  thermal_zone32
thermal_zone4   thermal_zone47  thermal_zone54  thermal_zone61
thermal_zone69  thermal_zone76  thermal_zone83
cooling_device11  cooling_device19  cooling_device26  cooling_device5
 thermal_zone11   thermal_zone19  thermal_zone26  thermal_zone33
thermal_zone40  thermal_zone48  thermal_zone55  thermal_zone62
thermal_zone7   thermal_zone77  thermal_zone84
cooling_device12  cooling_device2   cooling_device27  cooling_device6
 thermal_zone12   thermal_zone2   thermal_zone27  thermal_zone34
thermal_zone41  thermal_zone49  thermal_zone56  thermal_zone63
thermal_zone70  thermal_zone78  thermal_zone85
cooling_device13  cooling_device20  cooling_device28  cooling_device7
 thermal_zone13   thermal_zone20  thermal_zone28  thermal_zone35
thermal_zone42  thermal_zone5   thermal_zone57  thermal_zone64
thermal_zone71  thermal_zone79  thermal_zone86
cooling_device14  cooling_device21  cooling_device29  cooling_device8
 thermal_zone14   thermal_zone21  thermal_zone29  thermal_zone36
thermal_zone43  thermal_zone50  thermal_zone58  thermal_zone65
thermal_zone72  thermal_zone8   thermal_zone87
Quoted 1 line

But what userspace code needs to do this, and for what?

Quoted 12 lines

In Android, thermal daemon and thermal HAL as well as some init.rc script would use those thermal paths for managing and monitoring thermal. The daemon/HAL could have logic pipled in, however Android's init.rc script would be really tricky.
On a related note, we also create /dev/block/by-name links from userspace.

Thanks!
-Wei

Quoted 4 lines

thanks,

greg k-h

--
<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

author
#20 lore raw
Quoted 35 lines · Sandeep Patil

Hi Wei,

(resending because I accidentally replied only to you. Also adding everyone else back in the thread, not sure what happened before).

Quoted 28 lines · Wei Wang
Quoted 24 lines · Daniel Lezcano
Quoted 13 lines · Wei Wang
Quoted 11 lines · Greg Kroah-Hartman
Quoted 8 lines · Daniel Lezcano
Quoted 7 lines · Wei Wang
Quoted 5 lines · Daniel Lezcano
Quoted 3 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.

Quoted 5 lines
Quoted 2 lines

tree traversal is supposed to be done in userspace code :)

Yes, that can be done in userspace, but given the amount of thermal zones we have in some mobile devices, this will bring a lot of convenience.

But usually all these thermal zones are fixed and building the name<->tz association is just a question of a few lines of code from userspace, no? What would be the benefit of adding more ABI?

Quoted 7 lines
Quoted 3 lines

code reuse as tz mapping is per device tree setup and we try to use the same _type_ in thermal zone in shared client code for Android thermal API.

CMIIW, but the tz mapping today can still be done by parsing through the directory the other way? (i.e. tz->type) instead of type->tz?

Parsing can be done in clients, one particular place is for init, but still possible.
https://cs.android.com/search?q=tz-by-name%20file:%5C.rc
Yeah, as said, we can definitely do pasing in every userspace daemon, but AFAICT, all userspace codes (open source / closed source) are using thermal zones / cooling devices by "name/type". So I am probling the possibility of having this kind of support for convenience.
FYI, Chrome also feel this is useful
https://lore.kernel.org/lkml/[email protected]/

Quoted 6 lines

Also, I understand the resistance to add a new ABI since it will have to be kept around forever. What would be worth knowing is if these mappings are static (built once) or you have to go through and parse the directories every time? AFAIU, the parsing (map creation) seems to only happen once during boot, so we (Android) should be ok?

Those mapping are usually static nowadays, but could be changed if modules are unloaded. Right now we only do UEVENT for userspace governor,
https://github.com/torvalds/linux/blob/9c7db5004280767566e91a33445bf93aa479ef02/drivers/thermal/user_space.c#L37
Actually, I am about sending another RFC to add some more uevent to userspace. Without the UEVENT, I think we cannot guarantee the mapping is static.

Quoted 15 lines
Quoted 9 lines
Quoted 4 lines

If there is a thermal zone change, then a notification can be used, so the userspace code rebuild the name<->tz, no?

Assuming you meant for type<->tz, this mapping won't change in runtime in our case (not unloading things), but it could be changed due to device tree change, e.g. adding/removing thermal zones.

I actually think we might need this after all. This assumes all tz devices are registered when userspace (thermal hal/daemon) is coming up. However, that may soon not be the case and we have to adjust for devices being added in parallel with userspace?

Quoted 7 lines

So, may be we need to start monitoring uevents like we do for power supplies too? In that case, the mapping can also be done dynamically.

Lastly, I think we can make a case further if there are benefits we can share here other than convenience. Like startup time, handing hardware variations etc? (Also point to the current userspace code that uses this)

I replied above before reading this part, and I guess we had the same thought on this.

Quoted 62 lines

- ssp

Quoted 61 lines
Quoted 61 lines
Quoted 40 lines

e.g. this is on Pixel 4 XL:
coral:/ # ls /sys/devices/virtual/thermal/
cdev-by-name cooling_device15 cooling_device22 cooling_device3
cooling_device9 thermal_zone15 thermal_zone22 thermal_zone3

thermal_zone37  thermal_zone44  thermal_zone51  thermal_zone59
thermal_zone66  thermal_zone73  thermal_zone80  thermal_zone88
cooling_device0   cooling_device16  cooling_device23  cooling_device30
 thermal_zone0    thermal_zone16  thermal_zone23  thermal_zone30
thermal_zone38  thermal_zone45  thermal_zone52  thermal_zone6
thermal_zone67  thermal_zone74  thermal_zone81  thermal_zone9
cooling_device1   cooling_device17  cooling_device24  cooling_device31
 thermal_zone1    thermal_zone17  thermal_zone24  thermal_zone31
thermal_zone39  thermal_zone46  thermal_zone53  thermal_zone60
thermal_zone68  thermal_zone75  thermal_zone82  tz-by-name
cooling_device10  cooling_device18  cooling_device25  cooling_device4
 thermal_zone10   thermal_zone18  thermal_zone25  thermal_zone32
thermal_zone4   thermal_zone47  thermal_zone54  thermal_zone61
thermal_zone69  thermal_zone76  thermal_zone83
cooling_device11  cooling_device19  cooling_device26  cooling_device5
 thermal_zone11   thermal_zone19  thermal_zone26  thermal_zone33
thermal_zone40  thermal_zone48  thermal_zone55  thermal_zone62
thermal_zone7   thermal_zone77  thermal_zone84
cooling_device12  cooling_device2   cooling_device27  cooling_device6
 thermal_zone12   thermal_zone2   thermal_zone27  thermal_zone34
thermal_zone41  thermal_zone49  thermal_zone56  thermal_zone63
thermal_zone70  thermal_zone78  thermal_zone85
cooling_device13  cooling_device20  cooling_device28  cooling_device7
 thermal_zone13   thermal_zone20  thermal_zone28  thermal_zone35
thermal_zone42  thermal_zone5   thermal_zone57  thermal_zone64
thermal_zone71  thermal_zone79  thermal_zone86
cooling_device14  cooling_device21  cooling_device29  cooling_device8
 thermal_zone14   thermal_zone21  thermal_zone29  thermal_zone36
thermal_zone43  thermal_zone50  thermal_zone58  thermal_zone65
thermal_zone72  thermal_zone8   thermal_zone87
Quoted 1 line

But what userspace code needs to do this, and for what?

Quoted 12 lines

In Android, thermal daemon and thermal HAL as well as some init.rc script would use those thermal paths for managing and monitoring thermal. The daemon/HAL could have logic pipled in, however Android's init.rc script would be really tricky.
On a related note, we also create /dev/block/by-name links from userspace.

Thanks!
-Wei

Quoted 4 lines

thanks,

greg k-h

--
<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

#21 lore raw
In reply to Wei Wang
Quoted 13 lines · Wei Wang
Quoted 11 lines · Greg Kroah-Hartman
Quoted 8 lines · Daniel Lezcano
Quoted 7 lines · Wei Wang
Quoted 5 lines · Daniel Lezcano
Quoted 3 lines · Wei Wang

The paths thermal_zone%d and cooling_device%d are not intuitive and the numbers are subject to change due to device tree change. This usually leads to tree traversal in userspace code.

Quoted 5 lines
Quoted 2 lines

tree traversal is supposed to be done in userspace code :)

Yes, that can be done in userspace, but given the amount of thermal zones we have in some mobile devices, this will bring a lot of convenience.

Userspace is easier to write than kernel code. Kernel code and apis you have to retain for the next 20+ years, are you willing to do that?

Quoted 26 lines

e.g. this is on Pixel 4 XL:
coral:/ # ls /sys/devices/virtual/thermal/
cdev-by-name cooling_device15 cooling_device22 cooling_device3
cooling_device9 thermal_zone15 thermal_zone22 thermal_zone3

thermal_zone37  thermal_zone44  thermal_zone51  thermal_zone59
thermal_zone66  thermal_zone73  thermal_zone80  thermal_zone88
cooling_device0   cooling_device16  cooling_device23  cooling_device30
 thermal_zone0    thermal_zone16  thermal_zone23  thermal_zone30
thermal_zone38  thermal_zone45  thermal_zone52  thermal_zone6
thermal_zone67  thermal_zone74  thermal_zone81  thermal_zone9
cooling_device1   cooling_device17  cooling_device24  cooling_device31
 thermal_zone1    thermal_zone17  thermal_zone24  thermal_zone31
thermal_zone39  thermal_zone46  thermal_zone53  thermal_zone60
thermal_zone68  thermal_zone75  thermal_zone82  tz-by-name
cooling_device10  cooling_device18  cooling_device25  cooling_device4
 thermal_zone10   thermal_zone18  thermal_zone25  thermal_zone32
thermal_zone4   thermal_zone47  thermal_zone54  thermal_zone61
thermal_zone69  thermal_zone76  thermal_zone83
cooling_device11  cooling_device19  cooling_device26  cooling_device5
 thermal_zone11   thermal_zone19  thermal_zone26  thermal_zone33
thermal_zone40  thermal_zone48  thermal_zone55  thermal_zone62
thermal_zone7   thermal_zone77  thermal_zone84
cooling_device12  cooling_device2   cooling_device27  cooling_device6
 thermal_zone12   thermal_zone2   thermal_zone27  thermal_zone34
thermal_zone41  thermal_zone49  thermal_zone56  thermal_zone63
thermal_zone70  thermal_zone78  thermal_zone85
Quoted 8 lines

cooling_device13 cooling_device20 cooling_device28 cooling_device7
thermal_zone13 thermal_zone20 thermal_zone28 thermal_zone35

thermal_zone42  thermal_zone5   thermal_zone57  thermal_zone64
thermal_zone71  thermal_zone79  thermal_zone86
cooling_device14  cooling_device21  cooling_device29  cooling_device8
 thermal_zone14   thermal_zone21  thermal_zone29  thermal_zone36
thermal_zone43  thermal_zone50  thermal_zone58  thermal_zone65
thermal_zone72  thermal_zone8   thermal_zone87

Nice, you have a crazy SoC here, good luck! :)

Quoted 5 lines
Quoted 1 line

But what userspace code needs to do this, and for what?

In Android, thermal daemon and thermal HAL as well as some init.rc script would use those thermal paths for managing and monitoring thermal. The daemon/HAL could have logic pipled in, however Android's init.rc script would be really tricky.

Then put that logic in your userspace code.

Quoted 1 line

On a related note, we also create /dev/block/by-name links from userspace.

Yes we do, from userspace, not from within the kernel itself, so I think you have answered the question :)

thanks,

greg k-h