OpenCloudOS-Kernel/drivers/thermal/thermal_netlink.c

705 lines
18 KiB
C
Raw Normal View History

thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright 2020 Linaro Limited
*
* Author: Daniel Lezcano <daniel.lezcano@linaro.org>
*
* Generic netlink for thermal management framework
*/
#include <linux/module.h>
#include <linux/kernel.h>
#include <net/genetlink.h>
#include <uapi/linux/thermal.h>
#include "thermal_core.h"
static const struct genl_multicast_group thermal_genl_mcgrps[] = {
{ .name = THERMAL_GENL_SAMPLING_GROUP_NAME, },
{ .name = THERMAL_GENL_EVENT_GROUP_NAME, },
};
static const struct nla_policy thermal_genl_policy[THERMAL_GENL_ATTR_MAX + 1] = {
/* Thermal zone */
[THERMAL_GENL_ATTR_TZ] = { .type = NLA_NESTED },
[THERMAL_GENL_ATTR_TZ_ID] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_TEMP] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_TRIP] = { .type = NLA_NESTED },
[THERMAL_GENL_ATTR_TZ_TRIP_ID] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_TRIP_TEMP] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_TRIP_TYPE] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_TRIP_HYST] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_MODE] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_CDEV_WEIGHT] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_TZ_NAME] = { .type = NLA_STRING,
.len = THERMAL_NAME_LENGTH },
/* Governor(s) */
[THERMAL_GENL_ATTR_TZ_GOV] = { .type = NLA_NESTED },
[THERMAL_GENL_ATTR_TZ_GOV_NAME] = { .type = NLA_STRING,
.len = THERMAL_NAME_LENGTH },
/* Cooling devices */
[THERMAL_GENL_ATTR_CDEV] = { .type = NLA_NESTED },
[THERMAL_GENL_ATTR_CDEV_ID] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_CDEV_CUR_STATE] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_CDEV_MAX_STATE] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_CDEV_NAME] = { .type = NLA_STRING,
.len = THERMAL_NAME_LENGTH },
/* CPU capabilities */
[THERMAL_GENL_ATTR_CPU_CAPABILITY] = { .type = NLA_NESTED },
[THERMAL_GENL_ATTR_CPU_CAPABILITY_ID] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_CPU_CAPABILITY_PERFORMANCE] = { .type = NLA_U32 },
[THERMAL_GENL_ATTR_CPU_CAPABILITY_EFFICIENCY] = { .type = NLA_U32 },
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
};
struct param {
struct nlattr **attrs;
struct sk_buff *msg;
const char *name;
int tz_id;
int cdev_id;
int trip_id;
int trip_temp;
int trip_type;
int trip_hyst;
int temp;
int cdev_state;
int cdev_max_state;
struct thermal_genl_cpu_caps *cpu_capabilities;
int cpu_capabilities_count;
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
};
typedef int (*cb_t)(struct param *);
static struct genl_family thermal_gnl_family;
/************************** Sampling encoding *******************************/
int thermal_genl_sampling_temp(int id, int temp)
{
struct sk_buff *skb;
void *hdr;
skb = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
if (!skb)
return -ENOMEM;
hdr = genlmsg_put(skb, 0, 0, &thermal_gnl_family, 0,
THERMAL_GENL_SAMPLING_TEMP);
if (!hdr)
goto out_free;
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
if (nla_put_u32(skb, THERMAL_GENL_ATTR_TZ_ID, id))
goto out_cancel;
if (nla_put_u32(skb, THERMAL_GENL_ATTR_TZ_TEMP, temp))
goto out_cancel;
genlmsg_end(skb, hdr);
genlmsg_multicast(&thermal_gnl_family, skb, 0, 0, GFP_KERNEL);
return 0;
out_cancel:
genlmsg_cancel(skb, hdr);
out_free:
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
nlmsg_free(skb);
return -EMSGSIZE;
}
/**************************** Event encoding *********************************/
static int thermal_genl_event_tz_create(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_ID, p->tz_id) ||
nla_put_string(p->msg, THERMAL_GENL_ATTR_TZ_NAME, p->name))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_tz(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_ID, p->tz_id))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_tz_trip_up(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_ID, p->tz_id) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_TRIP_ID, p->trip_id) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_TEMP, p->temp))
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_tz_trip_add(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_ID, p->tz_id) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_TRIP_ID, p->trip_id) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_TRIP_TYPE, p->trip_type) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_TRIP_TEMP, p->trip_temp) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_TRIP_HYST, p->trip_hyst))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_tz_trip_delete(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_ID, p->tz_id) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_TRIP_ID, p->trip_id))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_cdev_add(struct param *p)
{
if (nla_put_string(p->msg, THERMAL_GENL_ATTR_CDEV_NAME,
p->name) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_CDEV_ID,
p->cdev_id) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_CDEV_MAX_STATE,
p->cdev_max_state))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_cdev_delete(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_CDEV_ID, p->cdev_id))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_cdev_state_update(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_CDEV_ID,
p->cdev_id) ||
nla_put_u32(p->msg, THERMAL_GENL_ATTR_CDEV_CUR_STATE,
p->cdev_state))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_gov_change(struct param *p)
{
if (nla_put_u32(p->msg, THERMAL_GENL_ATTR_TZ_ID, p->tz_id) ||
nla_put_string(p->msg, THERMAL_GENL_ATTR_GOV_NAME, p->name))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_event_cpu_capability_change(struct param *p)
{
struct thermal_genl_cpu_caps *cpu_cap = p->cpu_capabilities;
struct sk_buff *msg = p->msg;
struct nlattr *start_cap;
int i;
start_cap = nla_nest_start(msg, THERMAL_GENL_ATTR_CPU_CAPABILITY);
if (!start_cap)
return -EMSGSIZE;
for (i = 0; i < p->cpu_capabilities_count; ++i) {
if (nla_put_u32(msg, THERMAL_GENL_ATTR_CPU_CAPABILITY_ID,
cpu_cap->cpu))
goto out_cancel_nest;
if (nla_put_u32(msg, THERMAL_GENL_ATTR_CPU_CAPABILITY_PERFORMANCE,
cpu_cap->performance))
goto out_cancel_nest;
if (nla_put_u32(msg, THERMAL_GENL_ATTR_CPU_CAPABILITY_EFFICIENCY,
cpu_cap->efficiency))
goto out_cancel_nest;
++cpu_cap;
}
nla_nest_end(msg, start_cap);
return 0;
out_cancel_nest:
nla_nest_cancel(msg, start_cap);
return -EMSGSIZE;
}
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
int thermal_genl_event_tz_delete(struct param *p)
__attribute__((alias("thermal_genl_event_tz")));
int thermal_genl_event_tz_enable(struct param *p)
__attribute__((alias("thermal_genl_event_tz")));
int thermal_genl_event_tz_disable(struct param *p)
__attribute__((alias("thermal_genl_event_tz")));
int thermal_genl_event_tz_trip_down(struct param *p)
__attribute__((alias("thermal_genl_event_tz_trip_up")));
int thermal_genl_event_tz_trip_change(struct param *p)
__attribute__((alias("thermal_genl_event_tz_trip_add")));
static cb_t event_cb[] = {
[THERMAL_GENL_EVENT_TZ_CREATE] = thermal_genl_event_tz_create,
[THERMAL_GENL_EVENT_TZ_DELETE] = thermal_genl_event_tz_delete,
[THERMAL_GENL_EVENT_TZ_ENABLE] = thermal_genl_event_tz_enable,
[THERMAL_GENL_EVENT_TZ_DISABLE] = thermal_genl_event_tz_disable,
[THERMAL_GENL_EVENT_TZ_TRIP_UP] = thermal_genl_event_tz_trip_up,
[THERMAL_GENL_EVENT_TZ_TRIP_DOWN] = thermal_genl_event_tz_trip_down,
[THERMAL_GENL_EVENT_TZ_TRIP_CHANGE] = thermal_genl_event_tz_trip_change,
[THERMAL_GENL_EVENT_TZ_TRIP_ADD] = thermal_genl_event_tz_trip_add,
[THERMAL_GENL_EVENT_TZ_TRIP_DELETE] = thermal_genl_event_tz_trip_delete,
[THERMAL_GENL_EVENT_CDEV_ADD] = thermal_genl_event_cdev_add,
[THERMAL_GENL_EVENT_CDEV_DELETE] = thermal_genl_event_cdev_delete,
[THERMAL_GENL_EVENT_CDEV_STATE_UPDATE] = thermal_genl_event_cdev_state_update,
[THERMAL_GENL_EVENT_TZ_GOV_CHANGE] = thermal_genl_event_gov_change,
[THERMAL_GENL_EVENT_CPU_CAPABILITY_CHANGE] = thermal_genl_event_cpu_capability_change,
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
};
/*
* Generic netlink event encoding
*/
static int thermal_genl_send_event(enum thermal_genl_event event,
struct param *p)
{
struct sk_buff *msg;
int ret = -EMSGSIZE;
void *hdr;
msg = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
p->msg = msg;
hdr = genlmsg_put(msg, 0, 0, &thermal_gnl_family, 0, event);
if (!hdr)
goto out_free_msg;
ret = event_cb[event](p);
if (ret)
goto out_cancel_msg;
genlmsg_end(msg, hdr);
genlmsg_multicast(&thermal_gnl_family, msg, 0, 1, GFP_KERNEL);
return 0;
out_cancel_msg:
genlmsg_cancel(msg, hdr);
out_free_msg:
nlmsg_free(msg);
return ret;
}
int thermal_notify_tz_create(int tz_id, const char *name)
{
struct param p = { .tz_id = tz_id, .name = name };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_CREATE, &p);
}
int thermal_notify_tz_delete(int tz_id)
{
struct param p = { .tz_id = tz_id };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_DELETE, &p);
}
int thermal_notify_tz_enable(int tz_id)
{
struct param p = { .tz_id = tz_id };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_ENABLE, &p);
}
int thermal_notify_tz_disable(int tz_id)
{
struct param p = { .tz_id = tz_id };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_DISABLE, &p);
}
int thermal_notify_tz_trip_down(int tz_id, int trip_id, int temp)
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
{
struct param p = { .tz_id = tz_id, .trip_id = trip_id, .temp = temp };
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_TRIP_DOWN, &p);
}
int thermal_notify_tz_trip_up(int tz_id, int trip_id, int temp)
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
{
struct param p = { .tz_id = tz_id, .trip_id = trip_id, .temp = temp };
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_TRIP_UP, &p);
}
int thermal_notify_tz_trip_add(int tz_id, int trip_id, int trip_type,
int trip_temp, int trip_hyst)
{
struct param p = { .tz_id = tz_id, .trip_id = trip_id,
.trip_type = trip_type, .trip_temp = trip_temp,
.trip_hyst = trip_hyst };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_TRIP_ADD, &p);
}
int thermal_notify_tz_trip_delete(int tz_id, int trip_id)
{
struct param p = { .tz_id = tz_id, .trip_id = trip_id };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_TRIP_DELETE, &p);
}
int thermal_notify_tz_trip_change(int tz_id, int trip_id, int trip_type,
int trip_temp, int trip_hyst)
{
struct param p = { .tz_id = tz_id, .trip_id = trip_id,
.trip_type = trip_type, .trip_temp = trip_temp,
.trip_hyst = trip_hyst };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_TRIP_CHANGE, &p);
}
int thermal_notify_cdev_state_update(int cdev_id, int cdev_state)
{
struct param p = { .cdev_id = cdev_id, .cdev_state = cdev_state };
return thermal_genl_send_event(THERMAL_GENL_EVENT_CDEV_STATE_UPDATE, &p);
}
int thermal_notify_cdev_add(int cdev_id, const char *name, int cdev_max_state)
{
struct param p = { .cdev_id = cdev_id, .name = name,
.cdev_max_state = cdev_max_state };
return thermal_genl_send_event(THERMAL_GENL_EVENT_CDEV_ADD, &p);
}
int thermal_notify_cdev_delete(int cdev_id)
{
struct param p = { .cdev_id = cdev_id };
return thermal_genl_send_event(THERMAL_GENL_EVENT_CDEV_DELETE, &p);
}
int thermal_notify_tz_gov_change(int tz_id, const char *name)
{
struct param p = { .tz_id = tz_id, .name = name };
return thermal_genl_send_event(THERMAL_GENL_EVENT_TZ_GOV_CHANGE, &p);
}
int thermal_genl_cpu_capability_event(int count,
struct thermal_genl_cpu_caps *caps)
{
struct param p = { .cpu_capabilities_count = count, .cpu_capabilities = caps };
return thermal_genl_send_event(THERMAL_GENL_EVENT_CPU_CAPABILITY_CHANGE, &p);
}
EXPORT_SYMBOL_GPL(thermal_genl_cpu_capability_event);
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
/*************************** Command encoding ********************************/
static int __thermal_genl_cmd_tz_get_id(struct thermal_zone_device *tz,
void *data)
{
struct sk_buff *msg = data;
if (nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_ID, tz->id) ||
nla_put_string(msg, THERMAL_GENL_ATTR_TZ_NAME, tz->type))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_cmd_tz_get_id(struct param *p)
{
struct sk_buff *msg = p->msg;
struct nlattr *start_tz;
int ret;
start_tz = nla_nest_start(msg, THERMAL_GENL_ATTR_TZ);
if (!start_tz)
return -EMSGSIZE;
ret = for_each_thermal_zone(__thermal_genl_cmd_tz_get_id, msg);
if (ret)
goto out_cancel_nest;
nla_nest_end(msg, start_tz);
return 0;
out_cancel_nest:
nla_nest_cancel(msg, start_tz);
return ret;
}
static int thermal_genl_cmd_tz_get_trip(struct param *p)
{
struct sk_buff *msg = p->msg;
struct thermal_zone_device *tz;
struct nlattr *start_trip;
int i, id;
if (!p->attrs[THERMAL_GENL_ATTR_TZ_ID])
return -EINVAL;
id = nla_get_u32(p->attrs[THERMAL_GENL_ATTR_TZ_ID]);
tz = thermal_zone_get_by_id(id);
if (!tz)
return -EINVAL;
start_trip = nla_nest_start(msg, THERMAL_GENL_ATTR_TZ_TRIP);
if (!start_trip)
return -EMSGSIZE;
mutex_lock(&tz->lock);
for (i = 0; i < tz->num_trips; i++) {
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
enum thermal_trip_type type;
int temp, hyst = 0;
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
tz->ops->get_trip_type(tz, i, &type);
tz->ops->get_trip_temp(tz, i, &temp);
if (tz->ops->get_trip_hyst)
tz->ops->get_trip_hyst(tz, i, &hyst);
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
if (nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_TRIP_ID, i) ||
nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_TRIP_TYPE, type) ||
nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_TRIP_TEMP, temp) ||
nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_TRIP_HYST, hyst))
goto out_cancel_nest;
}
mutex_unlock(&tz->lock);
nla_nest_end(msg, start_trip);
return 0;
out_cancel_nest:
mutex_unlock(&tz->lock);
return -EMSGSIZE;
}
static int thermal_genl_cmd_tz_get_temp(struct param *p)
{
struct sk_buff *msg = p->msg;
struct thermal_zone_device *tz;
int temp, ret, id;
if (!p->attrs[THERMAL_GENL_ATTR_TZ_ID])
return -EINVAL;
id = nla_get_u32(p->attrs[THERMAL_GENL_ATTR_TZ_ID]);
tz = thermal_zone_get_by_id(id);
if (!tz)
return -EINVAL;
ret = thermal_zone_get_temp(tz, &temp);
if (ret)
return ret;
if (nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_ID, id) ||
nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_TEMP, temp))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_cmd_tz_get_gov(struct param *p)
{
struct sk_buff *msg = p->msg;
struct thermal_zone_device *tz;
int id, ret = 0;
if (!p->attrs[THERMAL_GENL_ATTR_TZ_ID])
return -EINVAL;
id = nla_get_u32(p->attrs[THERMAL_GENL_ATTR_TZ_ID]);
tz = thermal_zone_get_by_id(id);
if (!tz)
return -EINVAL;
mutex_lock(&tz->lock);
if (nla_put_u32(msg, THERMAL_GENL_ATTR_TZ_ID, id) ||
nla_put_string(msg, THERMAL_GENL_ATTR_TZ_GOV_NAME,
tz->governor->name))
ret = -EMSGSIZE;
mutex_unlock(&tz->lock);
return ret;
}
static int __thermal_genl_cmd_cdev_get(struct thermal_cooling_device *cdev,
void *data)
{
struct sk_buff *msg = data;
if (nla_put_u32(msg, THERMAL_GENL_ATTR_CDEV_ID, cdev->id))
return -EMSGSIZE;
if (nla_put_string(msg, THERMAL_GENL_ATTR_CDEV_NAME, cdev->type))
return -EMSGSIZE;
return 0;
}
static int thermal_genl_cmd_cdev_get(struct param *p)
{
struct sk_buff *msg = p->msg;
struct nlattr *start_cdev;
int ret;
start_cdev = nla_nest_start(msg, THERMAL_GENL_ATTR_CDEV);
if (!start_cdev)
return -EMSGSIZE;
ret = for_each_thermal_cooling_device(__thermal_genl_cmd_cdev_get, msg);
if (ret)
goto out_cancel_nest;
nla_nest_end(msg, start_cdev);
return 0;
out_cancel_nest:
nla_nest_cancel(msg, start_cdev);
return ret;
}
static cb_t cmd_cb[] = {
[THERMAL_GENL_CMD_TZ_GET_ID] = thermal_genl_cmd_tz_get_id,
[THERMAL_GENL_CMD_TZ_GET_TRIP] = thermal_genl_cmd_tz_get_trip,
[THERMAL_GENL_CMD_TZ_GET_TEMP] = thermal_genl_cmd_tz_get_temp,
[THERMAL_GENL_CMD_TZ_GET_GOV] = thermal_genl_cmd_tz_get_gov,
[THERMAL_GENL_CMD_CDEV_GET] = thermal_genl_cmd_cdev_get,
};
static int thermal_genl_cmd_dumpit(struct sk_buff *skb,
struct netlink_callback *cb)
{
struct param p = { .msg = skb };
const struct genl_dumpit_info *info = genl_dumpit_info(cb);
int cmd = info->op.cmd;
int ret;
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
void *hdr;
hdr = genlmsg_put(skb, 0, 0, &thermal_gnl_family, 0, cmd);
if (!hdr)
return -EMSGSIZE;
ret = cmd_cb[cmd](&p);
if (ret)
goto out_cancel_msg;
genlmsg_end(skb, hdr);
return 0;
out_cancel_msg:
genlmsg_cancel(skb, hdr);
return ret;
}
static int thermal_genl_cmd_doit(struct sk_buff *skb,
struct genl_info *info)
{
struct param p = { .attrs = info->attrs };
struct sk_buff *msg;
void *hdr;
int cmd = info->genlhdr->cmd;
int ret = -EMSGSIZE;
msg = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
p.msg = msg;
hdr = genlmsg_put_reply(msg, info, &thermal_gnl_family, 0, cmd);
if (!hdr)
goto out_free_msg;
ret = cmd_cb[cmd](&p);
if (ret)
goto out_cancel_msg;
genlmsg_end(msg, hdr);
return genlmsg_reply(msg, info);
out_cancel_msg:
genlmsg_cancel(msg, hdr);
out_free_msg:
nlmsg_free(msg);
return ret;
}
static const struct genl_small_ops thermal_genl_ops[] = {
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
{
.cmd = THERMAL_GENL_CMD_TZ_GET_ID,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.dumpit = thermal_genl_cmd_dumpit,
},
{
.cmd = THERMAL_GENL_CMD_TZ_GET_TRIP,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.doit = thermal_genl_cmd_doit,
},
{
.cmd = THERMAL_GENL_CMD_TZ_GET_TEMP,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.doit = thermal_genl_cmd_doit,
},
{
.cmd = THERMAL_GENL_CMD_TZ_GET_GOV,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.doit = thermal_genl_cmd_doit,
},
{
.cmd = THERMAL_GENL_CMD_CDEV_GET,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.dumpit = thermal_genl_cmd_dumpit,
},
};
static struct genl_family thermal_gnl_family __ro_after_init = {
.hdrsize = 0,
.name = THERMAL_GENL_FAMILY_NAME,
.version = THERMAL_GENL_VERSION,
.maxattr = THERMAL_GENL_ATTR_MAX,
.policy = thermal_genl_policy,
.small_ops = thermal_genl_ops,
.n_small_ops = ARRAY_SIZE(thermal_genl_ops),
.resv_start_op = THERMAL_GENL_CMD_CDEV_GET + 1,
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
.mcgrps = thermal_genl_mcgrps,
.n_mcgrps = ARRAY_SIZE(thermal_genl_mcgrps),
};
int __init thermal_netlink_init(void)
thermal: core: genetlink support for events/cmd/sampling Initially the thermal framework had a very simple notification mechanism to send generic netlink messages to the userspace. The notification function was never called from anywhere and the corresponding dead code was removed. It was probably a first attempt to introduce the netlink notification. At LPC2018, the presentation "Linux thermal: User kernel interface", proposed to create the notifications to the userspace via a kfifo. The advantage of the kfifo is the performance. It is usually used from a 1:1 communication channel where a driver captures data and sends it as fast as possible to a userspace process. The drawback is that only one process uses the notification channel exclusively, thus no other process is allowed to use the channel to get temperature or notifications. This patch defines a generic netlink API to discover the current thermal setup and adds event notifications as well as temperature sampling. As any genetlink protocol, it can evolve and the versioning allows to keep the backward compatibility. In order to prevent the user from getting flooded with data on a single channel, there are two multicast channels, one for the temperature sampling when the thermal zone is updated and another one for the events, so the user can get the events only without the thermal zone temperature sampling. Also, a list of commands to discover the thermal setup is added and can be extended when needed. Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Zhang Rui <rui.zhang@intel.com> Link: https://lore.kernel.org/r/20200706105538.2159-3-daniel.lezcano@linaro.org
2020-07-06 18:55:37 +08:00
{
return genl_register_family(&thermal_gnl_family);
}