2011-08-25 21:35:03 +08:00
|
|
|
#ifndef _LINUX_PM_QOS_H
|
|
|
|
|
#define _LINUX_PM_QOS_H
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
/* interface for the pm_qos_power infrastructure of the linux kernel.
|
|
|
|
|
*
|
2008-08-06 04:01:35 +08:00
|
|
|
* Mark Gross <mgross@linux.intel.com>
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
*/
|
2010-07-06 04:53:06 +08:00
|
|
|
#include <linux/plist.h>
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
#include <linux/notifier.h>
|
|
|
|
|
#include <linux/miscdevice.h>
|
2011-09-30 04:29:44 +08:00
|
|
|
#include <linux/device.h>
|
2012-03-29 05:31:24 +08:00
|
|
|
#include <linux/workqueue.h>
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
|
2012-01-30 03:39:25 +08:00
|
|
|
enum {
|
|
|
|
|
PM_QOS_RESERVED = 0,
|
|
|
|
|
PM_QOS_CPU_DMA_LATENCY,
|
|
|
|
|
PM_QOS_NETWORK_LATENCY,
|
|
|
|
|
PM_QOS_NETWORK_THROUGHPUT,
|
2014-09-03 23:49:32 +08:00
|
|
|
PM_QOS_MEMORY_BANDWIDTH,
|
2012-01-30 03:39:25 +08:00
|
|
|
|
|
|
|
|
/* insert new class ID */
|
|
|
|
|
PM_QOS_NUM_CLASSES,
|
|
|
|
|
};
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
|
2012-10-23 07:09:12 +08:00
|
|
|
enum pm_qos_flags_status {
|
|
|
|
|
PM_QOS_FLAGS_UNDEFINED = -1,
|
|
|
|
|
PM_QOS_FLAGS_NONE,
|
|
|
|
|
PM_QOS_FLAGS_SOME,
|
|
|
|
|
PM_QOS_FLAGS_ALL,
|
|
|
|
|
};
|
|
|
|
|
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
#define PM_QOS_DEFAULT_VALUE -1
|
|
|
|
|
|
2011-02-12 04:49:04 +08:00
|
|
|
#define PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC)
|
|
|
|
|
#define PM_QOS_NETWORK_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC)
|
|
|
|
|
#define PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE 0
|
2014-09-03 23:49:32 +08:00
|
|
|
#define PM_QOS_MEMORY_BANDWIDTH_DEFAULT_VALUE 0
|
2014-02-11 07:35:23 +08:00
|
|
|
#define PM_QOS_RESUME_LATENCY_DEFAULT_VALUE 0
|
PM / QoS: Introcuce latency tolerance device PM QoS type
Add a new latency tolerance device PM QoS type to be use for
specifying active state (RPM_ACTIVE) memory access (DMA) latency
tolerance requirements for devices. It may be used to prevent
hardware from choosing overly aggressive energy-saving operation
modes (causing too much latency to appear) for the whole platform.
This feature reqiures hardware support, so it only will be
available for devices having a new .set_latency_tolerance()
callback in struct dev_pm_info populated, in which case the
routine pointed to by it should implement whatever is necessary
to transfer the effective requirement value to the hardware.
Whenever the effective latency tolerance changes for the device,
its .set_latency_tolerance() callback will be executed and the
effective value will be passed to it. If that value is negative,
which means that the list of latency tolerance requirements for
the device is empty, the callback is expected to switch the
underlying hardware latency tolerance control mechanism to an
autonomous mode if available. If that value is PM_QOS_LATENCY_ANY,
in turn, and the hardware supports a special "no requirement"
setting, the callback is expected to use it. That allows software
to prevent the hardware from automatically updating the device's
latency tolerance in response to its power state changes (e.g. during
transitions from D3cold to D0), which generally may be done in the
autonomous latency tolerance control mode.
If .set_latency_tolerance() is present for the device, a new
pm_qos_latency_tolerance_us attribute will be present in the
devivce's power directory in sysfs. Then, user space can use
that attribute to specify its latency tolerance requirement for
the device, if any. Writing "any" to it means "no requirement, but
do not let the hardware control latency tolerance" and writing
"auto" to it allows the hardware to be switched to the autonomous
mode if there are no other requirements from the kernel side in the
device's list.
This changeset includes a fix from Mika Westerberg.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-11 07:35:38 +08:00
|
|
|
#define PM_QOS_LATENCY_TOLERANCE_DEFAULT_VALUE 0
|
|
|
|
|
#define PM_QOS_LATENCY_TOLERANCE_NO_CONSTRAINT (-1)
|
|
|
|
|
#define PM_QOS_LATENCY_ANY ((s32)(~(__u32)0 >> 1))
|
2011-02-12 04:49:04 +08:00
|
|
|
|
2012-10-24 08:08:18 +08:00
|
|
|
#define PM_QOS_FLAG_NO_POWER_OFF (1 << 0)
|
|
|
|
|
#define PM_QOS_FLAG_REMOTE_WAKEUP (1 << 1)
|
|
|
|
|
|
2011-08-25 21:35:12 +08:00
|
|
|
struct pm_qos_request {
|
|
|
|
|
struct plist_node node;
|
2010-07-06 04:53:06 +08:00
|
|
|
int pm_qos_class;
|
2012-03-29 05:31:24 +08:00
|
|
|
struct delayed_work work; /* for pm_qos_update_request_timeout */
|
2010-07-06 04:53:06 +08:00
|
|
|
};
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
|
2012-10-23 07:07:46 +08:00
|
|
|
struct pm_qos_flags_request {
|
|
|
|
|
struct list_head node;
|
|
|
|
|
s32 flags; /* Do not change to 64 bit */
|
|
|
|
|
};
|
|
|
|
|
|
2012-10-23 07:09:12 +08:00
|
|
|
enum dev_pm_qos_req_type {
|
2014-02-11 07:35:23 +08:00
|
|
|
DEV_PM_QOS_RESUME_LATENCY = 1,
|
PM / QoS: Introcuce latency tolerance device PM QoS type
Add a new latency tolerance device PM QoS type to be use for
specifying active state (RPM_ACTIVE) memory access (DMA) latency
tolerance requirements for devices. It may be used to prevent
hardware from choosing overly aggressive energy-saving operation
modes (causing too much latency to appear) for the whole platform.
This feature reqiures hardware support, so it only will be
available for devices having a new .set_latency_tolerance()
callback in struct dev_pm_info populated, in which case the
routine pointed to by it should implement whatever is necessary
to transfer the effective requirement value to the hardware.
Whenever the effective latency tolerance changes for the device,
its .set_latency_tolerance() callback will be executed and the
effective value will be passed to it. If that value is negative,
which means that the list of latency tolerance requirements for
the device is empty, the callback is expected to switch the
underlying hardware latency tolerance control mechanism to an
autonomous mode if available. If that value is PM_QOS_LATENCY_ANY,
in turn, and the hardware supports a special "no requirement"
setting, the callback is expected to use it. That allows software
to prevent the hardware from automatically updating the device's
latency tolerance in response to its power state changes (e.g. during
transitions from D3cold to D0), which generally may be done in the
autonomous latency tolerance control mode.
If .set_latency_tolerance() is present for the device, a new
pm_qos_latency_tolerance_us attribute will be present in the
devivce's power directory in sysfs. Then, user space can use
that attribute to specify its latency tolerance requirement for
the device, if any. Writing "any" to it means "no requirement, but
do not let the hardware control latency tolerance" and writing
"auto" to it allows the hardware to be switched to the autonomous
mode if there are no other requirements from the kernel side in the
device's list.
This changeset includes a fix from Mika Westerberg.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-11 07:35:38 +08:00
|
|
|
DEV_PM_QOS_LATENCY_TOLERANCE,
|
2012-10-23 07:09:12 +08:00
|
|
|
DEV_PM_QOS_FLAGS,
|
|
|
|
|
};
|
|
|
|
|
|
2011-08-25 21:35:41 +08:00
|
|
|
struct dev_pm_qos_request {
|
2012-10-23 07:09:12 +08:00
|
|
|
enum dev_pm_qos_req_type type;
|
2012-10-23 07:09:00 +08:00
|
|
|
union {
|
|
|
|
|
struct plist_node pnode;
|
2012-10-23 07:09:12 +08:00
|
|
|
struct pm_qos_flags_request flr;
|
2012-10-23 07:09:00 +08:00
|
|
|
} data;
|
2011-08-25 21:35:41 +08:00
|
|
|
struct device *dev;
|
|
|
|
|
};
|
|
|
|
|
|
2011-08-25 21:35:27 +08:00
|
|
|
enum pm_qos_type {
|
|
|
|
|
PM_QOS_UNITIALIZED,
|
|
|
|
|
PM_QOS_MAX, /* return the largest value */
|
2014-09-03 23:49:32 +08:00
|
|
|
PM_QOS_MIN, /* return the smallest value */
|
|
|
|
|
PM_QOS_SUM /* return the sum */
|
2011-08-25 21:35:27 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/*
|
2012-10-23 07:07:46 +08:00
|
|
|
* Note: The lockless read path depends on the CPU accessing target_value
|
|
|
|
|
* or effective_flags atomically. Atomic access is only guaranteed on all CPU
|
2011-08-25 21:35:27 +08:00
|
|
|
* types linux supports for 32 bit quantites
|
|
|
|
|
*/
|
|
|
|
|
struct pm_qos_constraints {
|
|
|
|
|
struct plist_head list;
|
|
|
|
|
s32 target_value; /* Do not change to 64 bit */
|
|
|
|
|
s32 default_value;
|
2014-02-11 07:35:29 +08:00
|
|
|
s32 no_constraint_value;
|
2011-08-25 21:35:27 +08:00
|
|
|
enum pm_qos_type type;
|
|
|
|
|
struct blocking_notifier_head *notifiers;
|
|
|
|
|
};
|
|
|
|
|
|
2012-10-23 07:07:46 +08:00
|
|
|
struct pm_qos_flags {
|
|
|
|
|
struct list_head list;
|
|
|
|
|
s32 effective_flags; /* Do not change to 64 bit */
|
|
|
|
|
};
|
|
|
|
|
|
2012-10-23 07:07:27 +08:00
|
|
|
struct dev_pm_qos {
|
2014-02-11 07:35:23 +08:00
|
|
|
struct pm_qos_constraints resume_latency;
|
PM / QoS: Introcuce latency tolerance device PM QoS type
Add a new latency tolerance device PM QoS type to be use for
specifying active state (RPM_ACTIVE) memory access (DMA) latency
tolerance requirements for devices. It may be used to prevent
hardware from choosing overly aggressive energy-saving operation
modes (causing too much latency to appear) for the whole platform.
This feature reqiures hardware support, so it only will be
available for devices having a new .set_latency_tolerance()
callback in struct dev_pm_info populated, in which case the
routine pointed to by it should implement whatever is necessary
to transfer the effective requirement value to the hardware.
Whenever the effective latency tolerance changes for the device,
its .set_latency_tolerance() callback will be executed and the
effective value will be passed to it. If that value is negative,
which means that the list of latency tolerance requirements for
the device is empty, the callback is expected to switch the
underlying hardware latency tolerance control mechanism to an
autonomous mode if available. If that value is PM_QOS_LATENCY_ANY,
in turn, and the hardware supports a special "no requirement"
setting, the callback is expected to use it. That allows software
to prevent the hardware from automatically updating the device's
latency tolerance in response to its power state changes (e.g. during
transitions from D3cold to D0), which generally may be done in the
autonomous latency tolerance control mode.
If .set_latency_tolerance() is present for the device, a new
pm_qos_latency_tolerance_us attribute will be present in the
devivce's power directory in sysfs. Then, user space can use
that attribute to specify its latency tolerance requirement for
the device, if any. Writing "any" to it means "no requirement, but
do not let the hardware control latency tolerance" and writing
"auto" to it allows the hardware to be switched to the autonomous
mode if there are no other requirements from the kernel side in the
device's list.
This changeset includes a fix from Mika Westerberg.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-11 07:35:38 +08:00
|
|
|
struct pm_qos_constraints latency_tolerance;
|
2012-10-23 07:09:12 +08:00
|
|
|
struct pm_qos_flags flags;
|
2014-02-11 07:35:23 +08:00
|
|
|
struct dev_pm_qos_request *resume_latency_req;
|
PM / QoS: Introcuce latency tolerance device PM QoS type
Add a new latency tolerance device PM QoS type to be use for
specifying active state (RPM_ACTIVE) memory access (DMA) latency
tolerance requirements for devices. It may be used to prevent
hardware from choosing overly aggressive energy-saving operation
modes (causing too much latency to appear) for the whole platform.
This feature reqiures hardware support, so it only will be
available for devices having a new .set_latency_tolerance()
callback in struct dev_pm_info populated, in which case the
routine pointed to by it should implement whatever is necessary
to transfer the effective requirement value to the hardware.
Whenever the effective latency tolerance changes for the device,
its .set_latency_tolerance() callback will be executed and the
effective value will be passed to it. If that value is negative,
which means that the list of latency tolerance requirements for
the device is empty, the callback is expected to switch the
underlying hardware latency tolerance control mechanism to an
autonomous mode if available. If that value is PM_QOS_LATENCY_ANY,
in turn, and the hardware supports a special "no requirement"
setting, the callback is expected to use it. That allows software
to prevent the hardware from automatically updating the device's
latency tolerance in response to its power state changes (e.g. during
transitions from D3cold to D0), which generally may be done in the
autonomous latency tolerance control mode.
If .set_latency_tolerance() is present for the device, a new
pm_qos_latency_tolerance_us attribute will be present in the
devivce's power directory in sysfs. Then, user space can use
that attribute to specify its latency tolerance requirement for
the device, if any. Writing "any" to it means "no requirement, but
do not let the hardware control latency tolerance" and writing
"auto" to it allows the hardware to be switched to the autonomous
mode if there are no other requirements from the kernel side in the
device's list.
This changeset includes a fix from Mika Westerberg.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-11 07:35:38 +08:00
|
|
|
struct dev_pm_qos_request *latency_tolerance_req;
|
2012-10-24 08:08:18 +08:00
|
|
|
struct dev_pm_qos_request *flags_req;
|
2012-10-23 07:07:27 +08:00
|
|
|
};
|
|
|
|
|
|
2011-08-25 21:35:34 +08:00
|
|
|
/* Action requested to pm_qos_update_target */
|
|
|
|
|
enum pm_qos_req_action {
|
|
|
|
|
PM_QOS_ADD_REQ, /* Add a new request */
|
|
|
|
|
PM_QOS_UPDATE_REQ, /* Update an existing request */
|
|
|
|
|
PM_QOS_REMOVE_REQ /* Remove an existing request */
|
|
|
|
|
};
|
|
|
|
|
|
2011-08-25 21:35:41 +08:00
|
|
|
static inline int dev_pm_qos_request_active(struct dev_pm_qos_request *req)
|
|
|
|
|
{
|
2012-07-18 04:38:18 +08:00
|
|
|
return req->dev != NULL;
|
2011-08-25 21:35:41 +08:00
|
|
|
}
|
|
|
|
|
|
2011-08-25 21:35:34 +08:00
|
|
|
int pm_qos_update_target(struct pm_qos_constraints *c, struct plist_node *node,
|
|
|
|
|
enum pm_qos_req_action action, int value);
|
2012-10-23 07:07:46 +08:00
|
|
|
bool pm_qos_update_flags(struct pm_qos_flags *pqf,
|
|
|
|
|
struct pm_qos_flags_request *req,
|
|
|
|
|
enum pm_qos_req_action action, s32 val);
|
2011-08-25 21:35:12 +08:00
|
|
|
void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class,
|
|
|
|
|
s32 value);
|
|
|
|
|
void pm_qos_update_request(struct pm_qos_request *req,
|
2011-08-25 21:35:03 +08:00
|
|
|
s32 new_value);
|
2012-03-29 05:31:24 +08:00
|
|
|
void pm_qos_update_request_timeout(struct pm_qos_request *req,
|
|
|
|
|
s32 new_value, unsigned long timeout_us);
|
2011-08-25 21:35:12 +08:00
|
|
|
void pm_qos_remove_request(struct pm_qos_request *req);
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
|
2010-05-06 07:59:26 +08:00
|
|
|
int pm_qos_request(int pm_qos_class);
|
|
|
|
|
int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier);
|
|
|
|
|
int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier);
|
2011-08-25 21:35:12 +08:00
|
|
|
int pm_qos_request_active(struct pm_qos_request *req);
|
2011-08-25 21:35:47 +08:00
|
|
|
s32 pm_qos_read_value(struct pm_qos_constraints *c);
|
2011-08-25 21:35:41 +08:00
|
|
|
|
2012-02-13 23:23:42 +08:00
|
|
|
#ifdef CONFIG_PM
|
2012-10-23 07:09:12 +08:00
|
|
|
enum pm_qos_flags_status __dev_pm_qos_flags(struct device *dev, s32 mask);
|
|
|
|
|
enum pm_qos_flags_status dev_pm_qos_flags(struct device *dev, s32 mask);
|
2011-12-01 07:01:31 +08:00
|
|
|
s32 __dev_pm_qos_read_value(struct device *dev);
|
2011-09-30 04:29:44 +08:00
|
|
|
s32 dev_pm_qos_read_value(struct device *dev);
|
2011-08-25 21:35:41 +08:00
|
|
|
int dev_pm_qos_add_request(struct device *dev, struct dev_pm_qos_request *req,
|
2012-10-23 07:09:12 +08:00
|
|
|
enum dev_pm_qos_req_type type, s32 value);
|
2011-08-25 21:35:41 +08:00
|
|
|
int dev_pm_qos_update_request(struct dev_pm_qos_request *req, s32 new_value);
|
|
|
|
|
int dev_pm_qos_remove_request(struct dev_pm_qos_request *req);
|
|
|
|
|
int dev_pm_qos_add_notifier(struct device *dev,
|
|
|
|
|
struct notifier_block *notifier);
|
|
|
|
|
int dev_pm_qos_remove_notifier(struct device *dev,
|
|
|
|
|
struct notifier_block *notifier);
|
2011-08-25 21:35:47 +08:00
|
|
|
int dev_pm_qos_add_global_notifier(struct notifier_block *notifier);
|
|
|
|
|
int dev_pm_qos_remove_global_notifier(struct notifier_block *notifier);
|
2011-08-25 21:35:41 +08:00
|
|
|
void dev_pm_qos_constraints_init(struct device *dev);
|
|
|
|
|
void dev_pm_qos_constraints_destroy(struct device *dev);
|
PM / QoS: Introduce dev_pm_qos_add_ancestor_request()
Some devices, like the I2C controller on SH7372, are not
necessary for providing power to their children or forwarding
wakeup signals (and generally interrupts) from them. They are
only needed by their children when there's some data to transfer,
so they may be suspended for the majority of time and resumed
on demand, when the children have data to send or receive. For this
purpose, however, their power.ignore_children flags have to be set,
or the PM core wouldn't allow them to be suspended while their
children were active.
Unfortunately, in some situations it may take too much time to
resume such devices so that they can assist their children in
transferring data. For example, if such a device belongs to a PM
domain which goes to the "power off" state when that device is
suspended, it may take too much time to restore power to the
domain in response to the request from one of the device's
children. In that case, if the parent's resume time is critical,
the domain should stay in the "power on" state, although it still may
be desirable to power manage the parent itself (e.g. by manipulating
its clock).
In general, device PM QoS may be used to address this problem.
Namely, if the device's children added PM QoS latency constraints
for it, they would be able to prevent it from being put into an
overly deep low-power state. However, in some cases the devices
needing to be serviced are not the immediate children of a
"children-ignoring" device, but its grandchildren or even less
direct descendants. In those cases, the entity wanting to add a
PM QoS request for a given device's ancestor that ignores its
children will have to find it in the first place, so introduce a new
helper function that may be used to achieve that. This function,
dev_pm_qos_add_ancestor_request(), will search for the first
ancestor of the given device whose power.ignore_children flag is
set and will add a device PM QoS latency request for that ancestor
on behalf of the caller. The request added this way may be removed
with the help of dev_pm_qos_remove_request() in the future, like
any other device PM QoS latency request.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
2011-12-23 08:23:52 +08:00
|
|
|
int dev_pm_qos_add_ancestor_request(struct device *dev,
|
2014-02-11 07:36:00 +08:00
|
|
|
struct dev_pm_qos_request *req,
|
|
|
|
|
enum dev_pm_qos_req_type type, s32 value);
|
2011-08-25 21:35:03 +08:00
|
|
|
#else
|
2012-10-23 07:09:12 +08:00
|
|
|
static inline enum pm_qos_flags_status __dev_pm_qos_flags(struct device *dev,
|
|
|
|
|
s32 mask)
|
|
|
|
|
{ return PM_QOS_FLAGS_UNDEFINED; }
|
|
|
|
|
static inline enum pm_qos_flags_status dev_pm_qos_flags(struct device *dev,
|
|
|
|
|
s32 mask)
|
|
|
|
|
{ return PM_QOS_FLAGS_UNDEFINED; }
|
2011-12-01 07:01:31 +08:00
|
|
|
static inline s32 __dev_pm_qos_read_value(struct device *dev)
|
|
|
|
|
{ return 0; }
|
2011-09-30 04:29:44 +08:00
|
|
|
static inline s32 dev_pm_qos_read_value(struct device *dev)
|
|
|
|
|
{ return 0; }
|
2011-08-25 21:35:41 +08:00
|
|
|
static inline int dev_pm_qos_add_request(struct device *dev,
|
|
|
|
|
struct dev_pm_qos_request *req,
|
2012-10-23 07:09:12 +08:00
|
|
|
enum dev_pm_qos_req_type type,
|
2011-08-25 21:35:41 +08:00
|
|
|
s32 value)
|
|
|
|
|
{ return 0; }
|
|
|
|
|
static inline int dev_pm_qos_update_request(struct dev_pm_qos_request *req,
|
|
|
|
|
s32 new_value)
|
|
|
|
|
{ return 0; }
|
|
|
|
|
static inline int dev_pm_qos_remove_request(struct dev_pm_qos_request *req)
|
|
|
|
|
{ return 0; }
|
|
|
|
|
static inline int dev_pm_qos_add_notifier(struct device *dev,
|
|
|
|
|
struct notifier_block *notifier)
|
|
|
|
|
{ return 0; }
|
|
|
|
|
static inline int dev_pm_qos_remove_notifier(struct device *dev,
|
|
|
|
|
struct notifier_block *notifier)
|
|
|
|
|
{ return 0; }
|
2011-08-25 21:35:47 +08:00
|
|
|
static inline int dev_pm_qos_add_global_notifier(
|
|
|
|
|
struct notifier_block *notifier)
|
|
|
|
|
{ return 0; }
|
|
|
|
|
static inline int dev_pm_qos_remove_global_notifier(
|
|
|
|
|
struct notifier_block *notifier)
|
|
|
|
|
{ return 0; }
|
2011-08-25 21:35:41 +08:00
|
|
|
static inline void dev_pm_qos_constraints_init(struct device *dev)
|
2011-09-30 04:29:44 +08:00
|
|
|
{
|
|
|
|
|
dev->power.power_state = PMSG_ON;
|
|
|
|
|
}
|
2011-08-25 21:35:41 +08:00
|
|
|
static inline void dev_pm_qos_constraints_destroy(struct device *dev)
|
2011-09-30 04:29:44 +08:00
|
|
|
{
|
|
|
|
|
dev->power.power_state = PMSG_INVALID;
|
|
|
|
|
}
|
PM / QoS: Introduce dev_pm_qos_add_ancestor_request()
Some devices, like the I2C controller on SH7372, are not
necessary for providing power to their children or forwarding
wakeup signals (and generally interrupts) from them. They are
only needed by their children when there's some data to transfer,
so they may be suspended for the majority of time and resumed
on demand, when the children have data to send or receive. For this
purpose, however, their power.ignore_children flags have to be set,
or the PM core wouldn't allow them to be suspended while their
children were active.
Unfortunately, in some situations it may take too much time to
resume such devices so that they can assist their children in
transferring data. For example, if such a device belongs to a PM
domain which goes to the "power off" state when that device is
suspended, it may take too much time to restore power to the
domain in response to the request from one of the device's
children. In that case, if the parent's resume time is critical,
the domain should stay in the "power on" state, although it still may
be desirable to power manage the parent itself (e.g. by manipulating
its clock).
In general, device PM QoS may be used to address this problem.
Namely, if the device's children added PM QoS latency constraints
for it, they would be able to prevent it from being put into an
overly deep low-power state. However, in some cases the devices
needing to be serviced are not the immediate children of a
"children-ignoring" device, but its grandchildren or even less
direct descendants. In those cases, the entity wanting to add a
PM QoS request for a given device's ancestor that ignores its
children will have to find it in the first place, so introduce a new
helper function that may be used to achieve that. This function,
dev_pm_qos_add_ancestor_request(), will search for the first
ancestor of the given device whose power.ignore_children flag is
set and will add a device PM QoS latency request for that ancestor
on behalf of the caller. The request added this way may be removed
with the help of dev_pm_qos_remove_request() in the future, like
any other device PM QoS latency request.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
2011-12-23 08:23:52 +08:00
|
|
|
static inline int dev_pm_qos_add_ancestor_request(struct device *dev,
|
2014-02-11 07:36:00 +08:00
|
|
|
struct dev_pm_qos_request *req,
|
|
|
|
|
enum dev_pm_qos_req_type type,
|
|
|
|
|
s32 value)
|
PM / QoS: Introduce dev_pm_qos_add_ancestor_request()
Some devices, like the I2C controller on SH7372, are not
necessary for providing power to their children or forwarding
wakeup signals (and generally interrupts) from them. They are
only needed by their children when there's some data to transfer,
so they may be suspended for the majority of time and resumed
on demand, when the children have data to send or receive. For this
purpose, however, their power.ignore_children flags have to be set,
or the PM core wouldn't allow them to be suspended while their
children were active.
Unfortunately, in some situations it may take too much time to
resume such devices so that they can assist their children in
transferring data. For example, if such a device belongs to a PM
domain which goes to the "power off" state when that device is
suspended, it may take too much time to restore power to the
domain in response to the request from one of the device's
children. In that case, if the parent's resume time is critical,
the domain should stay in the "power on" state, although it still may
be desirable to power manage the parent itself (e.g. by manipulating
its clock).
In general, device PM QoS may be used to address this problem.
Namely, if the device's children added PM QoS latency constraints
for it, they would be able to prevent it from being put into an
overly deep low-power state. However, in some cases the devices
needing to be serviced are not the immediate children of a
"children-ignoring" device, but its grandchildren or even less
direct descendants. In those cases, the entity wanting to add a
PM QoS request for a given device's ancestor that ignores its
children will have to find it in the first place, so introduce a new
helper function that may be used to achieve that. This function,
dev_pm_qos_add_ancestor_request(), will search for the first
ancestor of the given device whose power.ignore_children flag is
set and will add a device PM QoS latency request for that ancestor
on behalf of the caller. The request added this way may be removed
with the help of dev_pm_qos_remove_request() in the future, like
any other device PM QoS latency request.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
2011-12-23 08:23:52 +08:00
|
|
|
{ return 0; }
|
2011-08-25 21:35:03 +08:00
|
|
|
#endif
|
pm qos infrastructure and interface
The following patch is a generalization of the latency.c implementation done
by Arjan last year. It provides infrastructure for more than one parameter,
and exposes a user mode interface for processes to register pm_qos
expectations of processes.
This interface provides a kernel and user mode interface for registering
performance expectations by drivers, subsystems and user space applications on
one of the parameters.
Currently we have {cpu_dma_latency, network_latency, network_throughput} as
the initial set of pm_qos parameters.
The infrastructure exposes multiple misc device nodes one per implemented
parameter. The set of parameters implement is defined by pm_qos_power_init()
and pm_qos_params.h. This is done because having the available parameters
being runtime configurable or changeable from a driver was seen as too easy to
abuse.
For each parameter a list of performance requirements is maintained along with
an aggregated target value. The aggregated target value is updated with
changes to the requirement list or elements of the list. Typically the
aggregated target value is simply the max or min of the requirement values
held in the parameter list elements.
>From kernel mode the use of this interface is simple:
pm_qos_add_requirement(param_id, name, target_value):
Will insert a named element in the list for that identified PM_QOS
parameter with the target value. Upon change to this list the new target is
recomputed and any registered notifiers are called only if the target value
is now different.
pm_qos_update_requirement(param_id, name, new_target_value):
Will search the list identified by the param_id for the named list element
and then update its target value, calling the notification tree if the
aggregated target is changed. with that name is already registered.
pm_qos_remove_requirement(param_id, name):
Will search the identified list for the named element and remove it, after
removal it will update the aggregate target and call the notification tree
if the target was changed as a result of removing the named requirement.
>From user mode:
Only processes can register a pm_qos requirement. To provide for
automatic cleanup for process the interface requires the process to register
its parameter requirements in the following way:
To register the default pm_qos target for the specific parameter, the
process must open one of /dev/[cpu_dma_latency, network_latency,
network_throughput]
As long as the device node is held open that process has a registered
requirement on the parameter. The name of the requirement is
"process_<PID>" derived from the current->pid from within the open system
call.
To change the requested target value the process needs to write a s32
value to the open device node. This translates to a
pm_qos_update_requirement call.
To remove the user mode request for a target value simply close the device
node.
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: fix build again]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: mark gross <mgross@linux.intel.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: Len Brown <lenb@kernel.org>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 14:30:08 +08:00
|
|
|
|
PM / QoS: Make it possible to expose PM QoS latency constraints
A runtime suspend of a device (e.g. an MMC controller) belonging to
a power domain or, in a more complicated scenario, a runtime suspend
of another device in the same power domain, may cause power to be
removed from the entire domain. In that case, the amount of time
necessary to runtime-resume the given device (e.g. the MMC
controller) is often substantially greater than the time needed to
run its driver's runtime resume callback. That may hurt performance
in some situations, because user data may need to wait for the
device to become operational, so we should make it possible to
prevent that from happening.
For this reason, introduce a new sysfs attribute for devices,
power/pm_qos_resume_latency_us, allowing user space to specify the
upper bound of the time necessary to bring the (runtime-suspended)
device up after the resume of it has been requested. However, make
that attribute appear only for the devices whose drivers declare
support for it by calling the (new) dev_pm_qos_expose_latency_limit()
helper function with the appropriate initial value of the attribute.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Reviewed-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
2012-03-13 08:01:39 +08:00
|
|
|
#ifdef CONFIG_PM_RUNTIME
|
|
|
|
|
int dev_pm_qos_expose_latency_limit(struct device *dev, s32 value);
|
|
|
|
|
void dev_pm_qos_hide_latency_limit(struct device *dev);
|
2012-10-24 08:08:18 +08:00
|
|
|
int dev_pm_qos_expose_flags(struct device *dev, s32 value);
|
|
|
|
|
void dev_pm_qos_hide_flags(struct device *dev);
|
|
|
|
|
int dev_pm_qos_update_flags(struct device *dev, s32 mask, bool set);
|
PM / QoS: Introcuce latency tolerance device PM QoS type
Add a new latency tolerance device PM QoS type to be use for
specifying active state (RPM_ACTIVE) memory access (DMA) latency
tolerance requirements for devices. It may be used to prevent
hardware from choosing overly aggressive energy-saving operation
modes (causing too much latency to appear) for the whole platform.
This feature reqiures hardware support, so it only will be
available for devices having a new .set_latency_tolerance()
callback in struct dev_pm_info populated, in which case the
routine pointed to by it should implement whatever is necessary
to transfer the effective requirement value to the hardware.
Whenever the effective latency tolerance changes for the device,
its .set_latency_tolerance() callback will be executed and the
effective value will be passed to it. If that value is negative,
which means that the list of latency tolerance requirements for
the device is empty, the callback is expected to switch the
underlying hardware latency tolerance control mechanism to an
autonomous mode if available. If that value is PM_QOS_LATENCY_ANY,
in turn, and the hardware supports a special "no requirement"
setting, the callback is expected to use it. That allows software
to prevent the hardware from automatically updating the device's
latency tolerance in response to its power state changes (e.g. during
transitions from D3cold to D0), which generally may be done in the
autonomous latency tolerance control mode.
If .set_latency_tolerance() is present for the device, a new
pm_qos_latency_tolerance_us attribute will be present in the
devivce's power directory in sysfs. Then, user space can use
that attribute to specify its latency tolerance requirement for
the device, if any. Writing "any" to it means "no requirement, but
do not let the hardware control latency tolerance" and writing
"auto" to it allows the hardware to be switched to the autonomous
mode if there are no other requirements from the kernel side in the
device's list.
This changeset includes a fix from Mika Westerberg.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-11 07:35:38 +08:00
|
|
|
s32 dev_pm_qos_get_user_latency_tolerance(struct device *dev);
|
|
|
|
|
int dev_pm_qos_update_user_latency_tolerance(struct device *dev, s32 val);
|
2012-10-24 08:08:18 +08:00
|
|
|
|
2014-02-11 07:35:23 +08:00
|
|
|
static inline s32 dev_pm_qos_requested_resume_latency(struct device *dev)
|
2012-10-24 08:08:18 +08:00
|
|
|
{
|
2014-02-11 07:35:23 +08:00
|
|
|
return dev->power.qos->resume_latency_req->data.pnode.prio;
|
2012-10-24 08:08:18 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static inline s32 dev_pm_qos_requested_flags(struct device *dev)
|
|
|
|
|
{
|
|
|
|
|
return dev->power.qos->flags_req->data.flr.flags;
|
|
|
|
|
}
|
PM / QoS: Make it possible to expose PM QoS latency constraints
A runtime suspend of a device (e.g. an MMC controller) belonging to
a power domain or, in a more complicated scenario, a runtime suspend
of another device in the same power domain, may cause power to be
removed from the entire domain. In that case, the amount of time
necessary to runtime-resume the given device (e.g. the MMC
controller) is often substantially greater than the time needed to
run its driver's runtime resume callback. That may hurt performance
in some situations, because user data may need to wait for the
device to become operational, so we should make it possible to
prevent that from happening.
For this reason, introduce a new sysfs attribute for devices,
power/pm_qos_resume_latency_us, allowing user space to specify the
upper bound of the time necessary to bring the (runtime-suspended)
device up after the resume of it has been requested. However, make
that attribute appear only for the devices whose drivers declare
support for it by calling the (new) dev_pm_qos_expose_latency_limit()
helper function with the appropriate initial value of the attribute.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Reviewed-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
2012-03-13 08:01:39 +08:00
|
|
|
#else
|
|
|
|
|
static inline int dev_pm_qos_expose_latency_limit(struct device *dev, s32 value)
|
|
|
|
|
{ return 0; }
|
|
|
|
|
static inline void dev_pm_qos_hide_latency_limit(struct device *dev) {}
|
2012-10-24 08:08:18 +08:00
|
|
|
static inline int dev_pm_qos_expose_flags(struct device *dev, s32 value)
|
|
|
|
|
{ return 0; }
|
|
|
|
|
static inline void dev_pm_qos_hide_flags(struct device *dev) {}
|
|
|
|
|
static inline int dev_pm_qos_update_flags(struct device *dev, s32 m, bool set)
|
|
|
|
|
{ return 0; }
|
PM / QoS: Introcuce latency tolerance device PM QoS type
Add a new latency tolerance device PM QoS type to be use for
specifying active state (RPM_ACTIVE) memory access (DMA) latency
tolerance requirements for devices. It may be used to prevent
hardware from choosing overly aggressive energy-saving operation
modes (causing too much latency to appear) for the whole platform.
This feature reqiures hardware support, so it only will be
available for devices having a new .set_latency_tolerance()
callback in struct dev_pm_info populated, in which case the
routine pointed to by it should implement whatever is necessary
to transfer the effective requirement value to the hardware.
Whenever the effective latency tolerance changes for the device,
its .set_latency_tolerance() callback will be executed and the
effective value will be passed to it. If that value is negative,
which means that the list of latency tolerance requirements for
the device is empty, the callback is expected to switch the
underlying hardware latency tolerance control mechanism to an
autonomous mode if available. If that value is PM_QOS_LATENCY_ANY,
in turn, and the hardware supports a special "no requirement"
setting, the callback is expected to use it. That allows software
to prevent the hardware from automatically updating the device's
latency tolerance in response to its power state changes (e.g. during
transitions from D3cold to D0), which generally may be done in the
autonomous latency tolerance control mode.
If .set_latency_tolerance() is present for the device, a new
pm_qos_latency_tolerance_us attribute will be present in the
devivce's power directory in sysfs. Then, user space can use
that attribute to specify its latency tolerance requirement for
the device, if any. Writing "any" to it means "no requirement, but
do not let the hardware control latency tolerance" and writing
"auto" to it allows the hardware to be switched to the autonomous
mode if there are no other requirements from the kernel side in the
device's list.
This changeset includes a fix from Mika Westerberg.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-11 07:35:38 +08:00
|
|
|
static inline s32 dev_pm_qos_get_user_latency_tolerance(struct device *dev)
|
|
|
|
|
{ return PM_QOS_LATENCY_TOLERANCE_NO_CONSTRAINT; }
|
|
|
|
|
static inline int dev_pm_qos_update_user_latency_tolerance(struct device *dev, s32 val)
|
|
|
|
|
{ return 0; }
|
2012-10-24 08:08:18 +08:00
|
|
|
|
2014-02-11 07:35:23 +08:00
|
|
|
static inline s32 dev_pm_qos_requested_resume_latency(struct device *dev) { return 0; }
|
2012-10-24 08:08:18 +08:00
|
|
|
static inline s32 dev_pm_qos_requested_flags(struct device *dev) { return 0; }
|
PM / QoS: Make it possible to expose PM QoS latency constraints
A runtime suspend of a device (e.g. an MMC controller) belonging to
a power domain or, in a more complicated scenario, a runtime suspend
of another device in the same power domain, may cause power to be
removed from the entire domain. In that case, the amount of time
necessary to runtime-resume the given device (e.g. the MMC
controller) is often substantially greater than the time needed to
run its driver's runtime resume callback. That may hurt performance
in some situations, because user data may need to wait for the
device to become operational, so we should make it possible to
prevent that from happening.
For this reason, introduce a new sysfs attribute for devices,
power/pm_qos_resume_latency_us, allowing user space to specify the
upper bound of the time necessary to bring the (runtime-suspended)
device up after the resume of it has been requested. However, make
that attribute appear only for the devices whose drivers declare
support for it by calling the (new) dev_pm_qos_expose_latency_limit()
helper function with the appropriate initial value of the attribute.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Reviewed-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
2012-03-13 08:01:39 +08:00
|
|
|
#endif
|
|
|
|
|
|
2010-07-06 04:53:06 +08:00
|
|
|
#endif
|