aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--Documentation/ABI/testing/sysfs-bus-usb18
-rw-r--r--Documentation/usb/power-management.txt41
2 files changed, 25 insertions, 34 deletions
diff --git a/Documentation/ABI/testing/sysfs-bus-usb b/Documentation/ABI/testing/sysfs-bus-usb
index deb6b489e4e5..a07c0f366f91 100644
--- a/Documentation/ABI/testing/sysfs-bus-usb
+++ b/Documentation/ABI/testing/sysfs-bus-usb
@@ -21,25 +21,27 @@ Contact: Alan Stern <stern@rowland.harvard.edu>
21Description: 21Description:
22 Each USB device directory will contain a file named 22 Each USB device directory will contain a file named
23 power/level. This file holds a power-level setting for 23 power/level. This file holds a power-level setting for
24 the device, one of "on", "auto", or "suspend". 24 the device, either "on" or "auto".
25 25
26 "on" means that the device is not allowed to autosuspend, 26 "on" means that the device is not allowed to autosuspend,
27 although normal suspends for system sleep will still 27 although normal suspends for system sleep will still
28 be honored. "auto" means the device will autosuspend 28 be honored. "auto" means the device will autosuspend
29 and autoresume in the usual manner, according to the 29 and autoresume in the usual manner, according to the
30 capabilities of its driver. "suspend" means the device 30 capabilities of its driver.
31 is forced into a suspended state and it will not autoresume
32 in response to I/O requests. However remote-wakeup requests
33 from the device may still be enabled (the remote-wakeup
34 setting is controlled separately by the power/wakeup
35 attribute).
36 31
37 During normal use, devices should be left in the "auto" 32 During normal use, devices should be left in the "auto"
38 level. The other levels are meant for administrative uses. 33 level. The "on" level is meant for administrative uses.
39 If you want to suspend a device immediately but leave it 34 If you want to suspend a device immediately but leave it
40 free to wake up in response to I/O requests, you should 35 free to wake up in response to I/O requests, you should
41 write "0" to power/autosuspend. 36 write "0" to power/autosuspend.
42 37
38 Device not capable of proper suspend and resume should be
39 left in the "on" level. Although the USB spec requires
40 devices to support suspend/resume, many of them do not.
41 In fact so many don't that by default, the USB core
42 initializes all non-hub devices in the "on" level. Some
43 drivers may change this setting when they are bound.
44
43What: /sys/bus/usb/devices/.../power/persist 45What: /sys/bus/usb/devices/.../power/persist
44Date: May 2007 46Date: May 2007
45KernelVersion: 2.6.23 47KernelVersion: 2.6.23
diff --git a/Documentation/usb/power-management.txt b/Documentation/usb/power-management.txt
index c7c1dc2f8017..3bf6818c8cf5 100644
--- a/Documentation/usb/power-management.txt
+++ b/Documentation/usb/power-management.txt
@@ -71,12 +71,10 @@ being accessed through sysfs, then it definitely is idle.
71 Forms of dynamic PM 71 Forms of dynamic PM
72 ------------------- 72 -------------------
73 73
74Dynamic suspends can occur in two ways: manual and automatic. 74Dynamic suspends occur when the kernel decides to suspend an idle
75"Manual" means that the user has told the kernel to suspend a device, 75device. This is called "autosuspend" for short. In general, a device
76whereas "automatic" means that the kernel has decided all by itself to 76won't be autosuspended unless it has been idle for some minimum period
77suspend a device. Automatic suspend is called "autosuspend" for 77of time, the so-called idle-delay time.
78short. In general, a device won't be autosuspended unless it has been
79idle for some minimum period of time, the so-called idle-delay time.
80 78
81Of course, nothing the kernel does on its own initiative should 79Of course, nothing the kernel does on its own initiative should
82prevent the computer or its devices from working properly. If a 80prevent the computer or its devices from working properly. If a
@@ -96,10 +94,11 @@ idle.
96We can categorize power management events in two broad classes: 94We can categorize power management events in two broad classes:
97external and internal. External events are those triggered by some 95external and internal. External events are those triggered by some
98agent outside the USB stack: system suspend/resume (triggered by 96agent outside the USB stack: system suspend/resume (triggered by
99userspace), manual dynamic suspend/resume (also triggered by 97userspace), manual dynamic resume (also triggered by userspace), and
100userspace), and remote wakeup (triggered by the device). Internal 98remote wakeup (triggered by the device). Internal events are those
101events are those triggered within the USB stack: autosuspend and 99triggered within the USB stack: autosuspend and autoresume. Note that
102autoresume. 100all dynamic suspend events are internal; external agents are not
101allowed to issue dynamic suspends.
103 102
104 103
105 The user interface for dynamic PM 104 The user interface for dynamic PM
@@ -145,9 +144,9 @@ relevant attribute files are: wakeup, level, and autosuspend.
145 number of seconds the device should remain idle before 144 number of seconds the device should remain idle before
146 the kernel will autosuspend it (the idle-delay time). 145 the kernel will autosuspend it (the idle-delay time).
147 The default is 2. 0 means to autosuspend as soon as 146 The default is 2. 0 means to autosuspend as soon as
148 the device becomes idle, and -1 means never to 147 the device becomes idle, and negative values mean
149 autosuspend. You can write a number to the file to 148 never to autosuspend. You can write a number to the
150 change the autosuspend idle-delay time. 149 file to change the autosuspend idle-delay time.
151 150
152Writing "-1" to power/autosuspend and writing "on" to power/level do 151Writing "-1" to power/autosuspend and writing "on" to power/level do
153essentially the same thing -- they both prevent the device from being 152essentially the same thing -- they both prevent the device from being
@@ -377,9 +376,9 @@ the device hasn't been idle for long enough, a delayed workqueue
377routine is automatically set up to carry out the operation when the 376routine is automatically set up to carry out the operation when the
378autosuspend idle-delay has expired. 377autosuspend idle-delay has expired.
379 378
380Autoresume attempts also can fail. This will happen if power/level is 379Autoresume attempts also can fail, although failure would mean that
381set to "suspend" or if the device doesn't manage to resume properly. 380the device is no longer present or operating properly. Unlike
382Unlike autosuspend, there's no delay for an autoresume. 381autosuspend, there's no delay for an autoresume.
383 382
384 383
385 Other parts of the driver interface 384 Other parts of the driver interface
@@ -527,13 +526,3 @@ succeed, it may still remain active and thus cause the system to
527resume as soon as the system suspend is complete. Or the remote 526resume as soon as the system suspend is complete. Or the remote
528wakeup may fail and get lost. Which outcome occurs depends on timing 527wakeup may fail and get lost. Which outcome occurs depends on timing
529and on the hardware and firmware design. 528and on the hardware and firmware design.
530
531More interestingly, a device might undergo a manual resume or
532autoresume during system suspend. With current kernels this shouldn't
533happen, because manual resumes must be initiated by userspace and
534autoresumes happen in response to I/O requests, but all user processes
535and I/O should be quiescent during a system suspend -- thanks to the
536freezer. However there are plans to do away with the freezer, which
537would mean these things would become possible. If and when this comes
538about, the USB core will carefully arrange matters so that either type
539of resume will block until the entire system has resumed.