Alan Stern | eaafbc3 | 2007-03-13 16:39:15 -0400 | [diff] [blame] | 1 | What: /sys/bus/usb/devices/.../power/autosuspend |
| 2 | Date: March 2007 |
| 3 | KernelVersion: 2.6.21 |
| 4 | Contact: Alan Stern <stern@rowland.harvard.edu> |
| 5 | Description: |
| 6 | Each USB device directory will contain a file named |
| 7 | power/autosuspend. This file holds the time (in seconds) |
| 8 | the device must be idle before it will be autosuspended. |
| 9 | 0 means the device will be autosuspended as soon as |
| 10 | possible. Negative values will prevent the device from |
| 11 | being autosuspended at all, and writing a negative value |
| 12 | will resume the device if it is already suspended. |
| 13 | |
| 14 | The autosuspend delay for newly-created devices is set to |
| 15 | the value of the usbcore.autosuspend module parameter. |
Alan Stern | 2add522 | 2007-03-20 14:59:39 -0400 | [diff] [blame] | 16 | |
Alan Stern | b41a60e | 2007-05-30 15:39:33 -0400 | [diff] [blame] | 17 | What: /sys/bus/usb/devices/.../power/persist |
| 18 | Date: May 2007 |
| 19 | KernelVersion: 2.6.23 |
| 20 | Contact: Alan Stern <stern@rowland.harvard.edu> |
| 21 | Description: |
| 22 | If CONFIG_USB_PERSIST is set, then each USB device directory |
| 23 | will contain a file named power/persist. The file holds a |
| 24 | boolean value (0 or 1) indicating whether or not the |
| 25 | "USB-Persist" facility is enabled for the device. Since the |
| 26 | facility is inherently dangerous, it is disabled by default |
| 27 | for all devices except hubs. For more information, see |
| 28 | Documentation/usb/persist.txt. |
Sarah Sharp | 979e524 | 2008-01-17 10:24:38 -0800 | [diff] [blame] | 29 | |
| 30 | What: /sys/bus/usb/device/.../power/connected_duration |
| 31 | Date: January 2008 |
| 32 | KernelVersion: 2.6.25 |
| 33 | Contact: Sarah Sharp <sarah.a.sharp@intel.com> |
| 34 | Description: |
| 35 | If CONFIG_PM and CONFIG_USB_SUSPEND are enabled, then this file |
| 36 | is present. When read, it returns the total time (in msec) |
| 37 | that the USB device has been connected to the machine. This |
| 38 | file is read-only. |
| 39 | Users: |
| 40 | PowerTOP <power@bughost.org> |
| 41 | http://www.lesswatts.org/projects/powertop/ |
| 42 | |
| 43 | What: /sys/bus/usb/device/.../power/active_duration |
| 44 | Date: January 2008 |
| 45 | KernelVersion: 2.6.25 |
| 46 | Contact: Sarah Sharp <sarah.a.sharp@intel.com> |
| 47 | Description: |
| 48 | If CONFIG_PM and CONFIG_USB_SUSPEND are enabled, then this file |
| 49 | is present. When read, it returns the total time (in msec) |
| 50 | that the USB device has been active, i.e. not in a suspended |
| 51 | state. This file is read-only. |
| 52 | |
| 53 | Tools can use this file and the connected_duration file to |
| 54 | compute the percentage of time that a device has been active. |
| 55 | For example, |
| 56 | echo $((100 * `cat active_duration` / `cat connected_duration`)) |
| 57 | will give an integer percentage. Note that this does not |
| 58 | account for counter wrap. |
| 59 | Users: |
| 60 | PowerTOP <power@bughost.org> |
| 61 | http://www.lesswatts.org/projects/powertop/ |
David Vrabel | c8cf246 | 2008-09-17 16:34:41 +0100 | [diff] [blame] | 62 | |
Sarah Sharp | 49e7cc8 | 2008-10-06 14:45:46 -0700 | [diff] [blame] | 63 | What: /sys/bus/usb/device/<busnum>-<devnum>...:<config num>-<interface num>/supports_autosuspend |
| 64 | Date: January 2008 |
| 65 | KernelVersion: 2.6.27 |
| 66 | Contact: Sarah Sharp <sarah.a.sharp@intel.com> |
| 67 | Description: |
| 68 | When read, this file returns 1 if the interface driver |
| 69 | for this interface supports autosuspend. It also |
| 70 | returns 1 if no driver has claimed this interface, as an |
| 71 | unclaimed interface will not stop the device from being |
| 72 | autosuspended if all other interface drivers are idle. |
| 73 | The file returns 0 if autosuspend support has not been |
| 74 | added to the driver. |
| 75 | Users: |
| 76 | USB PM tool |
| 77 | git://git.moblin.org/users/sarah/usb-pm-tool/ |
David Vrabel | 61e0e79 | 2008-10-20 16:07:19 +0100 | [diff] [blame] | 78 | |
David Vrabel | c8cf246 | 2008-09-17 16:34:41 +0100 | [diff] [blame] | 79 | What: /sys/bus/usb/device/.../authorized |
| 80 | Date: July 2008 |
| 81 | KernelVersion: 2.6.26 |
| 82 | Contact: David Vrabel <david.vrabel@csr.com> |
| 83 | Description: |
| 84 | Authorized devices are available for use by device |
| 85 | drivers, non-authorized one are not. By default, wired |
| 86 | USB devices are authorized. |
| 87 | |
| 88 | Certified Wireless USB devices are not authorized |
| 89 | initially and should be (by writing 1) after the |
| 90 | device has been authenticated. |
| 91 | |
| 92 | What: /sys/bus/usb/device/.../wusb_cdid |
| 93 | Date: July 2008 |
| 94 | KernelVersion: 2.6.27 |
| 95 | Contact: David Vrabel <david.vrabel@csr.com> |
| 96 | Description: |
| 97 | For Certified Wireless USB devices only. |
| 98 | |
| 99 | A devices's CDID, as 16 space-separated hex octets. |
| 100 | |
| 101 | What: /sys/bus/usb/device/.../wusb_ck |
| 102 | Date: July 2008 |
| 103 | KernelVersion: 2.6.27 |
| 104 | Contact: David Vrabel <david.vrabel@csr.com> |
| 105 | Description: |
| 106 | For Certified Wireless USB devices only. |
| 107 | |
| 108 | Write the device's connection key (CK) to start the |
| 109 | authentication of the device. The CK is 16 |
| 110 | space-separated hex octets. |
| 111 | |
| 112 | What: /sys/bus/usb/device/.../wusb_disconnect |
| 113 | Date: July 2008 |
| 114 | KernelVersion: 2.6.27 |
| 115 | Contact: David Vrabel <david.vrabel@csr.com> |
| 116 | Description: |
| 117 | For Certified Wireless USB devices only. |
| 118 | |
| 119 | Write a 1 to force the device to disconnect |
| 120 | (equivalent to unplugging a wired USB device). |
CHENG Renquan | 0c7a2b7 | 2009-11-22 01:28:52 +0800 | [diff] [blame] | 121 | |
Josua Dietze | ff231db | 2011-10-23 14:22:29 +0200 | [diff] [blame] | 122 | What: /sys/bus/usb/drivers/.../new_id |
| 123 | Date: October 2011 |
| 124 | Contact: linux-usb@vger.kernel.org |
| 125 | Description: |
| 126 | Writing a device ID to this file will attempt to |
| 127 | dynamically add a new device ID to a USB device driver. |
| 128 | This may allow the driver to support more hardware than |
| 129 | was included in the driver's static device ID support |
| 130 | table at compile time. The format for the device ID is: |
| 131 | idVendor idProduct bInterfaceClass. |
| 132 | The vendor ID and device ID fields are required, the |
| 133 | interface class is optional. |
| 134 | Upon successfully adding an ID, the driver will probe |
| 135 | for the device and attempt to bind to it. For example: |
| 136 | # echo "8086 10f5" > /sys/bus/usb/drivers/foo/new_id |
| 137 | |
Bjørn Mork | e6bbcef | 2012-05-13 12:34:59 +0200 | [diff] [blame] | 138 | Reading from this file will list all dynamically added |
| 139 | device IDs in the same format, with one entry per |
| 140 | line. For example: |
| 141 | # cat /sys/bus/usb/drivers/foo/new_id |
| 142 | 8086 10f5 |
| 143 | dead beef 06 |
| 144 | f00d cafe |
| 145 | |
| 146 | The list will be truncated at PAGE_SIZE bytes due to |
| 147 | sysfs restrictions. |
| 148 | |
Josua Dietze | ff231db | 2011-10-23 14:22:29 +0200 | [diff] [blame] | 149 | What: /sys/bus/usb-serial/drivers/.../new_id |
| 150 | Date: October 2011 |
| 151 | Contact: linux-usb@vger.kernel.org |
| 152 | Description: |
| 153 | For serial USB drivers, this attribute appears under the |
| 154 | extra bus folder "usb-serial" in sysfs; apart from that |
| 155 | difference, all descriptions from the entry |
| 156 | "/sys/bus/usb/drivers/.../new_id" apply. |
| 157 | |
CHENG Renquan | 0c7a2b7 | 2009-11-22 01:28:52 +0800 | [diff] [blame] | 158 | What: /sys/bus/usb/drivers/.../remove_id |
| 159 | Date: November 2009 |
| 160 | Contact: CHENG Renquan <rqcheng@smu.edu.sg> |
| 161 | Description: |
| 162 | Writing a device ID to this file will remove an ID |
| 163 | that was dynamically added via the new_id sysfs entry. |
| 164 | The format for the device ID is: |
| 165 | idVendor idProduct. After successfully |
| 166 | removing an ID, the driver will no longer support the |
| 167 | device. This is useful to ensure auto probing won't |
| 168 | match the driver to the device. For example: |
| 169 | # echo "046d c315" > /sys/bus/usb/drivers/foo/remove_id |
Oliver Neukum | ef95534 | 2010-01-16 01:33:03 +0100 | [diff] [blame] | 170 | |
Bjørn Mork | e6bbcef | 2012-05-13 12:34:59 +0200 | [diff] [blame] | 171 | Reading from this file will list the dynamically added |
| 172 | device IDs, exactly like reading from the entry |
| 173 | "/sys/bus/usb/drivers/.../new_id" |
| 174 | |
Oliver Neukum | 1e63ef0 | 2010-03-12 11:27:21 +0100 | [diff] [blame] | 175 | What: /sys/bus/usb/device/.../avoid_reset_quirk |
Oliver Neukum | ef95534 | 2010-01-16 01:33:03 +0100 | [diff] [blame] | 176 | Date: December 2009 |
| 177 | Contact: Oliver Neukum <oliver@neukum.org> |
| 178 | Description: |
| 179 | Writing 1 to this file tells the kernel that this |
| 180 | device will morph into another mode when it is reset. |
| 181 | Drivers will not use reset for error handling for |
| 182 | such devices. |
| 183 | Users: |
| 184 | usb_modeswitch |
Andiry Xu | c1045e8 | 2011-09-23 14:19:53 -0700 | [diff] [blame] | 185 | |
| 186 | What: /sys/bus/usb/devices/.../power/usb2_hardware_lpm |
| 187 | Date: September 2011 |
| 188 | Contact: Andiry Xu <andiry.xu@amd.com> |
| 189 | Description: |
| 190 | If CONFIG_USB_SUSPEND is set and a USB 2.0 lpm-capable device |
| 191 | is plugged in to a xHCI host which support link PM, it will |
| 192 | perform a LPM test; if the test is passed and host supports |
| 193 | USB2 hardware LPM (xHCI 1.0 feature), USB2 hardware LPM will |
| 194 | be enabled for the device and the USB device directory will |
| 195 | contain a file named power/usb2_hardware_lpm. The file holds |
| 196 | a string value (enable or disable) indicating whether or not |
| 197 | USB2 hardware LPM is enabled for the device. Developer can |
| 198 | write y/Y/1 or n/N/0 to the file to enable/disable the |
| 199 | feature. |
Matthew Garrett | 0846e7e | 2012-02-03 17:11:54 -0500 | [diff] [blame] | 200 | |
| 201 | What: /sys/bus/usb/devices/.../removable |
| 202 | Date: February 2012 |
| 203 | Contact: Matthew Garrett <mjg@redhat.com> |
| 204 | Description: |
| 205 | Some information about whether a given USB device is |
| 206 | physically fixed to the platform can be inferred from a |
Masanari Iida | c94bed8e | 2012-04-10 00:22:13 +0900 | [diff] [blame] | 207 | combination of hub descriptor bits and platform-specific data |
Matthew Garrett | 0846e7e | 2012-02-03 17:11:54 -0500 | [diff] [blame] | 208 | such as ACPI. This file will read either "removable" or |
| 209 | "fixed" if the information is available, and "unknown" |
Masanari Iida | c94bed8e | 2012-04-10 00:22:13 +0900 | [diff] [blame] | 210 | otherwise. |
Sarah Sharp | 024f117 | 2012-07-05 17:17:24 -0700 | [diff] [blame] | 211 | |
| 212 | What: /sys/bus/usb/devices/.../ltm_capable |
| 213 | Date: July 2012 |
| 214 | Contact: Sarah Sharp <sarah.a.sharp@linux.intel.com> |
| 215 | Description: |
| 216 | USB 3.0 devices may optionally support Latency Tolerance |
| 217 | Messaging (LTM). They indicate their support by setting a bit |
| 218 | in the bmAttributes field of their SuperSpeed BOS descriptors. |
| 219 | If that bit is set for the device, ltm_capable will read "yes". |
| 220 | If the device doesn't support LTM, the file will read "no". |
| 221 | The file will be present for all speeds of USB devices, and will |
| 222 | always read "no" for USB 1.1 and USB 2.0 devices. |