acpi-lid.txt 4.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596
  1. Special Usage Model of the ACPI Control Method Lid Device
  2. Copyright (C) 2016, Intel Corporation
  3. Author: Lv Zheng <lv.zheng@intel.com>
  4. Abstract:
  5. Platforms containing lids convey lid state (open/close) to OSPMs using a
  6. control method lid device. To implement this, the AML tables issue
  7. Notify(lid_device, 0x80) to notify the OSPMs whenever the lid state has
  8. changed. The _LID control method for the lid device must be implemented to
  9. report the "current" state of the lid as either "opened" or "closed".
  10. For most platforms, both the _LID method and the lid notifications are
  11. reliable. However, there are exceptions. In order to work with these
  12. exceptional buggy platforms, special restrictions and expections should be
  13. taken into account. This document describes the restrictions and the
  14. expections of the Linux ACPI lid device driver.
  15. 1. Restrictions of the returning value of the _LID control method
  16. The _LID control method is described to return the "current" lid state.
  17. However the word of "current" has ambiguity, some buggy AML tables return
  18. the lid state upon the last lid notification instead of returning the lid
  19. state upon the last _LID evaluation. There won't be difference when the
  20. _LID control method is evaluated during the runtime, the problem is its
  21. initial returning value. When the AML tables implement this control method
  22. with cached value, the initial returning value is likely not reliable.
  23. There are platforms always retun "closed" as initial lid state.
  24. 2. Restrictions of the lid state change notifications
  25. There are buggy AML tables never notifying when the lid device state is
  26. changed to "opened". Thus the "opened" notification is not guaranteed. But
  27. it is guaranteed that the AML tables always notify "closed" when the lid
  28. state is changed to "closed". The "closed" notification is normally used to
  29. trigger some system power saving operations on Windows. Since it is fully
  30. tested, it is reliable from all AML tables.
  31. 3. Expections for the userspace users of the ACPI lid device driver
  32. The ACPI button driver exports the lid state to the userspace via the
  33. following file:
  34. /proc/acpi/button/lid/LID0/state
  35. This file actually calls the _LID control method described above. And given
  36. the previous explanation, it is not reliable enough on some platforms. So
  37. it is advised for the userspace program to not to solely rely on this file
  38. to determine the actual lid state.
  39. The ACPI button driver emits the following input event to the userspace:
  40. SW_LID
  41. The ACPI lid device driver is implemented to try to deliver the platform
  42. triggered events to the userspace. However, given the fact that the buggy
  43. firmware cannot make sure "opened"/"closed" events are paired, the ACPI
  44. button driver uses the following 3 modes in order not to trigger issues.
  45. If the userspace hasn't been prepared to ignore the unreliable "opened"
  46. events and the unreliable initial state notification, Linux users can use
  47. the following kernel parameters to handle the possible issues:
  48. A. button.lid_init_state=method:
  49. When this option is specified, the ACPI button driver reports the
  50. initial lid state using the returning value of the _LID control method
  51. and whether the "opened"/"closed" events are paired fully relies on the
  52. firmware implementation.
  53. This option can be used to fix some platforms where the returning value
  54. of the _LID control method is reliable but the initial lid state
  55. notification is missing.
  56. This option is the default behavior during the period the userspace
  57. isn't ready to handle the buggy AML tables.
  58. B. button.lid_init_state=open:
  59. When this option is specified, the ACPI button driver always reports the
  60. initial lid state as "opened" and whether the "opened"/"closed" events
  61. are paired fully relies on the firmware implementation.
  62. This may fix some platforms where the returning value of the _LID
  63. control method is not reliable and the initial lid state notification is
  64. missing.
  65. If the userspace has been prepared to ignore the unreliable "opened" events
  66. and the unreliable initial state notification, Linux users should always
  67. use the following kernel parameter:
  68. C. button.lid_init_state=ignore:
  69. When this option is specified, the ACPI button driver never reports the
  70. initial lid state and there is a compensation mechanism implemented to
  71. ensure that the reliable "closed" notifications can always be delievered
  72. to the userspace by always pairing "closed" input events with complement
  73. "opened" input events. But there is still no guarantee that the "opened"
  74. notifications can be delivered to the userspace when the lid is actually
  75. opens given that some AML tables do not send "opened" notifications
  76. reliably.
  77. In this mode, if everything is correctly implemented by the platform
  78. firmware, the old userspace programs should still work. Otherwise, the
  79. new userspace programs are required to work with the ACPI button driver.
  80. This option will be the default behavior after the userspace is ready to
  81. handle the buggy AML tables.