2
0
mirror of https://github.com/koreader/koreader synced 2024-11-10 01:10:34 +00:00
koreader/plugins/autosuspend.koplugin/main.lua

241 lines
9.2 KiB
Lua
Raw Normal View History

local Device = require("device")
if not Device:isCervantes() and
not Device:isKindle() and
not Device:isKobo() and
not Device:isRemarkable() and
not Device:isSDL() and
not Device:isSonyPRSTUX() and
not Device:isPocketBook() then
return { disabled = true, }
end
local PluginShare = require("pluginshare")
The great Input/GestureDetector/TimeVal spring cleanup (a.k.a., a saner main loop) (#7415) * ReaderDictionary: Port delay computations to TimeVal * ReaderHighlight: Port delay computations to TimeVal * ReaderView: Port delay computations to TimeVal * Android: Reset gesture detection state on APP_CMD_TERM_WINDOW. This prevents potentially being stuck in bogus gesture states when switching apps. * GestureDetector: * Port delay computations to TimeVal * Fixed delay computations to handle time warps (large and negative deltas). * Simplified timed callback handling to invalidate timers much earlier, preventing accumulating useless timers that no longer have any chance of ever detecting a gesture. * Fixed state clearing to handle the actual effective slots, instead of hard-coding slot 0 & slot 1. * Simplified timed callback handling in general, and added support for a timerfd backend for better performance and accuracy. * The improved timed callback handling allows us to detect and honor (as much as possible) the three possible clock sources usable by Linux evdev events. The only case where synthetic timestamps are used (and that only to handle timed callbacks) is limited to non-timerfd platforms where input events use a clock source that is *NOT* MONOTONIC. AFAICT, that's pretty much... PocketBook, and that's it? * Input: * Use the <linux/input.h> FFI module instead of re-declaring every constant * Fixed (verbose) debug logging of input events to actually translate said constants properly. * Completely reset gesture detection state on suspend. This should prevent bogus gesture detection on resume. * Refactored the waitEvent loop to make it easier to comprehend (hopefully) and much more efficient. Of specific note, it no longer does a crazy select spam every 100µs, instead computing and relying on sane timeouts, as afforded by switching the UI event/input loop to the MONOTONIC time base, and the refactored timed callbacks in GestureDetector. * reMarkable: Stopped enforcing synthetic timestamps on input events, as it should no longer be necessary. * TimeVal: * Refactored and simplified, especially as far as metamethods are concerned (based on <bsd/sys/time.h>). * Added a host of new methods to query the various POSIX clock sources, and made :now default to MONOTONIC. * Removed the debug guard in __sub, as time going backwards can be a perfectly normal occurrence. * New methods: * Clock sources: :realtime, :monotonic, :monotonic_coarse, :realtime_coarse, :boottime * Utility: :tonumber, :tousecs, :tomsecs, :fromnumber, :isPositive, :isZero * UIManager: * Ported event loop & scheduling to TimeVal, and switched to the MONOTONIC time base. This ensures reliable and consistent scheduling, as time is ensured never to go backwards. * Added a :getTime() method, that returns a cached TimeVal:now(), updated at the top of every UI frame. It's used throughout the codebase to cadge a syscall in circumstances where we are guaranteed that a syscall would return a mostly identical value, because very few time has passed. The only code left that does live syscalls does it because it's actually necessary for accuracy, and the only code left that does that in a REALTIME time base is code that *actually* deals with calendar time (e.g., Statistics). * DictQuickLookup: Port delay computations to TimeVal * FootNoteWidget: Port delay computations to TimeVal * HTMLBoxWidget: Port delay computations to TimeVal * Notification: Port delay computations to TimeVal * TextBoxWidget: Port delay computations to TimeVal * AutoSuspend: Port to TimeVal * AutoTurn: * Fix it so that settings are actually honored. * Port to TimeVal * BackgroundRunner: Port to TimeVal * Calibre: Port benchmarking code to TimeVal * BookInfoManager: Removed unnecessary yield in the metadata extraction subprocess now that subprocesses get scheduled properly. * All in all, these changes reduced the CPU cost of a single tap by a factor of ten (!), and got rid of an insane amount of weird poll/wakeup cycles that must have been hell on CPU schedulers and batteries..
2021-03-30 00:57:59 +00:00
local TimeVal = require("ui/timeval")
local UIManager = require("ui/uimanager")
local WidgetContainer = require("ui/widget/container/widgetcontainer")
local logger = require("logger")
local _ = require("gettext")
local T = require("ffi/util").template
local default_autoshutdown_timeout_seconds = 3*24*60*60
local default_auto_suspend_timeout_seconds = 60*60
local AutoSuspend = WidgetContainer:new{
name = "autosuspend",
is_doc_only = false,
autoshutdown_timeout_seconds = default_autoshutdown_timeout_seconds,
auto_suspend_timeout_seconds = default_auto_suspend_timeout_seconds,
last_action_tv = TimeVal.zero,
standby_prevented = false,
task = nil,
}
function AutoSuspend:_enabled()
return self.auto_suspend_timeout_seconds > 0
end
function AutoSuspend:_enabledShutdown()
return Device:canPowerOff() and self.autoshutdown_timeout_seconds > 0
end
function AutoSuspend:_schedule(shutdown_only)
if not self:_enabled() and (Device:canPowerOff() and not self:_enabledShutdown()) then
logger.dbg("AutoSuspend:_schedule is disabled")
return
end
local delay_suspend, delay_shutdown
if PluginShare.pause_auto_suspend or Device.standby_prevented or Device.powerd:isCharging() then
delay_suspend = self.auto_suspend_timeout_seconds
delay_shutdown = self.autoshutdown_timeout_seconds
else
The great Input/GestureDetector/TimeVal spring cleanup (a.k.a., a saner main loop) (#7415) * ReaderDictionary: Port delay computations to TimeVal * ReaderHighlight: Port delay computations to TimeVal * ReaderView: Port delay computations to TimeVal * Android: Reset gesture detection state on APP_CMD_TERM_WINDOW. This prevents potentially being stuck in bogus gesture states when switching apps. * GestureDetector: * Port delay computations to TimeVal * Fixed delay computations to handle time warps (large and negative deltas). * Simplified timed callback handling to invalidate timers much earlier, preventing accumulating useless timers that no longer have any chance of ever detecting a gesture. * Fixed state clearing to handle the actual effective slots, instead of hard-coding slot 0 & slot 1. * Simplified timed callback handling in general, and added support for a timerfd backend for better performance and accuracy. * The improved timed callback handling allows us to detect and honor (as much as possible) the three possible clock sources usable by Linux evdev events. The only case where synthetic timestamps are used (and that only to handle timed callbacks) is limited to non-timerfd platforms where input events use a clock source that is *NOT* MONOTONIC. AFAICT, that's pretty much... PocketBook, and that's it? * Input: * Use the <linux/input.h> FFI module instead of re-declaring every constant * Fixed (verbose) debug logging of input events to actually translate said constants properly. * Completely reset gesture detection state on suspend. This should prevent bogus gesture detection on resume. * Refactored the waitEvent loop to make it easier to comprehend (hopefully) and much more efficient. Of specific note, it no longer does a crazy select spam every 100µs, instead computing and relying on sane timeouts, as afforded by switching the UI event/input loop to the MONOTONIC time base, and the refactored timed callbacks in GestureDetector. * reMarkable: Stopped enforcing synthetic timestamps on input events, as it should no longer be necessary. * TimeVal: * Refactored and simplified, especially as far as metamethods are concerned (based on <bsd/sys/time.h>). * Added a host of new methods to query the various POSIX clock sources, and made :now default to MONOTONIC. * Removed the debug guard in __sub, as time going backwards can be a perfectly normal occurrence. * New methods: * Clock sources: :realtime, :monotonic, :monotonic_coarse, :realtime_coarse, :boottime * Utility: :tonumber, :tousecs, :tomsecs, :fromnumber, :isPositive, :isZero * UIManager: * Ported event loop & scheduling to TimeVal, and switched to the MONOTONIC time base. This ensures reliable and consistent scheduling, as time is ensured never to go backwards. * Added a :getTime() method, that returns a cached TimeVal:now(), updated at the top of every UI frame. It's used throughout the codebase to cadge a syscall in circumstances where we are guaranteed that a syscall would return a mostly identical value, because very few time has passed. The only code left that does live syscalls does it because it's actually necessary for accuracy, and the only code left that does that in a REALTIME time base is code that *actually* deals with calendar time (e.g., Statistics). * DictQuickLookup: Port delay computations to TimeVal * FootNoteWidget: Port delay computations to TimeVal * HTMLBoxWidget: Port delay computations to TimeVal * Notification: Port delay computations to TimeVal * TextBoxWidget: Port delay computations to TimeVal * AutoSuspend: Port to TimeVal * AutoTurn: * Fix it so that settings are actually honored. * Port to TimeVal * BackgroundRunner: Port to TimeVal * Calibre: Port benchmarking code to TimeVal * BookInfoManager: Removed unnecessary yield in the metadata extraction subprocess now that subprocesses get scheduled properly. * All in all, these changes reduced the CPU cost of a single tap by a factor of ten (!), and got rid of an insane amount of weird poll/wakeup cycles that must have been hell on CPU schedulers and batteries..
2021-03-30 00:57:59 +00:00
local now_tv = UIManager:getTime()
delay_suspend = self.last_action_tv + TimeVal:new{ sec = self.auto_suspend_timeout_seconds, usec = 0 } - now_tv
The great Input/GestureDetector/TimeVal spring cleanup (a.k.a., a saner main loop) (#7415) * ReaderDictionary: Port delay computations to TimeVal * ReaderHighlight: Port delay computations to TimeVal * ReaderView: Port delay computations to TimeVal * Android: Reset gesture detection state on APP_CMD_TERM_WINDOW. This prevents potentially being stuck in bogus gesture states when switching apps. * GestureDetector: * Port delay computations to TimeVal * Fixed delay computations to handle time warps (large and negative deltas). * Simplified timed callback handling to invalidate timers much earlier, preventing accumulating useless timers that no longer have any chance of ever detecting a gesture. * Fixed state clearing to handle the actual effective slots, instead of hard-coding slot 0 & slot 1. * Simplified timed callback handling in general, and added support for a timerfd backend for better performance and accuracy. * The improved timed callback handling allows us to detect and honor (as much as possible) the three possible clock sources usable by Linux evdev events. The only case where synthetic timestamps are used (and that only to handle timed callbacks) is limited to non-timerfd platforms where input events use a clock source that is *NOT* MONOTONIC. AFAICT, that's pretty much... PocketBook, and that's it? * Input: * Use the <linux/input.h> FFI module instead of re-declaring every constant * Fixed (verbose) debug logging of input events to actually translate said constants properly. * Completely reset gesture detection state on suspend. This should prevent bogus gesture detection on resume. * Refactored the waitEvent loop to make it easier to comprehend (hopefully) and much more efficient. Of specific note, it no longer does a crazy select spam every 100µs, instead computing and relying on sane timeouts, as afforded by switching the UI event/input loop to the MONOTONIC time base, and the refactored timed callbacks in GestureDetector. * reMarkable: Stopped enforcing synthetic timestamps on input events, as it should no longer be necessary. * TimeVal: * Refactored and simplified, especially as far as metamethods are concerned (based on <bsd/sys/time.h>). * Added a host of new methods to query the various POSIX clock sources, and made :now default to MONOTONIC. * Removed the debug guard in __sub, as time going backwards can be a perfectly normal occurrence. * New methods: * Clock sources: :realtime, :monotonic, :monotonic_coarse, :realtime_coarse, :boottime * Utility: :tonumber, :tousecs, :tomsecs, :fromnumber, :isPositive, :isZero * UIManager: * Ported event loop & scheduling to TimeVal, and switched to the MONOTONIC time base. This ensures reliable and consistent scheduling, as time is ensured never to go backwards. * Added a :getTime() method, that returns a cached TimeVal:now(), updated at the top of every UI frame. It's used throughout the codebase to cadge a syscall in circumstances where we are guaranteed that a syscall would return a mostly identical value, because very few time has passed. The only code left that does live syscalls does it because it's actually necessary for accuracy, and the only code left that does that in a REALTIME time base is code that *actually* deals with calendar time (e.g., Statistics). * DictQuickLookup: Port delay computations to TimeVal * FootNoteWidget: Port delay computations to TimeVal * HTMLBoxWidget: Port delay computations to TimeVal * Notification: Port delay computations to TimeVal * TextBoxWidget: Port delay computations to TimeVal * AutoSuspend: Port to TimeVal * AutoTurn: * Fix it so that settings are actually honored. * Port to TimeVal * BackgroundRunner: Port to TimeVal * Calibre: Port benchmarking code to TimeVal * BookInfoManager: Removed unnecessary yield in the metadata extraction subprocess now that subprocesses get scheduled properly. * All in all, these changes reduced the CPU cost of a single tap by a factor of ten (!), and got rid of an insane amount of weird poll/wakeup cycles that must have been hell on CPU schedulers and batteries..
2021-03-30 00:57:59 +00:00
delay_suspend = delay_suspend:tonumber()
delay_shutdown = self.last_action_tv + TimeVal:new{ sec = self.autoshutdown_timeout_seconds, usec = 0 } - now_tv
The great Input/GestureDetector/TimeVal spring cleanup (a.k.a., a saner main loop) (#7415) * ReaderDictionary: Port delay computations to TimeVal * ReaderHighlight: Port delay computations to TimeVal * ReaderView: Port delay computations to TimeVal * Android: Reset gesture detection state on APP_CMD_TERM_WINDOW. This prevents potentially being stuck in bogus gesture states when switching apps. * GestureDetector: * Port delay computations to TimeVal * Fixed delay computations to handle time warps (large and negative deltas). * Simplified timed callback handling to invalidate timers much earlier, preventing accumulating useless timers that no longer have any chance of ever detecting a gesture. * Fixed state clearing to handle the actual effective slots, instead of hard-coding slot 0 & slot 1. * Simplified timed callback handling in general, and added support for a timerfd backend for better performance and accuracy. * The improved timed callback handling allows us to detect and honor (as much as possible) the three possible clock sources usable by Linux evdev events. The only case where synthetic timestamps are used (and that only to handle timed callbacks) is limited to non-timerfd platforms where input events use a clock source that is *NOT* MONOTONIC. AFAICT, that's pretty much... PocketBook, and that's it? * Input: * Use the <linux/input.h> FFI module instead of re-declaring every constant * Fixed (verbose) debug logging of input events to actually translate said constants properly. * Completely reset gesture detection state on suspend. This should prevent bogus gesture detection on resume. * Refactored the waitEvent loop to make it easier to comprehend (hopefully) and much more efficient. Of specific note, it no longer does a crazy select spam every 100µs, instead computing and relying on sane timeouts, as afforded by switching the UI event/input loop to the MONOTONIC time base, and the refactored timed callbacks in GestureDetector. * reMarkable: Stopped enforcing synthetic timestamps on input events, as it should no longer be necessary. * TimeVal: * Refactored and simplified, especially as far as metamethods are concerned (based on <bsd/sys/time.h>). * Added a host of new methods to query the various POSIX clock sources, and made :now default to MONOTONIC. * Removed the debug guard in __sub, as time going backwards can be a perfectly normal occurrence. * New methods: * Clock sources: :realtime, :monotonic, :monotonic_coarse, :realtime_coarse, :boottime * Utility: :tonumber, :tousecs, :tomsecs, :fromnumber, :isPositive, :isZero * UIManager: * Ported event loop & scheduling to TimeVal, and switched to the MONOTONIC time base. This ensures reliable and consistent scheduling, as time is ensured never to go backwards. * Added a :getTime() method, that returns a cached TimeVal:now(), updated at the top of every UI frame. It's used throughout the codebase to cadge a syscall in circumstances where we are guaranteed that a syscall would return a mostly identical value, because very few time has passed. The only code left that does live syscalls does it because it's actually necessary for accuracy, and the only code left that does that in a REALTIME time base is code that *actually* deals with calendar time (e.g., Statistics). * DictQuickLookup: Port delay computations to TimeVal * FootNoteWidget: Port delay computations to TimeVal * HTMLBoxWidget: Port delay computations to TimeVal * Notification: Port delay computations to TimeVal * TextBoxWidget: Port delay computations to TimeVal * AutoSuspend: Port to TimeVal * AutoTurn: * Fix it so that settings are actually honored. * Port to TimeVal * BackgroundRunner: Port to TimeVal * Calibre: Port benchmarking code to TimeVal * BookInfoManager: Removed unnecessary yield in the metadata extraction subprocess now that subprocesses get scheduled properly. * All in all, these changes reduced the CPU cost of a single tap by a factor of ten (!), and got rid of an insane amount of weird poll/wakeup cycles that must have been hell on CPU schedulers and batteries..
2021-03-30 00:57:59 +00:00
delay_shutdown = delay_shutdown:tonumber()
end
-- Try to shutdown first, as we may have been woken up from suspend just for the sole purpose of doing that.
if delay_shutdown <= 0 then
logger.dbg("AutoSuspend: initiating shutdown")
UIManager:poweroff_action()
elseif delay_suspend <= 0 and not shutdown_only then
logger.dbg("AutoSuspend: will suspend the device")
UIManager:suspend()
else
if self:_enabled() and not shutdown_only then
logger.dbg("AutoSuspend: scheduling next suspend check in", delay_suspend)
UIManager:scheduleIn(delay_suspend, self.task)
end
if self:_enabledShutdown() then
logger.dbg("AutoSuspend: scheduling next shutdown check in", delay_shutdown)
UIManager:scheduleIn(delay_shutdown, self.task)
end
end
end
function AutoSuspend:_unschedule()
if self.task then
logger.dbg("AutoSuspend: unschedule")
UIManager:unschedule(self.task)
end
end
function AutoSuspend:_start()
if self:_enabled() or self:_enabledShutdown() then
The great Input/GestureDetector/TimeVal spring cleanup (a.k.a., a saner main loop) (#7415) * ReaderDictionary: Port delay computations to TimeVal * ReaderHighlight: Port delay computations to TimeVal * ReaderView: Port delay computations to TimeVal * Android: Reset gesture detection state on APP_CMD_TERM_WINDOW. This prevents potentially being stuck in bogus gesture states when switching apps. * GestureDetector: * Port delay computations to TimeVal * Fixed delay computations to handle time warps (large and negative deltas). * Simplified timed callback handling to invalidate timers much earlier, preventing accumulating useless timers that no longer have any chance of ever detecting a gesture. * Fixed state clearing to handle the actual effective slots, instead of hard-coding slot 0 & slot 1. * Simplified timed callback handling in general, and added support for a timerfd backend for better performance and accuracy. * The improved timed callback handling allows us to detect and honor (as much as possible) the three possible clock sources usable by Linux evdev events. The only case where synthetic timestamps are used (and that only to handle timed callbacks) is limited to non-timerfd platforms where input events use a clock source that is *NOT* MONOTONIC. AFAICT, that's pretty much... PocketBook, and that's it? * Input: * Use the <linux/input.h> FFI module instead of re-declaring every constant * Fixed (verbose) debug logging of input events to actually translate said constants properly. * Completely reset gesture detection state on suspend. This should prevent bogus gesture detection on resume. * Refactored the waitEvent loop to make it easier to comprehend (hopefully) and much more efficient. Of specific note, it no longer does a crazy select spam every 100µs, instead computing and relying on sane timeouts, as afforded by switching the UI event/input loop to the MONOTONIC time base, and the refactored timed callbacks in GestureDetector. * reMarkable: Stopped enforcing synthetic timestamps on input events, as it should no longer be necessary. * TimeVal: * Refactored and simplified, especially as far as metamethods are concerned (based on <bsd/sys/time.h>). * Added a host of new methods to query the various POSIX clock sources, and made :now default to MONOTONIC. * Removed the debug guard in __sub, as time going backwards can be a perfectly normal occurrence. * New methods: * Clock sources: :realtime, :monotonic, :monotonic_coarse, :realtime_coarse, :boottime * Utility: :tonumber, :tousecs, :tomsecs, :fromnumber, :isPositive, :isZero * UIManager: * Ported event loop & scheduling to TimeVal, and switched to the MONOTONIC time base. This ensures reliable and consistent scheduling, as time is ensured never to go backwards. * Added a :getTime() method, that returns a cached TimeVal:now(), updated at the top of every UI frame. It's used throughout the codebase to cadge a syscall in circumstances where we are guaranteed that a syscall would return a mostly identical value, because very few time has passed. The only code left that does live syscalls does it because it's actually necessary for accuracy, and the only code left that does that in a REALTIME time base is code that *actually* deals with calendar time (e.g., Statistics). * DictQuickLookup: Port delay computations to TimeVal * FootNoteWidget: Port delay computations to TimeVal * HTMLBoxWidget: Port delay computations to TimeVal * Notification: Port delay computations to TimeVal * TextBoxWidget: Port delay computations to TimeVal * AutoSuspend: Port to TimeVal * AutoTurn: * Fix it so that settings are actually honored. * Port to TimeVal * BackgroundRunner: Port to TimeVal * Calibre: Port benchmarking code to TimeVal * BookInfoManager: Removed unnecessary yield in the metadata extraction subprocess now that subprocesses get scheduled properly. * All in all, these changes reduced the CPU cost of a single tap by a factor of ten (!), and got rid of an insane amount of weird poll/wakeup cycles that must have been hell on CPU schedulers and batteries..
2021-03-30 00:57:59 +00:00
local now_tv = UIManager:getTime()
logger.dbg("AutoSuspend: start at", now_tv:tonumber())
self.last_action_tv = now_tv
self:_schedule()
end
end
-- Variant that only re-engages the shutdown timer for onUnexpectedWakeupLimit
function AutoSuspend:_restart()
if self:_enabledShutdown() then
The great Input/GestureDetector/TimeVal spring cleanup (a.k.a., a saner main loop) (#7415) * ReaderDictionary: Port delay computations to TimeVal * ReaderHighlight: Port delay computations to TimeVal * ReaderView: Port delay computations to TimeVal * Android: Reset gesture detection state on APP_CMD_TERM_WINDOW. This prevents potentially being stuck in bogus gesture states when switching apps. * GestureDetector: * Port delay computations to TimeVal * Fixed delay computations to handle time warps (large and negative deltas). * Simplified timed callback handling to invalidate timers much earlier, preventing accumulating useless timers that no longer have any chance of ever detecting a gesture. * Fixed state clearing to handle the actual effective slots, instead of hard-coding slot 0 & slot 1. * Simplified timed callback handling in general, and added support for a timerfd backend for better performance and accuracy. * The improved timed callback handling allows us to detect and honor (as much as possible) the three possible clock sources usable by Linux evdev events. The only case where synthetic timestamps are used (and that only to handle timed callbacks) is limited to non-timerfd platforms where input events use a clock source that is *NOT* MONOTONIC. AFAICT, that's pretty much... PocketBook, and that's it? * Input: * Use the <linux/input.h> FFI module instead of re-declaring every constant * Fixed (verbose) debug logging of input events to actually translate said constants properly. * Completely reset gesture detection state on suspend. This should prevent bogus gesture detection on resume. * Refactored the waitEvent loop to make it easier to comprehend (hopefully) and much more efficient. Of specific note, it no longer does a crazy select spam every 100µs, instead computing and relying on sane timeouts, as afforded by switching the UI event/input loop to the MONOTONIC time base, and the refactored timed callbacks in GestureDetector. * reMarkable: Stopped enforcing synthetic timestamps on input events, as it should no longer be necessary. * TimeVal: * Refactored and simplified, especially as far as metamethods are concerned (based on <bsd/sys/time.h>). * Added a host of new methods to query the various POSIX clock sources, and made :now default to MONOTONIC. * Removed the debug guard in __sub, as time going backwards can be a perfectly normal occurrence. * New methods: * Clock sources: :realtime, :monotonic, :monotonic_coarse, :realtime_coarse, :boottime * Utility: :tonumber, :tousecs, :tomsecs, :fromnumber, :isPositive, :isZero * UIManager: * Ported event loop & scheduling to TimeVal, and switched to the MONOTONIC time base. This ensures reliable and consistent scheduling, as time is ensured never to go backwards. * Added a :getTime() method, that returns a cached TimeVal:now(), updated at the top of every UI frame. It's used throughout the codebase to cadge a syscall in circumstances where we are guaranteed that a syscall would return a mostly identical value, because very few time has passed. The only code left that does live syscalls does it because it's actually necessary for accuracy, and the only code left that does that in a REALTIME time base is code that *actually* deals with calendar time (e.g., Statistics). * DictQuickLookup: Port delay computations to TimeVal * FootNoteWidget: Port delay computations to TimeVal * HTMLBoxWidget: Port delay computations to TimeVal * Notification: Port delay computations to TimeVal * TextBoxWidget: Port delay computations to TimeVal * AutoSuspend: Port to TimeVal * AutoTurn: * Fix it so that settings are actually honored. * Port to TimeVal * BackgroundRunner: Port to TimeVal * Calibre: Port benchmarking code to TimeVal * BookInfoManager: Removed unnecessary yield in the metadata extraction subprocess now that subprocesses get scheduled properly. * All in all, these changes reduced the CPU cost of a single tap by a factor of ten (!), and got rid of an insane amount of weird poll/wakeup cycles that must have been hell on CPU schedulers and batteries..
2021-03-30 00:57:59 +00:00
local now_tv = UIManager:getTime()
logger.dbg("AutoSuspend: restart at", now_tv:tonumber())
self.last_action_tv = now_tv
self:_schedule(true)
end
end
function AutoSuspend:init()
logger.dbg("AutoSuspend: init")
if Device:isPocketBook() and not Device:canSuspend() then return end
self.autoshutdown_timeout_seconds = G_reader_settings:readSetting("autoshutdown_timeout_seconds") or default_autoshutdown_timeout_seconds
self.auto_suspend_timeout_seconds = G_reader_settings:readSetting("auto_suspend_timeout_seconds") or default_auto_suspend_timeout_seconds
UIManager.event_hook:registerWidget("InputEvent", self)
-- We need an instance-specific function reference to schedule, because in some rare cases,
-- we may instantiate a new plugin instance *before* tearing down the old one.
self.task = function(shutdown_only)
self:_schedule(shutdown_only)
end
self:_start()
-- self.ui is nil in the testsuite
if not self.ui or not self.ui.menu then return end
self.ui.menu:registerToMainMenu(self)
end
-- For event_hook automagic deregistration purposes
function AutoSuspend:onCloseWidget()
logger.dbg("AutoSuspend: onCloseWidget")
if Device:isPocketBook() and not Device:canSuspend() then return end
self:_unschedule()
self.task = nil
end
function AutoSuspend:onInputEvent()
logger.dbg("AutoSuspend: onInputEvent")
The great Input/GestureDetector/TimeVal spring cleanup (a.k.a., a saner main loop) (#7415) * ReaderDictionary: Port delay computations to TimeVal * ReaderHighlight: Port delay computations to TimeVal * ReaderView: Port delay computations to TimeVal * Android: Reset gesture detection state on APP_CMD_TERM_WINDOW. This prevents potentially being stuck in bogus gesture states when switching apps. * GestureDetector: * Port delay computations to TimeVal * Fixed delay computations to handle time warps (large and negative deltas). * Simplified timed callback handling to invalidate timers much earlier, preventing accumulating useless timers that no longer have any chance of ever detecting a gesture. * Fixed state clearing to handle the actual effective slots, instead of hard-coding slot 0 & slot 1. * Simplified timed callback handling in general, and added support for a timerfd backend for better performance and accuracy. * The improved timed callback handling allows us to detect and honor (as much as possible) the three possible clock sources usable by Linux evdev events. The only case where synthetic timestamps are used (and that only to handle timed callbacks) is limited to non-timerfd platforms where input events use a clock source that is *NOT* MONOTONIC. AFAICT, that's pretty much... PocketBook, and that's it? * Input: * Use the <linux/input.h> FFI module instead of re-declaring every constant * Fixed (verbose) debug logging of input events to actually translate said constants properly. * Completely reset gesture detection state on suspend. This should prevent bogus gesture detection on resume. * Refactored the waitEvent loop to make it easier to comprehend (hopefully) and much more efficient. Of specific note, it no longer does a crazy select spam every 100µs, instead computing and relying on sane timeouts, as afforded by switching the UI event/input loop to the MONOTONIC time base, and the refactored timed callbacks in GestureDetector. * reMarkable: Stopped enforcing synthetic timestamps on input events, as it should no longer be necessary. * TimeVal: * Refactored and simplified, especially as far as metamethods are concerned (based on <bsd/sys/time.h>). * Added a host of new methods to query the various POSIX clock sources, and made :now default to MONOTONIC. * Removed the debug guard in __sub, as time going backwards can be a perfectly normal occurrence. * New methods: * Clock sources: :realtime, :monotonic, :monotonic_coarse, :realtime_coarse, :boottime * Utility: :tonumber, :tousecs, :tomsecs, :fromnumber, :isPositive, :isZero * UIManager: * Ported event loop & scheduling to TimeVal, and switched to the MONOTONIC time base. This ensures reliable and consistent scheduling, as time is ensured never to go backwards. * Added a :getTime() method, that returns a cached TimeVal:now(), updated at the top of every UI frame. It's used throughout the codebase to cadge a syscall in circumstances where we are guaranteed that a syscall would return a mostly identical value, because very few time has passed. The only code left that does live syscalls does it because it's actually necessary for accuracy, and the only code left that does that in a REALTIME time base is code that *actually* deals with calendar time (e.g., Statistics). * DictQuickLookup: Port delay computations to TimeVal * FootNoteWidget: Port delay computations to TimeVal * HTMLBoxWidget: Port delay computations to TimeVal * Notification: Port delay computations to TimeVal * TextBoxWidget: Port delay computations to TimeVal * AutoSuspend: Port to TimeVal * AutoTurn: * Fix it so that settings are actually honored. * Port to TimeVal * BackgroundRunner: Port to TimeVal * Calibre: Port benchmarking code to TimeVal * BookInfoManager: Removed unnecessary yield in the metadata extraction subprocess now that subprocesses get scheduled properly. * All in all, these changes reduced the CPU cost of a single tap by a factor of ten (!), and got rid of an insane amount of weird poll/wakeup cycles that must have been hell on CPU schedulers and batteries..
2021-03-30 00:57:59 +00:00
self.last_action_tv = UIManager:getTime()
end
function AutoSuspend:onSuspend()
logger.dbg("AutoSuspend: onSuspend")
-- We do not want auto suspend procedure to waste battery during suspend. So let's unschedule it
-- when suspending and restart it after resume.
self:_unschedule()
if self:_enabledShutdown() and Device.wakeup_mgr then
Device.wakeup_mgr:addTask(self.autoshutdown_timeout_seconds, UIManager.poweroff_action)
end
end
function AutoSuspend:onResume()
logger.dbg("AutoSuspend: onResume")
if self:_enabledShutdown() and Device.wakeup_mgr then
Device.wakeup_mgr:removeTask(nil, nil, UIManager.poweroff_action)
end
-- Unschedule in case we tripped onUnexpectedWakeupLimit first...
self:_unschedule()
self:_start()
end
function AutoSuspend:onUnexpectedWakeupLimit()
logger.dbg("AutoSuspend: onUnexpectedWakeupLimit")
-- Only re-engage the *shutdown* schedule to avoid doing the same dance indefinitely.
self:_restart()
end
function AutoSuspend:onAllowStandby()
self.standby_prevented = false
end
function AutoSuspend:onPreventStandby()
self.standby_prevented = true
end
function AutoSuspend:addToMainMenu(menu_items)
menu_items.autosuspend = {
sorting_hint = "device",
text = _("Autosuspend timeout"),
callback = function()
local InfoMessage = require("ui/widget/infomessage")
local Screen = Device.screen
local SpinWidget = require("ui/widget/spinwidget")
local autosuspend_spin = SpinWidget:new {
width = math.floor(Screen:getWidth() * 0.6),
value = self.auto_suspend_timeout_seconds / 60,
value_min = 5,
value_max = 240,
value_hold_step = 15,
ok_text = _("Set timeout"),
title_text = _("Timeout in minutes"),
callback = function(autosuspend_spin)
self.auto_suspend_timeout_seconds = autosuspend_spin.value * 60
G_reader_settings:saveSetting("auto_suspend_timeout_seconds", self.auto_suspend_timeout_seconds)
UIManager:show(InfoMessage:new{
text = T(_("The system will automatically suspend after %1 minutes of inactivity."),
string.format("%.2f", self.auto_suspend_timeout_seconds / 60)),
timeout = 3,
})
self:_unschedule()
self:_start()
end
}
UIManager:show(autosuspend_spin)
end,
}
if not (Device:canPowerOff() or Device:isEmulator()) then return end
menu_items.autoshutdown = {
sorting_hint = "device",
text = _("Autoshutdown timeout"),
callback = function()
local InfoMessage = require("ui/widget/infomessage")
local Screen = Device.screen
local SpinWidget = require("ui/widget/spinwidget")
local autosuspend_spin = SpinWidget:new {
width = math.floor(Screen:getWidth() * 0.6),
value = self.autoshutdown_timeout_seconds / 60 / 60,
-- About a minute, good for testing and battery life fanatics.
-- Just high enough to avoid an instant shutdown death scenario.
value_min = 0.017,
-- More than three weeks seems a bit excessive if you want to enable authoshutdown,
-- even if the battery can last up to three months.
value_max = 28*24,
value_hold_step = 24,
precision = "%.2f",
ok_text = _("Set timeout"),
title_text = _("Timeout in hours"),
callback = function(autosuspend_spin)
self.autoshutdown_timeout_seconds = math.floor(autosuspend_spin.value * 60 * 60)
G_reader_settings:saveSetting("autoshutdown_timeout_seconds", self.autoshutdown_timeout_seconds)
UIManager:show(InfoMessage:new{
text = T(_("The system will automatically shut down after %1 hours of inactivity."),
string.format("%.2f", self.autoshutdown_timeout_seconds / 60 / 60)),
timeout = 3,
})
self:_unschedule()
self:_start()
end
}
UIManager:show(autosuspend_spin)
end,
}
end
return AutoSuspend