mirror of
https://github.com/safing/portmaster
synced 2025-04-20 10:59:10 +00:00
* Move portbase into monorepo * Add new simple module mgr * [WIP] Switch to new simple module mgr * Add StateMgr and more worker variants * [WIP] Switch more modules * [WIP] Switch more modules * [WIP] swtich more modules * [WIP] switch all SPN modules * [WIP] switch all service modules * [WIP] Convert all workers to the new module system * [WIP] add new task system to module manager * [WIP] Add second take for scheduling workers * [WIP] Add FIXME for bugs in new scheduler * [WIP] Add minor improvements to scheduler * [WIP] Add new worker scheduler * [WIP] Fix more bug related to new module system * [WIP] Fix start handing of the new module system * [WIP] Improve startup process * [WIP] Fix minor issues * [WIP] Fix missing subsystem in settings * [WIP] Initialize managers in constructor * [WIP] Move module event initialization to constrictors * [WIP] Fix setting for enabling and disabling the SPN module * [WIP] Move API registeration into module construction * [WIP] Update states mgr for all modules * [WIP] Add CmdLine operation support * Add state helper methods to module group and instance * Add notification and module status handling to status package * Fix starting issues * Remove pilot widget and update security lock to new status data * Remove debug logs * Improve http server shutdown * Add workaround for cleanly shutting down firewall+netquery * Improve logging * Add syncing states with notifications for new module system * Improve starting, stopping, shutdown; resolve FIXMEs/TODOs * [WIP] Fix most unit tests * Review new module system and fix minor issues * Push shutdown and restart events again via API * Set sleep mode via interface * Update example/template module * [WIP] Fix spn/cabin unit test * Remove deprecated UI elements * Make log output more similar for the logging transition phase * Switch spn hub and observer cmds to new module system * Fix log sources * Make worker mgr less error prone * Fix tests and minor issues * Fix observation hub * Improve shutdown and restart handling * Split up big connection.go source file * Move varint and dsd packages to structures repo * Improve expansion test * Fix linter warnings * Fix interception module on windows * Fix linter errors --------- Co-authored-by: Vladimir Stoilov <vladimir@safing.io>
86 lines
2.9 KiB
Go
86 lines
2.9 KiB
Go
package utils
|
|
|
|
// This file is forked from https://github.com/golang/go/blob/bc593eac2dc63d979a575eccb16c7369a5ff81e0/src/sync/once.go.
|
|
|
|
import (
|
|
"sync"
|
|
"sync/atomic"
|
|
)
|
|
|
|
// OnceAgain is an object that will perform only one action "in flight". It's
|
|
// basically the same as sync.Once, but is automatically reused when the
|
|
// function was executed and everyone who waited has left.
|
|
// Important: This is somewhat racy when used heavily as it only resets _after_
|
|
// everyone who waited has left. So, while some goroutines are waiting to be
|
|
// activated again to leave the waiting state, other goroutines will call Do()
|
|
// without executing the function again.
|
|
type OnceAgain struct {
|
|
// done indicates whether the action has been performed.
|
|
// It is first in the struct because it is used in the hot path.
|
|
// The hot path is inlined at every call site.
|
|
// Placing done first allows more compact instructions on some architectures (amd64/x86),
|
|
// and fewer instructions (to calculate offset) on other architectures.
|
|
done uint32
|
|
|
|
// Number of waiters waiting for the function to finish. The last waiter resets done.
|
|
waiters int32
|
|
|
|
m sync.Mutex
|
|
}
|
|
|
|
// Do calls the function f if and only if Do is being called for the
|
|
// first time for this instance of Once. In other words, given
|
|
//
|
|
// var once Once
|
|
//
|
|
// if once.Do(f) is called multiple times, only the first call will invoke f,
|
|
// even if f has a different value in each invocation. A new instance of
|
|
// Once is required for each function to execute.
|
|
//
|
|
// Do is intended for initialization that must be run exactly once. Since f
|
|
// is niladic, it may be necessary to use a function literal to capture the
|
|
// arguments to a function to be invoked by Do:
|
|
//
|
|
// config.once.Do(func() { config.init(filename) })
|
|
//
|
|
// Because no call to Do returns until the one call to f returns, if f causes
|
|
// Do to be called, it will deadlock.
|
|
//
|
|
// If f panics, Do considers it to have returned; future calls of Do return
|
|
// without calling f.
|
|
func (o *OnceAgain) Do(f func()) {
|
|
// Note: Here is an incorrect implementation of Do:
|
|
//
|
|
// if atomic.CompareAndSwapUint32(&o.done, 0, 1) {
|
|
// f()
|
|
// }
|
|
//
|
|
// Do guarantees that when it returns, f has finished.
|
|
// This implementation would not implement that guarantee:
|
|
// given two simultaneous calls, the winner of the cas would
|
|
// call f, and the second would return immediately, without
|
|
// waiting for the first's call to f to complete.
|
|
// This is why the slow path falls back to a mutex, and why
|
|
// the atomic.StoreUint32 must be delayed until after f returns.
|
|
|
|
if atomic.LoadUint32(&o.done) == 0 {
|
|
// Outlined slow-path to allow inlining of the fast-path.
|
|
o.doSlow(f)
|
|
}
|
|
}
|
|
|
|
func (o *OnceAgain) doSlow(f func()) {
|
|
atomic.AddInt32(&o.waiters, 1)
|
|
defer func() {
|
|
if atomic.AddInt32(&o.waiters, -1) == 0 {
|
|
atomic.StoreUint32(&o.done, 0) // reset
|
|
}
|
|
}()
|
|
|
|
o.m.Lock()
|
|
defer o.m.Unlock()
|
|
if o.done == 0 {
|
|
defer atomic.StoreUint32(&o.done, 1)
|
|
f()
|
|
}
|
|
}
|