gotosocial/vendor/github.com/godbus/dbus/v5/sequential_handler.go
Daniele Sluijters acc333c40b
[feature] Inherit resource limits from cgroups (#1336)
When GTS is running in a container runtime which has configured CPU or
memory limits or under an init system that uses cgroups to impose CPU
and memory limits the values the Go runtime sees for GOMAXPROCS and
GOMEMLIMIT are still based on the host resources, not the cgroup.

At least for the throttling middlewares which use GOMAXPROCS to
configure their queue size, this can result in GTS running with values
too big compared to the resources that will actuall be available to it.

This introduces 2 dependencies which can pick up resource contraints
from the current cgroup and tune the Go runtime accordingly. This should
result in the different queues being appropriately sized and in general
more predictable performance. These dependencies are a no-op on
non-Linux systems or if running in a cgroup that doesn't set a limit on
CPU or memory.

The automatic tuning of GOMEMLIMIT can be disabled by either explicitly
setting GOMEMLIMIT yourself or by setting AUTOMEMLIMIT=off. The
automatic tuning of GOMAXPROCS can similarly be counteracted by setting
GOMAXPROCS yourself.
2023-01-17 20:59:04 +00:00

126 lines
2.6 KiB
Go

package dbus
import (
"sync"
)
// NewSequentialSignalHandler returns an instance of a new
// signal handler that guarantees sequential processing of signals. It is a
// guarantee of this signal handler that signals will be written to
// channels in the order they are received on the DBus connection.
func NewSequentialSignalHandler() SignalHandler {
return &sequentialSignalHandler{}
}
type sequentialSignalHandler struct {
mu sync.RWMutex
closed bool
signals []*sequentialSignalChannelData
}
func (sh *sequentialSignalHandler) DeliverSignal(intf, name string, signal *Signal) {
sh.mu.RLock()
defer sh.mu.RUnlock()
if sh.closed {
return
}
for _, scd := range sh.signals {
scd.deliver(signal)
}
}
func (sh *sequentialSignalHandler) Terminate() {
sh.mu.Lock()
defer sh.mu.Unlock()
if sh.closed {
return
}
for _, scd := range sh.signals {
scd.close()
close(scd.ch)
}
sh.closed = true
sh.signals = nil
}
func (sh *sequentialSignalHandler) AddSignal(ch chan<- *Signal) {
sh.mu.Lock()
defer sh.mu.Unlock()
if sh.closed {
return
}
sh.signals = append(sh.signals, newSequentialSignalChannelData(ch))
}
func (sh *sequentialSignalHandler) RemoveSignal(ch chan<- *Signal) {
sh.mu.Lock()
defer sh.mu.Unlock()
if sh.closed {
return
}
for i := len(sh.signals) - 1; i >= 0; i-- {
if ch == sh.signals[i].ch {
sh.signals[i].close()
copy(sh.signals[i:], sh.signals[i+1:])
sh.signals[len(sh.signals)-1] = nil
sh.signals = sh.signals[:len(sh.signals)-1]
}
}
}
type sequentialSignalChannelData struct {
ch chan<- *Signal
in chan *Signal
done chan struct{}
}
func newSequentialSignalChannelData(ch chan<- *Signal) *sequentialSignalChannelData {
scd := &sequentialSignalChannelData{
ch: ch,
in: make(chan *Signal),
done: make(chan struct{}),
}
go scd.bufferSignals()
return scd
}
func (scd *sequentialSignalChannelData) bufferSignals() {
defer close(scd.done)
// Ensure that signals are delivered to scd.ch in the same
// order they are received from scd.in.
var queue []*Signal
for {
if len(queue) == 0 {
signal, ok := <- scd.in
if !ok {
return
}
queue = append(queue, signal)
}
select {
case scd.ch <- queue[0]:
copy(queue, queue[1:])
queue[len(queue)-1] = nil
queue = queue[:len(queue)-1]
case signal, ok := <-scd.in:
if !ok {
return
}
queue = append(queue, signal)
}
}
}
func (scd *sequentialSignalChannelData) deliver(signal *Signal) {
scd.in <- signal
}
func (scd *sequentialSignalChannelData) close() {
close(scd.in)
// Ensure that bufferSignals() has exited and won't attempt
// any future sends on scd.ch
<-scd.done
}