core/bloombits: use single channel for shutdown (#20878)
This replaces the two-stage shutdown scheme with the one we use almost everywhere else: a single quit channel signalling termination. Co-authored-by: Felix Lange <fjl@twurst.com>
This commit is contained in:
parent
abf2d7d74f
commit
9e04c5ec83
@ -155,7 +155,6 @@ func (m *Matcher) Start(ctx context.Context, begin, end uint64, results chan uin
|
||||
session := &MatcherSession{
|
||||
matcher: m,
|
||||
quit: make(chan struct{}),
|
||||
kill: make(chan struct{}),
|
||||
ctx: ctx,
|
||||
}
|
||||
for _, scheduler := range m.schedulers {
|
||||
@ -386,10 +385,8 @@ func (m *Matcher) distributor(dist chan *request, session *MatcherSession) {
|
||||
requests = make(map[uint][]uint64) // Per-bit list of section requests, ordered by section number
|
||||
unallocs = make(map[uint]struct{}) // Bits with pending requests but not allocated to any retriever
|
||||
retrievers chan chan uint // Waiting retrievers (toggled to nil if unallocs is empty)
|
||||
)
|
||||
var (
|
||||
allocs int // Number of active allocations to handle graceful shutdown requests
|
||||
shutdown = session.quit // Shutdown request channel, will gracefully wait for pending requests
|
||||
allocs int // Number of active allocations to handle graceful shutdown requests
|
||||
shutdown = session.quit // Shutdown request channel, will gracefully wait for pending requests
|
||||
)
|
||||
|
||||
// assign is a helper method fo try to assign a pending bit an actively
|
||||
@ -409,15 +406,12 @@ func (m *Matcher) distributor(dist chan *request, session *MatcherSession) {
|
||||
for {
|
||||
select {
|
||||
case <-shutdown:
|
||||
// Graceful shutdown requested, wait until all pending requests are honoured
|
||||
// Shutdown requested. No more retrievers can be allocated,
|
||||
// but we still need to wait until all pending requests have returned.
|
||||
shutdown = nil
|
||||
if allocs == 0 {
|
||||
return
|
||||
}
|
||||
shutdown = nil
|
||||
|
||||
case <-session.kill:
|
||||
// Pending requests not honoured in time, hard terminate
|
||||
return
|
||||
|
||||
case req := <-dist:
|
||||
// New retrieval request arrived to be distributed to some fetcher process
|
||||
@ -499,8 +493,9 @@ func (m *Matcher) distributor(dist chan *request, session *MatcherSession) {
|
||||
assign(result.Bit)
|
||||
}
|
||||
}
|
||||
// If we're in the process of shutting down, terminate
|
||||
if allocs == 0 && shutdown == nil {
|
||||
|
||||
// End the session when all pending deliveries have arrived.
|
||||
if shutdown == nil && allocs == 0 {
|
||||
return
|
||||
}
|
||||
}
|
||||
@ -514,7 +509,6 @@ type MatcherSession struct {
|
||||
|
||||
closer sync.Once // Sync object to ensure we only ever close once
|
||||
quit chan struct{} // Quit channel to request pipeline termination
|
||||
kill chan struct{} // Term channel to signal non-graceful forced shutdown
|
||||
|
||||
ctx context.Context // Context used by the light client to abort filtering
|
||||
err atomic.Value // Global error to track retrieval failures deep in the chain
|
||||
@ -529,7 +523,6 @@ func (s *MatcherSession) Close() {
|
||||
s.closer.Do(func() {
|
||||
// Signal termination and wait for all goroutines to tear down
|
||||
close(s.quit)
|
||||
time.AfterFunc(time.Second, func() { close(s.kill) })
|
||||
s.pend.Wait()
|
||||
})
|
||||
}
|
||||
@ -542,10 +535,10 @@ func (s *MatcherSession) Error() error {
|
||||
return nil
|
||||
}
|
||||
|
||||
// AllocateRetrieval assigns a bloom bit index to a client process that can either
|
||||
// allocateRetrieval assigns a bloom bit index to a client process that can either
|
||||
// immediately request and fetch the section contents assigned to this bit or wait
|
||||
// a little while for more sections to be requested.
|
||||
func (s *MatcherSession) AllocateRetrieval() (uint, bool) {
|
||||
func (s *MatcherSession) allocateRetrieval() (uint, bool) {
|
||||
fetcher := make(chan uint)
|
||||
|
||||
select {
|
||||
@ -557,9 +550,9 @@ func (s *MatcherSession) AllocateRetrieval() (uint, bool) {
|
||||
}
|
||||
}
|
||||
|
||||
// PendingSections returns the number of pending section retrievals belonging to
|
||||
// pendingSections returns the number of pending section retrievals belonging to
|
||||
// the given bloom bit index.
|
||||
func (s *MatcherSession) PendingSections(bit uint) int {
|
||||
func (s *MatcherSession) pendingSections(bit uint) int {
|
||||
fetcher := make(chan uint)
|
||||
|
||||
select {
|
||||
@ -571,9 +564,9 @@ func (s *MatcherSession) PendingSections(bit uint) int {
|
||||
}
|
||||
}
|
||||
|
||||
// AllocateSections assigns all or part of an already allocated bit-task queue
|
||||
// allocateSections assigns all or part of an already allocated bit-task queue
|
||||
// to the requesting process.
|
||||
func (s *MatcherSession) AllocateSections(bit uint, count int) []uint64 {
|
||||
func (s *MatcherSession) allocateSections(bit uint, count int) []uint64 {
|
||||
fetcher := make(chan *Retrieval)
|
||||
|
||||
select {
|
||||
@ -589,14 +582,10 @@ func (s *MatcherSession) AllocateSections(bit uint, count int) []uint64 {
|
||||
}
|
||||
}
|
||||
|
||||
// DeliverSections delivers a batch of section bit-vectors for a specific bloom
|
||||
// deliverSections delivers a batch of section bit-vectors for a specific bloom
|
||||
// bit index to be injected into the processing pipeline.
|
||||
func (s *MatcherSession) DeliverSections(bit uint, sections []uint64, bitsets [][]byte) {
|
||||
select {
|
||||
case <-s.kill:
|
||||
return
|
||||
case s.matcher.deliveries <- &Retrieval{Bit: bit, Sections: sections, Bitsets: bitsets}:
|
||||
}
|
||||
func (s *MatcherSession) deliverSections(bit uint, sections []uint64, bitsets [][]byte) {
|
||||
s.matcher.deliveries <- &Retrieval{Bit: bit, Sections: sections, Bitsets: bitsets}
|
||||
}
|
||||
|
||||
// Multiplex polls the matcher session for retrieval tasks and multiplexes it into
|
||||
@ -608,17 +597,17 @@ func (s *MatcherSession) DeliverSections(bit uint, sections []uint64, bitsets []
|
||||
func (s *MatcherSession) Multiplex(batch int, wait time.Duration, mux chan chan *Retrieval) {
|
||||
for {
|
||||
// Allocate a new bloom bit index to retrieve data for, stopping when done
|
||||
bit, ok := s.AllocateRetrieval()
|
||||
bit, ok := s.allocateRetrieval()
|
||||
if !ok {
|
||||
return
|
||||
}
|
||||
// Bit allocated, throttle a bit if we're below our batch limit
|
||||
if s.PendingSections(bit) < batch {
|
||||
if s.pendingSections(bit) < batch {
|
||||
select {
|
||||
case <-s.quit:
|
||||
// Session terminating, we can't meaningfully service, abort
|
||||
s.AllocateSections(bit, 0)
|
||||
s.DeliverSections(bit, []uint64{}, [][]byte{})
|
||||
s.allocateSections(bit, 0)
|
||||
s.deliverSections(bit, []uint64{}, [][]byte{})
|
||||
return
|
||||
|
||||
case <-time.After(wait):
|
||||
@ -626,13 +615,13 @@ func (s *MatcherSession) Multiplex(batch int, wait time.Duration, mux chan chan
|
||||
}
|
||||
}
|
||||
// Allocate as much as we can handle and request servicing
|
||||
sections := s.AllocateSections(bit, batch)
|
||||
sections := s.allocateSections(bit, batch)
|
||||
request := make(chan *Retrieval)
|
||||
|
||||
select {
|
||||
case <-s.quit:
|
||||
// Session terminating, we can't meaningfully service, abort
|
||||
s.DeliverSections(bit, sections, make([][]byte, len(sections)))
|
||||
s.deliverSections(bit, sections, make([][]byte, len(sections)))
|
||||
return
|
||||
|
||||
case mux <- request:
|
||||
@ -644,7 +633,7 @@ func (s *MatcherSession) Multiplex(batch int, wait time.Duration, mux chan chan
|
||||
s.err.Store(result.Error)
|
||||
s.Close()
|
||||
}
|
||||
s.DeliverSections(result.Bit, result.Sections, result.Bitsets)
|
||||
s.deliverSections(result.Bit, result.Sections, result.Bitsets)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
Loading…
Reference in New Issue
Block a user