mirror of
https://github.com/QuilibriumNetwork/ceremonyclient.git
synced 2026-02-21 10:27:26 +08:00
* wip: conversion of hotstuff from flow into Q-oriented model * bulk of tests * remaining non-integration tests * add integration test, adjust log interface, small tweaks * further adjustments, restore full pacemaker shape * add component lifecycle management+supervisor * further refinements * resolve timeout hanging * mostly finalized state for consensus * bulk of engine swap out * lifecycle-ify most types * wiring nearly complete, missing needed hooks for proposals * plugged in, vetting message validation paths * global consensus, plugged in and verified * app shard now wired in too * do not decode empty keys.yml (#456) * remove obsolete engine.maxFrames config parameter (#454) * default to Info log level unless debug is enabled (#453) * respect config's "logging" section params, remove obsolete single-file logging (#452) * Trivial code cleanup aiming to reduce Go compiler warnings (#451) * simplify range traversal * simplify channel read for single select case * delete rand.Seed() deprecated in Go 1.20 and no-op as of Go 1.24 * simplify range traversal * simplify channel read for single select case * remove redundant type from array * simplify range traversal * simplify channel read for single select case * RC slate * finalize 2.1.0.5 * Update comments in StrictMonotonicCounter Fix comment formatting and clarify description. --------- Co-authored-by: Black Swan <3999712+blacks1ne@users.noreply.github.com>
150 lines
5.2 KiB
Go
150 lines
5.2 KiB
Go
package votecollector
|
|
|
|
import (
|
|
"errors"
|
|
"sync"
|
|
|
|
"source.quilibrium.com/quilibrium/monorepo/consensus"
|
|
"source.quilibrium.com/quilibrium/monorepo/consensus/models"
|
|
)
|
|
|
|
var (
|
|
// RepeatedVoteErr is emitted, when we receive a vote for the same state
|
|
// from the same voter multiple times. This error does _not_ indicate
|
|
// equivocation.
|
|
RepeatedVoteErr = errors.New("duplicated vote")
|
|
)
|
|
|
|
// voteContainer container stores the vote and in index representing
|
|
// the order in which the votes were received
|
|
type voteContainer[VoteT models.Unique] struct {
|
|
Vote *VoteT
|
|
index int
|
|
}
|
|
|
|
// VotesCache maintains a _concurrency safe_ cache of votes for one particular
|
|
// rank. The cache memorizes the order in which the votes were received. Votes
|
|
// are de-duplicated based on the following rules:
|
|
// - Vor each voter (i.e. SignerID), we store the _first_ vote v0.
|
|
// - For any subsequent vote v, we check whether v.Identifier == v0.Identifier.
|
|
// If this is the case, we consider the vote a duplicate and drop it.
|
|
// If v and v0 have different Identifiers, the voter is equivocating and
|
|
// we return a models.DoubleVoteError
|
|
type VotesCache[VoteT models.Unique] struct {
|
|
lock sync.RWMutex
|
|
rank uint64
|
|
votes map[models.Identity]voteContainer[VoteT] // signerID -> first vote
|
|
voteConsumers []consensus.VoteConsumer[VoteT]
|
|
}
|
|
|
|
// NewVotesCache instantiates a VotesCache for the given rank
|
|
func NewVotesCache[VoteT models.Unique](rank uint64) *VotesCache[VoteT] {
|
|
return &VotesCache[VoteT]{
|
|
rank: rank,
|
|
votes: make(map[models.Identity]voteContainer[VoteT]),
|
|
}
|
|
}
|
|
|
|
func (vc *VotesCache[VoteT]) Rank() uint64 { return vc.rank }
|
|
|
|
// AddVote stores a vote in the cache. The following errors are expected during
|
|
// normal operations:
|
|
// - nil: if the vote was successfully added
|
|
// - models.DoubleVoteError is returned if the voter is equivocating
|
|
// (i.e. voting in the same rank for different states).
|
|
// - RepeatedVoteErr is returned when adding a vote for the same state from
|
|
// the same voter multiple times.
|
|
// - IncompatibleRankErr is returned if the vote is for a different rank.
|
|
//
|
|
// When AddVote returns an error, the vote is _not_ stored.
|
|
func (vc *VotesCache[VoteT]) AddVote(vote *VoteT) error {
|
|
if (*vote).GetRank() != vc.rank {
|
|
return VoteForIncompatibleRankError
|
|
}
|
|
vc.lock.Lock()
|
|
defer vc.lock.Unlock()
|
|
|
|
// De-duplicated votes based on the following rules:
|
|
// * Vor each voter (i.e. SignerID), we store the _first_ vote v0.
|
|
// * For any subsequent vote v, we check whether
|
|
// v.Identifier == v0.Identifier.
|
|
// If this is the case, we consider the vote a duplicate and drop it.
|
|
// If v and v0 have different Identifiers, the voter is equivocating and
|
|
// we return a models.DoubleVoteError
|
|
firstVote, exists := vc.votes[(*vote).Identity()]
|
|
if exists {
|
|
if (*firstVote.Vote).Source() != (*vote).Source() {
|
|
return models.NewDoubleVoteErrorf(
|
|
firstVote.Vote,
|
|
vote,
|
|
"detected vote equivocation at rank: %d",
|
|
vc.rank,
|
|
)
|
|
}
|
|
return RepeatedVoteErr
|
|
}
|
|
|
|
// previously unknown vote: (1) store and (2) forward to consumers
|
|
vc.votes[(*vote).Identity()] = voteContainer[VoteT]{vote, len(vc.votes)}
|
|
for _, consumer := range vc.voteConsumers {
|
|
consumer(vote)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// GetVote returns the stored vote for the given `signerID`. Returns:
|
|
// - (vote, true) if a vote from signerID is known
|
|
// - (false, nil) no vote from signerID is known
|
|
func (vc *VotesCache[VoteT]) GetVote(signerID models.Identity) (*VoteT, bool) {
|
|
vc.lock.RLock()
|
|
container, exists := vc.votes[signerID] // if signerID is unknown, its `Vote` pointer is nil
|
|
vc.lock.RUnlock()
|
|
return container.Vote, exists
|
|
}
|
|
|
|
// Size returns the number of cached votes
|
|
func (vc *VotesCache[VoteT]) Size() int {
|
|
vc.lock.RLock()
|
|
s := len(vc.votes)
|
|
vc.lock.RUnlock()
|
|
return s
|
|
}
|
|
|
|
// RegisterVoteConsumer registers a VoteConsumer. Upon registration, the cache
|
|
// feeds all cached votes into the consumer in the order they arrived.
|
|
// CAUTION: a consumer _must_ be non-stateing and consume the votes without
|
|
// noteworthy delay. Otherwise, consensus speed is impacted.
|
|
//
|
|
// Expected usage patter: During happy-path operations, the state arrives in a
|
|
// timely manner. Hence, we expect that only a few votes are cached when a
|
|
// consumer is registered. For the purpose of forensics, we might register a
|
|
// consumer later, when already lots of votes are cached. However, this should
|
|
// be a rare occurrence (we except moderate performance overhead in this case).
|
|
func (vc *VotesCache[VoteT]) RegisterVoteConsumer(
|
|
consumer consensus.VoteConsumer[VoteT],
|
|
) {
|
|
vc.lock.Lock()
|
|
defer vc.lock.Unlock()
|
|
|
|
vc.voteConsumers = append(vc.voteConsumers, consumer)
|
|
for _, vote := range vc.all() { // feed the consumer with the cached votes
|
|
consumer(vote) // non-stateing per API contract
|
|
}
|
|
}
|
|
|
|
// All returns all currently cached votes. Concurrency safe.
|
|
func (vc *VotesCache[VoteT]) All() []*VoteT {
|
|
vc.lock.Lock()
|
|
defer vc.lock.Unlock()
|
|
return vc.all()
|
|
}
|
|
|
|
// all returns all currently cached votes. NOT concurrency safe
|
|
func (vc *VotesCache[VoteT]) all() []*VoteT {
|
|
orderedVotes := make([]*VoteT, len(vc.votes))
|
|
for _, v := range vc.votes {
|
|
orderedVotes[v.index] = v.Vote
|
|
}
|
|
return orderedVotes
|
|
}
|