perf(chatd): fix six scale bottlenecks identified by benchmarking (#22957)

## Summary

Scale-tested the `chatd` package with mock-based benchmarks to identify
performance bottlenecks. This PR fixes 6 of the 8 identified issues,
ranked by severity.

## Changes

### 1. Parallel tool execution (HIGH) — `chatloop.go`
`executeTools` ran tool calls sequentially. Now dispatches all calls
concurrently via goroutines with `sync.WaitGroup`. Results are
pre-allocated by index (no mutex needed). `onResult` callbacks fire as
each tool completes.

### 2. Pubsub-backed subagent await (HIGH) — `subagent.go`
`awaitSubagentCompletion` polled the DB every 200ms. Now subscribes to
the child chat's `ChatStreamNotifyChannel` via pubsub for near-instant
notifications. Fallback poll reduced to 5s. Falls back to 200ms only
when `pubsub == nil` (single-instance / in-memory).

### 3. Per-chat stream locking (MEDIUM) — `chatd.go`
Replaced single global `streamMu` + `map[uuid.UUID]*chatStreamState`
with `sync.Map` where each `chatStreamState` has its own `sync.Mutex`.
Zero cross-chat contention.

### 4. Batch chat acquisition (MEDIUM) — `chatd.go`
`processOnce` acquired 1 chat per tick. Now loops up to
`maxChatsPerAcquire = 10` per tick, avoiding idle time when many chats
are pending.

### 5. Reduced heartbeat frequency (LOW-MEDIUM) — `chatd.go`
`chatHeartbeatInterval` changed from 30s to 60s. Safe given the 5-minute
`DefaultInFlightChatStaleAfter`.

### 6. O(depth) descendant check (LOW) — `subagent.go`
Replaced top-down BFS (`O(total_descendants)` queries) with bottom-up
parent-chain walk (`O(depth)` queries). Includes cycle protection.

## Not addressed (intentionally)
- Message serialization overhead
- Buffer eviction (`buffer[1:]` pattern)
This commit is contained in:
Kyle Carberry
2026-03-11 14:00:08 -04:00
committed by GitHub
parent e5c19d0af4
commit 1f37df4db3
10 changed files with 304 additions and 223 deletions
+4 -4
View File
@@ -1512,13 +1512,13 @@ func (q *querier) authorizeProvisionerJob(ctx context.Context, job database.Prov
return nil
}
func (q *querier) AcquireChat(ctx context.Context, arg database.AcquireChatParams) (database.Chat, error) {
// AcquireChat is a system-level operation used by the chat processor.
func (q *querier) AcquireChats(ctx context.Context, arg database.AcquireChatsParams) ([]database.Chat, error) {
// AcquireChats is a system-level operation used by the chat processor.
// Authorization is done at the system level, not per-user.
if err := q.authorizeContext(ctx, policy.ActionUpdate, rbac.ResourceChat); err != nil {
return database.Chat{}, err
return nil, err
}
return q.db.AcquireChat(ctx, arg)
return q.db.AcquireChats(ctx, arg)
}
func (q *querier) AcquireLock(ctx context.Context, id int64) error {
+5 -4
View File
@@ -373,14 +373,15 @@ func (s *MethodTestSuite) TestConnectionLogs() {
}
func (s *MethodTestSuite) TestChats() {
s.Run("AcquireChat", s.Mocked(func(dbm *dbmock.MockStore, faker *gofakeit.Faker, check *expects) {
arg := database.AcquireChatParams{
s.Run("AcquireChats", s.Mocked(func(dbm *dbmock.MockStore, faker *gofakeit.Faker, check *expects) {
arg := database.AcquireChatsParams{
StartedAt: dbtime.Now(),
WorkerID: uuid.New(),
NumChats: 1,
}
chat := testutil.Fake(s.T(), faker, database.Chat{})
dbm.EXPECT().AcquireChat(gomock.Any(), arg).Return(chat, nil).AnyTimes()
check.Args(arg).Asserts(rbac.ResourceChat, policy.ActionUpdate).Returns(chat)
dbm.EXPECT().AcquireChats(gomock.Any(), arg).Return([]database.Chat{chat}, nil).AnyTimes()
check.Args(arg).Asserts(rbac.ResourceChat, policy.ActionUpdate).Returns([]database.Chat{chat})
}))
s.Run("DeleteAllChatQueuedMessages", s.Mocked(func(dbm *dbmock.MockStore, faker *gofakeit.Faker, check *expects) {
chat := testutil.Fake(s.T(), faker, database.Chat{})