[release-15.0] fix data race in join engine primitive olap streaming …

…mode execution (#14012) (#14014) Signed-off-by: Harshit Gangal <[email protected]> Co-authored-by: vitess-bot[bot] <108069721+vitess-bot[bot]@users.noreply.github.com> Co-authored-by: Harshit Gangal <[email protected]>
vitessio · Sep 19, 2023 · d777c3a · d777c3a
1 parent 69d003a
commit d777c3a
Show file tree

Hide file tree

Showing 2 changed files with 45 additions and 10 deletions.
diff --git a/go/vt/vtgate/engine/join.go b/go/vt/vtgate/engine/join.go
@@ -22,6 +22,7 @@ import (
 	"strings"
 
 	"vitess.io/vitess/go/sqltypes"
+	"vitess.io/vitess/go/sync2"
 	querypb "vitess.io/vitess/go/vt/proto/query"
 )
 
@@ -95,34 +96,36 @@ func (jn *Join) TryExecute(ctx context.Context, vcursor VCursor, bindVars map[st
 
 // TryStreamExecute performs a streaming exec.
 func (jn *Join) TryStreamExecute(ctx context.Context, vcursor VCursor, bindVars map[string]*querypb.BindVariable, wantfields bool, callback func(*sqltypes.Result) error) error {
-	joinVars := make(map[string]*querypb.BindVariable)
-	err := vcursor.StreamExecutePrimitive(ctx, jn.Left, bindVars, wantfields, func(lresult *sqltypes.Result) error {
+	var fieldNeeded sync2.AtomicBool
+	fieldNeeded.Set(wantfields)
+	err := vcursor.StreamExecutePrimitive(ctx, jn.Left, bindVars, fieldNeeded.Get(), func(lresult *sqltypes.Result) error {
+		joinVars := make(map[string]*querypb.BindVariable)
 		for _, lrow := range lresult.Rows {
 			for k, col := range jn.Vars {
 				joinVars[k] = sqltypes.ValueBindVariable(lrow[col])
 			}
-			rowSent := false
-			err := vcursor.StreamExecutePrimitive(ctx, jn.Right, combineVars(bindVars, joinVars), wantfields, func(rresult *sqltypes.Result) error {
+			var rowSent sync2.AtomicBool
+			err := vcursor.StreamExecutePrimitive(ctx, jn.Right, combineVars(bindVars, joinVars), fieldNeeded.Get(), func(rresult *sqltypes.Result) error {
 				result := &sqltypes.Result{}
-				if wantfields {
+				if fieldNeeded.Get() {
 					// This code is currently unreachable because the first result
 					// will always be just the field info, which will cause the outer
 					// wantfields code path to be executed. But this may change in the future.
-					wantfields = false
+					fieldNeeded.Set(false)
 					result.Fields = joinFields(lresult.Fields, rresult.Fields, jn.Cols)
 				}
 				for _, rrow := range rresult.Rows {
 					result.Rows = append(result.Rows, joinRows(lrow, rrow, jn.Cols))
 				}
 				if len(rresult.Rows) != 0 {
-					rowSent = true
+					rowSent.Set(true)
 				}
 				return callback(result)
 			})
 			if err != nil {
 				return err
 			}
-			if jn.Opcode == LeftJoin && !rowSent {
+			if jn.Opcode == LeftJoin && !rowSent.Get() {
 				result := &sqltypes.Result{}
 				result.Rows = [][]sqltypes.Value{joinRows(
 					lrow,
@@ -132,8 +135,8 @@ func (jn *Join) TryStreamExecute(ctx context.Context, vcursor VCursor, bindVars
 				return callback(result)
 			}
 		}
-		if wantfields {
-			wantfields = false
+		if fieldNeeded.Get() {
+			fieldNeeded.Set(false)
 			for k := range jn.Vars {
 				joinVars[k] = sqltypes.NullBindVariable
 			}

diff --git a/go/vt/vtgate/executor_select_test.go b/go/vt/vtgate/executor_select_test.go
@@ -3782,3 +3782,35 @@ func TestMain(m *testing.M) {
 	_flag.ParseFlagsForTest()
 	os.Exit(m.Run())
 }
+
+func TestStreamJoinQuery(t *testing.T) {
+	// Special setup: Don't use createExecutorEnv.
+	cell := "aa"
+	hc := discovery.NewFakeHealthCheck(nil)
+	u := createSandbox(KsTestUnsharded)
+	s := createSandbox(KsTestSharded)
+	s.VSchema = executorVSchema
+	u.VSchema = unshardedVSchema
+	serv := newSandboxForCells([]string{cell})
+	resolver := newTestResolver(hc, serv, cell)
+	shards := []string{"-20", "20-40", "40-60", "60-80", "80-a0", "a0-c0", "c0-e0", "e0-"}
+	for _, shard := range shards {
+		_ = hc.AddTestTablet(cell, shard, 1, "TestExecutor", shard, topodatapb.TabletType_PRIMARY, true, 1, nil)
+	}
+	executor := createExecutor(serv, cell, resolver)
+
+	sql := "select u.foo, u.apa, ue.bar, ue.apa from user u join user_extra ue on u.foo = ue.bar"
+	result, err := executorStream(executor, sql)
+	require.NoError(t, err)
+	wantResult := &sqltypes.Result{
+		Fields: append(sandboxconn.SingleRowResult.Fields, sandboxconn.SingleRowResult.Fields...),
+	}
+	wantRow := append(sandboxconn.StreamRowResult.Rows[0], sandboxconn.StreamRowResult.Rows[0]...)
+	for i := 0; i < 64; i++ {
+		wantResult.Rows = append(wantResult.Rows, wantRow)
+	}
+	require.Equal(t, len(wantResult.Rows), len(result.Rows))
+	for idx := 0; idx < 64; idx++ {
+		utils.MustMatch(t, wantResult.Rows[idx], result.Rows[idx], "mismatched on: ", strconv.Itoa(idx))
+	}
+}