statistics: fix wrong NDV in the global stats when to disable async-merge-global-stats

pkg/statistics/handle/globalstats/BUILD.bazel

@@ -45,7 +45,7 @@ go_test(
     ],
     embed = [":globalstats"],
     flaky = True,
-    shard_count = 26,
+    shard_count = 28,
     deps = [
         "//pkg/domain",
         "//pkg/parser/model",

pkg/statistics/handle/globalstats/global_stats.go

@@ -305,10 +305,15 @@ func blockingMergePartitionStats2GlobalStats(
 		}
 		// FMSketch use many memory, so we first deal with it and then destroy it.
 		// Merge FMSketch.
+		// NOTE: allFms maybe contain empty.
 		globalStats.Fms[i] = allFms[i][0]
 		for j := 1; j < len(allFms[i]); j++ {
-			globalStats.Fms[i].MergeFMSketch(allFms[i][j])
-			allFms[i][j].DestroyAndPutToPool()
+			if globalStats.Fms[i] == nil {
+				globalStats.Fms[i] = allFms[i][j]
+			} else {
+				globalStats.Fms[i].MergeFMSketch(allFms[i][j])
+				allFms[i][j].DestroyAndPutToPool()
+			}
 		}
 
 		// Update the global NDV.

pkg/statistics/handle/globalstats/global_stats_test.go

@@ -959,3 +959,75 @@ func TestGlobalStatsAndSQLBindingWithConcurrency(t *testing.T) {
 	tk.MustExec("set global tidb_merge_partition_stats_concurrency=2")
 	testGlobalStatsAndSQLBinding(tk)
 }
+
+func TestBlockCheckFMSketch(t *testing.T) {
+	store := testkit.CreateMockStore(t)
+	tk := testkit.NewTestKit(t, store)
+	tk.MustExec("use test")
+	tk.MustExec("set @@tidb_enable_async_merge_global_stats=OFF;")
+	defer func() {
+		tk.MustExec("set @@tidb_enable_async_merge_global_stats=ON;")
+	}()
+	checkFMSketch(tk)
+}
+
+func TestAsyncCheckFMSketch(t *testing.T) {
+	store := testkit.CreateMockStore(t)
+	tk := testkit.NewTestKit(t, store)
+	tk.MustExec("use test")
+	tk.MustExec("set @@tidb_enable_async_merge_global_stats=ON;")
+	checkFMSketch(tk)
+}
+
+func checkFMSketch(tk *testkit.TestKit) {
+	tk.MustExec(`CREATE TABLE employees  (id INT NOT NULL AUTO_INCREMENT PRIMARY KEY,fname VARCHAR(25) NOT NULL,lname VARCHAR(25) NOT NULL,store_id INT NOT NULL,department_id INT NOT NULL
+) PARTITION BY RANGE(id)  (
+    PARTITION p0 VALUES LESS THAN (5),
+    PARTITION p1 VALUES LESS THAN (10),
+    PARTITION p2 VALUES LESS THAN (15),
+    PARTITION p3 VALUES LESS THAN MAXVALUE
+);`)
+	tk.MustExec(`INSERT INTO employees(FNAME,LNAME,STORE_ID,DEPARTMENT_ID) VALUES
+    ('Bob', 'Taylor', 3, 2), ('Frank', 'Williams', 1, 2),
+    ('Ellen', 'Johnson', 3, 4), ('Jim', 'Smith', 2, 4),
+    ('Mary', 'Jones', 1, 1), ('Linda', 'Black', 2, 3),
+    ('Ed', 'Jones', 2, 1), ('June', 'Wilson', 3, 1),
+    ('Andy', 'Smith', 1, 3), ('Lou', 'Waters', 2, 4),
+    ('Jill', 'Stone', 1, 4), ('Roger', 'White', 3, 2),
+    ('Howard', 'Andrews', 1, 2), ('Fred', 'Goldberg', 3, 3),
+    ('Barbara', 'Brown', 2, 3), ('Alice', 'Rogers', 2, 2),
+    ('Mark', 'Morgan', 3, 3), ('Karen', 'Cole', 3, 2);`)
+	tk.MustExec("ANALYZE TABLE employees;")
+	tk.MustExec("select * from employees;")
+	tk.MustExec("alter table employees truncate   partition p0 ; ")
+	tk.MustExec("select * from employees;")
+	tk.MustExec("analyze table employees partition p3;")
+	tk.MustExec("select * from employees;")
+	tk.MustQuery("select distinct_count, null_count, tot_col_size from mysql.stats_histograms where is_index=0 order by table_id asc").Check(
+		testkit.Rows("14 0 14",
+			"14 0 76",
+			"13 0 92",
+			"3 0 14",
+			"4 0 14",
+			"4 0 4",
+			"4 0 20",
+			"4 0 30",
+			"3 0 4",
+			"2 0 4",
+			"5 0 5",
+			"5 0 24",
+			"4 0 31",
+			"3 0 5",
+			"2 0 5",
+			"5 0 5",
+			"5 0 27",
+			"5 0 36",
+			"3 0 5",
+			"3 0 5",
+			"4 0 4",
+			"4 0 25",
+			"4 0 25",
+			"2 0 4",
+			"2 0 4",
+		))
+}