commit 8545447decb54e4f8c54949b809aa2238892f261
Author: Zsolt Parragi <zsolt.parragi@percona.com>
Date:   Thu May 24 10:01:09 2018 +0200

    Fix bug 90351 / PS-3951 (Fix decreasing status counters)
    
    Status counters were cleaned in THD::release_resources, before an
    optional thread cleanup.  The cleanup could possibly perform a
    rollback, or other counter-increasing operations.  In this case, these
    additions to the thread local counters were lost during the
    destruction of the thread, but were visible for a short time, before
    the thread was removed from the global thread list.
    
    This change avoids this issue by moving the global counter update to
    the end of the release_resources method, ensuring that possible
    counter-changing operation happen before it.

diff --git a/mysql-test/r/bug90351.result b/mysql-test/r/bug90351.result
new file mode 100644
index 00000000000..0546793fd97
--- /dev/null
+++ b/mysql-test/r/bug90351.result
@@ -0,0 +1,6 @@
+CREATE TABLE t1 (i INTEGER);
+INSERT INTO t1 VALUES (1);
+SET AUTOCOMMIT=0;
+START TRANSACTION;
+DELETE FROM t1;
+DROP TABLE t1;
diff --git a/mysql-test/t/bug90351.test b/mysql-test/t/bug90351.test
new file mode 100644
index 00000000000..42a9d1d8fbd
--- /dev/null
+++ b/mysql-test/t/bug90351.test
@@ -0,0 +1,25 @@
+#
+# Bug 90351: status counters decrease on thread destruction
+#
+--source include/count_sessions.inc
+
+connect (con1,localhost,root,,);
+
+connection con1;
+CREATE TABLE t1 (i INTEGER);
+INSERT INTO t1 VALUES (1);
+let $con1_id= `SELECT connection_id()`;
+SET AUTOCOMMIT=0;
+START TRANSACTION;
+DELETE FROM t1;
+
+connection default;
+--disable_query_log
+eval KILL $con1_id;
+--enable_query_log
+
+connection default;
+DROP TABLE t1;
+
+disconnect con1;
+--source include/wait_until_count_sessions.inc
diff --git a/sql/handler.cc b/sql/handler.cc
index 0e54fba63ed..0f0fec26672 100644
--- a/sql/handler.cc
+++ b/sql/handler.cc
@@ -1347,6 +1347,7 @@ int ha_prepare(THD *thd) {
 
     while (ha_info) {
       handlerton *ht = ha_info->ht();
+      DBUG_ASSERT(!thd->status_var_aggregated);
       thd->status_var.ha_prepare_count++;
       if (ht->prepare) {
         DBUG_EXECUTE_IF("simulate_xa_failure_prepare", {
@@ -1809,6 +1810,7 @@ int ha_commit_low(THD *thd, bool all, bool run_after_commit) {
                  my_strerror(errbuf, MYSQL_ERRMSG_SIZE, err));
         error = 1;
       }
+      DBUG_ASSERT(!thd->status_var_aggregated);
       thd->status_var.ha_commit_count++;
       ha_info_next = ha_info->next();
       if (restore_backup_ha_data) reattach_engine_ha_data_to_thd(thd, ht);
@@ -1869,6 +1871,7 @@ int ha_rollback_low(THD *thd, bool all) {
                  my_strerror(errbuf, MYSQL_ERRMSG_SIZE, err));
         error = 1;
       }
+      DBUG_ASSERT(!thd->status_var_aggregated);
       thd->status_var.ha_rollback_count++;
       ha_info_next = ha_info->next();
       if (restore_backup_ha_data) reattach_engine_ha_data_to_thd(thd, ht);
@@ -2031,6 +2034,7 @@ int ha_commit_attachable(THD *thd) {
         DBUG_ASSERT(false);
         error = 1;
       }
+      DBUG_ASSERT(!thd->status_var_aggregated);
       thd->status_var.ha_commit_count++;
       ha_info_next = ha_info->next();
 
@@ -2119,6 +2123,7 @@ int ha_rollback_to_savepoint(THD *thd, SAVEPOINT *sv) {
                my_strerror(errbuf, MYSQL_ERRMSG_SIZE, err));
       error = 1;
     }
+    DBUG_ASSERT(!thd->status_var_aggregated);
     thd->status_var.ha_savepoint_rollback_count++;
     if (ht->prepare == 0) trn_ctx->set_no_2pc(trx_scope, true);
   }
@@ -2137,6 +2142,7 @@ int ha_rollback_to_savepoint(THD *thd, SAVEPOINT *sv) {
                my_strerror(errbuf, MYSQL_ERRMSG_SIZE, err));
       error = 1;
     }
+    DBUG_ASSERT(!thd->status_var_aggregated);
     thd->status_var.ha_rollback_count++;
     ha_info_next = ha_info->next();
     ha_info->reset(); /* keep it conveniently zero-filled */
@@ -2175,6 +2181,7 @@ int ha_prepare_low(THD *thd, bool all) {
                  my_strerror(errbuf, MYSQL_ERRMSG_SIZE, err));
         error = 1;
       }
+      DBUG_ASSERT(!thd->status_var_aggregated);
       thd->status_var.ha_prepare_count++;
     }
     DBUG_EXECUTE_IF("crash_commit_after_prepare", DBUG_SUICIDE(););
@@ -2215,6 +2222,7 @@ int ha_savepoint(THD *thd, SAVEPOINT *sv) {
                my_strerror(errbuf, MYSQL_ERRMSG_SIZE, err));
       error = 1;
     }
+    DBUG_ASSERT(!thd->status_var_aggregated);
     thd->status_var.ha_savepoint_count++;
   }
   /*
@@ -2529,6 +2537,7 @@ err:
 
 void handler::ha_statistic_increment(
     ulonglong System_status_var::*offset) const {
+  DBUG_ASSERT(!table->in_use->status_var_aggregated);
   (table->in_use->status_var.*offset)++;
 }
 
@@ -5300,7 +5309,10 @@ static int ha_discover(THD *thd, const char *db, const char *name,
                      &args))
     error = 0;
 
-  if (!error) thd->status_var.ha_discover_count++;
+  if (!error) {
+    DBUG_ASSERT(!thd->status_var_aggregated);
+    thd->status_var.ha_discover_count++;
+  }
   DBUG_RETURN(error);
 }
 
@@ -6134,7 +6146,10 @@ int DsMrr_impl::dsmrr_init(RANGE_SEQ_IF *seq_funcs, void *seq_init_param,
 
   is_mrr_assoc = !(mode & HA_MRR_NO_ASSOCIATION);
 
-  if (is_mrr_assoc) table->in_use->status_var.ha_multi_range_read_init_count++;
+  if (is_mrr_assoc) {
+    DBUG_ASSERT(!thd->status_var_aggregated);
+    table->in_use->status_var.ha_multi_range_read_init_count++;
+  }
 
   rowids_buf_end = buf->buffer_end;
   elem_size = h->ref_length + (int)is_mrr_assoc * sizeof(void *);
diff --git a/sql/rpl_master.cc b/sql/rpl_master.cc
index db9e1984221..bd26f6b8ed8 100644
--- a/sql/rpl_master.cc
+++ b/sql/rpl_master.cc
@@ -650,6 +650,7 @@ bool com_binlog_dump(THD *thd, char *packet, size_t packet_length) {
   const uchar *packet_position = (uchar *)packet;
   size_t packet_bytes_todo = packet_length;
 
+  DBUG_ASSERT(!thd->status_var_aggregated);
   thd->status_var.com_other++;
   thd->enable_slow_log = opt_log_slow_admin_statements;
   if (check_global_access(thd, REPL_SLAVE_ACL)) DBUG_RETURN(false);
@@ -700,6 +701,7 @@ bool com_binlog_dump_gtid(THD *thd, char *packet, size_t packet_length) {
       NULL /*no sid_lock because this is a completely local object*/);
   Gtid_set slave_gtid_executed(&sid_map);
 
+  DBUG_ASSERT(!thd->status_var_aggregated);
   thd->status_var.com_other++;
   thd->enable_slow_log = opt_log_slow_admin_statements;
   if (check_global_access(thd, REPL_SLAVE_ACL)) DBUG_RETURN(false);
diff --git a/sql/sql_class.cc b/sql/sql_class.cc
index 13ab695841c..d13f7b90a2d 100644
--- a/sql/sql_class.cc
+++ b/sql/sql_class.cc
@@ -969,18 +969,6 @@ void THD::release_resources() {
 
   Global_THD_manager::get_instance()->release_thread_id(m_thread_id);
 
-  mysql_mutex_lock(&LOCK_status);
-  add_to_status(&global_status_var, &status_var, false);
-  /*
-    Status queries after this point should not aggregate THD::status_var
-    since the values has been added to global_status_var.
-    The status values are not reset so that they can still be read
-    by performance schema.
-  */
-  status_var_aggregated = true;
-
-  mysql_mutex_unlock(&LOCK_status);
-
   /* Ensure that no one is using THD */
   mysql_mutex_lock(&LOCK_thd_data);
   mysql_mutex_lock(&LOCK_query_plan);
@@ -1028,6 +1016,18 @@ void THD::release_resources() {
 
   if (current_thd == this) restore_globals();
 
+  mysql_mutex_lock(&LOCK_status);
+  add_to_status(&global_status_var, &status_var, false);
+  /*
+    Status queries after this point should not aggregate THD::status_var
+    since the values has been added to global_status_var.
+    The status values are not reset so that they can still be read
+    by performance schema.
+  */
+  status_var_aggregated = true;
+
+  mysql_mutex_unlock(&LOCK_status);
+
   m_release_resources_done = true;
 }
 
@@ -1165,7 +1165,10 @@ void THD::awake(THD::killed_state state_to_set) {
   if (state_to_set != THD::NOT_KILLED) ha_kill_connection(this);
 
   if (state_to_set == THD::KILL_TIMEOUT)
+  {
+    DBUG_ASSERT(!status_var_aggregated);
     status_var.max_execution_time_exceeded++;
+  }
 
   /* Broadcast a condition to kick the target if it is waiting on it. */
   if (is_killable) {
@@ -2065,6 +2068,7 @@ void THD::inc_examined_row_count(ha_rows count) {
 }
 
 void THD::inc_status_created_tmp_disk_tables() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.created_tmp_disk_tables++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_created_tmp_disk_tables)(m_statement_psi, 1);
@@ -2072,6 +2076,7 @@ void THD::inc_status_created_tmp_disk_tables() {
 }
 
 void THD::inc_status_created_tmp_tables() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.created_tmp_tables++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_created_tmp_tables)(m_statement_psi, 1);
@@ -2079,6 +2084,7 @@ void THD::inc_status_created_tmp_tables() {
 }
 
 void THD::inc_status_select_full_join() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.select_full_join_count++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_select_full_join)(m_statement_psi, 1);
@@ -2086,6 +2092,7 @@ void THD::inc_status_select_full_join() {
 }
 
 void THD::inc_status_select_full_range_join() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.select_full_range_join_count++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_select_full_range_join)(m_statement_psi, 1);
@@ -2093,6 +2100,7 @@ void THD::inc_status_select_full_range_join() {
 }
 
 void THD::inc_status_select_range() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.select_range_count++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_select_range)(m_statement_psi, 1);
@@ -2100,6 +2108,7 @@ void THD::inc_status_select_range() {
 }
 
 void THD::inc_status_select_range_check() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.select_range_check_count++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_select_range_check)(m_statement_psi, 1);
@@ -2107,6 +2116,7 @@ void THD::inc_status_select_range_check() {
 }
 
 void THD::inc_status_select_scan() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.select_scan_count++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_select_scan)(m_statement_psi, 1);
@@ -2114,6 +2124,7 @@ void THD::inc_status_select_scan() {
 }
 
 void THD::inc_status_sort_merge_passes() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.filesort_merge_passes++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_sort_merge_passes)(m_statement_psi, 1);
@@ -2121,6 +2132,7 @@ void THD::inc_status_sort_merge_passes() {
 }
 
 void THD::inc_status_sort_range() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.filesort_range_count++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_sort_range)(m_statement_psi, 1);
@@ -2128,6 +2140,7 @@ void THD::inc_status_sort_range() {
 }
 
 void THD::inc_status_sort_rows(ha_rows count) {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.filesort_rows += count;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_sort_rows)
@@ -2136,6 +2149,7 @@ void THD::inc_status_sort_rows(ha_rows count) {
 }
 
 void THD::inc_status_sort_scan() {
+  DBUG_ASSERT(!status_var_aggregated);
   status_var.filesort_scan_count++;
 #ifdef HAVE_PSI_STATEMENT_INTERFACE
   PSI_STATEMENT_CALL(inc_statement_sort_scan)(m_statement_psi, 1);
diff --git a/sql/sql_class.h b/sql/sql_class.h
index d4ae7eac69c..a98e439f6f9 100644
--- a/sql/sql_class.h
+++ b/sql/sql_class.h
@@ -951,6 +951,7 @@ class THD : public MDL_context_owner,
     actually reports the previous query, not itself.
   */
   void save_current_query_costs() {
+    DBUG_ASSERT(!status_var_aggregated);
     status_var.last_query_cost = m_current_query_cost;
     status_var.last_query_partial_plans = m_current_query_partial_plans;
   }