MDEV-23475 InnoDB performance regression for write-heavy workloads

In commit fe39d02f51b96536dccca7ff89faf05e13548877 (MDEV-20638)
we removed some wake-up signaling of the master thread that should
have been there, to ensure a steady log checkpointing workload.

Common sense suggests that the commit omitted some necessary calls
to srv_inc_activity_count(). But, an attempt to add the call to
trx_flush_log_if_needed_low() as well as to reinstate the function
innobase_active_small() did not restore the performance for the
case where sync_binlog=1 is set.

Therefore, we will revert the entire commit in MariaDB Server 10.2.
In MariaDB Server 10.5, adding a srv_inc_activity_count() call to
trx_flush_log_if_needed_low() did restore the performance, so we
will not revert MDEV-20638 across all versions.

1 files changed, 6 insertions, 0 deletions

diff --git a/extra/mariabackup/xtrabackup.cc b/extra/mariabackup/xtrabackup.cc
index 9b3d9f9ea39..4845f51e6b2 100644
--- a/extra/mariabackup/xtrabackup.cc
+++ b/extra/mariabackup/xtrabackup.cc
@@ -254,6 +254,12 @@ my_bool innobase_locks_unsafe_for_binlog;
 my_bool innobase_rollback_on_timeout;
 my_bool innobase_create_status_file;
 
+/* The following counter is used to convey information to InnoDB
+about server activity: in selects it is not sensible to call
+srv_active_wake_master_thread after each fetch or search, we only do
+it every INNOBASE_WAKE_INTERVAL'th step. */
+
+#define INNOBASE_WAKE_INTERVAL	32
 ulong	innobase_active_counter	= 0;