MDEV-29515 innodb.deadlock_victim_race is unstable

The test is unstable because 'UPDATE t SET b = 100' latches a page and
waits for 'upd_cont' signal in lock_trx_handle_wait_enter sync point, then
purge requests RW_X_LATCH on the same page, and then 'SELECT * FROM t
WHERE a = 10 FOR UPDATE' requests RW_S_LATCH, waiting for RW_X_LATCH
requested by purge. 'UPDATE t SET b = 100' can't release page latch as
it waits for upd_cont signal, which must be emitted after 'SELECT * FROM
t WHERE a = 10 FOR UPDATE' acquired RW_S_LATCH. So we have a deadlock,
which is resolved by finishing the debug sync point wait by timeout, and
the 'UPDATE t SET b = 100' releases it's record locks rolling back the
transaction, and 'SELECT * FROM t WHERE a = 10 FOR UPDATE' is finished
successfully instead of finishing by lock wait timeout.

The fix is to forbid purging during the test by opening read view in a
separate connection before the first insert into the table.

Besides, 'lock_wait_end' syncpoint is not needed, as it enough to wait
the end of the SELECT execution to let the UPDATE to continue.

mysql-test/suite/innodb/r/deadlock_victim_race.result

@@ -1,3 +1,6 @@
+connect cancel_purge,localhost,root,,;
+START TRANSACTION WITH CONSISTENT SNAPSHOT;
+connection default;
 CREATE TABLE t (a int PRIMARY KEY, b int) engine = InnoDB;
 CREATE TABLE t2 (a int PRIMARY KEY) engine = InnoDB;
 INSERT INTO t VALUES (10, 10), (20, 20), (30, 30);
@@ -24,11 +27,11 @@ connection default;
 SET DEBUG_SYNC='now WAIT_FOR sel_locked';
 ROLLBACK;
 SET DEBUG_SYNC='now SIGNAL upd_cont';
-SET innodb_lock_wait_timeout=1;
 SET DEBUG_SYNC="now WAIT_FOR upd_locked";
-SET DEBUG_SYNC="lock_wait_end SIGNAL upd_cont";
+SET SESSION innodb_lock_wait_timeout=1;
 SELECT * FROM t WHERE a = 10 FOR UPDATE;
 ERROR HY000: Lock wait timeout exceeded; try restarting transaction
+SET DEBUG_SYNC="now SIGNAL upd_cont";
 connection con_3;
 a	b
 20	20
@@ -40,3 +43,4 @@ connection default;
 SET DEBUG_SYNC = 'RESET';
 DROP TABLE t;
 DROP TABLE t2;
+disconnect cancel_purge;

mysql-test/suite/innodb/t/deadlock_victim_race.test

@@ -2,6 +2,17 @@
 --source include/have_debug_sync.inc
 --source include/count_sessions.inc
 
+--connect(cancel_purge,localhost,root,,)
+# Purge can cause deadlock in the test, requesting page's RW_X_LATCH for trx
+# ids reseting, after trx 2 acqured RW_S_LATCH and suspended in debug sync point
+# lock_trx_handle_wait_enter, waiting for upd_cont signal, which must be
+# emitted after the last SELECT in this test. The last SELECT will hang waiting
+# for purge RW_X_LATCH releasing, and trx 2 will be rolled back by timeout.
+# The last SELECT will then be successfully executed instead of finishing by
+# lock wait timeout.
+START TRANSACTION WITH CONSISTENT SNAPSHOT;
+
+--connection default
 CREATE TABLE t (a int PRIMARY KEY, b int) engine = InnoDB;
 CREATE TABLE t2 (a int PRIMARY KEY) engine = InnoDB;
 
@@ -60,7 +71,6 @@ SET DEBUG_SYNC='now WAIT_FOR sel_locked';
 ROLLBACK;
 SET DEBUG_SYNC='now SIGNAL upd_cont';
 
-SET innodb_lock_wait_timeout=1;
 SET DEBUG_SYNC="now WAIT_FOR upd_locked";
 # Locking queue:
 # (10,10)             (20,20)                  (30,30)
@@ -72,11 +82,12 @@ SET DEBUG_SYNC="now WAIT_FOR upd_locked";
 # sequential read (with rr_sequential() read record function), and requested
 # lock on (30,30). But the deadlock has not been determined yet.
 
-SET DEBUG_SYNC="lock_wait_end SIGNAL upd_cont";
+SET SESSION innodb_lock_wait_timeout=1;
 --error ER_LOCK_WAIT_TIMEOUT
 # The deadlock will be determined in lock_wait() after lock wait timeout
 # expired.
 SELECT * FROM t WHERE a = 10 FOR UPDATE;
+SET DEBUG_SYNC="now SIGNAL upd_cont";
 
 --connection con_3
 --reap
@@ -98,5 +109,5 @@ SELECT * FROM t WHERE a = 10 FOR UPDATE;
 SET DEBUG_SYNC = 'RESET';
 DROP TABLE t;
 DROP TABLE t2;
-
+--disconnect cancel_purge
 --source include/wait_until_count_sessions.inc

storage/innobase/lock/lock0lock.cc

@@ -44,7 +44,6 @@ Created 5/7/1996 Heikki Tuuri
 #include "row0vers.h"
 #include "pars0pars.h"
 #include "srv0mon.h"
-#include "scope.h"
 
 #include <set>
 
@@ -1724,12 +1723,6 @@ dberr_t lock_wait(que_thr_t *thr)
   if (trx->mysql_thd)
     DEBUG_SYNC_C("lock_wait_start");
 
-  /* Create the sync point for any quit from the function. */
-  ut_d(SCOPE_EXIT([trx]() {
-    if (trx->mysql_thd)
-      DEBUG_SYNC_C("lock_wait_end");
-  }));
-
   /* InnoDB system transactions may use the global value of
   innodb_lock_wait_timeout, because trx->mysql_thd == NULL. */
   const ulong innodb_lock_wait_timeout= trx_lock_wait_timeout_get(trx);