SQL> recreate table s(id int, x int);
SQL> commit;
SQL> insert into s(id, x) select row_number()over(), 2 * row_number()over() from rdb$types rows 5;
SQL> commit;
SQL> select * from s;

          ID            X
============ ============
           1            2
           2            4
           3            6
           4            8
           5           10

SQL> recreate table t(id int primary key, x int);
SQL> commit;

SQL> set count on;
SQL> merge into t using s on s.id = t.id when not matched then insert values(s.id, s.x);
Records affected: 5
SQL> select * from t;

          ID            X
============ ============
           1            2
           2            4
           3            6
           4            8
           5           10

Records affected: 5
SQL> merge into t using s on s.id = t.id when matched then update set t.x = - 3 * s.x;
 Records affected: 0 
SQL> select * from t;

          ID            X
============ ============
           1           -6
           2          -12
           3          -18
           4          -24
           5          -30

Records affected: 5