问题是这样的：
1 sql> r
2 1 select count(*)
3 2 from t_edu_member_info
4 3* where status='1' and xs_zy='1'
01 执行计划
02 -------------------------------------------------------------------------------------------
03 0 select statement optimizer=all_rows (cost=1 card=1 bytes=10)
04 1 0 sort (aggregate)
05 2 1 filter
06 3 2 index (range scan) of 'ind_t_prebm4' (non-unique) (cost=1 card=5 bytes=50)
07
08 统计信息
09 ----------------------------------------------------------
10 25 recursive calls
11 14 db block gets
12 3 consistent gets
13 1 physical reads
14 972 redo size
15 375 bytes sent via sql*net to client
16 372 bytes received via sql*net from client
17 2 sql*net roundtrips to/from client
18 1 sorts (memory)
19 0 sorts (disk)
20 1 rows processed
这个结果已经跑了几次，但是 recursive calls 依然存在，后来从10046事件产生的trace文件中到了线索。用tkprof格式化trace文件后，可以看到：
1 select count(*)
2 from t_edu_member_info
3 where status='1'
4 and xs_zy='1'
01 call count cpu elapsed disk query current rows
02 ------- ------ -------- ---------- ---------- ---------- ---------- ----------
03 parse 1 0.01 0.00 0 0 0 0
04 execute 1 0.78 0.90
bitscn.com