Fixed bug#28404.

This patch adds cost estimation for the queries with ORDER BY / GROUP BY
and LIMIT. 
If there was a ref/range access to the table whose rows were required
to be ordered in the result set the optimizer always employed this access
though a scan by a different index that was compatible with the required 
order could be cheaper to produce the first L rows of the result set.
Now for such queries the optimizer makes a choice between the cheapest
ref/range accesses not compatible with the given order and index scans
compatible with it.

1 files changed, 1 insertions, 1 deletions

diff --git a/mysql-test/t/distinct.test b/mysql-test/t/distinct.test
index 7310f98cd16..bfdb5f8b9f8 100644
--- a/mysql-test/t/distinct.test
+++ b/mysql-test/t/distinct.test
@@ -97,7 +97,7 @@ explain SELECT t1.a from t1 group by a order by a desc;
 explain SELECT distinct t1.a from t1 order by a desc limit 1;
 explain SELECT distinct a from t3 order by a desc limit 2;
 explain SELECT distinct a,b from t3 order by a+1;
-explain SELECT distinct a,b from t3 order by a limit 10;
+explain SELECT distinct a,b from t3 order by a limit 2;
 explain SELECT a,b from t3 group by a,b order by a+1;
 
 drop table t1,t2,t3,t4;