Bug#10178 - failure to find a row in heap table by concurrent UPDATEs

Moved the key statistics update to info().
The table is not locked in open(). This made wrong stats possible.

No test case for the test suite.
This happens only with heavy concurrency.
A test script is added to the bug report.


mysql-test/r/heap_hash.result:
  Bug#10178 - failure to find a row in heap table by concurrent UPDATEs
  Updated test results to reflect the new  statistics behaviour.
mysql-test/t/heap_hash.test:
  Bug#10178 - failure to find a row in heap table by concurrent UPDATEs
  Added a FLUSH TABLES to avoid statistics differences between normal 
  and ps-protocol tests.
sql/ha_heap.cc:
  Bug#10178 - failure to find a row in heap table by concurrent UPDATEs
  Moved the key statistics update to info().
  The table is not locked in open(). This made wrong stats possible.
sql/ha_heap.h:
  Bug#10178 - failure to find a row in heap table by concurrent UPDATEs
  Added an element to track the validity of the key statistics.

1 files changed, 3 insertions, 1 deletions

diff --git a/sql/ha_heap.h b/sql/ha_heap.h
index 33de0156074..cbe2474492d 100644
--- a/sql/ha_heap.h
+++ b/sql/ha_heap.h
@@ -29,8 +29,10 @@ class ha_heap: public handler
   key_map btree_keys;
   /* number of records changed since last statistics update */
   uint    records_changed;
+  bool    key_stats_ok;
 public:
-  ha_heap(TABLE *table): handler(table), file(0), records_changed(0) {}
+  ha_heap(TABLE *table): handler(table), file(0), records_changed(0),
+      key_stats_ok(0) {}
   ~ha_heap() {}
   const char *table_type() const { return "HEAP"; }
   const char *index_type(uint inx)