Merge mskold@bk-internal.mysql.com:/home/bk/mysql-5.0

into  mysql.com:/home/marty/MySQL/mysql-5.0

mysql-test/r/ndb_lock.result

@@ -63,3 +63,83 @@ pk	u	o
 5	5	5
 insert into t1 values (1,1,1);
 drop table t1;
+create table t1 (x integer not null primary key, y varchar(32), z integer, key(z)) engine = ndb;
+insert into t1 values (1,'one',1), (2,'two',2),(3,"three",3);
+begin;
+select * from t1 where x = 1 for update;
+x	y	z
+1	one	1
+begin;
+select * from t1 where x = 2 for update;
+x	y	z
+2	two	2
+select * from t1 where x = 1 for update;
+ERROR HY000: Lock wait timeout exceeded; try restarting transaction
+rollback;
+commit;
+begin;
+select * from t1 where y = 'one' or y = 'three' for update;
+x	y	z
+3	three	3
+1	one	1
+begin;
+select * from t1 where x = 1 for update;
+ERROR HY000: Lock wait timeout exceeded; try restarting transaction
+rollback;
+commit;
+begin;
+select * from t1 where z > 1 and z < 3 for update;
+x	y	z
+2	two	2
+begin;
+select * from t1 where x = 1 for update;
+x	y	z
+1	one	1
+select * from t1 where x = 2 for update;
+ERROR HY000: Lock wait timeout exceeded; try restarting transaction
+rollback;
+commit;
+begin;
+select * from t1 where x = 1 lock in share mode;
+x	y	z
+1	one	1
+begin;
+select * from t1 where x = 1 lock in share mode;
+x	y	z
+1	one	1
+select * from t1 where x = 2 for update;
+x	y	z
+2	two	2
+select * from t1 where x = 1 for update;
+ERROR HY000: Lock wait timeout exceeded; try restarting transaction
+rollback;
+commit;
+begin;
+select * from t1 where y = 'one' or y = 'three' lock in share mode;
+x	y	z
+3	three	3
+1	one	1
+begin;
+select * from t1 where y = 'one' lock in share mode;
+x	y	z
+1	one	1
+select * from t1 where x = 1 for update;
+ERROR HY000: Lock wait timeout exceeded; try restarting transaction
+rollback;
+commit;
+begin;
+select * from t1 where z > 1 and z < 3 lock in share mode;
+x	y	z
+2	two	2
+begin;
+select * from t1 where z = 1 lock in share mode;
+x	y	z
+1	one	1
+select * from t1 where x = 1 for update;
+x	y	z
+1	one	1
+select * from t1 where x = 2 for update;
+ERROR HY000: Lock wait timeout exceeded; try restarting transaction
+rollback;
+commit;
+drop table t1;

mysql-test/r/ndb_truncate.result

-DROP TABLE IF EXISTS t2;
-CREATE TABLE t2 (
-a bigint unsigned NOT NULL PRIMARY KEY,
+DROP TABLE IF EXISTS t1, t2;
+CREATE TABLE t1 (
+a bigint unsigned NOT NULL AUTO_INCREMENT PRIMARY KEY,
 b int unsigned not null,
 c int unsigned
 ) engine=ndbcluster;
-select count(*) from t2;
+select count(*) from t1;
 count(*)
 5000
-truncate table t2;
-select count(*) from t2;
+select * from t1 order by a limit 2;
+a	b	c
+1	509	2500
+2	510	7
+truncate table t1;
+select count(*) from t1;
 count(*)
 0
-drop table t2;
+insert into t1 values(NULL,1,1),(NULL,2,2);
+select * from t1 order by a;
+a	b	c
+1	1	1
+2	2	2
+drop table t1;

mysql-test/t/ndb_lock.test

@@ -69,4 +69,117 @@ insert into t1 values (1,1,1);
 
 drop table t1;
 
+# Lock for update
+
+create table t1 (x integer not null primary key, y varchar(32), z integer, key(z)) engine = ndb;
+
+insert into t1 values (1,'one',1), (2,'two',2),(3,"three",3); 
+
+# PK access
+connection con1;
+begin;
+select * from t1 where x = 1 for update;
+
+connection con2;
+begin;
+select * from t1 where x = 2 for update;
+--error 1205
+select * from t1 where x = 1 for update;
+rollback;
+
+connection con1;
+commit;
+
+# table scan
+connection con1;
+begin;
+select * from t1 where y = 'one' or y = 'three' for update;
+
+connection con2;
+begin;
+# Have to check with pk access here since scans take locks on
+# all rows and then release them in chunks
+# Bug #20390 SELECT FOR UPDATE does not release locks of untouched rows in full table scans
+#select * from t1 where x = 2 for update;
+--error 1205
+select * from t1 where x = 1 for update;
+rollback;
+
+connection con1;
+commit;
+
+# index scan
+connection con1;
+begin;
+select * from t1 where z > 1 and z < 3 for update;
+
+connection con2;
+begin;
+# Have to check with pk access here since scans take locks on
+# all rows and then release them in chunks
+select * from t1 where x = 1 for update;
+--error 1205
+select * from t1 where x = 2 for update;
+rollback;
+
+connection con1;
+commit;
+
+# share locking
+
+# PK access
+connection con1;
+begin;
+select * from t1 where x = 1 lock in share mode;
+
+connection con2;
+begin;
+select * from t1 where x = 1 lock in share mode;
+select * from t1 where x = 2 for update;
+--error 1205
+select * from t1 where x = 1 for update;
+rollback;
+
+connection con1;
+commit;
+
+# table scan
+connection con1;
+begin;
+select * from t1 where y = 'one' or y = 'three' lock in share mode;
+
+connection con2;
+begin;
+select * from t1 where y = 'one' lock in share mode;
+# Have to check with pk access here since scans take locks on
+# all rows and then release them in chunks
+# Bug #20390 SELECT FOR UPDATE does not release locks of untouched rows in full table scans
+#select * from t1 where x = 2 for update;
+--error 1205
+select * from t1 where x = 1 for update;
+rollback;
+
+connection con1;
+commit;
+
+# index scan
+connection con1;
+begin;
+select * from t1 where z > 1 and z < 3 lock in share mode;
+
+connection con2;
+begin;
+select * from t1 where z = 1 lock in share mode;
+# Have to check with pk access here since scans take locks on
+# all rows and then release them in chunks
+select * from t1 where x = 1 for update;
+--error 1205
+select * from t1 where x = 2 for update;
+rollback;
+
+connection con1;
+commit;
+
+drop table t1;
+
 # End of 4.1 tests

mysql-test/t/ndb_truncate.test

@@ -2,12 +2,11 @@
 -- source include/not_embedded.inc
 
 --disable_warnings
-DROP TABLE IF EXISTS t2;
+DROP TABLE IF EXISTS t1, t2;
 --enable_warnings
 
-
-CREATE TABLE t2 (
-  a bigint unsigned NOT NULL PRIMARY KEY,
+CREATE TABLE t1 (
+  a bigint unsigned NOT NULL AUTO_INCREMENT PRIMARY KEY,
   b int unsigned not null,
   c int unsigned
 ) engine=ndbcluster;
@@ -20,17 +19,23 @@ let $1=500;
 disable_query_log;
 while ($1)
 {
- eval insert into t2 values($1*10, $1+9, 5*$1), ($1*10+1, $1+10, 7),($1*10+2, $1+10, 7*$1), ($1*10+3, $1+10, 10+$1), ($1*10+4, $1+10, 70*$1), ($1*10+5, $1+10, 7), ($1*10+6, $1+10, 9), ($1*10+7, $1+299, 899), ($1*10+8, $1+10, 12), ($1*10+9, $1+10, 14*$1);
+ eval insert into t1 values(NULL, $1+9, 5*$1), (NULL, $1+10, 7),(NULL, $1+10, 7*$1), (NULL, $1+10, 10+$1), (NULL, $1+10, 70*$1), (NULL, $1+10, 7), (NULL, $1+10, 9), (NULL, $1+299, 899), (NULL, $1+10, 12), (NULL, $1+10, 14*$1);
  dec $1;
 }
 enable_query_log;
 
-select count(*) from t2;
+select count(*) from t1;
+
+select * from t1 order by a limit 2;
+
+truncate table t1;
+
+select count(*) from t1;
 
-truncate table t2;
+insert into t1 values(NULL,1,1),(NULL,2,2);
 
-select count(*) from t2;
+select * from t1 order by a;
 
-drop table t2;
+drop table t1;
 
 # End of 4.1 tests

ndb/include/ndbapi/NdbIndexScanOperation.hpp

@@ -61,11 +61,14 @@ public:
                         Uint32 parallel,
                         bool order_by,
                         bool order_desc = false,
-                        bool read_range_no = false) {
+                        bool read_range_no = false,
+			bool keyinfo = false) {
     Uint32 scan_flags =
       (SF_OrderBy & -(Int32)order_by) |
       (SF_Descending & -(Int32)order_desc) |
-      (SF_ReadRangeNo & -(Int32)read_range_no);
+      (SF_ReadRangeNo & -(Int32)read_range_no) | 
+      (SF_KeyInfo & -(Int32)keyinfo);
+    
     return readTuples(lock_mode, scan_flags, parallel);
   }
 #endif

ndb/include/ndbapi/NdbScanOperation.hpp

@@ -44,7 +44,8 @@ public:
     SF_TupScan = (1 << 16),     // scan TUP - only LM_CommittedRead
     SF_OrderBy = (1 << 24),     // index scan in order
     SF_Descending = (2 << 24),  // index scan in descending order
-    SF_ReadRangeNo = (4 << 24)  // enable @ref get_range_no
+    SF_ReadRangeNo = (4 << 24), // enable @ref get_range_no
+    SF_KeyInfo = 1              // request KeyInfo to be sent back
   };
 
   /**
@@ -61,15 +62,14 @@ public:
 #ifndef DOXYGEN_SHOULD_SKIP_DEPRECATED
   /**
    * readTuples
-   * 
    * @param lock_mode Lock mode
    * @param batch No of rows to fetch from each fragment at a time
    * @param parallel No of fragments to scan in parallell
-   * @note specifying 0 for batch and parallall means max performance
+   * @note specifying 0 for batch and parallell means max performance
    */ 
 #ifdef ndb_readtuples_impossible_overload
   int readTuples(LockMode lock_mode = LM_Read, 
-		 Uint32 batch = 0, Uint32 parallel = 0);
+		 Uint32 batch = 0, Uint32 parallel = 0, bool keyinfo = false);
 #endif
   
   inline int readTuples(int parallell){
@@ -141,6 +141,20 @@ public:
    */
   void close(bool forceSend = false, bool releaseOp = false);
 
+  /**
+   * Lock current tuple
+   *
+   * @return an NdbOperation or NULL.
+   */
+  NdbOperation* lockCurrentTuple();
+  /**
+   * Lock current tuple
+   *
+   * @param lockTrans Transaction that should perform the lock
+   *
+   * @return an NdbOperation or NULL.
+   */
+  NdbOperation*	lockCurrentTuple(NdbTransaction* lockTrans);
   /**
    * Update current tuple
    *
@@ -249,6 +263,19 @@ protected:
   NdbRecAttr *m_curr_row; // Pointer to last returned row
 };
 
+inline
+NdbOperation* 
+NdbScanOperation::lockCurrentTuple(){
+  return lockCurrentTuple(m_transConnection);
+}
+
+inline
+NdbOperation* 
+NdbScanOperation::lockCurrentTuple(NdbTransaction* takeOverTrans){
+  return takeOverScanOp(NdbOperation::ReadRequest, 
+			takeOverTrans);
+}
+
 inline
 NdbOperation* 
 NdbScanOperation::updateCurrentTuple(){

ndb/src/ndbapi/NdbScanOperation.cpp

@@ -160,7 +160,7 @@ NdbScanOperation::readTuples(NdbScanOperation::LockMode lm,
     return -1;
   }
 
-  m_keyInfo = lockExcl ? 1 : 0;
+  m_keyInfo = ((scan_flags & SF_KeyInfo) || lockExcl) ? 1 : 0;
 
   bool rangeScan = false;
   if (m_accessTable->m_indexType == NdbDictionary::Index::OrderedIndex)
@@ -924,18 +924,28 @@ NdbScanOperation::takeOverScanOp(OperationType opType, NdbTransaction* pTrans)
     if (newOp == NULL){
       return NULL;
     }
+    if (!m_keyInfo)
+    {
+      // Cannot take over lock if no keyinfo was requested
+      setErrorCodeAbort(4604);
+      return NULL;
+    }
     pTrans->theSimpleState = 0;
     
     const Uint32 len = (tRecAttr->attrSize() * tRecAttr->arraySize() + 3)/4-1;
 
     newOp->theTupKeyLen = len;
     newOp->theOperationType = opType;
-    if (opType == DeleteRequest) {
+    switch (opType) {
+    case (ReadRequest):
+      newOp->theLockMode = theLockMode;
+      // Fall through
+    case (DeleteRequest):
       newOp->theStatus = GetValue;
-    } else {
+      break;
+    default:
       newOp->theStatus = SetValue;
     }
-    
     const Uint32 * src = (Uint32*)tRecAttr->aRef();
     const Uint32 tScanInfo = src[len] & 0x3FFFF;
     const Uint32 tTakeOverFragment = src[len] >> 20;

ndb/src/ndbapi/ndberror.c

@@ -290,7 +290,7 @@ ErrorBundle ErrorCodes[] = {
   { 4601, AE, "Transaction is not started"},
   { 4602, AE, "You must call getNdbOperation before executeScan" },
   { 4603, AE, "There can only be ONE operation in a scan transaction" },
-  { 4604, AE, "takeOverScanOp, opType must be UpdateRequest or DeleteRequest" },
+  { 4604, AE, "takeOverScanOp, to take over a scanned row one must explicitly request keyinfo in readTuples call" },
   { 4605, AE, "You may only call openScanRead or openScanExclusive once for each operation"},
   { 4607, AE, "There may only be one operation in a scan transaction"},
   { 4608, AE, "You can not takeOverScan unless you have used openScanExclusive"},

sql/ha_ndbcluster.cc

@@ -70,7 +70,7 @@ handlerton ndbcluster_hton = {
   NULL, /* create_cursor_read_view */
   NULL, /* set_cursor_read_view */
   NULL, /* close_cursor_read_view */
-  HTON_NO_FLAGS
+  HTON_CAN_RECREATE
 };
 
 #define NDB_AUTO_INCREMENT_RETRIES 10
@@ -1174,12 +1174,23 @@ void ha_ndbcluster::release_metadata()
 
 int ha_ndbcluster::get_ndb_lock_type(enum thr_lock_type type)
 {
+  DBUG_ENTER("ha_ndbcluster::get_ndb_lock_type");
   if (type >= TL_WRITE_ALLOW_WRITE)
-    return NdbOperation::LM_Exclusive;
-  else if (uses_blob_value(m_retrieve_all_fields))
-    return NdbOperation::LM_Read;
+  {
+    DBUG_PRINT("info", ("Using exclusive lock"));
+    DBUG_RETURN(NdbOperation::LM_Exclusive);
+  }
+  else if (type ==  TL_READ_WITH_SHARED_LOCKS ||
+	   uses_blob_value(m_retrieve_all_fields))
+  {
+    DBUG_PRINT("info", ("Using read lock"));
+    DBUG_RETURN(NdbOperation::LM_Read);
+  }
   else
-    return NdbOperation::LM_CommittedRead;
+  {
+    DBUG_PRINT("info", ("Using committed read"));
+    DBUG_RETURN(NdbOperation::LM_CommittedRead);
+  }
 }
 
 static const ulong index_type_flags[]=
@@ -1679,7 +1690,30 @@ inline int ha_ndbcluster::fetch_next(NdbScanOperation* cursor)
   int check;
   NdbTransaction *trans= m_active_trans;
   
-  bool contact_ndb= m_lock.type < TL_WRITE_ALLOW_WRITE;
+    if (m_lock_tuple)
+  {
+    /*
+      Lock level m_lock.type either TL_WRITE_ALLOW_WRITE
+      (SELECT FOR UPDATE) or TL_READ_WITH_SHARED_LOCKS (SELECT
+      LOCK WITH SHARE MODE) and row was not explictly unlocked 
+      with unlock_row() call
+    */
+      NdbConnection *trans= m_active_trans;
+      NdbOperation *op;
+      // Lock row
+      DBUG_PRINT("info", ("Keeping lock on scanned row"));
+      
+      if (!(op= m_active_cursor->lockCurrentTuple()))
+      {
+	m_lock_tuple= false;
+	ERR_RETURN(trans->getNdbError());
+      }
+      m_ops_pending++;
+  }
+  m_lock_tuple= false;
+
+  bool contact_ndb= m_lock.type < TL_WRITE_ALLOW_WRITE &&
+                    m_lock.type != TL_READ_WITH_SHARED_LOCKS;
   do {
     DBUG_PRINT("info", ("Call nextResult, contact_ndb: %d", contact_ndb));
     /*
@@ -1695,6 +1729,13 @@ inline int ha_ndbcluster::fetch_next(NdbScanOperation* cursor)
     
     if ((check= cursor->nextResult(contact_ndb, m_force_send)) == 0)
     {
+      /*
+	Explicitly lock tuple if "select for update" or
+	"select lock in share mode"
+      */
+      m_lock_tuple= (m_lock.type == TL_WRITE_ALLOW_WRITE
+		     || 
+		     m_lock.type == TL_READ_WITH_SHARED_LOCKS);
       DBUG_RETURN(0);
     } 
     else if (check == 1 || check == 2)
@@ -1983,10 +2024,11 @@ int ha_ndbcluster::ordered_index_scan(const key_range *start_key,
     restart= FALSE;
     NdbOperation::LockMode lm=
       (NdbOperation::LockMode)get_ndb_lock_type(m_lock.type);
+    bool need_pk = (lm == NdbOperation::LM_Read);
     if (!(op= trans->getNdbIndexScanOperation((NDBINDEX *)
                                               m_index[active_index].index, 
                                               (const NDBTAB *) m_table)) ||
-        op->readTuples(lm, 0, parallelism, sorted, descending))
+        op->readTuples(lm, 0, parallelism, sorted, descending, false, need_pk))
       ERR_RETURN(trans->getNdbError());
     m_active_cursor= op;
   } else {
@@ -2036,8 +2078,11 @@ int ha_ndbcluster::full_table_scan(byte *buf)
 
   NdbOperation::LockMode lm=
     (NdbOperation::LockMode)get_ndb_lock_type(m_lock.type);
+  bool need_pk = (lm == NdbOperation::LM_Read);
   if (!(op=trans->getNdbScanOperation((const NDBTAB *) m_table)) ||
-      op->readTuples(lm, 0, parallelism))
+      op->readTuples(lm, 
+		     (need_pk)?NdbScanOperation::SF_KeyInfo:0, 
+		     parallelism))
     ERR_RETURN(trans->getNdbError());
   m_active_cursor= op;
   if (generate_scan_filter(m_cond_stack, op))
@@ -2327,6 +2372,7 @@ int ha_ndbcluster::update_row(const byte *old_data, byte *new_data)
     DBUG_PRINT("info", ("Calling updateTuple on cursor"));
     if (!(op= cursor->updateCurrentTuple()))
       ERR_RETURN(trans->getNdbError());
+    m_lock_tuple= false;
     m_ops_pending++;
     if (uses_blob_value(FALSE))
       m_blobs_pending= TRUE;
@@ -2406,6 +2452,7 @@ int ha_ndbcluster::delete_row(const byte *record)
     DBUG_PRINT("info", ("Calling deleteTuple on cursor"));
     if (cursor->deleteCurrentTuple() != 0)
       ERR_RETURN(trans->getNdbError());     
+    m_lock_tuple= false;
     m_ops_pending++;
 
     no_uncommitted_rows_update(-1);
@@ -2531,7 +2578,7 @@ void ha_ndbcluster::unpack_record(byte* buf)
     DBUG_PRINT("hidden", ("%d: %s \"%llu\"", hidden_no, 
 			  hidden_col->getName(), rec->u_64_value()));
   } 
-  //print_results();
+  print_results();
 #endif
   DBUG_VOID_RETURN;
 }
@@ -2605,6 +2652,12 @@ int ha_ndbcluster::index_init(uint index)
 {
   DBUG_ENTER("ha_ndbcluster::index_init");
   DBUG_PRINT("enter", ("index: %u", index));
+ /*
+    Locks are are explicitly released in scan
+    unless m_lock.type == TL_READ_HIGH_PRIORITY
+    and no sub-sequent call to unlock_row()
+   */
+  m_lock_tuple= false;
   DBUG_RETURN(handler::index_init(index));
 }
 
@@ -3613,6 +3666,22 @@ int ha_ndbcluster::external_lock(THD *thd, int lock_type)
   DBUG_RETURN(error);
 }
 
+/*
+  Unlock the last row read in an open scan.
+  Rows are unlocked by default in ndb, but
+  for SELECT FOR UPDATE and SELECT LOCK WIT SHARE MODE
+  locks are kept if unlock_row() is not called.
+*/
+
+void ha_ndbcluster::unlock_row() 
+{
+  DBUG_ENTER("unlock_row");
+
+  DBUG_PRINT("info", ("Unlocking row"));
+  m_lock_tuple= false;
+  DBUG_VOID_RETURN;
+}
+
 /*
   Start a transaction for running a statement if one is not
   already running in a transaction. This will be the case in
@@ -4085,6 +4154,12 @@ int ha_ndbcluster::create(const char *name,
   set_dbname(name2);
   set_tabname(name2);    
 
+  if (current_thd->lex->sql_command == SQLCOM_TRUNCATE)
+  {
+    DBUG_PRINT("info", ("Dropping and re-creating table for TRUNCATE"));
+    if ((my_errno= delete_table(name)))
+      DBUG_RETURN(my_errno);
+  }
   if (create_from_engine)
   {
     /*
@@ -5891,6 +5966,7 @@ ha_ndbcluster::read_multi_range_first(KEY_MULTI_RANGE **found_range_p,
   byte *end_of_buffer= (byte*)buffer->buffer_end;
   NdbOperation::LockMode lm= 
     (NdbOperation::LockMode)get_ndb_lock_type(m_lock.type);
+  bool need_pk = (lm == NdbOperation::LM_Read);
   const NDBTAB *tab= (const NDBTAB *) m_table;
   const NDBINDEX *unique_idx= (NDBINDEX *) m_index[active_index].unique_index;
   const NDBINDEX *idx= (NDBINDEX *) m_index[active_index].index; 
@@ -5957,7 +6033,8 @@ ha_ndbcluster::read_multi_range_first(KEY_MULTI_RANGE **found_range_p,
           end_of_buffer -= reclength;
         }
         else if ((scanOp= m_active_trans->getNdbIndexScanOperation(idx, tab)) 
-                 &&!scanOp->readTuples(lm, 0, parallelism, sorted, FALSE, TRUE)
+                 &&!scanOp->readTuples(lm, 0, parallelism, sorted, 
+				       FALSE, TRUE, need_pk)
                  &&!generate_scan_filter(m_cond_stack, scanOp)
                  &&!define_read_attrs(end_of_buffer-reclength, scanOp))
         {

sql/ha_ndbcluster.h

@@ -503,6 +503,7 @@ class ha_ndbcluster: public handler
   int extra(enum ha_extra_function operation);
   int extra_opt(enum ha_extra_function operation, ulong cache_size);
   int external_lock(THD *thd, int lock_type);
+  void unlock_row();
   int start_stmt(THD *thd, thr_lock_type lock_type);
   const char * table_type() const;
   const char ** bas_ext() const;
@@ -684,6 +685,7 @@ static void set_tabname(const char *pathname, char *tabname);
   char m_tabname[FN_HEADLEN];
   ulong m_table_flags;
   THR_LOCK_DATA m_lock;
+  bool m_lock_tuple;
   NDB_SHARE *m_share;
   NDB_INDEX_DATA  m_index[MAX_KEY];
   // NdbRecAttr has no reference to blob