Implemented fragmented signal from API

primarily to enable creation of tables with more attributes than 91 (up to 128)

Intermediate push for testing
CHUNK_SZ will be set higher in real implementation and API_TRACE added

ndb/src/ndbapi/TransporterFacade.cpp

@@ -475,7 +475,8 @@ TransporterFacade::TransporterFacade() :
   theTransporterRegistry(0),
   theStopReceive(0),
   theSendThread(NULL),
-  theReceiveThread(NULL)
+  theReceiveThread(NULL),
+  m_fragmented_signal_id(0)
 {
   theOwnId = 0;
 
@@ -833,6 +834,105 @@ TransporterFacade::sendSignalUnCond(NdbApiSignal * aSignal, NodeId aNode){
   return (ss == SEND_OK ? 0 : -1);
 }
 
+#define CHUNK_SZ 100u
+int
+TransporterFacade::sendFragmentedSignal(NdbApiSignal* aSignal, NodeId aNode, 
+					LinearSectionPtr ptr[3], Uint32 secs){
+  NdbApiSignal tmp_signal(*(SignalHeader*)aSignal);
+  LinearSectionPtr tmp_ptr[3];
+  Uint32 unique_id= m_fragmented_signal_id++; // next unique id
+  unsigned i;
+  for (i= 0; i < secs; i++)
+    tmp_ptr[i]= ptr[i];
+
+  unsigned start_i= 0;
+  unsigned chunk_sz= 0;
+  unsigned fragment_info= 0;
+  Uint32 *tmp_data= tmp_signal.getDataPtrSend();
+  for (i= 0; i < secs;) {
+    unsigned save_sz= tmp_ptr[i].sz;
+    tmp_data[i-start_i]= i;
+    if (chunk_sz + save_sz > CHUNK_SZ) {
+      // truncate
+      unsigned send_sz= CHUNK_SZ - chunk_sz;
+      tmp_ptr[i].sz= send_sz;
+      if (fragment_info < 2)
+	fragment_info++;
+
+      // send tmp_signal
+      tmp_data[i-start_i+1]= unique_id;
+      tmp_signal.setLength(i-start_i+2);
+      tmp_signal.m_fragmentInfo= fragment_info;
+      // do prepare send
+      {
+	int ret;
+	if(getIsNodeSendable(aNode) == true){
+	  SendStatus ss = theTransporterRegistry->prepareSend
+	    (&tmp_signal, 
+	     1, // JBB
+	     tmp_signal.getDataPtrSend(),
+	     aNode, 
+	     &ptr[start_i]);
+	  assert(ss != SEND_MESSAGE_TOO_BIG);
+	  ret = (ss == SEND_OK ? 0 : -1);
+	} else
+	  ret = -1;
+	if (ret != SEND_OK)
+	  return ret;
+      }
+
+      // setup variables for next signal
+      start_i= i;
+      chunk_sz= 0;
+      tmp_ptr[i].sz= save_sz-send_sz;
+      tmp_ptr[i].p+= send_sz;
+    }
+    else
+    {
+      chunk_sz+= save_sz;
+      i++;
+    }
+  }
+
+  unsigned a_sz= aSignal->getLength();
+
+  if (fragment_info > 0) {
+    // update the original signal to include section info
+    Uint32 *a_data= aSignal->getDataPtrSend();
+    unsigned tmp_sz= i-start_i;
+    memcpy(a_data+a_sz,
+	   tmp_data,
+	   tmp_sz*sizeof(Uint32));
+    a_data[a_sz+tmp_sz]= unique_id;
+    aSignal->setLength(a_sz+tmp_sz+1);
+
+    // send last fragment
+    aSignal->m_fragmentInfo= 3;
+    aSignal->m_noOfSections= i-start_i;
+  } else {
+    aSignal->m_noOfSections= secs;
+  }
+
+  // send aSignal
+  int ret;
+  if(getIsNodeSendable(aNode) == true){
+    SendStatus ss = theTransporterRegistry->prepareSend
+      (aSignal, 
+       1, // JBB
+       aSignal->getDataPtrSend(),
+       aNode, 
+       &ptr[start_i]);
+    assert(ss != SEND_MESSAGE_TOO_BIG);
+    ret = (ss == SEND_OK ? 0 : -1);
+  } else
+    ret = -1;
+  aSignal->m_noOfSections = 0;
+  aSignal->m_fragmentInfo = 0;
+  aSignal->setLength(a_sz);
+  return ret;
+}
+
+#if 0
 int
 TransporterFacade::sendFragmentedSignal(NdbApiSignal* aSignal, NodeId aNode, 
 					LinearSectionPtr ptr[3], Uint32 secs){
@@ -864,7 +964,7 @@ TransporterFacade::sendFragmentedSignal(NdbApiSignal* aSignal, NodeId aNode,
   aSignal->m_noOfSections = 0;
   return -1;
 }
-
+#endif
 
 
 int
@@ -886,11 +986,10 @@ TransporterFacade::sendFragmentedSignalUnCond(NdbApiSignal* aSignal,
       aSignal->theSendersBlockRef = tmp;
   }
 #endif
-  SendStatus ss = theTransporterRegistry->prepareSend(aSignal, 
-						      1, // JBB
-						      aSignal->getDataPtrSend(),
-						      aNode, 
-						      ptr);
+  SendStatus ss =
+    theTransporterRegistry->prepareSend(aSignal, 1, // JBB
+					aSignal->getDataPtrSend(),
+					aNode, ptr);
   assert(ss != SEND_MESSAGE_TOO_BIG);
   aSignal->m_noOfSections = 0;
   return (ss == SEND_OK ? 0 : -1);

ndb/src/ndbapi/TransporterFacade.hpp

@@ -224,7 +224,8 @@ private:
   } m_threads;
   
   Uint32 m_max_trans_id;
-  
+  Uint32 m_fragmented_signal_id;
+
   /**
    * execute function
    */