sapi: Use one SMP lock for all chains

This partially reverts commit 1215fd4d9426a59d568560e9a485628560363133.

In order to support profiling of SMP locks and provide a future
compatible SMP locks API it is necessary to add an SMP lock destroy
function.  Since the commit above adds an SMP lock to each chain control
we would have to add a rtems_chain_destroy() function as well.  This
complicates the chain usage dramatically.  Thus revert the patch above.
A global SMP lock for all chains is used to implement the protected
chain operations.

Advantages:

* The SAPI chain API is now identical on SMP and non-SMP
  configurations.

* The size of the chain control is reduced and is then equal to the
  Score chains.

* The protected chain operations work correctly on SMP.

Disadvantage:

* Applications using many different chains and the protected operations
  may notice lock contention.

The chain control size drop is a huge benefit (SAPI chain controls are
66% larger than the Score chain controls).  The only disadvantage is not
really a problem since these applications can use specific interrupt
locks and unprotected chain operations to avoid this issue.

1 files changed, 8 insertions, 13 deletions

diff --git a/cpukit/posix/src/aio_misc.c b/cpukit/posix/src/aio_misc.c
index 656ba41bab..36785739d4 100644
--- a/cpukit/posix/src/aio_misc.c
+++ b/cpukit/posix/src/aio_misc.c
@@ -120,9 +120,7 @@ rtems_aio_search_fd (rtems_chain_control *chain, int fildes, int create)
       if (rtems_chain_is_empty (chain))
         rtems_chain_prepend (chain, &r_chain->next_fd);
       else
-        rtems_chain_explicit_insert (chain,
-                                     rtems_chain_previous (node),
-                                     &r_chain->next_fd);
+        rtems_chain_insert (rtems_chain_previous (node), &r_chain->next_fd);
 
       r_chain->new_fd = 1;
 	  r_chain->fildes = fildes;
@@ -159,9 +157,7 @@ rtems_aio_move_to_work (rtems_aio_request_chain *r_chain)
     temp = (rtems_aio_request_chain *) node;
   }
 
-  rtems_chain_explicit_insert (work_req_chain,
-                               rtems_chain_previous (node),
-                               &r_chain->next_fd);
+  rtems_chain_insert (rtems_chain_previous (node), &r_chain->next_fd);
 }
  
 
@@ -200,7 +196,7 @@ rtems_aio_insert_prio (rtems_chain_control *chain, rtems_aio_request *req)
       prio = ((rtems_aio_request *) node)->aiocbp->aio_reqprio;
     }
 
-    rtems_chain_explicit_insert (chain, node->previous, &req->next_prio);
+    rtems_chain_insert (node->previous, &req->next_prio);
 
   }
 }
@@ -226,7 +222,7 @@ void rtems_aio_remove_fd (rtems_aio_request_chain *r_chain)
   
   while (!rtems_chain_is_tail (chain, node))
     {
-      rtems_chain_explicit_extract (chain, node);
+      rtems_chain_extract (node);
       rtems_aio_request *req = (rtems_aio_request *) node;
       node = rtems_chain_next (node);
       req->aiocbp->error_code = ECANCELED;
@@ -270,7 +266,7 @@ int rtems_aio_remove_req (rtems_chain_control *chain, struct aiocb *aiocbp)
     return AIO_NOTCANCELED;
   else
     {
-      rtems_chain_explicit_extract (chain, node);
+      rtems_chain_extract (node);
       current->aiocbp->error_code = ECANCELED;
       current->aiocbp->return_value = -1;
       free (current); 
@@ -445,7 +441,7 @@ rtems_aio_handle (void *arg)
       param.sched_priority = req->priority;
       pthread_setschedparam (pthread_self(), req->policy, &param);
 
-      rtems_chain_explicit_extract (chain, node);
+      rtems_chain_extract (node);
 
       pthread_mutex_unlock (&r_chain->mutex);
 
@@ -511,8 +507,7 @@ rtems_aio_handle (void *arg)
 	  /* If no requests were added to the chain we delete the fd chain from 
 	     the queue and start working with idle fd chains */
 	  if (result == ETIMEDOUT) {
-	    rtems_chain_explicit_extract (&aio_request_queue.work_req,
-                                    &r_chain->next_fd);
+	    rtems_chain_extract (&r_chain->next_fd);
 	    pthread_mutex_destroy (&r_chain->mutex);
 	    pthread_cond_destroy (&r_chain->cond);
 	    free (r_chain);
@@ -548,7 +543,7 @@ rtems_aio_handle (void *arg)
 	    ++aio_request_queue.active_threads;
 
 	    node = rtems_chain_first (&aio_request_queue.idle_req);
-	    rtems_chain_explicit_extract (&aio_request_queue.idle_req, node);
+	    rtems_chain_extract (node);
 
 	    r_chain = (rtems_aio_request_chain *) node;
 	    rtems_aio_move_to_work (r_chain);