multi: remove PENDING + MSGSENT handles from the main linked list

As they are not driving transfers or any socket activity, the main loop
does not need to iterate over these handles. A performance improvement.

They are instead only held in their own separate lists.

'data->multi' is kept a pointer to the multi handle as long as the easy
handle is actually part of it even when the handle is moved to the
pending/msgsent lists. It needs to know which multi handle it belongs
to, if for example curl_easy_cleanup() is called before the handle is
removed from the multi handle.

Alll 'data->multi' pointers of handles still part of the multi handle
gets cleared by curl_multi_cleanup() which "orphans" all previously
attached easy handles.

This is take 2. The first version was reverted for the 8.0.1 release.

Assisted-by: Stefan Eissing
Closes #10801

6 files changed, 73 insertions, 36 deletions

diff --git a/lib/ftp.c b/lib/ftp.c
index caf33d214..ac09853f7 100644
--- a/lib/ftp.c
+++ b/lib/ftp.c
@@ -3977,8 +3977,10 @@ static CURLcode wc_statemach(struct Curl_easy *data)
     case CURLWC_DONE:
     case CURLWC_ERROR:
     case CURLWC_CLEAR:
-      if(wildcard->dtor)
+      if(wildcard->dtor) {
         wildcard->dtor(wildcard->ftpwc);
+        wildcard->ftpwc = NULL;
+      }
       return result;
     }
   }
diff --git a/lib/multi.c b/lib/multi.c
index 262321d9d..a011a36be 100644
--- a/lib/multi.c
+++ b/lib/multi.c
@@ -392,12 +392,10 @@ static void sh_init(struct Curl_hash *hash, int hashsize)
  * Called when a transfer is completed. Adds the given msg pointer to
  * the list kept in the multi handle.
  */
-static CURLMcode multi_addmsg(struct Curl_multi *multi,
-                              struct Curl_message *msg)
+static void multi_addmsg(struct Curl_multi *multi, struct Curl_message *msg)
 {
   Curl_llist_insert_next(&multi->msglist, multi->msglist.tail, msg,
                          &msg->list);
-  return CURLM_OK;
 }
 
 struct Curl_multi *Curl_multi_handle(int hashsize, /* socket hash */
@@ -420,6 +418,7 @@ struct Curl_multi *Curl_multi_handle(int hashsize, /* socket hash */
 
   Curl_llist_init(&multi->msglist, NULL);
   Curl_llist_init(&multi->pending, NULL);
+  Curl_llist_init(&multi->msgsent, NULL);
 
   multi->multiplexing = TRUE;
 
@@ -465,6 +464,14 @@ struct Curl_multi *curl_multi_init(void)
                            CURL_DNS_HASH_SIZE);
 }
 
+/* returns TRUE if the easy handle is supposed to be present in the main link
+   list */
+static bool in_main_list(struct Curl_easy *data)
+{
+  return ((data->mstate != MSTATE_PENDING) &&
+          (data->mstate != MSTATE_MSGSENT));
+}
+
 static void link_easy(struct Curl_multi *multi,
                       struct Curl_easy *data)
 {
@@ -498,6 +505,8 @@ static void unlink_easy(struct Curl_multi *multi,
     data->next->prev = data->prev;
   else
     multi->easylp = data->prev; /* point to last node */
+
+  data->prev = data->next = NULL;
 }
 
 
@@ -857,10 +866,16 @@ CURLMcode curl_multi_remove_handle(struct Curl_multi *multi,
      called. Do it after multi_done() in case that sets another time! */
   Curl_expire_clear(data);
 
-  if(data->connect_queue.ptr)
-    /* the handle was in the pending list waiting for an available connection,
-       so go ahead and remove it */
-    Curl_llist_remove(&multi->pending, &data->connect_queue, NULL);
+  if(data->connect_queue.ptr) {
+    /* the handle is in the pending or msgsent lists, so go ahead and remove
+       it */
+    if(data->mstate == MSTATE_PENDING)
+      Curl_llist_remove(&multi->pending, &data->connect_queue, NULL);
+    else
+      Curl_llist_remove(&multi->msgsent, &data->connect_queue, NULL);
+  }
+  if(in_main_list(data))
+    unlink_easy(multi, data);
 
   if(data->dns.hostcachetype == HCACHE_MULTI) {
     /* stop using the multi handle's DNS cache, *after* the possible
@@ -921,7 +936,6 @@ CURLMcode curl_multi_remove_handle(struct Curl_multi *multi,
 
   /* make sure there's no pending message in the queue sent from this easy
      handle */
-
   for(e = multi->msglist.head; e; e = e->next) {
     struct Curl_message *msg = e->ptr;
 
@@ -932,19 +946,6 @@ CURLMcode curl_multi_remove_handle(struct Curl_multi *multi,
     }
   }
 
-  /* Remove from the pending list if it is there. Otherwise this will
-     remain on the pending list forever due to the state change. */
-  for(e = multi->pending.head; e; e = e->next) {
-    struct Curl_easy *curr_data = e->ptr;
-
-    if(curr_data == data) {
-      Curl_llist_remove(&multi->pending, e, NULL);
-      break;
-    }
-  }
-
-  unlink_easy(multi, data);
-
   /* NOTE NOTE NOTE
      We do not touch the easy handle here! */
   multi->num_easy--; /* one less to care about now */
@@ -1952,11 +1953,6 @@ static CURLMcode multi_runsingle(struct Curl_multi *multi,
       }
       break;
 
-    case MSTATE_PENDING:
-      /* We will stay here until there is a connection available. Then
-         we try again in the MSTATE_CONNECT state. */
-      break;
-
     case MSTATE_CONNECT:
       /* Connect. We want to get a connection identifier filled in. */
       /* init this transfer. */
@@ -1980,6 +1976,8 @@ static CURLMcode multi_runsingle(struct Curl_multi *multi,
         /* add this handle to the list of connect-pending handles */
         Curl_llist_insert_next(&multi->pending, multi->pending.tail, data,
                                &data->connect_queue);
+        /* unlink from the main list */
+        unlink_easy(multi, data);
         result = CURLE_OK;
         break;
       }
@@ -2606,9 +2604,11 @@ static CURLMcode multi_runsingle(struct Curl_multi *multi,
     case MSTATE_COMPLETED:
       break;
 
+    case MSTATE_PENDING:
     case MSTATE_MSGSENT:
-      data->result = result;
-      return CURLM_OK; /* do nothing */
+      /* handles in these states should NOT be in this list */
+      DEBUGASSERT(0);
+      break;
 
     default:
       return CURLM_INTERNAL_ERROR;
@@ -2696,10 +2696,17 @@ static CURLMcode multi_runsingle(struct Curl_multi *multi,
         msg->extmsg.easy_handle = data;
         msg->extmsg.data.result = result;
 
-        rc = multi_addmsg(multi, msg);
+        multi_addmsg(multi, msg);
         DEBUGASSERT(!data->conn);
       }
       multistate(data, MSTATE_MSGSENT);
+
+      /* add this handle to the list of msgsent handles */
+      Curl_llist_insert_next(&multi->msgsent, multi->msgsent.tail, data,
+                             &data->connect_queue);
+      /* unlink from the main list */
+      unlink_easy(multi, data);
+      return CURLM_OK;
     }
   } while((rc == CURLM_CALL_MULTI_PERFORM) || multi_ischanged(multi, FALSE));
 
@@ -2730,6 +2737,9 @@ CURLMcode curl_multi_perform(struct Curl_multi *multi, int *running_handles)
     /* Do the loop and only alter the signal ignore state if the next handle
        has a different NO_SIGNAL state than the previous */
     do {
+      /* the current node might be unlinked in multi_runsingle(), get the next
+         pointer now */
+      struct Curl_easy *datanext = data->next;
       if(data->set.no_signal != nosig) {
         sigpipe_restore(&pipe_st);
         sigpipe_ignore(data, &pipe_st);
@@ -2738,7 +2748,7 @@ CURLMcode curl_multi_perform(struct Curl_multi *multi, int *running_handles)
       result = multi_runsingle(multi, &now, data);
       if(result)
         returncode = result;
-      data = data->next; /* operate on next handle */
+      data = datanext; /* operate on next handle */
     } while(data);
     sigpipe_restore(&pipe_st);
   }
@@ -2769,6 +2779,18 @@ CURLMcode curl_multi_perform(struct Curl_multi *multi, int *running_handles)
   return returncode;
 }
 
+/* unlink_all_msgsent_handles() detaches all those easy handles from this
+   multi handle */
+static void unlink_all_msgsent_handles(struct Curl_multi *multi)
+{
+  struct Curl_llist_element *e = multi->msgsent.head;
+  if(e) {
+    struct Curl_easy *data = e->ptr;
+    DEBUGASSERT(data->mstate == MSTATE_MSGSENT);
+    data->multi = NULL;
+  }
+}
+
 CURLMcode curl_multi_cleanup(struct Curl_multi *multi)
 {
   struct Curl_easy *data;
@@ -2780,6 +2802,8 @@ CURLMcode curl_multi_cleanup(struct Curl_multi *multi)
 
     multi->magic = 0; /* not good anymore */
 
+    unlink_all_msgsent_handles(multi);
+    process_pending_handles(multi);
     /* First remove all remaining easy handles */
     data = multi->easyp;
     while(data) {
@@ -3711,6 +3735,8 @@ void Curl_multiuse_state(struct Curl_easy *data,
   process_pending_handles(data->multi);
 }
 
+/* process_pending_handles() moves all handles from PENDING
+   back into the main list and change state to CONNECT */
 static void process_pending_handles(struct Curl_multi *multi)
 {
   struct Curl_llist_element *e = multi->pending.head;
@@ -3719,6 +3745,9 @@ static void process_pending_handles(struct Curl_multi *multi)
 
     DEBUGASSERT(data->mstate == MSTATE_PENDING);
 
+    /* put it back into the main list */
+    link_easy(multi, data);
+
     multistate(data, MSTATE_CONNECT);
 
     /* Remove this node from the list */
diff --git a/lib/multihandle.h b/lib/multihandle.h
index 6cda65d44..5b16bb605 100644
--- a/lib/multihandle.h
+++ b/lib/multihandle.h
@@ -101,6 +101,8 @@ struct Curl_multi {
 
   struct Curl_llist pending; /* Curl_easys that are in the
                                 MSTATE_PENDING state */
+  struct Curl_llist msgsent; /* Curl_easys that are in the
+                                MSTATE_MSGSENT state */
 
   /* callback function and user data pointer for the *socket() API */
   curl_socket_callback socket_cb;
diff --git a/lib/transfer.c b/lib/transfer.c
index a28395233..cb781833f 100644
--- a/lib/transfer.c
+++ b/lib/transfer.c
@@ -1408,7 +1408,12 @@ CURLcode Curl_pretransfer(struct Curl_easy *data)
           return CURLE_OUT_OF_MEMORY;
       }
       wc = data->wildcard;
-      if(wc->state < CURLWC_INIT) {
+      if((wc->state < CURLWC_INIT) ||
+         (wc->state >= CURLWC_CLEAN)) {
+        if(wc->ftpwc)
+          wc->dtor(wc->ftpwc);
+        Curl_safefree(wc->pattern);
+        Curl_safefree(wc->path);
         result = Curl_wildcard_init(wc); /* init wildcard structures */
         if(result)
           return CURLE_OUT_OF_MEMORY;
diff --git a/lib/url.c b/lib/url.c
index f7b4bbbe9..c1fdd1d67 100644
--- a/lib/url.c
+++ b/lib/url.c
@@ -346,7 +346,6 @@ static void up_free(struct Curl_easy *data)
 
 CURLcode Curl_close(struct Curl_easy **datap)
 {
-  struct Curl_multi *m;
   struct Curl_easy *data;
 
   if(!datap || !*datap)
@@ -360,8 +359,7 @@ CURLcode Curl_close(struct Curl_easy **datap)
   /* Detach connection if any is left. This should not be normal, but can be
      the case for example with CONNECT_ONLY + recv/send (test 556) */
   Curl_detach_connection(data);
-  m = data->multi;
-  if(m)
+  if(data->multi)
     /* This handle is still part of a multi handle, take care of this first
        and detach this handle from there. */
     curl_multi_remove_handle(data->multi, data);
diff --git a/lib/urldata.h b/lib/urldata.h
index 86b3da718..1fc9d91ff 100644
--- a/lib/urldata.h
+++ b/lib/urldata.h
@@ -1903,7 +1903,8 @@ struct Curl_easy {
   struct Curl_easy *prev;
 
   struct connectdata *conn;
-  struct Curl_llist_element connect_queue;
+  struct Curl_llist_element connect_queue; /* for the pending and msgsent
+                                              lists */
   struct Curl_llist_element conn_queue; /* list per connectdata */
 
   CURLMstate mstate;  /* the handle's state */