Fix bogus coding in ExecAppendAsyncEventWait().

No configured-by-FDW events would result in "return" directly out of a
PG_TRY block, making the exception stack dangling.  Repair.

Oversight in commit 501cfd07d; back-patch to v14, like that commit, but
as we do not have this issue in HEAD (cf. commit 50c67c201), no need to
apply this patch to it.

In passing, improve a comment about the handling of in-process requests
in a postgres_fdw.c function called from this function.

Alexander Pyhalov, with comment adjustment/improvement by me.

Discussion: https://postgr.es/m/425fa29a429b21b0332737c42a4fdc70%40postgrespro.ru

diff --git a/contrib/postgres_fdw/postgres_fdw.c b/contrib/postgres_fdw/postgres_fdw.c
index c5cada55fb732942ba586aeecefc395b646ba55c..23a57104e857aca55eb85b78968573ff11e56a1c 100644 (file)
--- a/contrib/postgres_fdw/postgres_fdw.c
+++ b/contrib/postgres_fdw/postgres_fdw.c
@@ -7174,14 +7174,16 @@ postgresForeignAsyncConfigureWait(AsyncRequest *areq)
    {
        /*
         * This is the case when the in-process request was made by another
-        * Append.  Note that it might be useless to process the request,
-        * because the query might not need tuples from that Append anymore.
-        * If there are any child subplans of the same parent that are ready
-        * for new requests, skip the given request.  Likewise, if there are
-        * any configured events other than the postmaster death event, skip
-        * it.  Otherwise, process the in-process request, then begin a fetch
-        * to configure the event below, because we might otherwise end up
-        * with no configured events other than the postmaster death event.
+        * Append.  Note that it might be useless to process the request made
+        * by that Append, because the query might not need tuples from that
+        * Append anymore; so we avoid processing it to begin a fetch for the
+        * given request if possible.  If there are any child subplans of the
+        * same parent that are ready for new requests, skip the given
+        * request.  Likewise, if there are any configured events other than
+        * the postmaster death event, skip it.  Otherwise, process the
+        * in-process request, then begin a fetch to configure the event
+        * below, because we might otherwise end up with no configured events
+        * other than the postmaster death event.
         */
        if (!bms_is_empty(requestor->as_needrequest))
            return;

diff --git a/src/backend/executor/nodeAppend.c b/src/backend/executor/nodeAppend.c
index 99818d3ebceebc6d7dda70b84482e0acfb572ae2..338484bac9ec41bf14aa99ce1c1253de2864369a 100644 (file)
--- a/src/backend/executor/nodeAppend.c
+++ b/src/backend/executor/nodeAppend.c
@@ -1043,26 +1043,25 @@ ExecAppendAsyncEventWait(AppendState *node)
        }
 
        /*
-        * No need for further processing if there are no configured events
-        * other than the postmaster death event.
+        * If there are no configured events other than the postmaster death
+        * event, we don't need to wait or poll.
         */
        if (GetNumRegisteredWaitEvents(node->as_eventset) == 1)
+           noccurred = 0;
+       else
        {
-           FreeWaitEventSet(node->as_eventset);
-           node->as_eventset = NULL;
-           return;
-       }
+           /* Return at most EVENT_BUFFER_SIZE events in one call. */
+           if (nevents > EVENT_BUFFER_SIZE)
+               nevents = EVENT_BUFFER_SIZE;
 
-       /* Return at most EVENT_BUFFER_SIZE events in one call. */
-       if (nevents > EVENT_BUFFER_SIZE)
-           nevents = EVENT_BUFFER_SIZE;
-
-       /*
-        * If the timeout is -1, wait until at least one event occurs.  If the
-        * timeout is 0, poll for events, but do not wait at all.
-        */
-       noccurred = WaitEventSetWait(node->as_eventset, timeout, occurred_event,
-                                    nevents, WAIT_EVENT_APPEND_READY);
+           /*
+            * If the timeout is -1, wait until at least one event occurs.  If
+            * the timeout is 0, poll for events, but do not wait at all.
+            */
+           noccurred = WaitEventSetWait(node->as_eventset, timeout,
+                                        occurred_event, nevents,
+                                        WAIT_EVENT_APPEND_READY);
+       }
    }
    PG_FINALLY();
    {