Use WL_SOCKET_CLOSED for client_connection_check_interval.

macdice · macdice · commit cba5b994c990 · 2022-02-14T16:52:23.000+13:00
Previously we used poll() directly to check for a POLLRDHUP event. Instead, use the WaitEventSet API to poll the socket for WL_SOCKET_CLOSED, which knows how to detect this condition on many more operating systems. Reviewed-by: Zhihong Yu <zyu@yugabyte.com> Reviewed-by: Maksim Milyutin <milyutinma@gmail.com> Reviewed-by: Andres Freund <andres@anarazel.de> Discussion: https://postgr.es/m/77def86b27e41f0efcba411460e929ae%40postgrespro.ru
diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
@@ -1012,9 +1012,9 @@ include_dir 'conf.d'
         the kernel reports that the connection is closed.
        </para>
        <para>
-        This option is currently available only on systems that support the
-        non-standard <symbol>POLLRDHUP</symbol> extension to the
-        <symbol>poll</symbol> system call, including Linux.
+        This option relies on kernel events exposed by Linux, macOS, illumos
+        and the BSD family of operating systems, and is not currently available
+        on other systems.
        </para>
        <para>
         If the value is specified without units, it is taken as milliseconds.
diff --git a/src/backend/libpq/pqcomm.c b/src/backend/libpq/pqcomm.c
@@ -204,7 +204,7 @@ pq_init(void)
 				(errmsg("could not set socket to nonblocking mode: %m")));
 #endif
 
-	FeBeWaitSet = CreateWaitEventSet(TopMemoryContext, 3);
+	FeBeWaitSet = CreateWaitEventSet(TopMemoryContext, FeBeWaitSetNEvents);
 	socket_pos = AddWaitEventToSet(FeBeWaitSet, WL_SOCKET_WRITEABLE,
 								   MyProcPort->sock, NULL, NULL);
 	latch_pos = AddWaitEventToSet(FeBeWaitSet, WL_LATCH_SET, PGINVALID_SOCKET,
@@ -1960,33 +1960,33 @@ pq_settcpusertimeout(int timeout, Port *port)
 bool
 pq_check_connection(void)
 {
-#if defined(POLLRDHUP)
-	/*
-	 * POLLRDHUP is a Linux extension to poll(2) to detect sockets closed by
-	 * the other end.  We don't have a portable way to do that without
-	 * actually trying to read or write data on other systems.  We don't want
-	 * to read because that would be confused by pipelined queries and COPY
-	 * data. Perhaps in future we'll try to write a heartbeat message instead.
-	 */
-	struct pollfd pollfd;
+	WaitEvent	events[FeBeWaitSetNEvents];
 	int			rc;
 
-	pollfd.fd = MyProcPort->sock;
-	pollfd.events = POLLOUT | POLLIN | POLLRDHUP;
-	pollfd.revents = 0;
-
-	rc = poll(&pollfd, 1, 0);
+	/*
+	 * It's OK to modify the socket event filter without restoring, because
+	 * all FeBeWaitSet socket wait sites do the same.
+	 */
+	ModifyWaitEvent(FeBeWaitSet, FeBeWaitSetSocketPos, WL_SOCKET_CLOSED, NULL);
 
-	if (rc < 0)
+retry:
+	rc = WaitEventSetWait(FeBeWaitSet, 0, events, lengthof(events), 0);
+	for (int i = 0; i < rc; ++i)
 	{
-		ereport(COMMERROR,
-				(errcode_for_socket_access(),
-				 errmsg("could not poll socket: %m")));
-		return false;
+		if (events[i].events & WL_SOCKET_CLOSED)
+			return false;
+		if (events[i].events & WL_LATCH_SET)
+		{
+			/*
+			 * A latch event might be preventing other events from being
+			 * reported.  Reset it and poll again.  No need to restore it
+			 * because no code should expect latches to survive across
+			 * CHECK_FOR_INTERRUPTS().
+			 */
+			 ResetLatch(MyLatch);
+			 goto retry;
+		}
 	}
-	else if (rc == 1 && (pollfd.revents & (POLLHUP | POLLRDHUP)))
-		return false;
-#endif
 
 	return true;
 }
diff --git a/src/backend/utils/misc/guc.c b/src/backend/utils/misc/guc.c
@@ -12192,14 +12192,11 @@ check_huge_page_size(int *newval, void **extra, GucSource source)
 static bool
 check_client_connection_check_interval(int *newval, void **extra, GucSource source)
 {
-#ifndef POLLRDHUP
-	/* Linux only, for now.  See pq_check_connection(). */
-	if (*newval != 0)
+	if (!WaitEventSetCanReportClosed() && *newval != 0)
 	{
-		GUC_check_errdetail("client_connection_check_interval must be set to 0 on platforms that lack POLLRDHUP.");
+		GUC_check_errdetail("client_connection_check_interval must be set to 0 on this platform");
 		return false;
 	}
-#endif
 	return true;
 }
 
diff --git a/src/include/libpq/libpq.h b/src/include/libpq/libpq.h
@@ -62,6 +62,7 @@ extern WaitEventSet *FeBeWaitSet;
 
 #define FeBeWaitSetSocketPos 0
 #define FeBeWaitSetLatchPos 1
+#define FeBeWaitSetNEvents 3
 
 extern int	StreamServerPort(int family, const char *hostName,
 							 unsigned short portNumber, const char *unixSocketDir,

Original file line number	Diff line number	Diff line change
`@@ -12192,14 +12192,11 @@ check_huge_page_size(int newval, void *extra, GucSource source)`
`12192`	`12192`	`static bool`
`12193`	`12193`	`check_client_connection_check_interval(int newval, void *extra, GucSource source)`
`12194`	`12194`	`{`
`12195`		`-#ifndef POLLRDHUP`
`12196`		`- /* Linux only, for now. See pq_check_connection(). */`
`12197`		`- if (*newval != 0)`
	`12195`	`+ if (!WaitEventSetCanReportClosed() && *newval != 0)`
`12198`	`12196`	`{`
`12199`		`- GUC_check_errdetail("client_connection_check_interval must be set to 0 on platforms that lack POLLRDHUP.");`
	`12197`	`+ GUC_check_errdetail("client_connection_check_interval must be set to 0 on this platform");`
`12200`	`12198`	`return false;`
`12201`	`12199`	`}`
`12202`		`-#endif`
`12203`	`12200`	`return true;`
`12204`	`12201`	`}`
`12205`	`12202`