From owner-freebsd-arch@FreeBSD.ORG  Tue Jul 15 09:00:34 2003
Return-Path: <owner-freebsd-arch@FreeBSD.ORG>
Delivered-To: freebsd-arch@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id B094437B404
	for <arch@freebsd.org>; Tue, 15 Jul 2003 09:00:34 -0700 (PDT)
Received: from comp.chem.msu.su (comp-ext.chem.msu.su [158.250.32.157])
	by mx1.FreeBSD.org (Postfix) with ESMTP id 0CA4843FA3
	for <arch@freebsd.org>; Tue, 15 Jul 2003 09:00:31 -0700 (PDT)
	(envelope-from yar@comp.chem.msu.su)
Received: from comp.chem.msu.su (localhost [127.0.0.1])
	by comp.chem.msu.su (8.12.3p2/8.12.3) with ESMTP id h6FG0Pgb044336
	for <arch@freebsd.org>; Tue, 15 Jul 2003 20:00:25 +0400 (MSD)
	(envelope-from yar@comp.chem.msu.su)
Received: (from yar@localhost)
	by comp.chem.msu.su (8.12.3p2/8.12.3/Submit) id h6FG0P0F044332
	for arch@freebsd.org; Tue, 15 Jul 2003 20:00:25 +0400 (MSD)
	(envelope-from yar)
Date: Tue, 15 Jul 2003 20:00:25 +0400
From: Yar Tikhiy <yar@freebsd.org>
To: arch@freebsd.org
Message-ID: <20030715160024.GA43108@comp.chem.msu.su>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.3i
Subject: Interrupted connect(2) and errno issue
X-BeenThere: freebsd-arch@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Discussion related to FreeBSD architecture
	<freebsd-arch.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-arch>,
	<mailto:freebsd-arch-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-arch>
List-Post: <mailto:freebsd-arch@freebsd.org>
List-Help: <mailto:freebsd-arch-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-arch>,
	<mailto:freebsd-arch-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Tue, 15 Jul 2003 16:00:35 -0000

Hi folks,

Background:
A blocking connect(2) call will be interrupted by any delivered
signal, returning EINTR.  Despites the call interrupted, the
connection establishment will be carried out asynchronously, as
though the socket were in non-blocking mode.

The issue:
Further calls to connect(2) on such a socket will return EADDRINUSE,
while those on a non-blocking socket with connecting in progress
will return EALREADY.  Is there a reason for this difference?  If
there isn't, what do you think of the attached patch?  I believe
it will improve our conformance to POSIX.

BTW, this patch could have been cleaner and not involved a new flag
variable if we were able to drop the splx() stuff, which is anyway
a no-op in CURRENT AFAIK.

-- 
Yar

--- uipc_syscalls.c.orig	Thu Jun 19 07:55:01 2003
+++ uipc_syscalls.c	Tue Jul 15 17:03:25 2003
@@ -480,11 +480,12 @@
 {
 	struct socket *so;
 	int error, s;
+	int interrupted = 0;
 
 	mtx_lock(&Giant);
 	if ((error = fgetsock(td, fd, &so, NULL)) != 0)
 		goto done2;
-	if ((so->so_state & SS_NBIO) && (so->so_state & SS_ISCONNECTING)) {
+	if (so->so_state & SS_ISCONNECTING) {
 		error = EALREADY;
 		goto done1;
 	}
@@ -503,6 +504,8 @@
 	s = splnet();
 	while ((so->so_state & SS_ISCONNECTING) && so->so_error == 0) {
 		error = tsleep(&so->so_timeo, PSOCK | PCATCH, "connec", 0);
+		if (error == EINTR || error == ERESTART)
+			interrupted = 1;
 		if (error)
 			break;
 	}
@@ -512,7 +515,8 @@
 	}
 	splx(s);
 bad:
-	so->so_state &= ~SS_ISCONNECTING;
+	if (!interrupted)
+		so->so_state &= ~SS_ISCONNECTING;
 	if (error == ERESTART)
 		error = EINTR;
 done1:
%%%