[ECOS] new_net compilation failure

[ECOS] new_net compilation failure

[Jani Monoses]

now that sys/param.h includes errno.h which #defines errno
net/bsd_tcpip/src/sys/netinet/in_pcb.c won't build because some functions
have errno as a parameter name.That expands to the errno define and gives
several syntax errors.

I am just using the plain new_net template with no further config.

The solution would be either reverting the change and including errno.h only in the two files which needed
it or renaming the parameters.

the functions are in_rtchange & in_pcnbotifyall
Jani.

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

Re: [ECOS] new_net compilation failure

[Gary Thomas]

On Wed, 2002-07-10 at 08:08, Jani Monoses wrote:
> now that sys/param.h includes errno.h which #defines errno
> net/bsd_tcpip/src/sys/netinet/in_pcb.c won't build because some functions
> have errno as a parameter name.That expands to the errno define and gives
> several syntax errors.
> 
> I am just using the plain new_net template with no further config.
> 
> The solution would be either reverting the change and including errno.h only in the two files which needed
> it or renaming the parameters.
> 

I just put it there to be closer to the old stack behavior.

> the functions are in_rtchange & in_pcnbotifyall

So, why didn't I get any errors on these?  It seems that different
compilers don't behave the same.

Anyway, I'll commit this change which is better anyway.

Index: net/bsd_tcpip/current/ChangeLog
===================================================================
RCS file: /misc/cvsfiles/ecos-opt/net/net/bsd_tcpip/current/ChangeLog,v
retrieving revision 1.5
diff -u -5 -p -r1.5 ChangeLog
--- net/bsd_tcpip/current/ChangeLog	8 Jul 2002 20:54:12 -0000	1.5
+++ net/bsd_tcpip/current/ChangeLog	10 Jul 2002 12:20:08 -0000
@@ -1,5 +1,12 @@
+2002-07-10  Gary Thomas  <gary@chez-thomas.org>
+
+	* src/sys/netinet/udp_usrreq.c: 
+	* src/sys/netinet/tcp_subr.c: 
+	* src/sys/netinet/in_pcb.c: Rename 'errno' function parameter
+	since some compilers have problems with this.
+
 2002-07-08  Gary Thomas  <gary@chez-thomas.org>
 
 	* include/sys/param.h: Need <errno.h> which may or may not be
 	already present, depending on configuration.
 
Index: net/bsd_tcpip/current/src/sys/netinet/in_pcb.c
===================================================================
RCS file: /misc/cvsfiles/ecos-opt/net/net/bsd_tcpip/current/src/sys/netinet/in_pcb.c,v
retrieving revision 1.1
diff -u -5 -p -r1.1 in_pcb.c
--- net/bsd_tcpip/current/src/sys/netinet/in_pcb.c	20 May 2002 22:25:02 -0000	1.1
+++ net/bsd_tcpip/current/src/sys/netinet/in_pcb.c	10 Jul 2002 12:15:31 -0000
@@ -579,11 +579,11 @@ in_setpeeraddr(so, nam)
 	*nam = (struct sockaddr *)sin;
 	return 0;
 }
 
 void
-in_pcbnotifyall(head, faddr, errno, notify)
+in_pcbnotifyall(head, faddr, _errno, notify)
 	struct inpcbhead *head;
 	struct in_addr faddr;
 	void (*notify) __P((struct inpcb *, int));
 {
 	struct inpcb *inp, *ninp;
@@ -597,11 +597,11 @@ in_pcbnotifyall(head, faddr, errno, noti
 			continue;
 #endif
 		if (inp->inp_faddr.s_addr != faddr.s_addr ||
 		    inp->inp_socket == NULL)
 				continue;
-		(*notify)(inp, errno);
+		(*notify)(inp, _errno);
 	}
 	splx(s);
 }
 
 void
@@ -678,13 +678,13 @@ in_losing(inp)
 /*
  * After a routing change, flush old routing
  * and allocate a (hopefully) better one.
  */
 void
-in_rtchange(inp, errno)
+in_rtchange(inp, _errno)
 	register struct inpcb *inp;
-	int errno;
+	int _errno;
 {
 	if (inp->inp_route.ro_rt) {
 		rtfree(inp->inp_route.ro_rt);
 		inp->inp_route.ro_rt = 0;
 		/*
Index: net/bsd_tcpip/current/src/sys/netinet/tcp_subr.c
===================================================================
RCS file: /misc/cvsfiles/ecos-opt/net/net/bsd_tcpip/current/src/sys/netinet/tcp_subr.c,v
retrieving revision 1.1
diff -u -5 -p -r1.1 tcp_subr.c
--- net/bsd_tcpip/current/src/sys/netinet/tcp_subr.c	20 May 2002 22:25:02 -0000	1.1
+++ net/bsd_tcpip/current/src/sys/netinet/tcp_subr.c	10 Jul 2002 12:16:59 -0000
@@ -531,25 +531,25 @@ tcp_newtcpcb(inp)
  * Drop a TCP connection, reporting
  * the specified error.  If connection is synchronized,
  * then send a RST to peer.
  */
 struct tcpcb *
-tcp_drop(tp, errno)
+tcp_drop(tp, _errno)
 	register struct tcpcb *tp;
-	int errno;
+	int _errno;
 {
 	struct socket *so = tp->t_inpcb->inp_socket;
 
 	if (TCPS_HAVERCVDSYN(tp->t_state)) {
 		tp->t_state = TCPS_CLOSED;
 		(void) tcp_output(tp);
 		tcpstat.tcps_drops++;
 	} else
 		tcpstat.tcps_conndrops++;
-	if (errno == ETIMEDOUT && tp->t_softerror)
-		errno = tp->t_softerror;
-	so->so_error = errno;
+	if (_errno == ETIMEDOUT && tp->t_softerror)
+		_errno = tp->t_softerror;
+	so->so_error = _errno;
 	return (tcp_close(tp));
 }
 
 /*
  * Close a TCP control block:
@@ -1000,13 +1000,13 @@ tcp_new_isn(tp)
 /*
  * When a source quench is received, close congestion window
  * to one segment.  We will gradually open it again as we proceed.
  */
 void
-tcp_quench(inp, errno)
+tcp_quench(inp, _errno)
 	struct inpcb *inp;
-	int errno;
+	int _errno;
 {
 	struct tcpcb *tp = intotcpcb(inp);
 
 	if (tp)
 		tp->snd_cwnd = tp->t_maxseg;
@@ -1016,30 +1016,30 @@ tcp_quench(inp, errno)
  * When a specific ICMP unreachable message is received and the
  * connection state is SYN-SENT, drop the connection.  This behavior
  * is controlled by the icmp_may_rst sysctl.
  */
 void
-tcp_drop_syn_sent(inp, errno)
+tcp_drop_syn_sent(inp, _errno)
 	struct inpcb *inp;
-	int errno;
+	int _errno;
 {
 	struct tcpcb *tp = intotcpcb(inp);
 
 	if (tp && tp->t_state == TCPS_SYN_SENT)
-		tcp_drop(tp, errno);
+		tcp_drop(tp, _errno);
 }
 
 /*
  * When `need fragmentation' ICMP is received, update our idea of the MSS
  * based on the new value in the route.  Also nudge TCP to send something,
  * since we know the packet we just sent was dropped.
  * This duplicates some code in the tcp_mss() function in tcp_input.c.
  */
 void
-tcp_mtudisc(inp, errno)
+tcp_mtudisc(inp, _errno)
 	struct inpcb *inp;
-	int errno;
+	int _errno;
 {
 	struct tcpcb *tp = intotcpcb(inp);
 	struct rtentry *rt;
 	struct rmxp_tao *taop;
 	struct socket *so = inp->inp_socket;
Index: net/bsd_tcpip/current/src/sys/netinet/udp_usrreq.c
===================================================================
RCS file: /misc/cvsfiles/ecos-opt/net/net/bsd_tcpip/current/src/sys/netinet/udp_usrreq.c,v
retrieving revision 1.1
diff -u -5 -p -r1.1 udp_usrreq.c
--- net/bsd_tcpip/current/src/sys/netinet/udp_usrreq.c	20 May 2002 22:25:02 -0000	1.1
+++ net/bsd_tcpip/current/src/sys/netinet/udp_usrreq.c	10 Jul 2002 12:16:03 -0000
@@ -512,15 +512,15 @@ udp_append(last, ip, n, off)
 /*
  * Notify a udp user of an asynchronous error;
  * just wake up so that he can collect error status.
  */
 void
-udp_notify(inp, errno)
+udp_notify(inp, _errno)
 	register struct inpcb *inp;
-	int errno;
+	int _errno;
 {
-	inp->inp_socket->so_error = errno;
+	inp->inp_socket->so_error = _errno;
 	sorwakeup(inp->inp_socket);
 	sowwakeup(inp->inp_socket);
 }
 
 void


-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

Re: [ECOS] new_net compilation failure

[Wolfgang Heppner]

Gary Thomas wrote:
> 
> ...
> 
> So, why didn't I get any errors on these?  It seems that different
> compilers don't behave the same.
> 

What compiler are you using?
I got the same problem with gcc 2.95.2, configured for powerpc-eabi.


> Anyway, I'll commit this change which is better anyway.
> 

Thanks,
Wolfgang

-- 
Wolfgang Heppner

High Speed Kamerasysteme  - Abt. Bildsensorik (ESY-B)
High Speed Camera Systems - Dept. Electronic Imaging (ESY-B)

Fraunhofer                            
Institut Integrierte Schaltungen
Angewandte Elektronik

Am Wolfsmantel 33 
91058 Erlangen       
(Germany)

Tel:   +49 (0) 9131 776 528
Fax:   +49 (0) 9131 776 598

email: hep@iis.fhg.de
web:   http://www.iis.fhg.de

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

Re: [ECOS] new_net compilation failure

[Jani Monoses]

> So, why didn't I get any errors on these?  It seems that different
> compilers don't behave the same.

Ok with this patch it builds.

gcc version 2.95.3 20010315 (release)

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

Re: [ECOS] new_net compilation failure

[Jani Monoses]

> So, why didn't I get any errors on these?  It seems that different
> compilers don't behave the same.

Maybe some installations of gcc have different paths for sys/params.h ?????
I don't know.

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss