RTNL assert

RTNL assert

[Stephen E. Clark]

When I configure in Tunneling I get the following error message. Is this
normal?

GRE over IPv4 tunneling driver
RTNL: assertion failed at devinet.c(775):inetdev_event
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

RTNL assert

[Stephen E. Clark]

When I configure in Tunneling I get the following error message. Is this
normal? This with 2.4test9pre5

GRE over IPv4 tunneling driver
RTNL: assertion failed at devinet.c(775):inetdev_event
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: RTNL assert

[Andrew Morton]

"Stephen E. Clark" wrote:
> 
> When I configure in Tunneling I get the following error message. Is this
> normal? This with 2.4test9pre5
> 
> GRE over IPv4 tunneling driver
> RTNL: assertion failed at devinet.c(775):inetdev_event

The rtnetlink lock needs to be taken around
register_netdevice().  There should be a function
which does these three common steps, but there isn't.


--- linux-2.4.0-test10-pre5/net/ipv4/ip_gre.c	Sat Sep  9 16:19:30 2000
+++ linux-akpm/net/ipv4/ip_gre.c	Sat Oct 28 21:44:23 2000
@@ -1266,7 +1266,9 @@
 #ifdef MODULE
 	register_netdev(&ipgre_fb_tunnel_dev);
 #else
+	rtnl_lock();
 	register_netdevice(&ipgre_fb_tunnel_dev);
+	rtnl_unlock();
 #endif
 
 	inet_add_protocol(&ipgre_protocol);
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: RTNL assert

[Stephen E. Clark]

Andrew Morton wrote:
> 
> "Stephen E. Clark" wrote:
> >
> > When I configure in Tunneling I get the following error message. Is this
> > normal? This with 2.4test9pre5
> >
> > GRE over IPv4 tunneling driver
> > RTNL: assertion failed at devinet.c(775):inetdev_event
> 
> The rtnetlink lock needs to be taken around
> register_netdevice().  There should be a function
> which does these three common steps, but there isn't.
> 
> --- linux-2.4.0-test10-pre5/net/ipv4/ip_gre.c   Sat Sep  9 16:19:30 2000
> +++ linux-akpm/net/ipv4/ip_gre.c        Sat Oct 28 21:44:23 2000
> @@ -1266,7 +1266,9 @@
>  #ifdef MODULE
>         register_netdev(&ipgre_fb_tunnel_dev);
>  #else
> +       rtnl_lock();
>         register_netdevice(&ipgre_fb_tunnel_dev);
> +       rtnl_unlock();
>  #endif
> 
>         inet_add_protocol(&ipgre_protocol);
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> Please read the FAQ at http://www.tux.org/lkml/

Thanks Andrew,

I also get the same error if I try to configure in normal IPV4
tunneling. I guess it needs the same kind of patch.

Oct 27 14:46:59 joker kernel: IPv4 over IPv4 tunneling driver 
Oct 27 14:46:59 joker kernel: RTNL: assertion failed at
devinet.c(775):inetdev_event 
Oct 27 14:46:59 joker kernel: GRE over IPv4 tunneling driver 
Oct 27 14:46:59 joker kernel: RTNL: assertion failed at
devinet.c(775):inetdev_event

Steve
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: RTNL assert

[Andrew Morton]

"Stephen E. Clark" wrote:
> 
> I also get the same error if I try to configure in normal IPV4
> tunneling. I guess it needs the same kind of patch.

Yep.

--- linux-2.4.0-test10-pre5/net/ipv4/ipip.c	Sat Sep  9 16:19:30 2000
+++ linux-akpm/net/ipv4/ipip.c	Sun Oct 29 03:17:38 2000
@@ -894,7 +894,9 @@
 #ifdef MODULE
 	register_netdev(&ipip_fb_tunnel_dev);
 #else
+	rtnl_lock();
 	register_netdevice(&ipip_fb_tunnel_dev);
+	rtnl_unlock();
 #endif
 
 	inet_add_protocol(&ipip_protocol);
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: RTNL assert

[David S. Miller]

There is even one more Andrew :-)  Here is the full patch:

--- ./net/ipv4/ip_gre.c.~1~	Thu Aug 24 18:48:54 2000
+++ ./net/ipv4/ip_gre.c	Sat Oct 28 09:59:43 2000
@@ -1266,7 +1266,9 @@
 #ifdef MODULE
 	register_netdev(&ipgre_fb_tunnel_dev);
 #else
+	rtnl_lock();
 	register_netdevice(&ipgre_fb_tunnel_dev);
+	rtnl_unlock();
 #endif
 
 	inet_add_protocol(&ipgre_protocol);
--- ./net/ipv4/ipip.c.~1~	Thu Aug 24 19:15:47 2000
+++ ./net/ipv4/ipip.c	Sat Oct 28 10:00:04 2000
@@ -894,7 +894,9 @@
 #ifdef MODULE
 	register_netdev(&ipip_fb_tunnel_dev);
 #else
+	rtnl_lock();
 	register_netdevice(&ipip_fb_tunnel_dev);
+	rtnl_unlock();
 #endif
 
 	inet_add_protocol(&ipip_protocol);
--- ./net/ipv6/sit.c.~1~	Mon Oct  9 21:36:50 2000
+++ ./net/ipv6/sit.c	Sat Oct 28 10:00:50 2000
@@ -829,7 +829,9 @@
 #ifdef MODULE
 	register_netdev(&ipip6_fb_tunnel_dev);
 #else
+	rtnl_lock();
 	register_netdevice(&ipip6_fb_tunnel_dev);
+	rtnl_unlock();
 #endif
 	inet_add_protocol(&sit_protocol);
 	return 0;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: RTNL assert

[David Woodhouse]

On Sat, 28 Oct 2000, Andrew Morton wrote:

> The rtnetlink lock needs to be taken around
> register_netdevice().  There should be a function
> which does these three common steps, but there isn't.

I thought the only difference between register_netdev() and
register_netdevice() was that one took the rtnl_lock and the other didn't?

But it's been a while since I played with networking code.

-- 
dwmw2


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: RTNL assert

[Jeff Garzik]

David Woodhouse wrote:
> 
> On Sat, 28 Oct 2000, Andrew Morton wrote:
> 
> > The rtnetlink lock needs to be taken around
> > register_netdevice().  There should be a function
> > which does these three common steps, but there isn't.
> 
> I thought the only difference between register_netdev() and
> register_netdevice() was that one took the rtnl_lock and the other didn't?

And, register_netdev allocates a name for you if necessary...

-- 
Jeff Garzik             | "Mind if I drive?"  -Sam
Building 1024           | "Not if you don't mind me clawing at the
MandrakeSoft            |  dash and shrieking like a cheerleader."
                        |                     -Max
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/