Change the curvnet variable from a global const struct vnet *,

previously always pointing to the default vnet context, to a
dynamically changing thread-local one.  The currvnet context
should be set on entry to networking code via CURVNET_SET() macros,
and reverted to previous state via CURVNET_RESTORE().  Recursions
on curvnet are permitted, though strongly discuouraged.

This change should have no functional impact on nooptions VIMAGE
kernel builds, where CURVNET_* macros expand to whitespace.

The curthread->td_vnet (aka curvnet) variable's purpose is to be an
indicator of the vnet context in which the current network-related
operation takes place, in case we cannot deduce the current vnet
context from any other source, such as by looking at mbuf's
m->m_pkthdr.rcvif->if_vnet, sockets's so->so_vnet etc.  Moreover, so
far curvnet has turned out to be an invaluable consistency checking
aid: it helps to catch cases when sockets, ifnets or any other
vnet-aware structures may have leaked from one vnet to another.

The exact placement of the CURVNET_SET() / CURVNET_RESTORE() macros
was a result of an empirical iterative process, whith an aim to
reduce recursions on CURVNET_SET() to a minimum, while still reducing
the scope of CURVNET_SET() to networking only operations - the
alternative would be calling CURVNET_SET() on each system call entry.
In general, curvnet has to be set in three typicall cases: when
processing socket-related requests from userspace or from within the
kernel; when processing inbound traffic flowing from device drivers
to upper layers of the networking stack, and when executing
timer-driven networking functions.

This change also introduces a DDB subcommand to show the list of all
vnet instances.

Approved by:	julian (mentor)

5 files changed, 14 insertions, 9 deletions

diff --git a/sys/netinet/if_ether.c b/sys/netinet/if_ether.c
index 38e420e..d24b459 100644
--- a/sys/netinet/if_ether.c
+++ b/sys/netinet/if_ether.c
@@ -143,10 +143,12 @@ arp_ifscrub(struct ifnet *ifp, uint32_t addr)
 	addr4.sin_len    = sizeof(addr4);
 	addr4.sin_family = AF_INET;
 	addr4.sin_addr.s_addr = addr;
+	CURVNET_SET(ifp->if_vnet);
 	IF_AFDATA_LOCK(ifp);
 	lla_lookup(LLTABLE(ifp), (LLE_DELETE | LLE_IFADDR),
 	    (struct sockaddr *)&addr4);
 	IF_AFDATA_UNLOCK(ifp);
+	CURVNET_RESTORE();
 }
 #endif
 
diff --git a/sys/netinet/igmp.c b/sys/netinet/igmp.c
index 2bbd10d..3fb3711 100644
--- a/sys/netinet/igmp.c
+++ b/sys/netinet/igmp.c
@@ -1117,6 +1117,9 @@ igmp_input_v3_group_query(struct in_multi *inm, struct igmp_ifinfo *igi,
 
 	nsrc = ntohs(igmpv3->igmp_numsrc);
 
+	if (!IS_DEFAULT_VNET(curvnet))
+		return (retval);
+
 	/*
 	 * Deal with group-specific queries upfront.
 	 * If any group query is already pending, purge any recorded
@@ -3372,7 +3375,7 @@ igmp_intr(struct mbuf *m)
 	 * indexes to guard against interface detach, they are
 	 * unique to each VIMAGE and must be retrieved.
 	 */
-	CURVNET_SET(m->m_pkthdr.header);
+	CURVNET_SET((struct vnet *)(m->m_pkthdr.header));
 	INIT_VNET_NET(curvnet);
 	INIT_VNET_INET(curvnet);
 	ifindex = igmp_restore_context(m);
@@ -3654,9 +3657,7 @@ igmp_modevent(module_t mod, int type, void *unused __unused)
 	break;
     case MOD_UNLOAD:
 #ifndef VIMAGE_GLOBALS
-#ifdef NOTYET
 	vnet_mod_deregister(&vnet_igmp_modinfo);
-#endif
 #else
 	vnet_igmp_idetach(NULL);
 #endif
diff --git a/sys/netinet/in_rmx.c b/sys/netinet/in_rmx.c
index 695f473..19cd5fb 100644
--- a/sys/netinet/in_rmx.c
+++ b/sys/netinet/in_rmx.c
@@ -250,14 +250,13 @@ static void in_rtqtimo_one(void *rock);
 static void
 in_rtqtimo(void *rock)
 {
+	CURVNET_SET((struct vnet *) rock);
 	INIT_VNET_NET(curvnet);
 	INIT_VNET_INET(curvnet);
 	int fibnum;
 	void *newrock;
 	struct timeval atv;
 
-	KASSERT((rock == (void *)V_rt_tables[0][AF_INET]),
-			("in_rtqtimo: unexpected arg"));
 	for (fibnum = 0; fibnum < rt_numfibs; fibnum++) {
 		if ((newrock = V_rt_tables[fibnum][AF_INET]) != NULL)
 			in_rtqtimo_one(newrock);
@@ -265,6 +264,7 @@ in_rtqtimo(void *rock)
 	atv.tv_usec = 0;
 	atv.tv_sec = V_rtq_timeout;
 	callout_reset(&V_rtq_timer, tvtohz(&atv), in_rtqtimo, rock);
+	CURVNET_RESTORE();
 }
 
 static void
@@ -377,7 +377,7 @@ in_inithead(void **head, int off)
 	rnh->rnh_close = in_clsroute;
 	if (_in_rt_was_here == 0 ) {
 		callout_init(&V_rtq_timer, CALLOUT_MPSAFE);
-		in_rtqtimo(rnh);	/* kick off timeout first time */
+		callout_reset(&V_rtq_timer, 1, in_rtqtimo, curvnet);
 		_in_rt_was_here = 1;
 	}
 	return 1;
diff --git a/sys/netinet/ip_input.c b/sys/netinet/ip_input.c
index 7ee5a13..973c392 100644
--- a/sys/netinet/ip_input.c
+++ b/sys/netinet/ip_input.c
@@ -339,7 +339,7 @@ ip_init(void)
 
 	/* Start ipport_tick. */
 	callout_init(&ipport_tick_callout, CALLOUT_MPSAFE);
-	ipport_tick(NULL);
+	callout_reset(&ipport_tick_callout, 1, ipport_tick, NULL);
 	EVENTHANDLER_REGISTER(shutdown_pre_sync, ip_fini, NULL,
 		SHUTDOWN_PRI_DEFAULT);
 	EVENTHANDLER_REGISTER(nmbclusters_change, ipq_zone_change,
diff --git a/sys/netinet/tcp_hostcache.c b/sys/netinet/tcp_hostcache.c
index 71f4626..84586db 100644
--- a/sys/netinet/tcp_hostcache.c
+++ b/sys/netinet/tcp_hostcache.c
@@ -227,7 +227,7 @@ tcp_hc_init(void)
 	 */
 	callout_init(&V_tcp_hc_callout, CALLOUT_MPSAFE);
 	callout_reset(&V_tcp_hc_callout, V_tcp_hostcache.prune * hz,
-	    tcp_hc_purge, 0);
+	    tcp_hc_purge, curvnet);
 }
 
 /*
@@ -634,9 +634,10 @@ sysctl_tcp_hc_list(SYSCTL_HANDLER_ARGS)
 static void
 tcp_hc_purge(void *arg)
 {
+	CURVNET_SET((struct vnet *) arg);
 	INIT_VNET_INET(curvnet);
 	struct hc_metrics *hc_entry, *hc_next;
-	int all = (intptr_t)arg;
+	int all = 0;	/* XXX was: (intptr_t)arg - makes no sense? */
 	int i;
 
 	if (V_tcp_hostcache.purgeall) {
@@ -662,4 +663,5 @@ tcp_hc_purge(void *arg)
 
 	callout_reset(&V_tcp_hc_callout, V_tcp_hostcache.prune * hz,
 	    tcp_hc_purge, arg);
+	CURVNET_RESTORE();
 }