freebsd-skq/sys/netinet/in_rmx.c
delphij f0a3248def Add a bandaid for a long-standing race condition during route entry
un-expiring.

The previous version of code have no locking when testing rt_refcnt.
The result of the lack of locking may result in a condition where
a routing entry have a reference count but at the same time have
RTPRF_OURS bit set and an expiration timer.  These would eventually
lead to a panic:

	panic: rtqkill route really not free

When the system have ICMP redirects accepted from local gateway
in a moderate frequency, for instance.

Commit this workaround for now until we have some better solution.

PR:		kern/149804
Reviewed by:	bz
Tested by:	Zhao Xin, Pete French
MFC after:	2 weeks
2010-09-27 19:26:56 +00:00

515 lines
14 KiB
C

/*-
* Copyright 1994, 1995 Massachusetts Institute of Technology
*
* Permission to use, copy, modify, and distribute this software and
* its documentation for any purpose and without fee is hereby
* granted, provided that both the above copyright notice and this
* permission notice appear in all copies, that both the above
* copyright notice and this permission notice appear in all
* supporting documentation, and that the name of M.I.T. not be used
* in advertising or publicity pertaining to distribution of the
* software without specific, written prior permission. M.I.T. makes
* no representations about the suitability of this software for any
* purpose. It is provided "as is" without express or implied
* warranty.
*
* THIS SOFTWARE IS PROVIDED BY M.I.T. ``AS IS''. M.I.T. DISCLAIMS
* ALL EXPRESS OR IMPLIED WARRANTIES WITH REGARD TO THIS SOFTWARE,
* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT
* SHALL M.I.T. BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
* USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
* OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
/*
* This code does two things necessary for the enhanced TCP metrics to
* function in a useful manner:
* 1) It marks all non-host routes as `cloning', thus ensuring that
* every actual reference to such a route actually gets turned
* into a reference to a host route to the specific destination
* requested.
* 2) When such routes lose all their references, it arranges for them
* to be deleted in some random collection of circumstances, so that
* a large quantity of stale routing data is not kept in kernel memory
* indefinitely. See in_rtqtimo() below for the exact mechanism.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/sysctl.h>
#include <sys/socket.h>
#include <sys/mbuf.h>
#include <sys/syslog.h>
#include <sys/callout.h>
#include <net/if.h>
#include <net/route.h>
#include <net/vnet.h>
#include <netinet/in.h>
#include <netinet/in_var.h>
#include <netinet/ip_var.h>
extern int in_inithead(void **head, int off);
#ifdef VIMAGE
extern int in_detachhead(void **head, int off);
#endif
#define RTPRF_OURS RTF_PROTO3 /* set on routes we manage */
/*
* Do what we need to do when inserting a route.
*/
static struct radix_node *
in_addroute(void *v_arg, void *n_arg, struct radix_node_head *head,
struct radix_node *treenodes)
{
struct rtentry *rt = (struct rtentry *)treenodes;
struct sockaddr_in *sin = (struct sockaddr_in *)rt_key(rt);
RADIX_NODE_HEAD_WLOCK_ASSERT(head);
/*
* A little bit of help for both IP output and input:
* For host routes, we make sure that RTF_BROADCAST
* is set for anything that looks like a broadcast address.
* This way, we can avoid an expensive call to in_broadcast()
* in ip_output() most of the time (because the route passed
* to ip_output() is almost always a host route).
*
* We also do the same for local addresses, with the thought
* that this might one day be used to speed up ip_input().
*
* We also mark routes to multicast addresses as such, because
* it's easy to do and might be useful (but this is much more
* dubious since it's so easy to inspect the address).
*/
if (rt->rt_flags & RTF_HOST) {
if (in_broadcast(sin->sin_addr, rt->rt_ifp)) {
rt->rt_flags |= RTF_BROADCAST;
} else if (satosin(rt->rt_ifa->ifa_addr)->sin_addr.s_addr ==
sin->sin_addr.s_addr) {
rt->rt_flags |= RTF_LOCAL;
}
}
if (IN_MULTICAST(ntohl(sin->sin_addr.s_addr)))
rt->rt_flags |= RTF_MULTICAST;
if (!rt->rt_rmx.rmx_mtu && rt->rt_ifp)
rt->rt_rmx.rmx_mtu = rt->rt_ifp->if_mtu;
return (rn_addroute(v_arg, n_arg, head, treenodes));
}
/*
* This code is the inverse of in_clsroute: on first reference, if we
* were managing the route, stop doing so and set the expiration timer
* back off again.
*/
static struct radix_node *
in_matroute(void *v_arg, struct radix_node_head *head)
{
struct radix_node *rn = rn_match(v_arg, head);
struct rtentry *rt = (struct rtentry *)rn;
if (rt) {
RT_LOCK(rt);
if (rt->rt_flags & RTPRF_OURS) {
rt->rt_flags &= ~RTPRF_OURS;
rt->rt_rmx.rmx_expire = 0;
}
RT_UNLOCK(rt);
}
return rn;
}
static VNET_DEFINE(int, rtq_reallyold) = 60*60; /* one hour is "really old" */
#define V_rtq_reallyold VNET(rtq_reallyold)
SYSCTL_VNET_INT(_net_inet_ip, IPCTL_RTEXPIRE, rtexpire, CTLFLAG_RW,
&VNET_NAME(rtq_reallyold), 0,
"Default expiration time on dynamically learned routes");
/* never automatically crank down to less */
static VNET_DEFINE(int, rtq_minreallyold) = 10;
#define V_rtq_minreallyold VNET(rtq_minreallyold)
SYSCTL_VNET_INT(_net_inet_ip, IPCTL_RTMINEXPIRE, rtminexpire, CTLFLAG_RW,
&VNET_NAME(rtq_minreallyold), 0,
"Minimum time to attempt to hold onto dynamically learned routes");
/* 128 cached routes is "too many" */
static VNET_DEFINE(int, rtq_toomany) = 128;
#define V_rtq_toomany VNET(rtq_toomany)
SYSCTL_VNET_INT(_net_inet_ip, IPCTL_RTMAXCACHE, rtmaxcache, CTLFLAG_RW,
&VNET_NAME(rtq_toomany), 0,
"Upper limit on dynamically learned routes");
/*
* On last reference drop, mark the route as belong to us so that it can be
* timed out.
*/
static void
in_clsroute(struct radix_node *rn, struct radix_node_head *head)
{
struct rtentry *rt = (struct rtentry *)rn;
RT_LOCK_ASSERT(rt);
if (!(rt->rt_flags & RTF_UP))
return; /* prophylactic measures */
if (rt->rt_flags & RTPRF_OURS)
return;
if (!(rt->rt_flags & RTF_DYNAMIC))
return;
/*
* If rtq_reallyold is 0, just delete the route without
* waiting for a timeout cycle to kill it.
*/
if (V_rtq_reallyold != 0) {
rt->rt_flags |= RTPRF_OURS;
rt->rt_rmx.rmx_expire = time_uptime + V_rtq_reallyold;
} else {
rtexpunge(rt);
}
}
struct rtqk_arg {
struct radix_node_head *rnh;
int draining;
int killed;
int found;
int updating;
time_t nextstop;
};
/*
* Get rid of old routes. When draining, this deletes everything, even when
* the timeout is not expired yet. When updating, this makes sure that
* nothing has a timeout longer than the current value of rtq_reallyold.
*/
static int
in_rtqkill(struct radix_node *rn, void *rock)
{
struct rtqk_arg *ap = rock;
struct rtentry *rt = (struct rtentry *)rn;
int err;
RADIX_NODE_HEAD_WLOCK_ASSERT(ap->rnh);
if (rt->rt_flags & RTPRF_OURS) {
ap->found++;
if (ap->draining || rt->rt_rmx.rmx_expire <= time_uptime) {
if (rt->rt_refcnt > 0)
panic("rtqkill route really not free");
err = in_rtrequest(RTM_DELETE,
(struct sockaddr *)rt_key(rt),
rt->rt_gateway, rt_mask(rt),
rt->rt_flags | RTF_RNH_LOCKED, 0,
rt->rt_fibnum);
if (err) {
log(LOG_WARNING, "in_rtqkill: error %d\n", err);
} else {
ap->killed++;
}
} else {
if (ap->updating &&
(rt->rt_rmx.rmx_expire - time_uptime >
V_rtq_reallyold)) {
rt->rt_rmx.rmx_expire =
time_uptime + V_rtq_reallyold;
}
ap->nextstop = lmin(ap->nextstop,
rt->rt_rmx.rmx_expire);
}
}
return 0;
}
#define RTQ_TIMEOUT 60*10 /* run no less than once every ten minutes */
static VNET_DEFINE(int, rtq_timeout) = RTQ_TIMEOUT;
static VNET_DEFINE(struct callout, rtq_timer);
#define V_rtq_timeout VNET(rtq_timeout)
#define V_rtq_timer VNET(rtq_timer)
static void in_rtqtimo_one(void *rock);
static void
in_rtqtimo(void *rock)
{
CURVNET_SET((struct vnet *) rock);
int fibnum;
void *newrock;
struct timeval atv;
for (fibnum = 0; fibnum < rt_numfibs; fibnum++) {
newrock = rt_tables_get_rnh(fibnum, AF_INET);
if (newrock != NULL)
in_rtqtimo_one(newrock);
}
atv.tv_usec = 0;
atv.tv_sec = V_rtq_timeout;
callout_reset(&V_rtq_timer, tvtohz(&atv), in_rtqtimo, rock);
CURVNET_RESTORE();
}
static void
in_rtqtimo_one(void *rock)
{
struct radix_node_head *rnh = rock;
struct rtqk_arg arg;
static time_t last_adjusted_timeout = 0;
arg.found = arg.killed = 0;
arg.rnh = rnh;
arg.nextstop = time_uptime + V_rtq_timeout;
arg.draining = arg.updating = 0;
RADIX_NODE_HEAD_LOCK(rnh);
rnh->rnh_walktree(rnh, in_rtqkill, &arg);
RADIX_NODE_HEAD_UNLOCK(rnh);
/*
* Attempt to be somewhat dynamic about this:
* If there are ``too many'' routes sitting around taking up space,
* then crank down the timeout, and see if we can't make some more
* go away. However, we make sure that we will never adjust more
* than once in rtq_timeout seconds, to keep from cranking down too
* hard.
*/
if ((arg.found - arg.killed > V_rtq_toomany) &&
(time_uptime - last_adjusted_timeout >= V_rtq_timeout) &&
V_rtq_reallyold > V_rtq_minreallyold) {
V_rtq_reallyold = 2 * V_rtq_reallyold / 3;
if (V_rtq_reallyold < V_rtq_minreallyold) {
V_rtq_reallyold = V_rtq_minreallyold;
}
last_adjusted_timeout = time_uptime;
#ifdef DIAGNOSTIC
log(LOG_DEBUG, "in_rtqtimo: adjusted rtq_reallyold to %d\n",
V_rtq_reallyold);
#endif
arg.found = arg.killed = 0;
arg.updating = 1;
RADIX_NODE_HEAD_LOCK(rnh);
rnh->rnh_walktree(rnh, in_rtqkill, &arg);
RADIX_NODE_HEAD_UNLOCK(rnh);
}
}
void
in_rtqdrain(void)
{
VNET_ITERATOR_DECL(vnet_iter);
struct radix_node_head *rnh;
struct rtqk_arg arg;
int fibnum;
VNET_LIST_RLOCK_NOSLEEP();
VNET_FOREACH(vnet_iter) {
CURVNET_SET(vnet_iter);
for ( fibnum = 0; fibnum < rt_numfibs; fibnum++) {
rnh = rt_tables_get_rnh(fibnum, AF_INET);
arg.found = arg.killed = 0;
arg.rnh = rnh;
arg.nextstop = 0;
arg.draining = 1;
arg.updating = 0;
RADIX_NODE_HEAD_LOCK(rnh);
rnh->rnh_walktree(rnh, in_rtqkill, &arg);
RADIX_NODE_HEAD_UNLOCK(rnh);
}
CURVNET_RESTORE();
}
VNET_LIST_RUNLOCK_NOSLEEP();
}
static int _in_rt_was_here;
/*
* Initialize our routing tree.
*/
int
in_inithead(void **head, int off)
{
struct radix_node_head *rnh;
/* XXX MRT
* This can be called from vfs_export.c too in which case 'off'
* will be 0. We know the correct value so just use that and
* return directly if it was 0.
* This is a hack that replaces an even worse hack on a bad hack
* on a bad design. After RELENG_7 this should be fixed but that
* will change the ABI, so for now do it this way.
*/
if (!rn_inithead(head, 32))
return 0;
if (off == 0) /* XXX MRT see above */
return 1; /* only do the rest for a real routing table */
rnh = *head;
rnh->rnh_addaddr = in_addroute;
rnh->rnh_matchaddr = in_matroute;
rnh->rnh_close = in_clsroute;
if (_in_rt_was_here == 0 ) {
callout_init(&V_rtq_timer, CALLOUT_MPSAFE);
callout_reset(&V_rtq_timer, 1, in_rtqtimo, curvnet);
_in_rt_was_here = 1;
}
return 1;
}
#ifdef VIMAGE
int
in_detachhead(void **head, int off)
{
callout_drain(&V_rtq_timer);
return (1);
}
#endif
/*
* This zaps old routes when the interface goes down or interface
* address is deleted. In the latter case, it deletes static routes
* that point to this address. If we don't do this, we may end up
* using the old address in the future. The ones we always want to
* get rid of are things like ARP entries, since the user might down
* the interface, walk over to a completely different network, and
* plug back in.
*/
struct in_ifadown_arg {
struct ifaddr *ifa;
int del;
};
static int
in_ifadownkill(struct radix_node *rn, void *xap)
{
struct in_ifadown_arg *ap = xap;
struct rtentry *rt = (struct rtentry *)rn;
RT_LOCK(rt);
if (rt->rt_ifa == ap->ifa &&
(ap->del || !(rt->rt_flags & RTF_STATIC))) {
/*
* Aquire a reference so that it can later be freed
* as the refcount would be 0 here in case of at least
* ap->del.
*/
RT_ADDREF(rt);
/*
* Disconnect it from the tree and permit protocols
* to cleanup.
*/
rtexpunge(rt);
/*
* At this point it is an rttrash node, and in case
* the above is the only reference we must free it.
* If we do not noone will have a pointer and the
* rtentry will be leaked forever.
* In case someone else holds a reference, we are
* fine as we only decrement the refcount. In that
* case if the other entity calls RT_REMREF, we
* will still be leaking but at least we tried.
*/
RTFREE_LOCKED(rt);
return (0);
}
RT_UNLOCK(rt);
return 0;
}
int
in_ifadown(struct ifaddr *ifa, int delete)
{
struct in_ifadown_arg arg;
struct radix_node_head *rnh;
int fibnum;
if (ifa->ifa_addr->sa_family != AF_INET)
return 1;
for ( fibnum = 0; fibnum < rt_numfibs; fibnum++) {
rnh = rt_tables_get_rnh(fibnum, AF_INET);
arg.ifa = ifa;
arg.del = delete;
RADIX_NODE_HEAD_LOCK(rnh);
rnh->rnh_walktree(rnh, in_ifadownkill, &arg);
RADIX_NODE_HEAD_UNLOCK(rnh);
ifa->ifa_flags &= ~IFA_ROUTE; /* XXXlocking? */
}
return 0;
}
/*
* inet versions of rt functions. These have fib extensions and
* for now will just reference the _fib variants.
* eventually this order will be reversed,
*/
void
in_rtalloc_ign(struct route *ro, u_long ignflags, u_int fibnum)
{
rtalloc_ign_fib(ro, ignflags, fibnum);
}
int
in_rtrequest( int req,
struct sockaddr *dst,
struct sockaddr *gateway,
struct sockaddr *netmask,
int flags,
struct rtentry **ret_nrt,
u_int fibnum)
{
return (rtrequest_fib(req, dst, gateway, netmask,
flags, ret_nrt, fibnum));
}
struct rtentry *
in_rtalloc1(struct sockaddr *dst, int report, u_long ignflags, u_int fibnum)
{
return (rtalloc1_fib(dst, report, ignflags, fibnum));
}
void
in_rtredirect(struct sockaddr *dst,
struct sockaddr *gateway,
struct sockaddr *netmask,
int flags,
struct sockaddr *src,
u_int fibnum)
{
rtredirect_fib(dst, gateway, netmask, flags, src, fibnum);
}
void
in_rtalloc(struct route *ro, u_int fibnum)
{
rtalloc_ign_fib(ro, 0UL, fibnum);
}
#if 0
int in_rt_getifa(struct rt_addrinfo *, u_int fibnum);
int in_rtioctl(u_long, caddr_t, u_int);
int in_rtrequest1(int, struct rt_addrinfo *, struct rtentry **, u_int);
#endif