Hello,
It has come to my attention that when bridge_fragment() has to fragment
an IP packet it can cause mbuf leaks if, for example m_pullup() ends up
allocating a new buffer at the front of the mbuf chain or if the calls
to M_PREPEND would fail due to the mbuf pool being depleted.
One can easily convince himself that something is wrong by observing
that, in bridge_pfil(), bridge_fragment() gets passed an mbuf pointer
and if the call to m_pullup() in bridge_fragment ends up allocating a
new mbuf at the front of the chain, then whatever happens next isn't
going to give bridge_pfil() the updated mbuf pointer. You can see below
this is why I had to change the arguments to bridge_fragment to take a
double pointer to the mbuf.
The problem actually goes a lot deeper since, after having called
ip_fragment(), bridge_fragment() simply goes over the list of packets it
got and carelessly calls into M_PREPEND() to make space for the ethernet
header. Now this isn't a problem if M_PREPEND() can find the space at
the beginning of the mbuf but will wreak havoc if it must take the least
used code path and prepend an mbuf to the chain. In this case we have m0
pointing to a newly allocated mbuf but the list (walked through
m_nextpkt) will never get updated to point to m0 (the previous mbuf's
m_nextpkt pointer should now point to the new m0).
And it goes on, the error case isn't' handled properly since if MPREPEND
deleted m0 we not have a list of packets that is potentially pointed to
an already freed mbuf. Nowhere can we see that m_nextpkt is updated to
take care of that case. This is why in my patch below I detach each m0
from the packet list so if a failure occurs in M_PREPEND of if a new
mbuf is added at the beginning of the chain I can keep the list of
packets updated.
Same thing with the goto out; at the end. If we lost a fragment while
adding ethernet header to it we must free the entire chain of packets or
we will leak mbufs. There is no point in sending fragments if we aren't
going to send them all.
To convince yourself of this problem, one can simple create, on his
favorite bridge one 'reass' rule in ipfw. I use the following rule:
00080 253 27878 reass ip from any to any { proto udp or proto gre }
Then send (assuming your MTU is 1500), a big udp packet using iperf. I
use this command:
iperf -c 10.10.73.2 -u -b16Mb -t 240 -i 10 -l 5000
After iperf has finished, look at your mbuf count with netstat -m.
I have made a patch to fix this problem (attached). The patch was tested
under load when we actually do run out of mbuf to trigger the error
cases as well as with all sorts of number of fragments (varying the
initial packet size).
I see this seems to affect all modern versions of FreeBSD, please feel
free to test or contest. I would like to see this added to FreeBSD
eventually in one shape or another.
Best regards,
Karim.
diff --git a/freebsd/sys/net/if_bridge.c b/freebsd/sys/net/if_bridge.c
index b9c3bca..a550bc1 100644
--- a/freebsd/sys/net/if_bridge.c
+++ b/freebsd/sys/net/if_bridge.c
@@ -344,7 +344,7 @@ static int bridge_ip_checkbasic(struct mbuf **mp);
#ifdef INET6
static int bridge_ip6_checkbasic(struct mbuf **mp);
#endif /* INET6 */
-static int bridge_fragment(struct ifnet *, struct mbuf *,
+static int bridge_fragment(struct ifnet *, struct mbuf **,
struct ether_header *, int, struct llc *);
static void bridge_linkstate(struct ifnet *ifp);
static void bridge_linkcheck(struct bridge_softc *sc);
@@ -4025,7 +4025,7 @@ ipfwpass:
if (pfil_member && ifp != NULL && dir == PFIL_OUT) {
i = (*mp)->m_pkthdr.len;
if (i > ifp->if_mtu) {
- error = bridge_fragment(ifp, *mp, &eh2, snap,
+ error = bridge_fragment(ifp, mp, &eh2, snap,
&llc1);
return (error);
}
@@ -4313,10 +4313,13 @@ bad:
* Return a fragmented mbuf chain.
*/
static int
-bridge_fragment(struct ifnet *ifp, struct mbuf *m, struct ether_header *eh,
+bridge_fragment(struct ifnet *ifp, struct mbuf **mp, struct ether_header *eh,
int snap, struct llc *llc)
{
struct mbuf *m0;
+ struct mbuf *prev = NULL;
+ struct mbuf *next;
+ struct mbuf *m = *mp;
struct ip *ip;
int error = -1;
@@ -4324,42 +4327,78 @@ bridge_fragment(struct ifnet *ifp, struct mbuf *m, struct ether_header *eh,
(m = m_pullup(m, sizeof(struct ip))) == NULL)
goto out;
ip = mtod(m, struct ip *);
+ /* m could have changed */
+ *mp = m;
error = ip_fragment(ip, &m, ifp->if_mtu, ifp->if_hwassist,
CSUM_DELAY_IP);
if (error)
goto out;
- /* walk the chain and re-add the Ethernet header */
- for (m0 = m; m0; m0 = m0->m_nextpkt) {
- if (error == 0) {
- if (snap) {
- M_PREPEND(m0, sizeof(struct llc), M_DONTWAIT);
- if (m0 == NULL) {
- error = ENOBUFS;
- continue;
- }
- bcopy(llc, mtod(m0, caddr_t),
- sizeof(struct llc));
- }
- M_PREPEND(m0, ETHER_HDR_LEN, M_DONTWAIT);
+ /* m could have changed */
+ *mp = m;
+ m0 = m;
+ next = m->m_nextpkt;
+ /*
+ * walk the chain and re-add the Ethernet header:
+ *
+ * We can't trust M_PREPEND to keep the list of packets consistent since it may
+ * allocate new pkt headers to be put in front and won't copy m_nextpkt over.
+ */
+ while (m0) {
+ /* disconnect from list of fragments the mbuf we may potentially free */
+ if (prev != NULL)
+ prev->m_nextpkt = NULL;
+
+ if (snap) {
+ M_PREPEND(m0, sizeof(struct llc), M_DONTWAIT);
if (m0 == NULL) {
error = ENOBUFS;
- continue;
+ if (prev) /* there are pkt fragments before the one we lost */
+ prev->m_nextpkt = next;
+ else
+ m = next;
+ goto out;
}
- bcopy(eh, mtod(m0, caddr_t), ETHER_HDR_LEN);
- } else
- m_freem(m);
+ bcopy(llc, mtod(m0, caddr_t),
+ sizeof(struct llc));
+ }
+ M_PREPEND(m0, ETHER_HDR_LEN, M_DONTWAIT);
+ if (m0 == NULL) {
+ error = ENOBUFS;
+ if (prev) /* there are pkt fragments before the one we lost */
+ prev->m_nextpkt = next;
+ else
+ m = next;
+ goto out;
+ }
+ bcopy(eh, mtod(m0, caddr_t), ETHER_HDR_LEN);
+ /* m0 could have changed due to m_prepend */
+ if (prev != NULL)
+ prev->m_nextpkt = m0;
+ else
+ m = m0; /* first fragment has no previous mbufs */
+ prev = m0;
+ m0 = next;
+ if (next)
+ next = next->m_nextpkt;
}
if (error == 0)
KMOD_IPSTAT_INC(ips_fragmented);
+ *mp = m;
return (error);
out:
- if (m != NULL)
+ /* free the whole list of potential fragments */
+ while (m) {
+ next = m->m_nextpkt;
m_freem(m);
+ m = next;
+ }
+
+ *mp = NULL;
return (error);
}
_______________________________________________
freebsd-net@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"