NetBSD/sys/net/if_faith.c
ozaki-r 6fb8880601 Make the routing table and rtcaches MP-safe
See the following descriptions for details.

Proposed on tech-kern and tech-net


Overview
--------

We protect the routing table with a rwock and protect
rtcaches with another rwlock. Each rtentry is protected
from being freed or updated via reference counting and psref.

Global rwlocks
--------------

There are two rwlocks; one for the routing table (rt_lock) and
the other for rtcaches (rtcache_lock). rtcache_lock covers
all existing rtcaches; there may have room for optimizations
(future work).

The locking order is rtcache_lock first and rt_lock is next.

rtentry references
------------------

References to an rtentry is managed with reference counting
and psref. Either of the two mechanisms is used depending on
where a rtentry is obtained. Reference counting is used when
we obtain a rtentry from the routing table directly via
rtalloc1 and rtrequest{,1} while psref is used when we obtain
a rtentry from a rtcache via rtcache_* APIs. In both cases,
a caller can sleep/block with holding an obtained rtentry.

The reasons why we use two different mechanisms are (i) only
using reference counting hurts the performance due to atomic
instructions (rtcache case) (ii) ease of implementation;
applying psref to APIs such rtaloc1 and rtrequest{,1} requires
additional works (adding a local variable and an argument).

We will finally migrate to use only psref but we can do it
when we have a lockless routing table alternative.

Reference counting for rtentry
------------------------------

rt_refcnt now doesn't count permanent references such as for
rt_timers and rtcaches, instead it is used only for temporal
references when obtaining a rtentry via rtalloc1 and rtrequest{,1}.
We can do so because destroying a rtentry always involves
removing references of rt_timers and rtcaches to the rtentry
and we don't need to track such references. This also makes
it easy to wait for readers to release references on deleting
or updating a rtentry, i.e., we can simply wait until the
reference counter is 0 or 1. (If there are permanent references
the counter can be arbitrary.)

rt_ref increments a reference counter of a rtentry and rt_unref
decrements it. rt_ref is called inside APIs (rtalloc1 and
rtrequest{,1} so users don't need to care about it while
users must call rt_unref to an obtained rtentry after using it.

rtfree is removed and we use rt_unref and rt_free instead.
rt_unref now just decrements the counter of a given rtentry
and rt_free just tries to destroy a given rtentry.

See the next section for destructions of rtentries by rt_free.

Destructions of rtentries
-------------------------

We destroy a rtentry only when we call rtrequst{,1}(RTM_DELETE);
the original implementation can destroy in any rtfree where it's
the last reference. If we use reference counting or psref, it's
easy to understand if the place that a rtentry is destroyed is
fixed.

rt_free waits for references to a given rtentry to be released
before actually destroying the rtentry. rt_free uses a condition
variable (cv_wait) (and psref_target_destroy for psref) to wait.

Unfortunately rtrequst{,1}(RTM_DELETE) can be called in softint
that we cannot use cv_wait. In that case, we have to defer the
destruction to a workqueue.

rtentry#rt_cv, rtentry#rt_psref and global variables
(see rt_free_global) are added to conduct the procedure.

Updates of rtentries
--------------------

One difficulty to use refcnt/psref instead of rwlock for rtentry
is updates of rtentries. We need an additional mechanism to
prevent readers from seeing inconsistency of a rtentry being
updated.

We introduce RTF_UPDATING flag to rtentries that are updating.
While the flag is set to a rtentry, users cannot acquire the
rtentry. By doing so, we avoid users to see inconsistent
rtentries.

There are two options when a user tries to acquire a rtentry
with the RTF_UPDATING flag; if a user runs in softint context
the user fails to acquire a rtentry (NULL is returned).
Otherwise a user waits until the update completes by waiting
on cv.

The procedure of a updater is simpler to destruction of
a rtentry. Wait on cv (and psref) and after all readers left,
proceed with the update.

Global variables (see rt_update_global) are added to conduct
the procedure.

Currently we apply the mechanism to only RTM_CHANGE in
rtsock.c. We would have to apply other codes. See
"Known issues" section.

psref for rtentry
-----------------

When we obtain a rtentry from a rtcache via rtcache_* APIs,
psref is used to reference to the rtentry.

rtcache_ref acquires a reference to a rtentry with psref
and rtcache_unref releases the reference after using it.
rtcache_ref is called inside rtcache_* APIs and users don't
need to take care of it while users must call rtcache_unref
to release the reference.

struct psref and int bound that is needed for psref is
embedded into struct route. By doing so we don't need to
add local variables and additional argument to APIs.

However this adds another constraint to psref other than
reference counting one's; holding a reference of an rtentry
via a rtcache is allowed by just one caller at the same time.
So we must not acquire a rtentry via a rtcache twice and
avoid a recursive use of a rtcache. And also a rtcache must
be arranged to be used by a LWP/softint at the same time
somehow. For IP forwarding case, we have per-CPU rtcaches
used in softint so the constraint is guaranteed. For a h
rtcache of a PCB case, the constraint is guaranteed by the
solock of each PCB. Any other cases (pf, ipf, stf and ipsec)
are currently guaranteed by only the existence of the global
locks (softnet_lock and/or KERNEL_LOCK). If we've found the
cases that we cannot guarantee the constraint, we would need
to introduce other rtcache APIs that use simple reference
counting.

psref of rtcache is created with IPL_SOFTNET and so rtcache
shouldn't used at an IPL higher than IPL_SOFTNET.

Note that rtcache_free is used to invalidate a given rtcache.
We don't need another care by my change; just keep them as
they are.

Performance impact
------------------

When NET_MPSAFE is disabled the performance drop is 3% while
when it's enabled the drop is increased to 11%. The difference
comes from that currently we don't take any global locks and
don't use psref if NET_MPSAFE is disabled.

We can optimize the performance of the case of NET_MPSAFE
on by reducing lookups of rtcache that uses psref;
currently we do two lookups but we should be able to trim
one of two. This is a future work.

Known issues
------------

There are two known issues to be solved; one is that
a caller of rtrequest(RTM_ADD) may change rtentry (see rtinit).
We need to prevent new references during the update. Or
we may be able to remove the code (perhaps, need more
investigations).

The other is rtredirect that updates a rtentry. We need
to apply our update mechanism, however it's not easy because
rtredirect is called in softint and we cannot apply our
mechanism simply. One solution is to defer rtredirect to
a workqueue but it requires some code restructuring.
2016-12-12 03:55:57 +00:00

336 lines
7.2 KiB
C

/* $NetBSD: if_faith.c,v 1.55 2016/12/12 03:55:57 ozaki-r Exp $ */
/* $KAME: if_faith.c,v 1.21 2001/02/20 07:59:26 itojun Exp $ */
/*
* Copyright (c) 1982, 1986, 1993
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS 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.
*/
/*
* derived from
* @(#)if_loop.c 8.1 (Berkeley) 6/10/93
* Id: if_loop.c,v 1.22 1996/06/19 16:24:10 wollman Exp
*/
/*
* IPv6-to-IPv4 TCP relay capturing interface
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: if_faith.c,v 1.55 2016/12/12 03:55:57 ozaki-r Exp $");
#ifdef _KERNEL_OPT
#include "opt_inet.h"
#endif
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/errno.h>
#include <sys/ioctl.h>
#include <sys/time.h>
#include <sys/queue.h>
#include <sys/device.h>
#include <sys/module.h>
#include <sys/atomic.h>
#include <sys/cpu.h>
#include <net/if.h>
#include <net/if_types.h>
#include <net/netisr.h>
#include <net/route.h>
#include <net/bpf.h>
#include <net/if_faith.h>
#ifdef INET
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/in_var.h>
#include <netinet/ip.h>
#endif
#ifdef INET6
#ifndef INET
#include <netinet/in.h>
#endif
#include <netinet6/in6_var.h>
#include <netinet/ip6.h>
#include <netinet6/ip6_var.h>
#endif
#include <net/net_osdep.h>
#include "ioconf.h"
static int faithioctl(struct ifnet *, u_long, void *);
static int faithoutput(struct ifnet *, struct mbuf *,
const struct sockaddr *, const struct rtentry *);
static void faithrtrequest(int, struct rtentry *,
const struct rt_addrinfo *);
static int faith_clone_create(struct if_clone *, int);
static int faith_clone_destroy(struct ifnet *);
static struct if_clone faith_cloner =
IF_CLONE_INITIALIZER("faith", faith_clone_create, faith_clone_destroy);
#define FAITHMTU 1500
static u_int faith_count;
/* ARGSUSED */
void
faithattach(int count)
{
/*
* Nothing to do here, initialization is handled by the
* module initialization code in faithinit() below).
*/
}
static void
faithinit(void)
{
if_clone_attach(&faith_cloner);
}
static int
faithdetach(void)
{
int error = 0;
if (faith_count != 0)
error = EBUSY;
if (error == 0)
if_clone_detach(&faith_cloner);
return error;
}
static int
faith_clone_create(struct if_clone *ifc, int unit)
{
struct ifnet *ifp;
ifp = if_alloc(IFT_FAITH);
if_initname(ifp, ifc->ifc_name, unit);
ifp->if_mtu = FAITHMTU;
/* Change to BROADCAST experimentaly to announce its prefix. */
ifp->if_flags = /* IFF_LOOPBACK */ IFF_BROADCAST | IFF_MULTICAST;
ifp->if_ioctl = faithioctl;
ifp->if_output = faithoutput;
ifp->if_type = IFT_FAITH;
ifp->if_hdrlen = 0;
ifp->if_addrlen = 0;
ifp->if_dlt = DLT_NULL;
if_attach(ifp);
if_alloc_sadl(ifp);
bpf_attach(ifp, DLT_NULL, sizeof(u_int));
atomic_inc_uint(&faith_count);
return (0);
}
int
faith_clone_destroy(struct ifnet *ifp)
{
bpf_detach(ifp);
if_detach(ifp);
if_free(ifp);
atomic_dec_uint(&faith_count);
return (0);
}
static int
faithoutput(struct ifnet *ifp, struct mbuf *m, const struct sockaddr *dst,
const struct rtentry *rt)
{
pktqueue_t *pktq;
size_t pktlen;
int s, error;
uint32_t af;
if ((m->m_flags & M_PKTHDR) == 0)
panic("faithoutput no HDR");
af = dst->sa_family;
/* BPF write needs to be handled specially */
if (af == AF_UNSPEC) {
af = *(mtod(m, int *));
m_adj(m, sizeof(int));
}
bpf_mtap_af(ifp, af, m);
if (rt && rt->rt_flags & (RTF_REJECT|RTF_BLACKHOLE)) {
m_freem(m);
return (rt->rt_flags & RTF_BLACKHOLE ? 0 :
rt->rt_flags & RTF_HOST ? EHOSTUNREACH : ENETUNREACH);
}
pktlen = m->m_pkthdr.len;
ifp->if_opackets++;
ifp->if_obytes += pktlen;
switch (af) {
#ifdef INET
case AF_INET:
pktq = ip_pktq;
break;
#endif
#ifdef INET6
case AF_INET6:
pktq = ip6_pktq;
break;
#endif
default:
m_freem(m);
return EAFNOSUPPORT;
}
/* XXX do we need more sanity checks? */
KASSERT(pktq != NULL);
m_set_rcvif(m, ifp);
s = splnet();
if (__predict_true(pktq_enqueue(pktq, m, 0))) {
ifp->if_ipackets++;
ifp->if_ibytes += pktlen;
error = 0;
} else {
m_freem(m);
error = ENOBUFS;
}
splx(s);
return error;
}
/* ARGSUSED */
static void
faithrtrequest(int cmd, struct rtentry *rt,
const struct rt_addrinfo *info)
{
if (rt)
rt->rt_rmx.rmx_mtu = rt->rt_ifp->if_mtu; /* for ISO */
}
/*
* Process an ioctl request.
*/
/* ARGSUSED */
static int
faithioctl(struct ifnet *ifp, u_long cmd, void *data)
{
struct ifaddr *ifa;
struct ifreq *ifr = (struct ifreq *)data;
int error = 0;
switch (cmd) {
case SIOCINITIFADDR:
ifp->if_flags |= IFF_UP | IFF_RUNNING;
ifa = (struct ifaddr *)data;
ifa->ifa_rtrequest = faithrtrequest;
/*
* Everything else is done at a higher level.
*/
break;
case SIOCADDMULTI:
case SIOCDELMULTI:
if (ifr == 0) {
error = EAFNOSUPPORT; /* XXX */
break;
}
switch (ifr->ifr_addr.sa_family) {
#ifdef INET
case AF_INET:
break;
#endif
#ifdef INET6
case AF_INET6:
break;
#endif
default:
error = EAFNOSUPPORT;
break;
}
break;
default:
if ((error = ifioctl_common(ifp, cmd, data)) == ENETRESET)
error = 0;
break;
}
return (error);
}
#ifdef INET6
/*
* XXX could be slow
* XXX could be layer violation to call sys/net from sys/netinet6
*/
int
faithprefix(struct in6_addr *in6)
{
struct rtentry *rt;
struct sockaddr_in6 sin6;
int ret;
if (ip6_keepfaith == 0)
return 0;
memset(&sin6, 0, sizeof(sin6));
sin6.sin6_family = AF_INET6;
sin6.sin6_len = sizeof(struct sockaddr_in6);
sin6.sin6_addr = *in6;
rt = rtalloc1((struct sockaddr *)&sin6, 0);
if (rt && rt->rt_ifp && rt->rt_ifp->if_type == IFT_FAITH &&
(rt->rt_ifp->if_flags & IFF_UP) != 0)
ret = 1;
else
ret = 0;
if (rt)
rt_unref(rt);
return ret;
}
#endif
/*
* Module infrastructure
*/
#include "if_module.h"
IF_MODULE(MODULE_CLASS_DRIVER, faith, "")