From owner-svn-src-all@freebsd.org Sun Apr 26 16:22:36 2020 Return-Path: Delivered-To: svn-src-all@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id 39EBA2B4A41; Sun, 26 Apr 2020 16:22:36 +0000 (UTC) (envelope-from kp@FreeBSD.org) Received: from mxrelay.nyi.freebsd.org (mxrelay.nyi.freebsd.org [IPv6:2610:1c1:1:606c::19:3]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) server-signature RSA-PSS (4096 bits) client-signature RSA-PSS (4096 bits) client-digest SHA256) (Client CN "mxrelay.nyi.freebsd.org", Issuer "Let's Encrypt Authority X3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 499Cqc0T0tz4gCl; Sun, 26 Apr 2020 16:22:36 +0000 (UTC) (envelope-from kp@FreeBSD.org) Received: from repo.freebsd.org (repo.freebsd.org [IPv6:2610:1c1:1:6068::e6a:0]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mxrelay.nyi.freebsd.org (Postfix) with ESMTPS id 0B36723AF6; Sun, 26 Apr 2020 16:22:36 +0000 (UTC) (envelope-from kp@FreeBSD.org) Received: from repo.freebsd.org ([127.0.1.37]) by repo.freebsd.org (8.15.2/8.15.2) with ESMTP id 03QGMZwu001812; Sun, 26 Apr 2020 16:22:35 GMT (envelope-from kp@FreeBSD.org) Received: (from kp@localhost) by repo.freebsd.org (8.15.2/8.15.2/Submit) id 03QGMZWt001811; Sun, 26 Apr 2020 16:22:35 GMT (envelope-from kp@FreeBSD.org) Message-Id: <202004261622.03QGMZWt001811@repo.freebsd.org> X-Authentication-Warning: repo.freebsd.org: kp set sender to kp@FreeBSD.org using -f From: Kristof Provost Date: Sun, 26 Apr 2020 16:22:35 +0000 (UTC) To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r360345 - head/sys/net X-SVN-Group: head X-SVN-Commit-Author: kp X-SVN-Commit-Paths: head/sys/net X-SVN-Commit-Revision: 360345 X-SVN-Commit-Repository: base MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 26 Apr 2020 16:22:36 -0000 Author: kp Date: Sun Apr 26 16:22:35 2020 New Revision: 360345 URL: https://svnweb.freebsd.org/changeset/base/360345 Log: bridge: epoch-ification Run the bridge datapath under epoch, rather than under the BRIDGE_LOCK(). We still take the BRIDGE_LOCK() whenever we insert or delete items in the relevant lists, but we use epoch callbacks to free items so that it's safe to iterate the lists without the BRIDGE_LOCK. Tests on mercat5/6 shows this increases bridge throughput significantly, from 3.7Mpps to 18.6Mpps. Reviewed by: emaste, philip, melifaro MFC after: 2 months Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D24250 Modified: head/sys/net/if_bridge.c Modified: head/sys/net/if_bridge.c ============================================================================== --- head/sys/net/if_bridge.c Sun Apr 26 16:16:39 2020 (r360344) +++ head/sys/net/if_bridge.c Sun Apr 26 16:22:35 2020 (r360345) @@ -189,41 +189,14 @@ extern void nd6_setmtu(struct ifnet *); */ #define BRIDGE_LOCK_INIT(_sc) do { \ mtx_init(&(_sc)->sc_mtx, "if_bridge", NULL, MTX_DEF); \ - cv_init(&(_sc)->sc_cv, "if_bridge_cv"); \ } while (0) #define BRIDGE_LOCK_DESTROY(_sc) do { \ mtx_destroy(&(_sc)->sc_mtx); \ - cv_destroy(&(_sc)->sc_cv); \ } while (0) #define BRIDGE_LOCK(_sc) mtx_lock(&(_sc)->sc_mtx) #define BRIDGE_UNLOCK(_sc) mtx_unlock(&(_sc)->sc_mtx) #define BRIDGE_LOCK_ASSERT(_sc) mtx_assert(&(_sc)->sc_mtx, MA_OWNED) #define BRIDGE_UNLOCK_ASSERT(_sc) mtx_assert(&(_sc)->sc_mtx, MA_NOTOWNED) -#define BRIDGE_LOCK2REF(_sc, _err) do { \ - mtx_assert(&(_sc)->sc_mtx, MA_OWNED); \ - if ((_sc)->sc_iflist_xcnt > 0) \ - (_err) = EBUSY; \ - else \ - (_sc)->sc_iflist_ref++; \ - mtx_unlock(&(_sc)->sc_mtx); \ -} while (0) -#define BRIDGE_UNREF(_sc) do { \ - mtx_lock(&(_sc)->sc_mtx); \ - (_sc)->sc_iflist_ref--; \ - if (((_sc)->sc_iflist_xcnt > 0) && ((_sc)->sc_iflist_ref == 0)) \ - cv_broadcast(&(_sc)->sc_cv); \ - mtx_unlock(&(_sc)->sc_mtx); \ -} while (0) -#define BRIDGE_XLOCK(_sc) do { \ - mtx_assert(&(_sc)->sc_mtx, MA_OWNED); \ - (_sc)->sc_iflist_xcnt++; \ - while ((_sc)->sc_iflist_ref > 0) \ - cv_wait(&(_sc)->sc_cv, &(_sc)->sc_mtx); \ -} while (0) -#define BRIDGE_XDROP(_sc) do { \ - mtx_assert(&(_sc)->sc_mtx, MA_OWNED); \ - (_sc)->sc_iflist_xcnt--; \ -} while (0) /* * Bridge interface list entry. @@ -237,6 +210,7 @@ struct bridge_iflist { uint32_t bif_addrmax; /* max # of addresses */ uint32_t bif_addrcnt; /* cur. # of addresses */ uint32_t bif_addrexceeded;/* # of address violations */ + struct epoch_context bif_epoch_ctx; }; /* @@ -250,6 +224,8 @@ struct bridge_rtnode { uint8_t brt_flags; /* address flags */ uint8_t brt_addr[ETHER_ADDR_LEN]; uint16_t brt_vlan; /* vlan id */ + struct vnet *brt_vnet; + struct epoch_context brt_epoch_ctx; }; #define brt_ifp brt_dst->bif_ifp @@ -260,13 +236,10 @@ struct bridge_softc { struct ifnet *sc_ifp; /* make this an interface */ LIST_ENTRY(bridge_softc) sc_list; struct mtx sc_mtx; - struct cv sc_cv; uint32_t sc_brtmax; /* max # of addresses */ uint32_t sc_brtcnt; /* cur. # of addresses */ uint32_t sc_brttimeout; /* rt timeout in seconds */ struct callout sc_brcallout; /* bridge callout */ - uint32_t sc_iflist_ref; /* refcount for sc_iflist */ - uint32_t sc_iflist_xcnt; /* refcount for sc_iflist */ CK_LIST_HEAD(, bridge_iflist) sc_iflist; /* member interface list */ CK_LIST_HEAD(, bridge_rtnode) *sc_rthash; /* our forwarding table */ CK_LIST_HEAD(, bridge_rtnode) sc_rtlist; /* list version of above */ @@ -276,6 +249,7 @@ struct bridge_softc { uint32_t sc_brtexceeded; /* # of cache drops */ struct ifnet *sc_ifaddr; /* member mac copied from */ struct ether_addr sc_defaddr; /* Default MAC address */ + struct epoch_context sc_epoch_ctx; }; VNET_DEFINE_STATIC(struct mtx, bridge_list_mtx); @@ -596,6 +570,10 @@ vnet_bridge_uninit(const void *unused __unused) if_clone_detach(V_bridge_cloner); V_bridge_cloner = NULL; BRIDGE_LIST_LOCK_DESTROY(); + + /* Callbacks may use the UMA zone. */ + epoch_drain_callbacks(net_epoch_preempt); + uma_zdestroy(V_bridge_rtnode_zone); } VNET_SYSUNINIT(vnet_bridge_uninit, SI_SUB_PSEUDO, SI_ORDER_ANY, @@ -722,6 +700,17 @@ bridge_clone_create(struct if_clone *ifc, int unit, ca return (0); } +static void +bridge_clone_destroy_cb(struct epoch_context *ctx) +{ + struct bridge_softc *sc; + + sc = __containerof(ctx, struct bridge_softc, sc_epoch_ctx); + + BRIDGE_LOCK_DESTROY(sc); + free(sc, M_DEVBUF); +} + /* * bridge_clone_destroy: * @@ -732,7 +721,9 @@ bridge_clone_destroy(struct ifnet *ifp) { struct bridge_softc *sc = ifp->if_softc; struct bridge_iflist *bif; + struct epoch_tracker et; + NET_EPOCH_ENTER(et); BRIDGE_LOCK(sc); bridge_stop(ifp, 1); @@ -757,11 +748,12 @@ bridge_clone_destroy(struct ifnet *ifp) BRIDGE_LIST_UNLOCK(); bstp_detach(&sc->sc_stp); + NET_EPOCH_EXIT(et); + ether_ifdetach(ifp); if_free(ifp); - BRIDGE_LOCK_DESTROY(sc); - free(sc, M_DEVBUF); + NET_EPOCH_CALL(bridge_clone_destroy_cb, &sc->sc_epoch_ctx); } /* @@ -787,7 +779,10 @@ bridge_ioctl(struct ifnet *ifp, u_long cmd, caddr_t da struct ifdrv *ifd = (struct ifdrv *) data; const struct bridge_control *bc; int error = 0, oldmtu; + struct epoch_tracker et; + NET_EPOCH_ENTER(et); + switch (cmd) { case SIOCADDMULTI: @@ -908,6 +903,8 @@ bridge_ioctl(struct ifnet *ifp, u_long cmd, caddr_t da break; } + NET_EPOCH_EXIT(et); + return (error); } @@ -922,6 +919,8 @@ bridge_mutecaps(struct bridge_softc *sc) struct bridge_iflist *bif; int enabled, mask; + BRIDGE_LOCK_ASSERT(sc); + /* Initial bitmask of capabilities to test */ mask = BRIDGE_IFCAPS_MASK; @@ -930,7 +929,6 @@ bridge_mutecaps(struct bridge_softc *sc) mask &= bif->bif_savedcaps; } - BRIDGE_XLOCK(sc); CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { enabled = bif->bif_ifp->if_capenable; enabled &= ~BRIDGE_IFCAPS_STRIP; @@ -941,8 +939,6 @@ bridge_mutecaps(struct bridge_softc *sc) bridge_set_ifcap(sc, bif, enabled); BRIDGE_LOCK(sc); } - BRIDGE_XDROP(sc); - } static void @@ -983,7 +979,7 @@ bridge_lookup_member(struct bridge_softc *sc, const ch struct bridge_iflist *bif; struct ifnet *ifp; - BRIDGE_LOCK_ASSERT(sc); + NET_EPOCH_ASSERT(); CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { ifp = bif->bif_ifp; @@ -1004,7 +1000,7 @@ bridge_lookup_member_if(struct bridge_softc *sc, struc { struct bridge_iflist *bif; - BRIDGE_LOCK_ASSERT(sc); + NET_EPOCH_ASSERT(); CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { if (bif->bif_ifp == member_ifp) @@ -1014,6 +1010,16 @@ bridge_lookup_member_if(struct bridge_softc *sc, struc return (NULL); } +static void +bridge_delete_member_cb(struct epoch_context *ctx) +{ + struct bridge_iflist *bif; + + bif = __containerof(ctx, struct bridge_iflist, bif_epoch_ctx); + + free(bif, M_DEVBUF); +} + /* * bridge_delete_member: * @@ -1033,9 +1039,7 @@ bridge_delete_member(struct bridge_softc *sc, struct b bstp_disable(&bif->bif_stp); ifs->if_bridge = NULL; - BRIDGE_XLOCK(sc); CK_LIST_REMOVE(bif, bif_next); - BRIDGE_XDROP(sc); /* * If removing the interface that gave the bridge its mac address, set @@ -1094,7 +1098,8 @@ bridge_delete_member(struct bridge_softc *sc, struct b } bstp_destroy(&bif->bif_stp); /* prepare to free */ BRIDGE_LOCK(sc); - free(bif, M_DEVBUF); + + NET_EPOCH_CALL(bridge_delete_member_cb, &bif->bif_epoch_ctx); } /* @@ -1111,7 +1116,8 @@ bridge_delete_span(struct bridge_softc *sc, struct bri ("%s: not a span interface", __func__)); CK_LIST_REMOVE(bif, bif_next); - free(bif, M_DEVBUF); + + NET_EPOCH_CALL(bridge_delete_member_cb, &bif->bif_epoch_ctx); } static int @@ -1167,7 +1173,6 @@ bridge_ioctl_add(struct bridge_softc *sc, void *arg) * If any, remove all inet6 addresses from the member * interfaces. */ - BRIDGE_XLOCK(sc); CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { if (in6ifa_llaonifp(bif->bif_ifp)) { BRIDGE_UNLOCK(sc); @@ -1180,7 +1185,6 @@ bridge_ioctl_add(struct bridge_softc *sc, void *arg) bif->bif_ifp->if_xname); } } - BRIDGE_XDROP(sc); if (in6ifa_llaonifp(ifs)) { BRIDGE_UNLOCK(sc); in6_ifdetach(ifs); @@ -1494,12 +1498,17 @@ bridge_ioctl_saddr(struct bridge_softc *sc, void *arg) struct bridge_iflist *bif; int error; + NET_EPOCH_ASSERT(); + bif = bridge_lookup_member(sc, req->ifba_ifsname); if (bif == NULL) return (ENOENT); + /* bridge_rtupdate() may acquire the lock. */ + BRIDGE_UNLOCK(sc); error = bridge_rtupdate(sc, req->ifba_dst, req->ifba_vlan, bif, 1, req->ifba_flags); + BRIDGE_LOCK(sc); return (error); } @@ -1838,6 +1847,7 @@ bridge_ifdetach(void *arg __unused, struct ifnet *ifp) { struct bridge_softc *sc = ifp->if_bridge; struct bridge_iflist *bif; + struct epoch_tracker et; if (ifp->if_flags & IFF_RENAMING) return; @@ -1848,6 +1858,7 @@ bridge_ifdetach(void *arg __unused, struct ifnet *ifp) */ return; } + NET_EPOCH_ENTER(et); /* Check if the interface is a bridge member */ if (sc != NULL) { BRIDGE_LOCK(sc); @@ -1857,6 +1868,7 @@ bridge_ifdetach(void *arg __unused, struct ifnet *ifp) bridge_delete_member(sc, bif, 1); BRIDGE_UNLOCK(sc); + NET_EPOCH_EXIT(et); return; } @@ -1873,6 +1885,7 @@ bridge_ifdetach(void *arg __unused, struct ifnet *ifp) BRIDGE_UNLOCK(sc); } BRIDGE_LIST_UNLOCK(); + NET_EPOCH_EXIT(et); } /* @@ -1909,6 +1922,7 @@ bridge_stop(struct ifnet *ifp, int disable) { struct bridge_softc *sc = ifp->if_softc; + NET_EPOCH_ASSERT(); BRIDGE_LOCK_ASSERT(sc); if ((ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) @@ -2032,6 +2046,8 @@ bridge_output(struct ifnet *ifp, struct mbuf *m, struc struct bridge_softc *sc; uint16_t vlan; + NET_EPOCH_ASSERT(); + if (m->m_len < ETHER_HDR_LEN) { m = m_pullup(m, ETHER_HDR_LEN); if (m == NULL) @@ -2042,7 +2058,6 @@ bridge_output(struct ifnet *ifp, struct mbuf *m, struc sc = ifp->if_bridge; vlan = VLANTAGOF(m); - BRIDGE_LOCK(sc); bifp = sc->sc_ifp; /* @@ -2069,16 +2084,10 @@ bridge_output(struct ifnet *ifp, struct mbuf *m, struc if (dst_if == NULL) { struct bridge_iflist *bif; struct mbuf *mc; - int error = 0, used = 0; + int used = 0; bridge_span(sc, m); - BRIDGE_LOCK2REF(sc, error); - if (error) { - m_freem(m); - return (0); - } - CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { dst_if = bif->bif_ifp; @@ -2112,7 +2121,6 @@ bridge_output(struct ifnet *ifp, struct mbuf *m, struc } if (used == 0) m_freem(m); - BRIDGE_UNREF(sc); return (0); } @@ -2124,11 +2132,9 @@ sendunicast: bridge_span(sc, m); if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) { m_freem(m); - BRIDGE_UNLOCK(sc); return (0); } - BRIDGE_UNLOCK(sc); bridge_enqueue(sc, dst_if, m); return (0); } @@ -2153,10 +2159,8 @@ bridge_transmit(struct ifnet *ifp, struct mbuf *m) eh = mtod(m, struct ether_header *); - BRIDGE_LOCK(sc); if (((m->m_flags & (M_BCAST|M_MCAST)) == 0) && (dst_if = bridge_rtlookup(sc, eh->ether_dhost, 1)) != NULL) { - BRIDGE_UNLOCK(sc); error = bridge_enqueue(sc, dst_if, m); } else bridge_broadcast(sc, ifp, m, 0); @@ -2190,6 +2194,8 @@ bridge_forward(struct bridge_softc *sc, struct bridge_ uint8_t *dst; int error; + NET_EPOCH_ASSERT(); + src_if = m->m_pkthdr.rcvif; ifp = sc->sc_ifp; @@ -2268,12 +2274,10 @@ bridge_forward(struct bridge_softc *sc, struct bridge_ || PFIL_HOOKED_IN(V_inet6_pfil_head) #endif ) { - BRIDGE_UNLOCK(sc); if (bridge_pfil(&m, ifp, src_if, PFIL_IN) != 0) return; if (m == NULL) return; - BRIDGE_LOCK(sc); } if (dst_if == NULL) { @@ -2301,8 +2305,6 @@ bridge_forward(struct bridge_softc *sc, struct bridge_ dbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) goto drop; - BRIDGE_UNLOCK(sc); - if (PFIL_HOOKED_OUT(V_inet_pfil_head) #ifdef INET6 || PFIL_HOOKED_OUT(V_inet6_pfil_head) @@ -2318,7 +2320,6 @@ bridge_forward(struct bridge_softc *sc, struct bridge_ return; drop: - BRIDGE_UNLOCK(sc); m_freem(m); } @@ -2339,6 +2340,8 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) uint16_t vlan; int error; + NET_EPOCH_ASSERT(); + if ((sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) return (m); @@ -2359,10 +2362,8 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) m_freem(m); return (NULL); } - BRIDGE_LOCK(sc); bif = bridge_lookup_member_if(sc, ifp); if (bif == NULL) { - BRIDGE_UNLOCK(sc); return (m); } @@ -2375,13 +2376,11 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) if (memcmp(eh->ether_dhost, bstp_etheraddr, ETHER_ADDR_LEN) == 0) { bstp_input(&bif->bif_stp, ifp, m); /* consumes mbuf */ - BRIDGE_UNLOCK(sc); return (NULL); } if ((bif->bif_flags & IFBIF_STP) && bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) { - BRIDGE_UNLOCK(sc); return (m); } @@ -2392,7 +2391,6 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) */ mc = m_dup(m, M_NOWAIT); if (mc == NULL) { - BRIDGE_UNLOCK(sc); return (m); } @@ -2424,7 +2422,6 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) if ((bif->bif_flags & IFBIF_STP) && bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) { - BRIDGE_UNLOCK(sc); return (m); } @@ -2458,7 +2455,6 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) error = bridge_rtupdate(sc, eh->ether_shost, \ vlan, bif, 0, IFBAF_DYNAMIC); \ if (error && bif->bif_addrmax) { \ - BRIDGE_UNLOCK(sc); \ m_freem(m); \ return (NULL); \ } \ @@ -2466,7 +2462,6 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) m->m_pkthdr.rcvif = iface; \ if ((iface) == ifp) { \ /* Skip bridge processing... src == dest */ \ - BRIDGE_UNLOCK(sc); \ return (m); \ } \ /* It's passing over or to the bridge, locally. */ \ @@ -2478,13 +2473,11 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) OR_PFIL_HOOKED_INET6)) { \ if (bridge_pfil(&m, NULL, ifp, \ PFIL_IN) != 0 || m == NULL) { \ - BRIDGE_UNLOCK(sc); \ return (NULL); \ } \ } \ if ((iface) != bifp) \ ETHER_BPF_MTAP(iface, m); \ - BRIDGE_UNLOCK(sc); \ return (m); \ } \ \ @@ -2492,7 +2485,6 @@ bridge_input(struct ifnet *ifp, struct mbuf *m) if (memcmp(IF_LLADDR((iface)), eh->ether_shost, ETHER_ADDR_LEN) == 0 \ OR_CARP_CHECK_WE_ARE_SRC((iface)) \ ) { \ - BRIDGE_UNLOCK(sc); \ m_freem(m); \ return (NULL); \ } @@ -2543,16 +2535,12 @@ bridge_broadcast(struct bridge_softc *sc, struct ifnet struct bridge_iflist *dbif, *sbif; struct mbuf *mc; struct ifnet *dst_if; - int error = 0, used = 0, i; + int used = 0, i; + NET_EPOCH_ASSERT(); + sbif = bridge_lookup_member_if(sc, src_if); - BRIDGE_LOCK2REF(sc, error); - if (error) { - m_freem(m); - return; - } - /* Filter on the bridge interface before broadcasting */ if (runfilt && (PFIL_HOOKED_OUT(V_inet_pfil_head) #ifdef INET6 @@ -2560,9 +2548,9 @@ bridge_broadcast(struct bridge_softc *sc, struct ifnet #endif )) { if (bridge_pfil(&m, sc->sc_ifp, NULL, PFIL_OUT) != 0) - goto out; + return; if (m == NULL) - goto out; + return; } CK_LIST_FOREACH(dbif, &sc->sc_iflist, bif_next) { @@ -2625,9 +2613,6 @@ bridge_broadcast(struct bridge_softc *sc, struct ifnet } if (used == 0) m_freem(m); - -out: - BRIDGE_UNREF(sc); } /* @@ -2643,6 +2628,8 @@ bridge_span(struct bridge_softc *sc, struct mbuf *m) struct ifnet *dst_if; struct mbuf *mc; + NET_EPOCH_ASSERT(); + if (CK_LIST_EMPTY(&sc->sc_spanlist)) return; @@ -2674,7 +2661,8 @@ bridge_rtupdate(struct bridge_softc *sc, const uint8_t struct bridge_rtnode *brt; int error; - BRIDGE_LOCK_ASSERT(sc); + NET_EPOCH_ASSERT(); + BRIDGE_UNLOCK_ASSERT(sc); /* Check the source address is valid and not multicast. */ if (ETHER_IS_MULTICAST(dst) || @@ -2691,13 +2679,24 @@ bridge_rtupdate(struct bridge_softc *sc, const uint8_t * update it, otherwise create a new one. */ if ((brt = bridge_rtnode_lookup(sc, dst, vlan)) == NULL) { + BRIDGE_LOCK(sc); + + /* Check again, now that we have the lock. There could have + * been a race and we only want to insert this once. */ + if ((brt = bridge_rtnode_lookup(sc, dst, vlan)) != NULL) { + BRIDGE_UNLOCK(sc); + return (0); + } + if (sc->sc_brtcnt >= sc->sc_brtmax) { sc->sc_brtexceeded++; + BRIDGE_UNLOCK(sc); return (ENOSPC); } /* Check per interface address limits (if enabled) */ if (bif->bif_addrmax && bif->bif_addrcnt >= bif->bif_addrmax) { bif->bif_addrexceeded++; + BRIDGE_UNLOCK(sc); return (ENOSPC); } @@ -2707,8 +2706,11 @@ bridge_rtupdate(struct bridge_softc *sc, const uint8_t * address. */ brt = uma_zalloc(V_bridge_rtnode_zone, M_NOWAIT | M_ZERO); - if (brt == NULL) + if (brt == NULL) { + BRIDGE_UNLOCK(sc); return (ENOMEM); + } + brt->brt_vnet = curvnet; if (bif->bif_flags & IFBIF_STICKY) brt->brt_flags = IFBAF_STICKY; @@ -2720,17 +2722,22 @@ bridge_rtupdate(struct bridge_softc *sc, const uint8_t if ((error = bridge_rtnode_insert(sc, brt)) != 0) { uma_zfree(V_bridge_rtnode_zone, brt); + BRIDGE_UNLOCK(sc); return (error); } brt->brt_dst = bif; bif->bif_addrcnt++; + + BRIDGE_UNLOCK(sc); } if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && brt->brt_dst != bif) { + BRIDGE_LOCK(sc); brt->brt_dst->bif_addrcnt--; brt->brt_dst = bif; brt->brt_dst->bif_addrcnt++; + BRIDGE_UNLOCK(sc); } if ((flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) @@ -2751,7 +2758,7 @@ bridge_rtlookup(struct bridge_softc *sc, const uint8_t { struct bridge_rtnode *brt; - BRIDGE_LOCK_ASSERT(sc); + NET_EPOCH_ASSERT(); if ((brt = bridge_rtnode_lookup(sc, addr, vlan)) == NULL) return (NULL); @@ -2771,6 +2778,7 @@ bridge_rttrim(struct bridge_softc *sc) { struct bridge_rtnode *brt, *nbrt; + NET_EPOCH_ASSERT(); BRIDGE_LOCK_ASSERT(sc); /* Make sure we actually need to do this. */ @@ -2800,7 +2808,9 @@ static void bridge_timer(void *arg) { struct bridge_softc *sc = arg; + struct epoch_tracker et; + NET_EPOCH_ENTER(et); BRIDGE_LOCK_ASSERT(sc); /* Destruction of rtnodes requires a proper vnet context */ @@ -2811,6 +2821,7 @@ bridge_timer(void *arg) callout_reset(&sc->sc_brcallout, bridge_rtable_prune_period * hz, bridge_timer, sc); CURVNET_RESTORE(); + NET_EPOCH_EXIT(et); } /* @@ -2823,6 +2834,7 @@ bridge_rtage(struct bridge_softc *sc) { struct bridge_rtnode *brt, *nbrt; + NET_EPOCH_ASSERT(); BRIDGE_LOCK_ASSERT(sc); CK_LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { @@ -2843,6 +2855,7 @@ bridge_rtflush(struct bridge_softc *sc, int full) { struct bridge_rtnode *brt, *nbrt; + NET_EPOCH_ASSERT(); BRIDGE_LOCK_ASSERT(sc); CK_LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { @@ -2862,6 +2875,7 @@ bridge_rtdaddr(struct bridge_softc *sc, const uint8_t struct bridge_rtnode *brt; int found = 0; + NET_EPOCH_ASSERT(); BRIDGE_LOCK_ASSERT(sc); /* @@ -2886,6 +2900,7 @@ bridge_rtdelete(struct bridge_softc *sc, struct ifnet { struct bridge_rtnode *brt, *nbrt; + NET_EPOCH_ASSERT(); BRIDGE_LOCK_ASSERT(sc); CK_LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { @@ -2990,7 +3005,7 @@ bridge_rtnode_lookup(struct bridge_softc *sc, const ui uint32_t hash; int dir; - BRIDGE_LOCK_ASSERT(sc); + NET_EPOCH_ASSERT(); hash = bridge_rthash(sc, addr); CK_LIST_FOREACH(brt, &sc->sc_rthash[hash], brt_hash) { @@ -3053,6 +3068,18 @@ out: return (0); } +static void +bridge_rtnode_destroy_cb(struct epoch_context *ctx) +{ + struct bridge_rtnode *brt; + + brt = __containerof(ctx, struct bridge_rtnode, brt_epoch_ctx); + + CURVNET_SET(brt->brt_vnet); + uma_zfree(V_bridge_rtnode_zone, brt); + CURVNET_RESTORE(); +} + /* * bridge_rtnode_destroy: * @@ -3061,6 +3088,7 @@ out: static void bridge_rtnode_destroy(struct bridge_softc *sc, struct bridge_rtnode *brt) { + NET_EPOCH_ASSERT(); BRIDGE_LOCK_ASSERT(sc); CK_LIST_REMOVE(brt, brt_hash); @@ -3068,7 +3096,8 @@ bridge_rtnode_destroy(struct bridge_softc *sc, struct CK_LIST_REMOVE(brt, brt_list); sc->sc_brtcnt--; brt->brt_dst->bif_addrcnt--; - uma_zfree(V_bridge_rtnode_zone, brt); + + NET_EPOCH_CALL(bridge_rtnode_destroy_cb, &brt->brt_epoch_ctx); } /* @@ -3081,7 +3110,9 @@ bridge_rtable_expire(struct ifnet *ifp, int age) { struct bridge_softc *sc = ifp->if_bridge; struct bridge_rtnode *brt; + struct epoch_tracker et; + NET_EPOCH_ENTER(et); CURVNET_SET(ifp->if_vnet); BRIDGE_LOCK(sc); @@ -3102,6 +3133,7 @@ bridge_rtable_expire(struct ifnet *ifp, int age) } BRIDGE_UNLOCK(sc); CURVNET_RESTORE(); + NET_EPOCH_EXIT(et); } /* @@ -3607,17 +3639,20 @@ bridge_linkstate(struct ifnet *ifp) { struct bridge_softc *sc = ifp->if_bridge; struct bridge_iflist *bif; + struct epoch_tracker et; - BRIDGE_LOCK(sc); + NET_EPOCH_ENTER(et); + bif = bridge_lookup_member_if(sc, ifp); if (bif == NULL) { - BRIDGE_UNLOCK(sc); + NET_EPOCH_EXIT(et); return; } bridge_linkcheck(sc); - BRIDGE_UNLOCK(sc); bstp_linkstate(&bif->bif_stp); + + NET_EPOCH_EXIT(et); } static void @@ -3626,7 +3661,8 @@ bridge_linkcheck(struct bridge_softc *sc) struct bridge_iflist *bif; int new_link, hasls; - BRIDGE_LOCK_ASSERT(sc); + NET_EPOCH_ASSERT(); + new_link = LINK_STATE_DOWN; hasls = 0; /* Our link is considered up if at least one of our ports is active */