diff options
Diffstat (limited to 'net')
107 files changed, 4289 insertions, 2491 deletions
diff --git a/net/9p/Kconfig b/net/9p/Kconfig index 3d11fec3a8dc..64468c49791f 100644 --- a/net/9p/Kconfig +++ b/net/9p/Kconfig @@ -4,7 +4,6 @@ # menuconfig NET_9P - depends on NET tristate "Plan 9 Resource Sharing Support (9P2000)" help If you say Y here, you will get experimental support for diff --git a/net/Makefile b/net/Makefile index d96b0aa8f39f..9ca9572188fe 100644 --- a/net/Makefile +++ b/net/Makefile @@ -6,20 +6,19 @@ # Rewritten to use lists instead of if-statements. # -obj-$(CONFIG_NET) := devres.o socket.o core/ +obj-y := devres.o socket.o core/ -tmp-$(CONFIG_COMPAT) := compat.o -obj-$(CONFIG_NET) += $(tmp-y) +obj-$(CONFIG_COMPAT) += compat.o # LLC has to be linked before the files in net/802/ obj-$(CONFIG_LLC) += llc/ -obj-$(CONFIG_NET) += ethernet/ 802/ sched/ netlink/ bpf/ ethtool/ +obj-y += ethernet/ 802/ sched/ netlink/ bpf/ ethtool/ obj-$(CONFIG_NETFILTER) += netfilter/ obj-$(CONFIG_INET) += ipv4/ obj-$(CONFIG_TLS) += tls/ obj-$(CONFIG_XFRM) += xfrm/ obj-$(CONFIG_UNIX_SCM) += unix/ -obj-$(CONFIG_NET) += ipv6/ +obj-y += ipv6/ obj-$(CONFIG_BPFILTER) += bpfilter/ obj-$(CONFIG_PACKET) += packet/ obj-$(CONFIG_NET_KEY) += key/ @@ -56,16 +55,12 @@ obj-$(CONFIG_SMC) += smc/ obj-$(CONFIG_RFKILL) += rfkill/ obj-$(CONFIG_NET_9P) += 9p/ obj-$(CONFIG_CAIF) += caif/ -ifneq ($(CONFIG_DCB),) -obj-y += dcb/ -endif +obj-$(CONFIG_DCB) += dcb/ obj-$(CONFIG_6LOWPAN) += 6lowpan/ obj-$(CONFIG_IEEE802154) += ieee802154/ obj-$(CONFIG_MAC802154) += mac802154/ -ifeq ($(CONFIG_NET),y) obj-$(CONFIG_SYSCTL) += sysctl_net.o -endif obj-$(CONFIG_DNS_RESOLVER) += dns_resolver/ obj-$(CONFIG_CEPH_LIB) += ceph/ obj-$(CONFIG_BATMAN_ADV) += batman-adv/ @@ -77,12 +72,8 @@ obj-$(CONFIG_VSOCKETS) += vmw_vsock/ obj-$(CONFIG_MPLS) += mpls/ obj-$(CONFIG_NET_NSH) += nsh/ obj-$(CONFIG_HSR) += hsr/ -ifneq ($(CONFIG_NET_SWITCHDEV),) -obj-y += switchdev/ -endif -ifneq ($(CONFIG_NET_L3_MASTER_DEV),) -obj-y += l3mdev/ -endif +obj-$(CONFIG_NET_SWITCHDEV) += switchdev/ +obj-$(CONFIG_NET_L3_MASTER_DEV) += l3mdev/ obj-$(CONFIG_QRTR) += qrtr/ obj-$(CONFIG_NET_NCSI) += ncsi/ obj-$(CONFIG_XDP_SOCKETS) += xdp/ diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig index 993afd5ff7bb..43ae3dcbbbeb 100644 --- a/net/batman-adv/Kconfig +++ b/net/batman-adv/Kconfig @@ -9,7 +9,6 @@ config BATMAN_ADV tristate "B.A.T.M.A.N. Advanced Meshing Protocol" - depends on NET select LIBCRC32C help B.A.T.M.A.N. (better approach to mobile ad-hoc networking) is diff --git a/net/bluetooth/Kconfig b/net/bluetooth/Kconfig index 64e669acd42f..400c5130dc0a 100644 --- a/net/bluetooth/Kconfig +++ b/net/bluetooth/Kconfig @@ -5,7 +5,7 @@ menuconfig BT tristate "Bluetooth subsystem support" - depends on NET && !S390 + depends on !S390 depends on RFKILL || !RFKILL select CRC16 select CRYPTO diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 23dfb2010ba6..58bcb8c849d5 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -272,7 +272,8 @@ int bpf_prog_test_run_raw_tp(struct bpf_prog *prog, kattr->test.repeat) return -EINVAL; - if (ctx_size_in < prog->aux->max_ctx_offset) + if (ctx_size_in < prog->aux->max_ctx_offset || + ctx_size_in > MAX_BPF_FUNC_ARGS * sizeof(u64)) return -EINVAL; if ((kattr->test.flags & BPF_F_TEST_RUN_ON_CPU) == 0 && cpu != 0) diff --git a/net/bpfilter/Kconfig b/net/bpfilter/Kconfig index 8ad0233ce497..3d4a21462458 100644 --- a/net/bpfilter/Kconfig +++ b/net/bpfilter/Kconfig @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only menuconfig BPFILTER bool "BPF based packet filtering framework (BPFILTER)" - depends on NET && BPF && INET + depends on BPF && INET select USERMODE_DRIVER help This builds experimental bpfilter framework that is aiming to diff --git a/net/bridge/Makefile b/net/bridge/Makefile index 4702702a74d3..7fb9a021873b 100644 --- a/net/bridge/Makefile +++ b/net/bridge/Makefile @@ -18,7 +18,7 @@ br_netfilter-y := br_netfilter_hooks.o br_netfilter-$(subst m,y,$(CONFIG_IPV6)) += br_netfilter_ipv6.o obj-$(CONFIG_BRIDGE_NETFILTER) += br_netfilter.o -bridge-$(CONFIG_BRIDGE_IGMP_SNOOPING) += br_multicast.o br_mdb.o +bridge-$(CONFIG_BRIDGE_IGMP_SNOOPING) += br_multicast.o br_mdb.o br_multicast_eht.o bridge-$(CONFIG_BRIDGE_VLAN_FILTERING) += br_vlan.o br_vlan_tunnel.o br_vlan_options.o diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 257ac4e25f6d..6f672eb7ff33 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -33,6 +33,7 @@ #endif #include "br_private.h" +#include "br_private_mcast_eht.h" static const struct rhashtable_params br_mdb_rht_params = { .head_offset = offsetof(struct net_bridge_mdb_entry, rhnode), @@ -441,7 +442,8 @@ static void br_multicast_fwd_src_add(struct net_bridge_group_src *src) br_multicast_sg_add_exclude_ports(star_mp, sg); } -static void br_multicast_fwd_src_remove(struct net_bridge_group_src *src) +static void br_multicast_fwd_src_remove(struct net_bridge_group_src *src, + bool fastleave) { struct net_bridge_port_group *p, *pg = src->pg; struct net_bridge_port_group __rcu **pp; @@ -466,6 +468,8 @@ static void br_multicast_fwd_src_remove(struct net_bridge_group_src *src) (p->flags & MDB_PG_FLAGS_PERMANENT)) break; + if (fastleave) + p->flags |= MDB_PG_FLAGS_FAST_LEAVE; br_multicast_del_pg(mp, p, pp); break; } @@ -559,11 +563,12 @@ static void br_multicast_destroy_group_src(struct net_bridge_mcast_gc *gc) kfree_rcu(src, rcu); } -static void br_multicast_del_group_src(struct net_bridge_group_src *src) +void br_multicast_del_group_src(struct net_bridge_group_src *src, + bool fastleave) { struct net_bridge *br = src->pg->key.port->br; - br_multicast_fwd_src_remove(src); + br_multicast_fwd_src_remove(src, fastleave); hlist_del_init_rcu(&src->node); src->pg->src_ents--; hlist_add_head(&src->mcast_gc.gc_node, &br->mcast_gc_list); @@ -593,8 +598,9 @@ void br_multicast_del_pg(struct net_bridge_mdb_entry *mp, rcu_assign_pointer(*pp, pg->next); hlist_del_init(&pg->mglist); + br_multicast_eht_clean_sets(pg); hlist_for_each_entry_safe(ent, tmp, &pg->src_list, node) - br_multicast_del_group_src(ent); + br_multicast_del_group_src(ent, false); br_mdb_notify(br->dev, mp, pg, RTM_DELMDB); if (!br_multicast_is_star_g(&mp->addr)) { rhashtable_remove_fast(&br->sg_port_tbl, &pg->rhnode, @@ -651,7 +657,7 @@ static void br_multicast_port_group_expired(struct timer_list *t) pg->filter_mode = MCAST_INCLUDE; hlist_for_each_entry_safe(src_ent, tmp, &pg->src_list, node) { if (!timer_pending(&src_ent->timer)) { - br_multicast_del_group_src(src_ent); + br_multicast_del_group_src(src_ent, false); changed = true; } } @@ -1078,7 +1084,7 @@ static void br_multicast_group_src_expired(struct timer_list *t) pg = src->pg; if (pg->filter_mode == MCAST_INCLUDE) { - br_multicast_del_group_src(src); + br_multicast_del_group_src(src, false); if (!hlist_empty(&pg->src_list)) goto out; br_multicast_find_del_pg(br, pg); @@ -1090,7 +1096,7 @@ out: spin_unlock(&br->multicast_lock); } -static struct net_bridge_group_src * +struct net_bridge_group_src * br_multicast_find_group_src(struct net_bridge_port_group *pg, struct br_ip *ip) { struct net_bridge_group_src *ent; @@ -1172,6 +1178,8 @@ struct net_bridge_port_group *br_multicast_new_port_group( p->flags = flags; p->filter_mode = filter_mode; p->rt_protocol = rt_protocol; + p->eht_host_tree = RB_ROOT; + p->eht_set_tree = RB_ROOT; p->mcast_gc.destroy = br_multicast_destroy_port_group; INIT_HLIST_HEAD(&p->src_list); @@ -1292,7 +1300,7 @@ static int br_multicast_add_group(struct net_bridge *br, pg = __br_multicast_add_group(br, port, group, src, filter_mode, igmpv2_mldv1, false); /* NULL is considered valid for host joined groups */ - err = IS_ERR(pg) ? PTR_ERR(pg) : 0; + err = PTR_ERR_OR_ZERO(pg); spin_unlock(&br->multicast_lock); return err; @@ -1600,6 +1608,7 @@ static void br_mc_disabled_update(struct net_device *dev, bool value) int br_multicast_add_port(struct net_bridge_port *port) { port->multicast_router = MDB_RTR_TYPE_TEMP_QUERY; + port->multicast_eht_hosts_limit = BR_MCAST_DEFAULT_EHT_HOSTS_LIMIT; timer_setup(&port->multicast_router_timer, br_multicast_router_expired, 0); @@ -1700,7 +1709,7 @@ static int __grp_src_delete_marked(struct net_bridge_port_group *pg) hlist_for_each_entry_safe(ent, tmp, &pg->src_list, node) if (ent->flags & BR_SGRP_F_DELETE) { - br_multicast_del_group_src(ent); + br_multicast_del_group_src(ent, false); deleted++; } @@ -1799,8 +1808,9 @@ static void __grp_send_query_and_rexmit(struct net_bridge_port_group *pg) * INCLUDE (A) ALLOW (B) INCLUDE (A+B) (B)=GMI * EXCLUDE (X,Y) ALLOW (A) EXCLUDE (X+A,Y-A) (A)=GMI */ -static bool br_multicast_isinc_allow(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool br_multicast_isinc_allow(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge *br = pg->key.port->br; struct net_bridge_group_src *ent; @@ -1812,7 +1822,7 @@ static bool br_multicast_isinc_allow(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (!ent) { ent = br_multicast_new_group_src(pg, &src_ip); @@ -1822,9 +1832,11 @@ static bool br_multicast_isinc_allow(struct net_bridge_port_group *pg, if (ent) __grp_src_mod_timer(ent, now + br_multicast_gmi(br)); - srcs += src_size; } + if (br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type)) + changed = true; + return changed; } @@ -1833,8 +1845,9 @@ static bool br_multicast_isinc_allow(struct net_bridge_port_group *pg, * Delete (A-B) * Group Timer=GMI */ -static void __grp_src_isexc_incl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static void __grp_src_isexc_incl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge_group_src *ent; struct br_ip src_ip; @@ -1846,7 +1859,7 @@ static void __grp_src_isexc_incl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (ent) ent->flags &= ~BR_SGRP_F_DELETE; @@ -1854,9 +1867,10 @@ static void __grp_src_isexc_incl(struct net_bridge_port_group *pg, ent = br_multicast_new_group_src(pg, &src_ip); if (ent) br_multicast_fwd_src_handle(ent); - srcs += src_size; } + br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type); + __grp_src_delete_marked(pg); } @@ -1866,8 +1880,9 @@ static void __grp_src_isexc_incl(struct net_bridge_port_group *pg, * Delete (Y-A) * Group Timer=GMI */ -static bool __grp_src_isexc_excl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool __grp_src_isexc_excl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge *br = pg->key.port->br; struct net_bridge_group_src *ent; @@ -1882,7 +1897,7 @@ static bool __grp_src_isexc_excl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (ent) { ent->flags &= ~BR_SGRP_F_DELETE; @@ -1894,29 +1909,34 @@ static bool __grp_src_isexc_excl(struct net_bridge_port_group *pg, changed = true; } } - srcs += src_size; } + if (br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type)) + changed = true; + if (__grp_src_delete_marked(pg)) changed = true; return changed; } -static bool br_multicast_isexc(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool br_multicast_isexc(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge *br = pg->key.port->br; bool changed = false; switch (pg->filter_mode) { case MCAST_INCLUDE: - __grp_src_isexc_incl(pg, srcs, nsrcs, src_size); + __grp_src_isexc_incl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); br_multicast_star_g_handle_mode(pg, MCAST_EXCLUDE); changed = true; break; case MCAST_EXCLUDE: - changed = __grp_src_isexc_excl(pg, srcs, nsrcs, src_size); + changed = __grp_src_isexc_excl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); break; } @@ -1930,8 +1950,9 @@ static bool br_multicast_isexc(struct net_bridge_port_group *pg, * INCLUDE (A) TO_IN (B) INCLUDE (A+B) (B)=GMI * Send Q(G,A-B) */ -static bool __grp_src_toin_incl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool __grp_src_toin_incl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge *br = pg->key.port->br; u32 src_idx, to_send = pg->src_ents; @@ -1946,7 +1967,7 @@ static bool __grp_src_toin_incl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (ent) { ent->flags &= ~BR_SGRP_F_SEND; @@ -1958,9 +1979,11 @@ static bool __grp_src_toin_incl(struct net_bridge_port_group *pg, } if (ent) __grp_src_mod_timer(ent, now + br_multicast_gmi(br)); - srcs += src_size; } + if (br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type)) + changed = true; + if (to_send) __grp_src_query_marked_and_rexmit(pg); @@ -1972,8 +1995,9 @@ static bool __grp_src_toin_incl(struct net_bridge_port_group *pg, * Send Q(G,X-A) * Send Q(G) */ -static bool __grp_src_toin_excl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool __grp_src_toin_excl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge *br = pg->key.port->br; u32 src_idx, to_send = pg->src_ents; @@ -1989,7 +2013,7 @@ static bool __grp_src_toin_excl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (ent) { if (timer_pending(&ent->timer)) { @@ -2003,9 +2027,11 @@ static bool __grp_src_toin_excl(struct net_bridge_port_group *pg, } if (ent) __grp_src_mod_timer(ent, now + br_multicast_gmi(br)); - srcs += src_size; } + if (br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type)) + changed = true; + if (to_send) __grp_src_query_marked_and_rexmit(pg); @@ -2014,20 +2040,32 @@ static bool __grp_src_toin_excl(struct net_bridge_port_group *pg, return changed; } -static bool br_multicast_toin(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool br_multicast_toin(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { bool changed = false; switch (pg->filter_mode) { case MCAST_INCLUDE: - changed = __grp_src_toin_incl(pg, srcs, nsrcs, src_size); + changed = __grp_src_toin_incl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); break; case MCAST_EXCLUDE: - changed = __grp_src_toin_excl(pg, srcs, nsrcs, src_size); + changed = __grp_src_toin_excl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); break; } + if (br_multicast_eht_should_del_pg(pg)) { + pg->flags |= MDB_PG_FLAGS_FAST_LEAVE; + br_multicast_find_del_pg(pg->key.port->br, pg); + /* a notification has already been sent and we shouldn't + * access pg after the delete so we have to return false + */ + changed = false; + } + return changed; } @@ -2037,8 +2075,9 @@ static bool br_multicast_toin(struct net_bridge_port_group *pg, * Send Q(G,A*B) * Group Timer=GMI */ -static void __grp_src_toex_incl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static void __grp_src_toex_incl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge_group_src *ent; u32 src_idx, to_send = 0; @@ -2050,7 +2089,7 @@ static void __grp_src_toex_incl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (ent) { ent->flags = (ent->flags & ~BR_SGRP_F_DELETE) | @@ -2061,9 +2100,10 @@ static void __grp_src_toex_incl(struct net_bridge_port_group *pg, } if (ent) br_multicast_fwd_src_handle(ent); - srcs += src_size; } + br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type); + __grp_src_delete_marked(pg); if (to_send) __grp_src_query_marked_and_rexmit(pg); @@ -2076,8 +2116,9 @@ static void __grp_src_toex_incl(struct net_bridge_port_group *pg, * Send Q(G,A-Y) * Group Timer=GMI */ -static bool __grp_src_toex_excl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool __grp_src_toex_excl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge_group_src *ent; u32 src_idx, to_send = 0; @@ -2090,7 +2131,7 @@ static bool __grp_src_toex_excl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (ent) { ent->flags &= ~BR_SGRP_F_DELETE; @@ -2105,9 +2146,11 @@ static bool __grp_src_toex_excl(struct net_bridge_port_group *pg, ent->flags |= BR_SGRP_F_SEND; to_send++; } - srcs += src_size; } + if (br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type)) + changed = true; + if (__grp_src_delete_marked(pg)) changed = true; if (to_send) @@ -2116,20 +2159,23 @@ static bool __grp_src_toex_excl(struct net_bridge_port_group *pg, return changed; } -static bool br_multicast_toex(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool br_multicast_toex(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, + int grec_type) { struct net_bridge *br = pg->key.port->br; bool changed = false; switch (pg->filter_mode) { case MCAST_INCLUDE: - __grp_src_toex_incl(pg, srcs, nsrcs, src_size); + __grp_src_toex_incl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); br_multicast_star_g_handle_mode(pg, MCAST_EXCLUDE); changed = true; break; case MCAST_EXCLUDE: - changed = __grp_src_toex_excl(pg, srcs, nsrcs, src_size); + changed = __grp_src_toex_excl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); break; } @@ -2142,11 +2188,12 @@ static bool br_multicast_toex(struct net_bridge_port_group *pg, /* State Msg type New state Actions * INCLUDE (A) BLOCK (B) INCLUDE (A) Send Q(G,A*B) */ -static void __grp_src_block_incl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool __grp_src_block_incl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, int grec_type) { struct net_bridge_group_src *ent; u32 src_idx, to_send = 0; + bool changed = false; struct br_ip src_ip; hlist_for_each_entry(ent, &pg->src_list, node) @@ -2155,28 +2202,29 @@ static void __grp_src_block_incl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (ent) { ent->flags |= BR_SGRP_F_SEND; to_send++; } - srcs += src_size; } + if (br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type)) + changed = true; + if (to_send) __grp_src_query_marked_and_rexmit(pg); - if (pg->filter_mode == MCAST_INCLUDE && hlist_empty(&pg->src_list)) - br_multicast_find_del_pg(pg->key.port->br, pg); + return changed; } /* State Msg type New state Actions * EXCLUDE (X,Y) BLOCK (A) EXCLUDE (X+(A-Y),Y) (A-X-Y)=Group Timer * Send Q(G,A-Y) */ -static bool __grp_src_block_excl(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool __grp_src_block_excl(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, int grec_type) { struct net_bridge_group_src *ent; u32 src_idx, to_send = 0; @@ -2189,7 +2237,7 @@ static bool __grp_src_block_excl(struct net_bridge_port_group *pg, memset(&src_ip, 0, sizeof(src_ip)); src_ip.proto = pg->key.addr.proto; for (src_idx = 0; src_idx < nsrcs; src_idx++) { - memcpy(&src_ip.src, srcs, src_size); + memcpy(&src_ip.src, srcs + (src_idx * addr_size), addr_size); ent = br_multicast_find_group_src(pg, &src_ip); if (!ent) { ent = br_multicast_new_group_src(pg, &src_ip); @@ -2202,29 +2250,44 @@ static bool __grp_src_block_excl(struct net_bridge_port_group *pg, ent->flags |= BR_SGRP_F_SEND; to_send++; } - srcs += src_size; } + if (br_multicast_eht_handle(pg, h_addr, srcs, nsrcs, addr_size, grec_type)) + changed = true; + if (to_send) __grp_src_query_marked_and_rexmit(pg); return changed; } -static bool br_multicast_block(struct net_bridge_port_group *pg, - void *srcs, u32 nsrcs, size_t src_size) +static bool br_multicast_block(struct net_bridge_port_group *pg, void *h_addr, + void *srcs, u32 nsrcs, size_t addr_size, int grec_type) { bool changed = false; switch (pg->filter_mode) { case MCAST_INCLUDE: - __grp_src_block_incl(pg, srcs, nsrcs, src_size); + changed = __grp_src_block_incl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); break; case MCAST_EXCLUDE: - changed = __grp_src_block_excl(pg, srcs, nsrcs, src_size); + changed = __grp_src_block_excl(pg, h_addr, srcs, nsrcs, addr_size, + grec_type); break; } + if ((pg->filter_mode == MCAST_INCLUDE && hlist_empty(&pg->src_list)) || + br_multicast_eht_should_del_pg(pg)) { + if (br_multicast_eht_should_del_pg(pg)) + pg->flags |= MDB_PG_FLAGS_FAST_LEAVE; + br_multicast_find_del_pg(pg->key.port->br, pg); + /* a notification has already been sent and we shouldn't + * access pg after the delete so we have to return false + */ + changed = false; + } + return changed; } @@ -2257,8 +2320,8 @@ static int br_ip4_multicast_igmp3_report(struct net_bridge *br, struct igmpv3_report *ih; struct igmpv3_grec *grec; int i, len, num, type; + __be32 group, *h_addr; bool changed = false; - __be32 group; int err = 0; u16 nsrcs; @@ -2318,32 +2381,33 @@ static int br_ip4_multicast_igmp3_report(struct net_bridge *br, pg = br_multicast_find_port(mdst, port, src); if (!pg || (pg->flags & MDB_PG_FLAGS_PERMANENT)) goto unlock_continue; - /* reload grec */ + /* reload grec and host addr */ grec = (void *)(skb->data + len - sizeof(*grec) - (nsrcs * 4)); + h_addr = &ip_hdr(skb)->saddr; switch (type) { case IGMPV3_ALLOW_NEW_SOURCES: - changed = br_multicast_isinc_allow(pg, grec->grec_src, - nsrcs, sizeof(__be32)); + changed = br_multicast_isinc_allow(pg, h_addr, grec->grec_src, + nsrcs, sizeof(__be32), type); break; case IGMPV3_MODE_IS_INCLUDE: - changed = br_multicast_isinc_allow(pg, grec->grec_src, nsrcs, - sizeof(__be32)); + changed = br_multicast_isinc_allow(pg, h_addr, grec->grec_src, + nsrcs, sizeof(__be32), type); break; case IGMPV3_MODE_IS_EXCLUDE: - changed = br_multicast_isexc(pg, grec->grec_src, nsrcs, - sizeof(__be32)); + changed = br_multicast_isexc(pg, h_addr, grec->grec_src, + nsrcs, sizeof(__be32), type); break; case IGMPV3_CHANGE_TO_INCLUDE: - changed = br_multicast_toin(pg, grec->grec_src, nsrcs, - sizeof(__be32)); + changed = br_multicast_toin(pg, h_addr, grec->grec_src, + nsrcs, sizeof(__be32), type); break; case IGMPV3_CHANGE_TO_EXCLUDE: - changed = br_multicast_toex(pg, grec->grec_src, nsrcs, - sizeof(__be32)); + changed = br_multicast_toex(pg, h_addr, grec->grec_src, + nsrcs, sizeof(__be32), type); break; case IGMPV3_BLOCK_OLD_SOURCES: - changed = br_multicast_block(pg, grec->grec_src, nsrcs, - sizeof(__be32)); + changed = br_multicast_block(pg, h_addr, grec->grec_src, + nsrcs, sizeof(__be32), type); break; } if (changed) @@ -2367,6 +2431,7 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, unsigned int nsrcs_offset; const unsigned char *src; struct icmp6hdr *icmp6h; + struct in6_addr *h_addr; struct mld2_grec *grec; unsigned int grec_len; bool changed = false; @@ -2445,31 +2510,43 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, pg = br_multicast_find_port(mdst, port, src); if (!pg || (pg->flags & MDB_PG_FLAGS_PERMANENT)) goto unlock_continue; + h_addr = &ipv6_hdr(skb)->saddr; switch (grec->grec_type) { case MLD2_ALLOW_NEW_SOURCES: - changed = br_multicast_isinc_allow(pg, grec->grec_src, - nsrcs, - sizeof(struct in6_addr)); + changed = br_multicast_isinc_allow(pg, h_addr, + grec->grec_src, nsrcs, + sizeof(struct in6_addr), + grec->grec_type); break; case MLD2_MODE_IS_INCLUDE: - changed = br_multicast_isinc_allow(pg, grec->grec_src, nsrcs, - sizeof(struct in6_addr)); + changed = br_multicast_isinc_allow(pg, h_addr, + grec->grec_src, nsrcs, + sizeof(struct in6_addr), + grec->grec_type); break; case MLD2_MODE_IS_EXCLUDE: - changed = br_multicast_isexc(pg, grec->grec_src, nsrcs, - sizeof(struct in6_addr)); + changed = br_multicast_isexc(pg, h_addr, + grec->grec_src, nsrcs, + sizeof(struct in6_addr), + grec->grec_type); break; case MLD2_CHANGE_TO_INCLUDE: - changed = br_multicast_toin(pg, grec->grec_src, nsrcs, - sizeof(struct in6_addr)); + changed = br_multicast_toin(pg, h_addr, + grec->grec_src, nsrcs, + sizeof(struct in6_addr), + grec->grec_type); break; case MLD2_CHANGE_TO_EXCLUDE: - changed = br_multicast_toex(pg, grec->grec_src, nsrcs, - sizeof(struct in6_addr)); + changed = br_multicast_toex(pg, h_addr, + grec->grec_src, nsrcs, + sizeof(struct in6_addr), + grec->grec_type); break; case MLD2_BLOCK_OLD_SOURCES: - changed = br_multicast_block(pg, grec->grec_src, nsrcs, - sizeof(struct in6_addr)); + changed = br_multicast_block(pg, h_addr, + grec->grec_src, nsrcs, + sizeof(struct in6_addr), + grec->grec_type); break; } if (changed) diff --git a/net/bridge/br_multicast_eht.c b/net/bridge/br_multicast_eht.c new file mode 100644 index 000000000000..fea38b9a7268 --- /dev/null +++ b/net/bridge/br_multicast_eht.c @@ -0,0 +1,878 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +// Copyright (c) 2020, Nikolay Aleksandrov <nikolay@nvidia.com> +#include <linux/err.h> +#include <linux/export.h> +#include <linux/if_ether.h> +#include <linux/igmp.h> +#include <linux/in.h> +#include <linux/jhash.h> +#include <linux/kernel.h> +#include <linux/log2.h> +#include <linux/netdevice.h> +#include <linux/netfilter_bridge.h> +#include <linux/random.h> +#include <linux/rculist.h> +#include <linux/skbuff.h> +#include <linux/slab.h> +#include <linux/timer.h> +#include <linux/inetdevice.h> +#include <linux/mroute.h> +#include <net/ip.h> +#include <net/switchdev.h> +#if IS_ENABLED(CONFIG_IPV6) +#include <linux/icmpv6.h> +#include <net/ipv6.h> +#include <net/mld.h> +#include <net/ip6_checksum.h> +#include <net/addrconf.h> +#endif + +#include "br_private.h" +#include "br_private_mcast_eht.h" + +static bool br_multicast_del_eht_set_entry(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *src_addr, + union net_bridge_eht_addr *h_addr); +static void br_multicast_create_eht_set_entry(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *src_addr, + union net_bridge_eht_addr *h_addr, + int filter_mode, + bool allow_zero_src); + +static struct net_bridge_group_eht_host * +br_multicast_eht_host_lookup(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr) +{ + struct rb_node *node = pg->eht_host_tree.rb_node; + + while (node) { + struct net_bridge_group_eht_host *this; + int result; + + this = rb_entry(node, struct net_bridge_group_eht_host, + rb_node); + result = memcmp(h_addr, &this->h_addr, sizeof(*h_addr)); + if (result < 0) + node = node->rb_left; + else if (result > 0) + node = node->rb_right; + else + return this; + } + + return NULL; +} + +static int br_multicast_eht_host_filter_mode(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr) +{ + struct net_bridge_group_eht_host *eht_host; + + eht_host = br_multicast_eht_host_lookup(pg, h_addr); + if (!eht_host) + return MCAST_INCLUDE; + + return eht_host->filter_mode; +} + +static struct net_bridge_group_eht_set_entry * +br_multicast_eht_set_entry_lookup(struct net_bridge_group_eht_set *eht_set, + union net_bridge_eht_addr *h_addr) +{ + struct rb_node *node = eht_set->entry_tree.rb_node; + + while (node) { + struct net_bridge_group_eht_set_entry *this; + int result; + + this = rb_entry(node, struct net_bridge_group_eht_set_entry, + rb_node); + result = memcmp(h_addr, &this->h_addr, sizeof(*h_addr)); + if (result < 0) + node = node->rb_left; + else if (result > 0) + node = node->rb_right; + else + return this; + } + + return NULL; +} + +static struct net_bridge_group_eht_set * +br_multicast_eht_set_lookup(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *src_addr) +{ + struct rb_node *node = pg->eht_set_tree.rb_node; + + while (node) { + struct net_bridge_group_eht_set *this; + int result; + + this = rb_entry(node, struct net_bridge_group_eht_set, + rb_node); + result = memcmp(src_addr, &this->src_addr, sizeof(*src_addr)); + if (result < 0) + node = node->rb_left; + else if (result > 0) + node = node->rb_right; + else + return this; + } + + return NULL; +} + +static void __eht_destroy_host(struct net_bridge_group_eht_host *eht_host) +{ + WARN_ON(!hlist_empty(&eht_host->set_entries)); + + br_multicast_eht_hosts_dec(eht_host->pg); + + rb_erase(&eht_host->rb_node, &eht_host->pg->eht_host_tree); + RB_CLEAR_NODE(&eht_host->rb_node); + kfree(eht_host); +} + +static void br_multicast_destroy_eht_set_entry(struct net_bridge_mcast_gc *gc) +{ + struct net_bridge_group_eht_set_entry *set_h; + + set_h = container_of(gc, struct net_bridge_group_eht_set_entry, mcast_gc); + WARN_ON(!RB_EMPTY_NODE(&set_h->rb_node)); + + del_timer_sync(&set_h->timer); + kfree(set_h); +} + +static void br_multicast_destroy_eht_set(struct net_bridge_mcast_gc *gc) +{ + struct net_bridge_group_eht_set *eht_set; + + eht_set = container_of(gc, struct net_bridge_group_eht_set, mcast_gc); + WARN_ON(!RB_EMPTY_NODE(&eht_set->rb_node)); + WARN_ON(!RB_EMPTY_ROOT(&eht_set->entry_tree)); + + del_timer_sync(&eht_set->timer); + kfree(eht_set); +} + +static void __eht_del_set_entry(struct net_bridge_group_eht_set_entry *set_h) +{ + struct net_bridge_group_eht_host *eht_host = set_h->h_parent; + union net_bridge_eht_addr zero_addr; + + rb_erase(&set_h->rb_node, &set_h->eht_set->entry_tree); + RB_CLEAR_NODE(&set_h->rb_node); + hlist_del_init(&set_h->host_list); + memset(&zero_addr, 0, sizeof(zero_addr)); + if (memcmp(&set_h->h_addr, &zero_addr, sizeof(zero_addr))) + eht_host->num_entries--; + hlist_add_head(&set_h->mcast_gc.gc_node, &set_h->br->mcast_gc_list); + queue_work(system_long_wq, &set_h->br->mcast_gc_work); + + if (hlist_empty(&eht_host->set_entries)) + __eht_destroy_host(eht_host); +} + +static void br_multicast_del_eht_set(struct net_bridge_group_eht_set *eht_set) +{ + struct net_bridge_group_eht_set_entry *set_h; + struct rb_node *node; + + while ((node = rb_first(&eht_set->entry_tree))) { + set_h = rb_entry(node, struct net_bridge_group_eht_set_entry, + rb_node); + __eht_del_set_entry(set_h); + } + + rb_erase(&eht_set->rb_node, &eht_set->pg->eht_set_tree); + RB_CLEAR_NODE(&eht_set->rb_node); + hlist_add_head(&eht_set->mcast_gc.gc_node, &eht_set->br->mcast_gc_list); + queue_work(system_long_wq, &eht_set->br->mcast_gc_work); +} + +void br_multicast_eht_clean_sets(struct net_bridge_port_group *pg) +{ + struct net_bridge_group_eht_set *eht_set; + struct rb_node *node; + + while ((node = rb_first(&pg->eht_set_tree))) { + eht_set = rb_entry(node, struct net_bridge_group_eht_set, + rb_node); + br_multicast_del_eht_set(eht_set); + } +} + +static void br_multicast_eht_set_entry_expired(struct timer_list *t) +{ + struct net_bridge_group_eht_set_entry *set_h = from_timer(set_h, t, timer); + struct net_bridge *br = set_h->br; + + spin_lock(&br->multicast_lock); + if (RB_EMPTY_NODE(&set_h->rb_node) || timer_pending(&set_h->timer)) + goto out; + + br_multicast_del_eht_set_entry(set_h->eht_set->pg, + &set_h->eht_set->src_addr, + &set_h->h_addr); +out: + spin_unlock(&br->multicast_lock); +} + +static void br_multicast_eht_set_expired(struct timer_list *t) +{ + struct net_bridge_group_eht_set *eht_set = from_timer(eht_set, t, + timer); + struct net_bridge *br = eht_set->br; + + spin_lock(&br->multicast_lock); + if (RB_EMPTY_NODE(&eht_set->rb_node) || timer_pending(&eht_set->timer)) + goto out; + + br_multicast_del_eht_set(eht_set); +out: + spin_unlock(&br->multicast_lock); +} + +static struct net_bridge_group_eht_host * +__eht_lookup_create_host(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + unsigned char filter_mode) +{ + struct rb_node **link = &pg->eht_host_tree.rb_node, *parent = NULL; + struct net_bridge_group_eht_host *eht_host; + + while (*link) { + struct net_bridge_group_eht_host *this; + int result; + + this = rb_entry(*link, struct net_bridge_group_eht_host, + rb_node); + result = memcmp(h_addr, &this->h_addr, sizeof(*h_addr)); + parent = *link; + if (result < 0) + link = &((*link)->rb_left); + else if (result > 0) + link = &((*link)->rb_right); + else + return this; + } + + if (br_multicast_eht_hosts_over_limit(pg)) + return NULL; + + eht_host = kzalloc(sizeof(*eht_host), GFP_ATOMIC); + if (!eht_host) + return NULL; + + memcpy(&eht_host->h_addr, h_addr, sizeof(*h_addr)); + INIT_HLIST_HEAD(&eht_host->set_entries); + eht_host->pg = pg; + eht_host->filter_mode = filter_mode; + + rb_link_node(&eht_host->rb_node, parent, link); + rb_insert_color(&eht_host->rb_node, &pg->eht_host_tree); + + br_multicast_eht_hosts_inc(pg); + + return eht_host; +} + +static struct net_bridge_group_eht_set_entry * +__eht_lookup_create_set_entry(struct net_bridge *br, + struct net_bridge_group_eht_set *eht_set, + struct net_bridge_group_eht_host *eht_host, + bool allow_zero_src) +{ + struct rb_node **link = &eht_set->entry_tree.rb_node, *parent = NULL; + struct net_bridge_group_eht_set_entry *set_h; + + while (*link) { + struct net_bridge_group_eht_set_entry *this; + int result; + + this = rb_entry(*link, struct net_bridge_group_eht_set_entry, + rb_node); + result = memcmp(&eht_host->h_addr, &this->h_addr, + sizeof(union net_bridge_eht_addr)); + parent = *link; + if (result < 0) + link = &((*link)->rb_left); + else if (result > 0) + link = &((*link)->rb_right); + else + return this; + } + + /* always allow auto-created zero entry */ + if (!allow_zero_src && eht_host->num_entries >= PG_SRC_ENT_LIMIT) + return NULL; + + set_h = kzalloc(sizeof(*set_h), GFP_ATOMIC); + if (!set_h) + return NULL; + + memcpy(&set_h->h_addr, &eht_host->h_addr, + sizeof(union net_bridge_eht_addr)); + set_h->mcast_gc.destroy = br_multicast_destroy_eht_set_entry; + set_h->eht_set = eht_set; + set_h->h_parent = eht_host; + set_h->br = br; + timer_setup(&set_h->timer, br_multicast_eht_set_entry_expired, 0); + + hlist_add_head(&set_h->host_list, &eht_host->set_entries); + rb_link_node(&set_h->rb_node, parent, link); + rb_insert_color(&set_h->rb_node, &eht_set->entry_tree); + /* we must not count the auto-created zero entry otherwise we won't be + * able to track the full list of PG_SRC_ENT_LIMIT entries + */ + if (!allow_zero_src) + eht_host->num_entries++; + + return set_h; +} + +static struct net_bridge_group_eht_set * +__eht_lookup_create_set(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *src_addr) +{ + struct rb_node **link = &pg->eht_set_tree.rb_node, *parent = NULL; + struct net_bridge_group_eht_set *eht_set; + + while (*link) { + struct net_bridge_group_eht_set *this; + int result; + + this = rb_entry(*link, struct net_bridge_group_eht_set, + rb_node); + result = memcmp(src_addr, &this->src_addr, sizeof(*src_addr)); + parent = *link; + if (result < 0) + link = &((*link)->rb_left); + else if (result > 0) + link = &((*link)->rb_right); + else + return this; + } + + eht_set = kzalloc(sizeof(*eht_set), GFP_ATOMIC); + if (!eht_set) + return NULL; + + memcpy(&eht_set->src_addr, src_addr, sizeof(*src_addr)); + eht_set->mcast_gc.destroy = br_multicast_destroy_eht_set; + eht_set->pg = pg; + eht_set->br = pg->key.port->br; + eht_set->entry_tree = RB_ROOT; + timer_setup(&eht_set->timer, br_multicast_eht_set_expired, 0); + + rb_link_node(&eht_set->rb_node, parent, link); + rb_insert_color(&eht_set->rb_node, &pg->eht_set_tree); + + return eht_set; +} + +static void br_multicast_ip_src_to_eht_addr(const struct br_ip *src, + union net_bridge_eht_addr *dest) +{ + switch (src->proto) { + case htons(ETH_P_IP): + dest->ip4 = src->src.ip4; + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + memcpy(&dest->ip6, &src->src.ip6, sizeof(struct in6_addr)); + break; +#endif + } +} + +static void br_eht_convert_host_filter_mode(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + int filter_mode) +{ + struct net_bridge_group_eht_host *eht_host; + union net_bridge_eht_addr zero_addr; + + eht_host = br_multicast_eht_host_lookup(pg, h_addr); + if (eht_host) + eht_host->filter_mode = filter_mode; + + memset(&zero_addr, 0, sizeof(zero_addr)); + switch (filter_mode) { + case MCAST_INCLUDE: + br_multicast_del_eht_set_entry(pg, &zero_addr, h_addr); + break; + case MCAST_EXCLUDE: + br_multicast_create_eht_set_entry(pg, &zero_addr, h_addr, + MCAST_EXCLUDE, + true); + break; + } +} + +static void br_multicast_create_eht_set_entry(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *src_addr, + union net_bridge_eht_addr *h_addr, + int filter_mode, + bool allow_zero_src) +{ + struct net_bridge_group_eht_set_entry *set_h; + struct net_bridge_group_eht_host *eht_host; + struct net_bridge *br = pg->key.port->br; + struct net_bridge_group_eht_set *eht_set; + union net_bridge_eht_addr zero_addr; + + memset(&zero_addr, 0, sizeof(zero_addr)); + if (!allow_zero_src && !memcmp(src_addr, &zero_addr, sizeof(zero_addr))) + return; + + eht_set = __eht_lookup_create_set(pg, src_addr); + if (!eht_set) + return; + + eht_host = __eht_lookup_create_host(pg, h_addr, filter_mode); + if (!eht_host) + goto fail_host; + + set_h = __eht_lookup_create_set_entry(br, eht_set, eht_host, + allow_zero_src); + if (!set_h) + goto fail_set_entry; + + mod_timer(&set_h->timer, jiffies + br_multicast_gmi(br)); + mod_timer(&eht_set->timer, jiffies + br_multicast_gmi(br)); + + return; + +fail_set_entry: + if (hlist_empty(&eht_host->set_entries)) + __eht_destroy_host(eht_host); +fail_host: + if (RB_EMPTY_ROOT(&eht_set->entry_tree)) + br_multicast_del_eht_set(eht_set); +} + +static bool br_multicast_del_eht_set_entry(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *src_addr, + union net_bridge_eht_addr *h_addr) +{ + struct net_bridge_group_eht_set_entry *set_h; + struct net_bridge_group_eht_set *eht_set; + bool set_deleted = false; + + eht_set = br_multicast_eht_set_lookup(pg, src_addr); + if (!eht_set) + goto out; + + set_h = br_multicast_eht_set_entry_lookup(eht_set, h_addr); + if (!set_h) + goto out; + + __eht_del_set_entry(set_h); + + if (RB_EMPTY_ROOT(&eht_set->entry_tree)) { + br_multicast_del_eht_set(eht_set); + set_deleted = true; + } + +out: + return set_deleted; +} + +static void br_multicast_del_eht_host(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr) +{ + struct net_bridge_group_eht_set_entry *set_h; + struct net_bridge_group_eht_host *eht_host; + struct hlist_node *tmp; + + eht_host = br_multicast_eht_host_lookup(pg, h_addr); + if (!eht_host) + return; + + hlist_for_each_entry_safe(set_h, tmp, &eht_host->set_entries, host_list) + br_multicast_del_eht_set_entry(set_h->eht_set->pg, + &set_h->eht_set->src_addr, + &set_h->h_addr); +} + +static void __eht_allow_incl(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size) +{ + union net_bridge_eht_addr eht_src_addr; + u32 src_idx; + + memset(&eht_src_addr, 0, sizeof(eht_src_addr)); + for (src_idx = 0; src_idx < nsrcs; src_idx++) { + memcpy(&eht_src_addr, srcs + (src_idx * addr_size), addr_size); + br_multicast_create_eht_set_entry(pg, &eht_src_addr, h_addr, + MCAST_INCLUDE, + false); + } +} + +static bool __eht_allow_excl(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size) +{ + bool changed = false, host_excl = false; + union net_bridge_eht_addr eht_src_addr; + struct net_bridge_group_src *src_ent; + struct br_ip src_ip; + u32 src_idx; + + host_excl = !!(br_multicast_eht_host_filter_mode(pg, h_addr) == MCAST_EXCLUDE); + memset(&eht_src_addr, 0, sizeof(eht_src_addr)); + for (src_idx = 0; src_idx < nsrcs; src_idx++) { + memcpy(&eht_src_addr, srcs + (src_idx * addr_size), addr_size); + if (!host_excl) { + br_multicast_create_eht_set_entry(pg, &eht_src_addr, h_addr, + MCAST_INCLUDE, + false); + } else { + if (!br_multicast_del_eht_set_entry(pg, &eht_src_addr, + h_addr)) + continue; + memcpy(&src_ip, srcs + (src_idx * addr_size), addr_size); + src_ent = br_multicast_find_group_src(pg, &src_ip); + if (!src_ent) + continue; + br_multicast_del_group_src(src_ent, true); + changed = true; + } + } + + return changed; +} + +static bool br_multicast_eht_allow(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size) +{ + bool changed = false; + + switch (br_multicast_eht_host_filter_mode(pg, h_addr)) { + case MCAST_INCLUDE: + __eht_allow_incl(pg, h_addr, srcs, nsrcs, addr_size); + break; + case MCAST_EXCLUDE: + changed = __eht_allow_excl(pg, h_addr, srcs, nsrcs, addr_size); + break; + } + + return changed; +} + +static bool __eht_block_incl(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size) +{ + union net_bridge_eht_addr eht_src_addr; + struct net_bridge_group_src *src_ent; + bool changed = false; + struct br_ip src_ip; + u32 src_idx; + + memset(&eht_src_addr, 0, sizeof(eht_src_addr)); + memset(&src_ip, 0, sizeof(src_ip)); + src_ip.proto = pg->key.addr.proto; + for (src_idx = 0; src_idx < nsrcs; src_idx++) { + memcpy(&eht_src_addr, srcs + (src_idx * addr_size), addr_size); + if (!br_multicast_del_eht_set_entry(pg, &eht_src_addr, h_addr)) + continue; + memcpy(&src_ip, srcs + (src_idx * addr_size), addr_size); + src_ent = br_multicast_find_group_src(pg, &src_ip); + if (!src_ent) + continue; + br_multicast_del_group_src(src_ent, true); + changed = true; + } + + return changed; +} + +static bool __eht_block_excl(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size) +{ + bool changed = false, host_excl = false; + union net_bridge_eht_addr eht_src_addr; + struct net_bridge_group_src *src_ent; + struct br_ip src_ip; + u32 src_idx; + + host_excl = !!(br_multicast_eht_host_filter_mode(pg, h_addr) == MCAST_EXCLUDE); + memset(&eht_src_addr, 0, sizeof(eht_src_addr)); + memset(&src_ip, 0, sizeof(src_ip)); + src_ip.proto = pg->key.addr.proto; + for (src_idx = 0; src_idx < nsrcs; src_idx++) { + memcpy(&eht_src_addr, srcs + (src_idx * addr_size), addr_size); + if (host_excl) { + br_multicast_create_eht_set_entry(pg, &eht_src_addr, h_addr, + MCAST_EXCLUDE, + false); + } else { + if (!br_multicast_del_eht_set_entry(pg, &eht_src_addr, + h_addr)) + continue; + memcpy(&src_ip, srcs + (src_idx * addr_size), addr_size); + src_ent = br_multicast_find_group_src(pg, &src_ip); + if (!src_ent) + continue; + br_multicast_del_group_src(src_ent, true); + changed = true; + } + } + + return changed; +} + +static bool br_multicast_eht_block(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size) +{ + bool changed = false; + + switch (br_multicast_eht_host_filter_mode(pg, h_addr)) { + case MCAST_INCLUDE: + changed = __eht_block_incl(pg, h_addr, srcs, nsrcs, addr_size); + break; + case MCAST_EXCLUDE: + changed = __eht_block_excl(pg, h_addr, srcs, nsrcs, addr_size); + break; + } + + return changed; +} + +/* flush_entries is true when changing mode */ +static bool __eht_inc_exc(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size, + unsigned char filter_mode, + bool to_report) +{ + bool changed = false, flush_entries = to_report; + union net_bridge_eht_addr eht_src_addr; + u32 src_idx; + + if (br_multicast_eht_host_filter_mode(pg, h_addr) != filter_mode) + flush_entries = true; + + memset(&eht_src_addr, 0, sizeof(eht_src_addr)); + /* if we're changing mode del host and its entries */ + if (flush_entries) + br_multicast_del_eht_host(pg, h_addr); + for (src_idx = 0; src_idx < nsrcs; src_idx++) { + memcpy(&eht_src_addr, srcs + (src_idx * addr_size), addr_size); + br_multicast_create_eht_set_entry(pg, &eht_src_addr, h_addr, + filter_mode, false); + } + /* we can be missing sets only if we've deleted some entries */ + if (flush_entries) { + struct net_bridge *br = pg->key.port->br; + struct net_bridge_group_eht_set *eht_set; + struct net_bridge_group_src *src_ent; + struct hlist_node *tmp; + + hlist_for_each_entry_safe(src_ent, tmp, &pg->src_list, node) { + br_multicast_ip_src_to_eht_addr(&src_ent->addr, + &eht_src_addr); + if (!br_multicast_eht_set_lookup(pg, &eht_src_addr)) { + br_multicast_del_group_src(src_ent, true); + changed = true; + continue; + } + /* this is an optimization for TO_INCLUDE where we lower + * the set's timeout to LMQT to catch timeout hosts: + * - host A (timing out): set entries X, Y + * - host B: set entry Z (new from current TO_INCLUDE) + * sends BLOCK Z after LMQT but host A's EHT + * entries still exist (unless lowered to LMQT + * so they can timeout with the S,Gs) + * => we wait another LMQT, when we can just delete the + * group immediately + */ + if (!(src_ent->flags & BR_SGRP_F_SEND) || + filter_mode != MCAST_INCLUDE || + !to_report) + continue; + eht_set = br_multicast_eht_set_lookup(pg, + &eht_src_addr); + if (!eht_set) + continue; + mod_timer(&eht_set->timer, jiffies + br_multicast_lmqt(br)); + } + } + + return changed; +} + +static bool br_multicast_eht_inc(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size, + bool to_report) +{ + bool changed; + + changed = __eht_inc_exc(pg, h_addr, srcs, nsrcs, addr_size, + MCAST_INCLUDE, to_report); + br_eht_convert_host_filter_mode(pg, h_addr, MCAST_INCLUDE); + + return changed; +} + +static bool br_multicast_eht_exc(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size, + bool to_report) +{ + bool changed; + + changed = __eht_inc_exc(pg, h_addr, srcs, nsrcs, addr_size, + MCAST_EXCLUDE, to_report); + br_eht_convert_host_filter_mode(pg, h_addr, MCAST_EXCLUDE); + + return changed; +} + +static bool __eht_ip4_handle(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + int grec_type) +{ + bool changed = false, to_report = false; + + switch (grec_type) { + case IGMPV3_ALLOW_NEW_SOURCES: + br_multicast_eht_allow(pg, h_addr, srcs, nsrcs, sizeof(__be32)); + break; + case IGMPV3_BLOCK_OLD_SOURCES: + changed = br_multicast_eht_block(pg, h_addr, srcs, nsrcs, + sizeof(__be32)); + break; + case IGMPV3_CHANGE_TO_INCLUDE: + to_report = true; + fallthrough; + case IGMPV3_MODE_IS_INCLUDE: + changed = br_multicast_eht_inc(pg, h_addr, srcs, nsrcs, + sizeof(__be32), to_report); + break; + case IGMPV3_CHANGE_TO_EXCLUDE: + to_report = true; + fallthrough; + case IGMPV3_MODE_IS_EXCLUDE: + changed = br_multicast_eht_exc(pg, h_addr, srcs, nsrcs, + sizeof(__be32), to_report); + break; + } + + return changed; +} + +#if IS_ENABLED(CONFIG_IPV6) +static bool __eht_ip6_handle(struct net_bridge_port_group *pg, + union net_bridge_eht_addr *h_addr, + void *srcs, + u32 nsrcs, + int grec_type) +{ + bool changed = false, to_report = false; + + switch (grec_type) { + case MLD2_ALLOW_NEW_SOURCES: + br_multicast_eht_allow(pg, h_addr, srcs, nsrcs, + sizeof(struct in6_addr)); + break; + case MLD2_BLOCK_OLD_SOURCES: + changed = br_multicast_eht_block(pg, h_addr, srcs, nsrcs, + sizeof(struct in6_addr)); + break; + case MLD2_CHANGE_TO_INCLUDE: + to_report = true; + fallthrough; + case MLD2_MODE_IS_INCLUDE: + changed = br_multicast_eht_inc(pg, h_addr, srcs, nsrcs, + sizeof(struct in6_addr), + to_report); + break; + case MLD2_CHANGE_TO_EXCLUDE: + to_report = true; + fallthrough; + case MLD2_MODE_IS_EXCLUDE: + changed = br_multicast_eht_exc(pg, h_addr, srcs, nsrcs, + sizeof(struct in6_addr), + to_report); + break; + } + + return changed; +} +#endif + +/* true means an entry was deleted */ +bool br_multicast_eht_handle(struct net_bridge_port_group *pg, + void *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size, + int grec_type) +{ + bool eht_enabled = !!(pg->key.port->flags & BR_MULTICAST_FAST_LEAVE); + union net_bridge_eht_addr eht_host_addr; + bool changed = false; + + if (!eht_enabled) + goto out; + + memset(&eht_host_addr, 0, sizeof(eht_host_addr)); + memcpy(&eht_host_addr, h_addr, addr_size); + if (addr_size == sizeof(__be32)) + changed = __eht_ip4_handle(pg, &eht_host_addr, srcs, nsrcs, + grec_type); +#if IS_ENABLED(CONFIG_IPV6) + else + changed = __eht_ip6_handle(pg, &eht_host_addr, srcs, nsrcs, + grec_type); +#endif + +out: + return changed; +} + +int br_multicast_eht_set_hosts_limit(struct net_bridge_port *p, + u32 eht_hosts_limit) +{ + struct net_bridge *br = p->br; + + if (!eht_hosts_limit) + return -EINVAL; + + spin_lock_bh(&br->multicast_lock); + p->multicast_eht_hosts_limit = eht_hosts_limit; + spin_unlock_bh(&br->multicast_lock); + + return 0; +} diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 762f273802cd..bd3962da345a 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -18,6 +18,7 @@ #include "br_private_stp.h" #include "br_private_cfm.h" #include "br_private_tunnel.h" +#include "br_private_mcast_eht.h" static int __get_num_vlan_infos(struct net_bridge_vlan_group *vg, u32 filter_mask) @@ -199,6 +200,8 @@ static inline size_t br_port_info_size(void) + nla_total_size(sizeof(u16)) /* IFLA_BRPORT_GROUP_FWD_MASK */ + nla_total_size(sizeof(u8)) /* IFLA_BRPORT_MRP_RING_OPEN */ + nla_total_size(sizeof(u8)) /* IFLA_BRPORT_MRP_IN_OPEN */ + + nla_total_size(sizeof(u32)) /* IFLA_BRPORT_MCAST_EHT_HOSTS_LIMIT */ + + nla_total_size(sizeof(u32)) /* IFLA_BRPORT_MCAST_EHT_HOSTS_CNT */ + 0; } @@ -283,7 +286,11 @@ static int br_port_fill_attrs(struct sk_buff *skb, #ifdef CONFIG_BRIDGE_IGMP_SNOOPING if (nla_put_u8(skb, IFLA_BRPORT_MULTICAST_ROUTER, - p->multicast_router)) + p->multicast_router) || + nla_put_u32(skb, IFLA_BRPORT_MCAST_EHT_HOSTS_LIMIT, + p->multicast_eht_hosts_limit) || + nla_put_u32(skb, IFLA_BRPORT_MCAST_EHT_HOSTS_CNT, + p->multicast_eht_hosts_cnt)) return -EMSGSIZE; #endif @@ -820,6 +827,7 @@ static const struct nla_policy br_port_policy[IFLA_BRPORT_MAX + 1] = { [IFLA_BRPORT_NEIGH_SUPPRESS] = { .type = NLA_U8 }, [IFLA_BRPORT_ISOLATED] = { .type = NLA_U8 }, [IFLA_BRPORT_BACKUP_PORT] = { .type = NLA_U32 }, + [IFLA_BRPORT_MCAST_EHT_HOSTS_LIMIT] = { .type = NLA_U32 }, }; /* Change the state of the port and notify spanning tree */ @@ -955,6 +963,15 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[]) if (err) return err; } + + if (tb[IFLA_BRPORT_MCAST_EHT_HOSTS_LIMIT]) { + u32 hlimit; + + hlimit = nla_get_u32(tb[IFLA_BRPORT_MCAST_EHT_HOSTS_LIMIT]); + err = br_multicast_eht_set_hosts_limit(p, hlimit); + if (err) + return err; + } #endif if (tb[IFLA_BRPORT_GROUP_FWD_MASK]) { diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index d62c6e1af64a..d242ba668e47 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -252,6 +252,8 @@ struct net_bridge_port_group { struct timer_list timer; struct timer_list rexmit_timer; struct hlist_node mglist; + struct rb_root eht_set_tree; + struct rb_root eht_host_tree; struct rhash_head rhnode; struct net_bridge_mcast_gc mcast_gc; @@ -308,6 +310,8 @@ struct net_bridge_port { #if IS_ENABLED(CONFIG_IPV6) struct bridge_mcast_own_query ip6_own_query; #endif /* IS_ENABLED(CONFIG_IPV6) */ + u32 multicast_eht_hosts_limit; + u32 multicast_eht_hosts_cnt; unsigned char multicast_router; struct bridge_mcast_stats __percpu *mcast_stats; struct timer_list multicast_router_timer; @@ -846,6 +850,10 @@ void br_multicast_star_g_handle_mode(struct net_bridge_port_group *pg, u8 filter_mode); void br_multicast_sg_add_exclude_ports(struct net_bridge_mdb_entry *star_mp, struct net_bridge_port_group *sg); +struct net_bridge_group_src * +br_multicast_find_group_src(struct net_bridge_port_group *pg, struct br_ip *ip); +void br_multicast_del_group_src(struct net_bridge_group_src *src, + bool fastleave); static inline bool br_group_is_l2(const struct br_ip *group) { diff --git a/net/bridge/br_private_mcast_eht.h b/net/bridge/br_private_mcast_eht.h new file mode 100644 index 000000000000..f89049f4892c --- /dev/null +++ b/net/bridge/br_private_mcast_eht.h @@ -0,0 +1,93 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later + * Copyright (c) 2020, Nikolay Aleksandrov <nikolay@nvidia.com> + */ +#ifndef _BR_PRIVATE_MCAST_EHT_H_ +#define _BR_PRIVATE_MCAST_EHT_H_ + +#define BR_MCAST_DEFAULT_EHT_HOSTS_LIMIT 512 + +union net_bridge_eht_addr { + __be32 ip4; +#if IS_ENABLED(CONFIG_IPV6) + struct in6_addr ip6; +#endif +}; + +/* single host's list of set entries and filter_mode */ +struct net_bridge_group_eht_host { + struct rb_node rb_node; + + union net_bridge_eht_addr h_addr; + struct hlist_head set_entries; + unsigned int num_entries; + unsigned char filter_mode; + struct net_bridge_port_group *pg; +}; + +/* (host, src entry) added to a per-src set and host's list */ +struct net_bridge_group_eht_set_entry { + struct rb_node rb_node; + struct hlist_node host_list; + + union net_bridge_eht_addr h_addr; + struct timer_list timer; + struct net_bridge *br; + struct net_bridge_group_eht_set *eht_set; + struct net_bridge_group_eht_host *h_parent; + struct net_bridge_mcast_gc mcast_gc; +}; + +/* per-src set */ +struct net_bridge_group_eht_set { + struct rb_node rb_node; + + union net_bridge_eht_addr src_addr; + struct rb_root entry_tree; + struct timer_list timer; + struct net_bridge_port_group *pg; + struct net_bridge *br; + struct net_bridge_mcast_gc mcast_gc; +}; + +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING +void br_multicast_eht_clean_sets(struct net_bridge_port_group *pg); +bool br_multicast_eht_handle(struct net_bridge_port_group *pg, + void *h_addr, + void *srcs, + u32 nsrcs, + size_t addr_size, + int grec_type); +int br_multicast_eht_set_hosts_limit(struct net_bridge_port *p, + u32 eht_hosts_limit); + +static inline bool +br_multicast_eht_should_del_pg(const struct net_bridge_port_group *pg) +{ + return !!((pg->key.port->flags & BR_MULTICAST_FAST_LEAVE) && + RB_EMPTY_ROOT(&pg->eht_host_tree)); +} + +static inline bool +br_multicast_eht_hosts_over_limit(const struct net_bridge_port_group *pg) +{ + const struct net_bridge_port *p = pg->key.port; + + return !!(p->multicast_eht_hosts_cnt >= p->multicast_eht_hosts_limit); +} + +static inline void br_multicast_eht_hosts_inc(struct net_bridge_port_group *pg) +{ + struct net_bridge_port *p = pg->key.port; + + p->multicast_eht_hosts_cnt++; +} + +static inline void br_multicast_eht_hosts_dec(struct net_bridge_port_group *pg) +{ + struct net_bridge_port *p = pg->key.port; + + p->multicast_eht_hosts_cnt--; +} +#endif /* CONFIG_BRIDGE_IGMP_SNOOPING */ + +#endif /* _BR_PRIVATE_MCAST_EHT_H_ */ diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c index 7a59cdddd3ce..b66305fae26b 100644 --- a/net/bridge/br_sysfs_if.c +++ b/net/bridge/br_sysfs_if.c @@ -16,6 +16,7 @@ #include <linux/sched/signal.h> #include "br_private.h" +#include "br_private_mcast_eht.h" struct brport_attribute { struct attribute attr; @@ -245,6 +246,29 @@ static int store_multicast_router(struct net_bridge_port *p, static BRPORT_ATTR(multicast_router, 0644, show_multicast_router, store_multicast_router); +static ssize_t show_multicast_eht_hosts_limit(struct net_bridge_port *p, + char *buf) +{ + return sprintf(buf, "%u\n", p->multicast_eht_hosts_limit); +} + +static int store_multicast_eht_hosts_limit(struct net_bridge_port *p, + unsigned long v) +{ + return br_multicast_eht_set_hosts_limit(p, v); +} +static BRPORT_ATTR(multicast_eht_hosts_limit, 0644, + show_multicast_eht_hosts_limit, + store_multicast_eht_hosts_limit); + +static ssize_t show_multicast_eht_hosts_cnt(struct net_bridge_port *p, + char *buf) +{ + return sprintf(buf, "%u\n", p->multicast_eht_hosts_cnt); +} +static BRPORT_ATTR(multicast_eht_hosts_cnt, 0444, show_multicast_eht_hosts_cnt, + NULL); + BRPORT_ATTR_FLAG(multicast_fast_leave, BR_MULTICAST_FAST_LEAVE); BRPORT_ATTR_FLAG(multicast_to_unicast, BR_MULTICAST_TO_UNICAST); #endif @@ -274,6 +298,8 @@ static const struct brport_attribute *brport_attrs[] = { &brport_attr_multicast_router, &brport_attr_multicast_fast_leave, &brport_attr_multicast_to_unicast, + &brport_attr_multicast_eht_hosts_limit, + &brport_attr_multicast_eht_hosts_cnt, #endif &brport_attr_proxyarp, &brport_attr_proxyarp_wifi, diff --git a/net/can/Kconfig b/net/can/Kconfig index 7c9958df91d3..a9ac5ffab286 100644 --- a/net/can/Kconfig +++ b/net/can/Kconfig @@ -4,7 +4,6 @@ # menuconfig CAN - depends on NET tristate "CAN bus subsystem support" help Controller Area Network (CAN) is a slow (up to 1Mbit/s) serial diff --git a/net/can/gw.c b/net/can/gw.c index 8598d9da0e5f..ba4124805602 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -225,7 +225,7 @@ static void mod_store_ccdlc(struct canfd_frame *cf) if (ccf->len <= CAN_MAX_DLEN) return; - /* potentially broken values are catched in can_can_gw_rcv() */ + /* potentially broken values are caught in can_can_gw_rcv() */ if (ccf->len > CAN_MAX_RAW_DLC) return; diff --git a/net/core/dev.c b/net/core/dev.c index 0332f2e8f7da..6df3f1bcdc68 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3878,6 +3878,7 @@ sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev) /* qdisc_skb_cb(skb)->pkt_len was already set by the caller. */ qdisc_skb_cb(skb)->mru = 0; + qdisc_skb_cb(skb)->post_ct = false; mini_qdisc_bstats_cpu_update(miniq, skb); switch (tcf_classify(skb, miniq->filter_list, &cl_res, false)) { @@ -4083,7 +4084,7 @@ static int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev) skb_reset_mac_header(skb); if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_SCHED_TSTAMP)) - __skb_tstamp_tx(skb, NULL, skb->sk, SCM_TSTAMP_SCHED); + __skb_tstamp_tx(skb, NULL, NULL, skb->sk, SCM_TSTAMP_SCHED); /* Disable soft irqs for various locks below. Also * stops preemption for RCU. @@ -4960,6 +4961,7 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, qdisc_skb_cb(skb)->pkt_len = skb->len; qdisc_skb_cb(skb)->mru = 0; + qdisc_skb_cb(skb)->post_ct = false; skb->tc_at_ingress = 1; mini_qdisc_bstats_cpu_update(miniq, skb); @@ -5709,7 +5711,7 @@ static void flush_all_backlogs(void) } /* we can have in flight packet[s] on the cpus we are not flushing, - * synchronize_net() in rollback_registered_many() will take care of + * synchronize_net() in unregister_netdevice_many() will take care of * them */ for_each_cpu(cpu, &flush_cpus) @@ -9459,106 +9461,6 @@ static void net_set_todo(struct net_device *dev) dev_net(dev)->dev_unreg_count++; } -static void rollback_registered_many(struct list_head *head) -{ - struct net_device *dev, *tmp; - LIST_HEAD(close_head); - - BUG_ON(dev_boot_phase); - ASSERT_RTNL(); - - list_for_each_entry_safe(dev, tmp, head, unreg_list) { - /* Some devices call without registering - * for initialization unwind. Remove those - * devices and proceed with the remaining. - */ - if (dev->reg_state == NETREG_UNINITIALIZED) { - pr_debug("unregister_netdevice: device %s/%p never was registered\n", - dev->name, dev); - - WARN_ON(1); - list_del(&dev->unreg_list); - continue; - } - dev->dismantle = true; - BUG_ON(dev->reg_state != NETREG_REGISTERED); - } - - /* If device is running, close it first. */ - list_for_each_entry(dev, head, unreg_list) - list_add_tail(&dev->close_list, &close_head); - dev_close_many(&close_head, true); - - list_for_each_entry(dev, head, unreg_list) { - /* And unlink it from device chain. */ - unlist_netdevice(dev); - - dev->reg_state = NETREG_UNREGISTERING; - } - flush_all_backlogs(); - - synchronize_net(); - - list_for_each_entry(dev, head, unreg_list) { - struct sk_buff *skb = NULL; - - /* Shutdown queueing discipline. */ - dev_shutdown(dev); - - dev_xdp_uninstall(dev); - - /* Notify protocols, that we are about to destroy - * this device. They should clean all the things. - */ - call_netdevice_notifiers(NETDEV_UNREGISTER, dev); - - if (!dev->rtnl_link_ops || - dev->rtnl_link_state == RTNL_LINK_INITIALIZED) - skb = rtmsg_ifinfo_build_skb(RTM_DELLINK, dev, ~0U, 0, - GFP_KERNEL, NULL, 0); - - /* - * Flush the unicast and multicast chains - */ - dev_uc_flush(dev); - dev_mc_flush(dev); - - netdev_name_node_alt_flush(dev); - netdev_name_node_free(dev->name_node); - - if (dev->netdev_ops->ndo_uninit) - dev->netdev_ops->ndo_uninit(dev); - - if (skb) - rtmsg_ifinfo_send(skb, dev, GFP_KERNEL); - - /* Notifier chain MUST detach us all upper devices. */ - WARN_ON(netdev_has_any_upper_dev(dev)); - WARN_ON(netdev_has_any_lower_dev(dev)); - - /* Remove entries from kobject tree */ - netdev_unregister_kobject(dev); -#ifdef CONFIG_XPS - /* Remove XPS queueing entries */ - netif_reset_xps_queues_gt(dev, 0); -#endif - } - - synchronize_net(); - - list_for_each_entry(dev, head, unreg_list) - dev_put(dev); -} - -static void rollback_registered(struct net_device *dev) -{ - LIST_HEAD(single); - - list_add(&dev->unreg_list, &single); - rollback_registered_many(&single); - list_del(&single); -} - static netdev_features_t netdev_sync_upper_features(struct net_device *lower, struct net_device *upper, netdev_features_t features) { @@ -9690,6 +9592,11 @@ static netdev_features_t netdev_fix_features(struct net_device *dev, } } + if ((features & NETIF_F_HW_TLS_RX) && !(features & NETIF_F_RXCSUM)) { + netdev_dbg(dev, "Dropping TLS RX HW offload feature since no RXCSUM feature.\n"); + features &= ~NETIF_F_HW_TLS_RX; + } + return features; } @@ -10103,8 +10010,7 @@ int register_netdevice(struct net_device *dev) if (ret) { /* Expect explicit free_netdev() on failure */ dev->needs_free_netdev = false; - rollback_registered(dev); - net_set_todo(dev); + unregister_netdevice_queue(dev, NULL); goto out; } /* @@ -10726,9 +10632,10 @@ void unregister_netdevice_queue(struct net_device *dev, struct list_head *head) if (head) { list_move_tail(&dev->unreg_list, head); } else { - rollback_registered(dev); - /* Finish processing unregister after unlock */ - net_set_todo(dev); + LIST_HEAD(single); + + list_add(&dev->unreg_list, &single); + unregister_netdevice_many(&single); } } EXPORT_SYMBOL(unregister_netdevice_queue); @@ -10742,14 +10649,100 @@ EXPORT_SYMBOL(unregister_netdevice_queue); */ void unregister_netdevice_many(struct list_head *head) { - struct net_device *dev; + struct net_device *dev, *tmp; + LIST_HEAD(close_head); + + BUG_ON(dev_boot_phase); + ASSERT_RTNL(); - if (!list_empty(head)) { - rollback_registered_many(head); - list_for_each_entry(dev, head, unreg_list) - net_set_todo(dev); - list_del(head); + if (list_empty(head)) + return; + + list_for_each_entry_safe(dev, tmp, head, unreg_list) { + /* Some devices call without registering + * for initialization unwind. Remove those + * devices and proceed with the remaining. + */ + if (dev->reg_state == NETREG_UNINITIALIZED) { + pr_debug("unregister_netdevice: device %s/%p never was registered\n", + dev->name, dev); + + WARN_ON(1); + list_del(&dev->unreg_list); + continue; + } + dev->dismantle = true; + BUG_ON(dev->reg_state != NETREG_REGISTERED); + } + + /* If device is running, close it first. */ + list_for_each_entry(dev, head, unreg_list) + list_add_tail(&dev->close_list, &close_head); + dev_close_many(&close_head, true); + + list_for_each_entry(dev, head, unreg_list) { + /* And unlink it from device chain. */ + unlist_netdevice(dev); + + dev->reg_state = NETREG_UNREGISTERING; } + flush_all_backlogs(); + + synchronize_net(); + + list_for_each_entry(dev, head, unreg_list) { + struct sk_buff *skb = NULL; + + /* Shutdown queueing discipline. */ + dev_shutdown(dev); + + dev_xdp_uninstall(dev); + + /* Notify protocols, that we are about to destroy + * this device. They should clean all the things. + */ + call_netdevice_notifiers(NETDEV_UNREGISTER, dev); + + if (!dev->rtnl_link_ops || + dev->rtnl_link_state == RTNL_LINK_INITIALIZED) + skb = rtmsg_ifinfo_build_skb(RTM_DELLINK, dev, ~0U, 0, + GFP_KERNEL, NULL, 0); + + /* + * Flush the unicast and multicast chains + */ + dev_uc_flush(dev); + dev_mc_flush(dev); + + netdev_name_node_alt_flush(dev); + netdev_name_node_free(dev->name_node); + + if (dev->netdev_ops->ndo_uninit) + dev->netdev_ops->ndo_uninit(dev); + + if (skb) + rtmsg_ifinfo_send(skb, dev, GFP_KERNEL); + + /* Notifier chain MUST detach us all upper devices. */ + WARN_ON(netdev_has_any_upper_dev(dev)); + WARN_ON(netdev_has_any_lower_dev(dev)); + + /* Remove entries from kobject tree */ + netdev_unregister_kobject(dev); +#ifdef CONFIG_XPS + /* Remove XPS queueing entries */ + netif_reset_xps_queues_gt(dev, 0); +#endif + } + + synchronize_net(); + + list_for_each_entry(dev, head, unreg_list) { + dev_put(dev); + net_set_todo(dev); + } + + list_del(head); } EXPORT_SYMBOL(unregister_netdevice_many); diff --git a/net/core/devlink.c b/net/core/devlink.c index 9f1be69bd5f8..737b61c2976e 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -4350,7 +4350,7 @@ out: static int devlink_nl_cmd_port_param_get_doit(struct sk_buff *skb, struct genl_info *info) { - struct devlink_port *devlink_port = info->user_ptr[0]; + struct devlink_port *devlink_port = info->user_ptr[1]; struct devlink_param_item *param_item; struct sk_buff *msg; int err; @@ -4379,7 +4379,7 @@ static int devlink_nl_cmd_port_param_get_doit(struct sk_buff *skb, static int devlink_nl_cmd_port_param_set_doit(struct sk_buff *skb, struct genl_info *info) { - struct devlink_port *devlink_port = info->user_ptr[0]; + struct devlink_port *devlink_port = info->user_ptr[1]; return __devlink_nl_cmd_param_set_doit(devlink_port->devlink, devlink_port->index, @@ -8867,6 +8867,10 @@ EXPORT_SYMBOL_GPL(devlink_dpipe_table_unregister); * @resource_id: resource's id * @parent_resource_id: resource's parent id * @size_params: size parameters + * + * Generic resources should reuse the same names across drivers. + * Please see the generic resources list at: + * Documentation/networking/devlink/devlink-resource.rst */ int devlink_resource_register(struct devlink *devlink, const char *resource_name, @@ -9758,6 +9762,7 @@ static const struct devlink_trap devlink_trap_generic[] = { DEVLINK_TRAP(GTP_PARSING, DROP), DEVLINK_TRAP(ESP_PARSING, DROP), DEVLINK_TRAP(BLACKHOLE_NEXTHOP, DROP), + DEVLINK_TRAP(DMAC_FILTER, DROP), }; #define DEVLINK_TRAP_GROUP(_id) \ diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 2d70ded389ae..c565c7a17091 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -237,9 +237,8 @@ skb_flow_dissect_set_enc_addr_type(enum flow_dissector_key_id type, void skb_flow_dissect_ct(const struct sk_buff *skb, struct flow_dissector *flow_dissector, - void *target_container, - u16 *ctinfo_map, - size_t mapsize) + void *target_container, u16 *ctinfo_map, + size_t mapsize, bool post_ct) { #if IS_ENABLED(CONFIG_NF_CONNTRACK) struct flow_dissector_key_ct *key; @@ -251,13 +250,19 @@ skb_flow_dissect_ct(const struct sk_buff *skb, return; ct = nf_ct_get(skb, &ctinfo); - if (!ct) + if (!ct && !post_ct) return; key = skb_flow_dissector_target(flow_dissector, FLOW_DISSECTOR_KEY_CT, target_container); + if (!ct) { + key->ct_state = TCA_FLOWER_KEY_CT_FLAGS_TRACKED | + TCA_FLOWER_KEY_CT_FLAGS_INVALID; + return; + } + if (ctinfo < mapsize) key->ct_state = ctinfo_map[ctinfo]; #if IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 80dbf2f4016e..8e582e29a41e 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -80,11 +80,11 @@ static void est_timer(struct timer_list *t) u64 rate, brate; est_fetch_counters(est, &b); - brate = (b.bytes - est->last_bytes) << (10 - est->ewma_log - est->intvl_log); - brate -= (est->avbps >> est->ewma_log); + brate = (b.bytes - est->last_bytes) << (10 - est->intvl_log); + brate = (brate >> est->ewma_log) - (est->avbps >> est->ewma_log); - rate = (b.packets - est->last_packets) << (10 - est->ewma_log - est->intvl_log); - rate -= (est->avpps >> est->ewma_log); + rate = (b.packets - est->last_packets) << (10 - est->intvl_log); + rate = (rate >> est->ewma_log) - (est->avpps >> est->ewma_log); write_seqcount_begin(&est->seq); est->avbps += brate; @@ -143,6 +143,9 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, if (parm->interval < -2 || parm->interval > 3) return -EINVAL; + if (parm->ewma_log == 0 || parm->ewma_log >= 31) + return -EINVAL; + est = kzalloc(sizeof(*est), GFP_KERNEL); if (!est) return -ENOBUFS; diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 105978604ffd..3fba429f1f57 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -3464,7 +3464,7 @@ static int pktgen_thread_worker(void *arg) struct pktgen_dev *pkt_dev = NULL; int cpu = t->cpu; - BUG_ON(smp_processor_id() != cpu); + WARN_ON(smp_processor_id() != cpu); init_waitqueue_head(&t->queue); complete(&t->start_done); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 3d6ab194d0f5..c313aaf2bce1 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -55,7 +55,7 @@ #include <net/net_namespace.h> #define RTNL_MAX_TYPE 50 -#define RTNL_SLAVE_MAX_TYPE 36 +#define RTNL_SLAVE_MAX_TYPE 40 struct rtnl_link { rtnl_doit_func doit; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index cf2c4dcf4257..2af12f7e170c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -437,7 +437,11 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int len, len += NET_SKB_PAD; - if ((len > SKB_WITH_OVERHEAD(PAGE_SIZE)) || + /* If requested length is either too small or too big, + * we use kmalloc() for skb->head allocation. + */ + if (len <= SKB_WITH_OVERHEAD(1024) || + len > SKB_WITH_OVERHEAD(PAGE_SIZE) || (gfp_mask & (__GFP_DIRECT_RECLAIM | GFP_DMA))) { skb = __alloc_skb(len, gfp_mask, SKB_ALLOC_RX, NUMA_NO_NODE); if (!skb) @@ -4717,6 +4721,7 @@ err: EXPORT_SYMBOL_GPL(skb_complete_tx_timestamp); void __skb_tstamp_tx(struct sk_buff *orig_skb, + const struct sk_buff *ack_skb, struct skb_shared_hwtstamps *hwtstamps, struct sock *sk, int tstype) { @@ -4739,7 +4744,8 @@ void __skb_tstamp_tx(struct sk_buff *orig_skb, if ((sk->sk_tsflags & SOF_TIMESTAMPING_OPT_STATS) && sk->sk_protocol == IPPROTO_TCP && sk->sk_type == SOCK_STREAM) { - skb = tcp_get_timestamping_opt_stats(sk, orig_skb); + skb = tcp_get_timestamping_opt_stats(sk, orig_skb, + ack_skb); opt_stats = true; } else #endif @@ -4768,7 +4774,7 @@ EXPORT_SYMBOL_GPL(__skb_tstamp_tx); void skb_tstamp_tx(struct sk_buff *orig_skb, struct skb_shared_hwtstamps *hwtstamps) { - return __skb_tstamp_tx(orig_skb, hwtstamps, orig_skb->sk, + return __skb_tstamp_tx(orig_skb, NULL, hwtstamps, orig_skb->sk, SCM_TSTAMP_SND); } EXPORT_SYMBOL_GPL(skb_tstamp_tx); diff --git a/net/core/sock.c b/net/core/sock.c index bbcd4b97eddd..4600e58828da 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1876,123 +1876,120 @@ static void sk_init_common(struct sock *sk) struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) { struct proto *prot = READ_ONCE(sk->sk_prot); - struct sock *newsk; + struct sk_filter *filter; bool is_charged = true; + struct sock *newsk; newsk = sk_prot_alloc(prot, priority, sk->sk_family); - if (newsk != NULL) { - struct sk_filter *filter; + if (!newsk) + goto out; - sock_copy(newsk, sk); + sock_copy(newsk, sk); - newsk->sk_prot_creator = prot; + newsk->sk_prot_creator = prot; - /* SANITY */ - if (likely(newsk->sk_net_refcnt)) - get_net(sock_net(newsk)); - sk_node_init(&newsk->sk_node); - sock_lock_init(newsk); - bh_lock_sock(newsk); - newsk->sk_backlog.head = newsk->sk_backlog.tail = NULL; - newsk->sk_backlog.len = 0; + /* SANITY */ + if (likely(newsk->sk_net_refcnt)) + get_net(sock_net(newsk)); + sk_node_init(&newsk->sk_node); + sock_lock_init(newsk); + bh_lock_sock(newsk); + newsk->sk_backlog.head = newsk->sk_backlog.tail = NULL; + newsk->sk_backlog.len = 0; - atomic_set(&newsk->sk_rmem_alloc, 0); - /* - * sk_wmem_alloc set to one (see sk_free() and sock_wfree()) - */ - refcount_set(&newsk->sk_wmem_alloc, 1); - atomic_set(&newsk->sk_omem_alloc, 0); - sk_init_common(newsk); + atomic_set(&newsk->sk_rmem_alloc, 0); - newsk->sk_dst_cache = NULL; - newsk->sk_dst_pending_confirm = 0; - newsk->sk_wmem_queued = 0; - newsk->sk_forward_alloc = 0; - atomic_set(&newsk->sk_drops, 0); - newsk->sk_send_head = NULL; - newsk->sk_userlocks = sk->sk_userlocks & ~SOCK_BINDPORT_LOCK; - atomic_set(&newsk->sk_zckey, 0); + /* sk_wmem_alloc set to one (see sk_free() and sock_wfree()) */ + refcount_set(&newsk->sk_wmem_alloc, 1); - sock_reset_flag(newsk, SOCK_DONE); + atomic_set(&newsk->sk_omem_alloc, 0); + sk_init_common(newsk); - /* sk->sk_memcg will be populated at accept() time */ - newsk->sk_memcg = NULL; + newsk->sk_dst_cache = NULL; + newsk->sk_dst_pending_confirm = 0; + newsk->sk_wmem_queued = 0; + newsk->sk_forward_alloc = 0; + atomic_set(&newsk->sk_drops, 0); + newsk->sk_send_head = NULL; + newsk->sk_userlocks = sk->sk_userlocks & ~SOCK_BINDPORT_LOCK; + atomic_set(&newsk->sk_zckey, 0); - cgroup_sk_clone(&newsk->sk_cgrp_data); + sock_reset_flag(newsk, SOCK_DONE); - rcu_read_lock(); - filter = rcu_dereference(sk->sk_filter); - if (filter != NULL) - /* though it's an empty new sock, the charging may fail - * if sysctl_optmem_max was changed between creation of - * original socket and cloning - */ - is_charged = sk_filter_charge(newsk, filter); - RCU_INIT_POINTER(newsk->sk_filter, filter); - rcu_read_unlock(); + /* sk->sk_memcg will be populated at accept() time */ + newsk->sk_memcg = NULL; - if (unlikely(!is_charged || xfrm_sk_clone_policy(newsk, sk))) { - /* We need to make sure that we don't uncharge the new - * socket if we couldn't charge it in the first place - * as otherwise we uncharge the parent's filter. - */ - if (!is_charged) - RCU_INIT_POINTER(newsk->sk_filter, NULL); - sk_free_unlock_clone(newsk); - newsk = NULL; - goto out; - } - RCU_INIT_POINTER(newsk->sk_reuseport_cb, NULL); + cgroup_sk_clone(&newsk->sk_cgrp_data); - if (bpf_sk_storage_clone(sk, newsk)) { - sk_free_unlock_clone(newsk); - newsk = NULL; - goto out; - } + rcu_read_lock(); + filter = rcu_dereference(sk->sk_filter); + if (filter != NULL) + /* though it's an empty new sock, the charging may fail + * if sysctl_optmem_max was changed between creation of + * original socket and cloning + */ + is_charged = sk_filter_charge(newsk, filter); + RCU_INIT_POINTER(newsk->sk_filter, filter); + rcu_read_unlock(); - /* Clear sk_user_data if parent had the pointer tagged - * as not suitable for copying when cloning. + if (unlikely(!is_charged || xfrm_sk_clone_policy(newsk, sk))) { + /* We need to make sure that we don't uncharge the new + * socket if we couldn't charge it in the first place + * as otherwise we uncharge the parent's filter. */ - if (sk_user_data_is_nocopy(newsk)) - newsk->sk_user_data = NULL; + if (!is_charged) + RCU_INIT_POINTER(newsk->sk_filter, NULL); + sk_free_unlock_clone(newsk); + newsk = NULL; + goto out; + } + RCU_INIT_POINTER(newsk->sk_reuseport_cb, NULL); - newsk->sk_err = 0; - newsk->sk_err_soft = 0; - newsk->sk_priority = 0; - newsk->sk_incoming_cpu = raw_smp_processor_id(); - if (likely(newsk->sk_net_refcnt)) - sock_inuse_add(sock_net(newsk), 1); + if (bpf_sk_storage_clone(sk, newsk)) { + sk_free_unlock_clone(newsk); + newsk = NULL; + goto out; + } - /* - * Before updating sk_refcnt, we must commit prior changes to memory - * (Documentation/RCU/rculist_nulls.rst for details) - */ - smp_wmb(); - refcount_set(&newsk->sk_refcnt, 2); + /* Clear sk_user_data if parent had the pointer tagged + * as not suitable for copying when cloning. + */ + if (sk_user_data_is_nocopy(newsk)) + newsk->sk_user_data = NULL; - /* - * Increment the counter in the same struct proto as the master - * sock (sk_refcnt_debug_inc uses newsk->sk_prot->socks, that - * is the same as sk->sk_prot->socks, as this field was copied - * with memcpy). - * - * This _changes_ the previous behaviour, where - * tcp_create_openreq_child always was incrementing the - * equivalent to tcp_prot->socks (inet_sock_nr), so this have - * to be taken into account in all callers. -acme - */ - sk_refcnt_debug_inc(newsk); - sk_set_socket(newsk, NULL); - sk_tx_queue_clear(newsk); - RCU_INIT_POINTER(newsk->sk_wq, NULL); + newsk->sk_err = 0; + newsk->sk_err_soft = 0; + newsk->sk_priority = 0; + newsk->sk_incoming_cpu = raw_smp_processor_id(); + if (likely(newsk->sk_net_refcnt)) + sock_inuse_add(sock_net(newsk), 1); - if (newsk->sk_prot->sockets_allocated) - sk_sockets_allocated_inc(newsk); + /* Before updating sk_refcnt, we must commit prior changes to memory + * (Documentation/RCU/rculist_nulls.rst for details) + */ + smp_wmb(); + refcount_set(&newsk->sk_refcnt, 2); - if (sock_needs_netstamp(sk) && - newsk->sk_flags & SK_FLAGS_TIMESTAMP) - net_enable_timestamp(); - } + /* Increment the counter in the same struct proto as the master + * sock (sk_refcnt_debug_inc uses newsk->sk_prot->socks, that + * is the same as sk->sk_prot->socks, as this field was copied + * with memcpy). + * + * This _changes_ the previous behaviour, where + * tcp_create_openreq_child always was incrementing the + * equivalent to tcp_prot->socks (inet_sock_nr), so this have + * to be taken into account in all callers. -acme + */ + sk_refcnt_debug_inc(newsk); + sk_set_socket(newsk, NULL); + sk_tx_queue_clear(newsk); + RCU_INIT_POINTER(newsk->sk_wq, NULL); + + if (newsk->sk_prot->sockets_allocated) + sk_sockets_allocated_inc(newsk); + + if (sock_needs_netstamp(sk) && newsk->sk_flags & SK_FLAGS_TIMESTAMP) + net_enable_timestamp(); out: return newsk; } diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index d86d8d11cfe4..4567de519603 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -309,7 +309,6 @@ proc_dolongvec_minmax_bpf_restricted(struct ctl_table *table, int write, #endif static struct ctl_table net_core_table[] = { -#ifdef CONFIG_NET { .procname = "wmem_max", .data = &sysctl_wmem_max, @@ -507,7 +506,6 @@ static struct ctl_table net_core_table[] = { .proc_handler = set_default_qdisc }, #endif -#endif /* CONFIG_NET */ { .procname = "netdev_budget", .data = &netdev_budget, diff --git a/net/dcb/Makefile b/net/dcb/Makefile index 3016e5a7716a..2c0fa16ee2a9 100644 --- a/net/dcb/Makefile +++ b/net/dcb/Makefile @@ -1,2 +1,2 @@ # SPDX-License-Identifier: GPL-2.0-only -obj-$(CONFIG_DCB) += dcbnl.o dcbevent.o +obj-y += dcbnl.o dcbevent.o diff --git a/net/dns_resolver/Kconfig b/net/dns_resolver/Kconfig index 255df9b6e9e8..155b06163409 100644 --- a/net/dns_resolver/Kconfig +++ b/net/dns_resolver/Kconfig @@ -4,7 +4,7 @@ # config DNS_RESOLVER tristate "DNS Resolver support" - depends on NET && KEYS + depends on KEYS help Saying Y here will include support for the DNS Resolver key type which can be used to make upcalls to perform DNS lookups in diff --git a/net/ethtool/common.c b/net/ethtool/common.c index 24036e3055a1..181220101a6e 100644 --- a/net/ethtool/common.c +++ b/net/ethtool/common.c @@ -68,6 +68,7 @@ const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] = { [NETIF_F_HW_TLS_RX_BIT] = "tls-hw-rx-offload", [NETIF_F_GRO_FRAGLIST_BIT] = "rx-gro-list", [NETIF_F_HW_MACSEC_BIT] = "macsec-hw-offload", + [NETIF_F_GRO_UDP_FWD_BIT] = "rx-udp-gro-forwarding", }; const char diff --git a/net/ife/Kconfig b/net/ife/Kconfig index bcf650564db4..de36a5b91e50 100644 --- a/net/ife/Kconfig +++ b/net/ife/Kconfig @@ -4,7 +4,6 @@ # menuconfig NET_IFE - depends on NET tristate "Inter-FE based on IETF ForCES InterFE LFB" default n help diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c index e0a246575887..10bc49bde9a1 100644 --- a/net/ipv4/gre_offload.c +++ b/net/ipv4/gre_offload.c @@ -15,10 +15,10 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, netdev_features_t features) { int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb); - bool need_csum, need_recompute_csum, gso_partial; struct sk_buff *segs = ERR_PTR(-EINVAL); u16 mac_offset = skb->mac_header; __be16 protocol = skb->protocol; + bool need_csum, gso_partial; u16 mac_len = skb->mac_len; int gre_offset, outer_hlen; @@ -41,10 +41,11 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, skb->protocol = skb->inner_protocol; need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_GRE_CSUM); - need_recompute_csum = skb->csum_not_inet; skb->encap_hdr_csum = need_csum; features &= skb->dev->hw_enc_features; + if (need_csum) + features &= ~NETIF_F_SCTP_CRC; /* segment inner packet. */ segs = skb_mac_gso_segment(skb, features); @@ -99,15 +100,7 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, } *(pcsum + 1) = 0; - if (need_recompute_csum && !skb_is_gso(skb)) { - __wsum csum; - - csum = skb_checksum(skb, gre_offset, - skb->len - gre_offset, 0); - *pcsum = csum_fold(csum); - } else { - *pcsum = gso_make_checksum(skb, 0); - } + *pcsum = gso_make_checksum(skb, 0); } while ((skb = skb->next)); out: return segs; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index fd8b8800a2c3..6bd7ca09af03 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -851,6 +851,7 @@ struct sock *inet_csk_clone_lock(const struct sock *sk, newicsk->icsk_retransmits = 0; newicsk->icsk_backoff = 0; newicsk->icsk_probes_out = 0; + newicsk->icsk_probes_tstamp = 0; /* Deinitialize accept_queue to trap illegal accesses. */ memset(&newicsk->icsk_accept_queue, 0, sizeof(newicsk->icsk_accept_queue)); diff --git a/net/ipv4/netfilter/ipt_rpfilter.c b/net/ipv4/netfilter/ipt_rpfilter.c index cc23f1ce239c..8cd3224d913e 100644 --- a/net/ipv4/netfilter/ipt_rpfilter.c +++ b/net/ipv4/netfilter/ipt_rpfilter.c @@ -76,7 +76,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par) flow.daddr = iph->saddr; flow.saddr = rpfilter_get_saddr(iph->daddr); flow.flowi4_mark = info->flags & XT_RPFILTER_VALID_MARK ? skb->mark : 0; - flow.flowi4_tos = RT_TOS(iph->tos); + flow.flowi4_tos = iph->tos & IPTOS_RT_MASK; flow.flowi4_scope = RT_SCOPE_UNIVERSE; flow.flowi4_oif = l3mdev_master_ifindex_rcu(xt_in(par)); diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c index e53e43aef785..e6dfca426242 100644 --- a/net/ipv4/nexthop.c +++ b/net/ipv4/nexthop.c @@ -22,7 +22,7 @@ static void remove_nexthop(struct net *net, struct nexthop *nh, #define NH_DEV_HASHBITS 8 #define NH_DEV_HASHSIZE (1U << NH_DEV_HASHBITS) -static const struct nla_policy rtm_nh_policy[NHA_MAX + 1] = { +static const struct nla_policy rtm_nh_policy_new[] = { [NHA_ID] = { .type = NLA_U32 }, [NHA_GROUP] = { .type = NLA_BINARY }, [NHA_GROUP_TYPE] = { .type = NLA_U16 }, @@ -31,6 +31,15 @@ static const struct nla_policy rtm_nh_policy[NHA_MAX + 1] = { [NHA_GATEWAY] = { .type = NLA_BINARY }, [NHA_ENCAP_TYPE] = { .type = NLA_U16 }, [NHA_ENCAP] = { .type = NLA_NESTED }, + [NHA_FDB] = { .type = NLA_FLAG }, +}; + +static const struct nla_policy rtm_nh_policy_get[] = { + [NHA_ID] = { .type = NLA_U32 }, +}; + +static const struct nla_policy rtm_nh_policy_dump[] = { + [NHA_OIF] = { .type = NLA_U32 }, [NHA_GROUPS] = { .type = NLA_FLAG }, [NHA_MASTER] = { .type = NLA_U32 }, [NHA_FDB] = { .type = NLA_FLAG }, @@ -565,7 +574,8 @@ static int nh_check_attr_fdb_group(struct nexthop *nh, u8 *nh_family, return 0; } -static int nh_check_attr_group(struct net *net, struct nlattr *tb[], +static int nh_check_attr_group(struct net *net, + struct nlattr *tb[], size_t tb_size, struct netlink_ext_ack *extack) { unsigned int len = nla_len(tb[NHA_GROUP]); @@ -624,7 +634,7 @@ static int nh_check_attr_group(struct net *net, struct nlattr *tb[], return -EINVAL; } } - for (i = NHA_GROUP_TYPE + 1; i < __NHA_MAX; ++i) { + for (i = NHA_GROUP_TYPE + 1; i < tb_size; ++i) { if (!tb[i]) continue; if (i == NHA_FDB) @@ -1643,11 +1653,12 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, struct netlink_ext_ack *extack) { struct nhmsg *nhm = nlmsg_data(nlh); - struct nlattr *tb[NHA_MAX + 1]; + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_new)]; int err; - err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy, - extack); + err = nlmsg_parse(nlh, sizeof(*nhm), tb, + ARRAY_SIZE(rtm_nh_policy_new) - 1, + rtm_nh_policy_new, extack); if (err < 0) return err; @@ -1674,11 +1685,6 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, goto out; } - if (tb[NHA_GROUPS] || tb[NHA_MASTER]) { - NL_SET_ERR_MSG(extack, "Invalid attributes in request"); - goto out; - } - memset(cfg, 0, sizeof(*cfg)); cfg->nlflags = nlh->nlmsg_flags; cfg->nlinfo.portid = NETLINK_CB(skb).portid; @@ -1720,7 +1726,7 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, NL_SET_ERR_MSG(extack, "Invalid group type"); goto out; } - err = nh_check_attr_group(net, tb, extack); + err = nh_check_attr_group(net, tb, ARRAY_SIZE(tb), extack); /* no other attributes should be set */ goto out; @@ -1842,28 +1848,16 @@ static int nh_valid_get_del_req(struct nlmsghdr *nlh, u32 *id, struct netlink_ext_ack *extack) { struct nhmsg *nhm = nlmsg_data(nlh); - struct nlattr *tb[NHA_MAX + 1]; - int err, i; + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_get)]; + int err; - err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy, - extack); + err = nlmsg_parse(nlh, sizeof(*nhm), tb, + ARRAY_SIZE(rtm_nh_policy_get) - 1, + rtm_nh_policy_get, extack); if (err < 0) return err; err = -EINVAL; - for (i = 0; i < __NHA_MAX; ++i) { - if (!tb[i]) - continue; - - switch (i) { - case NHA_ID: - break; - default: - NL_SET_ERR_MSG_ATTR(extack, tb[i], - "Unexpected attribute in request"); - goto out; - } - } if (nhm->nh_protocol || nhm->resvd || nhm->nh_scope || nhm->nh_flags) { NL_SET_ERR_MSG(extack, "Invalid values in header"); goto out; @@ -1991,48 +1985,35 @@ static int nh_valid_dump_req(const struct nlmsghdr *nlh, int *dev_idx, bool *fdb_filter, struct netlink_callback *cb) { struct netlink_ext_ack *extack = cb->extack; - struct nlattr *tb[NHA_MAX + 1]; + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_dump)]; struct nhmsg *nhm; - int err, i; + int err; u32 idx; - err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy, - NULL); + err = nlmsg_parse(nlh, sizeof(*nhm), tb, + ARRAY_SIZE(rtm_nh_policy_dump) - 1, + rtm_nh_policy_dump, NULL); if (err < 0) return err; - for (i = 0; i <= NHA_MAX; ++i) { - if (!tb[i]) - continue; - - switch (i) { - case NHA_OIF: - idx = nla_get_u32(tb[i]); - if (idx > INT_MAX) { - NL_SET_ERR_MSG(extack, "Invalid device index"); - return -EINVAL; - } - *dev_idx = idx; - break; - case NHA_MASTER: - idx = nla_get_u32(tb[i]); - if (idx > INT_MAX) { - NL_SET_ERR_MSG(extack, "Invalid master device index"); - return -EINVAL; - } - *master_idx = idx; - break; - case NHA_GROUPS: - *group_filter = true; - break; - case NHA_FDB: - *fdb_filter = true; - break; - default: - NL_SET_ERR_MSG(extack, "Unsupported attribute in dump request"); + if (tb[NHA_OIF]) { + idx = nla_get_u32(tb[NHA_OIF]); + if (idx > INT_MAX) { + NL_SET_ERR_MSG(extack, "Invalid device index"); + return -EINVAL; + } + *dev_idx = idx; + } + if (tb[NHA_MASTER]) { + idx = nla_get_u32(tb[NHA_MASTER]); + if (idx > INT_MAX) { + NL_SET_ERR_MSG(extack, "Invalid master device index"); return -EINVAL; } + *master_idx = idx; } + *group_filter = nla_get_flag(tb[NHA_GROUPS]); + *fdb_filter = nla_get_flag(tb[NHA_FDB]); nhm = nlmsg_data(nlh); if (nhm->nh_protocol || nhm->resvd || nhm->nh_scope || nhm->nh_flags) { diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 2267d21c73a6..e1a17c6b473c 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -280,6 +280,12 @@ #include <asm/ioctls.h> #include <net/busy_poll.h> +/* Track pending CMSGs. */ +enum { + TCP_CMSG_INQ = 1, + TCP_CMSG_TS = 2 +}; + struct percpu_counter tcp_orphan_count; EXPORT_SYMBOL_GPL(tcp_orphan_count); @@ -1739,6 +1745,20 @@ int tcp_set_rcvlowat(struct sock *sk, int val) } EXPORT_SYMBOL(tcp_set_rcvlowat); +static void tcp_update_recv_tstamps(struct sk_buff *skb, + struct scm_timestamping_internal *tss) +{ + if (skb->tstamp) + tss->ts[0] = ktime_to_timespec64(skb->tstamp); + else + tss->ts[0] = (struct timespec64) {0}; + + if (skb_hwtstamps(skb)->hwtstamp) + tss->ts[2] = ktime_to_timespec64(skb_hwtstamps(skb)->hwtstamp); + else + tss->ts[2] = (struct timespec64) {0}; +} + #ifdef CONFIG_MMU static const struct vm_operations_struct tcp_vm_ops = { }; @@ -1842,13 +1862,13 @@ static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, struct scm_timestamping_internal *tss, int *cmsg_flags); static int receive_fallback_to_copy(struct sock *sk, - struct tcp_zerocopy_receive *zc, int inq) + struct tcp_zerocopy_receive *zc, int inq, + struct scm_timestamping_internal *tss) { unsigned long copy_address = (unsigned long)zc->copybuf_address; - struct scm_timestamping_internal tss_unused; - int err, cmsg_flags_unused; struct msghdr msg = {}; struct iovec iov; + int err; zc->length = 0; zc->recv_skip_hint = 0; @@ -1862,7 +1882,7 @@ static int receive_fallback_to_copy(struct sock *sk, return err; err = tcp_recvmsg_locked(sk, &msg, inq, /*nonblock=*/1, /*flags=*/0, - &tss_unused, &cmsg_flags_unused); + tss, &zc->msg_flags); if (err < 0) return err; @@ -1903,21 +1923,27 @@ static int tcp_copy_straggler_data(struct tcp_zerocopy_receive *zc, return (__s32)copylen; } -static int tcp_zerocopy_handle_leftover_data(struct tcp_zerocopy_receive *zc, - struct sock *sk, - struct sk_buff *skb, - u32 *seq, - s32 copybuf_len) +static int tcp_zc_handle_leftover(struct tcp_zerocopy_receive *zc, + struct sock *sk, + struct sk_buff *skb, + u32 *seq, + s32 copybuf_len, + struct scm_timestamping_internal *tss) { u32 offset, copylen = min_t(u32, copybuf_len, zc->recv_skip_hint); if (!copylen) return 0; /* skb is null if inq < PAGE_SIZE. */ - if (skb) + if (skb) { offset = *seq - TCP_SKB_CB(skb)->seq; - else + } else { skb = tcp_recv_skb(sk, *seq, &offset); + if (TCP_SKB_CB(skb)->has_rxtstamp) { + tcp_update_recv_tstamps(skb, tss); + zc->msg_flags |= TCP_CMSG_TS; + } + } zc->copybuf_len = tcp_copy_straggler_data(zc, skb, copylen, &offset, seq); @@ -2004,9 +2030,37 @@ static int tcp_zerocopy_vm_insert_batch(struct vm_area_struct *vma, err); } +static void tcp_recv_timestamp(struct msghdr *msg, const struct sock *sk, + struct scm_timestamping_internal *tss); +static void tcp_zc_finalize_rx_tstamp(struct sock *sk, + struct tcp_zerocopy_receive *zc, + struct scm_timestamping_internal *tss) +{ + unsigned long msg_control_addr; + struct msghdr cmsg_dummy; + + msg_control_addr = (unsigned long)zc->msg_control; + cmsg_dummy.msg_control = (void *)msg_control_addr; + cmsg_dummy.msg_controllen = + (__kernel_size_t)zc->msg_controllen; + cmsg_dummy.msg_flags = in_compat_syscall() + ? MSG_CMSG_COMPAT : 0; + zc->msg_flags = 0; + if (zc->msg_control == msg_control_addr && + zc->msg_controllen == cmsg_dummy.msg_controllen) { + tcp_recv_timestamp(&cmsg_dummy, sk, tss); + zc->msg_control = (__u64) + ((uintptr_t)cmsg_dummy.msg_control); + zc->msg_controllen = + (__u64)cmsg_dummy.msg_controllen; + zc->msg_flags = (__u32)cmsg_dummy.msg_flags; + } +} + #define TCP_ZEROCOPY_PAGE_BATCH_SIZE 32 static int tcp_zerocopy_receive(struct sock *sk, - struct tcp_zerocopy_receive *zc) + struct tcp_zerocopy_receive *zc, + struct scm_timestamping_internal *tss) { u32 length = 0, offset, vma_len, avail_len, copylen = 0; unsigned long address = (unsigned long)zc->address; @@ -2023,6 +2077,7 @@ static int tcp_zerocopy_receive(struct sock *sk, int ret; zc->copybuf_len = 0; + zc->msg_flags = 0; if (address & (PAGE_SIZE - 1) || address != zc->address) return -EINVAL; @@ -2033,7 +2088,7 @@ static int tcp_zerocopy_receive(struct sock *sk, sock_rps_record_flow(sk); if (inq && inq <= copybuf_len) - return receive_fallback_to_copy(sk, zc, inq); + return receive_fallback_to_copy(sk, zc, inq, tss); if (inq < PAGE_SIZE) { zc->length = 0; @@ -2078,6 +2133,11 @@ static int tcp_zerocopy_receive(struct sock *sk, } else { skb = tcp_recv_skb(sk, seq, &offset); } + + if (TCP_SKB_CB(skb)->has_rxtstamp) { + tcp_update_recv_tstamps(skb, tss); + zc->msg_flags |= TCP_CMSG_TS; + } zc->recv_skip_hint = skb->len - offset; frags = skb_advance_to_frag(skb, offset, &offset_frag); if (!frags || offset_frag) @@ -2120,8 +2180,7 @@ out: mmap_read_unlock(current->mm); /* Try to copy straggler data. */ if (!ret) - copylen = tcp_zerocopy_handle_leftover_data(zc, sk, skb, &seq, - copybuf_len); + copylen = tcp_zc_handle_leftover(zc, sk, skb, &seq, copybuf_len, tss); if (length + copylen) { WRITE_ONCE(tp->copied_seq, seq); @@ -2142,20 +2201,6 @@ out: } #endif -static void tcp_update_recv_tstamps(struct sk_buff *skb, - struct scm_timestamping_internal *tss) -{ - if (skb->tstamp) - tss->ts[0] = ktime_to_timespec64(skb->tstamp); - else - tss->ts[0] = (struct timespec64) {0}; - - if (skb_hwtstamps(skb)->hwtstamp) - tss->ts[2] = ktime_to_timespec64(skb_hwtstamps(skb)->hwtstamp); - else - tss->ts[2] = (struct timespec64) {0}; -} - /* Similar to __sock_recv_timestamp, but does not require an skb */ static void tcp_recv_timestamp(struct msghdr *msg, const struct sock *sk, struct scm_timestamping_internal *tss) @@ -2272,7 +2317,7 @@ static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, goto out; if (tp->recvmsg_inq) - *cmsg_flags = 1; + *cmsg_flags = TCP_CMSG_INQ; timeo = sock_rcvtimeo(sk, nonblock); /* Urgent data needs to be handled specially. */ @@ -2453,7 +2498,7 @@ skip_copy: if (TCP_SKB_CB(skb)->has_rxtstamp) { tcp_update_recv_tstamps(skb, tss); - *cmsg_flags |= 2; + *cmsg_flags |= TCP_CMSG_TS; } if (used + offset < skb->len) @@ -2513,9 +2558,9 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, release_sock(sk); if (cmsg_flags && ret >= 0) { - if (cmsg_flags & 2) + if (cmsg_flags & TCP_CMSG_TS) tcp_recv_timestamp(msg, sk, &tss); - if (cmsg_flags & 1) { + if (cmsg_flags & TCP_CMSG_INQ) { inq = tcp_inq_hint(sk); put_cmsg(msg, SOL_TCP, TCP_CM_INQ, sizeof(inq), &inq); } @@ -2937,6 +2982,7 @@ int tcp_disconnect(struct sock *sk, int flags) icsk->icsk_backoff = 0; icsk->icsk_probes_out = 0; + icsk->icsk_probes_tstamp = 0; icsk->icsk_rto = TCP_TIMEOUT_INIT; icsk->icsk_rto_min = TCP_RTO_MIN; icsk->icsk_delack_max = TCP_DELACK_MAX; @@ -3766,11 +3812,24 @@ static size_t tcp_opt_stats_get_size(void) nla_total_size(sizeof(u16)) + /* TCP_NLA_TIMEOUT_REHASH */ nla_total_size(sizeof(u32)) + /* TCP_NLA_BYTES_NOTSENT */ nla_total_size_64bit(sizeof(u64)) + /* TCP_NLA_EDT */ + nla_total_size(sizeof(u8)) + /* TCP_NLA_TTL */ 0; } +/* Returns TTL or hop limit of an incoming packet from skb. */ +static u8 tcp_skb_ttl_or_hop_limit(const struct sk_buff *skb) +{ + if (skb->protocol == htons(ETH_P_IP)) + return ip_hdr(skb)->ttl; + else if (skb->protocol == htons(ETH_P_IPV6)) + return ipv6_hdr(skb)->hop_limit; + else + return 0; +} + struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk, - const struct sk_buff *orig_skb) + const struct sk_buff *orig_skb, + const struct sk_buff *ack_skb) { const struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *stats; @@ -3826,6 +3885,9 @@ struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk, max_t(int, 0, tp->write_seq - tp->snd_nxt)); nla_put_u64_64bit(stats, TCP_NLA_EDT, orig_skb->skb_mstamp_ns, TCP_NLA_PAD); + if (ack_skb) + nla_put_u8(stats, TCP_NLA_TTL, + tcp_skb_ttl_or_hop_limit(ack_skb)); return stats; } @@ -4082,6 +4144,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level, } #ifdef CONFIG_MMU case TCP_ZEROCOPY_RECEIVE: { + struct scm_timestamping_internal tss; struct tcp_zerocopy_receive zc = {}; int err; @@ -4097,11 +4160,18 @@ static int do_tcp_getsockopt(struct sock *sk, int level, if (copy_from_user(&zc, optval, len)) return -EFAULT; lock_sock(sk); - err = tcp_zerocopy_receive(sk, &zc); + err = tcp_zerocopy_receive(sk, &zc, &tss); release_sock(sk); - if (len >= offsetofend(struct tcp_zerocopy_receive, err)) - goto zerocopy_rcv_sk_err; + if (len >= offsetofend(struct tcp_zerocopy_receive, msg_flags)) + goto zerocopy_rcv_cmsg; switch (len) { + case offsetofend(struct tcp_zerocopy_receive, msg_flags): + goto zerocopy_rcv_cmsg; + case offsetofend(struct tcp_zerocopy_receive, msg_controllen): + case offsetofend(struct tcp_zerocopy_receive, msg_control): + case offsetofend(struct tcp_zerocopy_receive, flags): + case offsetofend(struct tcp_zerocopy_receive, copybuf_len): + case offsetofend(struct tcp_zerocopy_receive, copybuf_address): case offsetofend(struct tcp_zerocopy_receive, err): goto zerocopy_rcv_sk_err; case offsetofend(struct tcp_zerocopy_receive, inq): @@ -4110,6 +4180,11 @@ static int do_tcp_getsockopt(struct sock *sk, int level, default: goto zerocopy_rcv_out; } +zerocopy_rcv_cmsg: + if (zc.msg_flags & TCP_CMSG_TS) + tcp_zc_finalize_rx_tstamp(sk, &zc, &tss); + else + zc.msg_flags = 0; zerocopy_rcv_sk_err: if (!err) zc.err = sock_error(sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c7e16b0ed791..d4f66aba9fd8 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3145,7 +3145,7 @@ static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb) } static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, - u32 prior_snd_una) + const struct sk_buff *ack_skb, u32 prior_snd_una) { const struct skb_shared_info *shinfo; @@ -3157,7 +3157,7 @@ static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, if (!before(shinfo->tskey, prior_snd_una) && before(shinfo->tskey, tcp_sk(sk)->snd_una)) { tcp_skb_tsorted_save(skb) { - __skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK); + __skb_tstamp_tx(skb, ack_skb, NULL, sk, SCM_TSTAMP_ACK); } tcp_skb_tsorted_restore(skb); } } @@ -3166,8 +3166,8 @@ static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, * is before the ack sequence we can discard it as it's confirmed to have * arrived at the other end. */ -static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, - u32 prior_snd_una, +static int tcp_clean_rtx_queue(struct sock *sk, const struct sk_buff *ack_skb, + u32 prior_fack, u32 prior_snd_una, struct tcp_sacktag_state *sack, bool ece_ack) { const struct inet_connection_sock *icsk = inet_csk(sk); @@ -3256,7 +3256,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, if (!fully_acked) break; - tcp_ack_tstamp(sk, skb, prior_snd_una); + tcp_ack_tstamp(sk, skb, ack_skb, prior_snd_una); next = skb_rb_next(skb); if (unlikely(skb == tp->retransmit_skb_hint)) @@ -3274,7 +3274,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, tp->snd_up = tp->snd_una; if (skb) { - tcp_ack_tstamp(sk, skb, prior_snd_una); + tcp_ack_tstamp(sk, skb, ack_skb, prior_snd_una); if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) flag |= FLAG_SACK_RENEGING; } @@ -3384,6 +3384,7 @@ static void tcp_ack_probe(struct sock *sk) return; if (!after(TCP_SKB_CB(head)->end_seq, tcp_wnd_end(tp))) { icsk->icsk_backoff = 0; + icsk->icsk_probes_tstamp = 0; inet_csk_clear_xmit_timer(sk, ICSK_TIME_PROBE0); /* Socket must be waked up by subsequent tcp_data_snd_check(). * This function is not for random using! @@ -3808,8 +3809,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) goto no_queue; /* See if we can take anything off of the retransmit queue. */ - flag |= tcp_clean_rtx_queue(sk, prior_fack, prior_snd_una, &sack_state, - flag & FLAG_ECE); + flag |= tcp_clean_rtx_queue(sk, skb, prior_fack, prior_snd_una, + &sack_state, flag & FLAG_ECE); tcp_rack_update_reo_wnd(sk, &rs); @@ -4396,10 +4397,9 @@ static void tcp_rcv_spurious_retrans(struct sock *sk, const struct sk_buff *skb) * The receiver remembers and reflects via DSACKs. Leverage the * DSACK state and change the txhash to re-route speculatively. */ - if (TCP_SKB_CB(skb)->seq == tcp_sk(sk)->duplicate_sack[0].start_seq) { - sk_rethink_txhash(sk); + if (TCP_SKB_CB(skb)->seq == tcp_sk(sk)->duplicate_sack[0].start_seq && + sk_rethink_txhash(sk)) NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDUPLICATEDATAREHASH); - } } static void tcp_send_dupack(struct sock *sk, const struct sk_buff *skb) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 58207c7769d0..777306b5bc22 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1595,6 +1595,8 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, tcp_move_syn(newtp, req); ireq->ireq_opt = NULL; } else { + newinet->inet_opt = NULL; + if (!req_unhash && found_dup_sk) { /* This code path should only be executed in the * syncookie case only @@ -1602,8 +1604,6 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, bh_unlock_sock(newsk); sock_put(newsk); newsk = NULL; - } else { - newinet->inet_opt = NULL; } } return newsk; @@ -1760,6 +1760,7 @@ int tcp_v4_early_demux(struct sk_buff *skb) bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) { u32 limit = READ_ONCE(sk->sk_rcvbuf) + READ_ONCE(sk->sk_sndbuf); + u32 tail_gso_size, tail_gso_segs; struct skb_shared_info *shinfo; const struct tcphdr *th; struct tcphdr *thtail; @@ -1767,6 +1768,7 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) unsigned int hdrlen; bool fragstolen; u32 gso_segs; + u32 gso_size; int delta; /* In case all data was pulled from skb frags (in __pskb_pull_tail()), @@ -1792,13 +1794,6 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) */ th = (const struct tcphdr *)skb->data; hdrlen = th->doff * 4; - shinfo = skb_shinfo(skb); - - if (!shinfo->gso_size) - shinfo->gso_size = skb->len - hdrlen; - - if (!shinfo->gso_segs) - shinfo->gso_segs = 1; tail = sk->sk_backlog.tail; if (!tail) @@ -1821,6 +1816,15 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) goto no_coalesce; __skb_pull(skb, hdrlen); + + shinfo = skb_shinfo(skb); + gso_size = shinfo->gso_size ?: skb->len; + gso_segs = shinfo->gso_segs ?: 1; + + shinfo = skb_shinfo(tail); + tail_gso_size = shinfo->gso_size ?: (tail->len - hdrlen); + tail_gso_segs = shinfo->gso_segs ?: 1; + if (skb_try_coalesce(tail, skb, &fragstolen, &delta)) { TCP_SKB_CB(tail)->end_seq = TCP_SKB_CB(skb)->end_seq; @@ -1847,11 +1851,8 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) } /* Not as strict as GRO. We only need to carry mss max value */ - skb_shinfo(tail)->gso_size = max(shinfo->gso_size, - skb_shinfo(tail)->gso_size); - - gso_segs = skb_shinfo(tail)->gso_segs + shinfo->gso_segs; - skb_shinfo(tail)->gso_segs = min_t(u32, gso_segs, 0xFFFF); + shinfo->gso_size = max(gso_size, tail_gso_size); + shinfo->gso_segs = min_t(u32, gso_segs + tail_gso_segs, 0xFFFF); sk->sk_backlog.len += delta; __NET_INC_STATS(sock_net(sk), diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 899d053cb10e..b0a6b0b30a6b 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -4084,6 +4084,7 @@ void tcp_send_probe0(struct sock *sk) /* Cancel probe timer, if it is not required. */ icsk->icsk_probes_out = 0; icsk->icsk_backoff = 0; + icsk->icsk_probes_tstamp = 0; return; } diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 6c62b9ea1320..faa92948441b 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -219,14 +219,8 @@ static int tcp_write_timeout(struct sock *sk) int retry_until; if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) { - if (icsk->icsk_retransmits) { - dst_negative_advice(sk); - } else { - sk_rethink_txhash(sk); - tp->timeout_rehash++; - __NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPTIMEOUTREHASH); - } + if (icsk->icsk_retransmits) + __dst_negative_advice(sk); retry_until = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries; expired = icsk->icsk_retransmits >= retry_until; } else { @@ -234,12 +228,7 @@ static int tcp_write_timeout(struct sock *sk) /* Black hole detection */ tcp_mtu_probing(icsk, sk); - dst_negative_advice(sk); - } else { - sk_rethink_txhash(sk); - tp->timeout_rehash++; - __NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPTIMEOUTREHASH); + __dst_negative_advice(sk); } retry_until = net->ipv4.sysctl_tcp_retries2; @@ -270,6 +259,11 @@ static int tcp_write_timeout(struct sock *sk) return 1; } + if (sk_rethink_txhash(sk)) { + tp->timeout_rehash++; + __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPTIMEOUTREHASH); + } + return 0; } @@ -349,6 +343,7 @@ static void tcp_probe_timer(struct sock *sk) if (tp->packets_out || !skb) { icsk->icsk_probes_out = 0; + icsk->icsk_probes_tstamp = 0; return; } @@ -360,13 +355,12 @@ static void tcp_probe_timer(struct sock *sk) * corresponding system limit. We also implement similar policy when * we use RTO to probe window in tcp_retransmit_timer(). */ - if (icsk->icsk_user_timeout) { - u32 elapsed = tcp_model_timeout(sk, icsk->icsk_probes_out, - tcp_probe0_base(sk)); - - if (elapsed >= icsk->icsk_user_timeout) - goto abort; - } + if (!icsk->icsk_probes_tstamp) + icsk->icsk_probes_tstamp = tcp_jiffies32; + else if (icsk->icsk_user_timeout && + (s32)(tcp_jiffies32 - icsk->icsk_probes_tstamp) >= + msecs_to_jiffies(icsk->icsk_user_timeout)) + goto abort; max_probes = sock_net(sk)->ipv4.sysctl_tcp_retries2; if (sock_flag(sk, SOCK_DEAD)) { diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 7103b0a89756..69ea76578abb 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -2555,7 +2555,8 @@ int udp_v4_early_demux(struct sk_buff *skb) */ if (!inet_sk(sk)->inet_daddr && in_dev) return ip_mc_validate_source(skb, iph->daddr, - iph->saddr, iph->tos, + iph->saddr, + iph->tos & IPTOS_RT_MASK, skb->dev, in_dev, &itag); } return 0; diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index ff39e94781bf..41249705d9e9 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -68,8 +68,8 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb, (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM)))); features &= skb->dev->hw_enc_features; - /* CRC checksum can't be handled by HW when it's a UDP tunneling packet. */ - features &= ~NETIF_F_SCTP_CRC; + if (need_csum) + features &= ~NETIF_F_SCTP_CRC; /* The only checksum offload we care about from here on out is the * outer one so strip the existing checksum feature flags and @@ -460,7 +460,8 @@ struct sk_buff *udp_gro_receive(struct list_head *head, struct sk_buff *skb, if (skb->dev->features & NETIF_F_GRO_FRAGLIST) NAPI_GRO_CB(skb)->is_flist = sk ? !udp_sk(sk)->gro_enabled: 1; - if ((sk && udp_sk(sk)->gro_enabled) || NAPI_GRO_CB(skb)->is_flist) { + if ((!sk && (skb->dev->features & NETIF_F_GRO_UDP_FWD)) || + (sk && udp_sk(sk)->gro_enabled) || NAPI_GRO_CB(skb)->is_flist) { pp = call_gro_receive(udp_gro_receive_segment, head, skb); return pp; } diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index eff2cacd5209..f2337fb756ac 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -205,6 +205,7 @@ static struct ipv6_devconf ipv6_devconf __read_mostly = { .max_desync_factor = MAX_DESYNC_FACTOR, .max_addresses = IPV6_MAX_ADDRESSES, .accept_ra_defrtr = 1, + .ra_defrtr_metric = IP6_RT_PRIO_USER, .accept_ra_from_local = 0, .accept_ra_min_hop_limit= 1, .accept_ra_pinfo = 1, @@ -260,6 +261,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = { .max_desync_factor = MAX_DESYNC_FACTOR, .max_addresses = IPV6_MAX_ADDRESSES, .accept_ra_defrtr = 1, + .ra_defrtr_metric = IP6_RT_PRIO_USER, .accept_ra_from_local = 0, .accept_ra_min_hop_limit= 1, .accept_ra_pinfo = 1, @@ -2467,8 +2469,9 @@ static void addrconf_add_mroute(struct net_device *dev) .fc_ifindex = dev->ifindex, .fc_dst_len = 8, .fc_flags = RTF_UP, - .fc_type = RTN_UNICAST, + .fc_type = RTN_MULTICAST, .fc_nlinfo.nl_net = dev_net(dev), + .fc_protocol = RTPROT_KERNEL, }; ipv6_addr_set(&cfg.fc_dst, htonl(0xFF000000), 0, 0, 0); @@ -5475,6 +5478,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf, array[DEVCONF_MAX_DESYNC_FACTOR] = cnf->max_desync_factor; array[DEVCONF_MAX_ADDRESSES] = cnf->max_addresses; array[DEVCONF_ACCEPT_RA_DEFRTR] = cnf->accept_ra_defrtr; + array[DEVCONF_RA_DEFRTR_METRIC] = cnf->ra_defrtr_metric; array[DEVCONF_ACCEPT_RA_MIN_HOP_LIMIT] = cnf->accept_ra_min_hop_limit; array[DEVCONF_ACCEPT_RA_PINFO] = cnf->accept_ra_pinfo; #ifdef CONFIG_IPV6_ROUTER_PREF @@ -6668,6 +6672,14 @@ static const struct ctl_table addrconf_sysctl[] = { .proc_handler = proc_dointvec, }, { + .procname = "ra_defrtr_metric", + .data = &ipv6_devconf.ra_defrtr_metric, + .maxlen = sizeof(u32), + .mode = 0644, + .proc_handler = proc_douintvec_minmax, + .extra1 = (void *)SYSCTL_ONE, + }, + { .procname = "accept_ra_min_hop_limit", .data = &ipv6_devconf.accept_ra_min_hop_limit, .maxlen = sizeof(int), diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 76717478f173..c467c6419893 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1173,6 +1173,7 @@ static void ndisc_router_discovery(struct sk_buff *skb) struct neighbour *neigh = NULL; struct inet6_dev *in6_dev; struct fib6_info *rt = NULL; + u32 defrtr_usr_metric; struct net *net; int lifetime; struct ndisc_options ndopts; @@ -1303,18 +1304,21 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } } - if (rt && lifetime == 0) { + /* Set default route metric as specified by user */ + defrtr_usr_metric = in6_dev->cnf.ra_defrtr_metric; + /* delete the route if lifetime is 0 or if metric needs change */ + if (rt && (lifetime == 0 || rt->fib6_metric != defrtr_usr_metric)) { ip6_del_rt(net, rt, false); rt = NULL; } - ND_PRINTK(3, info, "RA: rt: %p lifetime: %d, for dev: %s\n", - rt, lifetime, skb->dev->name); + ND_PRINTK(3, info, "RA: rt: %p lifetime: %d, metric: %d, for dev: %s\n", + rt, lifetime, defrtr_usr_metric, skb->dev->name); if (!rt && lifetime) { ND_PRINTK(3, info, "RA: adding default router\n"); rt = rt6_add_dflt_router(net, &ipv6_hdr(skb)->saddr, - skb->dev, pref); + skb->dev, pref, defrtr_usr_metric); if (!rt) { ND_PRINTK(0, err, "RA: %s failed to add default route\n", diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 188e114b29b4..41d8f801b75f 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -4252,11 +4252,12 @@ struct fib6_info *rt6_get_dflt_router(struct net *net, struct fib6_info *rt6_add_dflt_router(struct net *net, const struct in6_addr *gwaddr, struct net_device *dev, - unsigned int pref) + unsigned int pref, + u32 defrtr_usr_metric) { struct fib6_config cfg = { .fc_table = l3mdev_fib_table(dev) ? : RT6_TABLE_DFLT, - .fc_metric = IP6_RT_PRIO_USER, + .fc_metric = defrtr_usr_metric, .fc_ifindex = dev->ifindex, .fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | RTF_UP | RTF_EXPIRES | RTF_PREF(pref), diff --git a/net/l3mdev/Makefile b/net/l3mdev/Makefile index 59755a9e2f9b..9e7da0acc58c 100644 --- a/net/l3mdev/Makefile +++ b/net/l3mdev/Makefile @@ -3,4 +3,4 @@ # Makefile for the L3 device API # -obj-$(CONFIG_NET_L3_MASTER_DEV) += l3mdev.o +obj-y += l3mdev.o diff --git a/net/llc/Kconfig b/net/llc/Kconfig index b0e646ac47eb..7f79f5e134f9 100644 --- a/net/llc/Kconfig +++ b/net/llc/Kconfig @@ -1,7 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only config LLC tristate - depends on NET config LLC2 tristate "ANSI/IEEE 802.2 LLC type 2 Support" diff --git a/net/mac80211/Makefile b/net/mac80211/Makefile index ad04c361cba5..23d25e8b2358 100644 --- a/net/mac80211/Makefile +++ b/net/mac80211/Makefile @@ -56,11 +56,9 @@ mac80211-$(CONFIG_PM) += pm.o CFLAGS_trace.o := -I$(src) rc80211_minstrel-y := \ - rc80211_minstrel.o \ rc80211_minstrel_ht.o rc80211_minstrel-$(CONFIG_MAC80211_DEBUGFS) += \ - rc80211_minstrel_debugfs.o \ rc80211_minstrel_ht_debugfs.o mac80211-$(CONFIG_MAC80211_RC_MINSTREL) += $(rc80211_minstrel-y) diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index 48f144f107d5..5296898875ff 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -120,18 +120,17 @@ static ssize_t aqm_write(struct file *file, { struct ieee80211_local *local = file->private_data; char buf[100]; - size_t len; - if (count > sizeof(buf)) + if (count >= sizeof(buf)) return -EINVAL; if (copy_from_user(buf, user_buf, count)) return -EFAULT; - buf[sizeof(buf) - 1] = '\0'; - len = strlen(buf); - if (len > 0 && buf[len-1] == '\n') - buf[len-1] = 0; + if (count && buf[count - 1] == '\n') + buf[count - 1] = '\0'; + else + buf[count] = '\0'; if (sscanf(buf, "fq_limit %u", &local->fq.limit) == 1) return count; @@ -177,18 +176,17 @@ static ssize_t airtime_flags_write(struct file *file, { struct ieee80211_local *local = file->private_data; char buf[16]; - size_t len; - if (count > sizeof(buf)) + if (count >= sizeof(buf)) return -EINVAL; if (copy_from_user(buf, user_buf, count)) return -EFAULT; - buf[sizeof(buf) - 1] = 0; - len = strlen(buf); - if (len > 0 && buf[len - 1] == '\n') - buf[len - 1] = 0; + if (count && buf[count - 1] == '\n') + buf[count - 1] = '\0'; + else + buf[count] = '\0'; if (kstrtou16(buf, 0, &local->airtime_flags)) return -EINVAL; @@ -237,20 +235,19 @@ static ssize_t aql_txq_limit_write(struct file *file, { struct ieee80211_local *local = file->private_data; char buf[100]; - size_t len; u32 ac, q_limit_low, q_limit_high, q_limit_low_old, q_limit_high_old; struct sta_info *sta; - if (count > sizeof(buf)) + if (count >= sizeof(buf)) return -EINVAL; if (copy_from_user(buf, user_buf, count)) return -EFAULT; - buf[sizeof(buf) - 1] = 0; - len = strlen(buf); - if (len > 0 && buf[len - 1] == '\n') - buf[len - 1] = 0; + if (count && buf[count - 1] == '\n') + buf[count - 1] = '\0'; + else + buf[count] = '\0'; if (sscanf(buf, "%u %u %u", &ac, &q_limit_low, &q_limit_high) != 3) return -EINVAL; @@ -284,6 +281,56 @@ static const struct file_operations aql_txq_limit_ops = { .llseek = default_llseek, }; +static ssize_t aql_enable_read(struct file *file, char __user *user_buf, + size_t count, loff_t *ppos) +{ + char buf[3]; + int len; + + len = scnprintf(buf, sizeof(buf), "%d\n", + !static_key_false(&aql_disable.key)); + + return simple_read_from_buffer(user_buf, count, ppos, buf, len); +} + +static ssize_t aql_enable_write(struct file *file, const char __user *user_buf, + size_t count, loff_t *ppos) +{ + bool aql_disabled = static_key_false(&aql_disable.key); + char buf[3]; + size_t len; + + if (count > sizeof(buf)) + return -EINVAL; + + if (copy_from_user(buf, user_buf, count)) + return -EFAULT; + + buf[sizeof(buf) - 1] = '\0'; + len = strlen(buf); + if (len > 0 && buf[len - 1] == '\n') + buf[len - 1] = 0; + + if (buf[0] == '0' && buf[1] == '\0') { + if (!aql_disabled) + static_branch_inc(&aql_disable); + } else if (buf[0] == '1' && buf[1] == '\0') { + if (aql_disabled) + static_branch_dec(&aql_disable); + } else { + return -EINVAL; + } + + return count; +} + +static const struct file_operations aql_enable_ops = { + .write = aql_enable_write, + .read = aql_enable_read, + .open = simple_open, + .llseek = default_llseek, +}; + static ssize_t force_tx_status_read(struct file *file, char __user *user_buf, size_t count, @@ -306,18 +353,17 @@ static ssize_t force_tx_status_write(struct file *file, { struct ieee80211_local *local = file->private_data; char buf[3]; - size_t len; - if (count > sizeof(buf)) + if (count >= sizeof(buf)) return -EINVAL; if (copy_from_user(buf, user_buf, count)) return -EFAULT; - buf[sizeof(buf) - 1] = '\0'; - len = strlen(buf); - if (len > 0 && buf[len - 1] == '\n') - buf[len - 1] = 0; + if (count && buf[count - 1] == '\n') + buf[count - 1] = '\0'; + else + buf[count] = '\0'; if (buf[0] == '0' && buf[1] == '\0') local->force_tx_status = 0; @@ -409,6 +455,7 @@ static const char *hw_flag_names[] = { FLAG(SUPPORTS_ONLY_HE_MULTI_BSSID), FLAG(AMPDU_KEYBORDER_SUPPORT), FLAG(SUPPORTS_TX_ENCAP_OFFLOAD), + FLAG(SUPPORTS_RX_DECAP_OFFLOAD), #undef FLAG }; @@ -572,6 +619,7 @@ void debugfs_hw_add(struct ieee80211_local *local) DEBUGFS_ADD(power); DEBUGFS_ADD(hw_conf); DEBUGFS_ADD_MODE(force_tx_status, 0600); + DEBUGFS_ADD_MODE(aql_enable, 0600); if (local->ops->wake_tx_queue) DEBUGFS_ADD_MODE(aqm, 0600); diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c index eb4bb79d936a..5a27c61a7b38 100644 --- a/net/mac80211/debugfs_sta.c +++ b/net/mac80211/debugfs_sta.c @@ -79,6 +79,7 @@ static const char * const sta_flag_names[] = { FLAG(MPSP_RECIPIENT), FLAG(PS_DELIVER), FLAG(USES_ENCRYPTION), + FLAG(DECAP_OFFLOAD), #undef FLAG }; diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index bcdfd19a596b..604ca59937f0 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -1413,4 +1413,20 @@ static inline void drv_sta_set_4addr(struct ieee80211_local *local, trace_drv_return_void(local); } +static inline void drv_sta_set_decap_offload(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta, + bool enabled) +{ + sdata = get_bss_sdata(sdata); + if (!check_sdata_in_driver(sdata)) + return; + + trace_drv_sta_set_decap_offload(local, sdata, sta, enabled); + if (local->ops->sta_set_decap_offload) + local->ops->sta_set_decap_offload(&local->hw, &sdata->vif, sta, + enabled); + trace_drv_return_void(local); +} + #endif /* __MAC80211_DRIVER_OPS */ diff --git a/net/mac80211/he.c b/net/mac80211/he.c index cc26f239838b..0c0b970835ce 100644 --- a/net/mac80211/he.c +++ b/net/mac80211/he.c @@ -52,6 +52,57 @@ ieee80211_update_from_he_6ghz_capa(const struct ieee80211_he_6ghz_capa *he_6ghz_ sta->sta.he_6ghz_capa = *he_6ghz_capa; } +static void ieee80211_he_mcs_disable(__le16 *he_mcs) +{ + u32 i; + + for (i = 0; i < 8; i++) + *he_mcs |= cpu_to_le16(IEEE80211_HE_MCS_NOT_SUPPORTED << i * 2); +} + +static void ieee80211_he_mcs_intersection(__le16 *he_own_rx, __le16 *he_peer_rx, + __le16 *he_own_tx, __le16 *he_peer_tx) +{ + u32 i; + u16 own_rx, own_tx, peer_rx, peer_tx; + + for (i = 0; i < 8; i++) { + own_rx = le16_to_cpu(*he_own_rx); + own_rx = (own_rx >> i * 2) & IEEE80211_HE_MCS_NOT_SUPPORTED; + + own_tx = le16_to_cpu(*he_own_tx); + own_tx = (own_tx >> i * 2) & IEEE80211_HE_MCS_NOT_SUPPORTED; + + peer_rx = le16_to_cpu(*he_peer_rx); + peer_rx = (peer_rx >> i * 2) & IEEE80211_HE_MCS_NOT_SUPPORTED; + + peer_tx = le16_to_cpu(*he_peer_tx); + peer_tx = (peer_tx >> i * 2) & IEEE80211_HE_MCS_NOT_SUPPORTED; + + if (peer_tx != IEEE80211_HE_MCS_NOT_SUPPORTED) { + if (own_rx == IEEE80211_HE_MCS_NOT_SUPPORTED) + peer_tx = IEEE80211_HE_MCS_NOT_SUPPORTED; + else if (own_rx < peer_tx) + peer_tx = own_rx; + } + + if (peer_rx != IEEE80211_HE_MCS_NOT_SUPPORTED) { + if (own_tx == IEEE80211_HE_MCS_NOT_SUPPORTED) + peer_rx = IEEE80211_HE_MCS_NOT_SUPPORTED; + else if (own_tx < peer_rx) + peer_rx = own_tx; + } + + *he_peer_rx &= + ~cpu_to_le16(IEEE80211_HE_MCS_NOT_SUPPORTED << i * 2); + *he_peer_rx |= cpu_to_le16(peer_rx << i * 2); + + *he_peer_tx &= + ~cpu_to_le16(IEEE80211_HE_MCS_NOT_SUPPORTED << i * 2); + *he_peer_tx |= cpu_to_le16(peer_tx << i * 2); + } +} + void ieee80211_he_cap_ie_to_sta_he_cap(struct ieee80211_sub_if_data *sdata, struct ieee80211_supported_band *sband, @@ -60,10 +111,12 @@ ieee80211_he_cap_ie_to_sta_he_cap(struct ieee80211_sub_if_data *sdata, struct sta_info *sta) { struct ieee80211_sta_he_cap *he_cap = &sta->sta.he_cap; + struct ieee80211_sta_he_cap own_he_cap = sband->iftype_data->he_cap; struct ieee80211_he_cap_elem *he_cap_ie_elem = (void *)he_cap_ie; u8 he_ppe_size; u8 mcs_nss_size; u8 he_total_size; + bool own_160, peer_160, own_80p80, peer_80p80; memset(he_cap, 0, sizeof(*he_cap)); @@ -101,6 +154,45 @@ ieee80211_he_cap_ie_to_sta_he_cap(struct ieee80211_sub_if_data *sdata, if (sband->band == NL80211_BAND_6GHZ && he_6ghz_capa) ieee80211_update_from_he_6ghz_capa(he_6ghz_capa, sta); + + ieee80211_he_mcs_intersection(&own_he_cap.he_mcs_nss_supp.rx_mcs_80, + &he_cap->he_mcs_nss_supp.rx_mcs_80, + &own_he_cap.he_mcs_nss_supp.tx_mcs_80, + &he_cap->he_mcs_nss_supp.tx_mcs_80); + + own_160 = own_he_cap.he_cap_elem.phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G; + peer_160 = he_cap->he_cap_elem.phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G; + + if (peer_160 && own_160) { + ieee80211_he_mcs_intersection(&own_he_cap.he_mcs_nss_supp.rx_mcs_160, + &he_cap->he_mcs_nss_supp.rx_mcs_160, + &own_he_cap.he_mcs_nss_supp.tx_mcs_160, + &he_cap->he_mcs_nss_supp.tx_mcs_160); + } else if (peer_160 && !own_160) { + ieee80211_he_mcs_disable(&he_cap->he_mcs_nss_supp.rx_mcs_160); + ieee80211_he_mcs_disable(&he_cap->he_mcs_nss_supp.tx_mcs_160); + he_cap->he_cap_elem.phy_cap_info[0] &= + ~IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G; + } + + own_80p80 = own_he_cap.he_cap_elem.phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G; + peer_80p80 = he_cap->he_cap_elem.phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G; + + if (peer_80p80 && own_80p80) { + ieee80211_he_mcs_intersection(&own_he_cap.he_mcs_nss_supp.rx_mcs_80p80, + &he_cap->he_mcs_nss_supp.rx_mcs_80p80, + &own_he_cap.he_mcs_nss_supp.tx_mcs_80p80, + &he_cap->he_mcs_nss_supp.tx_mcs_80p80); + } else if (peer_80p80 && !own_80p80) { + ieee80211_he_mcs_disable(&he_cap->he_mcs_nss_supp.rx_mcs_80p80); + ieee80211_he_mcs_disable(&he_cap->he_mcs_nss_supp.tx_mcs_80p80); + he_cap->he_cap_elem.phy_cap_info[0] &= + ~IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G; + } } void diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 8bf9c0e974d6..982fdc12abd9 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -848,7 +848,6 @@ enum txq_info_flags { */ struct txq_info { struct fq_tin tin; - struct fq_flow def_flow; struct codel_vars def_cvars; struct codel_stats cstats; struct sk_buff_head frags; @@ -1143,6 +1142,8 @@ enum mac80211_scan_state { SCAN_ABORT, }; +DECLARE_STATIC_KEY_FALSE(aql_disable); + struct ieee80211_local { /* embed the driver visible part. * don't cast (use the static inlines below), but we keep diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 3b9ec4ef81c3..4bc350cf4eee 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -357,11 +357,14 @@ static int ieee80211_open(struct net_device *dev) if (err) return err; - return ieee80211_do_open(&sdata->wdev, true); + wiphy_lock(sdata->local->hw.wiphy); + err = ieee80211_do_open(&sdata->wdev, true); + wiphy_unlock(sdata->local->hw.wiphy); + + return err; } -static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, - bool going_down) +static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_down) { struct ieee80211_local *local = sdata->local; unsigned long flags; @@ -637,7 +640,9 @@ static int ieee80211_stop(struct net_device *dev) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); + wiphy_lock(sdata->local->hw.wiphy); ieee80211_do_stop(sdata, true); + wiphy_unlock(sdata->local->hw.wiphy); return 0; } @@ -765,7 +770,7 @@ static const struct net_device_ops ieee80211_dataif_8023_ops = { .ndo_get_stats64 = ieee80211_get_stats64, }; -static bool ieee80211_iftype_supports_encap_offload(enum nl80211_iftype iftype) +static bool ieee80211_iftype_supports_hdr_offload(enum nl80211_iftype iftype) { switch (iftype) { /* P2P GO and client are mapped to AP/STATION types */ @@ -785,7 +790,7 @@ static bool ieee80211_set_sdata_offload_flags(struct ieee80211_sub_if_data *sdat flags = sdata->vif.offload_flags; if (ieee80211_hw_check(&local->hw, SUPPORTS_TX_ENCAP_OFFLOAD) && - ieee80211_iftype_supports_encap_offload(sdata->vif.type)) { + ieee80211_iftype_supports_hdr_offload(sdata->vif.type)) { flags |= IEEE80211_OFFLOAD_ENCAP_ENABLED; if (!ieee80211_hw_check(&local->hw, SUPPORTS_TX_FRAG) && @@ -798,10 +803,21 @@ static bool ieee80211_set_sdata_offload_flags(struct ieee80211_sub_if_data *sdat flags &= ~IEEE80211_OFFLOAD_ENCAP_ENABLED; } + if (ieee80211_hw_check(&local->hw, SUPPORTS_RX_DECAP_OFFLOAD) && + ieee80211_iftype_supports_hdr_offload(sdata->vif.type)) { + flags |= IEEE80211_OFFLOAD_DECAP_ENABLED; + + if (local->monitors) + flags &= ~IEEE80211_OFFLOAD_DECAP_ENABLED; + } else { + flags &= ~IEEE80211_OFFLOAD_DECAP_ENABLED; + } + if (sdata->vif.offload_flags == flags) return false; sdata->vif.offload_flags = flags; + ieee80211_check_fast_rx_iface(sdata); return true; } @@ -819,7 +835,7 @@ static void ieee80211_set_vif_encap_ops(struct ieee80211_sub_if_data *sdata) } if (!ieee80211_hw_check(&local->hw, SUPPORTS_TX_ENCAP_OFFLOAD) || - !ieee80211_iftype_supports_encap_offload(bss->vif.type)) + !ieee80211_iftype_supports_hdr_offload(bss->vif.type)) return; enabled = bss->vif.offload_flags & IEEE80211_OFFLOAD_ENCAP_ENABLED; @@ -1965,7 +1981,7 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name, ndev->min_mtu = 256; ndev->max_mtu = local->hw.max_mtu; - ret = register_netdevice(ndev); + ret = cfg80211_register_netdevice(ndev); if (ret) { free_netdev(ndev); return ret; @@ -1995,10 +2011,9 @@ void ieee80211_if_remove(struct ieee80211_sub_if_data *sdata) synchronize_rcu(); - if (sdata->dev) { - unregister_netdevice(sdata->dev); - } else { - cfg80211_unregister_wdev(&sdata->wdev); + cfg80211_unregister_wdev(&sdata->wdev); + + if (!sdata->dev) { ieee80211_teardown_sdata(sdata); kfree(sdata); } @@ -2047,13 +2062,16 @@ void ieee80211_remove_interfaces(struct ieee80211_local *local) list_add(&sdata->list, &wdev_list); } mutex_unlock(&local->iflist_mtx); + unregister_netdevice_many(&unreg_list); + wiphy_lock(local->hw.wiphy); list_for_each_entry_safe(sdata, tmp, &wdev_list, list) { list_del(&sdata->list); cfg80211_unregister_wdev(&sdata->wdev); kfree(sdata); } + wiphy_unlock(local->hw.wiphy); } static int netdev_notify(struct notifier_block *nb, diff --git a/net/mac80211/key.c b/net/mac80211/key.c index a4817aa4b171..56c068cb49c4 100644 --- a/net/mac80211/key.c +++ b/net/mac80211/key.c @@ -887,7 +887,7 @@ void ieee80211_reenable_keys(struct ieee80211_sub_if_data *sdata) struct ieee80211_key *key; struct ieee80211_sub_if_data *vlan; - ASSERT_RTNL(); + lockdep_assert_wiphy(sdata->local->hw.wiphy); mutex_lock(&sdata->local->key_mtx); @@ -924,7 +924,7 @@ void ieee80211_iter_keys(struct ieee80211_hw *hw, struct ieee80211_key *key, *tmp; struct ieee80211_sub_if_data *sdata; - ASSERT_RTNL(); + lockdep_assert_wiphy(hw->wiphy); mutex_lock(&local->key_mtx); if (vif) { diff --git a/net/mac80211/main.c b/net/mac80211/main.c index dee88ec566ad..4f3f8bb58e76 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -261,7 +261,9 @@ static void ieee80211_restart_work(struct work_struct *work) "%s called with hardware scan in progress\n", __func__); flush_work(&local->radar_detected_work); + /* we might do interface manipulations, so need both */ rtnl_lock(); + wiphy_lock(local->hw.wiphy); list_for_each_entry(sdata, &local->interfaces, list) { /* * XXX: there may be more work for other vif types and even @@ -293,6 +295,7 @@ static void ieee80211_restart_work(struct work_struct *work) synchronize_net(); ieee80211_reconfig(local); + wiphy_unlock(local->hw.wiphy); rtnl_unlock(); } @@ -1272,6 +1275,7 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) rate_control_add_debugfs(local); rtnl_lock(); + wiphy_lock(hw->wiphy); /* add one default STA interface if supported */ if (local->hw.wiphy->interface_modes & BIT(NL80211_IFTYPE_STATION) && @@ -1285,6 +1289,7 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) "Failed to add default virtual iface\n"); } + wiphy_unlock(hw->wiphy); rtnl_unlock(); #ifdef CONFIG_INET diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c index ae378a41c927..7809a906d7fe 100644 --- a/net/mac80211/pm.c +++ b/net/mac80211/pm.c @@ -1,4 +1,8 @@ // SPDX-License-Identifier: GPL-2.0 +/* + * Portions + * Copyright (C) 2020-2021 Intel Corporation + */ #include <net/mac80211.h> #include <net/rtnetlink.h> @@ -11,7 +15,7 @@ static void ieee80211_sched_scan_cancel(struct ieee80211_local *local) { if (ieee80211_request_sched_scan_stop(local)) return; - cfg80211_sched_scan_stopped_rtnl(local->hw.wiphy, 0); + cfg80211_sched_scan_stopped_locked(local->hw.wiphy, 0); } int __ieee80211_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan) diff --git a/net/mac80211/rc80211_minstrel.c b/net/mac80211/rc80211_minstrel.c deleted file mode 100644 index b13b1da19386..000000000000 --- a/net/mac80211/rc80211_minstrel.c +++ /dev/null @@ -1,574 +0,0 @@ -/* - * Copyright (C) 2008 Felix Fietkau <nbd@openwrt.org> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * Based on minstrel.c: - * Copyright (C) 2005-2007 Derek Smithies <derek@indranet.co.nz> - * Sponsored by Indranet Technologies Ltd - * - * Based on sample.c: - * Copyright (c) 2005 John Bicket - * 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, - * without modification. - * 2. Redistributions in binary form must reproduce at minimum a disclaimer - * similar to the "NO WARRANTY" disclaimer below ("Disclaimer") and any - * redistribution must be conditioned upon including a substantially - * similar Disclaimer requirement for further binary redistribution. - * 3. Neither the names of the above-listed copyright holders nor the names - * of any contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * Alternatively, this software may be distributed under the terms of the - * GNU General Public License ("GPL") version 2 as published by the Free - * Software Foundation. - * - * NO WARRANTY - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTIBILITY - * AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL - * THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR 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 DAMAGES. - */ -#include <linux/netdevice.h> -#include <linux/types.h> -#include <linux/skbuff.h> -#include <linux/debugfs.h> -#include <linux/random.h> -#include <linux/ieee80211.h> -#include <linux/slab.h> -#include <net/mac80211.h> -#include "rate.h" -#include "rc80211_minstrel.h" - -#define SAMPLE_TBL(_mi, _idx, _col) \ - _mi->sample_table[(_idx * SAMPLE_COLUMNS) + _col] - -/* convert mac80211 rate index to local array index */ -static inline int -rix_to_ndx(struct minstrel_sta_info *mi, int rix) -{ - int i = rix; - for (i = rix; i >= 0; i--) - if (mi->r[i].rix == rix) - break; - return i; -} - -/* return current EMWA throughput */ -int minstrel_get_tp_avg(struct minstrel_rate *mr, int prob_avg) -{ - int usecs; - - usecs = mr->perfect_tx_time; - if (!usecs) - usecs = 1000000; - - /* reset thr. below 10% success */ - if (mr->stats.prob_avg < MINSTREL_FRAC(10, 100)) - return 0; - - if (prob_avg > MINSTREL_FRAC(90, 100)) - return MINSTREL_TRUNC(100000 * (MINSTREL_FRAC(90, 100) / usecs)); - else - return MINSTREL_TRUNC(100000 * (prob_avg / usecs)); -} - -/* find & sort topmost throughput rates */ -static inline void -minstrel_sort_best_tp_rates(struct minstrel_sta_info *mi, int i, u8 *tp_list) -{ - int j; - struct minstrel_rate_stats *tmp_mrs; - struct minstrel_rate_stats *cur_mrs = &mi->r[i].stats; - - for (j = MAX_THR_RATES; j > 0; --j) { - tmp_mrs = &mi->r[tp_list[j - 1]].stats; - if (minstrel_get_tp_avg(&mi->r[i], cur_mrs->prob_avg) <= - minstrel_get_tp_avg(&mi->r[tp_list[j - 1]], tmp_mrs->prob_avg)) - break; - } - - if (j < MAX_THR_RATES - 1) - memmove(&tp_list[j + 1], &tp_list[j], MAX_THR_RATES - (j + 1)); - if (j < MAX_THR_RATES) - tp_list[j] = i; -} - -static void -minstrel_set_rate(struct minstrel_sta_info *mi, struct ieee80211_sta_rates *ratetbl, - int offset, int idx) -{ - struct minstrel_rate *r = &mi->r[idx]; - - ratetbl->rate[offset].idx = r->rix; - ratetbl->rate[offset].count = r->adjusted_retry_count; - ratetbl->rate[offset].count_cts = r->retry_count_cts; - ratetbl->rate[offset].count_rts = r->stats.retry_count_rtscts; -} - -static void -minstrel_update_rates(struct minstrel_priv *mp, struct minstrel_sta_info *mi) -{ - struct ieee80211_sta_rates *ratetbl; - int i = 0; - - ratetbl = kzalloc(sizeof(*ratetbl), GFP_ATOMIC); - if (!ratetbl) - return; - - /* Start with max_tp_rate */ - minstrel_set_rate(mi, ratetbl, i++, mi->max_tp_rate[0]); - - if (mp->hw->max_rates >= 3) { - /* At least 3 tx rates supported, use max_tp_rate2 next */ - minstrel_set_rate(mi, ratetbl, i++, mi->max_tp_rate[1]); - } - - if (mp->hw->max_rates >= 2) { - /* At least 2 tx rates supported, use max_prob_rate next */ - minstrel_set_rate(mi, ratetbl, i++, mi->max_prob_rate); - } - - /* Use lowest rate last */ - ratetbl->rate[i].idx = mi->lowest_rix; - ratetbl->rate[i].count = mp->max_retry; - ratetbl->rate[i].count_cts = mp->max_retry; - ratetbl->rate[i].count_rts = mp->max_retry; - - rate_control_set_rates(mp->hw, mi->sta, ratetbl); -} - -/* -* Recalculate statistics and counters of a given rate -*/ -void -minstrel_calc_rate_stats(struct minstrel_priv *mp, - struct minstrel_rate_stats *mrs) -{ - unsigned int cur_prob; - - if (unlikely(mrs->attempts > 0)) { - mrs->sample_skipped = 0; - cur_prob = MINSTREL_FRAC(mrs->success, mrs->attempts); - if (mp->new_avg) { - minstrel_filter_avg_add(&mrs->prob_avg, - &mrs->prob_avg_1, cur_prob); - } else if (unlikely(!mrs->att_hist)) { - mrs->prob_avg = cur_prob; - } else { - /*update exponential weighted moving avarage */ - mrs->prob_avg = minstrel_ewma(mrs->prob_avg, - cur_prob, - EWMA_LEVEL); - } - mrs->att_hist += mrs->attempts; - mrs->succ_hist += mrs->success; - } else { - mrs->sample_skipped++; - } - - mrs->last_success = mrs->success; - mrs->last_attempts = mrs->attempts; - mrs->success = 0; - mrs->attempts = 0; -} - -static void -minstrel_update_stats(struct minstrel_priv *mp, struct minstrel_sta_info *mi) -{ - u8 tmp_tp_rate[MAX_THR_RATES]; - u8 tmp_prob_rate = 0; - int i, tmp_cur_tp, tmp_prob_tp; - - for (i = 0; i < MAX_THR_RATES; i++) - tmp_tp_rate[i] = 0; - - for (i = 0; i < mi->n_rates; i++) { - struct minstrel_rate *mr = &mi->r[i]; - struct minstrel_rate_stats *mrs = &mi->r[i].stats; - struct minstrel_rate_stats *tmp_mrs = &mi->r[tmp_prob_rate].stats; - - /* Update statistics of success probability per rate */ - minstrel_calc_rate_stats(mp, mrs); - - /* Sample less often below the 10% chance of success. - * Sample less often above the 95% chance of success. */ - if (mrs->prob_avg > MINSTREL_FRAC(95, 100) || - mrs->prob_avg < MINSTREL_FRAC(10, 100)) { - mr->adjusted_retry_count = mrs->retry_count >> 1; - if (mr->adjusted_retry_count > 2) - mr->adjusted_retry_count = 2; - mr->sample_limit = 4; - } else { - mr->sample_limit = -1; - mr->adjusted_retry_count = mrs->retry_count; - } - if (!mr->adjusted_retry_count) - mr->adjusted_retry_count = 2; - - minstrel_sort_best_tp_rates(mi, i, tmp_tp_rate); - - /* To determine the most robust rate (max_prob_rate) used at - * 3rd mmr stage we distinct between two cases: - * (1) if any success probabilitiy >= 95%, out of those rates - * choose the maximum throughput rate as max_prob_rate - * (2) if all success probabilities < 95%, the rate with - * highest success probability is chosen as max_prob_rate */ - if (mrs->prob_avg >= MINSTREL_FRAC(95, 100)) { - tmp_cur_tp = minstrel_get_tp_avg(mr, mrs->prob_avg); - tmp_prob_tp = minstrel_get_tp_avg(&mi->r[tmp_prob_rate], - tmp_mrs->prob_avg); - if (tmp_cur_tp >= tmp_prob_tp) - tmp_prob_rate = i; - } else { - if (mrs->prob_avg >= tmp_mrs->prob_avg) - tmp_prob_rate = i; - } - } - - /* Assign the new rate set */ - memcpy(mi->max_tp_rate, tmp_tp_rate, sizeof(mi->max_tp_rate)); - mi->max_prob_rate = tmp_prob_rate; - -#ifdef CONFIG_MAC80211_DEBUGFS - /* use fixed index if set */ - if (mp->fixed_rate_idx != -1) { - mi->max_tp_rate[0] = mp->fixed_rate_idx; - mi->max_tp_rate[1] = mp->fixed_rate_idx; - mi->max_prob_rate = mp->fixed_rate_idx; - } -#endif - - /* Reset update timer */ - mi->last_stats_update = jiffies; - - minstrel_update_rates(mp, mi); -} - -static void -minstrel_tx_status(void *priv, struct ieee80211_supported_band *sband, - void *priv_sta, struct ieee80211_tx_status *st) -{ - struct ieee80211_tx_info *info = st->info; - struct minstrel_priv *mp = priv; - struct minstrel_sta_info *mi = priv_sta; - struct ieee80211_tx_rate *ar = info->status.rates; - int i, ndx; - int success; - - success = !!(info->flags & IEEE80211_TX_STAT_ACK); - - for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) { - if (ar[i].idx < 0 || !ar[i].count) - break; - - ndx = rix_to_ndx(mi, ar[i].idx); - if (ndx < 0) - continue; - - mi->r[ndx].stats.attempts += ar[i].count; - - if ((i != IEEE80211_TX_MAX_RATES - 1) && (ar[i + 1].idx < 0)) - mi->r[ndx].stats.success += success; - } - - if (time_after(jiffies, mi->last_stats_update + - mp->update_interval / (mp->new_avg ? 2 : 1))) - minstrel_update_stats(mp, mi); -} - - -static inline unsigned int -minstrel_get_retry_count(struct minstrel_rate *mr, - struct ieee80211_tx_info *info) -{ - u8 retry = mr->adjusted_retry_count; - - if (info->control.use_rts) - retry = max_t(u8, 2, min(mr->stats.retry_count_rtscts, retry)); - else if (info->control.use_cts_prot) - retry = max_t(u8, 2, min(mr->retry_count_cts, retry)); - return retry; -} - - -static int -minstrel_get_next_sample(struct minstrel_sta_info *mi) -{ - unsigned int sample_ndx; - sample_ndx = SAMPLE_TBL(mi, mi->sample_row, mi->sample_column); - mi->sample_row++; - if ((int) mi->sample_row >= mi->n_rates) { - mi->sample_row = 0; - mi->sample_column++; - if (mi->sample_column >= SAMPLE_COLUMNS) - mi->sample_column = 0; - } - return sample_ndx; -} - -static void -minstrel_get_rate(void *priv, struct ieee80211_sta *sta, - void *priv_sta, struct ieee80211_tx_rate_control *txrc) -{ - struct sk_buff *skb = txrc->skb; - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - struct minstrel_sta_info *mi = priv_sta; - struct minstrel_priv *mp = priv; - struct ieee80211_tx_rate *rate = &info->control.rates[0]; - struct minstrel_rate *msr, *mr; - unsigned int ndx; - bool mrr_capable; - bool prev_sample; - int delta; - int sampling_ratio; - - /* check multi-rate-retry capabilities & adjust lookaround_rate */ - mrr_capable = mp->has_mrr && - !txrc->rts && - !txrc->bss_conf->use_cts_prot; - if (mrr_capable) - sampling_ratio = mp->lookaround_rate_mrr; - else - sampling_ratio = mp->lookaround_rate; - - /* increase sum packet counter */ - mi->total_packets++; - -#ifdef CONFIG_MAC80211_DEBUGFS - if (mp->fixed_rate_idx != -1) - return; -#endif - - /* Don't use EAPOL frames for sampling on non-mrr hw */ - if (mp->hw->max_rates == 1 && - (info->control.flags & IEEE80211_TX_CTRL_PORT_CTRL_PROTO)) - return; - - delta = (mi->total_packets * sampling_ratio / 100) - - mi->sample_packets; - - /* delta < 0: no sampling required */ - prev_sample = mi->prev_sample; - mi->prev_sample = false; - if (delta < 0 || (!mrr_capable && prev_sample)) - return; - - if (mi->total_packets >= 10000) { - mi->sample_packets = 0; - mi->total_packets = 0; - } else if (delta > mi->n_rates * 2) { - /* With multi-rate retry, not every planned sample - * attempt actually gets used, due to the way the retry - * chain is set up - [max_tp,sample,prob,lowest] for - * sample_rate < max_tp. - * - * If there's too much sampling backlog and the link - * starts getting worse, minstrel would start bursting - * out lots of sampling frames, which would result - * in a large throughput loss. */ - mi->sample_packets += (delta - mi->n_rates * 2); - } - - /* get next random rate sample */ - ndx = minstrel_get_next_sample(mi); - msr = &mi->r[ndx]; - mr = &mi->r[mi->max_tp_rate[0]]; - - /* Decide if direct ( 1st mrr stage) or indirect (2nd mrr stage) - * rate sampling method should be used. - * Respect such rates that are not sampled for 20 interations. - */ - if (msr->perfect_tx_time < mr->perfect_tx_time || - msr->stats.sample_skipped >= 20) { - if (!msr->sample_limit) - return; - - mi->sample_packets++; - if (msr->sample_limit > 0) - msr->sample_limit--; - } - - /* If we're not using MRR and the sampling rate already - * has a probability of >95%, we shouldn't be attempting - * to use it, as this only wastes precious airtime */ - if (!mrr_capable && - (mi->r[ndx].stats.prob_avg > MINSTREL_FRAC(95, 100))) - return; - - mi->prev_sample = true; - - rate->idx = mi->r[ndx].rix; - rate->count = minstrel_get_retry_count(&mi->r[ndx], info); - info->flags |= IEEE80211_TX_CTL_RATE_CTRL_PROBE; -} - - -static void -calc_rate_durations(enum nl80211_band band, - struct minstrel_rate *d, - struct ieee80211_rate *rate, - struct cfg80211_chan_def *chandef) -{ - int erp = !!(rate->flags & IEEE80211_RATE_ERP_G); - int shift = ieee80211_chandef_get_shift(chandef); - - d->perfect_tx_time = ieee80211_frame_duration(band, 1200, - DIV_ROUND_UP(rate->bitrate, 1 << shift), erp, 1, - shift); - d->ack_time = ieee80211_frame_duration(band, 10, - DIV_ROUND_UP(rate->bitrate, 1 << shift), erp, 1, - shift); -} - -static void -init_sample_table(struct minstrel_sta_info *mi) -{ - unsigned int i, col, new_idx; - u8 rnd[8]; - - mi->sample_column = 0; - mi->sample_row = 0; - memset(mi->sample_table, 0xff, SAMPLE_COLUMNS * mi->n_rates); - - for (col = 0; col < SAMPLE_COLUMNS; col++) { - prandom_bytes(rnd, sizeof(rnd)); - for (i = 0; i < mi->n_rates; i++) { - new_idx = (i + rnd[i & 7]) % mi->n_rates; - while (SAMPLE_TBL(mi, new_idx, col) != 0xff) - new_idx = (new_idx + 1) % mi->n_rates; - - SAMPLE_TBL(mi, new_idx, col) = i; - } - } -} - -static void -minstrel_rate_init(void *priv, struct ieee80211_supported_band *sband, - struct cfg80211_chan_def *chandef, - struct ieee80211_sta *sta, void *priv_sta) -{ - struct minstrel_sta_info *mi = priv_sta; - struct minstrel_priv *mp = priv; - struct ieee80211_rate *ctl_rate; - unsigned int i, n = 0; - unsigned int t_slot = 9; /* FIXME: get real slot time */ - u32 rate_flags; - - mi->sta = sta; - mi->lowest_rix = rate_lowest_index(sband, sta); - ctl_rate = &sband->bitrates[mi->lowest_rix]; - mi->sp_ack_dur = ieee80211_frame_duration(sband->band, 10, - ctl_rate->bitrate, - !!(ctl_rate->flags & IEEE80211_RATE_ERP_G), 1, - ieee80211_chandef_get_shift(chandef)); - - rate_flags = ieee80211_chandef_rate_flags(&mp->hw->conf.chandef); - memset(mi->max_tp_rate, 0, sizeof(mi->max_tp_rate)); - mi->max_prob_rate = 0; - - for (i = 0; i < sband->n_bitrates; i++) { - struct minstrel_rate *mr = &mi->r[n]; - struct minstrel_rate_stats *mrs = &mi->r[n].stats; - unsigned int tx_time = 0, tx_time_cts = 0, tx_time_rtscts = 0; - unsigned int tx_time_single; - unsigned int cw = mp->cw_min; - int shift; - - if (!rate_supported(sta, sband->band, i)) - continue; - if ((rate_flags & sband->bitrates[i].flags) != rate_flags) - continue; - - n++; - memset(mr, 0, sizeof(*mr)); - memset(mrs, 0, sizeof(*mrs)); - - mr->rix = i; - shift = ieee80211_chandef_get_shift(chandef); - mr->bitrate = DIV_ROUND_UP(sband->bitrates[i].bitrate, - (1 << shift) * 5); - calc_rate_durations(sband->band, mr, &sband->bitrates[i], - chandef); - - /* calculate maximum number of retransmissions before - * fallback (based on maximum segment size) */ - mr->sample_limit = -1; - mrs->retry_count = 1; - mr->retry_count_cts = 1; - mrs->retry_count_rtscts = 1; - tx_time = mr->perfect_tx_time + mi->sp_ack_dur; - do { - /* add one retransmission */ - tx_time_single = mr->ack_time + mr->perfect_tx_time; - - /* contention window */ - tx_time_single += (t_slot * cw) >> 1; - cw = min((cw << 1) | 1, mp->cw_max); - - tx_time += tx_time_single; - tx_time_cts += tx_time_single + mi->sp_ack_dur; - tx_time_rtscts += tx_time_single + 2 * mi->sp_ack_dur; - if ((tx_time_cts < mp->segment_size) && - (mr->retry_count_cts < mp->max_retry)) - mr->retry_count_cts++; - if ((tx_time_rtscts < mp->segment_size) && - (mrs->retry_count_rtscts < mp->max_retry)) - mrs->retry_count_rtscts++; - } while ((tx_time < mp->segment_size) && - (++mr->stats.retry_count < mp->max_retry)); - mr->adjusted_retry_count = mrs->retry_count; - if (!(sband->bitrates[i].flags & IEEE80211_RATE_ERP_G)) - mr->retry_count_cts = mrs->retry_count; - } - - for (i = n; i < sband->n_bitrates; i++) { - struct minstrel_rate *mr = &mi->r[i]; - mr->rix = -1; - } - - mi->n_rates = n; - mi->last_stats_update = jiffies; - - init_sample_table(mi); - minstrel_update_rates(mp, mi); -} - -static u32 minstrel_get_expected_throughput(void *priv_sta) -{ - struct minstrel_sta_info *mi = priv_sta; - struct minstrel_rate_stats *tmp_mrs; - int idx = mi->max_tp_rate[0]; - int tmp_cur_tp; - - /* convert pkt per sec in kbps (1200 is the average pkt size used for - * computing cur_tp - */ - tmp_mrs = &mi->r[idx].stats; - tmp_cur_tp = minstrel_get_tp_avg(&mi->r[idx], tmp_mrs->prob_avg) * 10; - tmp_cur_tp = tmp_cur_tp * 1200 * 8 / 1024; - - return tmp_cur_tp; -} - -const struct rate_control_ops mac80211_minstrel = { - .tx_status_ext = minstrel_tx_status, - .get_rate = minstrel_get_rate, - .rate_init = minstrel_rate_init, - .get_expected_throughput = minstrel_get_expected_throughput, -}; diff --git a/net/mac80211/rc80211_minstrel.h b/net/mac80211/rc80211_minstrel.h deleted file mode 100644 index 86cd80b3ffde..000000000000 --- a/net/mac80211/rc80211_minstrel.h +++ /dev/null @@ -1,184 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (C) 2008 Felix Fietkau <nbd@openwrt.org> - */ - -#ifndef __RC_MINSTREL_H -#define __RC_MINSTREL_H - -#define EWMA_LEVEL 96 /* ewma weighting factor [/EWMA_DIV] */ -#define EWMA_DIV 128 -#define SAMPLE_COLUMNS 10 /* number of columns in sample table */ - -/* scaled fraction values */ -#define MINSTREL_SCALE 12 -#define MINSTREL_FRAC(val, div) (((val) << MINSTREL_SCALE) / div) -#define MINSTREL_TRUNC(val) ((val) >> MINSTREL_SCALE) - -/* number of highest throughput rates to consider*/ -#define MAX_THR_RATES 4 - -/* - * Coefficients for moving average with noise filter (period=16), - * scaled by 10 bits - * - * a1 = exp(-pi * sqrt(2) / period) - * coeff2 = 2 * a1 * cos(sqrt(2) * 2 * pi / period) - * coeff3 = -sqr(a1) - * coeff1 = 1 - coeff2 - coeff3 - */ -#define MINSTREL_AVG_COEFF1 (MINSTREL_FRAC(1, 1) - \ - MINSTREL_AVG_COEFF2 - \ - MINSTREL_AVG_COEFF3) -#define MINSTREL_AVG_COEFF2 0x00001499 -#define MINSTREL_AVG_COEFF3 -0x0000092e - -/* - * Perform EWMA (Exponentially Weighted Moving Average) calculation - */ -static inline int -minstrel_ewma(int old, int new, int weight) -{ - int diff, incr; - - diff = new - old; - incr = (EWMA_DIV - weight) * diff / EWMA_DIV; - - return old + incr; -} - -static inline int minstrel_filter_avg_add(u16 *prev_1, u16 *prev_2, s32 in) -{ - s32 out_1 = *prev_1; - s32 out_2 = *prev_2; - s32 val; - - if (!in) - in += 1; - - if (!out_1) { - val = out_1 = in; - goto out; - } - - val = MINSTREL_AVG_COEFF1 * in; - val += MINSTREL_AVG_COEFF2 * out_1; - val += MINSTREL_AVG_COEFF3 * out_2; - val >>= MINSTREL_SCALE; - - if (val > 1 << MINSTREL_SCALE) - val = 1 << MINSTREL_SCALE; - if (val < 0) - val = 1; - -out: - *prev_2 = out_1; - *prev_1 = val; - - return val; -} - -struct minstrel_rate_stats { - /* current / last sampling period attempts/success counters */ - u16 attempts, last_attempts; - u16 success, last_success; - - /* total attempts/success counters */ - u32 att_hist, succ_hist; - - /* prob_avg - moving average of prob */ - u16 prob_avg; - u16 prob_avg_1; - - /* maximum retry counts */ - u8 retry_count; - u8 retry_count_rtscts; - - u8 sample_skipped; - bool retry_updated; -}; - -struct minstrel_rate { - int bitrate; - - s8 rix; - u8 retry_count_cts; - u8 adjusted_retry_count; - - unsigned int perfect_tx_time; - unsigned int ack_time; - - int sample_limit; - - struct minstrel_rate_stats stats; -}; - -struct minstrel_sta_info { - struct ieee80211_sta *sta; - - unsigned long last_stats_update; - unsigned int sp_ack_dur; - unsigned int rate_avg; - - unsigned int lowest_rix; - - u8 max_tp_rate[MAX_THR_RATES]; - u8 max_prob_rate; - unsigned int total_packets; - unsigned int sample_packets; - - unsigned int sample_row; - unsigned int sample_column; - - int n_rates; - struct minstrel_rate *r; - bool prev_sample; - - /* sampling table */ - u8 *sample_table; -}; - -struct minstrel_priv { - struct ieee80211_hw *hw; - bool has_mrr; - bool new_avg; - u32 sample_switch; - unsigned int cw_min; - unsigned int cw_max; - unsigned int max_retry; - unsigned int segment_size; - unsigned int update_interval; - unsigned int lookaround_rate; - unsigned int lookaround_rate_mrr; - - u8 cck_rates[4]; - -#ifdef CONFIG_MAC80211_DEBUGFS - /* - * enable fixed rate processing per RC - * - write static index to debugfs:ieee80211/phyX/rc/fixed_rate_idx - * - write -1 to enable RC processing again - * - setting will be applied on next update - */ - u32 fixed_rate_idx; -#endif -}; - -struct minstrel_debugfs_info { - size_t len; - char buf[]; -}; - -extern const struct rate_control_ops mac80211_minstrel; -void minstrel_add_sta_debugfs(void *priv, void *priv_sta, struct dentry *dir); - -/* Recalculate success probabilities and counters for a given rate using EWMA */ -void minstrel_calc_rate_stats(struct minstrel_priv *mp, - struct minstrel_rate_stats *mrs); -int minstrel_get_tp_avg(struct minstrel_rate *mr, int prob_avg); - -/* debugfs */ -int minstrel_stats_open(struct inode *inode, struct file *file); -int minstrel_stats_csv_open(struct inode *inode, struct file *file); - -#endif diff --git a/net/mac80211/rc80211_minstrel_debugfs.c b/net/mac80211/rc80211_minstrel_debugfs.c deleted file mode 100644 index 9b8e0daeb7bb..000000000000 --- a/net/mac80211/rc80211_minstrel_debugfs.c +++ /dev/null @@ -1,172 +0,0 @@ -/* - * Copyright (C) 2008 Felix Fietkau <nbd@openwrt.org> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * Based on minstrel.c: - * Copyright (C) 2005-2007 Derek Smithies <derek@indranet.co.nz> - * Sponsored by Indranet Technologies Ltd - * - * Based on sample.c: - * Copyright (c) 2005 John Bicket - * 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, - * without modification. - * 2. Redistributions in binary form must reproduce at minimum a disclaimer - * similar to the "NO WARRANTY" disclaimer below ("Disclaimer") and any - * redistribution must be conditioned upon including a substantially - * similar Disclaimer requirement for further binary redistribution. - * 3. Neither the names of the above-listed copyright holders nor the names - * of any contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * Alternatively, this software may be distributed under the terms of the - * GNU General Public License ("GPL") version 2 as published by the Free - * Software Foundation. - * - * NO WARRANTY - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTIBILITY - * AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL - * THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR 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 DAMAGES. - */ -#include <linux/netdevice.h> -#include <linux/types.h> -#include <linux/skbuff.h> -#include <linux/debugfs.h> -#include <linux/ieee80211.h> -#include <linux/slab.h> -#include <linux/export.h> -#include <net/mac80211.h> -#include "rc80211_minstrel.h" - -int -minstrel_stats_open(struct inode *inode, struct file *file) -{ - struct minstrel_sta_info *mi = inode->i_private; - struct minstrel_debugfs_info *ms; - unsigned int i, tp_max, tp_avg, eprob; - char *p; - - ms = kmalloc(2048, GFP_KERNEL); - if (!ms) - return -ENOMEM; - - file->private_data = ms; - p = ms->buf; - p += sprintf(p, "\n"); - p += sprintf(p, - "best __________rate_________ ____statistics___ ____last_____ ______sum-of________\n"); - p += sprintf(p, - "rate [name idx airtime max_tp] [avg(tp) avg(prob)] [retry|suc|att] [#success | #attempts]\n"); - - for (i = 0; i < mi->n_rates; i++) { - struct minstrel_rate *mr = &mi->r[i]; - struct minstrel_rate_stats *mrs = &mi->r[i].stats; - - *(p++) = (i == mi->max_tp_rate[0]) ? 'A' : ' '; - *(p++) = (i == mi->max_tp_rate[1]) ? 'B' : ' '; - *(p++) = (i == mi->max_tp_rate[2]) ? 'C' : ' '; - *(p++) = (i == mi->max_tp_rate[3]) ? 'D' : ' '; - *(p++) = (i == mi->max_prob_rate) ? 'P' : ' '; - - p += sprintf(p, " %3u%s ", mr->bitrate / 2, - (mr->bitrate & 1 ? ".5" : " ")); - p += sprintf(p, "%3u ", i); - p += sprintf(p, "%6u ", mr->perfect_tx_time); - - tp_max = minstrel_get_tp_avg(mr, MINSTREL_FRAC(100,100)); - tp_avg = minstrel_get_tp_avg(mr, mrs->prob_avg); - eprob = MINSTREL_TRUNC(mrs->prob_avg * 1000); - - p += sprintf(p, "%4u.%1u %4u.%1u %3u.%1u" - " %3u %3u %-3u " - "%9llu %-9llu\n", - tp_max / 10, tp_max % 10, - tp_avg / 10, tp_avg % 10, - eprob / 10, eprob % 10, - mrs->retry_count, - mrs->last_success, - mrs->last_attempts, - (unsigned long long)mrs->succ_hist, - (unsigned long long)mrs->att_hist); - } - p += sprintf(p, "\nTotal packet count:: ideal %d " - "lookaround %d\n\n", - mi->total_packets - mi->sample_packets, - mi->sample_packets); - ms->len = p - ms->buf; - - WARN_ON(ms->len + sizeof(*ms) > 2048); - - return 0; -} - -int -minstrel_stats_csv_open(struct inode *inode, struct file *file) -{ - struct minstrel_sta_info *mi = inode->i_private; - struct minstrel_debugfs_info *ms; - unsigned int i, tp_max, tp_avg, eprob; - char *p; - - ms = kmalloc(2048, GFP_KERNEL); - if (!ms) - return -ENOMEM; - - file->private_data = ms; - p = ms->buf; - - for (i = 0; i < mi->n_rates; i++) { - struct minstrel_rate *mr = &mi->r[i]; - struct minstrel_rate_stats *mrs = &mi->r[i].stats; - - p += sprintf(p, "%s" ,((i == mi->max_tp_rate[0]) ? "A" : "")); - p += sprintf(p, "%s" ,((i == mi->max_tp_rate[1]) ? "B" : "")); - p += sprintf(p, "%s" ,((i == mi->max_tp_rate[2]) ? "C" : "")); - p += sprintf(p, "%s" ,((i == mi->max_tp_rate[3]) ? "D" : "")); - p += sprintf(p, "%s" ,((i == mi->max_prob_rate) ? "P" : "")); - - p += sprintf(p, ",%u%s", mr->bitrate / 2, - (mr->bitrate & 1 ? ".5," : ",")); - p += sprintf(p, "%u,", i); - p += sprintf(p, "%u,",mr->perfect_tx_time); - - tp_max = minstrel_get_tp_avg(mr, MINSTREL_FRAC(100,100)); - tp_avg = minstrel_get_tp_avg(mr, mrs->prob_avg); - eprob = MINSTREL_TRUNC(mrs->prob_avg * 1000); - - p += sprintf(p, "%u.%u,%u.%u,%u.%u,%u,%u,%u," - "%llu,%llu,%d,%d\n", - tp_max / 10, tp_max % 10, - tp_avg / 10, tp_avg % 10, - eprob / 10, eprob % 10, - mrs->retry_count, - mrs->last_success, - mrs->last_attempts, - (unsigned long long)mrs->succ_hist, - (unsigned long long)mrs->att_hist, - mi->total_packets - mi->sample_packets, - mi->sample_packets); - - } - ms->len = p - ms->buf; - - WARN_ON(ms->len + sizeof(*ms) > 2048); - - return 0; -} diff --git a/net/mac80211/rc80211_minstrel_ht.c b/net/mac80211/rc80211_minstrel_ht.c index b11a2af55b06..bfa550068de4 100644 --- a/net/mac80211/rc80211_minstrel_ht.c +++ b/net/mac80211/rc80211_minstrel_ht.c @@ -13,7 +13,6 @@ #include <net/mac80211.h> #include "rate.h" #include "sta_info.h" -#include "rc80211_minstrel.h" #include "rc80211_minstrel_ht.h" #define AVG_AMPDU_SIZE 16 @@ -136,20 +135,16 @@ __VHT_GROUP(_streams, _sgi, _bw, \ VHT_GROUP_SHIFT(_streams, _sgi, _bw)) -#define CCK_DURATION(_bitrate, _short, _len) \ +#define CCK_DURATION(_bitrate, _short) \ (1000 * (10 /* SIFS */ + \ (_short ? 72 + 24 : 144 + 48) + \ - (8 * (_len + 4) * 10) / (_bitrate))) - -#define CCK_ACK_DURATION(_bitrate, _short) \ - (CCK_DURATION((_bitrate > 10 ? 20 : 10), false, 60) + \ - CCK_DURATION(_bitrate, _short, AVG_PKT_SIZE)) + (8 * (AVG_PKT_SIZE + 4) * 10) / (_bitrate))) #define CCK_DURATION_LIST(_short, _s) \ - CCK_ACK_DURATION(10, _short) >> _s, \ - CCK_ACK_DURATION(20, _short) >> _s, \ - CCK_ACK_DURATION(55, _short) >> _s, \ - CCK_ACK_DURATION(110, _short) >> _s + CCK_DURATION(10, _short) >> _s, \ + CCK_DURATION(20, _short) >> _s, \ + CCK_DURATION(55, _short) >> _s, \ + CCK_DURATION(110, _short) >> _s #define __CCK_GROUP(_s) \ [MINSTREL_CCK_GROUP] = { \ @@ -163,10 +158,42 @@ } #define CCK_GROUP_SHIFT \ - GROUP_SHIFT(CCK_ACK_DURATION(10, false)) + GROUP_SHIFT(CCK_DURATION(10, false)) #define CCK_GROUP __CCK_GROUP(CCK_GROUP_SHIFT) +#define OFDM_DURATION(_bitrate) \ + (1000 * (16 /* SIFS + signal ext */ + \ + 16 /* T_PREAMBLE */ + \ + 4 /* T_SIGNAL */ + \ + 4 * (((16 + 80 * (AVG_PKT_SIZE + 4) + 6) / \ + ((_bitrate) * 4))))) + +#define OFDM_DURATION_LIST(_s) \ + OFDM_DURATION(60) >> _s, \ + OFDM_DURATION(90) >> _s, \ + OFDM_DURATION(120) >> _s, \ + OFDM_DURATION(180) >> _s, \ + OFDM_DURATION(240) >> _s, \ + OFDM_DURATION(360) >> _s, \ + OFDM_DURATION(480) >> _s, \ + OFDM_DURATION(540) >> _s + +#define __OFDM_GROUP(_s) \ + [MINSTREL_OFDM_GROUP] = { \ + .streams = 1, \ + .flags = 0, \ + .shift = _s, \ + .duration = { \ + OFDM_DURATION_LIST(_s), \ + } \ + } + +#define OFDM_GROUP_SHIFT \ + GROUP_SHIFT(OFDM_DURATION(60)) + +#define OFDM_GROUP __OFDM_GROUP(OFDM_GROUP_SHIFT) + static bool minstrel_vht_only = true; module_param(minstrel_vht_only, bool, 0644); @@ -203,6 +230,7 @@ const struct mcs_group minstrel_mcs_groups[] = { MCS_GROUP(4, 1, BW_40), CCK_GROUP, + OFDM_GROUP, VHT_GROUP(1, 0, BW_20), VHT_GROUP(2, 0, BW_20), @@ -235,6 +263,8 @@ const struct mcs_group minstrel_mcs_groups[] = { VHT_GROUP(4, 1, BW_80), }; +const s16 minstrel_cck_bitrates[4] = { 10, 20, 55, 110 }; +const s16 minstrel_ofdm_bitrates[8] = { 60, 90, 120, 180, 240, 360, 480, 540 }; static u8 sample_table[SAMPLE_COLUMNS][MCS_GROUP_RATES] __read_mostly; static void @@ -279,6 +309,13 @@ minstrel_get_valid_vht_rates(int bw, int nss, __le16 mcs_map) return 0x3ff & ~mask; } +static bool +minstrel_ht_is_legacy_group(int group) +{ + return group == MINSTREL_CCK_GROUP || + group == MINSTREL_OFDM_GROUP; +} + /* * Look up an MCS group index based on mac80211 rate information */ @@ -308,21 +345,34 @@ minstrel_ht_get_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, if (rate->flags & IEEE80211_TX_RC_MCS) { group = minstrel_ht_get_group_idx(rate); idx = rate->idx % 8; - } else if (rate->flags & IEEE80211_TX_RC_VHT_MCS) { + goto out; + } + + if (rate->flags & IEEE80211_TX_RC_VHT_MCS) { group = minstrel_vht_get_group_idx(rate); idx = ieee80211_rate_get_vht_mcs(rate); - } else { - group = MINSTREL_CCK_GROUP; + goto out; + } - for (idx = 0; idx < ARRAY_SIZE(mp->cck_rates); idx++) - if (rate->idx == mp->cck_rates[idx]) - break; + group = MINSTREL_CCK_GROUP; + for (idx = 0; idx < ARRAY_SIZE(mp->cck_rates); idx++) { + if (rate->idx != mp->cck_rates[idx]) + continue; /* short preamble */ if ((mi->supported[group] & BIT(idx + 4)) && (rate->flags & IEEE80211_TX_RC_USE_SHORT_PREAMBLE)) - idx += 4; + idx += 4; + goto out; } + + group = MINSTREL_OFDM_GROUP; + for (idx = 0; idx < ARRAY_SIZE(mp->ofdm_rates[0]); idx++) + if (rate->idx == mp->ofdm_rates[mi->band][idx]) + goto out; + + idx = 0; +out: return &mi->groups[group].rates[idx]; } @@ -332,13 +382,37 @@ minstrel_get_ratestats(struct minstrel_ht_sta *mi, int index) return &mi->groups[index / MCS_GROUP_RATES].rates[index % MCS_GROUP_RATES]; } +static inline int minstrel_get_duration(int index) +{ + const struct mcs_group *group = &minstrel_mcs_groups[index / MCS_GROUP_RATES]; + unsigned int duration = group->duration[index % MCS_GROUP_RATES]; + + return duration << group->shift; +} + static unsigned int minstrel_ht_avg_ampdu_len(struct minstrel_ht_sta *mi) { - if (!mi->avg_ampdu_len) - return AVG_AMPDU_SIZE; + int duration; - return MINSTREL_TRUNC(mi->avg_ampdu_len); + if (mi->avg_ampdu_len) + return MINSTREL_TRUNC(mi->avg_ampdu_len); + + if (minstrel_ht_is_legacy_group(mi->max_tp_rate[0] / MCS_GROUP_RATES)) + return 1; + + duration = minstrel_get_duration(mi->max_tp_rate[0]); + + if (duration > 400 * 1000) + return 2; + + if (duration > 250 * 1000) + return 4; + + if (duration > 150 * 1000) + return 8; + + return 16; } /* @@ -349,15 +423,19 @@ int minstrel_ht_get_tp_avg(struct minstrel_ht_sta *mi, int group, int rate, int prob_avg) { - unsigned int nsecs = 0; + unsigned int nsecs = 0, overhead = mi->overhead; + unsigned int ampdu_len = 1; /* do not account throughput if sucess prob is below 10% */ if (prob_avg < MINSTREL_FRAC(10, 100)) return 0; - if (group != MINSTREL_CCK_GROUP) - nsecs = 1000 * mi->overhead / minstrel_ht_avg_ampdu_len(mi); + if (minstrel_ht_is_legacy_group(group)) + overhead = mi->overhead_legacy; + else + ampdu_len = minstrel_ht_avg_ampdu_len(mi); + nsecs = 1000 * overhead / ampdu_len; nsecs += minstrel_mcs_groups[group].duration[rate] << minstrel_mcs_groups[group].shift; @@ -367,10 +445,9 @@ minstrel_ht_get_tp_avg(struct minstrel_ht_sta *mi, int group, int rate, * (prob is scaled - see MINSTREL_FRAC above) */ if (prob_avg > MINSTREL_FRAC(90, 100)) - return MINSTREL_TRUNC(100000 * ((MINSTREL_FRAC(90, 100) * 1000) - / nsecs)); - else - return MINSTREL_TRUNC(100000 * ((prob_avg * 1000) / nsecs)); + prob_avg = MINSTREL_FRAC(90, 100); + + return MINSTREL_TRUNC(100 * ((prob_avg * 1000000) / nsecs)); } /* @@ -417,12 +494,13 @@ minstrel_ht_sort_best_tp_rates(struct minstrel_ht_sta *mi, u16 index, * Find and set the topmost probability rate per sta and per group */ static void -minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 index) +minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 *dest, u16 index) { struct minstrel_mcs_group_data *mg; struct minstrel_rate_stats *mrs; int tmp_group, tmp_idx, tmp_tp_avg, tmp_prob; - int max_tp_group, cur_tp_avg, cur_group, cur_idx; + int max_tp_group, max_tp_idx, max_tp_prob; + int cur_tp_avg, cur_group, cur_idx; int max_gpr_group, max_gpr_idx; int max_gpr_tp_avg, max_gpr_prob; @@ -431,16 +509,24 @@ minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 index) mg = &mi->groups[index / MCS_GROUP_RATES]; mrs = &mg->rates[index % MCS_GROUP_RATES]; - tmp_group = mi->max_prob_rate / MCS_GROUP_RATES; - tmp_idx = mi->max_prob_rate % MCS_GROUP_RATES; + tmp_group = *dest / MCS_GROUP_RATES; + tmp_idx = *dest % MCS_GROUP_RATES; tmp_prob = mi->groups[tmp_group].rates[tmp_idx].prob_avg; tmp_tp_avg = minstrel_ht_get_tp_avg(mi, tmp_group, tmp_idx, tmp_prob); /* if max_tp_rate[0] is from MCS_GROUP max_prob_rate get selected from * MCS_GROUP as well as CCK_GROUP rates do not allow aggregation */ max_tp_group = mi->max_tp_rate[0] / MCS_GROUP_RATES; - if((index / MCS_GROUP_RATES == MINSTREL_CCK_GROUP) && - (max_tp_group != MINSTREL_CCK_GROUP)) + max_tp_idx = mi->max_tp_rate[0] % MCS_GROUP_RATES; + max_tp_prob = mi->groups[max_tp_group].rates[max_tp_idx].prob_avg; + + if (minstrel_ht_is_legacy_group(index / MCS_GROUP_RATES) && + !minstrel_ht_is_legacy_group(max_tp_group)) + return; + + /* skip rates faster than max tp rate with lower prob */ + if (minstrel_get_duration(mi->max_tp_rate[0]) > minstrel_get_duration(index) && + mrs->prob_avg < max_tp_prob) return; max_gpr_group = mg->max_group_prob_rate / MCS_GROUP_RATES; @@ -451,7 +537,7 @@ minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 index) cur_tp_avg = minstrel_ht_get_tp_avg(mi, cur_group, cur_idx, mrs->prob_avg); if (cur_tp_avg > tmp_tp_avg) - mi->max_prob_rate = index; + *dest = index; max_gpr_tp_avg = minstrel_ht_get_tp_avg(mi, max_gpr_group, max_gpr_idx, @@ -460,7 +546,7 @@ minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 index) mg->max_group_prob_rate = index; } else { if (mrs->prob_avg > tmp_prob) - mi->max_prob_rate = index; + *dest = index; if (mrs->prob_avg > max_gpr_prob) mg->max_group_prob_rate = index; } @@ -476,13 +562,13 @@ minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 index) static void minstrel_ht_assign_best_tp_rates(struct minstrel_ht_sta *mi, u16 tmp_mcs_tp_rate[MAX_THR_RATES], - u16 tmp_cck_tp_rate[MAX_THR_RATES]) + u16 tmp_legacy_tp_rate[MAX_THR_RATES]) { unsigned int tmp_group, tmp_idx, tmp_cck_tp, tmp_mcs_tp, tmp_prob; int i; - tmp_group = tmp_cck_tp_rate[0] / MCS_GROUP_RATES; - tmp_idx = tmp_cck_tp_rate[0] % MCS_GROUP_RATES; + tmp_group = tmp_legacy_tp_rate[0] / MCS_GROUP_RATES; + tmp_idx = tmp_legacy_tp_rate[0] % MCS_GROUP_RATES; tmp_prob = mi->groups[tmp_group].rates[tmp_idx].prob_avg; tmp_cck_tp = minstrel_ht_get_tp_avg(mi, tmp_group, tmp_idx, tmp_prob); @@ -493,7 +579,7 @@ minstrel_ht_assign_best_tp_rates(struct minstrel_ht_sta *mi, if (tmp_cck_tp > tmp_mcs_tp) { for(i = 0; i < MAX_THR_RATES; i++) { - minstrel_ht_sort_best_tp_rates(mi, tmp_cck_tp_rate[i], + minstrel_ht_sort_best_tp_rates(mi, tmp_legacy_tp_rate[i], tmp_mcs_tp_rate); } } @@ -511,6 +597,9 @@ minstrel_ht_prob_rate_reduce_streams(struct minstrel_ht_sta *mi) int tmp_max_streams, group, tmp_idx, tmp_prob; int tmp_tp = 0; + if (!mi->sta->ht_cap.ht_supported) + return; + tmp_max_streams = minstrel_mcs_groups[mi->max_tp_rate[0] / MCS_GROUP_RATES].streams; for (group = 0; group < ARRAY_SIZE(minstrel_mcs_groups); group++) { @@ -531,14 +620,6 @@ minstrel_ht_prob_rate_reduce_streams(struct minstrel_ht_sta *mi) } } -static inline int -minstrel_get_duration(int index) -{ - const struct mcs_group *group = &minstrel_mcs_groups[index / MCS_GROUP_RATES]; - unsigned int duration = group->duration[index % MCS_GROUP_RATES]; - return duration << group->shift; -} - static bool minstrel_ht_probe_group(struct minstrel_ht_sta *mi, const struct mcs_group *tp_group, int tp_idx, const struct mcs_group *group) @@ -658,6 +739,74 @@ out: mi->sample_mode = MINSTREL_SAMPLE_ACTIVE; } +static inline int +minstrel_ewma(int old, int new, int weight) +{ + int diff, incr; + + diff = new - old; + incr = (EWMA_DIV - weight) * diff / EWMA_DIV; + + return old + incr; +} + +static inline int minstrel_filter_avg_add(u16 *prev_1, u16 *prev_2, s32 in) +{ + s32 out_1 = *prev_1; + s32 out_2 = *prev_2; + s32 val; + + if (!in) + in += 1; + + if (!out_1) { + val = out_1 = in; + goto out; + } + + val = MINSTREL_AVG_COEFF1 * in; + val += MINSTREL_AVG_COEFF2 * out_1; + val += MINSTREL_AVG_COEFF3 * out_2; + val >>= MINSTREL_SCALE; + + if (val > 1 << MINSTREL_SCALE) + val = 1 << MINSTREL_SCALE; + if (val < 0) + val = 1; + +out: + *prev_2 = out_1; + *prev_1 = val; + + return val; +} + +/* +* Recalculate statistics and counters of a given rate +*/ +static void +minstrel_ht_calc_rate_stats(struct minstrel_priv *mp, + struct minstrel_rate_stats *mrs) +{ + unsigned int cur_prob; + + if (unlikely(mrs->attempts > 0)) { + mrs->sample_skipped = 0; + cur_prob = MINSTREL_FRAC(mrs->success, mrs->attempts); + minstrel_filter_avg_add(&mrs->prob_avg, + &mrs->prob_avg_1, cur_prob); + mrs->att_hist += mrs->attempts; + mrs->succ_hist += mrs->success; + } else { + mrs->sample_skipped++; + } + + mrs->last_success = mrs->success; + mrs->last_attempts = mrs->attempts; + mrs->success = 0; + mrs->attempts = 0; +} + /* * Update rate statistics and select new primary rates * @@ -675,7 +824,9 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, struct minstrel_rate_stats *mrs; int group, i, j, cur_prob; u16 tmp_mcs_tp_rate[MAX_THR_RATES], tmp_group_tp_rate[MAX_THR_RATES]; - u16 tmp_cck_tp_rate[MAX_THR_RATES], index; + u16 tmp_legacy_tp_rate[MAX_THR_RATES], tmp_max_prob_rate; + u16 index; + bool ht_supported = mi->sta->ht_cap.ht_supported; mi->sample_mode = MINSTREL_SAMPLE_IDLE; @@ -704,21 +855,30 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, mi->sample_count = 0; memset(tmp_mcs_tp_rate, 0, sizeof(tmp_mcs_tp_rate)); - memset(tmp_cck_tp_rate, 0, sizeof(tmp_cck_tp_rate)); + memset(tmp_legacy_tp_rate, 0, sizeof(tmp_legacy_tp_rate)); if (mi->supported[MINSTREL_CCK_GROUP]) - for (j = 0; j < ARRAY_SIZE(tmp_cck_tp_rate); j++) - tmp_cck_tp_rate[j] = MINSTREL_CCK_GROUP * MCS_GROUP_RATES; + for (j = 0; j < ARRAY_SIZE(tmp_legacy_tp_rate); j++) + tmp_legacy_tp_rate[j] = MINSTREL_CCK_GROUP * MCS_GROUP_RATES; + else if (mi->supported[MINSTREL_OFDM_GROUP]) + for (j = 0; j < ARRAY_SIZE(tmp_legacy_tp_rate); j++) + tmp_legacy_tp_rate[j] = MINSTREL_OFDM_GROUP * MCS_GROUP_RATES; if (mi->supported[MINSTREL_VHT_GROUP_0]) index = MINSTREL_VHT_GROUP_0 * MCS_GROUP_RATES; - else + else if (ht_supported) index = MINSTREL_HT_GROUP_0 * MCS_GROUP_RATES; + else if (mi->supported[MINSTREL_CCK_GROUP]) + index = MINSTREL_CCK_GROUP * MCS_GROUP_RATES; + else + index = MINSTREL_OFDM_GROUP * MCS_GROUP_RATES; + tmp_max_prob_rate = index; for (j = 0; j < ARRAY_SIZE(tmp_mcs_tp_rate); j++) tmp_mcs_tp_rate[j] = index; /* Find best rate sets within all MCS groups*/ for (group = 0; group < ARRAY_SIZE(minstrel_mcs_groups); group++) { + u16 *tp_rate = tmp_mcs_tp_rate; mg = &mi->groups[group]; if (!mi->supported[group]) @@ -730,6 +890,9 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, for(j = 0; j < MAX_THR_RATES; j++) tmp_group_tp_rate[j] = MCS_GROUP_RATES * group; + if (group == MINSTREL_CCK_GROUP && ht_supported) + tp_rate = tmp_legacy_tp_rate; + for (i = 0; i < MCS_GROUP_RATES; i++) { if (!(mi->supported[group] & BIT(i))) continue; @@ -738,27 +901,18 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, mrs = &mg->rates[i]; mrs->retry_updated = false; - minstrel_calc_rate_stats(mp, mrs); + minstrel_ht_calc_rate_stats(mp, mrs); cur_prob = mrs->prob_avg; if (minstrel_ht_get_tp_avg(mi, group, i, cur_prob) == 0) continue; /* Find max throughput rate set */ - if (group != MINSTREL_CCK_GROUP) { - minstrel_ht_sort_best_tp_rates(mi, index, - tmp_mcs_tp_rate); - } else if (group == MINSTREL_CCK_GROUP) { - minstrel_ht_sort_best_tp_rates(mi, index, - tmp_cck_tp_rate); - } + minstrel_ht_sort_best_tp_rates(mi, index, tp_rate); /* Find max throughput rate set within a group */ minstrel_ht_sort_best_tp_rates(mi, index, tmp_group_tp_rate); - - /* Find max probability rate per group and global */ - minstrel_ht_set_best_prob_rate(mi, index); } memcpy(mg->max_group_tp_rate, tmp_group_tp_rate, @@ -766,16 +920,36 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, } /* Assign new rate set per sta */ - minstrel_ht_assign_best_tp_rates(mi, tmp_mcs_tp_rate, tmp_cck_tp_rate); + minstrel_ht_assign_best_tp_rates(mi, tmp_mcs_tp_rate, + tmp_legacy_tp_rate); memcpy(mi->max_tp_rate, tmp_mcs_tp_rate, sizeof(mi->max_tp_rate)); + for (group = 0; group < ARRAY_SIZE(minstrel_mcs_groups); group++) { + if (!mi->supported[group]) + continue; + + mg = &mi->groups[group]; + mg->max_group_prob_rate = MCS_GROUP_RATES * group; + + for (i = 0; i < MCS_GROUP_RATES; i++) { + if (!(mi->supported[group] & BIT(i))) + continue; + + index = MCS_GROUP_RATES * group + i; + + /* Find max probability rate per group and global */ + minstrel_ht_set_best_prob_rate(mi, &tmp_max_prob_rate, + index); + } + } + + mi->max_prob_rate = tmp_max_prob_rate; + /* Try to increase robustness of max_prob_rate*/ minstrel_ht_prob_rate_reduce_streams(mi); - /* try to sample all available rates during each interval */ - mi->sample_count *= 8; - if (mp->new_avg) - mi->sample_count /= 2; + /* try to sample half of all available rates during each interval */ + mi->sample_count *= 4; if (sample) minstrel_ht_rate_sample_switch(mp, mi); @@ -795,8 +969,11 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, } static bool -minstrel_ht_txstat_valid(struct minstrel_priv *mp, struct ieee80211_tx_rate *rate) +minstrel_ht_txstat_valid(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, + struct ieee80211_tx_rate *rate) { + int i; + if (rate->idx < 0) return false; @@ -807,10 +984,15 @@ minstrel_ht_txstat_valid(struct minstrel_priv *mp, struct ieee80211_tx_rate *rat rate->flags & IEEE80211_TX_RC_VHT_MCS) return true; - return rate->idx == mp->cck_rates[0] || - rate->idx == mp->cck_rates[1] || - rate->idx == mp->cck_rates[2] || - rate->idx == mp->cck_rates[3]; + for (i = 0; i < ARRAY_SIZE(mp->cck_rates); i++) + if (rate->idx == mp->cck_rates[i]) + return true; + + for (i = 0; i < ARRAY_SIZE(mp->ofdm_rates[0]); i++) + if (rate->idx == mp->ofdm_rates[mi->band][i]) + return true; + + return false; } static void @@ -887,21 +1069,15 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, void *priv_sta, struct ieee80211_tx_status *st) { struct ieee80211_tx_info *info = st->info; - struct minstrel_ht_sta_priv *msp = priv_sta; - struct minstrel_ht_sta *mi = &msp->ht; + struct minstrel_ht_sta *mi = priv_sta; struct ieee80211_tx_rate *ar = info->status.rates; struct minstrel_rate_stats *rate, *rate2, *rate_sample = NULL; struct minstrel_priv *mp = priv; - u32 update_interval = mp->update_interval / 2; + u32 update_interval = mp->update_interval; bool last, update = false; bool sample_status = false; int i; - if (!msp->is_ht) - return mac80211_minstrel.tx_status_ext(priv, sband, - &msp->legacy, st); - - /* This packet was aggregated but doesn't carry status info */ if ((info->flags & IEEE80211_TX_CTL_AMPDU) && !(info->flags & IEEE80211_TX_STAT_AMPDU)) @@ -930,10 +1106,10 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, if (mi->sample_mode != MINSTREL_SAMPLE_IDLE) rate_sample = minstrel_get_ratestats(mi, mi->sample_rate); - last = !minstrel_ht_txstat_valid(mp, &ar[0]); + last = !minstrel_ht_txstat_valid(mp, mi, &ar[0]); for (i = 0; !last; i++) { last = (i == IEEE80211_TX_MAX_RATES - 1) || - !minstrel_ht_txstat_valid(mp, &ar[i + 1]); + !minstrel_ht_txstat_valid(mp, mi, &ar[i + 1]); rate = minstrel_ht_get_stats(mp, mi, &ar[i]); if (rate == rate_sample) @@ -947,9 +1123,8 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, switch (mi->sample_mode) { case MINSTREL_SAMPLE_IDLE: - if (mp->new_avg && - (mp->hw->max_rates > 1 || - mi->total_packets_cur < SAMPLE_SWITCH_THR)) + if (mp->hw->max_rates > 1 || + mi->total_packets_cur < SAMPLE_SWITCH_THR) update_interval /= 2; break; @@ -1031,7 +1206,10 @@ minstrel_calc_retransmit(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, ctime += (t_slot * cw) >> 1; cw = min((cw << 1) | 1, mp->cw_max); - if (index / MCS_GROUP_RATES != MINSTREL_CCK_GROUP) { + if (minstrel_ht_is_legacy_group(index / MCS_GROUP_RATES)) { + overhead = mi->overhead_legacy; + overhead_rtscts = mi->overhead_legacy_rtscts; + } else { overhead = mi->overhead; overhead_rtscts = mi->overhead_rtscts; } @@ -1061,7 +1239,8 @@ static void minstrel_ht_set_rate(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, struct ieee80211_sta_rates *ratetbl, int offset, int index) { - const struct mcs_group *group = &minstrel_mcs_groups[index / MCS_GROUP_RATES]; + int group_idx = index / MCS_GROUP_RATES; + const struct mcs_group *group = &minstrel_mcs_groups[group_idx]; struct minstrel_rate_stats *mrs; u8 idx; u16 flags = group->flags; @@ -1080,13 +1259,17 @@ minstrel_ht_set_rate(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, ratetbl->rate[offset].count_rts = mrs->retry_count_rtscts; } - if (index / MCS_GROUP_RATES == MINSTREL_CCK_GROUP) + index %= MCS_GROUP_RATES; + if (group_idx == MINSTREL_CCK_GROUP) idx = mp->cck_rates[index % ARRAY_SIZE(mp->cck_rates)]; + else if (group_idx == MINSTREL_OFDM_GROUP) + idx = mp->ofdm_rates[mi->band][index % + ARRAY_SIZE(mp->ofdm_rates[0])]; else if (flags & IEEE80211_TX_RC_VHT_MCS) idx = ((group->streams - 1) << 4) | - ((index % MCS_GROUP_RATES) & 0xF); + (index & 0xF); else - idx = index % MCS_GROUP_RATES + (group->streams - 1) * 8; + idx = index + (group->streams - 1) * 8; /* enable RTS/CTS if needed: * - if station is in dynamic SMPS (and streams > 1) @@ -1224,13 +1407,13 @@ minstrel_get_sample_rate(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) mrs = &mg->rates[sample_idx]; sample_idx += sample_group * MCS_GROUP_RATES; - /* Set tp_rate1, tp_rate2 to the highest / second highest max_tp_rate */ + tp_rate1 = mi->max_tp_rate[0]; + + /* Set tp_rate2 to the second highest max_tp_rate */ if (minstrel_get_duration(mi->max_tp_rate[0]) > minstrel_get_duration(mi->max_tp_rate[1])) { - tp_rate1 = mi->max_tp_rate[1]; tp_rate2 = mi->max_tp_rate[0]; } else { - tp_rate1 = mi->max_tp_rate[0]; tp_rate2 = mi->max_tp_rate[1]; } @@ -1296,16 +1479,12 @@ minstrel_ht_get_rate(void *priv, struct ieee80211_sta *sta, void *priv_sta, const struct mcs_group *sample_group; struct ieee80211_tx_info *info = IEEE80211_SKB_CB(txrc->skb); struct ieee80211_tx_rate *rate = &info->status.rates[0]; - struct minstrel_ht_sta_priv *msp = priv_sta; - struct minstrel_ht_sta *mi = &msp->ht; + struct minstrel_ht_sta *mi = priv_sta; struct minstrel_priv *mp = priv; int sample_idx; - if (!msp->is_ht) - return mac80211_minstrel.get_rate(priv, sta, &msp->legacy, txrc); - if (!(info->flags & IEEE80211_TX_CTL_AMPDU) && - mi->max_prob_rate / MCS_GROUP_RATES != MINSTREL_CCK_GROUP) + !minstrel_ht_is_legacy_group(mi->max_prob_rate / MCS_GROUP_RATES)) minstrel_aggr_check(sta, txrc->skb); info->flags |= mi->tx_flags; @@ -1346,6 +1525,9 @@ minstrel_ht_get_rate(void *priv, struct ieee80211_sta *sta, void *priv_sta, if (sample_group == &minstrel_mcs_groups[MINSTREL_CCK_GROUP]) { int idx = sample_idx % ARRAY_SIZE(mp->cck_rates); rate->idx = mp->cck_rates[idx]; + } else if (sample_group == &minstrel_mcs_groups[MINSTREL_OFDM_GROUP]) { + int idx = sample_idx % ARRAY_SIZE(mp->ofdm_rates[0]); + rate->idx = mp->ofdm_rates[mi->band][idx]; } else if (sample_group->flags & IEEE80211_TX_RC_VHT_MCS) { ieee80211_rate_set_vht(rate, sample_idx % MCS_GROUP_RATES, sample_group->streams); @@ -1366,44 +1548,59 @@ minstrel_ht_update_cck(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, if (sband->band != NL80211_BAND_2GHZ) return; - if (!ieee80211_hw_check(mp->hw, SUPPORTS_HT_CCK_RATES)) + if (sta->ht_cap.ht_supported && + !ieee80211_hw_check(mp->hw, SUPPORTS_HT_CCK_RATES)) return; - mi->cck_supported = 0; - mi->cck_supported_short = 0; for (i = 0; i < 4; i++) { - if (!rate_supported(sta, sband->band, mp->cck_rates[i])) + if (mp->cck_rates[i] == 0xff || + !rate_supported(sta, sband->band, mp->cck_rates[i])) continue; - mi->cck_supported |= BIT(i); + mi->supported[MINSTREL_CCK_GROUP] |= BIT(i); if (sband->bitrates[i].flags & IEEE80211_RATE_SHORT_PREAMBLE) - mi->cck_supported_short |= BIT(i); + mi->supported[MINSTREL_CCK_GROUP] |= BIT(i + 4); } +} + +static void +minstrel_ht_update_ofdm(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, + struct ieee80211_supported_band *sband, + struct ieee80211_sta *sta) +{ + const u8 *rates; + int i; - mi->supported[MINSTREL_CCK_GROUP] = mi->cck_supported; + if (sta->ht_cap.ht_supported) + return; + + rates = mp->ofdm_rates[sband->band]; + for (i = 0; i < ARRAY_SIZE(mp->ofdm_rates[0]); i++) { + if (rates[i] == 0xff || + !rate_supported(sta, sband->band, rates[i])) + continue; + + mi->supported[MINSTREL_OFDM_GROUP] |= BIT(i); + } } static void minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, struct cfg80211_chan_def *chandef, - struct ieee80211_sta *sta, void *priv_sta) + struct ieee80211_sta *sta, void *priv_sta) { struct minstrel_priv *mp = priv; - struct minstrel_ht_sta_priv *msp = priv_sta; - struct minstrel_ht_sta *mi = &msp->ht; + struct minstrel_ht_sta *mi = priv_sta; struct ieee80211_mcs_info *mcs = &sta->ht_cap.mcs; u16 ht_cap = sta->ht_cap.cap; struct ieee80211_sta_vht_cap *vht_cap = &sta->vht_cap; + const struct ieee80211_rate *ctl_rate; + bool ldpc, erp; int use_vht; int n_supported = 0; int ack_dur; int stbc; int i; - bool ldpc; - - /* fall back to the old minstrel for legacy stations */ - if (!sta->ht_cap.ht_supported) - goto use_legacy; BUILD_BUG_ON(ARRAY_SIZE(minstrel_mcs_groups) != MINSTREL_GROUPS_NB); @@ -1412,10 +1609,10 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, else use_vht = 0; - msp->is_ht = true; memset(mi, 0, sizeof(*mi)); mi->sta = sta; + mi->band = sband->band; mi->last_stats_update = jiffies; ack_dur = ieee80211_frame_duration(sband->band, 10, 60, 1, 1, 0); @@ -1423,6 +1620,14 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, mi->overhead += ack_dur; mi->overhead_rtscts = mi->overhead + 2 * ack_dur; + ctl_rate = &sband->bitrates[rate_lowest_index(sband, sta)]; + erp = ctl_rate->flags & IEEE80211_RATE_ERP_G; + ack_dur = ieee80211_frame_duration(sband->band, 10, + ctl_rate->bitrate, erp, 1, + ieee80211_chandef_get_shift(chandef)); + mi->overhead_legacy = ack_dur; + mi->overhead_legacy_rtscts = mi->overhead_legacy + 2 * ack_dur; + mi->avg_ampdu_len = MINSTREL_FRAC(1, 1); /* When using MRR, sample more on the first attempt, without delay */ @@ -1456,10 +1661,8 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, int bw, nss; mi->supported[i] = 0; - if (i == MINSTREL_CCK_GROUP) { - minstrel_ht_update_cck(mp, mi, sband, sta); + if (minstrel_ht_is_legacy_group(i)) continue; - } if (gflags & IEEE80211_TX_RC_SHORT_GI) { if (gflags & IEEE80211_TX_RC_40_MHZ_WIDTH) { @@ -1520,24 +1723,12 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, n_supported++; } - if (!n_supported) - goto use_legacy; - - mi->supported[MINSTREL_CCK_GROUP] |= mi->cck_supported_short << 4; + minstrel_ht_update_cck(mp, mi, sband, sta); + minstrel_ht_update_ofdm(mp, mi, sband, sta); /* create an initial rate table with the lowest supported rates */ minstrel_ht_update_stats(mp, mi, true); minstrel_ht_update_rates(mp, mi); - - return; - -use_legacy: - msp->is_ht = false; - memset(&msp->legacy, 0, sizeof(msp->legacy)); - msp->legacy.r = msp->ratelist; - msp->legacy.sample_table = msp->sample_table; - return mac80211_minstrel.rate_init(priv, sband, chandef, sta, - &msp->legacy); } static void @@ -1561,7 +1752,7 @@ static void * minstrel_ht_alloc_sta(void *priv, struct ieee80211_sta *sta, gfp_t gfp) { struct ieee80211_supported_band *sband; - struct minstrel_ht_sta_priv *msp; + struct minstrel_ht_sta *mi; struct minstrel_priv *mp = priv; struct ieee80211_hw *hw = mp->hw; int max_rates = 0; @@ -1573,72 +1764,80 @@ minstrel_ht_alloc_sta(void *priv, struct ieee80211_sta *sta, gfp_t gfp) max_rates = sband->n_bitrates; } - msp = kzalloc(sizeof(*msp), gfp); - if (!msp) - return NULL; - - msp->ratelist = kcalloc(max_rates, sizeof(struct minstrel_rate), gfp); - if (!msp->ratelist) - goto error; - - msp->sample_table = kmalloc_array(max_rates, SAMPLE_COLUMNS, gfp); - if (!msp->sample_table) - goto error1; - - return msp; - -error1: - kfree(msp->ratelist); -error: - kfree(msp); - return NULL; + return kzalloc(sizeof(*mi), gfp); } static void minstrel_ht_free_sta(void *priv, struct ieee80211_sta *sta, void *priv_sta) { - struct minstrel_ht_sta_priv *msp = priv_sta; - - kfree(msp->sample_table); - kfree(msp->ratelist); - kfree(msp); + kfree(priv_sta); } static void -minstrel_ht_init_cck_rates(struct minstrel_priv *mp) +minstrel_ht_fill_rate_array(u8 *dest, struct ieee80211_supported_band *sband, + const s16 *bitrates, int n_rates, u32 rate_flags) { - static const int bitrates[4] = { 10, 20, 55, 110 }; - struct ieee80211_supported_band *sband; - u32 rate_flags = ieee80211_chandef_rate_flags(&mp->hw->conf.chandef); int i, j; - sband = mp->hw->wiphy->bands[NL80211_BAND_2GHZ]; - if (!sband) - return; - for (i = 0; i < sband->n_bitrates; i++) { struct ieee80211_rate *rate = &sband->bitrates[i]; - if (rate->flags & IEEE80211_RATE_ERP_G) - continue; - if ((rate_flags & sband->bitrates[i].flags) != rate_flags) continue; - for (j = 0; j < ARRAY_SIZE(bitrates); j++) { + for (j = 0; j < n_rates; j++) { if (rate->bitrate != bitrates[j]) continue; - mp->cck_rates[j] = i; + dest[j] = i; break; } } } +static void +minstrel_ht_init_cck_rates(struct minstrel_priv *mp) +{ + static const s16 bitrates[4] = { 10, 20, 55, 110 }; + struct ieee80211_supported_band *sband; + u32 rate_flags = ieee80211_chandef_rate_flags(&mp->hw->conf.chandef); + + memset(mp->cck_rates, 0xff, sizeof(mp->cck_rates)); + sband = mp->hw->wiphy->bands[NL80211_BAND_2GHZ]; + if (!sband) + return; + + BUILD_BUG_ON(ARRAY_SIZE(mp->cck_rates) != ARRAY_SIZE(bitrates)); + minstrel_ht_fill_rate_array(mp->cck_rates, sband, + minstrel_cck_bitrates, + ARRAY_SIZE(minstrel_cck_bitrates), + rate_flags); +} + +static void +minstrel_ht_init_ofdm_rates(struct minstrel_priv *mp, enum nl80211_band band) +{ + static const s16 bitrates[8] = { 60, 90, 120, 180, 240, 360, 480, 540 }; + struct ieee80211_supported_band *sband; + u32 rate_flags = ieee80211_chandef_rate_flags(&mp->hw->conf.chandef); + + memset(mp->ofdm_rates[band], 0xff, sizeof(mp->ofdm_rates[band])); + sband = mp->hw->wiphy->bands[band]; + if (!sband) + return; + + BUILD_BUG_ON(ARRAY_SIZE(mp->ofdm_rates[band]) != ARRAY_SIZE(bitrates)); + minstrel_ht_fill_rate_array(mp->ofdm_rates[band], sband, + minstrel_ofdm_bitrates, + ARRAY_SIZE(minstrel_ofdm_bitrates), + rate_flags); +} + static void * minstrel_ht_alloc(struct ieee80211_hw *hw) { struct minstrel_priv *mp; + int i; mp = kzalloc(sizeof(struct minstrel_priv), GFP_ATOMIC); if (!mp) @@ -1652,12 +1851,6 @@ minstrel_ht_alloc(struct ieee80211_hw *hw) mp->cw_min = 15; mp->cw_max = 1023; - /* number of packets (in %) to use for sampling other rates - * sample less often for non-mrr packets, because the overhead - * is much higher than with mrr */ - mp->lookaround_rate = 5; - mp->lookaround_rate_mrr = 10; - /* maximum time that the hw is allowed to stay in one MRR segment */ mp->segment_size = 6000; @@ -1672,9 +1865,10 @@ minstrel_ht_alloc(struct ieee80211_hw *hw) mp->hw = hw; mp->update_interval = HZ / 10; - mp->new_avg = true; minstrel_ht_init_cck_rates(mp); + for (i = 0; i < ARRAY_SIZE(mp->hw->wiphy->bands); i++) + minstrel_ht_init_ofdm_rates(mp, i); return mp; } @@ -1690,8 +1884,6 @@ static void minstrel_ht_add_debugfs(struct ieee80211_hw *hw, void *priv, &mp->fixed_rate_idx); debugfs_create_u32("sample_switch", S_IRUGO | S_IWUSR, debugfsdir, &mp->sample_switch); - debugfs_create_bool("new_avg", S_IRUGO | S_IWUSR, debugfsdir, - &mp->new_avg); } #endif @@ -1703,13 +1895,9 @@ minstrel_ht_free(void *priv) static u32 minstrel_ht_get_expected_throughput(void *priv_sta) { - struct minstrel_ht_sta_priv *msp = priv_sta; - struct minstrel_ht_sta *mi = &msp->ht; + struct minstrel_ht_sta *mi = priv_sta; int i, j, prob, tp_avg; - if (!msp->is_ht) - return mac80211_minstrel.get_expected_throughput(priv_sta); - i = mi->max_tp_rate[0] / MCS_GROUP_RATES; j = mi->max_tp_rate[0] % MCS_GROUP_RATES; prob = mi->groups[i].rates[j].prob_avg; diff --git a/net/mac80211/rc80211_minstrel_ht.h b/net/mac80211/rc80211_minstrel_ht.h index 53ea3c29debf..7d6d0b720f6d 100644 --- a/net/mac80211/rc80211_minstrel_ht.h +++ b/net/mac80211/rc80211_minstrel_ht.h @@ -6,6 +6,33 @@ #ifndef __RC_MINSTREL_HT_H #define __RC_MINSTREL_HT_H +/* number of highest throughput rates to consider*/ +#define MAX_THR_RATES 4 +#define SAMPLE_COLUMNS 10 /* number of columns in sample table */ + +/* scaled fraction values */ +#define MINSTREL_SCALE 12 +#define MINSTREL_FRAC(val, div) (((val) << MINSTREL_SCALE) / div) +#define MINSTREL_TRUNC(val) ((val) >> MINSTREL_SCALE) + +#define EWMA_LEVEL 96 /* ewma weighting factor [/EWMA_DIV] */ +#define EWMA_DIV 128 + +/* + * Coefficients for moving average with noise filter (period=16), + * scaled by 10 bits + * + * a1 = exp(-pi * sqrt(2) / period) + * coeff2 = 2 * a1 * cos(sqrt(2) * 2 * pi / period) + * coeff3 = -sqr(a1) + * coeff1 = 1 - coeff2 - coeff3 + */ +#define MINSTREL_AVG_COEFF1 (MINSTREL_FRAC(1, 1) - \ + MINSTREL_AVG_COEFF2 - \ + MINSTREL_AVG_COEFF3) +#define MINSTREL_AVG_COEFF2 0x00001499 +#define MINSTREL_AVG_COEFF3 -0x0000092e + /* * The number of streams can be changed to 2 to reduce code * size and memory footprint. @@ -18,17 +45,43 @@ MINSTREL_HT_STREAM_GROUPS) #define MINSTREL_VHT_GROUPS_NB (MINSTREL_MAX_STREAMS * \ MINSTREL_VHT_STREAM_GROUPS) -#define MINSTREL_CCK_GROUPS_NB 1 +#define MINSTREL_LEGACY_GROUPS_NB 2 #define MINSTREL_GROUPS_NB (MINSTREL_HT_GROUPS_NB + \ MINSTREL_VHT_GROUPS_NB + \ - MINSTREL_CCK_GROUPS_NB) + MINSTREL_LEGACY_GROUPS_NB) #define MINSTREL_HT_GROUP_0 0 #define MINSTREL_CCK_GROUP (MINSTREL_HT_GROUP_0 + MINSTREL_HT_GROUPS_NB) -#define MINSTREL_VHT_GROUP_0 (MINSTREL_CCK_GROUP + 1) +#define MINSTREL_OFDM_GROUP (MINSTREL_CCK_GROUP + 1) +#define MINSTREL_VHT_GROUP_0 (MINSTREL_OFDM_GROUP + 1) #define MCS_GROUP_RATES 10 +struct minstrel_priv { + struct ieee80211_hw *hw; + bool has_mrr; + u32 sample_switch; + unsigned int cw_min; + unsigned int cw_max; + unsigned int max_retry; + unsigned int segment_size; + unsigned int update_interval; + + u8 cck_rates[4]; + u8 ofdm_rates[NUM_NL80211_BANDS][8]; + +#ifdef CONFIG_MAC80211_DEBUGFS + /* + * enable fixed rate processing per RC + * - write static index to debugfs:ieee80211/phyX/rc/fixed_rate_idx + * - write -1 to enable RC processing again + * - setting will be applied on next update + */ + u32 fixed_rate_idx; +#endif +}; + + struct mcs_group { u16 flags; u8 streams; @@ -37,8 +90,30 @@ struct mcs_group { u16 duration[MCS_GROUP_RATES]; }; +extern const s16 minstrel_cck_bitrates[4]; +extern const s16 minstrel_ofdm_bitrates[8]; extern const struct mcs_group minstrel_mcs_groups[]; +struct minstrel_rate_stats { + /* current / last sampling period attempts/success counters */ + u16 attempts, last_attempts; + u16 success, last_success; + + /* total attempts/success counters */ + u32 att_hist, succ_hist; + + /* prob_avg - moving average of prob */ + u16 prob_avg; + u16 prob_avg_1; + + /* maximum retry counts */ + u8 retry_count; + u8 retry_count_rtscts; + + u8 sample_skipped; + bool retry_updated; +}; + struct minstrel_mcs_group_data { u8 index; u8 column; @@ -77,6 +152,8 @@ struct minstrel_ht_sta { /* overhead time in usec for each frame */ unsigned int overhead; unsigned int overhead_rtscts; + unsigned int overhead_legacy; + unsigned int overhead_legacy_rtscts; unsigned int total_packets_last; unsigned int total_packets_cur; @@ -97,8 +174,7 @@ struct minstrel_ht_sta { /* current MCS group to be sampled */ u8 sample_group; - u8 cck_supported; - u8 cck_supported_short; + u8 band; /* Bitfield of supported MCS rates of all groups */ u16 supported[MINSTREL_GROUPS_NB]; @@ -107,16 +183,6 @@ struct minstrel_ht_sta { struct minstrel_mcs_group_data groups[MINSTREL_GROUPS_NB]; }; -struct minstrel_ht_sta_priv { - union { - struct minstrel_ht_sta ht; - struct minstrel_sta_info legacy; - }; - void *ratelist; - void *sample_table; - bool is_ht; -}; - void minstrel_ht_add_sta_debugfs(void *priv, void *priv_sta, struct dentry *dir); int minstrel_ht_get_tp_avg(struct minstrel_ht_sta *mi, int group, int rate, int prob_avg); diff --git a/net/mac80211/rc80211_minstrel_ht_debugfs.c b/net/mac80211/rc80211_minstrel_ht_debugfs.c index bebb71917742..3b7af242cde6 100644 --- a/net/mac80211/rc80211_minstrel_ht_debugfs.c +++ b/net/mac80211/rc80211_minstrel_ht_debugfs.c @@ -9,9 +9,13 @@ #include <linux/ieee80211.h> #include <linux/export.h> #include <net/mac80211.h> -#include "rc80211_minstrel.h" #include "rc80211_minstrel_ht.h" +struct minstrel_debugfs_info { + size_t len; + char buf[]; +}; + static ssize_t minstrel_stats_read(struct file *file, char __user *buf, size_t len, loff_t *ppos) { @@ -52,7 +56,6 @@ minstrel_ht_stats_dump(struct minstrel_ht_sta *mi, int i, char *p) for (j = 0; j < MCS_GROUP_RATES; j++) { struct minstrel_rate_stats *mrs = &mi->groups[i].rates[j]; - static const int bitrates[4] = { 10, 20, 55, 110 }; int idx = i * MCS_GROUP_RATES + j; unsigned int duration; @@ -67,6 +70,9 @@ minstrel_ht_stats_dump(struct minstrel_ht_sta *mi, int i, char *p) p += sprintf(p, "VHT%c0 ", htmode); p += sprintf(p, "%cGI ", gimode); p += sprintf(p, "%d ", mg->streams); + } else if (i == MINSTREL_OFDM_GROUP) { + p += sprintf(p, "OFDM "); + p += sprintf(p, "1 "); } else { p += sprintf(p, "CCK "); p += sprintf(p, "%cP ", j < 4 ? 'L' : 'S'); @@ -84,7 +90,12 @@ minstrel_ht_stats_dump(struct minstrel_ht_sta *mi, int i, char *p) } else if (gflags & IEEE80211_TX_RC_VHT_MCS) { p += sprintf(p, " MCS%-1u/%1u", j, mg->streams); } else { - int r = bitrates[j % 4]; + int r; + + if (i == MINSTREL_OFDM_GROUP) + r = minstrel_ofdm_bitrates[j % 8]; + else + r = minstrel_cck_bitrates[j % 4]; p += sprintf(p, " %2u.%1uM", r / 10, r % 10); } @@ -120,20 +131,11 @@ minstrel_ht_stats_dump(struct minstrel_ht_sta *mi, int i, char *p) static int minstrel_ht_stats_open(struct inode *inode, struct file *file) { - struct minstrel_ht_sta_priv *msp = inode->i_private; - struct minstrel_ht_sta *mi = &msp->ht; + struct minstrel_ht_sta *mi = inode->i_private; struct minstrel_debugfs_info *ms; unsigned int i; - int ret; char *p; - if (!msp->is_ht) { - inode->i_private = &msp->legacy; - ret = minstrel_stats_open(inode, file); - inode->i_private = msp; - return ret; - } - ms = kmalloc(32768, GFP_KERNEL); if (!ms) return -ENOMEM; @@ -199,7 +201,6 @@ minstrel_ht_stats_csv_dump(struct minstrel_ht_sta *mi, int i, char *p) for (j = 0; j < MCS_GROUP_RATES; j++) { struct minstrel_rate_stats *mrs = &mi->groups[i].rates[j]; - static const int bitrates[4] = { 10, 20, 55, 110 }; int idx = i * MCS_GROUP_RATES + j; unsigned int duration; @@ -214,6 +215,8 @@ minstrel_ht_stats_csv_dump(struct minstrel_ht_sta *mi, int i, char *p) p += sprintf(p, "VHT%c0,", htmode); p += sprintf(p, "%cGI,", gimode); p += sprintf(p, "%d,", mg->streams); + } else if (i == MINSTREL_OFDM_GROUP) { + p += sprintf(p, "OFDM,,1,"); } else { p += sprintf(p, "CCK,"); p += sprintf(p, "%cP,", j < 4 ? 'L' : 'S'); @@ -231,7 +234,13 @@ minstrel_ht_stats_csv_dump(struct minstrel_ht_sta *mi, int i, char *p) } else if (gflags & IEEE80211_TX_RC_VHT_MCS) { p += sprintf(p, ",MCS%-1u/%1u,", j, mg->streams); } else { - int r = bitrates[j % 4]; + int r; + + if (i == MINSTREL_OFDM_GROUP) + r = minstrel_ofdm_bitrates[j % 8]; + else + r = minstrel_cck_bitrates[j % 4]; + p += sprintf(p, ",%2u.%1uM,", r / 10, r % 10); } @@ -270,22 +279,12 @@ minstrel_ht_stats_csv_dump(struct minstrel_ht_sta *mi, int i, char *p) static int minstrel_ht_stats_csv_open(struct inode *inode, struct file *file) { - struct minstrel_ht_sta_priv *msp = inode->i_private; - struct minstrel_ht_sta *mi = &msp->ht; + struct minstrel_ht_sta *mi = inode->i_private; struct minstrel_debugfs_info *ms; unsigned int i; - int ret; char *p; - if (!msp->is_ht) { - inode->i_private = &msp->legacy; - ret = minstrel_stats_csv_open(inode, file); - inode->i_private = msp; - return ret; - } - ms = kmalloc(32768, GFP_KERNEL); - if (!ms) return -ENOMEM; @@ -316,10 +315,8 @@ static const struct file_operations minstrel_ht_stat_csv_fops = { void minstrel_ht_add_sta_debugfs(void *priv, void *priv_sta, struct dentry *dir) { - struct minstrel_ht_sta_priv *msp = priv_sta; - - debugfs_create_file("rc_stats", 0444, dir, msp, + debugfs_create_file("rc_stats", 0444, dir, priv_sta, &minstrel_ht_stat_fops); - debugfs_create_file("rc_stats_csv", 0444, dir, msp, + debugfs_create_file("rc_stats_csv", 0444, dir, priv_sta, &minstrel_ht_stat_csv_fops); } diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 13b9bcc4865d..c1343c028b76 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -4095,7 +4095,9 @@ void ieee80211_check_fast_rx(struct sta_info *sta) .vif_type = sdata->vif.type, .control_port_protocol = sdata->control_port_protocol, }, *old, *new = NULL; + bool set_offload = false; bool assign = false; + bool offload; /* use sparse to check that we don't return without updating */ __acquire(check_fast_rx); @@ -4176,6 +4178,8 @@ void ieee80211_check_fast_rx(struct sta_info *sta) rcu_read_lock(); key = rcu_dereference(sta->ptk[sta->ptk_idx]); + if (!key) + key = rcu_dereference(sdata->default_unicast_key); if (key) { switch (key->conf.cipher) { case WLAN_CIPHER_SUITE_TKIP: @@ -4206,6 +4210,17 @@ void ieee80211_check_fast_rx(struct sta_info *sta) if (assign) new = kmemdup(&fastrx, sizeof(fastrx), GFP_KERNEL); + offload = assign && + (sdata->vif.offload_flags & IEEE80211_OFFLOAD_DECAP_ENABLED); + + if (offload) + set_offload = !test_and_set_sta_flag(sta, WLAN_STA_DECAP_OFFLOAD); + else + set_offload = test_and_clear_sta_flag(sta, WLAN_STA_DECAP_OFFLOAD); + + if (set_offload) + drv_sta_set_decap_offload(local, sdata, &sta->sta, assign); + spin_lock_bh(&sta->lock); old = rcu_dereference_protected(sta->fast_rx, true); rcu_assign_pointer(sta->fast_rx, new); @@ -4252,6 +4267,104 @@ void ieee80211_check_fast_rx_iface(struct ieee80211_sub_if_data *sdata) mutex_unlock(&local->sta_mtx); } +static void ieee80211_rx_8023(struct ieee80211_rx_data *rx, + struct ieee80211_fast_rx *fast_rx, + int orig_len) +{ + struct ieee80211_sta_rx_stats *stats; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); + struct sta_info *sta = rx->sta; + struct sk_buff *skb = rx->skb; + void *sa = skb->data + ETH_ALEN; + void *da = skb->data; + + stats = &sta->rx_stats; + if (fast_rx->uses_rss) + stats = this_cpu_ptr(sta->pcpu_rx_stats); + + /* statistics part of ieee80211_rx_h_sta_process() */ + if (!(status->flag & RX_FLAG_NO_SIGNAL_VAL)) { + stats->last_signal = status->signal; + if (!fast_rx->uses_rss) + ewma_signal_add(&sta->rx_stats_avg.signal, + -status->signal); + } + + if (status->chains) { + int i; + + stats->chains = status->chains; + for (i = 0; i < ARRAY_SIZE(status->chain_signal); i++) { + int signal = status->chain_signal[i]; + + if (!(status->chains & BIT(i))) + continue; + + stats->chain_signal_last[i] = signal; + if (!fast_rx->uses_rss) + ewma_signal_add(&sta->rx_stats_avg.chain_signal[i], + -signal); + } + } + /* end of statistics */ + + stats->last_rx = jiffies; + stats->last_rate = sta_stats_encode_rate(status); + + stats->fragments++; + stats->packets++; + + skb->dev = fast_rx->dev; + + dev_sw_netstats_rx_add(fast_rx->dev, skb->len); + + /* The seqno index has the same property as needed + * for the rx_msdu field, i.e. it is IEEE80211_NUM_TIDS + * for non-QoS-data frames. Here we know it's a data + * frame, so count MSDUs. + */ + u64_stats_update_begin(&stats->syncp); + stats->msdu[rx->seqno_idx]++; + stats->bytes += orig_len; + u64_stats_update_end(&stats->syncp); + + if (fast_rx->internal_forward) { + struct sk_buff *xmit_skb = NULL; + if (is_multicast_ether_addr(da)) { + xmit_skb = skb_copy(skb, GFP_ATOMIC); + } else if (!ether_addr_equal(da, sa) && + sta_info_get(rx->sdata, da)) { + xmit_skb = skb; + skb = NULL; + } + + if (xmit_skb) { + /* + * Send to wireless media and increase priority by 256 + * to keep the received priority instead of + * reclassifying the frame (see cfg80211_classify8021d). + */ + xmit_skb->priority += 256; + xmit_skb->protocol = htons(ETH_P_802_3); + skb_reset_network_header(xmit_skb); + skb_reset_mac_header(xmit_skb); + dev_queue_xmit(xmit_skb); + } + + if (!skb) + return; + } + + /* deliver to local stack */ + skb->protocol = eth_type_trans(skb, fast_rx->dev); + memset(skb->cb, 0, sizeof(skb->cb)); + if (rx->list) + list_add_tail(&skb->list, rx->list); + else + netif_receive_skb(skb); + +} + static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, struct ieee80211_fast_rx *fast_rx) { @@ -4272,9 +4385,6 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, } addrs __aligned(2); struct ieee80211_sta_rx_stats *stats = &sta->rx_stats; - if (fast_rx->uses_rss) - stats = this_cpu_ptr(sta->pcpu_rx_stats); - /* for parallel-rx, we need to have DUP_VALIDATED, otherwise we write * to a common data structure; drivers can implement that per queue * but we don't have that information in mac80211 @@ -4348,32 +4458,6 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, pskb_trim(skb, skb->len - fast_rx->icv_len)) goto drop; - /* statistics part of ieee80211_rx_h_sta_process() */ - if (!(status->flag & RX_FLAG_NO_SIGNAL_VAL)) { - stats->last_signal = status->signal; - if (!fast_rx->uses_rss) - ewma_signal_add(&sta->rx_stats_avg.signal, - -status->signal); - } - - if (status->chains) { - int i; - - stats->chains = status->chains; - for (i = 0; i < ARRAY_SIZE(status->chain_signal); i++) { - int signal = status->chain_signal[i]; - - if (!(status->chains & BIT(i))) - continue; - - stats->chain_signal_last[i] = signal; - if (!fast_rx->uses_rss) - ewma_signal_add(&sta->rx_stats_avg.chain_signal[i], - -signal); - } - } - /* end of statistics */ - if (rx->key && !ieee80211_has_protected(hdr->frame_control)) goto drop; @@ -4385,12 +4469,6 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, return true; } - stats->last_rx = jiffies; - stats->last_rate = sta_stats_encode_rate(status); - - stats->fragments++; - stats->packets++; - /* do the header conversion - first grab the addresses */ ether_addr_copy(addrs.da, skb->data + fast_rx->da_offs); ether_addr_copy(addrs.sa, skb->data + fast_rx->sa_offs); @@ -4399,58 +4477,14 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, /* push the addresses in front */ memcpy(skb_push(skb, sizeof(addrs)), &addrs, sizeof(addrs)); - skb->dev = fast_rx->dev; - - dev_sw_netstats_rx_add(fast_rx->dev, skb->len); - - /* The seqno index has the same property as needed - * for the rx_msdu field, i.e. it is IEEE80211_NUM_TIDS - * for non-QoS-data frames. Here we know it's a data - * frame, so count MSDUs. - */ - u64_stats_update_begin(&stats->syncp); - stats->msdu[rx->seqno_idx]++; - stats->bytes += orig_len; - u64_stats_update_end(&stats->syncp); - - if (fast_rx->internal_forward) { - struct sk_buff *xmit_skb = NULL; - if (is_multicast_ether_addr(addrs.da)) { - xmit_skb = skb_copy(skb, GFP_ATOMIC); - } else if (!ether_addr_equal(addrs.da, addrs.sa) && - sta_info_get(rx->sdata, addrs.da)) { - xmit_skb = skb; - skb = NULL; - } - - if (xmit_skb) { - /* - * Send to wireless media and increase priority by 256 - * to keep the received priority instead of - * reclassifying the frame (see cfg80211_classify8021d). - */ - xmit_skb->priority += 256; - xmit_skb->protocol = htons(ETH_P_802_3); - skb_reset_network_header(xmit_skb); - skb_reset_mac_header(xmit_skb); - dev_queue_xmit(xmit_skb); - } - - if (!skb) - return true; - } - - /* deliver to local stack */ - skb->protocol = eth_type_trans(skb, fast_rx->dev); - memset(skb->cb, 0, sizeof(skb->cb)); - if (rx->list) - list_add_tail(&skb->list, rx->list); - else - netif_receive_skb(skb); + ieee80211_rx_8023(rx, fast_rx, orig_len); return true; drop: dev_kfree_skb(skb); + if (fast_rx->uses_rss) + stats = this_cpu_ptr(sta->pcpu_rx_stats); + stats->dropped++; return true; } @@ -4504,6 +4538,43 @@ static bool ieee80211_prepare_and_rx_handle(struct ieee80211_rx_data *rx, return true; } +static void __ieee80211_rx_handle_8023(struct ieee80211_hw *hw, + struct ieee80211_sta *pubsta, + struct sk_buff *skb, + struct list_head *list) +{ + struct ieee80211_local *local = hw_to_local(hw); + struct ieee80211_fast_rx *fast_rx; + struct ieee80211_rx_data rx; + + memset(&rx, 0, sizeof(rx)); + rx.skb = skb; + rx.local = local; + rx.list = list; + + I802_DEBUG_INC(local->dot11ReceivedFragmentCount); + + /* drop frame if too short for header */ + if (skb->len < sizeof(struct ethhdr)) + goto drop; + + if (!pubsta) + goto drop; + + rx.sta = container_of(pubsta, struct sta_info, sta); + rx.sdata = rx.sta->sdata; + + fast_rx = rcu_dereference(rx.sta->fast_rx); + if (!fast_rx) + goto drop; + + ieee80211_rx_8023(&rx, fast_rx, skb->len); + return; + +drop: + dev_kfree_skb(skb); +} + /* * This is the actual Rx frames handler. as it belongs to Rx path it must * be called with rcu_read_lock protection. @@ -4735,13 +4806,17 @@ void ieee80211_rx_list(struct ieee80211_hw *hw, struct ieee80211_sta *pubsta, * if it was previously present. * Also, frames with less than 16 bytes are dropped. */ - skb = ieee80211_rx_monitor(local, skb, rate); + if (!(status->flag & RX_FLAG_8023)) + skb = ieee80211_rx_monitor(local, skb, rate); if (skb) { ieee80211_tpt_led_trig_rx(local, ((struct ieee80211_hdr *)skb->data)->frame_control, skb->len); - __ieee80211_rx_handle_packet(hw, pubsta, skb, list); + if (status->flag & RX_FLAG_8023) + __ieee80211_rx_handle_8023(hw, pubsta, skb, list); + else + __ieee80211_rx_handle_packet(hw, pubsta, skb, list); } kcov_remote_stop(); diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 7afd07636b81..78b9d0c7cc58 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -71,6 +71,7 @@ * until pending frames are delivered * @WLAN_STA_USES_ENCRYPTION: This station was configured for encryption, * so drop all packets without a key later. + * @WLAN_STA_DECAP_OFFLOAD: This station uses rx decap offload * * @NUM_WLAN_STA_FLAGS: number of defined flags */ @@ -102,6 +103,7 @@ enum ieee80211_sta_info_flags { WLAN_STA_MPSP_RECIPIENT, WLAN_STA_PS_DELIVER, WLAN_STA_USES_ENCRYPTION, + WLAN_STA_DECAP_OFFLOAD, NUM_WLAN_STA_FLAGS, }; diff --git a/net/mac80211/tdls.c b/net/mac80211/tdls.c index e01e4daeb8cd..f91d02b81b92 100644 --- a/net/mac80211/tdls.c +++ b/net/mac80211/tdls.c @@ -1927,7 +1927,7 @@ ieee80211_process_tdls_channel_switch(struct ieee80211_sub_if_data *sdata, struct ieee80211_tdls_data *tf = (void *)skb->data; struct wiphy *wiphy = sdata->local->hw.wiphy; - ASSERT_RTNL(); + lockdep_assert_wiphy(wiphy); /* make sure the driver supports it */ if (!(wiphy->features & NL80211_FEATURE_TDLS_CHANNEL_SWITCH)) @@ -1979,7 +1979,7 @@ void ieee80211_tdls_chsw_work(struct work_struct *wk) struct sk_buff *skb; struct ieee80211_tdls_data *tf; - rtnl_lock(); + wiphy_lock(local->hw.wiphy); while ((skb = skb_dequeue(&local->skb_queue_tdls_chsw))) { tf = (struct ieee80211_tdls_data *)skb->data; list_for_each_entry(sdata, &local->interfaces, list) { @@ -1994,7 +1994,7 @@ void ieee80211_tdls_chsw_work(struct work_struct *wk) kfree_skb(skb); } - rtnl_unlock(); + wiphy_unlock(local->hw.wiphy); } void ieee80211_tdls_handle_disconnect(struct ieee80211_sub_if_data *sdata, diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h index 601322e16957..8fcc39056402 100644 --- a/net/mac80211/trace.h +++ b/net/mac80211/trace.h @@ -2761,7 +2761,7 @@ DEFINE_EVENT(local_sdata_addr_evt, drv_update_vif_offload, TP_ARGS(local, sdata) ); -TRACE_EVENT(drv_sta_set_4addr, +DECLARE_EVENT_CLASS(sta_flag_evt, TP_PROTO(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata, struct ieee80211_sta *sta, bool enabled), @@ -2788,6 +2788,22 @@ TRACE_EVENT(drv_sta_set_4addr, ) ); +DEFINE_EVENT(sta_flag_evt, drv_sta_set_4addr, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta, bool enabled), + + TP_ARGS(local, sdata, sta, enabled) +); + +DEFINE_EVENT(sta_flag_evt, drv_sta_set_decap_offload, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta, bool enabled), + + TP_ARGS(local, sdata, sta, enabled) +); + #endif /* !__MAC80211_DRIVER_TRACE || TRACE_HEADER_MULTI_READ */ #undef TRACE_INCLUDE_PATH diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 6422da6690f7..d626e6808bef 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -649,7 +649,7 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) if (!skip_hw && tx->key && tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) info->control.hw_key = &tx->key->conf; - } else if (!ieee80211_is_mgmt(hdr->frame_control) && tx->sta && + } else if (ieee80211_is_data_present(hdr->frame_control) && tx->sta && test_sta_flag(tx->sta, WLAN_STA_USES_ENCRYPTION)) { return TX_DROP; } @@ -1309,7 +1309,7 @@ static struct sk_buff *codel_dequeue_func(struct codel_vars *cvars, fq = &local->fq; if (cvars == &txqi->def_cvars) - flow = &txqi->def_flow; + flow = &txqi->tin.default_flow; else flow = &fq->flows[cvars - local->cvars]; @@ -1352,7 +1352,7 @@ static struct sk_buff *fq_tin_dequeue_func(struct fq *fq, cparams = &local->cparams; } - if (flow == &txqi->def_flow) + if (flow == &tin->default_flow) cvars = &txqi->def_cvars; else cvars = &local->cvars[flow - fq->flows]; @@ -1379,17 +1379,6 @@ static void fq_skb_free_func(struct fq *fq, ieee80211_free_txskb(&local->hw, skb); } -static struct fq_flow *fq_flow_get_default_func(struct fq *fq, - struct fq_tin *tin, - int idx, - struct sk_buff *skb) -{ - struct txq_info *txqi; - - txqi = container_of(tin, struct txq_info, tin); - return &txqi->def_flow; -} - static void ieee80211_txq_enqueue(struct ieee80211_local *local, struct txq_info *txqi, struct sk_buff *skb) @@ -1402,8 +1391,7 @@ static void ieee80211_txq_enqueue(struct ieee80211_local *local, spin_lock_bh(&fq->lock); fq_tin_enqueue(fq, tin, flow_idx, skb, - fq_skb_free_func, - fq_flow_get_default_func); + fq_skb_free_func); spin_unlock_bh(&fq->lock); } @@ -1446,7 +1434,6 @@ void ieee80211_txq_init(struct ieee80211_sub_if_data *sdata, struct txq_info *txqi, int tid) { fq_tin_init(&txqi->tin); - fq_flow_init(&txqi->def_flow); codel_vars_init(&txqi->def_cvars); codel_stats_init(&txqi->cstats); __skb_queue_head_init(&txqi->frags); @@ -2133,6 +2120,10 @@ bool ieee80211_parse_tx_radiotap(struct sk_buff *skb, if (mcs_known & IEEE80211_RADIOTAP_MCS_HAVE_BW && mcs_bw == IEEE80211_RADIOTAP_MCS_BW_40) rate_flags |= IEEE80211_TX_RC_40_MHZ_WIDTH; + + if (mcs_known & IEEE80211_RADIOTAP_MCS_HAVE_FEC && + mcs_flags & IEEE80211_RADIOTAP_MCS_FEC_LDPC) + info->flags |= IEEE80211_TX_CTL_LDPC; break; case IEEE80211_RADIOTAP_VHT: @@ -3283,8 +3274,7 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, */ tin = &txqi->tin; - flow = fq_flow_classify(fq, tin, flow_idx, skb, - fq_flow_get_default_func); + flow = fq_flow_classify(fq, tin, flow_idx, skb); head = skb_peek_tail(&flow->queue); if (!head || skb_is_gso(head)) goto out; @@ -3351,8 +3341,6 @@ out_recalc: if (head->len != orig_len) { flow->backlog += head->len - orig_len; tin->backlog_bytes += head->len - orig_len; - - fq_recalc_backlog(fq, tin, flow); } out: spin_unlock_bh(&fq->lock); @@ -3809,7 +3797,7 @@ void __ieee80211_schedule_txq(struct ieee80211_hw *hw, * get immediately moved to the back of the list on the next * call to ieee80211_next_txq(). */ - if (txqi->txq.sta && + if (txqi->txq.sta && local->airtime_flags && wiphy_ext_feature_isset(local->hw.wiphy, NL80211_EXT_FEATURE_AIRTIME_FAIRNESS)) list_add(&txqi->schedule_order, @@ -3823,6 +3811,8 @@ void __ieee80211_schedule_txq(struct ieee80211_hw *hw, } EXPORT_SYMBOL(__ieee80211_schedule_txq); +DEFINE_STATIC_KEY_FALSE(aql_disable); + bool ieee80211_txq_airtime_check(struct ieee80211_hw *hw, struct ieee80211_txq *txq) { @@ -3832,6 +3822,9 @@ bool ieee80211_txq_airtime_check(struct ieee80211_hw *hw, if (!wiphy_ext_feature_isset(local->hw.wiphy, NL80211_EXT_FEATURE_AQL)) return true; + if (static_branch_unlikely(&aql_disable)) + return true; + if (!txq->sta) return true; @@ -4251,7 +4244,6 @@ netdev_tx_t ieee80211_subif_start_xmit_8023(struct sk_buff *skb, struct ethhdr *ehdr = (struct ethhdr *)skb->data; struct ieee80211_key *key; struct sta_info *sta; - bool offload = true; if (unlikely(skb->len < ETH_HLEN)) { kfree_skb(skb); @@ -4267,18 +4259,22 @@ netdev_tx_t ieee80211_subif_start_xmit_8023(struct sk_buff *skb, if (unlikely(IS_ERR_OR_NULL(sta) || !sta->uploaded || !test_sta_flag(sta, WLAN_STA_AUTHORIZED) || - sdata->control_port_protocol == ehdr->h_proto)) - offload = false; - else if ((key = rcu_dereference(sta->ptk[sta->ptk_idx])) && - (!(key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) || - key->conf.cipher == WLAN_CIPHER_SUITE_TKIP)) - offload = false; - - if (offload) - ieee80211_8023_xmit(sdata, dev, sta, key, skb); - else - ieee80211_subif_start_xmit(skb, dev); + sdata->control_port_protocol == ehdr->h_proto)) + goto skip_offload; + + key = rcu_dereference(sta->ptk[sta->ptk_idx]); + if (!key) + key = rcu_dereference(sdata->default_unicast_key); + + if (key && (!(key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) || + key->conf.cipher == WLAN_CIPHER_SUITE_TKIP)) + goto skip_offload; + + ieee80211_8023_xmit(sdata, dev, sta, key, skb); + goto out; +skip_offload: + ieee80211_subif_start_xmit(skb, dev); out: rcu_read_unlock(); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 8d3ae6b2f95f..f080fcf60e45 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -832,7 +832,7 @@ void ieee80211_iterate_active_interfaces_atomic( } EXPORT_SYMBOL_GPL(ieee80211_iterate_active_interfaces_atomic); -void ieee80211_iterate_active_interfaces_rtnl( +void ieee80211_iterate_active_interfaces_mtx( struct ieee80211_hw *hw, u32 iter_flags, void (*iterator)(void *data, u8 *mac, struct ieee80211_vif *vif), @@ -840,12 +840,12 @@ void ieee80211_iterate_active_interfaces_rtnl( { struct ieee80211_local *local = hw_to_local(hw); - ASSERT_RTNL(); + lockdep_assert_wiphy(hw->wiphy); __iterate_interfaces(local, iter_flags | IEEE80211_IFACE_ITER_ACTIVE, iterator, data); } -EXPORT_SYMBOL_GPL(ieee80211_iterate_active_interfaces_rtnl); +EXPORT_SYMBOL_GPL(ieee80211_iterate_active_interfaces_mtx); static void __iterate_stations(struct ieee80211_local *local, void (*iterator)(void *data, @@ -2595,7 +2595,7 @@ int ieee80211_reconfig(struct ieee80211_local *local) mutex_unlock(&local->mtx); if (sched_scan_stopped) - cfg80211_sched_scan_stopped_rtnl(local->hw.wiphy, 0); + cfg80211_sched_scan_stopped_locked(local->hw.wiphy, 0); wake_up: @@ -3811,7 +3811,7 @@ void ieee80211_dfs_cac_cancel(struct ieee80211_local *local) struct cfg80211_chan_def chandef; /* for interface list, to avoid linking iflist_mtx and chanctx_mtx */ - ASSERT_RTNL(); + lockdep_assert_wiphy(local->hw.wiphy); mutex_lock(&local->mtx); list_for_each_entry(sdata, &local->interfaces, list) { @@ -3851,9 +3851,9 @@ void ieee80211_dfs_radar_detected_work(struct work_struct *work) } mutex_unlock(&local->chanctx_mtx); - rtnl_lock(); + wiphy_lock(local->hw.wiphy); ieee80211_dfs_cac_cancel(local); - rtnl_unlock(); + wiphy_unlock(local->hw.wiphy); if (num_chanctx > 1) /* XXX: multi-channel is not supported yet */ diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c index c3ca97373774..e856f9092137 100644 --- a/net/mac80211/vht.c +++ b/net/mac80211/vht.c @@ -484,6 +484,7 @@ enum ieee80211_sta_rx_bandwidth ieee80211_sta_cur_vht_bw(struct sta_info *sta) void ieee80211_sta_set_rx_nss(struct sta_info *sta) { u8 ht_rx_nss = 0, vht_rx_nss = 0, he_rx_nss = 0, rx_nss; + bool support_160; /* if we received a notification already don't overwrite it */ if (sta->sta.rx_nss) @@ -514,7 +515,13 @@ void ieee80211_sta_set_rx_nss(struct sta_info *sta) } } - he_rx_nss = min(rx_mcs_80, rx_mcs_160); + support_160 = he_cap->he_cap_elem.phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G; + + if (support_160) + he_rx_nss = min(rx_mcs_80, rx_mcs_160); + else + he_rx_nss = rx_mcs_80; } if (sta->sta.ht_cap.ht_supported) { diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c index 9b1f6298bbdb..83976b9ee99b 100644 --- a/net/mptcp/pm_netlink.c +++ b/net/mptcp/pm_netlink.c @@ -60,15 +60,20 @@ static bool addresses_equal(const struct mptcp_addr_info *a, { bool addr_equals = false; - if (a->family != b->family) - return false; - - if (a->family == AF_INET) - addr_equals = a->addr.s_addr == b->addr.s_addr; + if (a->family == b->family) { + if (a->family == AF_INET) + addr_equals = a->addr.s_addr == b->addr.s_addr; #if IS_ENABLED(CONFIG_MPTCP_IPV6) - else - addr_equals = !ipv6_addr_cmp(&a->addr6, &b->addr6); + else + addr_equals = !ipv6_addr_cmp(&a->addr6, &b->addr6); + } else if (a->family == AF_INET) { + if (ipv6_addr_v4mapped(&b->addr6)) + addr_equals = a->addr.s_addr == b->addr6.s6_addr32[3]; + } else if (b->family == AF_INET) { + if (ipv6_addr_v4mapped(&a->addr6)) + addr_equals = a->addr6.s6_addr32[3] == b->addr.s_addr; #endif + } if (!addr_equals) return false; @@ -137,6 +142,7 @@ select_local_address(const struct pm_nl_pernet *pernet, struct mptcp_sock *msk) { struct mptcp_pm_addr_entry *entry, *ret = NULL; + struct sock *sk = (struct sock *)msk; rcu_read_lock(); __mptcp_flush_join_list(msk); @@ -144,11 +150,20 @@ select_local_address(const struct pm_nl_pernet *pernet, if (!(entry->addr.flags & MPTCP_PM_ADDR_FLAG_SUBFLOW)) continue; + if (entry->addr.family != sk->sk_family) { +#if IS_ENABLED(CONFIG_MPTCP_IPV6) + if ((entry->addr.family == AF_INET && + !ipv6_addr_v4mapped(&sk->sk_v6_daddr)) || + (sk->sk_family == AF_INET && + !ipv6_addr_v4mapped(&entry->addr.addr6))) +#endif + continue; + } + /* avoid any address already in use by subflows and * pending join */ - if (entry->addr.family == ((struct sock *)msk)->sk_family && - !lookup_subflow_by_saddr(&msk->conn_list, &entry->addr)) { + if (!lookup_subflow_by_saddr(&msk->conn_list, &entry->addr)) { ret = entry; break; } @@ -310,7 +325,6 @@ void mptcp_pm_free_anno_list(struct mptcp_sock *msk) static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk) { - struct mptcp_addr_info remote = { 0 }; struct sock *sk = (struct sock *)msk; struct mptcp_pm_addr_entry *local; struct pm_nl_pernet *pernet; @@ -344,13 +358,14 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk) /* check if should create a new subflow */ if (msk->pm.local_addr_used < msk->pm.local_addr_max && msk->pm.subflows < msk->pm.subflows_max) { - remote_address((struct sock_common *)sk, &remote); - local = select_local_address(pernet, msk); if (local) { + struct mptcp_addr_info remote = { 0 }; + msk->pm.local_addr_used++; msk->pm.subflows++; check_work_pending(msk); + remote_address((struct sock_common *)sk, &remote); spin_unlock_bh(&msk->pm.lock); __mptcp_subflow_connect(sk, &local->addr, &remote); spin_lock_bh(&msk->pm.lock); diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index f998a077c7dd..a033bf9c26ee 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -45,6 +45,9 @@ static struct percpu_counter mptcp_sockets_allocated; static void __mptcp_destroy_sock(struct sock *sk); static void __mptcp_check_send_data_fin(struct sock *sk); +DEFINE_PER_CPU(struct mptcp_delegated_action, mptcp_delegated_actions); +static struct net_device mptcp_napi_dev; + /* If msk has an initial subflow socket, and the MP_CAPABLE handshake has not * completed yet or has failed, return the subflow socket. * Otherwise return NULL. @@ -114,11 +117,7 @@ static int __mptcp_socket_create(struct mptcp_sock *msk) list_add(&subflow->node, &msk->conn_list); sock_hold(ssock->sk); subflow->request_mptcp = 1; - - /* accept() will wait on first subflow sk_wq, and we always wakes up - * via msk->sk_socket - */ - RCU_INIT_POINTER(msk->first->sk_wq, &sk->sk_socket->wq); + mptcp_sock_graft(msk->first, sk->sk_socket); return 0; } @@ -734,10 +733,14 @@ wake: void __mptcp_flush_join_list(struct mptcp_sock *msk) { + struct mptcp_subflow_context *subflow; + if (likely(list_empty(&msk->join_list))) return; spin_lock_bh(&msk->join_list_lock); + list_for_each_entry(subflow, &msk->join_list, node) + mptcp_propagate_sndbuf((struct sock *)msk, mptcp_subflow_tcp_sock(subflow)); list_splice_tail_init(&msk->join_list, &msk->conn_list); spin_unlock_bh(&msk->join_list_lock); } @@ -1037,13 +1040,6 @@ out: __mptcp_update_wmem(sk); sk_mem_reclaim_partial(sk); } - - if (sk_stream_is_writeable(sk)) { - /* pairs with memory barrier in mptcp_poll */ - smp_mb(); - if (test_and_clear_bit(MPTCP_NOSPACE, &msk->flags)) - sk_stream_write_space(sk); - } } if (snd_una == READ_ONCE(msk->snd_nxt)) { @@ -1362,8 +1358,7 @@ struct subflow_send_info { u64 ratio; }; -static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk, - u32 *sndbuf) +static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk) { struct subflow_send_info send_info[2]; struct mptcp_subflow_context *subflow; @@ -1374,24 +1369,17 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk, sock_owned_by_me((struct sock *)msk); - *sndbuf = 0; if (__mptcp_check_fallback(msk)) { if (!msk->first) return NULL; - *sndbuf = msk->first->sk_sndbuf; return sk_stream_memory_free(msk->first) ? msk->first : NULL; } /* re-use last subflow, if the burst allow that */ if (msk->last_snd && msk->snd_burst > 0 && sk_stream_memory_free(msk->last_snd) && - mptcp_subflow_active(mptcp_subflow_ctx(msk->last_snd))) { - mptcp_for_each_subflow(msk, subflow) { - ssk = mptcp_subflow_tcp_sock(subflow); - *sndbuf = max(tcp_sk(ssk)->snd_wnd, *sndbuf); - } + mptcp_subflow_active(mptcp_subflow_ctx(msk->last_snd))) return msk->last_snd; - } /* pick the subflow with the lower wmem/wspace ratio */ for (i = 0; i < 2; ++i) { @@ -1404,8 +1392,7 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk, continue; nr_active += !subflow->backup; - *sndbuf = max(tcp_sk(ssk)->snd_wnd, *sndbuf); - if (!sk_stream_memory_free(subflow->tcp_sock)) + if (!sk_stream_memory_free(subflow->tcp_sock) || !tcp_sk(ssk)->snd_wnd) continue; pace = READ_ONCE(ssk->sk_pacing_rate); @@ -1431,9 +1418,10 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk, if (send_info[0].ssk) { msk->last_snd = send_info[0].ssk; msk->snd_burst = min_t(int, MPTCP_SEND_BURST_SIZE, - sk_stream_wspace(msk->last_snd)); + tcp_sk(msk->last_snd)->snd_wnd); return msk->last_snd; } + return NULL; } @@ -1454,7 +1442,6 @@ static void mptcp_push_pending(struct sock *sk, unsigned int flags) }; struct mptcp_data_frag *dfrag; int len, copied = 0; - u32 sndbuf; while ((dfrag = mptcp_send_head(sk))) { info.sent = dfrag->already_sent; @@ -1465,12 +1452,7 @@ static void mptcp_push_pending(struct sock *sk, unsigned int flags) prev_ssk = ssk; __mptcp_flush_join_list(msk); - ssk = mptcp_subflow_get_send(msk, &sndbuf); - - /* do auto tuning */ - if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK) && - sndbuf > READ_ONCE(sk->sk_sndbuf)) - WRITE_ONCE(sk->sk_sndbuf, sndbuf); + ssk = mptcp_subflow_get_send(msk); /* try to keep the subflow socket lock across * consecutive xmit on the same socket @@ -1527,7 +1509,9 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk) struct mptcp_sock *msk = mptcp_sk(sk); struct mptcp_sendmsg_info info; struct mptcp_data_frag *dfrag; + struct sock *xmit_ssk; int len, copied = 0; + bool first = true; info.flags = 0; while ((dfrag = mptcp_send_head(sk))) { @@ -1537,10 +1521,17 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk) while (len > 0) { int ret = 0; - /* do auto tuning */ - if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK) && - ssk->sk_sndbuf > READ_ONCE(sk->sk_sndbuf)) - WRITE_ONCE(sk->sk_sndbuf, ssk->sk_sndbuf); + /* the caller already invoked the packet scheduler, + * check for a different subflow usage only after + * spooling the first chunk of data + */ + xmit_ssk = first ? ssk : mptcp_subflow_get_send(mptcp_sk(sk)); + if (!xmit_ssk) + goto out; + if (xmit_ssk != ssk) { + mptcp_subflow_delegate(mptcp_subflow_ctx(xmit_ssk)); + goto out; + } if (unlikely(mptcp_must_reclaim_memory(sk, ssk))) { __mptcp_update_wmem(sk); @@ -1560,6 +1551,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk) msk->tx_pending_data -= ret; copied += ret; len -= ret; + first = false; } WRITE_ONCE(msk->first_pending, mptcp_send_next(sk)); } @@ -1579,6 +1571,15 @@ out: } } +static void mptcp_set_nospace(struct sock *sk) +{ + /* enable autotune */ + set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); + + /* will be cleared on avail space */ + set_bit(MPTCP_NOSPACE, &mptcp_sk(sk)->flags); +} + static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) { struct mptcp_sock *msk = mptcp_sk(sk); @@ -1680,7 +1681,7 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) continue; wait_for_memory: - set_bit(MPTCP_NOSPACE, &msk->flags); + mptcp_set_nospace(sk); mptcp_push_pending(sk, msg->msg_flags); ret = sk_stream_wait_memory(sk, &timeo); if (ret) @@ -2116,9 +2117,6 @@ static struct sock *mptcp_subflow_get_retrans(const struct mptcp_sock *msk) void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, struct mptcp_subflow_context *subflow) { - bool dispose_socket = false; - struct socket *sock; - list_del(&subflow->node); lock_sock_nested(ssk, SINGLE_DEPTH_NESTING); @@ -2126,11 +2124,8 @@ void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, /* if we are invoked by the msk cleanup code, the subflow is * already orphaned */ - sock = ssk->sk_socket; - if (sock) { - dispose_socket = sock != sk->sk_socket; + if (ssk->sk_socket) sock_orphan(ssk); - } subflow->disposable = 1; @@ -2148,8 +2143,6 @@ void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, __sock_put(ssk); } release_sock(ssk); - if (dispose_socket) - iput(SOCK_INODE(sock)); sock_put(ssk); } @@ -2536,6 +2529,12 @@ static void __mptcp_destroy_sock(struct sock *sk) pr_debug("msk=%p", msk); + /* dispose the ancillatory tcp socket, if any */ + if (msk->subflow) { + iput(SOCK_INODE(msk->subflow)); + msk->subflow = NULL; + } + /* be sure to always acquire the join list lock, to sync vs * mptcp_finish_join(). */ @@ -2586,20 +2585,10 @@ cleanup: inet_csk(sk)->icsk_mtup.probe_timestamp = tcp_jiffies32; list_for_each_entry(subflow, &mptcp_sk(sk)->conn_list, node) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); - bool slow, dispose_socket; - struct socket *sock; + bool slow = lock_sock_fast(ssk); - slow = lock_sock_fast(ssk); - sock = ssk->sk_socket; - dispose_socket = sock && sock != sk->sk_socket; sock_orphan(ssk); unlock_sock_fast(ssk, slow); - - /* for the outgoing subflows we additionally need to free - * the associated socket - */ - if (dispose_socket) - iput(SOCK_INODE(sock)); } sock_orphan(sk); @@ -2928,10 +2917,16 @@ void __mptcp_check_push(struct sock *sk, struct sock *ssk) if (!mptcp_send_head(sk)) return; - if (!sock_owned_by_user(sk)) - __mptcp_subflow_push_pending(sk, ssk); - else + if (!sock_owned_by_user(sk)) { + struct sock *xmit_ssk = mptcp_subflow_get_send(mptcp_sk(sk)); + + if (xmit_ssk == ssk) + __mptcp_subflow_push_pending(sk, ssk); + else if (xmit_ssk) + mptcp_subflow_delegate(mptcp_subflow_ctx(xmit_ssk)); + } else { set_bit(MPTCP_PUSH_PENDING, &mptcp_sk(sk)->flags); + } } #define MPTCP_DEFERRED_ALL (TCPF_WRITE_TIMER_DEFERRED) @@ -2979,6 +2974,20 @@ static void mptcp_release_cb(struct sock *sk) } } +void mptcp_subflow_process_delegated(struct sock *ssk) +{ + struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); + struct sock *sk = subflow->conn; + + mptcp_data_lock(sk); + if (!sock_owned_by_user(sk)) + __mptcp_subflow_push_pending(sk, ssk); + else + set_bit(MPTCP_PUSH_PENDING, &mptcp_sk(sk)->flags); + mptcp_data_unlock(sk); + mptcp_subflow_delegated_done(subflow); +} + static int mptcp_hash(struct sock *sk) { /* should never be called, @@ -3041,7 +3050,7 @@ void mptcp_finish_connect(struct sock *ssk) mptcp_rcv_space_init(msk, ssk); } -static void mptcp_sock_graft(struct sock *sk, struct socket *parent) +void mptcp_sock_graft(struct sock *sk, struct socket *parent) { write_lock_bh(&sk->sk_callback_lock); rcu_assign_pointer(sk->sk_wq, &parent->wq); @@ -3284,6 +3293,7 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, mptcp_copy_inaddrs(newsk, msk->first); mptcp_rcv_space_init(msk, msk->first); + mptcp_propagate_sndbuf(newsk, msk->first); /* set ssk->sk_socket of accept()ed flows to mptcp socket. * This is needed so NOSPACE flag can be set from tcp stack. @@ -3324,7 +3334,7 @@ static __poll_t mptcp_check_writeable(struct mptcp_sock *msk) if (sk_stream_is_writeable(sk)) return EPOLLOUT | EPOLLWRNORM; - set_bit(MPTCP_NOSPACE, &msk->flags); + mptcp_set_nospace(sk); smp_mb__after_atomic(); /* msk->flags is changed by write_space cb */ if (sk_stream_is_writeable(sk)) return EPOLLOUT | EPOLLWRNORM; @@ -3388,13 +3398,58 @@ static struct inet_protosw mptcp_protosw = { .flags = INET_PROTOSW_ICSK, }; +static int mptcp_napi_poll(struct napi_struct *napi, int budget) +{ + struct mptcp_delegated_action *delegated; + struct mptcp_subflow_context *subflow; + int work_done = 0; + + delegated = container_of(napi, struct mptcp_delegated_action, napi); + while ((subflow = mptcp_subflow_delegated_next(delegated)) != NULL) { + struct sock *ssk = mptcp_subflow_tcp_sock(subflow); + + bh_lock_sock_nested(ssk); + if (!sock_owned_by_user(ssk) && + mptcp_subflow_has_delegated_action(subflow)) + mptcp_subflow_process_delegated(ssk); + /* ... elsewhere tcp_release_cb_override already processed + * the action or will do at next release_sock(). + * In both case must dequeue the subflow here - on the same + * CPU that scheduled it. + */ + bh_unlock_sock(ssk); + sock_put(ssk); + + if (++work_done == budget) + return budget; + } + + /* always provide a 0 'work_done' argument, so that napi_complete_done + * will not try accessing the NULL napi->dev ptr + */ + napi_complete_done(napi, 0); + return work_done; +} + void __init mptcp_proto_init(void) { + struct mptcp_delegated_action *delegated; + int cpu; + mptcp_prot.h.hashinfo = tcp_prot.h.hashinfo; if (percpu_counter_init(&mptcp_sockets_allocated, 0, GFP_KERNEL)) panic("Failed to allocate MPTCP pcpu counter\n"); + init_dummy_netdev(&mptcp_napi_dev); + for_each_possible_cpu(cpu) { + delegated = per_cpu_ptr(&mptcp_delegated_actions, cpu); + INIT_LIST_HEAD(&delegated->head); + netif_tx_napi_add(&mptcp_napi_dev, &delegated->napi, mptcp_napi_poll, + NAPI_POLL_WEIGHT); + napi_enable(&delegated->napi); + } + mptcp_subflow_init(); mptcp_pm_init(); mptcp_token_init(); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index d6400ad2d615..1460705aaad0 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -378,6 +378,15 @@ enum mptcp_data_avail { MPTCP_SUBFLOW_OOO_DATA }; +struct mptcp_delegated_action { + struct napi_struct napi; + struct list_head head; +}; + +DECLARE_PER_CPU(struct mptcp_delegated_action, mptcp_delegated_actions); + +#define MPTCP_DELEGATE_SEND 0 + /* MPTCP subflow context */ struct mptcp_subflow_context { struct list_head node;/* conn_list of subflows */ @@ -415,6 +424,9 @@ struct mptcp_subflow_context { u8 local_id; u8 remote_id; + long delegated_status; + struct list_head delegated_node; /* link into delegated_action, protected by local BH */ + struct sock *tcp_sock; /* tcp sk backpointer */ struct sock *conn; /* parent mptcp_sock */ const struct inet_connection_sock_af_ops *icsk_af_ops; @@ -463,6 +475,61 @@ static inline void mptcp_add_pending_subflow(struct mptcp_sock *msk, spin_unlock_bh(&msk->join_list_lock); } +void mptcp_subflow_process_delegated(struct sock *ssk); + +static inline void mptcp_subflow_delegate(struct mptcp_subflow_context *subflow) +{ + struct mptcp_delegated_action *delegated; + bool schedule; + + /* The implied barrier pairs with mptcp_subflow_delegated_done(), and + * ensures the below list check sees list updates done prior to status + * bit changes + */ + if (!test_and_set_bit(MPTCP_DELEGATE_SEND, &subflow->delegated_status)) { + /* still on delegated list from previous scheduling */ + if (!list_empty(&subflow->delegated_node)) + return; + + /* the caller held the subflow bh socket lock */ + lockdep_assert_in_softirq(); + + delegated = this_cpu_ptr(&mptcp_delegated_actions); + schedule = list_empty(&delegated->head); + list_add_tail(&subflow->delegated_node, &delegated->head); + sock_hold(mptcp_subflow_tcp_sock(subflow)); + if (schedule) + napi_schedule(&delegated->napi); + } +} + +static inline struct mptcp_subflow_context * +mptcp_subflow_delegated_next(struct mptcp_delegated_action *delegated) +{ + struct mptcp_subflow_context *ret; + + if (list_empty(&delegated->head)) + return NULL; + + ret = list_first_entry(&delegated->head, struct mptcp_subflow_context, delegated_node); + list_del_init(&ret->delegated_node); + return ret; +} + +static inline bool mptcp_subflow_has_delegated_action(const struct mptcp_subflow_context *subflow) +{ + return test_bit(MPTCP_DELEGATE_SEND, &subflow->delegated_status); +} + +static inline void mptcp_subflow_delegated_done(struct mptcp_subflow_context *subflow) +{ + /* pairs with mptcp_subflow_delegate, ensures delegate_node is updated before + * touching the status bit + */ + smp_wmb(); + clear_bit(MPTCP_DELEGATE_SEND, &subflow->delegated_status); +} + int mptcp_is_enabled(struct net *net); unsigned int mptcp_get_add_addr_timeout(struct net *net); void mptcp_subflow_fully_established(struct mptcp_subflow_context *subflow, @@ -473,6 +540,7 @@ void mptcp_subflow_shutdown(struct sock *sk, struct sock *ssk, int how); void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, struct mptcp_subflow_context *subflow); void mptcp_subflow_reset(struct sock *ssk); +void mptcp_sock_graft(struct sock *sk, struct socket *parent); /* called with sk socket lock held */ int __mptcp_subflow_connect(struct sock *sk, const struct mptcp_addr_info *loc, @@ -521,6 +589,25 @@ static inline bool mptcp_data_fin_enabled(const struct mptcp_sock *msk) READ_ONCE(msk->write_seq) == READ_ONCE(msk->snd_nxt); } +static inline bool mptcp_propagate_sndbuf(struct sock *sk, struct sock *ssk) +{ + if ((sk->sk_userlocks & SOCK_SNDBUF_LOCK) || ssk->sk_sndbuf <= READ_ONCE(sk->sk_sndbuf)) + return false; + + WRITE_ONCE(sk->sk_sndbuf, ssk->sk_sndbuf); + return true; +} + +static inline void mptcp_write_space(struct sock *sk) +{ + if (sk_stream_is_writeable(sk)) { + /* pairs with memory barrier in mptcp_poll */ + smp_mb(); + if (test_and_clear_bit(MPTCP_NOSPACE, &mptcp_sk(sk)->flags)) + sk_stream_write_space(sk); + } +} + void mptcp_destroy_common(struct mptcp_sock *msk); void __init mptcp_token_init(void); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 278cbe3e539e..586156281e5a 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -18,12 +18,15 @@ #include <net/tcp.h> #if IS_ENABLED(CONFIG_MPTCP_IPV6) #include <net/ip6_route.h> +#include <net/transp_v6.h> #endif #include <net/mptcp.h> #include <uapi/linux/mptcp.h> #include "protocol.h" #include "mib.h" +static void mptcp_subflow_ops_undo_override(struct sock *ssk); + static void SUBFLOW_REQ_INC_STATS(struct request_sock *req, enum linux_mptcp_mib_field field) { @@ -343,6 +346,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) if (subflow->conn_finished) return; + mptcp_propagate_sndbuf(parent, sk); subflow->rel_write_seq = 1; subflow->conn_finished = 1; subflow->ssn_offset = TCP_SKB_CB(skb)->seq; @@ -427,6 +431,7 @@ drop: static struct tcp_request_sock_ops subflow_request_sock_ipv6_ops; static struct inet_connection_sock_af_ops subflow_v6_specific; static struct inet_connection_sock_af_ops subflow_v6m_specific; +static struct proto tcpv6_prot_override; static int subflow_v6_conn_request(struct sock *sk, struct sk_buff *skb) { @@ -508,6 +513,8 @@ static void subflow_ulp_fallback(struct sock *sk, icsk->icsk_ulp_ops = NULL; rcu_assign_pointer(icsk->icsk_ulp_data, NULL); tcp_sk(sk)->is_mptcp = 0; + + mptcp_subflow_ops_undo_override(sk); } static void subflow_drop_ctx(struct sock *ssk) @@ -681,6 +688,7 @@ dispose_child: } static struct inet_connection_sock_af_ops subflow_specific; +static struct proto tcp_prot_override; enum mapping_status { MAPPING_OK, @@ -1040,7 +1048,10 @@ static void subflow_data_ready(struct sock *sk) static void subflow_write_space(struct sock *ssk) { - /* we take action in __mptcp_clean_una() */ + struct sock *sk = mptcp_subflow_ctx(ssk)->conn; + + mptcp_propagate_sndbuf(sk, ssk); + mptcp_write_space(sk); } static struct inet_connection_sock_af_ops * @@ -1074,21 +1085,31 @@ void mptcpv6_handle_mapped(struct sock *sk, bool mapped) #endif static void mptcp_info2sockaddr(const struct mptcp_addr_info *info, - struct sockaddr_storage *addr) + struct sockaddr_storage *addr, + unsigned short family) { memset(addr, 0, sizeof(*addr)); - addr->ss_family = info->family; + addr->ss_family = family; if (addr->ss_family == AF_INET) { struct sockaddr_in *in_addr = (struct sockaddr_in *)addr; - in_addr->sin_addr = info->addr; + if (info->family == AF_INET) + in_addr->sin_addr = info->addr; +#if IS_ENABLED(CONFIG_MPTCP_IPV6) + else if (ipv6_addr_v4mapped(&info->addr6)) + in_addr->sin_addr.s_addr = info->addr6.s6_addr32[3]; +#endif in_addr->sin_port = info->port; } #if IS_ENABLED(CONFIG_MPTCP_IPV6) else if (addr->ss_family == AF_INET6) { struct sockaddr_in6 *in6_addr = (struct sockaddr_in6 *)addr; - in6_addr->sin6_addr = info->addr6; + if (info->family == AF_INET) + ipv6_addr_set_v4mapped(info->addr.s_addr, + &in6_addr->sin6_addr); + else + in6_addr->sin6_addr = info->addr6; in6_addr->sin6_port = info->port; } #endif @@ -1132,11 +1153,11 @@ int __mptcp_subflow_connect(struct sock *sk, const struct mptcp_addr_info *loc, subflow->remote_key = msk->remote_key; subflow->local_key = msk->local_key; subflow->token = msk->token; - mptcp_info2sockaddr(loc, &addr); + mptcp_info2sockaddr(loc, &addr, ssk->sk_family); addrlen = sizeof(struct sockaddr_in); #if IS_ENABLED(CONFIG_MPTCP_IPV6) - if (loc->family == AF_INET6) + if (addr.ss_family == AF_INET6) addrlen = sizeof(struct sockaddr_in6); #endif ssk->sk_bound_dev_if = loc->ifindex; @@ -1152,13 +1173,16 @@ int __mptcp_subflow_connect(struct sock *sk, const struct mptcp_addr_info *loc, subflow->remote_id = remote_id; subflow->request_join = 1; subflow->request_bkup = !!(loc->flags & MPTCP_PM_ADDR_FLAG_BACKUP); - mptcp_info2sockaddr(remote, &addr); + mptcp_info2sockaddr(remote, &addr, ssk->sk_family); mptcp_add_pending_subflow(msk, subflow); err = kernel_connect(sf, (struct sockaddr *)&addr, addrlen, O_NONBLOCK); if (err && err != -EINPROGRESS) goto failed_unlink; + /* discard the subflow socket */ + mptcp_sock_graft(ssk, sk->sk_socket); + iput(SOCK_INODE(sf)); return err; failed_unlink: @@ -1196,6 +1220,25 @@ static void mptcp_attach_cgroup(struct sock *parent, struct sock *child) #endif /* CONFIG_SOCK_CGROUP_DATA */ } +static void mptcp_subflow_ops_override(struct sock *ssk) +{ +#if IS_ENABLED(CONFIG_MPTCP_IPV6) + if (ssk->sk_prot == &tcpv6_prot) + ssk->sk_prot = &tcpv6_prot_override; + else +#endif + ssk->sk_prot = &tcp_prot_override; +} + +static void mptcp_subflow_ops_undo_override(struct sock *ssk) +{ +#if IS_ENABLED(CONFIG_MPTCP_IPV6) + if (ssk->sk_prot == &tcpv6_prot_override) + ssk->sk_prot = &tcpv6_prot; + else +#endif + ssk->sk_prot = &tcp_prot; +} int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock) { struct mptcp_subflow_context *subflow; @@ -1251,6 +1294,7 @@ int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock) *new_sock = sf; sock_hold(sk); subflow->conn = sk; + mptcp_subflow_ops_override(sf->sk); return 0; } @@ -1267,6 +1311,7 @@ static struct mptcp_subflow_context *subflow_create_ctx(struct sock *sk, rcu_assign_pointer(icsk->icsk_ulp_data, ctx); INIT_LIST_HEAD(&ctx->node); + INIT_LIST_HEAD(&ctx->delegated_node); pr_debug("subflow=%p", ctx); @@ -1299,6 +1344,7 @@ static void subflow_state_change(struct sock *sk) __subflow_state_change(sk); if (subflow_simultaneous_connect(sk)) { + mptcp_propagate_sndbuf(parent, sk); mptcp_do_fallback(sk); mptcp_rcv_space_init(mptcp_sk(parent), sk); pr_fallback(mptcp_sk(parent)); @@ -1378,6 +1424,7 @@ static void subflow_ulp_release(struct sock *ssk) sock_put(sk); } + mptcp_subflow_ops_undo_override(ssk); if (release) kfree_rcu(ctx, rcu); } @@ -1431,6 +1478,16 @@ static void subflow_ulp_clone(const struct request_sock *req, } } +static void tcp_release_cb_override(struct sock *ssk) +{ + struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); + + if (mptcp_subflow_has_delegated_action(subflow)) + mptcp_subflow_process_delegated(ssk); + + tcp_release_cb(ssk); +} + static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = { .name = "mptcp", .owner = THIS_MODULE, @@ -1471,6 +1528,9 @@ void __init mptcp_subflow_init(void) subflow_specific.syn_recv_sock = subflow_syn_recv_sock; subflow_specific.sk_rx_dst_set = subflow_finish_connect; + tcp_prot_override = tcp_prot; + tcp_prot_override.release_cb = tcp_release_cb_override; + #if IS_ENABLED(CONFIG_MPTCP_IPV6) subflow_request_sock_ipv6_ops = tcp_request_sock_ipv6_ops; subflow_request_sock_ipv6_ops.route_req = subflow_v6_route_req; @@ -1486,6 +1546,9 @@ void __init mptcp_subflow_init(void) subflow_v6m_specific.net_header_len = ipv4_specific.net_header_len; subflow_v6m_specific.mtu_reduced = ipv4_specific.mtu_reduced; subflow_v6m_specific.net_frag_header_len = 0; + + tcpv6_prot_override = tcpv6_prot; + tcpv6_prot_override.release_cb = tcp_release_cb_override; #endif mptcp_diag_subflow_init(&subflow_ulp_ops); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 49fbef0d99be..1a92063c73a4 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only menu "Core Netfilter Configuration" - depends on NET && INET && NETFILTER + depends on INET && NETFILTER config NETFILTER_INGRESS bool "Netfilter ingress support" diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig index eb0e329f9b8d..c39a1e35c104 100644 --- a/net/netfilter/ipvs/Kconfig +++ b/net/netfilter/ipvs/Kconfig @@ -4,7 +4,7 @@ # menuconfig IP_VS tristate "IP virtual server support" - depends on NET && INET && NETFILTER + depends on INET && NETFILTER depends on (NF_CONNTRACK || NF_CONNTRACK=n) help IP Virtual Server support will let you build a high-performance diff --git a/net/nfc/Kconfig b/net/nfc/Kconfig index 96b91674dd37..466a0279b93e 100644 --- a/net/nfc/Kconfig +++ b/net/nfc/Kconfig @@ -4,7 +4,6 @@ # menuconfig NFC - depends on NET depends on RFKILL || !RFKILL tristate "NFC subsystem support" default n diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c index 79bebf4b0796..59257400697d 100644 --- a/net/nfc/nci/core.c +++ b/net/nfc/nci/core.c @@ -508,7 +508,7 @@ static int nci_open_device(struct nci_dev *ndev) }; unsigned long opt = 0; - if (!(ndev->nci_ver & NCI_VER_2_MASK)) + if (ndev->nci_ver & NCI_VER_2_MASK) opt = (unsigned long)&nci_init_v2_cmd; rc = __nci_request(ndev, nci_init_req, opt, diff --git a/net/psample/Kconfig b/net/psample/Kconfig index 028f514a9c60..be0b839209ba 100644 --- a/net/psample/Kconfig +++ b/net/psample/Kconfig @@ -4,7 +4,6 @@ # menuconfig PSAMPLE - depends on NET tristate "Packet-sampling netlink channel" default n help diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c index 83a5c6722a06..b3442078aabc 100644 --- a/net/sched/act_ct.c +++ b/net/sched/act_ct.c @@ -1030,6 +1030,7 @@ out_push: out: tcf_action_update_bstats(&c->common, skb); + qdisc_skb_cb(skb)->post_ct = true; if (defrag) qdisc_skb_cb(skb)->pkt_len = skb->len; return retval; diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 1319986693fc..4a9297a89c77 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -302,6 +302,7 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res) { struct cls_fl_head *head = rcu_dereference_bh(tp->root); + bool post_ct = qdisc_skb_cb(skb)->post_ct; struct fl_flow_key skb_key; struct fl_flow_mask *mask; struct cls_fl_filter *f; @@ -318,7 +319,8 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp, skb_flow_dissect_tunnel_info(skb, &mask->dissector, &skb_key); skb_flow_dissect_ct(skb, &mask->dissector, &skb_key, fl_ct_info_to_flower_map, - ARRAY_SIZE(fl_ct_info_to_flower_map)); + ARRAY_SIZE(fl_ct_info_to_flower_map), + post_ct); skb_flow_dissect_hash(skb, &mask->dissector, &skb_key); skb_flow_dissect(skb, &mask->dissector, &skb_key, 0); @@ -1272,6 +1274,10 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, nla_opt_msk = nla_data(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]); msk_depth = nla_len(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]); + if (!nla_ok(nla_opt_msk, msk_depth)) { + NL_SET_ERR_MSG(extack, "Invalid nested attribute for masks"); + return -EINVAL; + } } nla_for_each_attr(nla_opt_key, nla_enc_key, @@ -1307,9 +1313,6 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, NL_SET_ERR_MSG(extack, "Key and mask miss aligned"); return -EINVAL; } - - if (msk_depth) - nla_opt_msk = nla_next(nla_opt_msk, &msk_depth); break; case TCA_FLOWER_KEY_ENC_OPTS_VXLAN: if (key->enc_opts.dst_opt_type) { @@ -1340,9 +1343,6 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, NL_SET_ERR_MSG(extack, "Key and mask miss aligned"); return -EINVAL; } - - if (msk_depth) - nla_opt_msk = nla_next(nla_opt_msk, &msk_depth); break; case TCA_FLOWER_KEY_ENC_OPTS_ERSPAN: if (key->enc_opts.dst_opt_type) { @@ -1373,14 +1373,20 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, NL_SET_ERR_MSG(extack, "Key and mask miss aligned"); return -EINVAL; } - - if (msk_depth) - nla_opt_msk = nla_next(nla_opt_msk, &msk_depth); break; default: NL_SET_ERR_MSG(extack, "Unknown tunnel option type"); return -EINVAL; } + + if (!msk_depth) + continue; + + if (!nla_ok(nla_opt_msk, msk_depth)) { + NL_SET_ERR_MSG(extack, "A mask attribute is invalid"); + return -EINVAL; + } + nla_opt_msk = nla_next(nla_opt_msk, &msk_depth); } return 0; diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c index 78bec347b8b6..c4007b9cd16d 100644 --- a/net/sched/cls_tcindex.c +++ b/net/sched/cls_tcindex.c @@ -366,9 +366,13 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, if (tb[TCA_TCINDEX_MASK]) cp->mask = nla_get_u16(tb[TCA_TCINDEX_MASK]); - if (tb[TCA_TCINDEX_SHIFT]) + if (tb[TCA_TCINDEX_SHIFT]) { cp->shift = nla_get_u32(tb[TCA_TCINDEX_SHIFT]); - + if (cp->shift > 16) { + err = -EINVAL; + goto errout; + } + } if (!cp->hash) { /* Hash not specified, use perfect hash if the upper limit * of the hashing index is below the threshold. diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 51cb553e4317..e2e4353db8a7 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -412,7 +412,8 @@ struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, { struct qdisc_rate_table *rtab; - if (tab == NULL || r->rate == 0 || r->cell_log == 0 || + if (tab == NULL || r->rate == 0 || + r->cell_log == 0 || r->cell_log >= 32 || nla_len(tab) != TC_RTAB_SIZE) { NL_SET_ERR_MSG(extack, "Invalid rate table parameters for searching"); return NULL; @@ -1865,7 +1866,8 @@ static int tclass_notify(struct net *net, struct sk_buff *oskb, static int tclass_del_notify(struct net *net, const struct Qdisc_class_ops *cops, struct sk_buff *oskb, struct nlmsghdr *n, - struct Qdisc *q, unsigned long cl) + struct Qdisc *q, unsigned long cl, + struct netlink_ext_ack *extack) { u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; struct sk_buff *skb; @@ -1884,7 +1886,7 @@ static int tclass_del_notify(struct net *net, return -EINVAL; } - err = cops->delete(q, cl); + err = cops->delete(q, cl, extack); if (err) { kfree_skb(skb); return err; @@ -2087,7 +2089,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, goto out; break; case RTM_DELTCLASS: - err = tclass_del_notify(net, cops, skb, n, q, cl); + err = tclass_del_notify(net, cops, skb, n, q, cl, extack); /* Unbind the class with flilters with 0 */ tc_bind_tclass(q, portid, clid, 0); goto out; diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 007bd2d9f1ff..d0c9a57398fc 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -320,7 +320,8 @@ err_out: return error; } -static int atm_tc_delete(struct Qdisc *sch, unsigned long arg) +static int atm_tc_delete(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) { struct atm_qdisc_data *p = qdisc_priv(sch); struct atm_flow_data *flow = (struct atm_flow_data *)arg; diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 53d45e029c36..320b3d31fa97 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1675,7 +1675,8 @@ failure: return err; } -static int cbq_delete(struct Qdisc *sch, unsigned long arg) +static int cbq_delete(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) { struct cbq_sched_data *q = qdisc_priv(sch); struct cbq_class *cl = (struct cbq_class *)arg; diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c index dde564670ad8..fc1e47069593 100644 --- a/net/sched/sch_drr.c +++ b/net/sched/sch_drr.c @@ -146,7 +146,8 @@ static void drr_destroy_class(struct Qdisc *sch, struct drr_class *cl) kfree(cl); } -static int drr_delete_class(struct Qdisc *sch, unsigned long arg) +static int drr_delete_class(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) { struct drr_sched *q = qdisc_priv(sch); struct drr_class *cl = (struct drr_class *)arg; diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 2b88710994d7..cd2748e2d4a2 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -150,7 +150,8 @@ errout: return err; } -static int dsmark_delete(struct Qdisc *sch, unsigned long arg) +static int dsmark_delete(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) { struct dsmark_qdisc_data *p = qdisc_priv(sch); diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index d1902fca9844..bf0034c66e35 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1090,7 +1090,8 @@ hfsc_destroy_class(struct Qdisc *sch, struct hfsc_class *cl) } static int -hfsc_delete_class(struct Qdisc *sch, unsigned long arg) +hfsc_delete_class(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) { struct hfsc_sched *q = qdisc_priv(sch); struct hfsc_class *cl = (struct hfsc_class *)arg; diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index cd70dbcbd72f..dff3adf5a915 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -114,6 +114,7 @@ struct htb_class { * Written often fields */ struct gnet_stats_basic_packed bstats; + struct gnet_stats_basic_packed bstats_bias; struct tc_htb_xstats xstats; /* our special stats */ /* token bucket parameters */ @@ -174,6 +175,11 @@ struct htb_sched { int row_mask[TC_HTB_MAXDEPTH]; struct htb_level hlevel[TC_HTB_MAXDEPTH]; + + struct Qdisc **direct_qdiscs; + unsigned int num_direct_qdiscs; + + bool offload; }; /* find class in global hash table using given handle */ @@ -957,7 +963,7 @@ static void htb_reset(struct Qdisc *sch) if (cl->level) memset(&cl->inner, 0, sizeof(cl->inner)); else { - if (cl->leaf.q) + if (cl->leaf.q && !q->offload) qdisc_reset(cl->leaf.q); } cl->prio_activity = 0; @@ -980,6 +986,7 @@ static const struct nla_policy htb_policy[TCA_HTB_MAX + 1] = { [TCA_HTB_DIRECT_QLEN] = { .type = NLA_U32 }, [TCA_HTB_RATE64] = { .type = NLA_U64 }, [TCA_HTB_CEIL64] = { .type = NLA_U64 }, + [TCA_HTB_OFFLOAD] = { .type = NLA_FLAG }, }; static void htb_work_func(struct work_struct *work) @@ -992,12 +999,27 @@ static void htb_work_func(struct work_struct *work) rcu_read_unlock(); } +static void htb_set_lockdep_class_child(struct Qdisc *q) +{ + static struct lock_class_key child_key; + + lockdep_set_class(qdisc_lock(q), &child_key); +} + +static int htb_offload(struct net_device *dev, struct tc_htb_qopt_offload *opt) +{ + return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_HTB, opt); +} + static int htb_init(struct Qdisc *sch, struct nlattr *opt, struct netlink_ext_ack *extack) { + struct net_device *dev = qdisc_dev(sch); + struct tc_htb_qopt_offload offload_opt; struct htb_sched *q = qdisc_priv(sch); struct nlattr *tb[TCA_HTB_MAX + 1]; struct tc_htb_glob *gopt; + unsigned int ntx; int err; qdisc_watchdog_init(&q->watchdog, sch); @@ -1022,9 +1044,26 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt, if (gopt->version != HTB_VER >> 16) return -EINVAL; + q->offload = nla_get_flag(tb[TCA_HTB_OFFLOAD]); + + if (q->offload) { + if (sch->parent != TC_H_ROOT) + return -EOPNOTSUPP; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return -EOPNOTSUPP; + + q->num_direct_qdiscs = dev->real_num_tx_queues; + q->direct_qdiscs = kcalloc(q->num_direct_qdiscs, + sizeof(*q->direct_qdiscs), + GFP_KERNEL); + if (!q->direct_qdiscs) + return -ENOMEM; + } + err = qdisc_class_hash_init(&q->clhash); if (err < 0) - return err; + goto err_free_direct_qdiscs; qdisc_skb_head_init(&q->direct_queue); @@ -1037,7 +1076,107 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt, q->rate2quantum = 1; q->defcls = gopt->defcls; + if (!q->offload) + return 0; + + for (ntx = 0; ntx < q->num_direct_qdiscs; ntx++) { + struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, ntx); + struct Qdisc *qdisc; + + qdisc = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, + TC_H_MAKE(sch->handle, 0), extack); + if (!qdisc) { + err = -ENOMEM; + goto err_free_qdiscs; + } + + htb_set_lockdep_class_child(qdisc); + q->direct_qdiscs[ntx] = qdisc; + qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + } + + sch->flags |= TCQ_F_MQROOT; + + offload_opt = (struct tc_htb_qopt_offload) { + .command = TC_HTB_CREATE, + .parent_classid = TC_H_MAJ(sch->handle) >> 16, + .classid = TC_H_MIN(q->defcls), + .extack = extack, + }; + err = htb_offload(dev, &offload_opt); + if (err) + goto err_free_qdiscs; + return 0; + +err_free_qdiscs: + /* TC_HTB_CREATE call failed, avoid any further calls to the driver. */ + q->offload = false; + + for (ntx = 0; ntx < q->num_direct_qdiscs && q->direct_qdiscs[ntx]; + ntx++) + qdisc_put(q->direct_qdiscs[ntx]); + + qdisc_class_hash_destroy(&q->clhash); + /* Prevent use-after-free and double-free when htb_destroy gets called. + */ + q->clhash.hash = NULL; + q->clhash.hashsize = 0; + +err_free_direct_qdiscs: + kfree(q->direct_qdiscs); + q->direct_qdiscs = NULL; + return err; +} + +static void htb_attach_offload(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct htb_sched *q = qdisc_priv(sch); + unsigned int ntx; + + for (ntx = 0; ntx < q->num_direct_qdiscs; ntx++) { + struct Qdisc *old, *qdisc = q->direct_qdiscs[ntx]; + + old = dev_graft_qdisc(qdisc->dev_queue, qdisc); + qdisc_put(old); + qdisc_hash_add(qdisc, false); + } + for (ntx = q->num_direct_qdiscs; ntx < dev->num_tx_queues; ntx++) { + struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, ntx); + struct Qdisc *old = dev_graft_qdisc(dev_queue, NULL); + + qdisc_put(old); + } + + kfree(q->direct_qdiscs); + q->direct_qdiscs = NULL; +} + +static void htb_attach_software(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + unsigned int ntx; + + /* Resemble qdisc_graft behavior. */ + for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { + struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, ntx); + struct Qdisc *old = dev_graft_qdisc(dev_queue, sch); + + qdisc_refcount_inc(sch); + + qdisc_put(old); + } +} + +static void htb_attach(struct Qdisc *sch) +{ + struct htb_sched *q = qdisc_priv(sch); + + if (q->offload) + htb_attach_offload(sch); + else + htb_attach_software(sch); } static int htb_dump(struct Qdisc *sch, struct sk_buff *skb) @@ -1046,6 +1185,11 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb) struct nlattr *nest; struct tc_htb_glob gopt; + if (q->offload) + sch->flags |= TCQ_F_OFFLOADED; + else + sch->flags &= ~TCQ_F_OFFLOADED; + sch->qstats.overlimits = q->overlimits; /* Its safe to not acquire qdisc lock. As we hold RTNL, * no change can happen on the qdisc parameters. @@ -1063,6 +1207,8 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb) if (nla_put(skb, TCA_HTB_INIT, sizeof(gopt), &gopt) || nla_put_u32(skb, TCA_HTB_DIRECT_QLEN, q->direct_qlen)) goto nla_put_failure; + if (q->offload && nla_put_flag(skb, TCA_HTB_OFFLOAD)) + goto nla_put_failure; return nla_nest_end(skb, nest); @@ -1075,6 +1221,7 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg, struct sk_buff *skb, struct tcmsg *tcm) { struct htb_class *cl = (struct htb_class *)arg; + struct htb_sched *q = qdisc_priv(sch); struct nlattr *nest; struct tc_htb_opt opt; @@ -1101,6 +1248,8 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg, opt.level = cl->level; if (nla_put(skb, TCA_HTB_PARMS, sizeof(opt), &opt)) goto nla_put_failure; + if (q->offload && nla_put_flag(skb, TCA_HTB_OFFLOAD)) + goto nla_put_failure; if ((cl->rate.rate_bytes_ps >= (1ULL << 32)) && nla_put_u64_64bit(skb, TCA_HTB_RATE64, cl->rate.rate_bytes_ps, TCA_HTB_PAD)) @@ -1117,10 +1266,39 @@ nla_put_failure: return -1; } +static void htb_offload_aggregate_stats(struct htb_sched *q, + struct htb_class *cl) +{ + struct htb_class *c; + unsigned int i; + + memset(&cl->bstats, 0, sizeof(cl->bstats)); + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(c, &q->clhash.hash[i], common.hnode) { + struct htb_class *p = c; + + while (p && p->level < cl->level) + p = p->parent; + + if (p != cl) + continue; + + cl->bstats.bytes += c->bstats_bias.bytes; + cl->bstats.packets += c->bstats_bias.packets; + if (c->level == 0) { + cl->bstats.bytes += c->leaf.q->bstats.bytes; + cl->bstats.packets += c->leaf.q->bstats.packets; + } + } + } +} + static int htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d) { struct htb_class *cl = (struct htb_class *)arg; + struct htb_sched *q = qdisc_priv(sch); struct gnet_stats_queue qs = { .drops = cl->drops, .overlimits = cl->overlimits, @@ -1135,6 +1313,19 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d) cl->xstats.ctokens = clamp_t(s64, PSCHED_NS2TICKS(cl->ctokens), INT_MIN, INT_MAX); + if (q->offload) { + if (!cl->level) { + if (cl->leaf.q) + cl->bstats = cl->leaf.q->bstats; + else + memset(&cl->bstats, 0, sizeof(cl->bstats)); + cl->bstats.bytes += cl->bstats_bias.bytes; + cl->bstats.packets += cl->bstats_bias.packets; + } else { + htb_offload_aggregate_stats(q, cl); + } + } + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d, NULL, &cl->bstats) < 0 || gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || @@ -1144,19 +1335,97 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d) return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats)); } +static struct netdev_queue * +htb_select_queue(struct Qdisc *sch, struct tcmsg *tcm) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_htb_qopt_offload offload_opt; + int err; + + offload_opt = (struct tc_htb_qopt_offload) { + .command = TC_HTB_LEAF_QUERY_QUEUE, + .classid = TC_H_MIN(tcm->tcm_parent), + }; + err = htb_offload(dev, &offload_opt); + if (err || offload_opt.qid >= dev->num_tx_queues) + return NULL; + return netdev_get_tx_queue(dev, offload_opt.qid); +} + +static struct Qdisc * +htb_graft_helper(struct netdev_queue *dev_queue, struct Qdisc *new_q) +{ + struct net_device *dev = dev_queue->dev; + struct Qdisc *old_q; + + if (dev->flags & IFF_UP) + dev_deactivate(dev); + old_q = dev_graft_qdisc(dev_queue, new_q); + if (new_q) + new_q->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + if (dev->flags & IFF_UP) + dev_activate(dev); + + return old_q; +} + +static void htb_offload_move_qdisc(struct Qdisc *sch, u16 qid_old, u16 qid_new) +{ + struct netdev_queue *queue_old, *queue_new; + struct net_device *dev = qdisc_dev(sch); + struct Qdisc *qdisc; + + queue_old = netdev_get_tx_queue(dev, qid_old); + queue_new = netdev_get_tx_queue(dev, qid_new); + + if (dev->flags & IFF_UP) + dev_deactivate(dev); + qdisc = dev_graft_qdisc(queue_old, NULL); + qdisc->dev_queue = queue_new; + qdisc = dev_graft_qdisc(queue_new, qdisc); + if (dev->flags & IFF_UP) + dev_activate(dev); + + WARN_ON(!(qdisc->flags & TCQ_F_BUILTIN)); +} + static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, struct Qdisc **old, struct netlink_ext_ack *extack) { + struct netdev_queue *dev_queue = sch->dev_queue; struct htb_class *cl = (struct htb_class *)arg; + struct htb_sched *q = qdisc_priv(sch); + struct Qdisc *old_q; if (cl->level) return -EINVAL; - if (new == NULL && - (new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, - cl->common.classid, extack)) == NULL) - return -ENOBUFS; + + if (q->offload) { + dev_queue = new->dev_queue; + WARN_ON(dev_queue != cl->leaf.q->dev_queue); + } + + if (!new) { + new = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, + cl->common.classid, extack); + if (!new) + return -ENOBUFS; + } + + if (q->offload) { + htb_set_lockdep_class_child(new); + /* One ref for cl->leaf.q, the other for dev_queue->qdisc. */ + qdisc_refcount_inc(new); + old_q = htb_graft_helper(dev_queue, new); + } *old = qdisc_replace(sch, new, &cl->leaf.q); + + if (q->offload) { + WARN_ON(old_q != *old); + qdisc_put(old_q); + } + return 0; } @@ -1184,9 +1453,10 @@ static inline int htb_parent_last_child(struct htb_class *cl) return 1; } -static void htb_parent_to_leaf(struct htb_sched *q, struct htb_class *cl, +static void htb_parent_to_leaf(struct Qdisc *sch, struct htb_class *cl, struct Qdisc *new_q) { + struct htb_sched *q = qdisc_priv(sch); struct htb_class *parent = cl->parent; WARN_ON(cl->level || !cl->leaf.q || cl->prio_activity); @@ -1204,6 +1474,76 @@ static void htb_parent_to_leaf(struct htb_sched *q, struct htb_class *cl, parent->cmode = HTB_CAN_SEND; } +static void htb_parent_to_leaf_offload(struct Qdisc *sch, + struct netdev_queue *dev_queue, + struct Qdisc *new_q) +{ + struct Qdisc *old_q; + + /* One ref for cl->leaf.q, the other for dev_queue->qdisc. */ + qdisc_refcount_inc(new_q); + old_q = htb_graft_helper(dev_queue, new_q); + WARN_ON(!(old_q->flags & TCQ_F_BUILTIN)); +} + +static int htb_destroy_class_offload(struct Qdisc *sch, struct htb_class *cl, + bool last_child, bool destroying, + struct netlink_ext_ack *extack) +{ + struct tc_htb_qopt_offload offload_opt; + struct Qdisc *q = cl->leaf.q; + struct Qdisc *old = NULL; + int err; + + if (cl->level) + return -EINVAL; + + WARN_ON(!q); + if (!destroying) { + /* On destroy of HTB, two cases are possible: + * 1. q is a normal qdisc, but q->dev_queue has noop qdisc. + * 2. q is a noop qdisc (for nodes that were inner), + * q->dev_queue is noop_netdev_queue. + */ + old = htb_graft_helper(q->dev_queue, NULL); + WARN_ON(!old); + WARN_ON(old != q); + } + + if (cl->parent) { + cl->parent->bstats_bias.bytes += q->bstats.bytes; + cl->parent->bstats_bias.packets += q->bstats.packets; + } + + offload_opt = (struct tc_htb_qopt_offload) { + .command = !last_child ? TC_HTB_LEAF_DEL : + destroying ? TC_HTB_LEAF_DEL_LAST_FORCE : + TC_HTB_LEAF_DEL_LAST, + .classid = cl->common.classid, + .extack = extack, + }; + err = htb_offload(qdisc_dev(sch), &offload_opt); + + if (!err || destroying) + qdisc_put(old); + else + htb_graft_helper(q->dev_queue, old); + + if (last_child) + return err; + + if (!err && offload_opt.moved_qid != 0) { + if (destroying) + q->dev_queue = netdev_get_tx_queue(qdisc_dev(sch), + offload_opt.qid); + else + htb_offload_move_qdisc(sch, offload_opt.moved_qid, + offload_opt.qid); + } + + return err; +} + static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl) { if (!cl->level) { @@ -1217,8 +1557,11 @@ static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl) static void htb_destroy(struct Qdisc *sch) { + struct net_device *dev = qdisc_dev(sch); + struct tc_htb_qopt_offload offload_opt; struct htb_sched *q = qdisc_priv(sch); struct hlist_node *next; + bool nonempty, changed; struct htb_class *cl; unsigned int i; @@ -1237,21 +1580,68 @@ static void htb_destroy(struct Qdisc *sch) cl->block = NULL; } } - for (i = 0; i < q->clhash.hashsize; i++) { - hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], - common.hnode) - htb_destroy_class(sch, cl); - } + + do { + nonempty = false; + changed = false; + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], + common.hnode) { + bool last_child; + + if (!q->offload) { + htb_destroy_class(sch, cl); + continue; + } + + nonempty = true; + + if (cl->level) + continue; + + changed = true; + + last_child = htb_parent_last_child(cl); + htb_destroy_class_offload(sch, cl, last_child, + true, NULL); + qdisc_class_hash_remove(&q->clhash, + &cl->common); + if (cl->parent) + cl->parent->children--; + if (last_child) + htb_parent_to_leaf(sch, cl, NULL); + htb_destroy_class(sch, cl); + } + } + } while (changed); + WARN_ON(nonempty); + qdisc_class_hash_destroy(&q->clhash); __qdisc_reset_queue(&q->direct_queue); + + if (!q->offload) + return; + + offload_opt = (struct tc_htb_qopt_offload) { + .command = TC_HTB_DESTROY, + }; + htb_offload(dev, &offload_opt); + + if (!q->direct_qdiscs) + return; + for (i = 0; i < q->num_direct_qdiscs && q->direct_qdiscs[i]; i++) + qdisc_put(q->direct_qdiscs[i]); + kfree(q->direct_qdiscs); } -static int htb_delete(struct Qdisc *sch, unsigned long arg) +static int htb_delete(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) { struct htb_sched *q = qdisc_priv(sch); struct htb_class *cl = (struct htb_class *)arg; struct Qdisc *new_q = NULL; int last_child = 0; + int err; /* TODO: why don't allow to delete subtree ? references ? does * tc subsys guarantee us that in htb_destroy it holds no class @@ -1260,11 +1650,28 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg) if (cl->children || cl->filter_cnt) return -EBUSY; - if (!cl->level && htb_parent_last_child(cl)) { - new_q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + if (!cl->level && htb_parent_last_child(cl)) + last_child = 1; + + if (q->offload) { + err = htb_destroy_class_offload(sch, cl, last_child, false, + extack); + if (err) + return err; + } + + if (last_child) { + struct netdev_queue *dev_queue; + + dev_queue = q->offload ? cl->leaf.q->dev_queue : sch->dev_queue; + new_q = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, cl->parent->common.classid, NULL); - last_child = 1; + if (q->offload) { + if (new_q) + htb_set_lockdep_class_child(new_q); + htb_parent_to_leaf_offload(sch, dev_queue, new_q); + } } sch_tree_lock(sch); @@ -1285,7 +1692,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg) &q->hlevel[cl->level].wait_pq); if (last_child) - htb_parent_to_leaf(q, cl, new_q); + htb_parent_to_leaf(sch, cl, new_q); sch_tree_unlock(sch); @@ -1300,9 +1707,11 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, int err = -EINVAL; struct htb_sched *q = qdisc_priv(sch); struct htb_class *cl = (struct htb_class *)*arg, *parent; + struct tc_htb_qopt_offload offload_opt; struct nlattr *opt = tca[TCA_OPTIONS]; struct nlattr *tb[TCA_HTB_MAX + 1]; struct Qdisc *parent_qdisc = NULL; + struct netdev_queue *dev_queue; struct tc_htb_opt *hopt; u64 rate64, ceil64; int warn = 0; @@ -1335,8 +1744,12 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, qdisc_put_rtab(qdisc_get_rtab(&hopt->ceil, tb[TCA_HTB_CTAB], NULL)); + rate64 = tb[TCA_HTB_RATE64] ? nla_get_u64(tb[TCA_HTB_RATE64]) : 0; + ceil64 = tb[TCA_HTB_CEIL64] ? nla_get_u64(tb[TCA_HTB_CEIL64]) : 0; + if (!cl) { /* new class */ - struct Qdisc *new_q; + struct net_device *dev = qdisc_dev(sch); + struct Qdisc *new_q, *old_q; int prio; struct { struct nlattr nla; @@ -1379,11 +1792,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, NULL, qdisc_root_sleeping_running(sch), tca[TCA_RATE] ? : &est.nla); - if (err) { - tcf_block_put(cl->block); - kfree(cl); - goto failure; - } + if (err) + goto err_block_put; } cl->children = 0; @@ -1392,12 +1802,76 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, for (prio = 0; prio < TC_HTB_NUMPRIO; prio++) RB_CLEAR_NODE(&cl->node[prio]); + cl->common.classid = classid; + + /* Make sure nothing interrupts us in between of two + * ndo_setup_tc calls. + */ + ASSERT_RTNL(); + /* create leaf qdisc early because it uses kmalloc(GFP_KERNEL) * so that can't be used inside of sch_tree_lock * -- thanks to Karlis Peisenieks */ - new_q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + if (!q->offload) { + dev_queue = sch->dev_queue; + } else if (!(parent && !parent->level)) { + /* Assign a dev_queue to this classid. */ + offload_opt = (struct tc_htb_qopt_offload) { + .command = TC_HTB_LEAF_ALLOC_QUEUE, + .classid = cl->common.classid, + .parent_classid = parent ? + TC_H_MIN(parent->common.classid) : + TC_HTB_CLASSID_ROOT, + .rate = max_t(u64, hopt->rate.rate, rate64), + .ceil = max_t(u64, hopt->ceil.rate, ceil64), + .extack = extack, + }; + err = htb_offload(dev, &offload_opt); + if (err) { + pr_err("htb: TC_HTB_LEAF_ALLOC_QUEUE failed with err = %d\n", + err); + goto err_kill_estimator; + } + dev_queue = netdev_get_tx_queue(dev, offload_opt.qid); + } else { /* First child. */ + dev_queue = parent->leaf.q->dev_queue; + old_q = htb_graft_helper(dev_queue, NULL); + WARN_ON(old_q != parent->leaf.q); + offload_opt = (struct tc_htb_qopt_offload) { + .command = TC_HTB_LEAF_TO_INNER, + .classid = cl->common.classid, + .parent_classid = + TC_H_MIN(parent->common.classid), + .rate = max_t(u64, hopt->rate.rate, rate64), + .ceil = max_t(u64, hopt->ceil.rate, ceil64), + .extack = extack, + }; + err = htb_offload(dev, &offload_opt); + if (err) { + pr_err("htb: TC_HTB_LEAF_TO_INNER failed with err = %d\n", + err); + htb_graft_helper(dev_queue, old_q); + goto err_kill_estimator; + } + parent->bstats_bias.bytes += old_q->bstats.bytes; + parent->bstats_bias.packets += old_q->bstats.packets; + qdisc_put(old_q); + } + new_q = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, classid, NULL); + if (q->offload) { + if (new_q) { + htb_set_lockdep_class_child(new_q); + /* One ref for cl->leaf.q, the other for + * dev_queue->qdisc. + */ + qdisc_refcount_inc(new_q); + } + old_q = htb_graft_helper(dev_queue, new_q); + /* No qdisc_put needed. */ + WARN_ON(!(old_q->flags & TCQ_F_BUILTIN)); + } sch_tree_lock(sch); if (parent && !parent->level) { /* turn parent into inner node */ @@ -1415,10 +1889,10 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, : TC_HTB_MAXDEPTH) - 1; memset(&parent->inner, 0, sizeof(parent->inner)); } + /* leaf (we) needs elementary qdisc */ cl->leaf.q = new_q ? new_q : &noop_qdisc; - cl->common.classid = classid; cl->parent = parent; /* set class to be in HTB_CAN_SEND state */ @@ -1444,12 +1918,30 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, if (err) return err; } - sch_tree_lock(sch); - } - rate64 = tb[TCA_HTB_RATE64] ? nla_get_u64(tb[TCA_HTB_RATE64]) : 0; + if (q->offload) { + struct net_device *dev = qdisc_dev(sch); + + offload_opt = (struct tc_htb_qopt_offload) { + .command = TC_HTB_NODE_MODIFY, + .classid = cl->common.classid, + .rate = max_t(u64, hopt->rate.rate, rate64), + .ceil = max_t(u64, hopt->ceil.rate, ceil64), + .extack = extack, + }; + err = htb_offload(dev, &offload_opt); + if (err) + /* Estimator was replaced, and rollback may fail + * as well, so we don't try to recover it, and + * the estimator won't work property with the + * offload anyway, because bstats are updated + * only when the stats are queried. + */ + return err; + } - ceil64 = tb[TCA_HTB_CEIL64] ? nla_get_u64(tb[TCA_HTB_CEIL64]) : 0; + sch_tree_lock(sch); + } psched_ratecfg_precompute(&cl->rate, &hopt->rate, rate64); psched_ratecfg_precompute(&cl->ceil, &hopt->ceil, ceil64); @@ -1492,6 +1984,11 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, *arg = (unsigned long)cl; return 0; +err_kill_estimator: + gen_kill_estimator(&cl->rate_est); +err_block_put: + tcf_block_put(cl->block); + kfree(cl); failure: return err; } @@ -1557,6 +2054,7 @@ static void htb_walk(struct Qdisc *sch, struct qdisc_walker *arg) } static const struct Qdisc_class_ops htb_class_ops = { + .select_queue = htb_select_queue, .graft = htb_graft, .leaf = htb_leaf, .qlen_notify = htb_qlen_notify, @@ -1579,6 +2077,7 @@ static struct Qdisc_ops htb_qdisc_ops __read_mostly = { .dequeue = htb_dequeue, .peek = qdisc_peek_dequeued, .init = htb_init, + .attach = htb_attach, .reset = htb_reset, .destroy = htb_destroy, .dump = htb_dump, diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 6335230a971e..1db9d4a2ef5e 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -529,7 +529,8 @@ static void qfq_destroy_class(struct Qdisc *sch, struct qfq_class *cl) kfree(cl); } -static int qfq_delete_class(struct Qdisc *sch, unsigned long arg) +static int qfq_delete_class(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) { struct qfq_sched *q = qdisc_priv(sch); struct qfq_class *cl = (struct qfq_class *)arg; diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c index da047a37a3bf..dde829d4b9f8 100644 --- a/net/sched/sch_sfb.c +++ b/net/sched/sch_sfb.c @@ -649,7 +649,8 @@ static int sfb_change_class(struct Qdisc *sch, u32 classid, u32 parentid, return -ENOSYS; } -static int sfb_delete(struct Qdisc *sch, unsigned long cl) +static int sfb_delete(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) { return -ENOSYS; } diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 5fb9164aa690..dcc50ae54550 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -857,6 +857,7 @@ int svc_recv(struct svc_rqst *rqstp, long timeout) err = -EAGAIN; if (len <= 0) goto out_release; + trace_svc_xdr_recvfrom(&rqstp->rq_arg); clear_bit(XPT_OLD, &xprt->xpt_flags); @@ -866,7 +867,6 @@ int svc_recv(struct svc_rqst *rqstp, long timeout) if (serv->sv_stats) serv->sv_stats->netcnt++; - trace_svc_xdr_recvfrom(rqstp, &rqstp->rq_arg); return len; out_release: rqstp->rq_res.len = 0; @@ -904,7 +904,7 @@ int svc_send(struct svc_rqst *rqstp) xb->len = xb->head[0].iov_len + xb->page_len + xb->tail[0].iov_len; - trace_svc_xdr_sendto(rqstp, xb); + trace_svc_xdr_sendto(rqstp->rq_xid, xb); trace_svc_stats_latency(rqstp); len = xprt->xpt_ops->xpo_sendto(rqstp); diff --git a/net/switchdev/Makefile b/net/switchdev/Makefile index bd69a3136e76..c5561d7f3a7c 100644 --- a/net/switchdev/Makefile +++ b/net/switchdev/Makefile @@ -3,4 +3,4 @@ # Makefile for the Switch device API # -obj-$(CONFIG_NET_SWITCHDEV) += switchdev.o +obj-y += switchdev.o diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 2aca86021df5..e9263280a2d4 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -117,10 +117,6 @@ struct sk_buff *tipc_msg_create(uint user, uint type, msg_set_origport(msg, oport); msg_set_destport(msg, dport); msg_set_errcode(msg, errcode); - if (hdr_sz > SHORT_H_SIZE) { - msg_set_orignode(msg, onode); - msg_set_destnode(msg, dnode); - } return buf; } diff --git a/net/wireless/chan.c b/net/wireless/chan.c index e4030f1fbc60..285b8076054b 100644 --- a/net/wireless/chan.c +++ b/net/wireless/chan.c @@ -1093,7 +1093,7 @@ static bool cfg80211_ir_permissive_chan(struct wiphy *wiphy, struct wireless_dev *wdev; struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy); - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); if (!IS_ENABLED(CONFIG_CFG80211_REG_RELAX_NO_IR) || !(wiphy->regulatory_flags & REGULATORY_ENABLE_RELAX_NO_IR)) @@ -1216,9 +1216,10 @@ bool cfg80211_reg_can_beacon_relax(struct wiphy *wiphy, struct cfg80211_chan_def *chandef, enum nl80211_iftype iftype) { + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy); bool check_no_ir; - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); /* * Under certain conditions suggested by some regulatory bodies a diff --git a/net/wireless/core.c b/net/wireless/core.c index 4b1f35e976e7..200cd9f5fd5f 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -222,7 +222,7 @@ static void cfg80211_rfkill_poll(struct rfkill *rfkill, void *data) void cfg80211_stop_p2p_device(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev) { - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); if (WARN_ON(wdev->iftype != NL80211_IFTYPE_P2P_DEVICE)) return; @@ -247,7 +247,7 @@ void cfg80211_stop_p2p_device(struct cfg80211_registered_device *rdev, void cfg80211_stop_nan(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev) { - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); if (WARN_ON(wdev->iftype != NL80211_IFTYPE_NAN)) return; @@ -273,7 +273,11 @@ void cfg80211_shutdown_all_interfaces(struct wiphy *wiphy) dev_close(wdev->netdev); continue; } + /* otherwise, check iftype */ + + wiphy_lock(wiphy); + switch (wdev->iftype) { case NL80211_IFTYPE_P2P_DEVICE: cfg80211_stop_p2p_device(rdev, wdev); @@ -284,6 +288,8 @@ void cfg80211_shutdown_all_interfaces(struct wiphy *wiphy) default: break; } + + wiphy_unlock(wiphy); } } EXPORT_SYMBOL_GPL(cfg80211_shutdown_all_interfaces); @@ -318,9 +324,9 @@ static void cfg80211_event_work(struct work_struct *work) rdev = container_of(work, struct cfg80211_registered_device, event_work); - rtnl_lock(); + wiphy_lock(&rdev->wiphy); cfg80211_process_rdev_events(rdev); - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); } void cfg80211_destroy_ifaces(struct cfg80211_registered_device *rdev) @@ -475,6 +481,7 @@ use_default_name: } } + mutex_init(&rdev->wiphy.mtx); INIT_LIST_HEAD(&rdev->wiphy.wdev_list); INIT_LIST_HEAD(&rdev->beacon_registrations); spin_lock_init(&rdev->beacon_registrations_lock); @@ -1007,15 +1014,16 @@ void wiphy_unregister(struct wiphy *wiphy) wait_event(rdev->dev_wait, ({ int __count; - rtnl_lock(); + wiphy_lock(&rdev->wiphy); __count = rdev->opencount; - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); __count == 0; })); if (rdev->rfkill) rfkill_unregister(rdev->rfkill); rtnl_lock(); + wiphy_lock(&rdev->wiphy); nl80211_notify_wiphy(rdev, NL80211_CMD_DEL_WIPHY); rdev->wiphy.registered = false; @@ -1038,6 +1046,7 @@ void wiphy_unregister(struct wiphy *wiphy) cfg80211_rdev_list_generation++; device_del(&rdev->wiphy.dev); + wiphy_unlock(&rdev->wiphy); rtnl_unlock(); flush_work(&rdev->scan_done_wk); @@ -1070,6 +1079,7 @@ void cfg80211_dev_free(struct cfg80211_registered_device *rdev) } list_for_each_entry_safe(scan, tmp, &rdev->bss_list, list) cfg80211_put_bss(&rdev->wiphy, &scan->pub); + mutex_destroy(&rdev->wiphy.mtx); kfree(rdev); } @@ -1094,19 +1104,28 @@ void cfg80211_cqm_config_free(struct wireless_dev *wdev) wdev->cqm_config = NULL; } -static void __cfg80211_unregister_wdev(struct wireless_dev *wdev, bool sync) +static void _cfg80211_unregister_wdev(struct wireless_dev *wdev, + bool unregister_netdev) { struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); flush_work(&wdev->pmsr_free_wk); nl80211_notify_iface(rdev, wdev, NL80211_CMD_DEL_INTERFACE); + wdev->registered = false; + + if (wdev->netdev) { + sysfs_remove_link(&wdev->netdev->dev.kobj, "phy80211"); + if (unregister_netdev) + unregister_netdevice(wdev->netdev); + } + list_del_rcu(&wdev->list); - if (sync) - synchronize_rcu(); + synchronize_net(); rdev->devlist_generation++; cfg80211_mlme_purge_registrations(wdev); @@ -1131,14 +1150,23 @@ static void __cfg80211_unregister_wdev(struct wireless_dev *wdev, bool sync) flush_work(&wdev->disconnect_wk); cfg80211_cqm_config_free(wdev); + + /* + * Ensure that all events have been processed and + * freed. + */ + cfg80211_process_wdev_events(wdev); + + if (WARN_ON(wdev->current_bss)) { + cfg80211_unhold_bss(wdev->current_bss); + cfg80211_put_bss(wdev->wiphy, &wdev->current_bss->pub); + wdev->current_bss = NULL; + } } void cfg80211_unregister_wdev(struct wireless_dev *wdev) { - if (WARN_ON(wdev->netdev)) - return; - - __cfg80211_unregister_wdev(wdev, true); + _cfg80211_unregister_wdev(wdev, true); } EXPORT_SYMBOL(cfg80211_unregister_wdev); @@ -1149,7 +1177,7 @@ static const struct device_type wiphy_type = { void cfg80211_update_iface_num(struct cfg80211_registered_device *rdev, enum nl80211_iftype iftype, int num) { - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); rdev->num_running_ifaces += num; if (iftype == NL80211_IFTYPE_MONITOR) @@ -1162,7 +1190,7 @@ void __cfg80211_leave(struct cfg80211_registered_device *rdev, struct net_device *dev = wdev->netdev; struct cfg80211_sched_scan_request *pos, *tmp; - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); ASSERT_WDEV_LOCK(wdev); cfg80211_pmsr_wdev_down(wdev); @@ -1279,6 +1307,9 @@ void cfg80211_init_wdev(struct wireless_dev *wdev) void cfg80211_register_wdev(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev) { + ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); + /* * We get here also when the interface changes network namespaces, * as it's registered into the new one, but we don't want it to @@ -1290,10 +1321,49 @@ void cfg80211_register_wdev(struct cfg80211_registered_device *rdev, wdev->identifier = ++rdev->wdev_id; list_add_rcu(&wdev->list, &rdev->wiphy.wdev_list); rdev->devlist_generation++; + wdev->registered = true; nl80211_notify_iface(rdev, wdev, NL80211_CMD_NEW_INTERFACE); } +int cfg80211_register_netdevice(struct net_device *dev) +{ + struct wireless_dev *wdev = dev->ieee80211_ptr; + struct cfg80211_registered_device *rdev; + int ret; + + ASSERT_RTNL(); + + if (WARN_ON(!wdev)) + return -EINVAL; + + rdev = wiphy_to_rdev(wdev->wiphy); + + lockdep_assert_held(&rdev->wiphy.mtx); + + /* we'll take care of this */ + wdev->registered = true; + ret = register_netdevice(dev); + if (ret) + goto out; + + if (sysfs_create_link(&dev->dev.kobj, &rdev->wiphy.dev.kobj, + "phy80211")) { + pr_err("failed to add phy80211 symlink to netdev!\n"); + unregister_netdevice(dev); + ret = -EINVAL; + goto out; + } + + cfg80211_register_wdev(rdev, wdev); + ret = 0; +out: + if (ret) + wdev->registered = false; + return ret; +} +EXPORT_SYMBOL(cfg80211_register_netdevice); + static int cfg80211_netdev_notifier_call(struct notifier_block *nb, unsigned long state, void *ptr) { @@ -1319,22 +1389,30 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, cfg80211_init_wdev(wdev); break; case NETDEV_REGISTER: + if (!wdev->registered) { + wiphy_lock(&rdev->wiphy); + cfg80211_register_wdev(rdev, wdev); + wiphy_unlock(&rdev->wiphy); + } + break; + case NETDEV_UNREGISTER: /* - * NB: cannot take rdev->mtx here because this may be - * called within code protected by it when interfaces - * are added with nl80211. + * It is possible to get NETDEV_UNREGISTER multiple times, + * so check wdev->registered. */ - if (sysfs_create_link(&dev->dev.kobj, &rdev->wiphy.dev.kobj, - "phy80211")) { - pr_err("failed to add phy80211 symlink to netdev!\n"); + if (wdev->registered) { + wiphy_lock(&rdev->wiphy); + _cfg80211_unregister_wdev(wdev, false); + wiphy_unlock(&rdev->wiphy); } - - cfg80211_register_wdev(rdev, wdev); break; case NETDEV_GOING_DOWN: + wiphy_lock(&rdev->wiphy); cfg80211_leave(rdev, wdev); + wiphy_unlock(&rdev->wiphy); break; case NETDEV_DOWN: + wiphy_lock(&rdev->wiphy); cfg80211_update_iface_num(rdev, wdev->iftype, -1); if (rdev->scan_req && rdev->scan_req->wdev == wdev) { if (WARN_ON(!rdev->scan_req->notified && @@ -1351,9 +1429,11 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, } rdev->opencount--; + wiphy_unlock(&rdev->wiphy); wake_up(&rdev->dev_wait); break; case NETDEV_UP: + wiphy_lock(&rdev->wiphy); cfg80211_update_iface_num(rdev, wdev->iftype, 1); wdev_lock(wdev); switch (wdev->iftype) { @@ -1400,38 +1480,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, /* assume this means it's off */ wdev->ps = false; } - break; - case NETDEV_UNREGISTER: - /* - * It is possible to get NETDEV_UNREGISTER - * multiple times. To detect that, check - * that the interface is still on the list - * of registered interfaces, and only then - * remove and clean it up. - */ - if (!list_empty(&wdev->list)) { - __cfg80211_unregister_wdev(wdev, false); - sysfs_remove_link(&dev->dev.kobj, "phy80211"); - } - /* - * synchronise (so that we won't find this netdev - * from other code any more) and then clear the list - * head so that the above code can safely check for - * !list_empty() to avoid double-cleanup. - */ - synchronize_rcu(); - INIT_LIST_HEAD(&wdev->list); - /* - * Ensure that all events have been processed and - * freed. - */ - cfg80211_process_wdev_events(wdev); - - if (WARN_ON(wdev->current_bss)) { - cfg80211_unhold_bss(wdev->current_bss); - cfg80211_put_bss(wdev->wiphy, &wdev->current_bss->pub); - wdev->current_bss = NULL; - } + wiphy_unlock(&rdev->wiphy); break; case NETDEV_PRE_UP: if (!cfg80211_iftype_allowed(wdev->wiphy, wdev->iftype, diff --git a/net/wireless/core.h b/net/wireless/core.h index 7df91f940212..a7d19b4b40ac 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -231,7 +231,7 @@ static inline void wdev_unlock(struct wireless_dev *wdev) static inline bool cfg80211_has_monitors_only(struct cfg80211_registered_device *rdev) { - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); return rdev->num_running_ifaces == rdev->num_running_monitor_ifaces && rdev->num_running_ifaces > 0; diff --git a/net/wireless/debugfs.c b/net/wireless/debugfs.c index 76b845f68ac8..aab43469a2f0 100644 --- a/net/wireless/debugfs.c +++ b/net/wireless/debugfs.c @@ -73,8 +73,6 @@ static ssize_t ht40allow_map_read(struct file *file, if (!buf) return -ENOMEM; - rtnl_lock(); - for (band = 0; band < NUM_NL80211_BANDS; band++) { sband = wiphy->bands[band]; if (!sband) @@ -84,8 +82,6 @@ static ssize_t ht40allow_map_read(struct file *file, buf, buf_size, offset); } - rtnl_unlock(); - r = simple_read_from_buffer(user_buf, count, ppos, buf, offset); kfree(buf); diff --git a/net/wireless/ibss.c b/net/wireless/ibss.c index a0621bb76d8e..8f98e546becf 100644 --- a/net/wireless/ibss.c +++ b/net/wireless/ibss.c @@ -3,6 +3,7 @@ * Some IBSS support code for cfg80211. * * Copyright 2009 Johannes Berg <johannes@sipsolutions.net> + * Copyright (C) 2020-2021 Intel Corporation */ #include <linux/etherdevice.h> @@ -92,7 +93,7 @@ int __cfg80211_join_ibss(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev = dev->ieee80211_ptr; int err; - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); ASSERT_WDEV_LOCK(wdev); if (wdev->ssid_len) diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c index e1e90761dc00..3aa69b375a10 100644 --- a/net/wireless/mlme.c +++ b/net/wireless/mlme.c @@ -450,7 +450,7 @@ static void cfg80211_mgmt_registrations_update(struct wireless_dev *wdev) struct cfg80211_mgmt_registration *reg; struct mgmt_frame_regs upd = {}; - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); spin_lock_bh(&wdev->mgmt_registrations_lock); if (!wdev->mgmt_registrations_need_update) { @@ -492,10 +492,10 @@ void cfg80211_mgmt_registrations_update_wk(struct work_struct *wk) rdev = container_of(wk, struct cfg80211_registered_device, mgmt_registrations_update_wk); - rtnl_lock(); + wiphy_lock(&rdev->wiphy); list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) cfg80211_mgmt_registrations_update(wdev); - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); } int cfg80211_mlme_register_mgmt(struct wireless_dev *wdev, u32 snd_portid, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 775d0c4d86c3..e5e9d889f00f 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -64,9 +64,9 @@ static const struct genl_multicast_group nl80211_mcgrps[] = { /* returns ERR_PTR values */ static struct wireless_dev * -__cfg80211_wdev_from_attrs(struct net *netns, struct nlattr **attrs) +__cfg80211_wdev_from_attrs(struct cfg80211_registered_device *rdev, + struct net *netns, struct nlattr **attrs) { - struct cfg80211_registered_device *rdev; struct wireless_dev *result = NULL; bool have_ifidx = attrs[NL80211_ATTR_IFINDEX]; bool have_wdev_id = attrs[NL80211_ATTR_WDEV]; @@ -74,8 +74,6 @@ __cfg80211_wdev_from_attrs(struct net *netns, struct nlattr **attrs) int wiphy_idx = -1; int ifidx = -1; - ASSERT_RTNL(); - if (!have_ifidx && !have_wdev_id) return ERR_PTR(-EINVAL); @@ -86,6 +84,28 @@ __cfg80211_wdev_from_attrs(struct net *netns, struct nlattr **attrs) wiphy_idx = wdev_id >> 32; } + if (rdev) { + struct wireless_dev *wdev; + + lockdep_assert_held(&rdev->wiphy.mtx); + + list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) { + if (have_ifidx && wdev->netdev && + wdev->netdev->ifindex == ifidx) { + result = wdev; + break; + } + if (have_wdev_id && wdev->identifier == (u32)wdev_id) { + result = wdev; + break; + } + } + + return result ?: ERR_PTR(-ENODEV); + } + + ASSERT_RTNL(); + list_for_each_entry(rdev, &cfg80211_rdev_list, list) { struct wireless_dev *wdev; @@ -914,22 +934,31 @@ int nl80211_prepare_wdev_dump(struct netlink_callback *cb, return err; } - *wdev = __cfg80211_wdev_from_attrs(sock_net(cb->skb->sk), + rtnl_lock(); + *wdev = __cfg80211_wdev_from_attrs(NULL, sock_net(cb->skb->sk), attrbuf); kfree(attrbuf); - if (IS_ERR(*wdev)) + if (IS_ERR(*wdev)) { + rtnl_unlock(); return PTR_ERR(*wdev); + } *rdev = wiphy_to_rdev((*wdev)->wiphy); + mutex_lock(&(*rdev)->wiphy.mtx); + rtnl_unlock(); /* 0 is the first index - add 1 to parse only once */ cb->args[0] = (*rdev)->wiphy_idx + 1; cb->args[1] = (*wdev)->identifier; } else { /* subtract the 1 again here */ - struct wiphy *wiphy = wiphy_idx_to_wiphy(cb->args[0] - 1); + struct wiphy *wiphy; struct wireless_dev *tmp; - if (!wiphy) + rtnl_lock(); + wiphy = wiphy_idx_to_wiphy(cb->args[0] - 1); + if (!wiphy) { + rtnl_unlock(); return -ENODEV; + } *rdev = wiphy_to_rdev(wiphy); *wdev = NULL; @@ -940,8 +969,12 @@ int nl80211_prepare_wdev_dump(struct netlink_callback *cb, } } - if (!*wdev) + if (!*wdev) { + rtnl_unlock(); return -ENODEV; + } + mutex_lock(&(*rdev)->wiphy.mtx); + rtnl_unlock(); } return 0; @@ -3141,7 +3174,7 @@ static int nl80211_set_channel(struct sk_buff *skb, struct genl_info *info) static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = NULL; struct net_device *netdev = NULL; struct wireless_dev *wdev; int result = 0, rem_txq_params = 0; @@ -3152,8 +3185,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) u8 coverage_class = 0; u32 txq_limit = 0, txq_memory_limit = 0, txq_quantum = 0; - ASSERT_RTNL(); - + rtnl_lock(); /* * Try to find the wiphy and netdev. Normally this * function shouldn't need the netdev, but this is @@ -3177,14 +3209,19 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) if (!netdev) { rdev = __cfg80211_rdev_from_attrs(genl_info_net(info), info->attrs); - if (IS_ERR(rdev)) + if (IS_ERR(rdev)) { + rtnl_unlock(); return PTR_ERR(rdev); + } wdev = NULL; netdev = NULL; result = 0; } else wdev = netdev->ieee80211_ptr; + wiphy_lock(&rdev->wiphy); + rtnl_unlock(); + /* * end workaround code, by now the rdev is available * and locked, and wdev may or may not be NULL. @@ -3195,24 +3232,32 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) rdev, nla_data(info->attrs[NL80211_ATTR_WIPHY_NAME])); if (result) - return result; + goto out; if (info->attrs[NL80211_ATTR_WIPHY_TXQ_PARAMS]) { struct ieee80211_txq_params txq_params; struct nlattr *tb[NL80211_TXQ_ATTR_MAX + 1]; - if (!rdev->ops->set_txq_params) - return -EOPNOTSUPP; + if (!rdev->ops->set_txq_params) { + result = -EOPNOTSUPP; + goto out; + } - if (!netdev) - return -EINVAL; + if (!netdev) { + result = -EINVAL; + goto out; + } if (netdev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && - netdev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) - return -EINVAL; + netdev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) { + result = -EINVAL; + goto out; + } - if (!netif_running(netdev)) - return -ENETDOWN; + if (!netif_running(netdev)) { + result = -ENETDOWN; + goto out; + } nla_for_each_nested(nl_txq_params, info->attrs[NL80211_ATTR_WIPHY_TXQ_PARAMS], @@ -3223,15 +3268,15 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) txq_params_policy, info->extack); if (result) - return result; + goto out; result = parse_txq_params(tb, &txq_params); if (result) - return result; + goto out; result = rdev_set_txq_params(rdev, netdev, &txq_params); if (result) - return result; + goto out; } } @@ -3241,7 +3286,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) nl80211_can_set_dev_channel(wdev) ? netdev : NULL, info); if (result) - return result; + goto out; } if (info->attrs[NL80211_ATTR_WIPHY_TX_POWER_SETTING]) { @@ -3252,15 +3297,19 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) if (!(rdev->wiphy.features & NL80211_FEATURE_VIF_TXPOWER)) txp_wdev = NULL; - if (!rdev->ops->set_tx_power) - return -EOPNOTSUPP; + if (!rdev->ops->set_tx_power) { + result = -EOPNOTSUPP; + goto out; + } idx = NL80211_ATTR_WIPHY_TX_POWER_SETTING; type = nla_get_u32(info->attrs[idx]); if (!info->attrs[NL80211_ATTR_WIPHY_TX_POWER_LEVEL] && - (type != NL80211_TX_POWER_AUTOMATIC)) - return -EINVAL; + (type != NL80211_TX_POWER_AUTOMATIC)) { + result = -EINVAL; + goto out; + } if (type != NL80211_TX_POWER_AUTOMATIC) { idx = NL80211_ATTR_WIPHY_TX_POWER_LEVEL; @@ -3269,7 +3318,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) result = rdev_set_tx_power(rdev, txp_wdev, type, mbm); if (result) - return result; + goto out; } if (info->attrs[NL80211_ATTR_WIPHY_ANTENNA_TX] && @@ -3278,8 +3327,10 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) if ((!rdev->wiphy.available_antennas_tx && !rdev->wiphy.available_antennas_rx) || - !rdev->ops->set_antenna) - return -EOPNOTSUPP; + !rdev->ops->set_antenna) { + result = -EOPNOTSUPP; + goto out; + } tx_ant = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_ANTENNA_TX]); rx_ant = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_ANTENNA_RX]); @@ -3287,15 +3338,17 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) /* reject antenna configurations which don't match the * available antenna masks, except for the "all" mask */ if ((~tx_ant && (tx_ant & ~rdev->wiphy.available_antennas_tx)) || - (~rx_ant && (rx_ant & ~rdev->wiphy.available_antennas_rx))) - return -EINVAL; + (~rx_ant && (rx_ant & ~rdev->wiphy.available_antennas_rx))) { + result = -EINVAL; + goto out; + } tx_ant = tx_ant & rdev->wiphy.available_antennas_tx; rx_ant = rx_ant & rdev->wiphy.available_antennas_rx; result = rdev_set_antenna(rdev, tx_ant, rx_ant); if (result) - return result; + goto out; } changed = 0; @@ -3317,8 +3370,10 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_WIPHY_FRAG_THRESHOLD]) { frag_threshold = nla_get_u32( info->attrs[NL80211_ATTR_WIPHY_FRAG_THRESHOLD]); - if (frag_threshold < 256) - return -EINVAL; + if (frag_threshold < 256) { + result = -EINVAL; + goto out; + } if (frag_threshold != (u32) -1) { /* @@ -3339,8 +3394,10 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) } if (info->attrs[NL80211_ATTR_WIPHY_COVERAGE_CLASS]) { - if (info->attrs[NL80211_ATTR_WIPHY_DYN_ACK]) - return -EINVAL; + if (info->attrs[NL80211_ATTR_WIPHY_DYN_ACK]) { + result = -EINVAL; + goto out; + } coverage_class = nla_get_u8( info->attrs[NL80211_ATTR_WIPHY_COVERAGE_CLASS]); @@ -3348,16 +3405,20 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) } if (info->attrs[NL80211_ATTR_WIPHY_DYN_ACK]) { - if (!(rdev->wiphy.features & NL80211_FEATURE_ACKTO_ESTIMATION)) - return -EOPNOTSUPP; + if (!(rdev->wiphy.features & NL80211_FEATURE_ACKTO_ESTIMATION)) { + result = -EOPNOTSUPP; + goto out; + } changed |= WIPHY_PARAM_DYN_ACK; } if (info->attrs[NL80211_ATTR_TXQ_LIMIT]) { if (!wiphy_ext_feature_isset(&rdev->wiphy, - NL80211_EXT_FEATURE_TXQS)) - return -EOPNOTSUPP; + NL80211_EXT_FEATURE_TXQS)) { + result = -EOPNOTSUPP; + goto out; + } txq_limit = nla_get_u32( info->attrs[NL80211_ATTR_TXQ_LIMIT]); changed |= WIPHY_PARAM_TXQ_LIMIT; @@ -3365,8 +3426,10 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_TXQ_MEMORY_LIMIT]) { if (!wiphy_ext_feature_isset(&rdev->wiphy, - NL80211_EXT_FEATURE_TXQS)) - return -EOPNOTSUPP; + NL80211_EXT_FEATURE_TXQS)) { + result = -EOPNOTSUPP; + goto out; + } txq_memory_limit = nla_get_u32( info->attrs[NL80211_ATTR_TXQ_MEMORY_LIMIT]); changed |= WIPHY_PARAM_TXQ_MEMORY_LIMIT; @@ -3374,8 +3437,10 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_TXQ_QUANTUM]) { if (!wiphy_ext_feature_isset(&rdev->wiphy, - NL80211_EXT_FEATURE_TXQS)) - return -EOPNOTSUPP; + NL80211_EXT_FEATURE_TXQS)) { + result = -EOPNOTSUPP; + goto out; + } txq_quantum = nla_get_u32( info->attrs[NL80211_ATTR_TXQ_QUANTUM]); changed |= WIPHY_PARAM_TXQ_QUANTUM; @@ -3387,8 +3452,10 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) u8 old_coverage_class; u32 old_txq_limit, old_txq_memory_limit, old_txq_quantum; - if (!rdev->ops->set_wiphy_params) - return -EOPNOTSUPP; + if (!rdev->ops->set_wiphy_params) { + result = -EOPNOTSUPP; + goto out; + } old_retry_short = rdev->wiphy.retry_short; old_retry_long = rdev->wiphy.retry_long; @@ -3426,10 +3493,15 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) rdev->wiphy.txq_limit = old_txq_limit; rdev->wiphy.txq_memory_limit = old_txq_memory_limit; rdev->wiphy.txq_quantum = old_txq_quantum; - return result; + goto out; } } - return 0; + + result = 0; + +out: + wiphy_unlock(&rdev->wiphy); + return result; } static int nl80211_send_chandef(struct sk_buff *msg, @@ -3960,6 +4032,17 @@ static int nl80211_del_interface(struct sk_buff *skb, struct genl_info *info) return -EOPNOTSUPP; /* + * We hold RTNL, so this is safe, without RTNL opencount cannot + * reach 0, and thus the rdev cannot be deleted. + * + * We need to do it for the dev_close(), since that will call + * the netdev notifiers, and we need to acquire the mutex there + * but don't know if we get there from here or from some other + * place (e.g. "ip link set ... down"). + */ + mutex_unlock(&rdev->wiphy.mtx); + + /* * If we remove a wireless device without a netdev then clear * user_ptr[1] so that nl80211_post_doit won't dereference it * to check if it needs to do dev_put(). Otherwise it crashes @@ -3968,6 +4051,10 @@ static int nl80211_del_interface(struct sk_buff *skb, struct genl_info *info) */ if (!wdev->netdev) info->user_ptr[1] = NULL; + else + dev_close(wdev->netdev); + + mutex_lock(&rdev->wiphy.mtx); return rdev_del_virtual_intf(rdev, wdev); } @@ -5884,10 +5971,11 @@ static int nl80211_dump_station(struct sk_buff *skb, int sta_idx = cb->args[2]; int err; - rtnl_lock(); err = nl80211_prepare_wdev_dump(cb, &rdev, &wdev); if (err) - goto out_err; + return err; + /* nl80211_prepare_wdev_dump acquired it in the successful case */ + __acquire(&rdev->wiphy.mtx); if (!wdev->netdev) { err = -EINVAL; @@ -5922,7 +6010,7 @@ static int nl80211_dump_station(struct sk_buff *skb, cb->args[2] = sta_idx; err = skb->len; out_err: - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); return err; } @@ -6780,10 +6868,11 @@ static int nl80211_dump_mpath(struct sk_buff *skb, int path_idx = cb->args[2]; int err; - rtnl_lock(); err = nl80211_prepare_wdev_dump(cb, &rdev, &wdev); if (err) - goto out_err; + return err; + /* nl80211_prepare_wdev_dump acquired it in the successful case */ + __acquire(&rdev->wiphy.mtx); if (!rdev->ops->dump_mpath) { err = -EOPNOTSUPP; @@ -6816,7 +6905,7 @@ static int nl80211_dump_mpath(struct sk_buff *skb, cb->args[2] = path_idx; err = skb->len; out_err: - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); return err; } @@ -6979,10 +7068,11 @@ static int nl80211_dump_mpp(struct sk_buff *skb, int path_idx = cb->args[2]; int err; - rtnl_lock(); err = nl80211_prepare_wdev_dump(cb, &rdev, &wdev); if (err) - goto out_err; + return err; + /* nl80211_prepare_wdev_dump acquired it in the successful case */ + __acquire(&rdev->wiphy.mtx); if (!rdev->ops->dump_mpp) { err = -EOPNOTSUPP; @@ -7015,7 +7105,7 @@ static int nl80211_dump_mpp(struct sk_buff *skb, cb->args[2] = path_idx; err = skb->len; out_err: - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); return err; } @@ -7634,12 +7724,15 @@ static int nl80211_get_reg_do(struct sk_buff *skb, struct genl_info *info) if (!hdr) goto put_failure; + rtnl_lock(); + if (info->attrs[NL80211_ATTR_WIPHY]) { bool self_managed; rdev = cfg80211_get_dev_from_info(genl_info_net(info), info); if (IS_ERR(rdev)) { nlmsg_free(msg); + rtnl_unlock(); return PTR_ERR(rdev); } @@ -7651,6 +7744,7 @@ static int nl80211_get_reg_do(struct sk_buff *skb, struct genl_info *info) /* a self-managed-reg device must have a private regdom */ if (WARN_ON(!regdom && self_managed)) { nlmsg_free(msg); + rtnl_unlock(); return -EINVAL; } @@ -7675,11 +7769,13 @@ static int nl80211_get_reg_do(struct sk_buff *skb, struct genl_info *info) rcu_read_unlock(); genlmsg_end(msg, hdr); + rtnl_unlock(); return genlmsg_reply(msg, info); nla_put_failure_rcu: rcu_read_unlock(); nla_put_failure: + rtnl_unlock(); put_failure: nlmsg_free(msg); return -EMSGSIZE; @@ -7842,12 +7938,17 @@ static int nl80211_set_reg(struct sk_buff *skb, struct genl_info *info) return -EINVAL; } - if (!reg_is_valid_request(alpha2)) - return -EINVAL; + rtnl_lock(); + if (!reg_is_valid_request(alpha2)) { + r = -EINVAL; + goto out; + } rd = kzalloc(struct_size(rd, reg_rules, num_rules), GFP_KERNEL); - if (!rd) - return -ENOMEM; + if (!rd) { + r = -ENOMEM; + goto out; + } rd->n_reg_rules = num_rules; rd->alpha2[0] = alpha2[0]; @@ -7879,10 +7980,13 @@ static int nl80211_set_reg(struct sk_buff *skb, struct genl_info *info) } } + r = set_regdom(rd, REGD_SOURCE_CRDA); /* set_regdom takes ownership of rd */ - return set_regdom(rd, REGD_SOURCE_CRDA); + rd = NULL; bad_reg: kfree(rd); + out: + rtnl_unlock(); return r; } #endif /* CONFIG_CFG80211_CRDA_SUPPORT */ @@ -9050,10 +9154,7 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) struct net_device *dev = info->user_ptr[1]; struct wireless_dev *wdev = dev->ieee80211_ptr; struct cfg80211_csa_settings params; - /* csa_attrs is defined static to avoid waste of stack size - this - * function is called under RTNL lock, so this should not be a problem. - */ - static struct nlattr *csa_attrs[NL80211_ATTR_MAX+1]; + struct nlattr **csa_attrs = NULL; int err; bool need_new_beacon = false; bool need_handle_dfs_flag = true; @@ -9118,28 +9219,39 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) if (err) return err; + csa_attrs = kcalloc(NL80211_ATTR_MAX + 1, sizeof(*csa_attrs), + GFP_KERNEL); + if (!csa_attrs) + return -ENOMEM; + err = nla_parse_nested_deprecated(csa_attrs, NL80211_ATTR_MAX, info->attrs[NL80211_ATTR_CSA_IES], nl80211_policy, info->extack); if (err) - return err; + goto free; err = nl80211_parse_beacon(rdev, csa_attrs, ¶ms.beacon_csa); if (err) - return err; + goto free; - if (!csa_attrs[NL80211_ATTR_CNTDWN_OFFS_BEACON]) - return -EINVAL; + if (!csa_attrs[NL80211_ATTR_CNTDWN_OFFS_BEACON]) { + err = -EINVAL; + goto free; + } len = nla_len(csa_attrs[NL80211_ATTR_CNTDWN_OFFS_BEACON]); - if (!len || (len % sizeof(u16))) - return -EINVAL; + if (!len || (len % sizeof(u16))) { + err = -EINVAL; + goto free; + } params.n_counter_offsets_beacon = len / sizeof(u16); if (rdev->wiphy.max_num_csa_counters && (params.n_counter_offsets_beacon > - rdev->wiphy.max_num_csa_counters)) - return -EINVAL; + rdev->wiphy.max_num_csa_counters)) { + err = -EINVAL; + goto free; + } params.counter_offsets_beacon = nla_data(csa_attrs[NL80211_ATTR_CNTDWN_OFFS_BEACON]); @@ -9148,23 +9260,31 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) for (i = 0; i < params.n_counter_offsets_beacon; i++) { u16 offset = params.counter_offsets_beacon[i]; - if (offset >= params.beacon_csa.tail_len) - return -EINVAL; + if (offset >= params.beacon_csa.tail_len) { + err = -EINVAL; + goto free; + } - if (params.beacon_csa.tail[offset] != params.count) - return -EINVAL; + if (params.beacon_csa.tail[offset] != params.count) { + err = -EINVAL; + goto free; + } } if (csa_attrs[NL80211_ATTR_CNTDWN_OFFS_PRESP]) { len = nla_len(csa_attrs[NL80211_ATTR_CNTDWN_OFFS_PRESP]); - if (!len || (len % sizeof(u16))) - return -EINVAL; + if (!len || (len % sizeof(u16))) { + err = -EINVAL; + goto free; + } params.n_counter_offsets_presp = len / sizeof(u16); if (rdev->wiphy.max_num_csa_counters && (params.n_counter_offsets_presp > - rdev->wiphy.max_num_csa_counters)) - return -EINVAL; + rdev->wiphy.max_num_csa_counters)) { + err = -EINVAL; + goto free; + } params.counter_offsets_presp = nla_data(csa_attrs[NL80211_ATTR_CNTDWN_OFFS_PRESP]); @@ -9173,35 +9293,42 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) for (i = 0; i < params.n_counter_offsets_presp; i++) { u16 offset = params.counter_offsets_presp[i]; - if (offset >= params.beacon_csa.probe_resp_len) - return -EINVAL; + if (offset >= params.beacon_csa.probe_resp_len) { + err = -EINVAL; + goto free; + } if (params.beacon_csa.probe_resp[offset] != - params.count) - return -EINVAL; + params.count) { + err = -EINVAL; + goto free; + } } } skip_beacons: err = nl80211_parse_chandef(rdev, info, ¶ms.chandef); if (err) - return err; + goto free; if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, ¶ms.chandef, - wdev->iftype)) - return -EINVAL; + wdev->iftype)) { + err = -EINVAL; + goto free; + } err = cfg80211_chandef_dfs_required(wdev->wiphy, ¶ms.chandef, wdev->iftype); if (err < 0) - return err; + goto free; if (err > 0) { params.radar_required = true; if (need_handle_dfs_flag && !nla_get_flag(info->attrs[NL80211_ATTR_HANDLE_DFS])) { - return -EINVAL; + err = -EINVAL; + goto free; } } @@ -9212,6 +9339,8 @@ skip_beacons: err = rdev_channel_switch(rdev, dev, ¶ms); wdev_unlock(wdev); +free: + kfree(csa_attrs); return err; } @@ -9362,12 +9491,11 @@ static int nl80211_dump_scan(struct sk_buff *skb, struct netlink_callback *cb) int start = cb->args[2], idx = 0; int err; - rtnl_lock(); err = nl80211_prepare_wdev_dump(cb, &rdev, &wdev); - if (err) { - rtnl_unlock(); + if (err) return err; - } + /* nl80211_prepare_wdev_dump acquired it in the successful case */ + __acquire(&rdev->wiphy.mtx); wdev_lock(wdev); spin_lock_bh(&rdev->bss_lock); @@ -9398,7 +9526,7 @@ static int nl80211_dump_scan(struct sk_buff *skb, struct netlink_callback *cb) wdev_unlock(wdev); cb->args[2] = idx; - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); return skb->len; } @@ -9496,10 +9624,13 @@ static int nl80211_dump_survey(struct sk_buff *skb, struct netlink_callback *cb) if (!attrbuf) return -ENOMEM; - rtnl_lock(); res = nl80211_prepare_wdev_dump(cb, &rdev, &wdev); - if (res) - goto out_err; + if (res) { + kfree(attrbuf); + return res; + } + /* nl80211_prepare_wdev_dump acquired it in the successful case */ + __acquire(&rdev->wiphy.mtx); /* prepare_wdev_dump parsed the attributes */ radio_stats = attrbuf[NL80211_ATTR_SURVEY_RADIO_STATS]; @@ -9541,7 +9672,7 @@ static int nl80211_dump_survey(struct sk_buff *skb, struct netlink_callback *cb) res = skb->len; out_err: kfree(attrbuf); - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); return res; } @@ -10403,10 +10534,14 @@ EXPORT_SYMBOL(__cfg80211_send_event_skb); static int nl80211_testmode_do(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *rdev = info->user_ptr[0]; - struct wireless_dev *wdev = - __cfg80211_wdev_from_attrs(genl_info_net(info), info->attrs); + struct wireless_dev *wdev; int err; + lockdep_assert_held(&rdev->wiphy.mtx); + + wdev = __cfg80211_wdev_from_attrs(rdev, genl_info_net(info), + info->attrs); + if (!rdev->ops->testmode_cmd) return -EOPNOTSUPP; @@ -13591,7 +13726,8 @@ static int nl80211_vendor_cmd(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct wireless_dev *wdev = - __cfg80211_wdev_from_attrs(genl_info_net(info), info->attrs); + __cfg80211_wdev_from_attrs(rdev, genl_info_net(info), + info->attrs); int i, err; u32 vid, subcmd; @@ -13715,7 +13851,7 @@ static int nl80211_prepare_vendor_dump(struct sk_buff *skb, goto out; } - *wdev = __cfg80211_wdev_from_attrs(sock_net(skb->sk), attrbuf); + *wdev = __cfg80211_wdev_from_attrs(NULL, sock_net(skb->sk), attrbuf); if (IS_ERR(*wdev)) *wdev = NULL; @@ -14650,31 +14786,24 @@ bad_tid_conf: static int nl80211_pre_doit(const struct genl_ops *ops, struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = NULL; struct wireless_dev *wdev; struct net_device *dev; - bool rtnl = ops->internal_flags & NL80211_FLAG_NEED_RTNL; - - if (rtnl) - rtnl_lock(); + rtnl_lock(); if (ops->internal_flags & NL80211_FLAG_NEED_WIPHY) { rdev = cfg80211_get_dev_from_info(genl_info_net(info), info); if (IS_ERR(rdev)) { - if (rtnl) - rtnl_unlock(); + rtnl_unlock(); return PTR_ERR(rdev); } info->user_ptr[0] = rdev; } else if (ops->internal_flags & NL80211_FLAG_NEED_NETDEV || ops->internal_flags & NL80211_FLAG_NEED_WDEV) { - ASSERT_RTNL(); - - wdev = __cfg80211_wdev_from_attrs(genl_info_net(info), + wdev = __cfg80211_wdev_from_attrs(NULL, genl_info_net(info), info->attrs); if (IS_ERR(wdev)) { - if (rtnl) - rtnl_unlock(); + rtnl_unlock(); return PTR_ERR(wdev); } @@ -14683,8 +14812,7 @@ static int nl80211_pre_doit(const struct genl_ops *ops, struct sk_buff *skb, if (ops->internal_flags & NL80211_FLAG_NEED_NETDEV) { if (!dev) { - if (rtnl) - rtnl_unlock(); + rtnl_unlock(); return -EINVAL; } @@ -14695,8 +14823,7 @@ static int nl80211_pre_doit(const struct genl_ops *ops, struct sk_buff *skb, if (ops->internal_flags & NL80211_FLAG_CHECK_NETDEV_UP && !wdev_running(wdev)) { - if (rtnl) - rtnl_unlock(); + rtnl_unlock(); return -ENETDOWN; } @@ -14706,6 +14833,14 @@ static int nl80211_pre_doit(const struct genl_ops *ops, struct sk_buff *skb, info->user_ptr[0] = rdev; } + if (rdev) { + wiphy_lock(&rdev->wiphy); + /* we keep the mutex locked until post_doit */ + __release(&rdev->wiphy.mtx); + } + if (!(ops->internal_flags & NL80211_FLAG_NEED_RTNL)) + rtnl_unlock(); + return 0; } @@ -14723,6 +14858,14 @@ static void nl80211_post_doit(const struct genl_ops *ops, struct sk_buff *skb, } } + if (info->user_ptr[0]) { + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + + /* we kept the mutex locked since pre_doit */ + __acquire(&rdev->wiphy.mtx); + wiphy_unlock(&rdev->wiphy); + } + if (ops->internal_flags & NL80211_FLAG_NEED_RTNL) rtnl_unlock(); @@ -14851,8 +14994,7 @@ static const struct genl_ops nl80211_ops[] = { .dumpit = nl80211_dump_wiphy, .done = nl80211_dump_wiphy_done, /* can be retrieved by unprivileged users */ - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, }; @@ -14862,7 +15004,6 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_wiphy, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_INTERFACE, @@ -14870,8 +15011,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_get_interface, .dumpit = nl80211_dump_interface, /* can be retrieved by unprivileged users */ - .internal_flags = NL80211_FLAG_NEED_WDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV, }, { .cmd = NL80211_CMD_SET_INTERFACE, @@ -14902,8 +15042,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_key, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_KEY, @@ -14911,7 +15050,6 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_set_key, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | NL80211_FLAG_CLEAR_SKB, }, { @@ -14920,7 +15058,6 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_new_key, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | NL80211_FLAG_CLEAR_SKB, }, { @@ -14928,64 +15065,56 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_del_key, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_BEACON, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .flags = GENL_UNS_ADMIN_PERM, .doit = nl80211_set_beacon, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_START_AP, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .flags = GENL_UNS_ADMIN_PERM, .doit = nl80211_start_ap, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_STOP_AP, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .flags = GENL_UNS_ADMIN_PERM, .doit = nl80211_stop_ap, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_GET_STATION, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_station, .dumpit = nl80211_dump_station, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_SET_STATION, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_station, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_NEW_STATION, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_new_station, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_DEL_STATION, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_del_station, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_GET_MPATH, @@ -14993,8 +15122,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_get_mpath, .dumpit = nl80211_dump_mpath, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_GET_MPP, @@ -15002,47 +15130,42 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_get_mpp, .dumpit = nl80211_dump_mpp, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_MPATH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_mpath, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_NEW_MPATH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_new_mpath, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_DEL_MPATH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_del_mpath, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_BSS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_bss, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_GET_REG, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_reg_do, .dumpit = nl80211_get_reg_dump, - .internal_flags = NL80211_FLAG_NEED_RTNL, + .internal_flags = 0, /* can be retrieved by unprivileged users */ }, #ifdef CONFIG_CFG80211_CRDA_SUPPORT @@ -15051,7 +15174,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_reg, .flags = GENL_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_RTNL, + .internal_flags = 0, }, #endif { @@ -15071,32 +15194,28 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_mesh_config, /* can be retrieved by unprivileged users */ - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_MESH_CONFIG, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_update_mesh_config, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_TRIGGER_SCAN, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_trigger_scan, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_ABORT_SCAN, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_abort_scan, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_GET_SCAN, @@ -15108,16 +15227,14 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_start_sched_scan, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_STOP_SCHED_SCAN, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_stop_sched_scan, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_AUTHENTICATE, @@ -15125,7 +15242,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_authenticate, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { @@ -15134,7 +15251,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_associate, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { @@ -15142,32 +15259,28 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_deauthenticate, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_DISASSOCIATE, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_disassociate, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_JOIN_IBSS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_join_ibss, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_LEAVE_IBSS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_leave_ibss, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, #ifdef CONFIG_NL80211_TESTMODE { @@ -15176,8 +15289,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_testmode_do, .dumpit = nl80211_testmode_dump, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, #endif { @@ -15186,7 +15298,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_connect, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { @@ -15195,7 +15307,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_update_connect_params, .flags = GENL_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { @@ -15203,16 +15315,14 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_disconnect, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_WIPHY_NETNS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_wiphy_netns, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, { .cmd = NL80211_CMD_GET_SURVEY, @@ -15225,7 +15335,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_setdel_pmksa, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { @@ -15233,128 +15343,112 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_setdel_pmksa, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_FLUSH_PMKSA, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_flush_pmksa, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_REMAIN_ON_CHANNEL, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_remain_on_channel, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_cancel_remain_on_channel, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_SET_TX_BITRATE_MASK, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_tx_bitrate_mask, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_REGISTER_FRAME, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_register_mgmt, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV, }, { .cmd = NL80211_CMD_FRAME, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_tx_mgmt, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_FRAME_WAIT_CANCEL, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_tx_mgmt_cancel_wait, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_SET_POWER_SAVE, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_power_save, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_GET_POWER_SAVE, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_power_save, /* can be retrieved by unprivileged users */ - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_SET_CQM, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_cqm, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_SET_CHANNEL, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_channel, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_JOIN_MESH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_join_mesh, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_LEAVE_MESH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_leave_mesh, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_JOIN_OCB, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_join_ocb, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_LEAVE_OCB, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_leave_ocb, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, #ifdef CONFIG_PM { @@ -15362,16 +15456,14 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_wowlan, /* can be retrieved by unprivileged users */ - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, { .cmd = NL80211_CMD_SET_WOWLAN, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_wowlan, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, #endif { @@ -15380,7 +15472,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .doit = nl80211_set_rekey_data, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { @@ -15388,48 +15480,42 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_tdls_mgmt, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_TDLS_OPER, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_tdls_oper, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_UNEXPECTED_FRAME, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_register_unexpected_frame, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_PROBE_CLIENT, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_probe_client, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_REGISTER_BEACONS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_register_beacons, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, { .cmd = NL80211_CMD_SET_NOACK_MAP, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_noack_map, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_START_P2P_DEVICE, @@ -15468,48 +15554,42 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_nan_add_func, .flags = GENL_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_DEL_NAN_FUNCTION, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_nan_del_func, .flags = GENL_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_CHANGE_NAN_CONFIG, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_nan_change_config, .flags = GENL_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_SET_MCAST_RATE, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_mcast_rate, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_SET_MAC_ACL, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_mac_acl, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_RADAR_DETECT, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_start_radar_detection, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_GET_PROTOCOL_FEATURES, @@ -15521,47 +15601,41 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_update_ft_ies, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_CRIT_PROTOCOL_START, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_crit_protocol_start, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_CRIT_PROTOCOL_STOP, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_crit_protocol_stop, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_GET_COALESCE, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_coalesce, - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, { .cmd = NL80211_CMD_SET_COALESCE, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_coalesce, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, { .cmd = NL80211_CMD_CHANNEL_SWITCH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_channel_switch, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_VENDOR, @@ -15570,7 +15644,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .dumpit = nl80211_vendor_cmd_dump, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { @@ -15578,123 +15652,108 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_qos_map, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_ADD_TX_TS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_add_tx_ts, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_DEL_TX_TS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_del_tx_ts, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_TDLS_CHANNEL_SWITCH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_tdls_channel_switch, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_TDLS_CANCEL_CHANNEL_SWITCH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_tdls_cancel_channel_switch, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_MULTICAST_TO_UNICAST, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_multicast_to_unicast, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_SET_PMK, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_pmk, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL | + 0 | NL80211_FLAG_CLEAR_SKB, }, { .cmd = NL80211_CMD_DEL_PMK, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_del_pmk, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_EXTERNAL_AUTH, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_external_auth, .flags = GENL_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_CONTROL_PORT_FRAME, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_tx_control_port, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_GET_FTM_RESPONDER_STATS, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_get_ftm_responder_stats, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_PEER_MEASUREMENT_START, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_pmsr_start, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_WDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP, }, { .cmd = NL80211_CMD_NOTIFY_RADAR, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_notify_radar_detection, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_UPDATE_OWE_INFO, .doit = nl80211_update_owe_info, .flags = GENL_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_PROBE_MESH_LINK, .doit = nl80211_probe_mesh_link, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP, }, { .cmd = NL80211_CMD_SET_TID_CONFIG, .doit = nl80211_set_tid_config, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = NL80211_FLAG_NEED_NETDEV | - NL80211_FLAG_NEED_RTNL, + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_SET_SAR_SPECS, diff --git a/net/wireless/reg.c b/net/wireless/reg.c index bb72447ad960..452b698f42be 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -5,7 +5,7 @@ * Copyright 2008-2011 Luis R. Rodriguez <mcgrof@qca.qualcomm.com> * Copyright 2013-2014 Intel Mobile Communications GmbH * Copyright 2017 Intel Deutschland GmbH - * Copyright (C) 2018 - 2019 Intel Corporation + * Copyright (C) 2018 - 2021 Intel Corporation * * Permission to use, copy, modify, and/or distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -139,10 +139,18 @@ static const struct ieee80211_regdomain *get_cfg80211_regdom(void) return rcu_dereference_rtnl(cfg80211_regdomain); } +/* + * Returns the regulatory domain associated with the wiphy. + * + * Requires any of RTNL, wiphy mutex or RCU protection. + */ const struct ieee80211_regdomain *get_wiphy_regdom(struct wiphy *wiphy) { - return rcu_dereference_rtnl(wiphy->regd); + return rcu_dereference_check(wiphy->regd, + lockdep_is_held(&wiphy->mtx) || + lockdep_rtnl_is_held()); } +EXPORT_SYMBOL(get_wiphy_regdom); static const char *reg_dfs_region_str(enum nl80211_dfs_regions dfs_region) { @@ -164,7 +172,9 @@ enum nl80211_dfs_regions reg_get_dfs_region(struct wiphy *wiphy) const struct ieee80211_regdomain *regd = NULL; const struct ieee80211_regdomain *wiphy_regd = NULL; + rcu_read_lock(); regd = get_cfg80211_regdom(); + if (!wiphy) goto out; @@ -181,6 +191,8 @@ enum nl80211_dfs_regions reg_get_dfs_region(struct wiphy *wiphy) reg_dfs_region_str(regd->dfs_region)); out: + rcu_read_unlock(); + return regd->dfs_region; } @@ -2571,9 +2583,15 @@ void wiphy_apply_custom_regulatory(struct wiphy *wiphy, if (IS_ERR(new_regd)) return; + rtnl_lock(); + wiphy_lock(wiphy); + tmp = get_wiphy_regdom(wiphy); rcu_assign_pointer(wiphy->regd, new_regd); rcu_free_regdom(tmp); + + wiphy_unlock(wiphy); + rtnl_unlock(); } EXPORT_SYMBOL(wiphy_apply_custom_regulatory); @@ -2735,7 +2753,10 @@ reg_process_hint_driver(struct wiphy *wiphy, return REG_REQ_IGNORE; tmp = get_wiphy_regdom(wiphy); + ASSERT_RTNL(); + wiphy_lock(wiphy); rcu_assign_pointer(wiphy->regd, regd); + wiphy_unlock(wiphy); rcu_free_regdom(tmp); } @@ -3067,41 +3088,52 @@ static void reg_process_pending_beacon_hints(void) spin_unlock_bh(®_pending_beacons_lock); } -static void reg_process_self_managed_hints(void) +static void reg_process_self_managed_hint(struct wiphy *wiphy) { - struct cfg80211_registered_device *rdev; - struct wiphy *wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy); const struct ieee80211_regdomain *tmp; const struct ieee80211_regdomain *regd; enum nl80211_band band; struct regulatory_request request = {}; - list_for_each_entry(rdev, &cfg80211_rdev_list, list) { - wiphy = &rdev->wiphy; + ASSERT_RTNL(); + lockdep_assert_wiphy(wiphy); - spin_lock(®_requests_lock); - regd = rdev->requested_regd; - rdev->requested_regd = NULL; - spin_unlock(®_requests_lock); + spin_lock(®_requests_lock); + regd = rdev->requested_regd; + rdev->requested_regd = NULL; + spin_unlock(®_requests_lock); - if (regd == NULL) - continue; + if (!regd) + return; - tmp = get_wiphy_regdom(wiphy); - rcu_assign_pointer(wiphy->regd, regd); - rcu_free_regdom(tmp); + tmp = get_wiphy_regdom(wiphy); + rcu_assign_pointer(wiphy->regd, regd); + rcu_free_regdom(tmp); - for (band = 0; band < NUM_NL80211_BANDS; band++) - handle_band_custom(wiphy, wiphy->bands[band], regd); + for (band = 0; band < NUM_NL80211_BANDS; band++) + handle_band_custom(wiphy, wiphy->bands[band], regd); + + reg_process_ht_flags(wiphy); - reg_process_ht_flags(wiphy); + request.wiphy_idx = get_wiphy_idx(wiphy); + request.alpha2[0] = regd->alpha2[0]; + request.alpha2[1] = regd->alpha2[1]; + request.initiator = NL80211_REGDOM_SET_BY_DRIVER; - request.wiphy_idx = get_wiphy_idx(wiphy); - request.alpha2[0] = regd->alpha2[0]; - request.alpha2[1] = regd->alpha2[1]; - request.initiator = NL80211_REGDOM_SET_BY_DRIVER; + nl80211_send_wiphy_reg_change_event(&request); +} - nl80211_send_wiphy_reg_change_event(&request); +static void reg_process_self_managed_hints(void) +{ + struct cfg80211_registered_device *rdev; + + ASSERT_RTNL(); + + list_for_each_entry(rdev, &cfg80211_rdev_list, list) { + wiphy_lock(&rdev->wiphy); + reg_process_self_managed_hint(&rdev->wiphy); + wiphy_unlock(&rdev->wiphy); } reg_check_channels(); @@ -3780,14 +3812,21 @@ static int reg_set_rd_driver(const struct ieee80211_regdomain *rd, return -ENODEV; if (!driver_request->intersect) { - if (request_wiphy->regd) + ASSERT_RTNL(); + wiphy_lock(request_wiphy); + if (request_wiphy->regd) { + wiphy_unlock(request_wiphy); return -EALREADY; + } regd = reg_copy_regd(rd); - if (IS_ERR(regd)) + if (IS_ERR(regd)) { + wiphy_unlock(request_wiphy); return PTR_ERR(regd); + } rcu_assign_pointer(request_wiphy->regd, regd); + wiphy_unlock(request_wiphy); reset_regdomains(false, rd); return 0; } @@ -3969,8 +4008,8 @@ int regulatory_set_wiphy_regd(struct wiphy *wiphy, } EXPORT_SYMBOL(regulatory_set_wiphy_regd); -int regulatory_set_wiphy_regd_sync_rtnl(struct wiphy *wiphy, - struct ieee80211_regdomain *rd) +int regulatory_set_wiphy_regd_sync(struct wiphy *wiphy, + struct ieee80211_regdomain *rd) { int ret; @@ -3981,10 +4020,11 @@ int regulatory_set_wiphy_regd_sync_rtnl(struct wiphy *wiphy, return ret; /* process the request immediately */ - reg_process_self_managed_hints(); + reg_process_self_managed_hint(wiphy); + reg_check_channels(); return 0; } -EXPORT_SYMBOL(regulatory_set_wiphy_regd_sync_rtnl); +EXPORT_SYMBOL(regulatory_set_wiphy_regd_sync); void wiphy_regulatory_register(struct wiphy *wiphy) { diff --git a/net/wireless/reg.h b/net/wireless/reg.h index f9e83031a40a..f3707f729024 100644 --- a/net/wireless/reg.h +++ b/net/wireless/reg.h @@ -63,7 +63,6 @@ unsigned int reg_get_max_bandwidth(const struct ieee80211_regdomain *rd, const struct ieee80211_reg_rule *rule); bool reg_last_request_cell_base(void); -const struct ieee80211_regdomain *get_wiphy_regdom(struct wiphy *wiphy); /** * regulatory_hint_found_beacon - hints a beacon was found on a channel diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 1b7fec3b53cd..019952d4fc7d 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -918,7 +918,7 @@ void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, union iwreq_data wrqu; #endif - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); if (rdev->scan_msg) { nl80211_send_scan_msg(rdev, rdev->scan_msg); @@ -987,9 +987,9 @@ void __cfg80211_scan_done(struct work_struct *wk) rdev = container_of(wk, struct cfg80211_registered_device, scan_done_wk); - rtnl_lock(); + wiphy_lock(&rdev->wiphy); ___cfg80211_scan_done(rdev, true); - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); } void cfg80211_scan_done(struct cfg80211_scan_request *request, @@ -1022,7 +1022,7 @@ EXPORT_SYMBOL(cfg80211_scan_done); void cfg80211_add_sched_scan_req(struct cfg80211_registered_device *rdev, struct cfg80211_sched_scan_request *req) { - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); list_add_rcu(&req->list, &rdev->sched_scan_req_list); } @@ -1030,7 +1030,7 @@ void cfg80211_add_sched_scan_req(struct cfg80211_registered_device *rdev, static void cfg80211_del_sched_scan_req(struct cfg80211_registered_device *rdev, struct cfg80211_sched_scan_request *req) { - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); list_del_rcu(&req->list); kfree_rcu(req, rcu_head); @@ -1042,7 +1042,7 @@ cfg80211_find_sched_scan_req(struct cfg80211_registered_device *rdev, u64 reqid) struct cfg80211_sched_scan_request *pos; list_for_each_entry_rcu(pos, &rdev->sched_scan_req_list, list, - lockdep_rtnl_is_held()) { + lockdep_is_held(&rdev->wiphy.mtx)) { if (pos->reqid == reqid) return pos; } @@ -1090,7 +1090,7 @@ void cfg80211_sched_scan_results_wk(struct work_struct *work) rdev = container_of(work, struct cfg80211_registered_device, sched_scan_res_wk); - rtnl_lock(); + wiphy_lock(&rdev->wiphy); list_for_each_entry_safe(req, tmp, &rdev->sched_scan_req_list, list) { if (req->report_results) { req->report_results = false; @@ -1105,7 +1105,7 @@ void cfg80211_sched_scan_results_wk(struct work_struct *work) NL80211_CMD_SCHED_SCAN_RESULTS); } } - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); } void cfg80211_sched_scan_results(struct wiphy *wiphy, u64 reqid) @@ -1126,23 +1126,23 @@ void cfg80211_sched_scan_results(struct wiphy *wiphy, u64 reqid) } EXPORT_SYMBOL(cfg80211_sched_scan_results); -void cfg80211_sched_scan_stopped_rtnl(struct wiphy *wiphy, u64 reqid) +void cfg80211_sched_scan_stopped_locked(struct wiphy *wiphy, u64 reqid) { struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy); - ASSERT_RTNL(); + lockdep_assert_held(&wiphy->mtx); trace_cfg80211_sched_scan_stopped(wiphy, reqid); __cfg80211_stop_sched_scan(rdev, reqid, true); } -EXPORT_SYMBOL(cfg80211_sched_scan_stopped_rtnl); +EXPORT_SYMBOL(cfg80211_sched_scan_stopped_locked); void cfg80211_sched_scan_stopped(struct wiphy *wiphy, u64 reqid) { - rtnl_lock(); - cfg80211_sched_scan_stopped_rtnl(wiphy, reqid); - rtnl_unlock(); + wiphy_lock(wiphy); + cfg80211_sched_scan_stopped_locked(wiphy, reqid); + wiphy_unlock(wiphy); } EXPORT_SYMBOL(cfg80211_sched_scan_stopped); @@ -1150,7 +1150,7 @@ int cfg80211_stop_sched_scan_req(struct cfg80211_registered_device *rdev, struct cfg80211_sched_scan_request *req, bool driver_initiated) { - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); if (!driver_initiated) { int err = rdev_sched_scan_stop(rdev, req->dev, req->reqid); @@ -1170,7 +1170,7 @@ int __cfg80211_stop_sched_scan(struct cfg80211_registered_device *rdev, { struct cfg80211_sched_scan_request *sched_scan_req; - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); sched_scan_req = cfg80211_find_sched_scan_req(rdev, reqid); if (!sched_scan_req) @@ -2774,6 +2774,8 @@ int cfg80211_wext_siwscan(struct net_device *dev, eth_broadcast_addr(creq->bssid); + wiphy_lock(&rdev->wiphy); + rdev->scan_req = creq; err = rdev_scan(rdev, creq); if (err) { @@ -2785,6 +2787,7 @@ int cfg80211_wext_siwscan(struct net_device *dev, creq = NULL; dev_hold(dev); } + wiphy_unlock(&rdev->wiphy); out: kfree(creq); return err; diff --git a/net/wireless/sme.c b/net/wireless/sme.c index 38df713f2e2e..07756ca5e3b5 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -67,7 +67,6 @@ static int cfg80211_conn_scan(struct wireless_dev *wdev) struct cfg80211_scan_request *request; int n_channels, err; - ASSERT_RTNL(); ASSERT_WDEV_LOCK(wdev); if (rdev->scan_req || rdev->scan_msg) @@ -233,7 +232,7 @@ void cfg80211_conn_work(struct work_struct *work) u8 bssid_buf[ETH_ALEN], *bssid = NULL; enum nl80211_timeout_reason treason; - rtnl_lock(); + wiphy_lock(&rdev->wiphy); list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) { if (!wdev->netdev) @@ -266,7 +265,7 @@ void cfg80211_conn_work(struct work_struct *work) wdev_unlock(wdev); } - rtnl_unlock(); + wiphy_unlock(&rdev->wiphy); } /* Returned bss is reference counted and must be cleaned up appropriately. */ diff --git a/net/wireless/sysfs.c b/net/wireless/sysfs.c index 3ac1f48195d2..043762354a66 100644 --- a/net/wireless/sysfs.c +++ b/net/wireless/sysfs.c @@ -5,6 +5,7 @@ * * Copyright 2005-2006 Jiri Benc <jbenc@suse.cz> * Copyright 2006 Johannes Berg <johannes@sipsolutions.net> + * Copyright (C) 2020-2021 Intel Corporation */ #include <linux/device.h> @@ -104,6 +105,7 @@ static int wiphy_suspend(struct device *dev) rdev->suspend_at = ktime_get_boottime_seconds(); rtnl_lock(); + wiphy_lock(&rdev->wiphy); if (rdev->wiphy.registered) { if (!rdev->wiphy.wowlan_config) { cfg80211_leave_all(rdev); @@ -118,6 +120,7 @@ static int wiphy_suspend(struct device *dev) ret = rdev_suspend(rdev, NULL); } } + wiphy_unlock(&rdev->wiphy); rtnl_unlock(); return ret; @@ -132,8 +135,10 @@ static int wiphy_resume(struct device *dev) cfg80211_bss_age(rdev, ktime_get_boottime_seconds() - rdev->suspend_at); rtnl_lock(); + wiphy_lock(&rdev->wiphy); if (rdev->wiphy.registered && rdev->ops->resume) ret = rdev_resume(rdev); + wiphy_unlock(&rdev->wiphy); rtnl_unlock(); return ret; diff --git a/net/wireless/util.c b/net/wireless/util.c index b4acc805114b..1bf0200f562a 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -997,7 +997,7 @@ void cfg80211_process_rdev_events(struct cfg80211_registered_device *rdev) { struct wireless_dev *wdev; - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) cfg80211_process_wdev_events(wdev); @@ -1010,7 +1010,7 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev, int err; enum nl80211_iftype otype = dev->ieee80211_ptr->iftype; - ASSERT_RTNL(); + lockdep_assert_held(&rdev->wiphy.mtx); /* don't support changing VLANs, you just re-create them */ if (otype == NL80211_IFTYPE_AP_VLAN) @@ -1188,6 +1188,25 @@ static u32 cfg80211_calculate_bitrate_dmg(struct rate_info *rate) return __mcs2bitrate[rate->mcs]; } +static u32 cfg80211_calculate_bitrate_extended_sc_dmg(struct rate_info *rate) +{ + static const u32 __mcs2bitrate[] = { + [6 - 6] = 26950, /* MCS 9.1 : 2695.0 mbps */ + [7 - 6] = 50050, /* MCS 12.1 */ + [8 - 6] = 53900, + [9 - 6] = 57750, + [10 - 6] = 63900, + [11 - 6] = 75075, + [12 - 6] = 80850, + }; + + /* Extended SC MCS not defined for base MCS below 6 or above 12 */ + if (WARN_ON_ONCE(rate->mcs < 6 || rate->mcs > 12)) + return 0; + + return __mcs2bitrate[rate->mcs - 6]; +} + static u32 cfg80211_calculate_bitrate_edmg(struct rate_info *rate) { static const u32 __mcs2bitrate[] = { @@ -1224,7 +1243,7 @@ static u32 cfg80211_calculate_bitrate_edmg(struct rate_info *rate) static u32 cfg80211_calculate_bitrate_vht(struct rate_info *rate) { - static const u32 base[4][10] = { + static const u32 base[4][12] = { { 6500000, 13000000, 19500000, @@ -1235,7 +1254,9 @@ static u32 cfg80211_calculate_bitrate_vht(struct rate_info *rate) 65000000, 78000000, /* not in the spec, but some devices use this: */ - 86500000, + 86700000, + 97500000, + 108300000, }, { 13500000, 27000000, @@ -1247,6 +1268,8 @@ static u32 cfg80211_calculate_bitrate_vht(struct rate_info *rate) 135000000, 162000000, 180000000, + 202500000, + 225000000, }, { 29300000, 58500000, @@ -1258,6 +1281,8 @@ static u32 cfg80211_calculate_bitrate_vht(struct rate_info *rate) 292500000, 351000000, 390000000, + 438800000, + 487500000, }, { 58500000, 117000000, @@ -1269,12 +1294,14 @@ static u32 cfg80211_calculate_bitrate_vht(struct rate_info *rate) 585000000, 702000000, 780000000, + 877500000, + 975000000, }, }; u32 bitrate; int idx; - if (rate->mcs > 9) + if (rate->mcs > 11) goto warn; switch (rate->bw) { @@ -1398,6 +1425,8 @@ u32 cfg80211_calculate_bitrate(struct rate_info *rate) return cfg80211_calculate_bitrate_ht(rate); if (rate->flags & RATE_INFO_FLAGS_DMG) return cfg80211_calculate_bitrate_dmg(rate); + if (rate->flags & RATE_INFO_FLAGS_EXTENDED_SC_DMG) + return cfg80211_calculate_bitrate_extended_sc_dmg(rate); if (rate->flags & RATE_INFO_FLAGS_EDMG) return cfg80211_calculate_bitrate_edmg(rate); if (rate->flags & RATE_INFO_FLAGS_VHT_MCS) diff --git a/net/wireless/wext-compat.c b/net/wireless/wext-compat.c index fd9ad74972fb..2e35cb78221e 100644 --- a/net/wireless/wext-compat.c +++ b/net/wireless/wext-compat.c @@ -7,7 +7,7 @@ * we directly assign the wireless handlers of wireless interfaces. * * Copyright 2008-2009 Johannes Berg <johannes@sipsolutions.net> - * Copyright (C) 2019 Intel Corporation + * Copyright (C) 2019-2021 Intel Corporation */ #include <linux/export.h> @@ -253,17 +253,23 @@ int cfg80211_wext_siwrts(struct net_device *dev, u32 orts = wdev->wiphy->rts_threshold; int err; - if (rts->disabled || !rts->fixed) + wiphy_lock(&rdev->wiphy); + if (rts->disabled || !rts->fixed) { wdev->wiphy->rts_threshold = (u32) -1; - else if (rts->value < 0) - return -EINVAL; - else + } else if (rts->value < 0) { + err = -EINVAL; + goto out; + } else { wdev->wiphy->rts_threshold = rts->value; + } err = rdev_set_wiphy_params(rdev, WIPHY_PARAM_RTS_THRESHOLD); + if (err) wdev->wiphy->rts_threshold = orts; +out: + wiphy_unlock(&rdev->wiphy); return err; } EXPORT_WEXT_HANDLER(cfg80211_wext_siwrts); @@ -291,11 +297,13 @@ int cfg80211_wext_siwfrag(struct net_device *dev, u32 ofrag = wdev->wiphy->frag_threshold; int err; - if (frag->disabled || !frag->fixed) + wiphy_lock(&rdev->wiphy); + if (frag->disabled || !frag->fixed) { wdev->wiphy->frag_threshold = (u32) -1; - else if (frag->value < 256) - return -EINVAL; - else { + } else if (frag->value < 256) { + err = -EINVAL; + goto out; + } else { /* Fragment length must be even, so strip LSB. */ wdev->wiphy->frag_threshold = frag->value & ~0x1; } @@ -303,6 +311,8 @@ int cfg80211_wext_siwfrag(struct net_device *dev, err = rdev_set_wiphy_params(rdev, WIPHY_PARAM_FRAG_THRESHOLD); if (err) wdev->wiphy->frag_threshold = ofrag; +out: + wiphy_unlock(&rdev->wiphy); return err; } @@ -337,6 +347,7 @@ static int cfg80211_wext_siwretry(struct net_device *dev, (retry->flags & IW_RETRY_TYPE) != IW_RETRY_LIMIT) return -EINVAL; + wiphy_lock(&rdev->wiphy); if (retry->flags & IW_RETRY_LONG) { wdev->wiphy->retry_long = retry->value; changed |= WIPHY_PARAM_RETRY_LONG; @@ -355,6 +366,7 @@ static int cfg80211_wext_siwretry(struct net_device *dev, wdev->wiphy->retry_short = oshort; wdev->wiphy->retry_long = olong; } + wiphy_unlock(&rdev->wiphy); return err; } @@ -577,15 +589,18 @@ static int cfg80211_wext_siwencode(struct net_device *dev, !rdev->ops->set_default_key) return -EOPNOTSUPP; + wiphy_lock(&rdev->wiphy); idx = erq->flags & IW_ENCODE_INDEX; if (idx == 0) { idx = wdev->wext.default_key; if (idx < 0) idx = 0; - } else if (idx < 1 || idx > 4) - return -EINVAL; - else + } else if (idx < 1 || idx > 4) { + err = -EINVAL; + goto out; + } else { idx--; + } if (erq->flags & IW_ENCODE_DISABLED) remove = true; @@ -599,22 +614,28 @@ static int cfg80211_wext_siwencode(struct net_device *dev, if (!err) wdev->wext.default_key = idx; wdev_unlock(wdev); - return err; + goto out; } memset(¶ms, 0, sizeof(params)); params.key = keybuf; params.key_len = erq->length; - if (erq->length == 5) + if (erq->length == 5) { params.cipher = WLAN_CIPHER_SUITE_WEP40; - else if (erq->length == 13) + } else if (erq->length == 13) { params.cipher = WLAN_CIPHER_SUITE_WEP104; - else if (!remove) - return -EINVAL; + } else if (!remove) { + err = -EINVAL; + goto out; + } + + err = cfg80211_set_encryption(rdev, dev, false, NULL, remove, + wdev->wext.default_key == -1, + idx, ¶ms); +out: + wiphy_unlock(&rdev->wiphy); - return cfg80211_set_encryption(rdev, dev, false, NULL, remove, - wdev->wext.default_key == -1, - idx, ¶ms); + return err; } static int cfg80211_wext_siwencodeext(struct net_device *dev, @@ -754,38 +775,61 @@ static int cfg80211_wext_siwfreq(struct net_device *dev, struct cfg80211_chan_def chandef = { .width = NL80211_CHAN_WIDTH_20_NOHT, }; - int freq; + int freq, ret; + + wiphy_lock(&rdev->wiphy); switch (wdev->iftype) { case NL80211_IFTYPE_STATION: - return cfg80211_mgd_wext_siwfreq(dev, info, wextfreq, extra); + ret = cfg80211_mgd_wext_siwfreq(dev, info, wextfreq, extra); + break; case NL80211_IFTYPE_ADHOC: - return cfg80211_ibss_wext_siwfreq(dev, info, wextfreq, extra); + ret = cfg80211_ibss_wext_siwfreq(dev, info, wextfreq, extra); + break; case NL80211_IFTYPE_MONITOR: freq = cfg80211_wext_freq(wextfreq); - if (freq < 0) - return freq; - if (freq == 0) - return -EINVAL; + if (freq < 0) { + ret = freq; + break; + } + if (freq == 0) { + ret = -EINVAL; + break; + } chandef.center_freq1 = freq; chandef.chan = ieee80211_get_channel(&rdev->wiphy, freq); - if (!chandef.chan) - return -EINVAL; - return cfg80211_set_monitor_channel(rdev, &chandef); + if (!chandef.chan) { + ret = -EINVAL; + break; + } + ret = cfg80211_set_monitor_channel(rdev, &chandef); + break; case NL80211_IFTYPE_MESH_POINT: freq = cfg80211_wext_freq(wextfreq); - if (freq < 0) - return freq; - if (freq == 0) - return -EINVAL; + if (freq < 0) { + ret = freq; + break; + } + if (freq == 0) { + ret = -EINVAL; + break; + } chandef.center_freq1 = freq; chandef.chan = ieee80211_get_channel(&rdev->wiphy, freq); - if (!chandef.chan) - return -EINVAL; - return cfg80211_set_mesh_channel(rdev, wdev, &chandef); + if (!chandef.chan) { + ret = -EINVAL; + break; + } + ret = cfg80211_set_mesh_channel(rdev, wdev, &chandef); + break; default: - return -EOPNOTSUPP; + ret = -EOPNOTSUPP; + break; } + + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_giwfreq(struct net_device *dev, @@ -797,24 +841,35 @@ static int cfg80211_wext_giwfreq(struct net_device *dev, struct cfg80211_chan_def chandef = {}; int ret; + wiphy_lock(&rdev->wiphy); switch (wdev->iftype) { case NL80211_IFTYPE_STATION: - return cfg80211_mgd_wext_giwfreq(dev, info, freq, extra); + ret = cfg80211_mgd_wext_giwfreq(dev, info, freq, extra); + break; case NL80211_IFTYPE_ADHOC: - return cfg80211_ibss_wext_giwfreq(dev, info, freq, extra); + ret = cfg80211_ibss_wext_giwfreq(dev, info, freq, extra); + break; case NL80211_IFTYPE_MONITOR: - if (!rdev->ops->get_channel) - return -EINVAL; + if (!rdev->ops->get_channel) { + ret = -EINVAL; + break; + } ret = rdev_get_channel(rdev, wdev, &chandef); if (ret) - return ret; + break; freq->m = chandef.chan->center_freq; freq->e = 6; - return 0; + ret = 0; + break; default: - return -EINVAL; + ret = -EINVAL; + break; } + + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_siwtxpower(struct net_device *dev, @@ -825,6 +880,7 @@ static int cfg80211_wext_siwtxpower(struct net_device *dev, struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); enum nl80211_tx_power_setting type; int dbm = 0; + int ret; if ((data->txpower.flags & IW_TXPOW_TYPE) != IW_TXPOW_DBM) return -EINVAL; @@ -866,7 +922,11 @@ static int cfg80211_wext_siwtxpower(struct net_device *dev, return 0; } - return rdev_set_tx_power(rdev, wdev, type, DBM_TO_MBM(dbm)); + wiphy_lock(&rdev->wiphy); + ret = rdev_set_tx_power(rdev, wdev, type, DBM_TO_MBM(dbm)); + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_giwtxpower(struct net_device *dev, @@ -885,7 +945,9 @@ static int cfg80211_wext_giwtxpower(struct net_device *dev, if (!rdev->ops->get_tx_power) return -EOPNOTSUPP; + wiphy_lock(&rdev->wiphy); err = rdev_get_tx_power(rdev, wdev, &val); + wiphy_unlock(&rdev->wiphy); if (err) return err; @@ -1125,7 +1187,9 @@ static int cfg80211_wext_siwpower(struct net_device *dev, timeout = wrq->value / 1000; } + wiphy_lock(&rdev->wiphy); err = rdev_set_power_mgmt(rdev, dev, ps, timeout); + wiphy_unlock(&rdev->wiphy); if (err) return err; @@ -1156,7 +1220,7 @@ static int cfg80211_wext_siwrate(struct net_device *dev, struct cfg80211_bitrate_mask mask; u32 fixed, maxrate; struct ieee80211_supported_band *sband; - int band, ridx; + int band, ridx, ret; bool match = false; if (!rdev->ops->set_bitrate_mask) @@ -1195,7 +1259,11 @@ static int cfg80211_wext_siwrate(struct net_device *dev, if (!match) return -EINVAL; - return rdev_set_bitrate_mask(rdev, dev, NULL, &mask); + wiphy_lock(&rdev->wiphy); + ret = rdev_set_bitrate_mask(rdev, dev, NULL, &mask); + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_giwrate(struct net_device *dev, @@ -1224,7 +1292,9 @@ static int cfg80211_wext_giwrate(struct net_device *dev, if (err) return err; + wiphy_lock(&rdev->wiphy); err = rdev_get_station(rdev, dev, addr, &sinfo); + wiphy_unlock(&rdev->wiphy); if (err) return err; @@ -1249,6 +1319,7 @@ static struct iw_statistics *cfg80211_wireless_stats(struct net_device *dev) static struct iw_statistics wstats; static struct station_info sinfo = {}; u8 bssid[ETH_ALEN]; + int ret; if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) return NULL; @@ -1267,7 +1338,11 @@ static struct iw_statistics *cfg80211_wireless_stats(struct net_device *dev) memset(&sinfo, 0, sizeof(sinfo)); - if (rdev_get_station(rdev, dev, bssid, &sinfo)) + wiphy_lock(&rdev->wiphy); + ret = rdev_get_station(rdev, dev, bssid, &sinfo); + wiphy_unlock(&rdev->wiphy); + + if (ret) return NULL; memset(&wstats, 0, sizeof(wstats)); @@ -1318,15 +1393,24 @@ static int cfg80211_wext_siwap(struct net_device *dev, struct sockaddr *ap_addr, char *extra) { struct wireless_dev *wdev = dev->ieee80211_ptr; + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); + int ret; + wiphy_lock(&rdev->wiphy); switch (wdev->iftype) { case NL80211_IFTYPE_ADHOC: - return cfg80211_ibss_wext_siwap(dev, info, ap_addr, extra); + ret = cfg80211_ibss_wext_siwap(dev, info, ap_addr, extra); + break; case NL80211_IFTYPE_STATION: - return cfg80211_mgd_wext_siwap(dev, info, ap_addr, extra); + ret = cfg80211_mgd_wext_siwap(dev, info, ap_addr, extra); + break; default: - return -EOPNOTSUPP; + ret = -EOPNOTSUPP; + break; } + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_giwap(struct net_device *dev, @@ -1334,15 +1418,24 @@ static int cfg80211_wext_giwap(struct net_device *dev, struct sockaddr *ap_addr, char *extra) { struct wireless_dev *wdev = dev->ieee80211_ptr; + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); + int ret; + wiphy_lock(&rdev->wiphy); switch (wdev->iftype) { case NL80211_IFTYPE_ADHOC: - return cfg80211_ibss_wext_giwap(dev, info, ap_addr, extra); + ret = cfg80211_ibss_wext_giwap(dev, info, ap_addr, extra); + break; case NL80211_IFTYPE_STATION: - return cfg80211_mgd_wext_giwap(dev, info, ap_addr, extra); + ret = cfg80211_mgd_wext_giwap(dev, info, ap_addr, extra); + break; default: - return -EOPNOTSUPP; + ret = -EOPNOTSUPP; + break; } + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_siwessid(struct net_device *dev, @@ -1350,15 +1443,24 @@ static int cfg80211_wext_siwessid(struct net_device *dev, struct iw_point *data, char *ssid) { struct wireless_dev *wdev = dev->ieee80211_ptr; + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); + int ret; + wiphy_lock(&rdev->wiphy); switch (wdev->iftype) { case NL80211_IFTYPE_ADHOC: - return cfg80211_ibss_wext_siwessid(dev, info, data, ssid); + ret = cfg80211_ibss_wext_siwessid(dev, info, data, ssid); + break; case NL80211_IFTYPE_STATION: - return cfg80211_mgd_wext_siwessid(dev, info, data, ssid); + ret = cfg80211_mgd_wext_siwessid(dev, info, data, ssid); + break; default: - return -EOPNOTSUPP; + ret = -EOPNOTSUPP; + break; } + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_giwessid(struct net_device *dev, @@ -1366,18 +1468,27 @@ static int cfg80211_wext_giwessid(struct net_device *dev, struct iw_point *data, char *ssid) { struct wireless_dev *wdev = dev->ieee80211_ptr; + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); + int ret; data->flags = 0; data->length = 0; + wiphy_lock(&rdev->wiphy); switch (wdev->iftype) { case NL80211_IFTYPE_ADHOC: - return cfg80211_ibss_wext_giwessid(dev, info, data, ssid); + ret = cfg80211_ibss_wext_giwessid(dev, info, data, ssid); + break; case NL80211_IFTYPE_STATION: - return cfg80211_mgd_wext_giwessid(dev, info, data, ssid); + ret = cfg80211_mgd_wext_giwessid(dev, info, data, ssid); + break; default: - return -EOPNOTSUPP; + ret = -EOPNOTSUPP; + break; } + wiphy_unlock(&rdev->wiphy); + + return ret; } static int cfg80211_wext_siwpmksa(struct net_device *dev, @@ -1388,6 +1499,7 @@ static int cfg80211_wext_siwpmksa(struct net_device *dev, struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); struct cfg80211_pmksa cfg_pmksa; struct iw_pmksa *pmksa = (struct iw_pmksa *)extra; + int ret; memset(&cfg_pmksa, 0, sizeof(struct cfg80211_pmksa)); @@ -1397,28 +1509,39 @@ static int cfg80211_wext_siwpmksa(struct net_device *dev, cfg_pmksa.bssid = pmksa->bssid.sa_data; cfg_pmksa.pmkid = pmksa->pmkid; + wiphy_lock(&rdev->wiphy); switch (pmksa->cmd) { case IW_PMKSA_ADD: - if (!rdev->ops->set_pmksa) - return -EOPNOTSUPP; - - return rdev_set_pmksa(rdev, dev, &cfg_pmksa); + if (!rdev->ops->set_pmksa) { + ret = -EOPNOTSUPP; + break; + } + ret = rdev_set_pmksa(rdev, dev, &cfg_pmksa); + break; case IW_PMKSA_REMOVE: - if (!rdev->ops->del_pmksa) - return -EOPNOTSUPP; - - return rdev_del_pmksa(rdev, dev, &cfg_pmksa); + if (!rdev->ops->del_pmksa) { + ret = -EOPNOTSUPP; + break; + } + ret = rdev_del_pmksa(rdev, dev, &cfg_pmksa); + break; case IW_PMKSA_FLUSH: - if (!rdev->ops->flush_pmksa) - return -EOPNOTSUPP; - - return rdev_flush_pmksa(rdev, dev); + if (!rdev->ops->flush_pmksa) { + ret = -EOPNOTSUPP; + break; + } + ret = rdev_flush_pmksa(rdev, dev); + break; default: - return -EOPNOTSUPP; + ret = -EOPNOTSUPP; + break; } + wiphy_unlock(&rdev->wiphy); + + return ret; } #define DEFINE_WEXT_COMPAT_STUB(func, type) \ diff --git a/net/wireless/wext-sme.c b/net/wireless/wext-sme.c index 73df23570d43..193a18a53142 100644 --- a/net/wireless/wext-sme.c +++ b/net/wireless/wext-sme.c @@ -3,7 +3,7 @@ * cfg80211 wext compat for managed mode. * * Copyright 2009 Johannes Berg <johannes@sipsolutions.net> - * Copyright (C) 2009 Intel Corporation. All rights reserved. + * Copyright (C) 2009, 2020-2021 Intel Corporation. */ #include <linux/export.h> @@ -379,6 +379,7 @@ int cfg80211_wext_siwmlme(struct net_device *dev, if (mlme->addr.sa_family != ARPHRD_ETHER) return -EINVAL; + wiphy_lock(&rdev->wiphy); wdev_lock(wdev); switch (mlme->cmd) { case IW_MLME_DEAUTH: @@ -390,6 +391,7 @@ int cfg80211_wext_siwmlme(struct net_device *dev, break; } wdev_unlock(wdev); + wiphy_unlock(&rdev->wiphy); return err; } diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 8037b04a9edd..4a83117507f5 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -108,9 +108,9 @@ EXPORT_SYMBOL(xsk_get_pool_from_qid); void xsk_clear_pool_at_qid(struct net_device *dev, u16 queue_id) { - if (queue_id < dev->real_num_rx_queues) + if (queue_id < dev->num_rx_queues) dev->_rx[queue_id].pool = NULL; - if (queue_id < dev->real_num_tx_queues) + if (queue_id < dev->num_tx_queues) dev->_tx[queue_id].pool = NULL; } |