aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rw-r--r--Documentation/networking/xfrm_device.txt3
-rw-r--r--drivers/net/ethernet/sfc/ptp.c2
-rw-r--r--include/linux/netdevice.h1
-rw-r--r--include/net/xfrm.h12
-rw-r--r--net/ipv6/route.c140
-rw-r--r--net/smc/af_smc.c152
-rw-r--r--net/smc/smc.h5
-rw-r--r--net/smc/smc_cdc.c20
-rw-r--r--net/smc/smc_close.c96
-rw-r--r--net/smc/smc_close.h1
-rw-r--r--net/smc/smc_core.c6
-rw-r--r--net/smc/smc_ib.c38
-rw-r--r--net/xfrm/xfrm_device.c11
-rw-r--r--net/xfrm/xfrm_replay.c2
14 files changed, 322 insertions, 167 deletions
diff --git a/Documentation/networking/xfrm_device.txt b/Documentation/networking/xfrm_device.txt
index 2d9d588cd34b..50c34ca65efe 100644
--- a/Documentation/networking/xfrm_device.txt
+++ b/Documentation/networking/xfrm_device.txt
@@ -41,6 +41,7 @@ struct xfrmdev_ops {
void (*xdo_dev_state_free) (struct xfrm_state *x);
bool (*xdo_dev_offload_ok) (struct sk_buff *skb,
struct xfrm_state *x);
+ void (*xdo_dev_state_advance_esn) (struct xfrm_state *x);
};
The NIC driver offering ipsec offload will need to implement these
@@ -117,6 +118,8 @@ the stack in xfrm_input().
hand the packet to napi_gro_receive() as usual
+In ESN mode, xdo_dev_state_advance_esn() is called from xfrm_replay_advance_esn().
+Driver will check packet seq number and update HW ESN state machine if needed.
When the SA is removed by the user, the driver's xdo_dev_state_delete()
is asked to disable the offload. Later, xdo_dev_state_free() is called
diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index 433d29d6bc95..f1cc2ed76029 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -643,7 +643,7 @@ static int efx_ptp_get_attributes(struct efx_nic *efx)
case MC_CMD_PTP_OUT_GET_ATTRIBUTES_SECONDS_QTR_NANOSECONDS:
ptp->ns_to_nic_time = efx_ptp_ns_to_s_qns;
ptp->nic_to_kernel_time = efx_ptp_s_qns_to_ktime_correction;
- ptp->nic_time.minor_max = 4000000000;
+ ptp->nic_time.minor_max = 4000000000UL;
ptp->nic_time.sync_event_minor_shift = 24;
break;
default:
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 24a62d590350..cd46d3d63aa0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -851,6 +851,7 @@ struct xfrmdev_ops {
void (*xdo_dev_state_free) (struct xfrm_state *x);
bool (*xdo_dev_offload_ok) (struct sk_buff *skb,
struct xfrm_state *x);
+ void (*xdo_dev_state_advance_esn) (struct xfrm_state *x);
};
#endif
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 2e6d4fe6b0ba..7d2077665c0b 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1904,6 +1904,14 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x,
struct xfrm_user_offload *xuo);
bool xfrm_dev_offload_ok(struct sk_buff *skb, struct xfrm_state *x);
+static inline void xfrm_dev_state_advance_esn(struct xfrm_state *x)
+{
+ struct xfrm_state_offload *xso = &x->xso;
+
+ if (xso->dev && xso->dev->xfrmdev_ops->xdo_dev_state_advance_esn)
+ xso->dev->xfrmdev_ops->xdo_dev_state_advance_esn(x);
+}
+
static inline bool xfrm_dst_offload_ok(struct dst_entry *dst)
{
struct xfrm_state *x = dst->xfrm;
@@ -1974,6 +1982,10 @@ static inline bool xfrm_dev_offload_ok(struct sk_buff *skb, struct xfrm_state *x
return false;
}
+static inline void xfrm_dev_state_advance_esn(struct xfrm_state *x)
+{
+}
+
static inline bool xfrm_dst_offload_ok(struct dst_entry *dst)
{
return false;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index aa4411c81e7e..fe3966a9c999 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2440,7 +2440,8 @@ static int ip6_convert_metrics(struct mx6_config *mxc,
static struct rt6_info *ip6_nh_lookup_table(struct net *net,
struct fib6_config *cfg,
- const struct in6_addr *gw_addr)
+ const struct in6_addr *gw_addr,
+ u32 tbid, int flags)
{
struct flowi6 fl6 = {
.flowi6_oif = cfg->fc_ifindex,
@@ -2449,15 +2450,15 @@ static struct rt6_info *ip6_nh_lookup_table(struct net *net,
};
struct fib6_table *table;
struct rt6_info *rt;
- int flags = RT6_LOOKUP_F_IFACE | RT6_LOOKUP_F_IGNORE_LINKSTATE;
- table = fib6_get_table(net, cfg->fc_table);
+ table = fib6_get_table(net, tbid);
if (!table)
return NULL;
if (!ipv6_addr_any(&cfg->fc_prefsrc))
flags |= RT6_LOOKUP_F_HAS_SADDR;
+ flags |= RT6_LOOKUP_F_IGNORE_LINKSTATE;
rt = ip6_pol_route(net, table, cfg->fc_ifindex, &fl6, flags);
/* if table lookup failed, fall back to full lookup */
@@ -2469,6 +2470,82 @@ static struct rt6_info *ip6_nh_lookup_table(struct net *net,
return rt;
}
+static int ip6_route_check_nh_onlink(struct net *net,
+ struct fib6_config *cfg,
+ struct net_device *dev,
+ struct netlink_ext_ack *extack)
+{
+ u32 tbid = l3mdev_fib_table(dev) ? : RT_TABLE_LOCAL;
+ const struct in6_addr *gw_addr = &cfg->fc_gateway;
+ u32 flags = RTF_LOCAL | RTF_ANYCAST | RTF_REJECT;
+ struct rt6_info *grt;
+ int err;
+
+ err = 0;
+ grt = ip6_nh_lookup_table(net, cfg, gw_addr, tbid, 0);
+ if (grt) {
+ if (grt->rt6i_flags & flags || dev != grt->dst.dev) {
+ NL_SET_ERR_MSG(extack, "Nexthop has invalid gateway");
+ err = -EINVAL;
+ }
+
+ ip6_rt_put(grt);
+ }
+
+ return err;
+}
+
+static int ip6_route_check_nh(struct net *net,
+ struct fib6_config *cfg,
+ struct net_device **_dev,
+ struct inet6_dev **idev)
+{
+ const struct in6_addr *gw_addr = &cfg->fc_gateway;
+ struct net_device *dev = _dev ? *_dev : NULL;
+ struct rt6_info *grt = NULL;
+ int err = -EHOSTUNREACH;
+
+ if (cfg->fc_table) {
+ int flags = RT6_LOOKUP_F_IFACE;
+
+ grt = ip6_nh_lookup_table(net, cfg, gw_addr,
+ cfg->fc_table, flags);
+ if (grt) {
+ if (grt->rt6i_flags & RTF_GATEWAY ||
+ (dev && dev != grt->dst.dev)) {
+ ip6_rt_put(grt);
+ grt = NULL;
+ }
+ }
+ }
+
+ if (!grt)
+ grt = rt6_lookup(net, gw_addr, NULL, cfg->fc_ifindex, 1);
+
+ if (!grt)
+ goto out;
+
+ if (dev) {
+ if (dev != grt->dst.dev) {
+ ip6_rt_put(grt);
+ goto out;
+ }
+ } else {
+ *_dev = dev = grt->dst.dev;
+ *idev = grt->rt6i_idev;
+ dev_hold(dev);
+ in6_dev_hold(grt->rt6i_idev);
+ }
+
+ if (!(grt->rt6i_flags & RTF_GATEWAY))
+ err = 0;
+
+ ip6_rt_put(grt);
+
+out:
+ return err;
+}
+
static struct rt6_info *ip6_route_info_create(struct fib6_config *cfg,
struct netlink_ext_ack *extack)
{
@@ -2520,6 +2597,21 @@ static struct rt6_info *ip6_route_info_create(struct fib6_config *cfg,
if (cfg->fc_metric == 0)
cfg->fc_metric = IP6_RT_PRIO_USER;
+ if (cfg->fc_flags & RTNH_F_ONLINK) {
+ if (!dev) {
+ NL_SET_ERR_MSG(extack,
+ "Nexthop device required for onlink");
+ err = -ENODEV;
+ goto out;
+ }
+
+ if (!(dev->flags & IFF_UP)) {
+ NL_SET_ERR_MSG(extack, "Nexthop device is not up");
+ err = -ENETDOWN;
+ goto out;
+ }
+ }
+
err = -ENOBUFS;
if (cfg->fc_nlinfo.nlh &&
!(cfg->fc_nlinfo.nlh->nlmsg_flags & NLM_F_CREATE)) {
@@ -2664,8 +2756,6 @@ static struct rt6_info *ip6_route_info_create(struct fib6_config *cfg,
rt->rt6i_gateway = *gw_addr;
if (gwa_type != (IPV6_ADDR_LINKLOCAL|IPV6_ADDR_UNICAST)) {
- struct rt6_info *grt = NULL;
-
/* IPv6 strictly inhibits using not link-local
addresses as nexthop address.
Otherwise, router will not able to send redirects.
@@ -2682,40 +2772,12 @@ static struct rt6_info *ip6_route_info_create(struct fib6_config *cfg,
goto out;
}
- if (cfg->fc_table) {
- grt = ip6_nh_lookup_table(net, cfg, gw_addr);
-
- if (grt) {
- if (grt->rt6i_flags & RTF_GATEWAY ||
- (dev && dev != grt->dst.dev)) {
- ip6_rt_put(grt);
- grt = NULL;
- }
- }
- }
-
- if (!grt)
- grt = rt6_lookup(net, gw_addr, NULL,
- cfg->fc_ifindex, 1);
-
- err = -EHOSTUNREACH;
- if (!grt)
- goto out;
- if (dev) {
- if (dev != grt->dst.dev) {
- ip6_rt_put(grt);
- goto out;
- }
+ if (cfg->fc_flags & RTNH_F_ONLINK) {
+ err = ip6_route_check_nh_onlink(net, cfg, dev,
+ extack);
} else {
- dev = grt->dst.dev;
- idev = grt->rt6i_idev;
- dev_hold(dev);
- in6_dev_hold(grt->rt6i_idev);
+ err = ip6_route_check_nh(net, cfg, &dev, &idev);
}
- if (!(grt->rt6i_flags & RTF_GATEWAY))
- err = 0;
- ip6_rt_put(grt);
-
if (err)
goto out;
}
@@ -2757,6 +2819,7 @@ install_route:
if (!(rt->rt6i_flags & (RTF_LOCAL | RTF_ANYCAST)) &&
!netif_carrier_ok(dev))
rt->rt6i_nh_flags |= RTNH_F_LINKDOWN;
+ rt->rt6i_nh_flags |= (cfg->fc_flags & RTNH_F_ONLINK);
rt->dst.dev = dev;
rt->rt6i_idev = idev;
rt->rt6i_table = table;
@@ -3826,6 +3889,8 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
if (rtm->rtm_flags & RTM_F_CLONED)
cfg->fc_flags |= RTF_CACHE;
+ cfg->fc_flags |= (rtm->rtm_flags & RTNH_F_ONLINK);
+
cfg->fc_nlinfo.portid = NETLINK_CB(skb).portid;
cfg->fc_nlinfo.nlh = nlh;
cfg->fc_nlinfo.nl_net = sock_net(skb->sk);
@@ -4231,6 +4296,7 @@ static int rt6_nexthop_info(struct sk_buff *skb, struct rt6_info *rt,
goto nla_put_failure;
}
+ *flags |= (rt->rt6i_nh_flags & RTNH_F_ONLINK);
if (rt->rt6i_nh_flags & RTNH_F_OFFLOAD)
*flags |= RTNH_F_OFFLOAD;
diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index cf0e11978b66..267e68379110 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -115,7 +115,6 @@ static int smc_release(struct socket *sock)
goto out;
smc = smc_sk(sk);
- sock_hold(sk);
if (sk->sk_state == SMC_LISTEN)
/* smc_close_non_accepted() is called and acquires
* sock lock for child sockets again
@@ -124,10 +123,7 @@ static int smc_release(struct socket *sock)
else
lock_sock(sk);
- if (smc->use_fallback) {
- sk->sk_state = SMC_CLOSED;
- sk->sk_state_change(sk);
- } else {
+ if (!smc->use_fallback) {
rc = smc_close_active(smc);
sock_set_flag(sk, SOCK_DEAD);
sk->sk_shutdown |= SHUTDOWN_MASK;
@@ -136,20 +132,21 @@ static int smc_release(struct socket *sock)
sock_release(smc->clcsock);
smc->clcsock = NULL;
}
+ if (smc->use_fallback) {
+ sock_put(sk); /* passive closing */
+ sk->sk_state = SMC_CLOSED;
+ sk->sk_state_change(sk);
+ }
/* detach socket */
sock_orphan(sk);
sock->sk = NULL;
- if (smc->use_fallback) {
- schedule_delayed_work(&smc->sock_put_work, TCP_TIMEWAIT_LEN);
- } else if (sk->sk_state == SMC_CLOSED) {
+ if (!smc->use_fallback && sk->sk_state == SMC_CLOSED)
smc_conn_free(&smc->conn);
- schedule_delayed_work(&smc->sock_put_work,
- SMC_CLOSE_SOCK_PUT_DELAY);
- }
release_sock(sk);
- sock_put(sk);
+ sk->sk_prot->unhash(sk);
+ sock_put(sk); /* final sock_put */
out:
return rc;
}
@@ -181,7 +178,6 @@ static struct sock *smc_sock_alloc(struct net *net, struct socket *sock)
INIT_WORK(&smc->tcp_listen_work, smc_tcp_listen_work);
INIT_LIST_HEAD(&smc->accept_q);
spin_lock_init(&smc->accept_q_lock);
- INIT_DELAYED_WORK(&smc->sock_put_work, smc_close_sock_put_work);
sk->sk_prot->hash(sk);
sk_refcnt_debug_inc(sk);
@@ -399,6 +395,8 @@ static int smc_connect_rdma(struct smc_sock *smc)
int rc = 0;
u8 ibport;
+ sock_hold(&smc->sk); /* sock put in passive closing */
+
if (!tcp_sk(smc->clcsock->sk)->syn_smc) {
/* peer has not signalled SMC-capability */
smc->use_fallback = true;
@@ -542,6 +540,8 @@ out_err_unlock:
mutex_unlock(&smc_create_lgr_pending);
smc_conn_free(&smc->conn);
out_err:
+ if (smc->sk.sk_state == SMC_INIT)
+ sock_put(&smc->sk); /* passive closing */
return rc;
}
@@ -620,7 +620,7 @@ static int smc_clcsock_accept(struct smc_sock *lsmc, struct smc_sock **new_smc)
new_sk->sk_state = SMC_CLOSED;
sock_set_flag(new_sk, SOCK_DEAD);
new_sk->sk_prot->unhash(new_sk);
- sock_put(new_sk);
+ sock_put(new_sk); /* final */
*new_smc = NULL;
goto out;
}
@@ -637,7 +637,7 @@ static void smc_accept_enqueue(struct sock *parent, struct sock *sk)
{
struct smc_sock *par = smc_sk(parent);
- sock_hold(sk);
+ sock_hold(sk); /* sock_put in smc_accept_unlink () */
spin_lock(&par->accept_q_lock);
list_add_tail(&smc_sk(sk)->accept_q, &par->accept_q);
spin_unlock(&par->accept_q_lock);
@@ -653,7 +653,7 @@ static void smc_accept_unlink(struct sock *sk)
list_del_init(&smc_sk(sk)->accept_q);
spin_unlock(&par->accept_q_lock);
sk_acceptq_removed(&smc_sk(sk)->listen_smc->sk);
- sock_put(sk);
+ sock_put(sk); /* sock_hold in smc_accept_enqueue */
}
/* remove a sock from the accept queue to bind it to a new socket created
@@ -670,8 +670,12 @@ struct sock *smc_accept_dequeue(struct sock *parent,
smc_accept_unlink(new_sk);
if (new_sk->sk_state == SMC_CLOSED) {
+ if (isk->clcsock) {
+ sock_release(isk->clcsock);
+ isk->clcsock = NULL;
+ }
new_sk->sk_prot->unhash(new_sk);
- sock_put(new_sk);
+ sock_put(new_sk); /* final */
continue;
}
if (new_sock)
@@ -686,14 +690,11 @@ void smc_close_non_accepted(struct sock *sk)
{
struct smc_sock *smc = smc_sk(sk);
- sock_hold(sk);
lock_sock(sk);
if (!sk->sk_lingertime)
/* wait for peer closing */
sk->sk_lingertime = SMC_MAX_STREAM_WAIT_TIMEOUT;
- if (smc->use_fallback) {
- sk->sk_state = SMC_CLOSED;
- } else {
+ if (!smc->use_fallback) {
smc_close_active(smc);
sock_set_flag(sk, SOCK_DEAD);
sk->sk_shutdown |= SHUTDOWN_MASK;
@@ -706,14 +707,15 @@ void smc_close_non_accepted(struct sock *sk)
sock_release(tcp);
}
if (smc->use_fallback) {
- schedule_delayed_work(&smc->sock_put_work, TCP_TIMEWAIT_LEN);
- } else if (sk->sk_state == SMC_CLOSED) {
- smc_conn_free(&smc->conn);
- schedule_delayed_work(&smc->sock_put_work,
- SMC_CLOSE_SOCK_PUT_DELAY);
+ sock_put(sk); /* passive closing */
+ sk->sk_state = SMC_CLOSED;
+ } else {
+ if (sk->sk_state == SMC_CLOSED)
+ smc_conn_free(&smc->conn);
}
release_sock(sk);
- sock_put(sk);
+ sk->sk_prot->unhash(sk);
+ sock_put(sk); /* final sock_put */
}
static int smc_serv_conf_first_link(struct smc_sock *smc)
@@ -937,6 +939,8 @@ out_err_unlock:
smc_lgr_forget(new_smc->conn.lgr);
mutex_unlock(&smc_create_lgr_pending);
out_err:
+ if (newsmcsk->sk_state == SMC_INIT)
+ sock_put(&new_smc->sk); /* passive closing */
newsmcsk->sk_state = SMC_CLOSED;
smc_conn_free(&new_smc->conn);
goto enqueue; /* queue new sock with sk_err set */
@@ -963,12 +967,22 @@ static void smc_tcp_listen_work(struct work_struct *work)
sock_hold(lsk); /* sock_put in smc_listen_work */
INIT_WORK(&new_smc->smc_listen_work, smc_listen_work);
smc_copy_sock_settings_to_smc(new_smc);
- schedule_work(&new_smc->smc_listen_work);
+ sock_hold(&new_smc->sk); /* sock_put in passive closing */
+ if (!schedule_work(&new_smc->smc_listen_work))
+ sock_put(&new_smc->sk);
}
out:
+ if (lsmc->clcsock) {
+ sock_release(lsmc->clcsock);
+ lsmc->clcsock = NULL;
+ }
release_sock(lsk);
- lsk->sk_data_ready(lsk); /* no more listening, wake accept */
+ /* no more listening, wake up smc_close_wait_listen_clcsock and
+ * accept
+ */
+ lsk->sk_state_change(lsk);
+ sock_put(&lsmc->sk); /* sock_hold in smc_listen */
}
static int smc_listen(struct socket *sock, int backlog)
@@ -1002,7 +1016,9 @@ static int smc_listen(struct socket *sock, int backlog)
sk->sk_ack_backlog = 0;
sk->sk_state = SMC_LISTEN;
INIT_WORK(&smc->tcp_listen_work, smc_tcp_listen_work);
- schedule_work(&smc->tcp_listen_work);
+ sock_hold(sk); /* sock_hold in tcp_listen_worker */
+ if (!schedule_work(&smc->tcp_listen_work))
+ sock_put(sk);
out:
release_sock(sk);
@@ -1019,6 +1035,7 @@ static int smc_accept(struct socket *sock, struct socket *new_sock,
int rc = 0;
lsmc = smc_sk(sk);
+ sock_hold(sk); /* sock_put below */
lock_sock(sk);
if (lsmc->sk.sk_state != SMC_LISTEN) {
@@ -1053,6 +1070,7 @@ static int smc_accept(struct socket *sock, struct socket *new_sock,
out:
release_sock(sk);
+ sock_put(sk); /* sock_hold above */
return rc;
}
@@ -1122,21 +1140,15 @@ out:
static unsigned int smc_accept_poll(struct sock *parent)
{
- struct smc_sock *isk;
- struct sock *sk;
+ struct smc_sock *isk = smc_sk(parent);
+ int mask = 0;
- lock_sock(parent);
- list_for_each_entry(isk, &smc_sk(parent)->accept_q, accept_q) {
- sk = (struct sock *)isk;
+ spin_lock(&isk->accept_q_lock);
+ if (!list_empty(&isk->accept_q))
+ mask = POLLIN | POLLRDNORM;
+ spin_unlock(&isk->accept_q_lock);
- if (sk->sk_state == SMC_ACTIVE) {
- release_sock(parent);
- return POLLIN | POLLRDNORM;
- }
- }
- release_sock(parent);
-
- return 0;
+ return mask;
}
static unsigned int smc_poll(struct file *file, struct socket *sock,
@@ -1147,9 +1159,15 @@ static unsigned int smc_poll(struct file *file, struct socket *sock,
struct smc_sock *smc;
int rc;
+ if (!sk)
+ return POLLNVAL;
+
smc = smc_sk(sock->sk);
+ sock_hold(sk);
+ lock_sock(sk);
if ((sk->sk_state == SMC_INIT) || smc->use_fallback) {
/* delegate to CLC child sock */
+ release_sock(sk);
mask = smc->clcsock->ops->poll(file, smc->clcsock, wait);
/* if non-blocking connect finished ... */
lock_sock(sk);
@@ -1161,37 +1179,43 @@ static unsigned int smc_poll(struct file *file, struct socket *sock,
rc = smc_connect_rdma(smc);
if (rc < 0)
mask |= POLLERR;
- else
- /* success cases including fallback */
- mask |= POLLOUT | POLLWRNORM;
+ /* success cases including fallback */
+ mask |= POLLOUT | POLLWRNORM;
}
}
- release_sock(sk);
} else {
- sock_poll_wait(file, sk_sleep(sk), wait);
- if (sk->sk_state == SMC_LISTEN)
- /* woken up by sk_data_ready in smc_listen_work() */
- mask |= smc_accept_poll(sk);
+ if (sk->sk_state != SMC_CLOSED) {
+ release_sock(sk);
+ sock_poll_wait(file, sk_sleep(sk), wait);
+ lock_sock(sk);
+ }
if (sk->sk_err)
mask |= POLLERR;
- if (atomic_read(&smc->conn.sndbuf_space) ||
- (sk->sk_shutdown & SEND_SHUTDOWN)) {
- mask |= POLLOUT | POLLWRNORM;
- } else {
- sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk);
- set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
- }
- if (atomic_read(&smc->conn.bytes_to_rcv))
- mask |= POLLIN | POLLRDNORM;
if ((sk->sk_shutdown == SHUTDOWN_MASK) ||
(sk->sk_state == SMC_CLOSED))
mask |= POLLHUP;
- if (sk->sk_shutdown & RCV_SHUTDOWN)
- mask |= POLLIN | POLLRDNORM | POLLRDHUP;
- if (sk->sk_state == SMC_APPCLOSEWAIT1)
- mask |= POLLIN;
+ if (sk->sk_state == SMC_LISTEN) {
+ /* woken up by sk_data_ready in smc_listen_work() */
+ mask = smc_accept_poll(sk);
+ } else {
+ if (atomic_read(&smc->conn.sndbuf_space) ||
+ sk->sk_shutdown & SEND_SHUTDOWN) {
+ mask |= POLLOUT | POLLWRNORM;
+ } else {
+ sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+ set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+ }
+ if (atomic_read(&smc->conn.bytes_to_rcv))
+ mask |= POLLIN | POLLRDNORM;
+ if (sk->sk_shutdown & RCV_SHUTDOWN)
+ mask |= POLLIN | POLLRDNORM | POLLRDHUP;
+ if (sk->sk_state == SMC_APPCLOSEWAIT1)
+ mask |= POLLIN;
+ }
}
+ release_sock(sk);
+ sock_put(sk);
return mask;
}
diff --git a/net/smc/smc.h b/net/smc/smc.h
index 0bee9d16cf29..9518986c97b1 100644
--- a/net/smc/smc.h
+++ b/net/smc/smc.h
@@ -178,7 +178,6 @@ struct smc_sock { /* smc sock container */
struct work_struct smc_listen_work;/* prepare new accept socket */
struct list_head accept_q; /* sockets to be accepted */
spinlock_t accept_q_lock; /* protects accept_q */
- struct delayed_work sock_put_work; /* final socket freeing */
bool use_fallback; /* fallback to tcp */
u8 wait_close_tx_prepared : 1;
/* shutdown wr or close
@@ -253,12 +252,12 @@ static inline int smc_uncompress_bufsize(u8 compressed)
static inline bool using_ipsec(struct smc_sock *smc)
{
return (smc->clcsock->sk->sk_policy[0] ||
- smc->clcsock->sk->sk_policy[1]) ? 1 : 0;
+ smc->clcsock->sk->sk_policy[1]) ? true : false;
}
#else
static inline bool using_ipsec(struct smc_sock *smc)
{
- return 0;
+ return false;
}
#endif
diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c
index 6e8f5fbe0f09..3cd086e5bd28 100644
--- a/net/smc/smc_cdc.c
+++ b/net/smc/smc_cdc.c
@@ -212,6 +212,14 @@ static void smc_cdc_msg_recv_action(struct smc_sock *smc,
smc->sk.sk_data_ready(&smc->sk);
}
+ /* piggy backed tx info */
+ /* trigger sndbuf consumer: RDMA write into peer RMBE and CDC */
+ if (diff_cons && smc_tx_prepared_sends(conn)) {
+ smc_tx_sndbuf_nonempty(conn);
+ /* trigger socket release if connection closed */
+ smc_close_wake_tx_prepared(smc);
+ }
+
if (conn->local_rx_ctrl.conn_state_flags.peer_conn_abort) {
smc->sk.sk_err = ECONNRESET;
conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
@@ -221,15 +229,9 @@ static void smc_cdc_msg_recv_action(struct smc_sock *smc,
if (smc->clcsock && smc->clcsock->sk)
smc->clcsock->sk->sk_shutdown |= RCV_SHUTDOWN;
sock_set_flag(&smc->sk, SOCK_DONE);
- schedule_work(&conn->close_work);
- }
-
- /* piggy backed tx info */
- /* trigger sndbuf consumer: RDMA write into peer RMBE and CDC */
- if (diff_cons && smc_tx_prepared_sends(conn)) {
- smc_tx_sndbuf_nonempty(conn);
- /* trigger socket release if connection closed */
- smc_close_wake_tx_prepared(smc);
+ sock_hold(&smc->sk); /* sock_put in close_work */
+ if (!schedule_work(&conn->close_work))
+ sock_put(&smc->sk);
}
}
diff --git a/net/smc/smc_close.c b/net/smc/smc_close.c
index babe05d385e7..e339c0186dcf 100644
--- a/net/smc/smc_close.c
+++ b/net/smc/smc_close.c
@@ -19,6 +19,8 @@
#include "smc_cdc.h"
#include "smc_close.h"
+#define SMC_CLOSE_WAIT_LISTEN_CLCSOCK_TIME (5 * HZ)
+
static void smc_close_cleanup_listen(struct sock *parent)
{
struct sock *sk;
@@ -28,6 +30,27 @@ static void smc_close_cleanup_listen(struct sock *parent)
smc_close_non_accepted(sk);
}
+static void smc_close_wait_listen_clcsock(struct smc_sock *smc)
+{
+ DEFINE_WAIT_FUNC(wait, woken_wake_function);
+ struct sock *sk = &smc->sk;
+ signed long timeout;
+
+ timeout = SMC_CLOSE_WAIT_LISTEN_CLCSOCK_TIME;
+ add_wait_queue(sk_sleep(sk), &wait);
+ do {
+ release_sock(sk);
+ if (smc->clcsock)
+ timeout = wait_woken(&wait, TASK_UNINTERRUPTIBLE,
+ timeout);
+ sched_annotate_sleep();
+ lock_sock(sk);
+ if (!smc->clcsock)
+ break;
+ } while (timeout);
+ remove_wait_queue(sk_sleep(sk), &wait);
+}
+
/* wait for sndbuf data being transmitted */
static void smc_close_stream_wait(struct smc_sock *smc, long timeout)
{
@@ -110,10 +133,10 @@ static void smc_close_active_abort(struct smc_sock *smc)
release_sock(sk);
cancel_delayed_work_sync(&smc->conn.tx_work);
lock_sock(sk);
+ sock_put(sk); /* passive closing */
break;
case SMC_APPCLOSEWAIT1:
case SMC_APPCLOSEWAIT2:
- sock_release(smc->clcsock);
if (!smc_cdc_rxed_any_close(&smc->conn))
sk->sk_state = SMC_PEERABORTWAIT;
else
@@ -125,19 +148,20 @@ static void smc_close_active_abort(struct smc_sock *smc)
case SMC_PEERCLOSEWAIT1:
case SMC_PEERCLOSEWAIT2:
if (!txflags->peer_conn_closed) {
+ /* just SHUTDOWN_SEND done */
sk->sk_state = SMC_PEERABORTWAIT;
- sock_release(smc->clcsock);
} else {
sk->sk_state = SMC_CLOSED;
}
+ sock_put(sk); /* passive closing */
break;
case SMC_PROCESSABORT:
case SMC_APPFINCLOSEWAIT:
- if (!txflags->peer_conn_closed)
- sock_release(smc->clcsock);
sk->sk_state = SMC_CLOSED;
break;
case SMC_PEERFINCLOSEWAIT:
+ sock_put(sk); /* passive closing */
+ break;
case SMC_PEERABORTWAIT:
case SMC_CLOSED:
break;
@@ -172,8 +196,6 @@ again:
switch (sk->sk_state) {
case SMC_INIT:
sk->sk_state = SMC_CLOSED;
- if (smc->smc_listen_work.func)
- cancel_work_sync(&smc->smc_listen_work);
break;
case SMC_LISTEN:
sk->sk_state = SMC_CLOSED;
@@ -182,11 +204,9 @@ again:
rc = kernel_sock_shutdown(smc->clcsock, SHUT_RDWR);
/* wake up kernel_accept of smc_tcp_listen_worker */
smc->clcsock->sk->sk_data_ready(smc->clcsock->sk);
+ smc_close_wait_listen_clcsock(smc);
}
- release_sock(sk);
smc_close_cleanup_listen(sk);
- cancel_work_sync(&smc->smc_listen_work);
- lock_sock(sk);
break;
case SMC_ACTIVE:
smc_close_stream_wait(smc, timeout);
@@ -229,12 +249,14 @@ again:
rc = smc_close_final(conn);
if (rc)
break;
- if (smc_cdc_rxed_any_close(conn))
+ if (smc_cdc_rxed_any_close(conn)) {
/* peer has closed the socket already */
sk->sk_state = SMC_CLOSED;
- else
+ sock_put(sk); /* postponed passive closing */
+ } else {
/* peer has just issued a shutdown write */
sk->sk_state = SMC_PEERFINCLOSEWAIT;
+ }
break;
case SMC_PEERCLOSEWAIT1:
case SMC_PEERCLOSEWAIT2:
@@ -272,27 +294,33 @@ static void smc_close_passive_abort_received(struct smc_sock *smc)
struct sock *sk = &smc->sk;
switch (sk->sk_state) {
+ case SMC_INIT:
case SMC_ACTIVE:
- case SMC_APPFINCLOSEWAIT:
case SMC_APPCLOSEWAIT1:
- case SMC_APPCLOSEWAIT2:
+ sk->sk_state = SMC_PROCESSABORT;
+ sock_put(sk); /* passive closing */
+ break;
+ case SMC_APPFINCLOSEWAIT:
sk->sk_state = SMC_PROCESSABORT;
break;
case SMC_PEERCLOSEWAIT1:
case SMC_PEERCLOSEWAIT2:
if (txflags->peer_done_writing &&
- !smc_close_sent_any_close(&smc->conn)) {
+ !smc_close_sent_any_close(&smc->conn))
/* just shutdown, but not yet closed locally */
sk->sk_state = SMC_PROCESSABORT;
- } else {
+ else
sk->sk_state = SMC_CLOSED;
- }
+ sock_put(sk); /* passive closing */
break;
+ case SMC_APPCLOSEWAIT2:
case SMC_PEERFINCLOSEWAIT:
+ sk->sk_state = SMC_CLOSED;
+ sock_put(sk); /* passive closing */
+ break;
case SMC_PEERABORTWAIT:
sk->sk_state = SMC_CLOSED;
break;
- case SMC_INIT:
case SMC_PROCESSABORT:
/* nothing to do, add tracing in future patch */
break;
@@ -336,13 +364,18 @@ static void smc_close_passive_work(struct work_struct *work)
case SMC_INIT:
if (atomic_read(&conn->bytes_to_rcv) ||
(rxflags->peer_done_writing &&
- !smc_cdc_rxed_any_close(conn)))
+ !smc_cdc_rxed_any_close(conn))) {
sk->sk_state = SMC_APPCLOSEWAIT1;
- else
+ } else {
sk->sk_state = SMC_CLOSED;
+ sock_put(sk); /* passive closing */
+ }
break;
case SMC_ACTIVE:
sk->sk_state = SMC_APPCLOSEWAIT1;
+ /* postpone sock_put() for passive closing to cover
+ * received SEND_SHUTDOWN as well
+ */
break;
case SMC_PEERCLOSEWAIT1:
if (rxflags->peer_done_writing)
@@ -360,13 +393,20 @@ static void smc_close_passive_work(struct work_struct *work)
/* just shutdown, but not yet closed locally */
sk->sk_state = SMC_APPFINCLOSEWAIT;
}
+ sock_put(sk); /* passive closing */
break;
case SMC_PEERFINCLOSEWAIT:
- if (smc_cdc_rxed_any_close(conn))
+ if (smc_cdc_rxed_any_close(conn)) {
sk->sk_state = SMC_CLOSED;
+ sock_put(sk); /* passive closing */
+ }
break;
case SMC_APPCLOSEWAIT1:
case SMC_APPCLOSEWAIT2:
+ /* postpone sock_put() for passive closing to cover
+ * received SEND_SHUTDOWN as well
+ */
+ break;
case SMC_APPFINCLOSEWAIT:
case SMC_PEERABORTWAIT:
case SMC_PROCESSABORT:
@@ -382,23 +422,11 @@ wakeup:
if (old_state != sk->sk_state) {
sk->sk_state_change(sk);
if ((sk->sk_state == SMC_CLOSED) &&
- (sock_flag(sk, SOCK_DEAD) || !sk->sk_socket)) {
+ (sock_flag(sk, SOCK_DEAD) || !sk->sk_socket))
smc_conn_free(conn);
- schedule_delayed_work(&smc->sock_put_work,
- SMC_CLOSE_SOCK_PUT_DELAY);
- }
}
release_sock(sk);
-}
-
-void smc_close_sock_put_work(struct work_struct *work)
-{
- struct smc_sock *smc = container_of(to_delayed_work(work),
- struct smc_sock,
- sock_put_work);
-
- smc->sk.sk_prot->unhash(&smc->sk);
- sock_put(&smc->sk);
+ sock_put(sk); /* sock_hold done by schedulers of close_work */
}
int smc_close_shutdown_write(struct smc_sock *smc)
diff --git a/net/smc/smc_close.h b/net/smc/smc_close.h
index 8c498885d758..19eb6a211c23 100644
--- a/net/smc/smc_close.h
+++ b/net/smc/smc_close.h
@@ -21,7 +21,6 @@
void smc_close_wake_tx_prepared(struct smc_sock *smc);
int smc_close_active(struct smc_sock *smc);
-void smc_close_sock_put_work(struct work_struct *work);
int smc_close_shutdown_write(struct smc_sock *smc);
void smc_close_init(struct smc_sock *smc);
diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c
index ed5b46d1fe41..2424c7100aaf 100644
--- a/net/smc/smc_core.c
+++ b/net/smc/smc_core.c
@@ -328,13 +328,13 @@ void smc_lgr_terminate(struct smc_link_group *lgr)
while (node) {
conn = rb_entry(node, struct smc_connection, alert_node);
smc = container_of(conn, struct smc_sock, conn);
- sock_hold(&smc->sk);
+ sock_hold(&smc->sk); /* sock_put in close work */
conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
__smc_lgr_unregister_conn(conn);
write_unlock_bh(&lgr->conns_lock);
- schedule_work(&conn->close_work);
+ if (!schedule_work(&conn->close_work))
+ sock_put(&smc->sk);
write_lock_bh(&lgr->conns_lock);
- sock_put(&smc->sk);
node = rb_first(&lgr->conns_all);
}
write_unlock_bh(&lgr->conns_lock);
diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c
index 90f1a7f9085c..2a8957bd6d38 100644
--- a/net/smc/smc_ib.c
+++ b/net/smc/smc_ib.c
@@ -141,6 +141,17 @@ out:
return rc;
}
+static void smc_ib_port_terminate(struct smc_ib_device *smcibdev, u8 ibport)
+{
+ struct smc_link_group *lgr, *l;
+
+ list_for_each_entry_safe(lgr, l, &smc_lgr_list.list, list) {
+ if (lgr->lnk[SMC_SINGLE_LINK].smcibdev == smcibdev &&
+ lgr->lnk[SMC_SINGLE_LINK].ibport == ibport)
+ smc_lgr_terminate(lgr);
+ }
+}
+
/* process context wrapper for might_sleep smc_ib_remember_port_attr */
static void smc_ib_port_event_work(struct work_struct *work)
{
@@ -151,6 +162,8 @@ static void smc_ib_port_event_work(struct work_struct *work)
for_each_set_bit(port_idx, &smcibdev->port_event_mask, SMC_MAX_PORTS) {
smc_ib_remember_port_attr(smcibdev, port_idx + 1);
clear_bit(port_idx, &smcibdev->port_event_mask);
+ if (!smc_ib_port_active(smcibdev, port_idx + 1))
+ smc_ib_port_terminate(smcibdev, port_idx + 1);
}
}
@@ -165,15 +178,7 @@ static void smc_ib_global_event_handler(struct ib_event_handler *handler,
switch (ibevent->event) {
case IB_EVENT_PORT_ERR:
- port_idx = ibevent->element.port_num - 1;
- set_bit(port_idx, &smcibdev->port_event_mask);
- schedule_work(&smcibdev->port_event_work);
- /* fall through */
case IB_EVENT_DEVICE_FATAL:
- /* tbd in follow-on patch:
- * abnormal close of corresponding connections
- */
- break;
case IB_EVENT_PORT_ACTIVE:
port_idx = ibevent->element.port_num - 1;
set_bit(port_idx, &smcibdev->port_event_mask);
@@ -186,7 +191,8 @@ static void smc_ib_global_event_handler(struct ib_event_handler *handler,
void smc_ib_dealloc_protection_domain(struct smc_link *lnk)
{
- ib_dealloc_pd(lnk->roce_pd);
+ if (lnk->roce_pd)
+ ib_dealloc_pd(lnk->roce_pd);
lnk->roce_pd = NULL;
}
@@ -203,14 +209,18 @@ int smc_ib_create_protection_domain(struct smc_link *lnk)
static void smc_ib_qp_event_handler(struct ib_event *ibevent, void *priv)
{
+ struct smc_ib_device *smcibdev =
+ (struct smc_ib_device *)ibevent->device;
+ u8 port_idx;
+
switch (ibevent->event) {
case IB_EVENT_DEVICE_FATAL:
case IB_EVENT_GID_CHANGE:
case IB_EVENT_PORT_ERR:
case IB_EVENT_QP_ACCESS_ERR:
- /* tbd in follow-on patch:
- * abnormal close of corresponding connections
- */
+ port_idx = ibevent->element.port_num - 1;
+ set_bit(port_idx, &smcibdev->port_event_mask);
+ schedule_work(&smcibdev->port_event_work);
break;
default:
break;
@@ -219,7 +229,8 @@ static void smc_ib_qp_event_handler(struct ib_event *ibevent, void *priv)
void smc_ib_destroy_queue_pair(struct smc_link *lnk)
{
- ib_destroy_qp(lnk->roce_qp);
+ if (lnk->roce_qp)
+ ib_destroy_qp(lnk->roce_qp);
lnk->roce_qp = NULL;
}
@@ -462,6 +473,7 @@ static void smc_ib_cleanup_per_ibdev(struct smc_ib_device *smcibdev)
{
if (!smcibdev->initialized)
return;
+ smcibdev->initialized = 0;
smc_wr_remove_dev(smcibdev);
ib_unregister_event_handler(&smcibdev->event_handler);
ib_destroy_cq(smcibdev->roce_cq_recv);
diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c
index 92b4648e75ca..8e70291e586a 100644
--- a/net/xfrm/xfrm_device.c
+++ b/net/xfrm/xfrm_device.c
@@ -147,8 +147,8 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x,
if (!x->type_offload)
return -EINVAL;
- /* We don't yet support UDP encapsulation, TFC padding and ESN. */
- if (x->encap || x->tfcpad || (x->props.flags & XFRM_STATE_ESN))
+ /* We don't yet support UDP encapsulation and TFC padding. */
+ if (x->encap || x->tfcpad)
return -EINVAL;
dev = dev_get_by_index(net, xuo->ifindex);
@@ -178,6 +178,13 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x,
return 0;
}
+ if (x->props.flags & XFRM_STATE_ESN &&
+ !dev->xfrmdev_ops->xdo_dev_state_advance_esn) {
+ xso->dev = NULL;
+ dev_put(dev);
+ return -EINVAL;
+ }
+
xso->dev = dev;
xso->num_exthdrs = 1;
xso->flags = xuo->flags;
diff --git a/net/xfrm/xfrm_replay.c b/net/xfrm/xfrm_replay.c
index 02501817227b..1d38c6acf8af 100644
--- a/net/xfrm/xfrm_replay.c
+++ b/net/xfrm/xfrm_replay.c
@@ -551,6 +551,8 @@ static void xfrm_replay_advance_esn(struct xfrm_state *x, __be32 net_seq)
bitnr = replay_esn->replay_window - (diff - pos);
}
+ xfrm_dev_state_advance_esn(x);
+
nr = bitnr >> 5;
bitnr = bitnr & 0x1F;
replay_esn->bmp[nr] |= (1U << bitnr);
Powered by cgit, Gruvboxed by Blaster4385.