LU-13510 lnd: Allow independent socklnd timeout

Allow the socklnd timeout to be set independent of
lnet_transaction_timeout and retry_count.

Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: Iaa76e77990c8c5ce79193ae8d1f7b3a7db6b433f
Reviewed-on: https://review.whamcloud.com/38460
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Amir Shehata <ashehata@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

diff --git a/lnet/klnds/socklnd/socklnd.c b/lnet/klnds/socklnd/socklnd.c
index b50b8b3..d750eaa 100644 (file)
--- a/lnet/klnds/socklnd/socklnd.c
+++ b/lnet/klnds/socklnd/socklnd.c
@@ -1311,7 +1311,7 @@ ksocknal_create_conn(struct lnet_ni *ni, struct ksock_route *route,
        /* Set the deadline for the outgoing HELLO to drain */
        conn->ksnc_tx_bufnob = sock->sk->sk_wmem_queued;
        conn->ksnc_tx_deadline = ktime_get_seconds() +
-                                lnet_get_lnd_timeout();
+                                ksocknal_timeout();
        smp_mb();   /* order with adding to peer_ni's conn list */
 
        list_add(&conn->ksnc_list, &peer_ni->ksnp_conns);
@@ -1688,7 +1688,7 @@ ksocknal_destroy_conn(struct ksock_conn *conn)
         switch (conn->ksnc_rx_state) {
         case SOCKNAL_RX_LNET_PAYLOAD:
                 last_rcv = conn->ksnc_rx_deadline -
-                          lnet_get_lnd_timeout();
+                          ksocknal_timeout();
                CERROR("Completing partial receive from %s[%d], "
                       "ip %pI4h:%d, with error, wanted: %d, left: %d, "
                       "last alive is %lld secs ago\n",

diff --git a/lnet/klnds/socklnd/socklnd.h b/lnet/klnds/socklnd/socklnd.h
index 59127c4..950c756 100644 (file)
--- a/lnet/klnds/socklnd/socklnd.h
+++ b/lnet/klnds/socklnd/socklnd.h
@@ -567,6 +567,13 @@ ksocknal_peer_decref(struct ksock_peer_ni *peer_ni)
                ksocknal_destroy_peer(peer_ni);
 }
 
+static inline int ksocknal_timeout(void)
+{
+       return *ksocknal_tunables.ksnd_timeout ?
+               *ksocknal_tunables.ksnd_timeout :
+               lnet_get_lnd_timeout();
+}
+
 int ksocknal_startup(struct lnet_ni *ni);
 void ksocknal_shutdown(struct lnet_ni *ni);
 int ksocknal_ctl(struct lnet_ni *ni, unsigned int cmd, void *arg);

diff --git a/lnet/klnds/socklnd/socklnd_cb.c b/lnet/klnds/socklnd/socklnd_cb.c
index 9641431..aa9f747 100644 (file)
--- a/lnet/klnds/socklnd/socklnd_cb.c
+++ b/lnet/klnds/socklnd/socklnd_cb.c
@@ -222,7 +222,7 @@ ksocknal_transmit(struct ksock_conn *conn, struct ksock_tx *tx,
                        /* allocated send buffer bytes < computed; infer
                         * something got ACKed */
                        conn->ksnc_tx_deadline = ktime_get_seconds() +
-                                                lnet_get_lnd_timeout();
+                                                ksocknal_timeout();
                        conn->ksnc_peer->ksnp_last_alive = ktime_get_seconds();
                        conn->ksnc_tx_bufnob = bufnob;
                        smp_mb();
@@ -271,7 +271,7 @@ ksocknal_recv_iov(struct ksock_conn *conn, struct kvec *scratchiov)
 
        conn->ksnc_peer->ksnp_last_alive = ktime_get_seconds();
        conn->ksnc_rx_deadline = ktime_get_seconds() +
-                                lnet_get_lnd_timeout();
+                                ksocknal_timeout();
        smp_mb();                       /* order with setting rx_started */
        conn->ksnc_rx_started = 1;
 
@@ -316,7 +316,7 @@ ksocknal_recv_kiov(struct ksock_conn *conn, struct page **rx_scratch_pgs,
 
        conn->ksnc_peer->ksnp_last_alive = ktime_get_seconds();
        conn->ksnc_rx_deadline = ktime_get_seconds() +
-                                lnet_get_lnd_timeout();
+                                ksocknal_timeout();
        smp_mb();                       /* order with setting rx_started */
        conn->ksnc_rx_started = 1;
 
@@ -490,7 +490,7 @@ ksocknal_check_zc_req(struct ksock_tx *tx)
 
         /* ZC_REQ is going to be pinned to the peer_ni */
        tx->tx_deadline = ktime_get_seconds() +
-                         lnet_get_lnd_timeout();
+                         ksocknal_timeout();
 
         LASSERT (tx->tx_msg.ksm_zc_cookies[0] == 0);
 
@@ -784,7 +784,7 @@ ksocknal_queue_tx_locked(struct ksock_tx *tx, struct ksock_conn *conn)
        if (list_empty(&conn->ksnc_tx_queue) && bufnob == 0) {
                /* First packet starts the timeout */
                conn->ksnc_tx_deadline = ktime_get_seconds() +
-                                        lnet_get_lnd_timeout();
+                                        ksocknal_timeout();
                if (conn->ksnc_tx_bufnob > 0) /* something got ACKed */
                        conn->ksnc_peer->ksnp_last_alive = ktime_get_seconds();
                conn->ksnc_tx_bufnob = 0;
@@ -961,7 +961,7 @@ ksocknal_launch_packet(struct lnet_ni *ni, struct ksock_tx *tx,
             ksocknal_find_connecting_route_locked (peer_ni) != NULL) {
                 /* the message is going to be pinned to the peer_ni */
                tx->tx_deadline = ktime_get_seconds() +
-                                 lnet_get_lnd_timeout();
+                                 ksocknal_timeout();
 
                 /* Queue the message until a connection is established */
                list_add_tail(&tx->tx_list, &peer_ni->ksnp_tx_queue);
@@ -1782,7 +1782,7 @@ ksocknal_recv_hello(struct lnet_ni *ni, struct ksock_conn *conn,
        /* socket type set on active connections - not set on passive */
        LASSERT(!active == !(conn->ksnc_type != SOCKLND_CONN_NONE));
 
-       timeout = active ? lnet_get_lnd_timeout() :
+       timeout = active ? ksocknal_timeout() :
                            lnet_acceptor_timeout();
 
        rc = lnet_sock_read(sock, &hello->kshm_magic,
@@ -1917,7 +1917,7 @@ ksocknal_connect(struct ksock_route *route)
         int               retry_later = 0;
         int               rc = 0;
 
-       deadline = ktime_get_seconds() + lnet_get_lnd_timeout();
+       deadline = ktime_get_seconds() + ksocknal_timeout();
 
        write_lock_bh(&ksocknal_data.ksnd_global_lock);
 
@@ -2662,7 +2662,7 @@ int ksocknal_reaper(void *arg)
                         * times the timeout interval.
                         */
 
-                       lnd_timeout = lnet_get_lnd_timeout();
+                       lnd_timeout = ksocknal_timeout();
                        if (lnd_timeout > n * p)
                                chunk = (chunk * n * p) / lnd_timeout;
                        if (chunk == 0)

diff --git a/lnet/klnds/socklnd/socklnd_modparams.c b/lnet/klnds/socklnd/socklnd_modparams.c
index 737cb88..fb8b95c 100644 (file)
--- a/lnet/klnds/socklnd/socklnd_modparams.c
+++ b/lnet/klnds/socklnd/socklnd_modparams.c
@@ -26,7 +26,7 @@
 #include <asm/hypervisor.h>
 #endif
 
-static int sock_timeout = 50;
+static int sock_timeout;
 module_param(sock_timeout, int, 0644);
 MODULE_PARM_DESC(sock_timeout, "dead socket timeout (seconds)");