/*
* Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
*
- * Copyright (c) 2011, 2016, Intel Corporation.
+ * Copyright (c) 2011, 2017, Intel Corporation.
*
- * This file is part of Lustre, https://wiki.hpdd.intel.com/
+ * This file is part of Lustre, https://wiki.whamcloud.com/
*
* Portals is free software; you can redistribute it and/or
* modify it under the terms of version 2 of the GNU General Public
return net->net_tunables.lct_peer_tx_credits;
}
-/* forward ref's */
-static int lnet_router_checker(void *);
-
static int check_routers_before_use;
module_param(check_routers_before_use, int, 0444);
MODULE_PARM_DESC(check_routers_before_use, "Assume routers are down and ping them before use");
void
lnet_notify_locked(struct lnet_peer_ni *lp, int notifylnd, int alive,
- cfs_time_t when)
+ time64_t when)
{
- if (cfs_time_before(when, lp->lpni_timestamp)) { /* out of date information */
+ if (lp->lpni_timestamp > when) { /* out of date information */
CDEBUG(D_NET, "Out of date\n");
return;
}
*/
spin_lock(&lp->lpni_lock);
- lp->lpni_timestamp = when; /* update timestamp */
+ lp->lpni_timestamp = when; /* update timestamp */
lp->lpni_ping_deadline = 0; /* disable ping timeout */
if (lp->lpni_alive_count != 0 && /* got old news */
* This function will always be called with lp->lpni_cpt lock held.
*/
static void
-lnet_ni_notify_locked(lnet_ni_t *ni, struct lnet_peer_ni *lp)
+lnet_ni_notify_locked(struct lnet_ni *ni, struct lnet_peer_ni *lp)
{
int alive;
int notifylnd;
/* a simple insertion sort */
list_for_each_prev(pos, &the_lnet.ln_routers) {
- struct lnet_peer_ni *rtr =
- list_entry(pos, struct lnet_peer_ni,
- lpni_rtr_list);
+ struct lnet_peer_ni *rtr;
+ rtr = list_entry(pos, struct lnet_peer_ni,
+ lpni_rtr_list);
if (rtr->lpni_nid < lp->lpni_nid)
break;
}
}
}
-lnet_remotenet_t *
+struct lnet_remotenet *
lnet_find_rnet_locked(__u32 net)
{
- lnet_remotenet_t *rnet;
+ struct lnet_remotenet *rnet;
struct list_head *tmp;
struct list_head *rn_list;
- LASSERT(!the_lnet.ln_shutdown);
+ LASSERT(the_lnet.ln_state == LNET_STATE_RUNNING);
rn_list = lnet_net2rnethash(net);
list_for_each(tmp, rn_list) {
- rnet = list_entry(tmp, lnet_remotenet_t, lrn_list);
+ rnet = list_entry(tmp, struct lnet_remotenet, lrn_list);
if (rnet->lrn_net == net)
return rnet;
__u32 lnd_type;
__u32 seed[2];
struct timespec64 ts;
- lnet_ni_t *ni = NULL;
+ struct lnet_ni *ni = NULL;
if (seeded)
return;
/* NB expects LNET_LOCK held */
static void
-lnet_add_route_to_rnet(lnet_remotenet_t *rnet, lnet_route_t *route)
+lnet_add_route_to_rnet(struct lnet_remotenet *rnet, struct lnet_route *route)
{
unsigned int len = 0;
unsigned int offset = 0;
unsigned int priority)
{
struct list_head *e;
- lnet_remotenet_t *rnet;
- lnet_remotenet_t *rnet2;
- lnet_route_t *route;
- lnet_ni_t *ni;
+ struct lnet_remotenet *rnet;
+ struct lnet_remotenet *rnet2;
+ struct lnet_route *route;
+ struct lnet_ni *ni;
struct lnet_peer_ni *lpni;
int add_route;
int rc;
return rc;
}
route->lr_gateway = lpni;
- LASSERT(!the_lnet.ln_shutdown);
+ LASSERT(the_lnet.ln_state == LNET_STATE_RUNNING);
rnet2 = lnet_find_rnet_locked(net);
if (rnet2 == NULL) {
/* Search for a duplicate route (it's a NOOP if it is) */
add_route = 1;
list_for_each(e, &rnet2->lrn_routes) {
- lnet_route_t *route2 = list_entry(e, lnet_route_t, lr_list);
+ struct lnet_route *route2;
+ route2 = list_entry(e, struct lnet_route, lr_list);
if (route2->lr_gateway == route->lr_gateway) {
add_route = 0;
break;
if (rnet != rnet2)
LIBCFS_FREE(rnet, sizeof(*rnet));
- /* indicate to startup the router checker if configured */
- wake_up(&the_lnet.ln_rc_waitq);
+ /* kick start the monitor thread to handle the added route */
+ wake_up(&the_lnet.ln_mt_waitq);
return rc;
}
int
lnet_check_routes(void)
{
- lnet_remotenet_t *rnet;
- lnet_route_t *route;
- lnet_route_t *route2;
+ struct lnet_remotenet *rnet;
+ struct lnet_route *route;
+ struct lnet_route *route2;
struct list_head *e1;
struct list_head *e2;
int cpt;
for (i = 0; i < LNET_REMOTE_NETS_HASH_SIZE; i++) {
rn_list = &the_lnet.ln_remote_nets_hash[i];
list_for_each(e1, rn_list) {
- rnet = list_entry(e1, lnet_remotenet_t, lrn_list);
+ rnet = list_entry(e1, struct lnet_remotenet, lrn_list);
route2 = NULL;
list_for_each(e2, &rnet->lrn_routes) {
lnet_nid_t nid2;
int net;
- route = list_entry(e2, lnet_route_t,
+ route = list_entry(e2, struct lnet_route,
lr_list);
if (route2 == NULL) {
lnet_del_route(__u32 net, lnet_nid_t gw_nid)
{
struct lnet_peer_ni *gateway;
- lnet_remotenet_t *rnet;
- lnet_route_t *route;
+ struct lnet_remotenet *rnet;
+ struct lnet_route *route;
struct list_head *e1;
struct list_head *e2;
int rc = -ENOENT;
again:
list_for_each(e1, rn_list) {
- rnet = list_entry(e1, lnet_remotenet_t, lrn_list);
+ rnet = list_entry(e1, struct lnet_remotenet, lrn_list);
if (!(net == LNET_NIDNET(LNET_NID_ANY) ||
net == rnet->lrn_net))
continue;
list_for_each(e2, &rnet->lrn_routes) {
- route = list_entry(e2, lnet_route_t, lr_list);
+ route = list_entry(e2, struct lnet_route, lr_list);
gateway = route->lr_gateway;
if (!(gw_nid == LNET_NID_ANY ||
return rc;
for (i = 0; i < LNET_NRBPOOLS; i++) {
- lnet_rtrbufpool_t *rbp;
+ struct lnet_rtrbufpool *rbp;
lnet_net_lock(LNET_LOCK_EX);
cfs_percpt_for_each(rbp, j, the_lnet.ln_rtrpools) {
{
struct list_head *e1;
struct list_head *e2;
- lnet_remotenet_t *rnet;
- lnet_route_t *route;
+ struct lnet_remotenet *rnet;
+ struct lnet_route *route;
int cpt;
int i;
struct list_head *rn_list;
for (i = 0; i < LNET_REMOTE_NETS_HASH_SIZE; i++) {
rn_list = &the_lnet.ln_remote_nets_hash[i];
list_for_each(e1, rn_list) {
- rnet = list_entry(e1, lnet_remotenet_t, lrn_list);
+ rnet = list_entry(e1, struct lnet_remotenet, lrn_list);
list_for_each(e2, &rnet->lrn_routes) {
- route = list_entry(e2, lnet_route_t,
+ route = list_entry(e2, struct lnet_route,
lr_list);
if (idx-- == 0) {
}
void
-lnet_swap_pinginfo(struct lnet_ping_info *info)
+lnet_swap_pinginfo(struct lnet_ping_buffer *pbuf)
{
- int i;
struct lnet_ni_status *stat;
+ int nnis;
+ int i;
- __swab32s(&info->pi_magic);
- __swab32s(&info->pi_features);
- __swab32s(&info->pi_pid);
- __swab32s(&info->pi_nnis);
- for (i = 0; i < info->pi_nnis && i < LNET_MAX_RTR_NIS; i++) {
- stat = &info->pi_ni[i];
+ __swab32s(&pbuf->pb_info.pi_magic);
+ __swab32s(&pbuf->pb_info.pi_features);
+ __swab32s(&pbuf->pb_info.pi_pid);
+ __swab32s(&pbuf->pb_info.pi_nnis);
+ nnis = pbuf->pb_info.pi_nnis;
+ if (nnis > pbuf->pb_nnis)
+ nnis = pbuf->pb_nnis;
+ for (i = 0; i < nnis; i++) {
+ stat = &pbuf->pb_info.pi_ni[i];
__swab64s(&stat->ns_nid);
__swab32s(&stat->ns_status);
}
* networks on that router.
*/
static void
-lnet_parse_rc_info(lnet_rc_data_t *rcd)
+lnet_parse_rc_info(struct lnet_rc_data *rcd)
{
- struct lnet_ping_info *info = rcd->rcd_pinginfo;
+ struct lnet_ping_buffer *pbuf = rcd->rcd_pingbuffer;
struct lnet_peer_ni *gw = rcd->rcd_gateway;
- lnet_route_t *rte;
+ struct lnet_route *rte;
+ int nnis;
- if (!gw->lpni_alive)
+ if (!gw->lpni_alive || !pbuf)
return;
/*
*/
spin_lock(&gw->lpni_lock);
- if (info->pi_magic == __swab32(LNET_PROTO_PING_MAGIC))
- lnet_swap_pinginfo(info);
+ if (pbuf->pb_info.pi_magic == __swab32(LNET_PROTO_PING_MAGIC))
+ lnet_swap_pinginfo(pbuf);
/* NB always racing with network! */
- if (info->pi_magic != LNET_PROTO_PING_MAGIC) {
+ if (pbuf->pb_info.pi_magic != LNET_PROTO_PING_MAGIC) {
CDEBUG(D_NET, "%s: Unexpected magic %08x\n",
- libcfs_nid2str(gw->lpni_nid), info->pi_magic);
+ libcfs_nid2str(gw->lpni_nid), pbuf->pb_info.pi_magic);
gw->lpni_ping_feats = LNET_PING_FEAT_INVAL;
- spin_unlock(&gw->lpni_lock);
- return;
+ goto out;
}
- gw->lpni_ping_feats = info->pi_features;
- if ((gw->lpni_ping_feats & LNET_PING_FEAT_MASK) == 0) {
- CDEBUG(D_NET, "%s: Unexpected features 0x%x\n",
- libcfs_nid2str(gw->lpni_nid), gw->lpni_ping_feats);
- spin_unlock(&gw->lpni_lock);
- return; /* nothing I can understand */
- }
+ gw->lpni_ping_feats = pbuf->pb_info.pi_features;
- if ((gw->lpni_ping_feats & LNET_PING_FEAT_NI_STATUS) == 0) {
- spin_unlock(&gw->lpni_lock);
- return; /* can't carry NI status info */
+ /* Without NI status info there's nothing more to do. */
+ if ((gw->lpni_ping_feats & LNET_PING_FEAT_NI_STATUS) == 0)
+ goto out;
+
+ /* Determine the number of NIs for which there is data. */
+ nnis = pbuf->pb_info.pi_nnis;
+ if (pbuf->pb_nnis < nnis) {
+ if (rcd->rcd_nnis < nnis)
+ rcd->rcd_nnis = nnis;
+ nnis = pbuf->pb_nnis;
}
list_for_each_entry(rte, &gw->lpni_routes, lr_gwlist) {
int up = 0;
int i;
+ /* If routing disabled then the route is down. */
if ((gw->lpni_ping_feats & LNET_PING_FEAT_RTE_DISABLED) != 0) {
rte->lr_downis = 1;
continue;
}
- for (i = 0; i < info->pi_nnis && i < LNET_MAX_RTR_NIS; i++) {
- struct lnet_ni_status *stat = &info->pi_ni[i];
+ for (i = 0; i < nnis; i++) {
+ struct lnet_ni_status *stat = &pbuf->pb_info.pi_ni[i];
lnet_nid_t nid = stat->ns_nid;
if (nid == LNET_NID_ANY) {
CDEBUG(D_NET, "%s: unexpected LNET_NID_ANY\n",
libcfs_nid2str(gw->lpni_nid));
gw->lpni_ping_feats = LNET_PING_FEAT_INVAL;
- spin_unlock(&gw->lpni_lock);
- return;
+ goto out;
}
if (LNET_NETTYP(LNET_NIDNET(nid)) == LOLND)
CDEBUG(D_NET, "%s: Unexpected status 0x%x\n",
libcfs_nid2str(gw->lpni_nid), stat->ns_status);
gw->lpni_ping_feats = LNET_PING_FEAT_INVAL;
- spin_unlock(&gw->lpni_lock);
- return;
+ goto out;
}
if (up) { /* ignore downed NIs if NI for dest network is up */
rte->lr_downis = down;
}
-
+out:
spin_unlock(&gw->lpni_lock);
}
static void
-lnet_router_checker_event(lnet_event_t *event)
+lnet_router_checker_event(struct lnet_event *event)
{
- lnet_rc_data_t *rcd = event->md.user_ptr;
+ struct lnet_rc_data *rcd = event->md.user_ptr;
struct lnet_peer_ni *lp;
LASSERT(rcd != NULL);
if (event->unlinked) {
- LNetInvalidateHandle(&rcd->rcd_mdh);
+ LNetInvalidateMDHandle(&rcd->rcd_mdh);
return;
}
* we ping alive routers to try to detect router death before
* apps get burned). */
- lnet_notify_locked(lp, 1, (event->status == 0), cfs_time_current());
+ lnet_notify_locked(lp, 1, !event->status, ktime_get_seconds());
/* The router checker will wake up very shortly and do the
* actual notification.
* XXX If 'lp' stops being a router before then, it will still
struct list_head *entry;
int all_known;
- LASSERT(the_lnet.ln_rc_state == LNET_RC_STATE_RUNNING);
+ LASSERT(the_lnet.ln_mt_state == LNET_MT_STATE_RUNNING);
for (;;) {
int cpt = lnet_net_lock_current();
void
lnet_router_ni_update_locked(struct lnet_peer_ni *gw, __u32 net)
{
- lnet_route_t *rte;
+ struct lnet_route *rte;
if ((gw->lpni_ping_feats & LNET_PING_FEAT_NI_STATUS) != 0) {
list_for_each_entry(rte, &gw->lpni_routes, lr_gwlist) {
static void
lnet_update_ni_status_locked(void)
{
- lnet_ni_t *ni = NULL;
- time64_t now;
- int timeout;
+ struct lnet_ni *ni = NULL;
+ time64_t now;
+ time64_t timeout;
LASSERT(the_lnet.ln_routing);
LASSERT(ni->ni_status != NULL);
if (ni->ni_status->ns_status != LNET_NI_STATUS_DOWN) {
- CDEBUG(D_NET, "NI(%s:%d) status changed to down\n",
+ CDEBUG(D_NET, "NI(%s:%lld) status changed to down\n",
libcfs_nid2str(ni->ni_nid), timeout);
/* NB: so far, this is the only place to set
* NI status to "down" */
}
static void
-lnet_destroy_rc_data(lnet_rc_data_t *rcd)
+lnet_destroy_rc_data(struct lnet_rc_data *rcd)
{
LASSERT(list_empty(&rcd->rcd_list));
/* detached from network */
- LASSERT(LNetHandleIsInvalid(rcd->rcd_mdh));
+ LASSERT(LNetMDHandleIsInvalid(rcd->rcd_mdh));
if (rcd->rcd_gateway != NULL) {
int cpt = rcd->rcd_gateway->lpni_cpt;
lnet_net_unlock(cpt);
}
- if (rcd->rcd_pinginfo != NULL)
- LIBCFS_FREE(rcd->rcd_pinginfo, LNET_PINGINFO_SIZE);
+ if (rcd->rcd_pingbuffer != NULL)
+ lnet_ping_buffer_decref(rcd->rcd_pingbuffer);
LIBCFS_FREE(rcd, sizeof(*rcd));
}
-static lnet_rc_data_t *
-lnet_create_rc_data_locked(struct lnet_peer_ni *gateway)
+static struct lnet_rc_data *
+lnet_update_rc_data_locked(struct lnet_peer_ni *gateway)
{
- lnet_rc_data_t *rcd = NULL;
- struct lnet_ping_info *pi;
- int rc;
- int i;
+ struct lnet_handle_md mdh;
+ struct lnet_rc_data *rcd;
+ struct lnet_ping_buffer *pbuf = NULL;
+ int nnis = LNET_INTERFACES_MIN;
+ int rc;
+ int i;
+
+ rcd = gateway->lpni_rcd;
+ if (rcd) {
+ nnis = rcd->rcd_nnis;
+ mdh = rcd->rcd_mdh;
+ LNetInvalidateMDHandle(&rcd->rcd_mdh);
+ pbuf = rcd->rcd_pingbuffer;
+ rcd->rcd_pingbuffer = NULL;
+ } else {
+ LNetInvalidateMDHandle(&mdh);
+ }
lnet_net_unlock(gateway->lpni_cpt);
- LIBCFS_ALLOC(rcd, sizeof(*rcd));
- if (rcd == NULL)
- goto out;
+ if (rcd) {
+ LNetMDUnlink(mdh);
+ lnet_ping_buffer_decref(pbuf);
+ } else {
+ LIBCFS_ALLOC(rcd, sizeof(*rcd));
+ if (rcd == NULL)
+ goto out;
- LNetInvalidateHandle(&rcd->rcd_mdh);
- INIT_LIST_HEAD(&rcd->rcd_list);
+ LNetInvalidateMDHandle(&rcd->rcd_mdh);
+ INIT_LIST_HEAD(&rcd->rcd_list);
+ rcd->rcd_nnis = nnis;
+ }
- LIBCFS_ALLOC(pi, LNET_PINGINFO_SIZE);
- if (pi == NULL)
+ pbuf = lnet_ping_buffer_alloc(nnis, GFP_NOFS);
+ if (pbuf == NULL)
goto out;
- for (i = 0; i < LNET_MAX_RTR_NIS; i++) {
- pi->pi_ni[i].ns_nid = LNET_NID_ANY;
- pi->pi_ni[i].ns_status = LNET_NI_STATUS_INVALID;
+ for (i = 0; i < nnis; i++) {
+ pbuf->pb_info.pi_ni[i].ns_nid = LNET_NID_ANY;
+ pbuf->pb_info.pi_ni[i].ns_status = LNET_NI_STATUS_INVALID;
}
- rcd->rcd_pinginfo = pi;
+ rcd->rcd_pingbuffer = pbuf;
- LASSERT(!LNetHandleIsInvalid(the_lnet.ln_rc_eqh));
- rc = LNetMDBind((lnet_md_t){.start = pi,
+ LASSERT(!LNetEQHandleIsInvalid(the_lnet.ln_rc_eqh));
+ rc = LNetMDBind((struct lnet_md){.start = &pbuf->pb_info,
.user_ptr = rcd,
- .length = LNET_PINGINFO_SIZE,
+ .length = LNET_PING_INFO_SIZE(nnis),
.threshold = LNET_MD_THRESH_INF,
.options = LNET_MD_TRUNCATE,
.eq_handle = the_lnet.ln_rc_eqh},
&rcd->rcd_mdh);
if (rc < 0) {
CERROR("Can't bind MD: %d\n", rc);
- goto out;
+ goto out_ping_buffer_decref;
}
LASSERT(rc == 0);
lnet_net_lock(gateway->lpni_cpt);
- /* router table changed or someone has created rcd for this gateway */
- if (!lnet_isrouter(gateway) || gateway->lpni_rcd != NULL) {
- lnet_net_unlock(gateway->lpni_cpt);
- goto out;
+ /* Check if this is still a router. */
+ if (!lnet_isrouter(gateway))
+ goto out_unlock;
+ /* Check if someone else installed router data. */
+ if (gateway->lpni_rcd && gateway->lpni_rcd != rcd)
+ goto out_unlock;
+
+ /* Install and/or update the router data. */
+ if (!gateway->lpni_rcd) {
+ lnet_peer_ni_addref_locked(gateway);
+ rcd->rcd_gateway = gateway;
+ gateway->lpni_rcd = rcd;
}
-
- lnet_peer_ni_addref_locked(gateway);
- rcd->rcd_gateway = gateway;
- gateway->lpni_rcd = rcd;
gateway->lpni_ping_notsent = 0;
return rcd;
+out_unlock:
+ lnet_net_unlock(gateway->lpni_cpt);
+ rc = LNetMDUnlink(mdh);
+ LASSERT(rc == 0);
+out_ping_buffer_decref:
+ lnet_ping_buffer_decref(pbuf);
out:
- if (rcd != NULL) {
- if (!LNetHandleIsInvalid(rcd->rcd_mdh)) {
- rc = LNetMDUnlink(rcd->rcd_mdh);
- LASSERT(rc == 0);
- }
+ if (rcd && rcd != gateway->lpni_rcd)
lnet_destroy_rc_data(rcd);
- }
-
lnet_net_lock(gateway->lpni_cpt);
return gateway->lpni_rcd;
}
static int
-lnet_router_check_interval (struct lnet_peer_ni *rtr)
+lnet_router_check_interval(struct lnet_peer_ni *rtr)
{
int secs;
}
static void
-lnet_ping_router_locked (struct lnet_peer_ni *rtr)
+lnet_ping_router_locked(struct lnet_peer_ni *rtr)
{
- lnet_rc_data_t *rcd = NULL;
- cfs_time_t now = cfs_time_current();
- int secs;
- struct lnet_ni *ni;
+ struct lnet_rc_data *rcd = NULL;
+ time64_t now = ktime_get_seconds();
+ time64_t secs;
+ struct lnet_ni *ni;
lnet_peer_ni_addref_locked(rtr);
if (rtr->lpni_ping_deadline != 0 && /* ping timed out? */
- cfs_time_after(now, rtr->lpni_ping_deadline))
+ now > rtr->lpni_ping_deadline)
lnet_notify_locked(rtr, 1, 0, now);
/* Run any outstanding notifications */
lnet_ni_notify_locked(ni, rtr);
if (!lnet_isrouter(rtr) ||
- the_lnet.ln_rc_state != LNET_RC_STATE_RUNNING) {
+ the_lnet.ln_mt_state != LNET_MT_STATE_RUNNING) {
/* router table changed or router checker is shutting down */
lnet_peer_ni_decref_locked(rtr);
return;
}
- rcd = rtr->lpni_rcd != NULL ?
- rtr->lpni_rcd : lnet_create_rc_data_locked(rtr);
+ rcd = rtr->lpni_rcd;
+ /*
+ * The response to the router checker ping could've timed out and
+ * the mdh might've been invalidated, so we need to update it
+ * again.
+ */
+ if (!rcd || rcd->rcd_nnis > rcd->rcd_pingbuffer->pb_nnis ||
+ LNetMDHandleIsInvalid(rcd->rcd_mdh))
+ rcd = lnet_update_rc_data_locked(rtr);
if (rcd == NULL)
return;
secs = lnet_router_check_interval(rtr);
CDEBUG(D_NET,
- "rtr %s %d: deadline %lu ping_notsent %d alive %d "
- "alive_count %d lpni_ping_timestamp %lu\n",
+ "rtr %s %lld: deadline %lld ping_notsent %d alive %d "
+ "alive_count %d lpni_ping_timestamp %lld\n",
libcfs_nid2str(rtr->lpni_nid), secs,
rtr->lpni_ping_deadline, rtr->lpni_ping_notsent,
rtr->lpni_alive, rtr->lpni_alive_count, rtr->lpni_ping_timestamp);
if (secs != 0 && !rtr->lpni_ping_notsent &&
- cfs_time_after(now, cfs_time_add(rtr->lpni_ping_timestamp,
- cfs_time_seconds(secs)))) {
+ now > rtr->lpni_ping_timestamp + secs) {
int rc;
- lnet_process_id_t id;
- lnet_handle_md_t mdh;
+ struct lnet_process_id id;
+ struct lnet_handle_md mdh;
id.nid = rtr->lpni_nid;
id.pid = LNET_PID_LUSTRE;
mdh = rcd->rcd_mdh;
if (rtr->lpni_ping_deadline == 0) {
- rtr->lpni_ping_deadline =
- cfs_time_shift(router_ping_timeout);
+ rtr->lpni_ping_deadline = ktime_get_seconds() +
+ router_ping_timeout;
}
lnet_net_unlock(rtr->lpni_cpt);
rc = LNetGet(LNET_NID_ANY, mdh, id, LNET_RESERVED_PORTAL,
- LNET_PROTO_PING_MATCHBITS, 0);
+ LNET_PROTO_PING_MATCHBITS, 0, false);
lnet_net_lock(rtr->lpni_cpt);
if (rc != 0)
return;
}
-int
-lnet_router_checker_start(void)
+int lnet_router_pre_mt_start(void)
{
- int rc;
- int eqsz = 0;
- struct task_struct *task;
-
- LASSERT(the_lnet.ln_rc_state == LNET_RC_STATE_SHUTDOWN);
+ int rc;
if (check_routers_before_use &&
dead_router_check_interval <= 0) {
return -EINVAL;
}
- sema_init(&the_lnet.ln_rc_signal, 0);
-
rc = LNetEQAlloc(0, lnet_router_checker_event, &the_lnet.ln_rc_eqh);
if (rc != 0) {
- CERROR("Can't allocate EQ(%d): %d\n", eqsz, rc);
+ CERROR("Can't allocate EQ(0): %d\n", rc);
return -ENOMEM;
}
- the_lnet.ln_rc_state = LNET_RC_STATE_RUNNING;
- task = kthread_run(lnet_router_checker, NULL, "router_checker");
- if (IS_ERR(task)) {
- rc = PTR_ERR(task);
- CERROR("Can't start router checker thread: %d\n", rc);
- /* block until event callback signals exit */
- down(&the_lnet.ln_rc_signal);
- rc = LNetEQFree(the_lnet.ln_rc_eqh);
- LASSERT(rc == 0);
- the_lnet.ln_rc_state = LNET_RC_STATE_SHUTDOWN;
- return -ENOMEM;
- }
+ return 0;
+}
+void lnet_router_post_mt_start(void)
+{
if (check_routers_before_use) {
/* Note that a helpful side-effect of pinging all known routers
* at startup is that it makes them drop stale connections they
* may have to a previous instance of me. */
lnet_wait_known_routerstate();
}
-
- return 0;
}
void
-lnet_router_checker_stop (void)
+lnet_router_cleanup(void)
{
int rc;
- if (the_lnet.ln_rc_state == LNET_RC_STATE_SHUTDOWN)
- return;
-
- LASSERT (the_lnet.ln_rc_state == LNET_RC_STATE_RUNNING);
- the_lnet.ln_rc_state = LNET_RC_STATE_STOPPING;
- /* wakeup the RC thread if it's sleeping */
- wake_up(&the_lnet.ln_rc_waitq);
-
- /* block until event callback signals exit */
- down(&the_lnet.ln_rc_signal);
- LASSERT(the_lnet.ln_rc_state == LNET_RC_STATE_SHUTDOWN);
-
rc = LNetEQFree(the_lnet.ln_rc_eqh);
LASSERT(rc == 0);
return;
}
-static void
+void
lnet_prune_rc_data(int wait_unlink)
{
- lnet_rc_data_t *rcd;
- lnet_rc_data_t *tmp;
- struct lnet_peer_ni *lp;
- struct list_head head;
- int i = 2;
+ struct lnet_rc_data *rcd;
+ struct lnet_rc_data *tmp;
+ struct lnet_peer_ni *lp;
+ struct list_head head;
+ int i = 2;
- if (likely(the_lnet.ln_rc_state == LNET_RC_STATE_RUNNING &&
+ if (likely(the_lnet.ln_mt_state == LNET_MT_STATE_RUNNING &&
list_empty(&the_lnet.ln_rcd_deathrow) &&
list_empty(&the_lnet.ln_rcd_zombie)))
return;
lnet_net_lock(LNET_LOCK_EX);
- if (the_lnet.ln_rc_state != LNET_RC_STATE_RUNNING) {
+ if (the_lnet.ln_mt_state != LNET_MT_STATE_RUNNING) {
/* router checker is stopping, prune all */
list_for_each_entry(lp, &the_lnet.ln_routers,
lpni_rtr_list) {
while (!list_empty(&the_lnet.ln_rcd_zombie)) {
list_for_each_entry_safe(rcd, tmp, &the_lnet.ln_rcd_zombie,
rcd_list) {
- if (LNetHandleIsInvalid(rcd->rcd_mdh))
+ if (LNetMDHandleIsInvalid(rcd->rcd_mdh))
list_move(&rcd->rcd_list, &head);
}
while (!list_empty(&head)) {
rcd = list_entry(head.next,
- lnet_rc_data_t, rcd_list);
+ struct lnet_rc_data, rcd_list);
list_del_init(&rcd->rcd_list);
lnet_destroy_rc_data(rcd);
}
}
/*
- * This function is called to check if the RC should block indefinitely.
- * It's called from lnet_router_checker() as well as being passed to
- * wait_event_interruptible() to avoid the lost wake_up problem.
- *
- * When it's called from wait_event_interruptible() it is necessary to
- * also not sleep if the rc state is not running to avoid a deadlock
- * when the system is shutting down
+ * This function is called from the monitor thread to check if there are
+ * any active routers that need to be checked.
*/
-static inline bool
+inline bool
lnet_router_checker_active(void)
{
- if (the_lnet.ln_rc_state != LNET_RC_STATE_RUNNING)
+ if (the_lnet.ln_mt_state != LNET_MT_STATE_RUNNING)
return true;
/* Router Checker thread needs to run when routing is enabled in
if (the_lnet.ln_routing)
return true;
+ /* if there are routers that need to be cleaned up then do so */
+ if (!list_empty(&the_lnet.ln_rcd_deathrow) ||
+ !list_empty(&the_lnet.ln_rcd_zombie))
+ return true;
+
return !list_empty(&the_lnet.ln_routers) &&
(live_router_check_interval > 0 ||
dead_router_check_interval > 0);
}
-static int
-lnet_router_checker(void *arg)
+void
+lnet_check_routers(void)
{
struct lnet_peer_ni *rtr;
struct list_head *entry;
+ __u64 version;
+ int cpt;
+ int cpt2;
- cfs_block_allsigs();
-
- while (the_lnet.ln_rc_state == LNET_RC_STATE_RUNNING) {
- __u64 version;
- int cpt;
- int cpt2;
-
- cpt = lnet_net_lock_current();
+ cpt = lnet_net_lock_current();
rescan:
- version = the_lnet.ln_routers_version;
-
- list_for_each(entry, &the_lnet.ln_routers) {
- rtr = list_entry(entry, struct lnet_peer_ni,
- lpni_rtr_list);
+ version = the_lnet.ln_routers_version;
- cpt2 = rtr->lpni_cpt;
- if (cpt != cpt2) {
- lnet_net_unlock(cpt);
- cpt = cpt2;
- lnet_net_lock(cpt);
- /* the routers list has changed */
- if (version != the_lnet.ln_routers_version)
- goto rescan;
- }
+ list_for_each(entry, &the_lnet.ln_routers) {
+ rtr = list_entry(entry, struct lnet_peer_ni,
+ lpni_rtr_list);
- lnet_ping_router_locked(rtr);
-
- /* NB dropped lock */
- if (version != the_lnet.ln_routers_version) {
- /* the routers list has changed */
+ cpt2 = rtr->lpni_cpt;
+ if (cpt != cpt2) {
+ lnet_net_unlock(cpt);
+ cpt = cpt2;
+ lnet_net_lock(cpt);
+ /* the routers list has changed */
+ if (version != the_lnet.ln_routers_version)
goto rescan;
- }
}
- if (the_lnet.ln_routing)
- lnet_update_ni_status_locked();
-
- lnet_net_unlock(cpt);
-
- lnet_prune_rc_data(0); /* don't wait for UNLINK */
+ lnet_ping_router_locked(rtr);
- /* Call schedule_timeout() here always adds 1 to load average
- * because kernel counts # active tasks as nr_running
- * + nr_uninterruptible. */
- /* if there are any routes then wakeup every second. If
- * there are no routes then sleep indefinitely until woken
- * up by a user adding a route */
- if (!lnet_router_checker_active())
- wait_event_interruptible(the_lnet.ln_rc_waitq,
- lnet_router_checker_active());
- else
- wait_event_interruptible_timeout(the_lnet.ln_rc_waitq,
- false,
- cfs_time_seconds(1));
+ /* NB dropped lock */
+ if (version != the_lnet.ln_routers_version) {
+ /* the routers list has changed */
+ goto rescan;
+ }
}
- lnet_prune_rc_data(1); /* wait for UNLINK */
+ if (the_lnet.ln_routing)
+ lnet_update_ni_status_locked();
- the_lnet.ln_rc_state = LNET_RC_STATE_SHUTDOWN;
- up(&the_lnet.ln_rc_signal);
- /* The unlink event callback will signal final completion */
- return 0;
+ lnet_net_unlock(cpt);
+
+ lnet_prune_rc_data(0); /* don't wait for UNLINK */
}
void
-lnet_destroy_rtrbuf(lnet_rtrbuf_t *rb, int npages)
+lnet_destroy_rtrbuf(struct lnet_rtrbuf *rb, int npages)
{
- int sz = offsetof(lnet_rtrbuf_t, rb_kiov[npages]);
+ int sz = offsetof(struct lnet_rtrbuf, rb_kiov[npages]);
while (--npages >= 0)
__free_page(rb->rb_kiov[npages].kiov_page);
LIBCFS_FREE(rb, sz);
}
-static lnet_rtrbuf_t *
-lnet_new_rtrbuf(lnet_rtrbufpool_t *rbp, int cpt)
+static struct lnet_rtrbuf *
+lnet_new_rtrbuf(struct lnet_rtrbufpool *rbp, int cpt)
{
int npages = rbp->rbp_npages;
- int sz = offsetof(lnet_rtrbuf_t, rb_kiov[npages]);
+ int sz = offsetof(struct lnet_rtrbuf, rb_kiov[npages]);
struct page *page;
- lnet_rtrbuf_t *rb;
+ struct lnet_rtrbuf *rb;
int i;
LIBCFS_CPT_ALLOC(rb, lnet_cpt_table(), cpt, sz);
}
static void
-lnet_rtrpool_free_bufs(lnet_rtrbufpool_t *rbp, int cpt)
+lnet_rtrpool_free_bufs(struct lnet_rtrbufpool *rbp, int cpt)
{
- int npages = rbp->rbp_npages;
- lnet_rtrbuf_t *rb;
+ int npages = rbp->rbp_npages;
+ struct lnet_rtrbuf *rb;
struct list_head tmp;
if (rbp->rbp_nbuffers == 0) /* not initialized or already freed */
/* Free buffers on the free list. */
while (!list_empty(&tmp)) {
- rb = list_entry(tmp.next, lnet_rtrbuf_t, rb_list);
+ rb = list_entry(tmp.next, struct lnet_rtrbuf, rb_list);
list_del(&rb->rb_list);
lnet_destroy_rtrbuf(rb, npages);
}
}
static int
-lnet_rtrpool_adjust_bufs(lnet_rtrbufpool_t *rbp, int nbufs, int cpt)
+lnet_rtrpool_adjust_bufs(struct lnet_rtrbufpool *rbp, int nbufs, int cpt)
{
struct list_head rb_list;
- lnet_rtrbuf_t *rb;
+ struct lnet_rtrbuf *rb;
int num_rb;
int num_buffers = 0;
int old_req_nbufs;
failed:
while (!list_empty(&rb_list)) {
- rb = list_entry(rb_list.next, lnet_rtrbuf_t, rb_list);
+ rb = list_entry(rb_list.next, struct lnet_rtrbuf, rb_list);
list_del(&rb->rb_list);
lnet_destroy_rtrbuf(rb, npages);
}
}
static void
-lnet_rtrpool_init(lnet_rtrbufpool_t *rbp, int npages)
+lnet_rtrpool_init(struct lnet_rtrbufpool *rbp, int npages)
{
INIT_LIST_HEAD(&rbp->rbp_msgs);
INIT_LIST_HEAD(&rbp->rbp_bufs);
void
lnet_rtrpools_free(int keep_pools)
{
- lnet_rtrbufpool_t *rtrp;
+ struct lnet_rtrbufpool *rtrp;
int i;
if (the_lnet.ln_rtrpools == NULL) /* uninitialized or freed */
int
lnet_rtrpools_alloc(int im_a_router)
{
- lnet_rtrbufpool_t *rtrp;
+ struct lnet_rtrbufpool *rtrp;
int nrb_tiny;
int nrb_small;
int nrb_large;
the_lnet.ln_rtrpools = cfs_percpt_alloc(lnet_cpt_table(),
LNET_NRBPOOLS *
- sizeof(lnet_rtrbufpool_t));
+ sizeof(struct lnet_rtrbufpool));
if (the_lnet.ln_rtrpools == NULL) {
LCONSOLE_ERROR_MSG(0x10c,
"Failed to initialize router buffe pool\n");
int nrb = 0;
int rc = 0;
int i;
- lnet_rtrbufpool_t *rtrp;
+ struct lnet_rtrbufpool *rtrp;
/* If the provided values for each buffer pool are different than the
* configured values, we need to take action. */
lnet_net_lock(LNET_LOCK_EX);
the_lnet.ln_routing = 1;
- the_lnet.ln_ping_info->pi_features &= ~LNET_PING_FEAT_RTE_DISABLED;
+ the_lnet.ln_ping_target->pb_info.pi_features &=
+ ~LNET_PING_FEAT_RTE_DISABLED;
lnet_net_unlock(LNET_LOCK_EX);
return rc;
lnet_net_lock(LNET_LOCK_EX);
the_lnet.ln_routing = 0;
- the_lnet.ln_ping_info->pi_features |= LNET_PING_FEAT_RTE_DISABLED;
+ the_lnet.ln_ping_target->pb_info.pi_features |=
+ LNET_PING_FEAT_RTE_DISABLED;
tiny_router_buffers = 0;
small_router_buffers = 0;
}
int
-lnet_notify(lnet_ni_t *ni, lnet_nid_t nid, int alive, cfs_time_t when)
+lnet_notify(struct lnet_ni *ni, lnet_nid_t nid, int alive, time64_t when)
{
struct lnet_peer_ni *lp = NULL;
- cfs_time_t now = cfs_time_current();
+ time64_t now = ktime_get_seconds();
int cpt = lnet_cpt_of_nid(nid, ni);
LASSERT (!in_interrupt ());
}
/* can't do predictions... */
- if (cfs_time_after(when, now)) {
+ if (when > now) {
CWARN("Ignoring prediction from %s of %s %s "
- "%ld seconds in the future\n",
+ "%lld seconds in the future\n",
(ni == NULL) ? "userspace" : libcfs_nid2str(ni->ni_nid),
- libcfs_nid2str(nid), alive ? "up" : "down",
- cfs_duration_sec(cfs_time_sub(when, now)));
+ libcfs_nid2str(nid), alive ? "up" : "down", when - now);
return -EINVAL;
}
lnet_net_lock(cpt);
- if (the_lnet.ln_shutdown) {
+ if (the_lnet.ln_state != LNET_STATE_RUNNING) {
lnet_net_unlock(cpt);
return -ESHUTDOWN;
}