X-Git-Url: https://git.whamcloud.com/?a=blobdiff_plain;f=lnet%2Flnet%2Fapi-ni.c;h=1a5aa3436ca995724d903c4627a5262345b1f407;hb=b0ab95d6133e783acacc6329c025d17fb282775e;hp=4910d3ae9ddc683b646c7897ae8187d06f392b46;hpb=8cbb8cd3e771e7f7e0f99cafc19fad32770dc015;p=fs%2Flustre-release.git diff --git a/lnet/lnet/api-ni.c b/lnet/lnet/api-ni.c index 4910d3a..1a5aa34 100644 --- a/lnet/lnet/api-ni.c +++ b/lnet/lnet/api-ni.c @@ -38,7 +38,7 @@ #define D_LNI D_CONSOLE -lnet_t the_lnet; /* THE state of the network */ +struct lnet the_lnet; /* THE state of the network */ EXPORT_SYMBOL(the_lnet); static char *ip2nets = ""; @@ -62,8 +62,22 @@ module_param(use_tcp_bonding, int, 0444); MODULE_PARM_DESC(use_tcp_bonding, "Set to 1 to use socklnd bonding. 0 to use Multi-Rail"); -static int lnet_ping(lnet_process_id_t id, signed long timeout, - lnet_process_id_t __user *ids, int n_ids); +unsigned int lnet_numa_range = 0; +module_param(lnet_numa_range, uint, 0444); +MODULE_PARM_DESC(lnet_numa_range, + "NUMA range to consider during Multi-Rail selection"); + +/* + * This sequence number keeps track of how many times DLC was used to + * update the local NIs. It is incremented when a NI is added or + * removed and checked when sending a message to determine if there is + * a need to re-run the selection algorithm. See lnet_select_pathway() + * for more details on its usage. + */ +static atomic_t lnet_dlc_seq_no = ATOMIC_INIT(0); + +static int lnet_ping(struct lnet_process_id id, signed long timeout, + struct lnet_process_id __user *ids, int n_ids); static char * lnet_get_routes(void) @@ -119,7 +133,7 @@ lnet_descriptor_setup(void) /* create specific kmem_cache for MEs and small MDs (i.e., originally * allocated in kmem_cache). */ - lnet_mes_cachep = kmem_cache_create("lnet_MEs", sizeof(lnet_me_t), + lnet_mes_cachep = kmem_cache_create("lnet_MEs", sizeof(struct lnet_me), 0, 0, NULL); if (!lnet_mes_cachep) return -ENOMEM; @@ -245,83 +259,83 @@ static void lnet_assert_wire_constants(void) CLASSERT((int)offsetof(struct lnet_handle_wire, wh_object_cookie) == 8); CLASSERT((int)sizeof(((struct lnet_handle_wire *)0)->wh_object_cookie) == 8); - /* Checks for struct lnet_magicversion_t */ - CLASSERT((int)sizeof(lnet_magicversion_t) == 8); - CLASSERT((int)offsetof(lnet_magicversion_t, magic) == 0); - CLASSERT((int)sizeof(((lnet_magicversion_t *)0)->magic) == 4); - CLASSERT((int)offsetof(lnet_magicversion_t, version_major) == 4); - CLASSERT((int)sizeof(((lnet_magicversion_t *)0)->version_major) == 2); - CLASSERT((int)offsetof(lnet_magicversion_t, version_minor) == 6); - CLASSERT((int)sizeof(((lnet_magicversion_t *)0)->version_minor) == 2); - - /* Checks for struct lnet_hdr_t */ - CLASSERT((int)sizeof(lnet_hdr_t) == 72); - CLASSERT((int)offsetof(lnet_hdr_t, dest_nid) == 0); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->dest_nid) == 8); - CLASSERT((int)offsetof(lnet_hdr_t, src_nid) == 8); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->src_nid) == 8); - CLASSERT((int)offsetof(lnet_hdr_t, dest_pid) == 16); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->dest_pid) == 4); - CLASSERT((int)offsetof(lnet_hdr_t, src_pid) == 20); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->src_pid) == 4); - CLASSERT((int)offsetof(lnet_hdr_t, type) == 24); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->type) == 4); - CLASSERT((int)offsetof(lnet_hdr_t, payload_length) == 28); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->payload_length) == 4); - CLASSERT((int)offsetof(lnet_hdr_t, msg) == 32); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg) == 40); + /* Checks for struct struct lnet_magicversion */ + CLASSERT((int)sizeof(struct lnet_magicversion) == 8); + CLASSERT((int)offsetof(struct lnet_magicversion, magic) == 0); + CLASSERT((int)sizeof(((struct lnet_magicversion *)0)->magic) == 4); + CLASSERT((int)offsetof(struct lnet_magicversion, version_major) == 4); + CLASSERT((int)sizeof(((struct lnet_magicversion *)0)->version_major) == 2); + CLASSERT((int)offsetof(struct lnet_magicversion, version_minor) == 6); + CLASSERT((int)sizeof(((struct lnet_magicversion *)0)->version_minor) == 2); + + /* Checks for struct struct lnet_hdr */ + CLASSERT((int)sizeof(struct lnet_hdr) == 72); + CLASSERT((int)offsetof(struct lnet_hdr, dest_nid) == 0); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->dest_nid) == 8); + CLASSERT((int)offsetof(struct lnet_hdr, src_nid) == 8); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->src_nid) == 8); + CLASSERT((int)offsetof(struct lnet_hdr, dest_pid) == 16); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->dest_pid) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, src_pid) == 20); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->src_pid) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, type) == 24); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->type) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, payload_length) == 28); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->payload_length) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg) == 32); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg) == 40); /* Ack */ - CLASSERT((int)offsetof(lnet_hdr_t, msg.ack.dst_wmd) == 32); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.ack.dst_wmd) == 16); - CLASSERT((int)offsetof(lnet_hdr_t, msg.ack.match_bits) == 48); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.ack.match_bits) == 8); - CLASSERT((int)offsetof(lnet_hdr_t, msg.ack.mlength) == 56); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.ack.mlength) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg.ack.dst_wmd) == 32); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.ack.dst_wmd) == 16); + CLASSERT((int)offsetof(struct lnet_hdr, msg.ack.match_bits) == 48); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.ack.match_bits) == 8); + CLASSERT((int)offsetof(struct lnet_hdr, msg.ack.mlength) == 56); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.ack.mlength) == 4); /* Put */ - CLASSERT((int)offsetof(lnet_hdr_t, msg.put.ack_wmd) == 32); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.put.ack_wmd) == 16); - CLASSERT((int)offsetof(lnet_hdr_t, msg.put.match_bits) == 48); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.put.match_bits) == 8); - CLASSERT((int)offsetof(lnet_hdr_t, msg.put.hdr_data) == 56); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.put.hdr_data) == 8); - CLASSERT((int)offsetof(lnet_hdr_t, msg.put.ptl_index) == 64); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.put.ptl_index) == 4); - CLASSERT((int)offsetof(lnet_hdr_t, msg.put.offset) == 68); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.put.offset) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg.put.ack_wmd) == 32); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.put.ack_wmd) == 16); + CLASSERT((int)offsetof(struct lnet_hdr, msg.put.match_bits) == 48); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.put.match_bits) == 8); + CLASSERT((int)offsetof(struct lnet_hdr, msg.put.hdr_data) == 56); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.put.hdr_data) == 8); + CLASSERT((int)offsetof(struct lnet_hdr, msg.put.ptl_index) == 64); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.put.ptl_index) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg.put.offset) == 68); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.put.offset) == 4); /* Get */ - CLASSERT((int)offsetof(lnet_hdr_t, msg.get.return_wmd) == 32); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.get.return_wmd) == 16); - CLASSERT((int)offsetof(lnet_hdr_t, msg.get.match_bits) == 48); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.get.match_bits) == 8); - CLASSERT((int)offsetof(lnet_hdr_t, msg.get.ptl_index) == 56); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.get.ptl_index) == 4); - CLASSERT((int)offsetof(lnet_hdr_t, msg.get.src_offset) == 60); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.get.src_offset) == 4); - CLASSERT((int)offsetof(lnet_hdr_t, msg.get.sink_length) == 64); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.get.sink_length) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg.get.return_wmd) == 32); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.get.return_wmd) == 16); + CLASSERT((int)offsetof(struct lnet_hdr, msg.get.match_bits) == 48); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.get.match_bits) == 8); + CLASSERT((int)offsetof(struct lnet_hdr, msg.get.ptl_index) == 56); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.get.ptl_index) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg.get.src_offset) == 60); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.get.src_offset) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg.get.sink_length) == 64); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.get.sink_length) == 4); /* Reply */ - CLASSERT((int)offsetof(lnet_hdr_t, msg.reply.dst_wmd) == 32); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.reply.dst_wmd) == 16); + CLASSERT((int)offsetof(struct lnet_hdr, msg.reply.dst_wmd) == 32); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.reply.dst_wmd) == 16); /* Hello */ - CLASSERT((int)offsetof(lnet_hdr_t, msg.hello.incarnation) == 32); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.hello.incarnation) == 8); - CLASSERT((int)offsetof(lnet_hdr_t, msg.hello.type) == 40); - CLASSERT((int)sizeof(((lnet_hdr_t *)0)->msg.hello.type) == 4); + CLASSERT((int)offsetof(struct lnet_hdr, msg.hello.incarnation) == 32); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.hello.incarnation) == 8); + CLASSERT((int)offsetof(struct lnet_hdr, msg.hello.type) == 40); + CLASSERT((int)sizeof(((struct lnet_hdr *)0)->msg.hello.type) == 4); } -static lnd_t *lnet_find_lnd_by_type(__u32 type) +static struct lnet_lnd *lnet_find_lnd_by_type(__u32 type) { - lnd_t *lnd; + struct lnet_lnd *lnd; struct list_head *tmp; /* holding lnd mutex */ list_for_each(tmp, &the_lnet.ln_lnds) { - lnd = list_entry(tmp, lnd_t, lnd_list); + lnd = list_entry(tmp, struct lnet_lnd, lnd_list); if (lnd->lnd_type == type) return lnd; @@ -330,7 +344,7 @@ static lnd_t *lnet_find_lnd_by_type(__u32 type) } void -lnet_register_lnd (lnd_t *lnd) +lnet_register_lnd(struct lnet_lnd *lnd) { mutex_lock(&the_lnet.ln_lnd_mutex); @@ -347,7 +361,7 @@ lnet_register_lnd (lnd_t *lnd) EXPORT_SYMBOL(lnet_register_lnd); void -lnet_unregister_lnd (lnd_t *lnd) +lnet_unregister_lnd(struct lnet_lnd *lnd) { mutex_lock(&the_lnet.ln_lnd_mutex); @@ -362,9 +376,9 @@ lnet_unregister_lnd (lnd_t *lnd) EXPORT_SYMBOL(lnet_unregister_lnd); void -lnet_counters_get(lnet_counters_t *counters) +lnet_counters_get(struct lnet_counters *counters) { - lnet_counters_t *ctr; + struct lnet_counters *ctr; int i; memset(counters, 0, sizeof(*counters)); @@ -392,13 +406,13 @@ EXPORT_SYMBOL(lnet_counters_get); void lnet_counters_reset(void) { - lnet_counters_t *counters; + struct lnet_counters *counters; int i; lnet_net_lock(LNET_LOCK_EX); cfs_percpt_for_each(counters, i, the_lnet.ln_counters) - memset(counters, 0, sizeof(lnet_counters_t)); + memset(counters, 0, sizeof(struct lnet_counters)); lnet_net_unlock(LNET_LOCK_EX); } @@ -431,10 +445,10 @@ lnet_res_container_cleanup(struct lnet_res_container *rec) list_del_init(e); if (rec->rec_type == LNET_COOKIE_TYPE_EQ) { - lnet_eq_free(list_entry(e, lnet_eq_t, eq_list)); + lnet_eq_free(list_entry(e, struct lnet_eq, eq_list)); } else if (rec->rec_type == LNET_COOKIE_TYPE_MD) { - lnet_md_free(list_entry(e, lnet_libmd_t, md_list)); + lnet_md_free(list_entry(e, struct lnet_libmd, md_list)); } else { /* NB: Active MEs should be attached on portals */ LBUG(); @@ -530,12 +544,12 @@ lnet_res_containers_create(int type) return recs; } -lnet_libhandle_t * +struct lnet_libhandle * lnet_res_lh_lookup(struct lnet_res_container *rec, __u64 cookie) { /* ALWAYS called with lnet_res_lock held */ struct list_head *head; - lnet_libhandle_t *lh; + struct lnet_libhandle *lh; unsigned int hash; if ((cookie & LNET_COOKIE_MASK) != rec->rec_type) @@ -553,7 +567,8 @@ lnet_res_lh_lookup(struct lnet_res_container *rec, __u64 cookie) } void -lnet_res_lh_initialize(struct lnet_res_container *rec, lnet_libhandle_t *lh) +lnet_res_lh_initialize(struct lnet_res_container *rec, + struct lnet_libhandle *lh) { /* ALWAYS called with lnet_res_lock held */ unsigned int ibits = LNET_COOKIE_TYPE_BITS + LNET_CPT_BITS; @@ -589,6 +604,8 @@ lnet_prepare(lnet_pid_t requested_pid) the_lnet.ln_pid = requested_pid; INIT_LIST_HEAD(&the_lnet.ln_test_peers); + INIT_LIST_HEAD(&the_lnet.ln_peers); + INIT_LIST_HEAD(&the_lnet.ln_remote_peer_ni_list); INIT_LIST_HEAD(&the_lnet.ln_nets); INIT_LIST_HEAD(&the_lnet.ln_routers); INIT_LIST_HEAD(&the_lnet.ln_drop_rules); @@ -609,7 +626,7 @@ lnet_prepare(lnet_pid_t requested_pid) the_lnet.ln_interface_cookie = ktime_get_real_ns(); the_lnet.ln_counters = cfs_percpt_alloc(lnet_cpt_table(), - sizeof(lnet_counters_t)); + sizeof(struct lnet_counters)); if (the_lnet.ln_counters == NULL) { CERROR("Failed to allocate counters for LNet\n"); rc = -ENOMEM; @@ -687,7 +704,7 @@ lnet_unprepare (void) lnet_res_container_cleanup(&the_lnet.ln_eq_container); lnet_msg_containers_destroy(); - lnet_peer_tables_destroy(); + lnet_peer_uninit(); lnet_rtrpools_free(0); if (the_lnet.ln_counters != NULL) { @@ -700,7 +717,7 @@ lnet_unprepare (void) return 0; } -lnet_ni_t * +struct lnet_ni * lnet_net2ni_locked(__u32 net_id, int cpt) { struct lnet_ni *ni; @@ -719,18 +736,20 @@ lnet_net2ni_locked(__u32 net_id, int cpt) return NULL; } -lnet_ni_t * -lnet_net2ni(__u32 net) +struct lnet_ni * +lnet_net2ni_addref(__u32 net) { - lnet_ni_t *ni; + struct lnet_ni *ni; lnet_net_lock(0); ni = lnet_net2ni_locked(net, 0); + if (ni) + lnet_ni_addref_locked(ni, 0); lnet_net_unlock(0); return ni; } -EXPORT_SYMBOL(lnet_net2ni); +EXPORT_SYMBOL(lnet_net2ni_addref); struct lnet_net * lnet_get_net_locked(__u32 net_id) @@ -745,7 +764,7 @@ lnet_get_net_locked(__u32 net_id) return NULL; } -static unsigned int +unsigned int lnet_nid_cpt_hash(lnet_nid_t nid, unsigned int number) { __u64 key = nid; @@ -845,7 +864,7 @@ lnet_is_ni_healthy_locked(struct lnet_ni *ni) return false; } -lnet_ni_t * +struct lnet_ni * lnet_nid2ni_locked(lnet_nid_t nid, int cpt) { struct lnet_net *net; @@ -863,10 +882,10 @@ lnet_nid2ni_locked(lnet_nid_t nid, int cpt) return NULL; } -lnet_ni_t * +struct lnet_ni * lnet_nid2ni_addref(lnet_nid_t nid) { - lnet_ni_t *ni; + struct lnet_ni *ni; lnet_net_lock(0); ni = lnet_nid2ni_locked(nid, 0); @@ -946,6 +965,18 @@ lnet_get_net_ni_count_locked(struct lnet_net *net) } static inline int +lnet_get_net_ni_count_pre(struct lnet_net *net) +{ + struct lnet_ni *ni; + int count = 0; + + list_for_each_entry(ni, &net->net_ni_added, ni_netlist) + count++; + + return count; +} + +static inline int lnet_get_ni_count(void) { struct lnet_ni *ni; @@ -995,7 +1026,7 @@ lnet_ping_info_destroy(void) } static void -lnet_ping_event_handler(lnet_event_t *event) +lnet_ping_event_handler(struct lnet_event *event) { struct lnet_ping_info *pinfo = event->md.user_ptr; @@ -1004,13 +1035,17 @@ lnet_ping_event_handler(lnet_event_t *event) } static int -lnet_ping_info_setup(struct lnet_ping_info **ppinfo, lnet_handle_md_t *md_handle, +lnet_ping_info_setup(struct lnet_ping_info **ppinfo, + struct lnet_handle_md *md_handle, int ni_count, bool set_eq) { - lnet_handle_me_t me_handle; - lnet_process_id_t id = {LNET_NID_ANY, LNET_PID_ANY}; - lnet_md_t md = {NULL}; - int rc, rc2; + struct lnet_process_id id = { + .nid = LNET_NID_ANY, + .pid = LNET_PID_ANY + }; + struct lnet_handle_me me_handle; + struct lnet_md md = { NULL }; + int rc, rc2; if (set_eq) { rc = LNetEQAlloc(0, lnet_ping_event_handler, @@ -1069,12 +1104,12 @@ failed_0: } static void -lnet_ping_md_unlink(struct lnet_ping_info *pinfo, lnet_handle_md_t *md_handle) +lnet_ping_md_unlink(struct lnet_ping_info *pinfo, struct lnet_handle_md *md_handle) { sigset_t blocked = cfs_block_allsigs(); LNetMDUnlink(*md_handle); - LNetInvalidateHandle(md_handle); + LNetInvalidateMDHandle(md_handle); /* NB md could be busy; this just starts the unlink */ while (pinfo->pi_features != LNET_PING_FEAT_INVAL) { @@ -1117,10 +1152,11 @@ lnet_ping_info_install_locked(struct lnet_ping_info *ping_info) } static void -lnet_ping_target_update(struct lnet_ping_info *pinfo, lnet_handle_md_t md_handle) +lnet_ping_target_update(struct lnet_ping_info *pinfo, + struct lnet_handle_md md_handle) { struct lnet_ping_info *old_pinfo = NULL; - lnet_handle_md_t old_md; + struct lnet_handle_md old_md; /* switch the NIs to point to the new ping info created */ lnet_net_lock(LNET_LOCK_EX); @@ -1160,7 +1196,7 @@ lnet_ping_target_fini(void) } static int -lnet_ni_tq_credits(lnet_ni_t *ni) +lnet_ni_tq_credits(struct lnet_ni *ni) { int credits; @@ -1177,7 +1213,7 @@ lnet_ni_tq_credits(lnet_ni_t *ni) } static void -lnet_ni_unlink_locked(lnet_ni_t *ni) +lnet_ni_unlink_locked(struct lnet_ni *ni) { if (!list_empty(&ni->ni_cptlist)) { list_del_init(&ni->ni_cptlist); @@ -1195,7 +1231,7 @@ lnet_clear_zombies_nis_locked(struct lnet_net *net) { int i; int islo; - lnet_ni_t *ni; + struct lnet_ni *ni; struct list_head *zombie_list = &net->net_ni_zombie; /* @@ -1208,7 +1244,7 @@ lnet_clear_zombies_nis_locked(struct lnet_net *net) int j; ni = list_entry(zombie_list->next, - lnet_ni_t, ni_netlist); + struct lnet_ni, ni_netlist); list_del_init(&ni->ni_netlist); /* the ni should be in deleting state. If it's not it's * a bug */ @@ -1262,15 +1298,13 @@ lnet_shutdown_lndni(struct lnet_ni *ni) lnet_net_lock(LNET_LOCK_EX); ni->ni_state = LNET_NI_STATE_DELETING; lnet_ni_unlink_locked(ni); + lnet_incr_dlc_seq(); lnet_net_unlock(LNET_LOCK_EX); /* clear messages for this NI on the lazy portal */ for (i = 0; i < the_lnet.ln_nportals; i++) lnet_clear_lazy_portal(ni, i, "Shutting down NI"); - /* Do peer table cleanup for this ni */ - lnet_peer_tables_cleanup(ni); - lnet_net_lock(LNET_LOCK_EX); lnet_clear_zombies_nis_locked(net); lnet_net_unlock(LNET_LOCK_EX); @@ -1289,12 +1323,18 @@ lnet_shutdown_lndnet(struct lnet_net *net) while (!list_empty(&net->net_ni_list)) { ni = list_entry(net->net_ni_list.next, - lnet_ni_t, ni_netlist); + struct lnet_ni, ni_netlist); lnet_net_unlock(LNET_LOCK_EX); lnet_shutdown_lndni(ni); lnet_net_lock(LNET_LOCK_EX); } + lnet_net_unlock(LNET_LOCK_EX); + + /* Do peer table cleanup for this net */ + lnet_peer_tables_cleanup(net); + + lnet_net_lock(LNET_LOCK_EX); /* * decrement ref count on lnd only when the entire network goes * away @@ -1314,11 +1354,11 @@ lnet_shutdown_lndnets(void) /* NB called holding the global mutex */ /* All quiet on the API front */ - LASSERT(!the_lnet.ln_shutdown); + LASSERT(the_lnet.ln_state == LNET_STATE_RUNNING); LASSERT(the_lnet.ln_refcount == 0); lnet_net_lock(LNET_LOCK_EX); - the_lnet.ln_shutdown = 1; /* flag shutdown */ + the_lnet.ln_state = LNET_STATE_STOPPING; while (!list_empty(&the_lnet.ln_nets)) { /* @@ -1346,7 +1386,7 @@ lnet_shutdown_lndnets(void) } lnet_net_lock(LNET_LOCK_EX); - the_lnet.ln_shutdown = 0; + the_lnet.ln_state = LNET_STATE_SHUTDOWN; lnet_net_unlock(LNET_LOCK_EX); } @@ -1411,6 +1451,9 @@ lnet_startup_lndni(struct lnet_ni *ni, struct lnet_lnd_tunables *tun) tq->tq_credits = lnet_ni_tq_credits(ni); } + atomic_set(&ni->ni_tx_credits, + lnet_ni_tq_credits(ni) * ni->ni_ncpts); + CDEBUG(D_LNI, "Added LNI %s [%d/%d/%d/%d]\n", libcfs_nid2str(ni->ni_nid), ni->ni_net->net_tunables.lct_peer_tx_credits, @@ -1427,13 +1470,13 @@ failed0: static int lnet_startup_lndnet(struct lnet_net *net, struct lnet_lnd_tunables *tun) { - struct lnet_ni *ni; - struct lnet_net *net_l = NULL; + struct lnet_ni *ni; + struct lnet_net *net_l = NULL; struct list_head local_ni_list; int rc; int ni_count = 0; __u32 lnd_type; - lnd_t *lnd; + struct lnet_lnd *lnd; int peer_timeout = net->net_tunables.lct_peer_timeout; int maxtxcredits = @@ -1455,13 +1498,6 @@ lnet_startup_lndnet(struct lnet_net *net, struct lnet_lnd_tunables *tun) LASSERT(libcfs_isknown_lnd(lnd_type)); - if (lnd_type == CIBLND || lnd_type == OPENIBLND || - lnd_type == IIBLND || lnd_type == VIBLND) { - CERROR("LND %s obsoleted\n", libcfs_lnd2str(lnd_type)); - rc = -EINVAL; - goto failed0; - } - mutex_lock(&the_lnet.ln_lnd_mutex); lnd = lnet_find_lnd_by_type(lnd_type); @@ -1564,6 +1600,7 @@ lnet_startup_lndnet(struct lnet_net *net, struct lnet_lnd_tunables *tun) lnet_net_lock(LNET_LOCK_EX); list_splice_tail(&local_ni_list, &net_l->net_ni_list); + lnet_incr_dlc_seq(); lnet_net_unlock(LNET_LOCK_EX); /* if the network is not unique then we don't want to keep @@ -1620,6 +1657,15 @@ lnet_startup_lndnets(struct list_head *netlist) int rc; int ni_count = 0; + /* + * Change to running state before bringing up the LNDs. This + * allows lnet_shutdown_lndnets() to assert that we've passed + * through here. + */ + lnet_net_lock(LNET_LOCK_EX); + the_lnet.ln_state = LNET_STATE_RUNNING; + lnet_net_unlock(LNET_LOCK_EX); + while (!list_empty(netlist)) { net = list_entry(netlist->next, struct lnet_net, net_list); list_del_init(&net->net_list); @@ -1680,7 +1726,7 @@ int lnet_lib_init(void) } the_lnet.ln_refcount = 0; - LNetInvalidateHandle(&the_lnet.ln_rc_eqh); + LNetInvalidateEQHandle(&the_lnet.ln_rc_eqh); INIT_LIST_HEAD(&the_lnet.ln_lnds); INIT_LIST_HEAD(&the_lnet.ln_net_zombie); INIT_LIST_HEAD(&the_lnet.ln_rcd_zombie); @@ -1715,7 +1761,7 @@ void lnet_lib_exit(void) while (!list_empty(&the_lnet.ln_lnds)) lnet_unregister_lnd(list_entry(the_lnet.ln_lnds.next, - lnd_t, lnd_list)); + struct lnet_lnd, lnd_list)); lnet_destroy_locks(); } @@ -1741,7 +1787,7 @@ LNetNIInit(lnet_pid_t requested_pid) int rc; int ni_count; struct lnet_ping_info *pinfo; - lnet_handle_md_t md_handle; + struct lnet_handle_md md_handle; struct list_head net_head; struct lnet_net *net; @@ -1904,16 +1950,82 @@ EXPORT_SYMBOL(LNetNIFini); * parameters * * \param[in] ni network interface structure - * \param[out] cpt_count the number of cpts the ni is on - * \param[out] nid Network Interface ID - * \param[out] peer_timeout NI peer timeout - * \param[out] peer_tx_crdits NI peer transmit credits - * \param[out] peer_rtr_credits NI peer router credits - * \param[out] max_tx_credits NI max transmit credit - * \param[out] net_config Network configuration + * \param[out] cfg_ni NI config information + * \param[out] tun network and LND tunables + */ +static void +lnet_fill_ni_info(struct lnet_ni *ni, struct lnet_ioctl_config_ni *cfg_ni, + struct lnet_ioctl_config_lnd_tunables *tun, + struct lnet_ioctl_element_stats *stats, + __u32 tun_size) +{ + size_t min_size = 0; + int i; + + if (!ni || !cfg_ni || !tun) + return; + + if (ni->ni_interfaces[0] != NULL) { + for (i = 0; i < ARRAY_SIZE(ni->ni_interfaces); i++) { + if (ni->ni_interfaces[i] != NULL) { + strncpy(cfg_ni->lic_ni_intf[i], + ni->ni_interfaces[i], + sizeof(cfg_ni->lic_ni_intf[i])); + } + } + } + + cfg_ni->lic_nid = ni->ni_nid; + cfg_ni->lic_status = ni->ni_status->ns_status; + cfg_ni->lic_tcp_bonding = use_tcp_bonding; + cfg_ni->lic_dev_cpt = ni->ni_dev_cpt; + + memcpy(&tun->lt_cmn, &ni->ni_net->net_tunables, sizeof(tun->lt_cmn)); + + if (stats) { + stats->iel_send_count = atomic_read(&ni->ni_stats.send_count); + stats->iel_recv_count = atomic_read(&ni->ni_stats.recv_count); + } + + /* + * tun->lt_tun will always be present, but in order to be + * backwards compatible, we need to deal with the cases when + * tun->lt_tun is smaller than what the kernel has, because it + * comes from an older version of a userspace program, then we'll + * need to copy as much information as we have available space. + */ + min_size = tun_size - sizeof(tun->lt_cmn); + memcpy(&tun->lt_tun, &ni->ni_lnd_tunables, min_size); + + /* copy over the cpts */ + if (ni->ni_ncpts == LNET_CPT_NUMBER && + ni->ni_cpts == NULL) { + for (i = 0; i < ni->ni_ncpts; i++) + cfg_ni->lic_cpts[i] = i; + } else { + for (i = 0; + ni->ni_cpts != NULL && i < ni->ni_ncpts && + i < LNET_MAX_SHOW_NUM_CPT; + i++) + cfg_ni->lic_cpts[i] = ni->ni_cpts[i]; + } + cfg_ni->lic_ncpts = ni->ni_ncpts; +} + +/** + * NOTE: This is a legacy function left in the code to be backwards + * compatible with older userspace programs. It should eventually be + * removed. + * + * Grabs the ni data from the ni structure and fills the out + * parameters + * + * \param[in] ni network interface structure + * \param[out] config config information */ static void -lnet_fill_ni_info(struct lnet_ni *ni, struct lnet_ioctl_config_data *config) +lnet_fill_ni_info_legacy(struct lnet_ni *ni, + struct lnet_ioctl_config_data *config) { struct lnet_ioctl_net_config *net_config; struct lnet_ioctl_config_lnd_tunables *lnd_cfg = NULL; @@ -2060,7 +2172,7 @@ lnet_get_net_config(struct lnet_ioctl_config_data *config) if (ni != NULL) { rc = 0; lnet_ni_lock(ni); - lnet_fill_ni_info(ni, config); + lnet_fill_ni_info_legacy(ni, config); lnet_ni_unlock(ni); } @@ -2069,49 +2181,56 @@ lnet_get_net_config(struct lnet_ioctl_config_data *config) } int -lnet_dyn_add_ni(lnet_pid_t requested_pid, struct lnet_ioctl_config_data *conf) +lnet_get_ni_config(struct lnet_ioctl_config_ni *cfg_ni, + struct lnet_ioctl_config_lnd_tunables *tun, + struct lnet_ioctl_element_stats *stats, + __u32 tun_size) { - char *nets = conf->cfg_config_u.cfg_net.net_intf; - struct lnet_ping_info *pinfo; - lnet_handle_md_t md_handle; - struct lnet_net *net; - struct list_head net_head; - int rc; - lnet_remotenet_t *rnet; - int net_ni_count; - int num_acceptor_nets; - __u32 net_type; - struct lnet_ioctl_config_lnd_tunables *lnd_tunables = NULL; - - INIT_LIST_HEAD(&net_head); + struct lnet_ni *ni; + int cpt; + int rc = -ENOENT; - if (conf && conf->cfg_hdr.ioc_len > sizeof(*conf)) - lnd_tunables = (struct lnet_ioctl_config_lnd_tunables *)conf->cfg_bulk; + if (!cfg_ni || !tun || !stats) + return -EINVAL; - /* Create a net/ni structures for the network string */ - rc = lnet_parse_networks(&net_head, nets, use_tcp_bonding); - if (rc <= 0) - return rc == 0 ? -EINVAL : rc; + cpt = lnet_net_lock_current(); - mutex_lock(&the_lnet.ln_api_mutex); + ni = lnet_get_ni_idx_locked(cfg_ni->lic_idx); - if (rc > 1) { - rc = -EINVAL; /* only add one network per call */ - goto failed0; + if (ni) { + rc = 0; + lnet_ni_lock(ni); + lnet_fill_ni_info(ni, cfg_ni, tun, stats, tun_size); + lnet_ni_unlock(ni); } - net = list_entry(net_head.next, struct lnet_net, net_list); + lnet_net_unlock(cpt); + return rc; +} + +static int lnet_add_net_common(struct lnet_net *net, + struct lnet_ioctl_config_lnd_tunables *tun) +{ + __u32 net_id; + struct lnet_ping_info *pinfo; + struct lnet_handle_md md_handle; + int rc; + struct lnet_remotenet *rnet; + int net_ni_count; + int num_acceptor_nets; lnet_net_lock(LNET_LOCK_EX); rnet = lnet_find_rnet_locked(net->net_id); lnet_net_unlock(LNET_LOCK_EX); - /* make sure that the net added doesn't invalidate the current - * configuration LNet is keeping */ - if (rnet != NULL) { + /* + * make sure that the net added doesn't invalidate the current + * configuration LNet is keeping + */ + if (rnet) { CERROR("Adding net %s will invalidate routing configuration\n", - nets); - rc = -EUSERS; - goto failed0; + libcfs_net2str(net->net_id)); + lnet_net_free(net); + return -EUSERS; } /* @@ -2120,22 +2239,24 @@ lnet_dyn_add_ni(lnet_pid_t requested_pid, struct lnet_ioctl_config_data *conf) * we should allocate enough slots to accomodate the number of NIs * which will be added. * - * We can use lnet_get_net_ni_count_locked() since the net is not - * on a public list yet, so locking is not a problem + * since ni hasn't been configured yet, use + * lnet_get_net_ni_count_pre() which checks the net_ni_added list */ - net_ni_count = lnet_get_net_ni_count_locked(net); + net_ni_count = lnet_get_net_ni_count_pre(net); rc = lnet_ping_info_setup(&pinfo, &md_handle, net_ni_count + lnet_get_ni_count(), false); - if (rc != 0) - goto failed0; - - list_del_init(&net->net_list); + if (rc < 0) { + lnet_net_free(net); + return rc; + } - if (lnd_tunables) + if (tun) memcpy(&net->net_tunables, - &lnd_tunables->lt_cmn, sizeof(lnd_tunables->lt_cmn)); + &tun->lt_cmn, sizeof(net->net_tunables)); + else + memset(&net->net_tunables, -1, sizeof(net->net_tunables)); /* * before starting this network get a count of the current TCP @@ -2145,49 +2266,77 @@ lnet_dyn_add_ni(lnet_pid_t requested_pid, struct lnet_ioctl_config_data *conf) */ num_acceptor_nets = lnet_count_acceptor_nets(); - /* - * lnd_startup_lndnet() can deallocate 'net' even if it it returns - * success, because we endded up adding interfaces to an existing - * network. So grab the net_type now - */ - net_type = LNET_NETTYP(net->net_id); + net_id = net->net_id; rc = lnet_startup_lndnet(net, - (lnd_tunables) ? &lnd_tunables->lt_tun : NULL); + (tun) ? &tun->lt_tun : NULL); if (rc < 0) - goto failed1; + goto failed; + + lnet_net_lock(LNET_LOCK_EX); + net = lnet_get_net_locked(net_id); + lnet_net_unlock(LNET_LOCK_EX); + + LASSERT(net); /* * Start the acceptor thread if this is the first network * being added that requires the thread. */ - if (net_type == SOCKLND && num_acceptor_nets == 0) - { + if (net->net_lnd->lnd_accept && num_acceptor_nets == 0) { rc = lnet_acceptor_start(); if (rc < 0) { /* shutdown the net that we just started */ CERROR("Failed to start up acceptor thread\n"); - /* - * Note that if we needed to start the acceptor - * thread, then 'net' must have been the first TCP - * network, therefore was unique, and therefore - * wasn't deallocated by lnet_startup_lndnet() - */ lnet_shutdown_lndnet(net); - goto failed1; + goto failed; } } + lnet_net_lock(LNET_LOCK_EX); + lnet_peer_net_added(net); + lnet_net_unlock(LNET_LOCK_EX); + lnet_ping_target_update(pinfo, md_handle); - mutex_unlock(&the_lnet.ln_api_mutex); return 0; -failed1: +failed: lnet_ping_md_unlink(pinfo, &md_handle); lnet_ping_info_free(pinfo); -failed0: + return rc; +} + +static int lnet_handle_legacy_ip2nets(char *ip2nets, + struct lnet_ioctl_config_lnd_tunables *tun) +{ + struct lnet_net *net; + char *nets; + int rc; + struct list_head net_head; + + INIT_LIST_HEAD(&net_head); + + rc = lnet_parse_ip2nets(&nets, ip2nets); + if (rc < 0) + return rc; + + rc = lnet_parse_networks(&net_head, nets, use_tcp_bonding); + if (rc < 0) + return rc; + + mutex_lock(&the_lnet.ln_api_mutex); + while (!list_empty(&net_head)) { + net = list_entry(net_head.next, struct lnet_net, net_list); + list_del_init(&net->net_list); + rc = lnet_add_net_common(net, tun); + if (rc < 0) + goto out; + } + +out: mutex_unlock(&the_lnet.ln_api_mutex); + while (!list_empty(&net_head)) { net = list_entry(net_head.next, struct lnet_net, net_list); list_del_init(&net->net_list); @@ -2196,12 +2345,204 @@ failed0: return rc; } +int lnet_dyn_add_ni(struct lnet_ioctl_config_ni *conf) +{ + struct lnet_net *net; + struct lnet_ni *ni; + struct lnet_ioctl_config_lnd_tunables *tun = NULL; + int rc, i; + __u32 net_id; + + /* get the tunables if they are available */ + if (conf->lic_cfg_hdr.ioc_len >= + sizeof(*conf) + sizeof(*tun)) + tun = (struct lnet_ioctl_config_lnd_tunables *) + conf->lic_bulk; + + /* handle legacy ip2nets from DLC */ + if (conf->lic_legacy_ip2nets[0] != '\0') + return lnet_handle_legacy_ip2nets(conf->lic_legacy_ip2nets, + tun); + + net_id = LNET_NIDNET(conf->lic_nid); + + net = lnet_net_alloc(net_id, NULL); + if (!net) + return -ENOMEM; + + for (i = 0; i < conf->lic_ncpts; i++) { + if (conf->lic_cpts[i] >= LNET_CPT_NUMBER) + return -EINVAL; + } + + ni = lnet_ni_alloc_w_cpt_array(net, conf->lic_cpts, conf->lic_ncpts, + conf->lic_ni_intf[0]); + if (!ni) + return -ENOMEM; + + mutex_lock(&the_lnet.ln_api_mutex); + + rc = lnet_add_net_common(net, tun); + + mutex_unlock(&the_lnet.ln_api_mutex); + + return rc; +} + +int lnet_dyn_del_ni(struct lnet_ioctl_config_ni *conf) +{ + struct lnet_net *net; + struct lnet_ni *ni; + __u32 net_id = LNET_NIDNET(conf->lic_nid); + struct lnet_ping_info *pinfo; + struct lnet_handle_md md_handle; + int rc; + int net_count; + __u32 addr; + + /* don't allow userspace to shutdown the LOLND */ + if (LNET_NETTYP(net_id) == LOLND) + return -EINVAL; + + mutex_lock(&the_lnet.ln_api_mutex); + + lnet_net_lock(0); + + net = lnet_get_net_locked(net_id); + if (!net) { + CERROR("net %s not found\n", + libcfs_net2str(net_id)); + rc = -ENOENT; + goto net_unlock; + } + + addr = LNET_NIDADDR(conf->lic_nid); + if (addr == 0) { + /* remove the entire net */ + net_count = lnet_get_net_ni_count_locked(net); + + lnet_net_unlock(0); + + /* create and link a new ping info, before removing the old one */ + rc = lnet_ping_info_setup(&pinfo, &md_handle, + lnet_get_ni_count() - net_count, + false); + if (rc != 0) + goto out; + + lnet_shutdown_lndnet(net); + + if (lnet_count_acceptor_nets() == 0) + lnet_acceptor_stop(); + + lnet_ping_target_update(pinfo, md_handle); + + goto out; + } + + ni = lnet_nid2ni_locked(conf->lic_nid, 0); + if (!ni) { + CERROR("nid %s not found \n", + libcfs_nid2str(conf->lic_nid)); + rc = -ENOENT; + goto net_unlock; + } + + net_count = lnet_get_net_ni_count_locked(net); + + lnet_net_unlock(0); + + /* create and link a new ping info, before removing the old one */ + rc = lnet_ping_info_setup(&pinfo, &md_handle, + lnet_get_ni_count() - 1, false); + if (rc != 0) + goto out; + + lnet_shutdown_lndni(ni); + + if (lnet_count_acceptor_nets() == 0) + lnet_acceptor_stop(); + + lnet_ping_target_update(pinfo, md_handle); + + /* check if the net is empty and remove it if it is */ + if (net_count == 1) + lnet_shutdown_lndnet(net); + + goto out; + +net_unlock: + lnet_net_unlock(0); +out: + mutex_unlock(&the_lnet.ln_api_mutex); + + return rc; +} + +/* + * lnet_dyn_add_net and lnet_dyn_del_net are now deprecated. + * They are only expected to be called for unique networks. + * That can be as a result of older DLC library + * calls. Multi-Rail DLC and beyond no longer uses these APIs. + */ +int +lnet_dyn_add_net(struct lnet_ioctl_config_data *conf) +{ + struct lnet_net *net; + struct list_head net_head; + int rc; + struct lnet_ioctl_config_lnd_tunables tun; + char *nets = conf->cfg_config_u.cfg_net.net_intf; + + INIT_LIST_HEAD(&net_head); + + /* Create a net/ni structures for the network string */ + rc = lnet_parse_networks(&net_head, nets, use_tcp_bonding); + if (rc <= 0) + return rc == 0 ? -EINVAL : rc; + + mutex_lock(&the_lnet.ln_api_mutex); + + if (rc > 1) { + rc = -EINVAL; /* only add one network per call */ + goto out_unlock_clean; + } + + net = list_entry(net_head.next, struct lnet_net, net_list); + list_del_init(&net->net_list); + + LASSERT(lnet_net_unique(net->net_id, &the_lnet.ln_nets, NULL)); + + memset(&tun, 0, sizeof(tun)); + + tun.lt_cmn.lct_peer_timeout = + conf->cfg_config_u.cfg_net.net_peer_timeout; + tun.lt_cmn.lct_peer_tx_credits = + conf->cfg_config_u.cfg_net.net_peer_tx_credits; + tun.lt_cmn.lct_peer_rtr_credits = + conf->cfg_config_u.cfg_net.net_peer_rtr_credits; + tun.lt_cmn.lct_max_tx_credits = + conf->cfg_config_u.cfg_net.net_max_tx_credits; + + rc = lnet_add_net_common(net, &tun); + +out_unlock_clean: + mutex_unlock(&the_lnet.ln_api_mutex); + while (!list_empty(&net_head)) { + /* net_head list is empty in success case */ + net = list_entry(net_head.next, struct lnet_net, net_list); + list_del_init(&net->net_list); + lnet_net_free(net); + } + return rc; +} + int -lnet_dyn_del_ni(__u32 net_id) +lnet_dyn_del_net(__u32 net_id) { struct lnet_net *net; struct lnet_ping_info *pinfo; - lnet_handle_md_t md_handle; + struct lnet_handle_md md_handle; int rc; int net_ni_count; @@ -2215,6 +2556,7 @@ lnet_dyn_del_ni(__u32 net_id) net = lnet_get_net_locked(net_id); if (net == NULL) { + lnet_net_unlock(0); rc = -EINVAL; goto out; } @@ -2242,6 +2584,16 @@ out: return rc; } +void lnet_incr_dlc_seq(void) +{ + atomic_inc(&lnet_dlc_seq_no); +} + +__u32 lnet_get_dlc_seq_locked(void) +{ + return atomic_read(&lnet_dlc_seq_no); +} + /** * LNet ioctl handler. * @@ -2251,8 +2603,8 @@ LNetCtl(unsigned int cmd, void *arg) { struct libcfs_ioctl_data *data = arg; struct lnet_ioctl_config_data *config; - lnet_process_id_t id = {0}; - lnet_ni_t *ni; + struct lnet_process_id id = {0}; + struct lnet_ni *ni; int rc; BUILD_BUG_ON(sizeof(struct lnet_ioctl_net_config) + @@ -2305,13 +2657,42 @@ LNetCtl(unsigned int cmd, void *arg) if (config->cfg_hdr.ioc_len < sizeof(*config)) return -EINVAL; - return lnet_get_route(config->cfg_count, - &config->cfg_net, - &config->cfg_config_u.cfg_route.rtr_hop, - &config->cfg_nid, - &config->cfg_config_u.cfg_route.rtr_flags, - &config->cfg_config_u.cfg_route. + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_get_route(config->cfg_count, + &config->cfg_net, + &config->cfg_config_u.cfg_route.rtr_hop, + &config->cfg_nid, + &config->cfg_config_u.cfg_route.rtr_flags, + &config->cfg_config_u.cfg_route. rtr_priority); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; + + case IOC_LIBCFS_GET_LOCAL_NI: { + struct lnet_ioctl_config_ni *cfg_ni; + struct lnet_ioctl_config_lnd_tunables *tun = NULL; + struct lnet_ioctl_element_stats *stats; + __u32 tun_size; + + cfg_ni = arg; + /* get the tunables if they are available */ + if (cfg_ni->lic_cfg_hdr.ioc_len < + sizeof(*cfg_ni) + sizeof(*stats)+ sizeof(*tun)) + return -EINVAL; + + stats = (struct lnet_ioctl_element_stats *) + cfg_ni->lic_bulk; + tun = (struct lnet_ioctl_config_lnd_tunables *) + (cfg_ni->lic_bulk + sizeof(*stats)); + + tun_size = cfg_ni->lic_cfg_hdr.ioc_len - sizeof(*cfg_ni) - + sizeof(*stats); + + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_get_ni_config(cfg_ni, tun, stats, tun_size); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; + } case IOC_LIBCFS_GET_NET: { size_t total = sizeof(*config) + @@ -2321,7 +2702,10 @@ LNetCtl(unsigned int cmd, void *arg) if (config->cfg_hdr.ioc_len < total) return -EINVAL; - return lnet_get_net_config(config); + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_get_net_config(config); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; } case IOC_LIBCFS_GET_LNET_STATS: @@ -2331,7 +2715,9 @@ LNetCtl(unsigned int cmd, void *arg) if (lnet_stats->st_hdr.ioc_len < sizeof(*lnet_stats)) return -EINVAL; + mutex_lock(&the_lnet.ln_api_mutex); lnet_counters_get(&lnet_stats->st_cntrs); + mutex_unlock(&the_lnet.ln_api_mutex); return 0; } @@ -2367,6 +2753,26 @@ LNetCtl(unsigned int cmd, void *arg) mutex_unlock(&the_lnet.ln_api_mutex); return rc; + case IOC_LIBCFS_SET_NUMA_RANGE: { + struct lnet_ioctl_numa_range *numa; + numa = arg; + if (numa->nr_hdr.ioc_len != sizeof(*numa)) + return -EINVAL; + mutex_lock(&the_lnet.ln_api_mutex); + lnet_numa_range = numa->nr_range; + mutex_unlock(&the_lnet.ln_api_mutex); + return 0; + } + + case IOC_LIBCFS_GET_NUMA_RANGE: { + struct lnet_ioctl_numa_range *numa; + numa = arg; + if (numa->nr_hdr.ioc_len != sizeof(*numa)) + return -EINVAL; + numa->nr_range = lnet_numa_range; + return 0; + } + case IOC_LIBCFS_GET_BUF: { struct lnet_ioctl_pool_cfg *pool_cfg; size_t total = sizeof(*config) + sizeof(*pool_cfg); @@ -2377,7 +2783,38 @@ LNetCtl(unsigned int cmd, void *arg) return -EINVAL; pool_cfg = (struct lnet_ioctl_pool_cfg *)config->cfg_bulk; - return lnet_get_rtr_pool_cfg(config->cfg_count, pool_cfg); + + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_get_rtr_pool_cfg(config->cfg_count, pool_cfg); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; + } + + case IOC_LIBCFS_ADD_PEER_NI: { + struct lnet_ioctl_peer_cfg *cfg = arg; + + if (cfg->prcfg_hdr.ioc_len < sizeof(*cfg)) + return -EINVAL; + + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_add_peer_ni_to_peer(cfg->prcfg_prim_nid, + cfg->prcfg_cfg_nid, + cfg->prcfg_mr); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; + } + + case IOC_LIBCFS_DEL_PEER_NI: { + struct lnet_ioctl_peer_cfg *cfg = arg; + + if (cfg->prcfg_hdr.ioc_len < sizeof(*cfg)) + return -EINVAL; + + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_del_peer_ni_from_peer(cfg->prcfg_prim_nid, + cfg->prcfg_cfg_nid); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; } case IOC_LIBCFS_GET_PEER_INFO: { @@ -2386,7 +2823,8 @@ LNetCtl(unsigned int cmd, void *arg) if (peer_info->pr_hdr.ioc_len < sizeof(*peer_info)) return -EINVAL; - return lnet_get_peer_info( + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_get_peer_ni_info( peer_info->pr_count, &peer_info->pr_nid, peer_info->pr_lnd_u.pr_peer_credits.cr_aliveness, @@ -2395,8 +2833,31 @@ LNetCtl(unsigned int cmd, void *arg) &peer_info->pr_lnd_u.pr_peer_credits.cr_ni_peer_tx_credits, &peer_info->pr_lnd_u.pr_peer_credits.cr_peer_tx_credits, &peer_info->pr_lnd_u.pr_peer_credits.cr_peer_rtr_credits, - &peer_info->pr_lnd_u.pr_peer_credits.cr_peer_min_rtr_credits, + &peer_info->pr_lnd_u.pr_peer_credits.cr_peer_min_tx_credits, &peer_info->pr_lnd_u.pr_peer_credits.cr_peer_tx_qnob); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; + } + + case IOC_LIBCFS_GET_PEER_NI: { + struct lnet_ioctl_peer_cfg *cfg = arg; + struct lnet_peer_ni_credit_info __user *lpni_cri; + struct lnet_ioctl_element_stats __user *lpni_stats; + size_t usr_size = sizeof(*lpni_cri) + sizeof(*lpni_stats); + + if ((cfg->prcfg_hdr.ioc_len != sizeof(*cfg)) || + (cfg->prcfg_size != usr_size)) + return -EINVAL; + + lpni_cri = cfg->prcfg_bulk; + lpni_stats = cfg->prcfg_bulk + sizeof(*lpni_cri); + + mutex_lock(&the_lnet.ln_api_mutex); + rc = lnet_get_peer_info(cfg->prcfg_count, &cfg->prcfg_prim_nid, + &cfg->prcfg_cfg_nid, &cfg->prcfg_mr, + lpni_cri, lpni_stats); + mutex_unlock(&the_lnet.ln_api_mutex); + return rc; } case IOC_LIBCFS_NOTIFY_ROUTER: { @@ -2443,14 +2904,15 @@ LNetCtl(unsigned int cmd, void *arg) timeout = msecs_to_jiffies(data->ioc_u32[1]); rc = lnet_ping(id, timeout, data->ioc_pbuf1, - data->ioc_plen1 / sizeof(lnet_process_id_t)); + data->ioc_plen1 / sizeof(struct lnet_process_id)); if (rc < 0) return rc; data->ioc_count = rc; return 0; } + default: - ni = lnet_net2ni(data->ioc_net); + ni = lnet_net2ni_addref(data->ioc_net); if (ni == NULL) return -EINVAL; @@ -2459,31 +2921,61 @@ LNetCtl(unsigned int cmd, void *arg) else rc = ni->ni_net->net_lnd->lnd_ctl(ni, cmd, arg); + lnet_ni_decref(ni); return rc; } /* not reached */ } EXPORT_SYMBOL(LNetCtl); -void LNetDebugPeer(lnet_process_id_t id) +void LNetDebugPeer(struct lnet_process_id id) { lnet_debug_peer(id.nid); } EXPORT_SYMBOL(LNetDebugPeer); /** - * Retrieve the lnet_process_id_t ID of LNet interface at \a index. Note that - * all interfaces share a same PID, as requested by LNetNIInit(). + * Determine if the specified peer \a nid is on the local node. + * + * \param nid peer nid to check + * + * \retval true If peer NID is on the local node. + * \retval false If peer NID is not on the local node. + */ +bool LNetIsPeerLocal(lnet_nid_t nid) +{ + struct lnet_net *net; + struct lnet_ni *ni; + int cpt; + + cpt = lnet_net_lock_current(); + list_for_each_entry(net, &the_lnet.ln_nets, net_list) { + list_for_each_entry(ni, &net->net_ni_list, ni_netlist) { + if (ni->ni_nid == nid) { + lnet_net_unlock(cpt); + return true; + } + } + } + lnet_net_unlock(cpt); + + return false; +} +EXPORT_SYMBOL(LNetIsPeerLocal); + +/** + * Retrieve the struct lnet_process_id ID of LNet interface at \a index. + * Note that all interfaces share a same PID, as requested by LNetNIInit(). * * \param index Index of the interface to look up. * \param id On successful return, this location will hold the - * lnet_process_id_t ID of the interface. + * struct lnet_process_id ID of the interface. * * \retval 0 If an interface exists at \a index. * \retval -ENOENT If no interface has been found. */ int -LNetGetId(unsigned int index, lnet_process_id_t *id) +LNetGetId(unsigned int index, struct lnet_process_id *id) { struct lnet_ni *ni; struct lnet_net *net; @@ -2511,31 +3003,20 @@ LNetGetId(unsigned int index, lnet_process_id_t *id) } EXPORT_SYMBOL(LNetGetId); -/** - * Print a string representation of handle \a h into buffer \a str of - * \a len bytes. - */ -void -LNetSnprintHandle(char *str, int len, lnet_handle_any_t h) -{ - snprintf(str, len, "%#llx", h.cookie); -} -EXPORT_SYMBOL(LNetSnprintHandle); - -static int lnet_ping(lnet_process_id_t id, signed long timeout, - lnet_process_id_t __user *ids, int n_ids) +static int lnet_ping(struct lnet_process_id id, signed long timeout, + struct lnet_process_id __user *ids, int n_ids) { - lnet_handle_eq_t eqh; - lnet_handle_md_t mdh; - lnet_event_t event; - lnet_md_t md = { NULL }; + struct lnet_handle_eq eqh; + struct lnet_handle_md mdh; + struct lnet_event event; + struct lnet_md md = { NULL }; int which; int unlinked = 0; int replied = 0; const signed long a_long_time = msecs_to_jiffies(60 * MSEC_PER_SEC); int infosz; - struct lnet_ping_info *info; - lnet_process_id_t tmpid; + struct lnet_ping_info *info; + struct lnet_process_id tmpid; int i; int nob; int rc;