Whamcloud - gitweb
LU-13004 ptlrpc: Allow BULK_BUF_KIOV to accept a kvec
[fs/lustre-release.git] / lustre / ptlrpc / ptlrpc_internal.h
index 42648cf..19d0487 100644 (file)
@@ -1,23 +1,33 @@
-/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*-
- * vim:expandtab:shiftwidth=8:tabstop=8:
+/*
+ * GPL HEADER START
  *
- *  Copyright (C) 2003 Cluster File Systems, Inc.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
- *   This file is part of Lustre, http://www.lustre.org.
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 only,
+ * as published by the Free Software Foundation.
  *
- *   Lustre is free software; you can redistribute it and/or
- *   modify it under the terms of version 2 of the GNU General Public
- *   License as published by the Free Software Foundation.
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License version 2 for more details (a copy is included
+ * in the LICENSE file that accompanied this code).
  *
- *   Lustre is distributed in the hope that it will be useful,
- *   but WITHOUT ANY WARRANTY; without even the implied warranty of
- *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *   GNU General Public License for more details.
+ * You should have received a copy of the GNU General Public License
+ * version 2 along with this program; If not, see
+ * http://www.gnu.org/licenses/gpl-2.0.html
  *
- *   You should have received a copy of the GNU General Public License
- *   along with Lustre; if not, write to the Free Software
- *   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ * GPL HEADER END
+ */
+/*
+ * Copyright (c) 2003, 2010, Oracle and/or its affiliates. All rights reserved.
+ * Use is subject to license terms.
  *
+ * Copyright (c) 2011, 2017, Intel Corporation.
+ */
+/*
+ * This file is part of Lustre, http://www.lustre.org/
+ * Lustre is a trademark of Sun Microsystems, Inc.
  */
 
 /* Intramodule declarations for ptlrpc. */
@@ -31,97 +41,378 @@ struct ldlm_namespace;
 struct obd_import;
 struct ldlm_res_id;
 struct ptlrpc_request_set;
+extern int test_req_buffer_pressure;
+extern struct list_head ptlrpc_all_services;
+extern struct mutex ptlrpc_all_services_mutex;
+extern struct ptlrpc_nrs_pol_conf nrs_conf_fifo;
+
+#ifdef HAVE_SERVER_SUPPORT
+extern struct ptlrpc_nrs_pol_conf nrs_conf_crrn;
+extern struct ptlrpc_nrs_pol_conf nrs_conf_orr;
+extern struct ptlrpc_nrs_pol_conf nrs_conf_trr;
+extern struct ptlrpc_nrs_pol_conf nrs_conf_tbf;
+extern struct ptlrpc_nrs_pol_conf nrs_conf_delay;
+#endif /* HAVE_SERVER_SUPPORT */
+
+/**
+ * \addtogoup nrs
+ * @{
+ */
+extern struct nrs_core nrs_core;
+
+extern struct mutex ptlrpcd_mutex;
+extern struct mutex pinger_mutex;
 
-void ptlrpc_daemonize(void);
+int ptlrpc_start_thread(struct ptlrpc_service_part *svcpt, int wait);
+/* ptlrpcd.c */
+int ptlrpcd_start(struct ptlrpcd_ctl *pc);
+
+/* client.c */
+void ptlrpc_at_adj_net_latency(struct ptlrpc_request *req,
+                              unsigned int service_time);
+struct ptlrpc_bulk_desc *ptlrpc_new_bulk(unsigned npages, unsigned max_brw,
+                                        enum ptlrpc_bulk_op_type type,
+                                        unsigned portal,
+                                        const struct ptlrpc_bulk_frag_ops
+                                               *ops);
+int ptlrpc_request_cache_init(void);
+void ptlrpc_request_cache_fini(void);
+struct ptlrpc_request *ptlrpc_request_cache_alloc(gfp_t flags);
+void ptlrpc_request_cache_free(struct ptlrpc_request *req);
+void ptlrpc_init_xid(void);
+void ptlrpc_set_add_new_req(struct ptlrpcd_ctl *pc,
+                           struct ptlrpc_request *req);
+int ptlrpc_expired_set(void *data);
+time64_t ptlrpc_set_next_timeout(struct ptlrpc_request_set *);
+void ptlrpc_resend_req(struct ptlrpc_request *request);
+void ptlrpc_set_bulk_mbits(struct ptlrpc_request *req);
+void ptlrpc_assign_next_xid_nolock(struct ptlrpc_request *req);
+__u64 ptlrpc_known_replied_xid(struct obd_import *imp);
+void ptlrpc_add_unreplied(struct ptlrpc_request *req);
+
+/* events.c */
+int ptlrpc_init_portals(void);
+void ptlrpc_exit_portals(void);
 
 void ptlrpc_request_handle_notconn(struct ptlrpc_request *);
 void lustre_assert_wire_constants(void);
 int ptlrpc_import_in_recovery(struct obd_import *imp);
-int ptlrpc_set_import_discon(struct obd_import *imp);
+int ptlrpc_set_import_discon(struct obd_import *imp, __u32 conn_cnt,
+                            bool invalid);
 void ptlrpc_handle_failed_import(struct obd_import *imp);
 int ptlrpc_replay_next(struct obd_import *imp, int *inflight);
 void ptlrpc_initiate_recovery(struct obd_import *imp);
 
+int lustre_unpack_req_ptlrpc_body(struct ptlrpc_request *req, int offset);
+int lustre_unpack_rep_ptlrpc_body(struct ptlrpc_request *req, int offset);
+
+int ptlrpc_sysfs_register_service(struct kset *parent,
+                                 struct ptlrpc_service *svc);
+void ptlrpc_sysfs_unregister_service(struct ptlrpc_service *svc);
 
-#ifdef __KERNEL__
-void ptlrpc_lprocfs_register_service(struct proc_dir_entry *proc_entry,
-                                     struct ptlrpc_service *svc);
+void ptlrpc_ldebugfs_register_service(struct dentry *debugfs_entry,
+                                     struct ptlrpc_service *svc);
+#ifdef CONFIG_PROC_FS
 void ptlrpc_lprocfs_unregister_service(struct ptlrpc_service *svc);
-void ptlrpc_lprocfs_rpc_sent(struct ptlrpc_request *req);
+void ptlrpc_lprocfs_rpc_sent(struct ptlrpc_request *req, long amount);
 void ptlrpc_lprocfs_do_request_stat (struct ptlrpc_request *req,
                                      long q_usec, long work_usec);
 #else
-#define ptlrpc_lprocfs_register_service(params...) do{}while(0)
 #define ptlrpc_lprocfs_unregister_service(params...) do{}while(0)
 #define ptlrpc_lprocfs_rpc_sent(params...) do{}while(0)
 #define ptlrpc_lprocfs_do_request_stat(params...) do{}while(0)
-#endif /* __KERNEL__ */
+#endif /* CONFIG_PROC_FS */
 
-/* recovd_thread.c */
-int llog_init_commit_master(void);
-int llog_cleanup_commit_master(int force);
-
-static inline int opcode_offset(__u32 opc) {
-        if (opc < OST_LAST_OPC) {
-                 /* OST opcode */
-                return (opc - OST_FIRST_OPC);
-        } else if (opc < MDS_LAST_OPC) {
-                /* MDS opcode */
-                return (opc - MDS_FIRST_OPC +
-                        (OST_LAST_OPC - OST_FIRST_OPC));
-        } else if (opc < LDLM_LAST_OPC) {
-                /* LDLM Opcode */
-                return (opc - LDLM_FIRST_OPC +
-                        (MDS_LAST_OPC - MDS_FIRST_OPC) +
-                        (OST_LAST_OPC - OST_FIRST_OPC));
-        } else if (opc < PTLBD_LAST_OPC) {
-                /* Portals Block Device */
-                return (opc - PTLBD_FIRST_OPC +
-                        (LDLM_LAST_OPC - LDLM_FIRST_OPC) +
-                        (MDS_LAST_OPC - MDS_FIRST_OPC) +
-                        (OST_LAST_OPC - OST_FIRST_OPC));
-        } else if (opc < OBD_LAST_OPC) {
-                /* OBD Ping */
-                return (opc - OBD_FIRST_OPC +
-                        (PTLBD_LAST_OPC - PTLBD_FIRST_OPC) +
-                        (LDLM_LAST_OPC - LDLM_FIRST_OPC) +
-                        (MDS_LAST_OPC - MDS_FIRST_OPC) +
-                        (OST_LAST_OPC - OST_FIRST_OPC));
-        } else {
-                /* Unknown Opcode */
-                return -1;
-        }
-}
-
-#define LUSTRE_MAX_OPCODES ((PTLBD_LAST_OPC - PTLBD_FIRST_OPC) + \
-                            (LDLM_LAST_OPC - LDLM_FIRST_OPC)   + \
-                            (MDS_LAST_OPC - MDS_FIRST_OPC)     + \
-                            (OST_LAST_OPC - OST_FIRST_OPC)     + \
-                            (OBD_LAST_OPC - OBD_FIRST_OPC))
-
-enum {
-        PTLRPC_REQWAIT_CNTR = 0,
-        PTLRPC_REQQDEPTH_CNTR,
-        PTLRPC_REQACTIVE_CNTR,
-        PTLRPC_LAST_CNTR
+/* NRS */
+
+/**
+ * NRS core object.
+ *
+ * Holds NRS core fields.
+ */
+struct nrs_core {
+       /**
+        * Protects nrs_core::nrs_policies, serializes external policy
+        * registration/unregistration, and NRS core lprocfs operations.
+        */
+       struct mutex nrs_mutex;
+       /**
+        * List of all policy descriptors registered with NRS core; protected
+        * by nrs_core::nrs_mutex.
+        */
+       struct list_head nrs_policies;
 };
 
-int ptlrpc_expire_one_request(struct ptlrpc_request *req);
+int ptlrpc_service_nrs_setup(struct ptlrpc_service *svc);
+void ptlrpc_service_nrs_cleanup(struct ptlrpc_service *svc);
+
+void ptlrpc_nrs_req_initialize(struct ptlrpc_service_part *svcpt,
+                              struct ptlrpc_request *req, bool hp);
+void ptlrpc_nrs_req_finalize(struct ptlrpc_request *req);
+void ptlrpc_nrs_req_stop_nolock(struct ptlrpc_request *req);
+void ptlrpc_nrs_req_add(struct ptlrpc_service_part *svcpt,
+                       struct ptlrpc_request *req, bool hp);
+
+struct ptlrpc_request *
+ptlrpc_nrs_req_get_nolock0(struct ptlrpc_service_part *svcpt, bool hp,
+                          bool peek, bool force);
+
+static inline struct ptlrpc_request *
+ptlrpc_nrs_req_get_nolock(struct ptlrpc_service_part *svcpt, bool hp,
+                         bool force)
+{
+       return ptlrpc_nrs_req_get_nolock0(svcpt, hp, false, force);
+}
+
+static inline struct ptlrpc_request *
+ptlrpc_nrs_req_peek_nolock(struct ptlrpc_service_part *svcpt, bool hp)
+{
+       return ptlrpc_nrs_req_get_nolock0(svcpt, hp, true, false);
+}
+
+void ptlrpc_nrs_req_del_nolock(struct ptlrpc_request *req);
+bool ptlrpc_nrs_req_pending_nolock(struct ptlrpc_service_part *svcpt, bool hp);
+bool ptlrpc_nrs_req_throttling_nolock(struct ptlrpc_service_part *svcpt,
+                                     bool hp);
+
+int ptlrpc_nrs_policy_control(const struct ptlrpc_service *svc,
+                             enum ptlrpc_nrs_queue_type queue, char *name,
+                             enum ptlrpc_nrs_ctl opc, bool single, void *arg);
 
-#if !defined(__KERNEL__) && defined(CRAY_PORTALS)
-/* forward ref in events.c */
-static void cray_portals_callback(ptl_event_t *ev);
-#endif
+int ptlrpc_nrs_init(void);
+void ptlrpc_nrs_fini(void);
+
+static inline bool nrs_svcpt_has_hp(const struct ptlrpc_service_part *svcpt)
+{
+       return svcpt->scp_nrs_hp != NULL;
+}
+
+static inline bool nrs_svc_has_hp(const struct ptlrpc_service *svc)
+{
+       /**
+        * If the first service partition has an HP NRS head, all service
+        * partitions will.
+        */
+       return nrs_svcpt_has_hp(svc->srv_parts[0]);
+}
+
+static inline
+struct ptlrpc_nrs *nrs_svcpt2nrs(struct ptlrpc_service_part *svcpt, bool hp)
+{
+       LASSERT(ergo(hp, nrs_svcpt_has_hp(svcpt)));
+       return hp ? svcpt->scp_nrs_hp : &svcpt->scp_nrs_reg;
+}
+
+static inline int nrs_pol2cptid(const struct ptlrpc_nrs_policy *policy)
+{
+       return policy->pol_nrs->nrs_svcpt->scp_cpt;
+}
+
+static inline
+struct ptlrpc_service *nrs_pol2svc(struct ptlrpc_nrs_policy *policy)
+{
+       return policy->pol_nrs->nrs_svcpt->scp_service;
+}
+
+static inline
+struct ptlrpc_service_part *nrs_pol2svcpt(struct ptlrpc_nrs_policy *policy)
+{
+       return policy->pol_nrs->nrs_svcpt;
+}
+
+static inline
+struct cfs_cpt_table *nrs_pol2cptab(struct ptlrpc_nrs_policy *policy)
+{
+       return nrs_pol2svc(policy)->srv_cptable;
+}
+
+static inline struct ptlrpc_nrs_resource *
+nrs_request_resource(struct ptlrpc_nrs_request *nrq)
+{
+       LASSERT(nrq->nr_initialized);
+       LASSERT(!nrq->nr_finalized);
+
+       return nrq->nr_res_ptrs[nrq->nr_res_idx];
+}
+
+static inline
+struct ptlrpc_nrs_policy *nrs_request_policy(struct ptlrpc_nrs_request *nrq)
+{
+       return nrs_request_resource(nrq)->res_policy;
+}
+
+#define NRS_LPROCFS_QUANTUM_NAME_REG   "reg_quantum:"
+#define NRS_LPROCFS_QUANTUM_NAME_HP    "hp_quantum:"
+
+/**
+ * the maximum size of nrs_crrn_client::cc_quantum and nrs_orr_data::od_quantum.
+ */
+#define LPROCFS_NRS_QUANTUM_MAX                65535
+
+/**
+ * Max valid command string is the size of the labels, plus "65535" twice, plus
+ * a separating space character.
+ */
+#define LPROCFS_NRS_WR_QUANTUM_MAX_CMD                                        \
+ sizeof(NRS_LPROCFS_QUANTUM_NAME_REG __stringify(LPROCFS_NRS_QUANTUM_MAX) " "  \
+        NRS_LPROCFS_QUANTUM_NAME_HP __stringify(LPROCFS_NRS_QUANTUM_MAX))
+
+/* recovd_thread.c */
+
+int ptlrpc_expire_one_request(struct ptlrpc_request *req, int async_unlink);
 
 /* pers.c */
-void ptlrpc_fill_bulk_md(ptl_md_t *md, struct ptlrpc_bulk_desc *desc);
-void ptlrpc_add_bulk_page(struct ptlrpc_bulk_desc *desc, struct page *page, 
-                          int pageoffset, int len);
+void ptlrpc_fill_bulk_md(struct lnet_md *md, struct ptlrpc_bulk_desc *desc,
+                        int mdcnt);
+
+/* pack_generic.c */
+struct ptlrpc_reply_state *
+lustre_get_emerg_rs(struct ptlrpc_service_part *svcpt);
+void lustre_put_emerg_rs(struct ptlrpc_reply_state *rs);
 
 /* pinger.c */
 int ptlrpc_start_pinger(void);
 int ptlrpc_stop_pinger(void);
 void ptlrpc_pinger_sending_on_import(struct obd_import *imp);
+void ptlrpc_pinger_commit_expected(struct obd_import *imp);
 void ptlrpc_pinger_wake_up(void);
+void ptlrpc_ping_import_soon(struct obd_import *imp);
+int ping_evictor_wake(struct obd_export *exp);
+
+/* sec_null.c */
+int  sptlrpc_null_init(void);
+void sptlrpc_null_fini(void);
+
+/* sec_plain.c */
+int  sptlrpc_plain_init(void);
+void sptlrpc_plain_fini(void);
+
+/* sec_bulk.c */
+int  sptlrpc_enc_pool_init(void);
+void sptlrpc_enc_pool_fini(void);
+int sptlrpc_proc_enc_pool_seq_show(struct seq_file *m, void *v);
+
+/* sec_lproc.c */
+int  sptlrpc_lproc_init(void);
+void sptlrpc_lproc_fini(void);
+
+/* sec_gc.c */
+int sptlrpc_gc_init(void);
+void sptlrpc_gc_fini(void);
+
+/* sec_config.c */
+void sptlrpc_conf_choose_flavor(enum lustre_sec_part from,
+                                enum lustre_sec_part to,
+                                struct obd_uuid *target,
+                                lnet_nid_t nid,
+                                struct sptlrpc_flavor *sf);
+int  sptlrpc_conf_init(void);
+void sptlrpc_conf_fini(void);
+
+/* sec.c */
+int  sptlrpc_init(void);
+void sptlrpc_fini(void);
+
+/* layout.c */
+__u32 __req_capsule_offset(const struct req_capsule *pill,
+                          const struct req_msg_field *field,
+                          enum req_location loc);
+
+static inline bool ptlrpc_recoverable_error(int rc)
+{
+       return (rc == -ENOTCONN || rc == -ENODEV);
+}
+
+#ifdef HAVE_SERVER_SUPPORT
+int tgt_mod_init(void);
+void tgt_mod_exit(void);
+int nodemap_mod_init(void);
+void nodemap_mod_exit(void);
+#else /* HAVE_SERVER_SUPPORT */
+static inline int tgt_mod_init(void)
+{
+       return 0;
+}
+
+static inline void tgt_mod_exit(void)
+{
+       return;
+}
+
+static inline int nodemap_mod_init(void)
+{
+       return 0;
+}
+
+static inline void nodemap_mod_exit(void)
+{
+       return;
+}
+#endif /* !HAVE_SERVER_SUPPORT */
+
+static inline void ptlrpc_reqset_put(struct ptlrpc_request_set *set)
+{
+       if (atomic_dec_and_test(&set->set_refcount))
+               OBD_FREE_PTR(set);
+}
+
+/** initialise ptlrpc common fields */
+static inline void ptlrpc_req_comm_init(struct ptlrpc_request *req)
+{
+       spin_lock_init(&req->rq_lock);
+       spin_lock_init(&req->rq_early_free_lock);
+       atomic_set(&req->rq_refcount, 1);
+       INIT_LIST_HEAD(&req->rq_list);
+       INIT_LIST_HEAD(&req->rq_replay_list);
+}
+
+/** initialise client side ptlrpc request */
+static inline void ptlrpc_cli_req_init(struct ptlrpc_request *req)
+{
+       struct ptlrpc_cli_req *cr = &req->rq_cli;
+
+       ptlrpc_req_comm_init(req);
+
+       req->rq_receiving_reply = 0;
+       req->rq_req_unlinked = req->rq_reply_unlinked = 1;
+
+       INIT_LIST_HEAD(&cr->cr_set_chain);
+       INIT_LIST_HEAD(&cr->cr_ctx_chain);
+       INIT_LIST_HEAD(&cr->cr_unreplied_list);
+       init_waitqueue_head(&cr->cr_reply_waitq);
+       init_waitqueue_head(&cr->cr_set_waitq);
+}
+
+/** initialise server side ptlrpc request */
+static inline void ptlrpc_srv_req_init(struct ptlrpc_request *req)
+{
+       struct ptlrpc_srv_req *sr = &req->rq_srv;
+
+       ptlrpc_req_comm_init(req);
+       req->rq_srv_req = 1;
+       INIT_LIST_HEAD(&sr->sr_exp_list);
+       INIT_LIST_HEAD(&sr->sr_timed_list);
+       INIT_LIST_HEAD(&sr->sr_hist_list);
+}
+
+static inline bool ptlrpc_req_is_connect(struct ptlrpc_request *req)
+{
+       if (lustre_msg_get_opc(req->rq_reqmsg) == MDS_CONNECT ||
+           lustre_msg_get_opc(req->rq_reqmsg) == OST_CONNECT ||
+           lustre_msg_get_opc(req->rq_reqmsg) == MGS_CONNECT)
+               return true;
+       else
+               return false;
+}
+
+static inline bool ptlrpc_req_is_disconnect(struct ptlrpc_request *req)
+{
+       if (lustre_msg_get_opc(req->rq_reqmsg) == MDS_DISCONNECT ||
+           lustre_msg_get_opc(req->rq_reqmsg) == OST_DISCONNECT ||
+           lustre_msg_get_opc(req->rq_reqmsg) == MGS_DISCONNECT)
+               return true;
+       else
+               return false;
+}
 
 #endif /* PTLRPC_INTERNAL_H */