* Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
- * Copyright (c) 2011, 2013, Intel Corporation.
+ * Copyright (c) 2011, 2014, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
#ifndef __OBD_H
#define __OBD_H
-#ifndef __KERNEL__
-# error "userspace should not include <obd.h>"
-#endif
-#if defined(__linux__)
-#include <linux/obd.h>
-#elif defined(__APPLE__)
-#include <darwin/obd.h>
-#elif defined(__WINNT__)
-#include <winnt/obd.h>
-#else
-#error Unsupported operating system.
-#endif
+#include <linux/spinlock.h>
#include <lustre/lustre_idl.h>
#include <lustre_lib.h>
#include <lustre_export.h>
#include <lustre_fid.h>
#include <lustre_fld.h>
+#include <lustre_handles.h>
+#include <lustre_intent.h>
#include <lustre_capa.h>
#include <lvfs.h>
{
}
+/* If we are unable to get the maximum object size from the OST in
+ * ocd_maxbytes using OBD_CONNECT_MAXBYTES, then we fall back to using
+ * the old maximum object size from ext3. */
+#define LUSTRE_EXT3_STRIPE_MAXBYTES 0x1fffffff000ULL
+
struct lov_stripe_md {
atomic_t lsm_refc;
spinlock_t lsm_lock;
/* maximum possible file size, might change as OSTs status changes,
* e.g. disconnected, deactivated */
- __u64 lsm_maxbytes;
- struct {
- /* Public members. */
- struct ost_id lw_object_oi; /* lov object id/seq */
-
- /* LOV-private members start here -- only for use in lov/. */
- __u32 lw_magic;
- __u32 lw_stripe_size; /* size of the stripe */
- __u32 lw_pattern; /* striping pattern (RAID0, RAID1) */
- __u16 lw_stripe_count; /* number of objects being striped over */
- __u16 lw_layout_gen; /* generation of the layout */
- char lw_pool_name[LOV_MAXPOOLNAME]; /* pool name */
- } lsm_wire;
-
- struct lov_oinfo *lsm_oinfo[0];
+ loff_t lsm_maxbytes;
+ struct ost_id lsm_oi;
+ __u32 lsm_magic;
+ __u32 lsm_stripe_size;
+ __u32 lsm_pattern; /* RAID0, RAID1, released, ... */
+ __u16 lsm_stripe_count;
+ __u16 lsm_layout_gen;
+ char lsm_pool_name[LOV_MAXPOOLNAME + 1];
+ struct lov_oinfo *lsm_oinfo[0];
};
-#define lsm_oi lsm_wire.lw_object_oi
-#define lsm_magic lsm_wire.lw_magic
-#define lsm_layout_gen lsm_wire.lw_layout_gen
-#define lsm_stripe_size lsm_wire.lw_stripe_size
-#define lsm_pattern lsm_wire.lw_pattern
-#define lsm_stripe_count lsm_wire.lw_stripe_count
-#define lsm_pool_name lsm_wire.lw_pool_name
-
static inline bool lsm_is_released(struct lov_stripe_md *lsm)
{
return !!(lsm->lsm_pattern & LOV_PATTERN_F_RELEASED);
- while setattr, the flags used for distinguish punch operation
*/
__u64 oi_flags;
- /* Lock handle specific for every OSC lock. */
- struct lustre_handle *oi_lockh;
/* lsm data specific for every OSC. */
struct lov_stripe_md *oi_md;
/* obdo data specific for every OSC, if needed at all. */
/* statfs data specific for every OSC, if needed at all. */
struct obd_statfs *oi_osfs;
/* An update callback which is called to update some data on upper
- * level. E.g. it is used for update lsm->lsm_oinfo at every recieved
+ * level. E.g. it is used for update lsm->lsm_oinfo at every received
* request in osc level for enqueue requests. It is also possible to
* update some caller data from LOV layer if needed. */
obd_enqueue_update_f oi_cb_up;
/* oss capability, its type is obd_capa in client to avoid copy.
* in contrary its type is lustre_capa in OSS. */
void *oi_capa;
- /* transfer jobid from ost_sync() to filter_sync()... */
- char *oi_jobid;
};
struct obd_type {
- struct list_head typ_chain;
- struct obd_ops *typ_dt_ops;
- struct md_ops *typ_md_ops;
- struct proc_dir_entry *typ_procroot;
- struct proc_dir_entry *typ_procsym;
- __u32 typ_sym_filter;
- char *typ_name;
- int typ_refcnt;
- struct lu_device_type *typ_lu;
- spinlock_t obd_type_lock;
+ struct list_head typ_chain;
+ struct obd_ops *typ_dt_ops;
+ struct md_ops *typ_md_ops;
+ struct proc_dir_entry *typ_procroot;
+ struct proc_dir_entry *typ_procsym;
+ __u32 typ_sym_filter;
+ char *typ_name;
+ int typ_refcnt;
+ struct lu_device_type *typ_lu;
+ spinlock_t obd_type_lock;
};
struct brw_page {
- obd_off off;
- struct page *pg;
- int count;
- obd_flag flag;
-};
-
-/* llog contexts */
-enum llog_ctxt_id {
- LLOG_CONFIG_ORIG_CTXT = 0,
- LLOG_CONFIG_REPL_CTXT,
- LLOG_MDS_OST_ORIG_CTXT,
- LLOG_MDS_OST_REPL_CTXT,
- LLOG_SIZE_ORIG_CTXT,
- LLOG_SIZE_REPL_CTXT,
- LLOG_RD1_ORIG_CTXT,
- LLOG_RD1_REPL_CTXT,
- LLOG_TEST_ORIG_CTXT,
- LLOG_TEST_REPL_CTXT,
- LLOG_LOVEA_ORIG_CTXT,
- LLOG_LOVEA_REPL_CTXT,
- LLOG_CHANGELOG_ORIG_CTXT, /**< changelog generation on mdd */
- LLOG_CHANGELOG_REPL_CTXT, /**< changelog access on clients */
- LLOG_CHANGELOG_USER_ORIG_CTXT, /**< for multiple changelog consumers */
- LLOG_AGENT_ORIG_CTXT, /**< agent requests generation on cdt */
- LLOG_MAX_CTXTS
+ u64 off;
+ struct page *pg;
+ u32 count;
+ u32 flag;
};
struct timeout_item {
- enum timeout_event ti_event;
- cfs_time_t ti_timeout;
- timeout_cb_t ti_cb;
- void *ti_cb_data;
- cfs_list_t ti_obd_list;
- cfs_list_t ti_chain;
+ enum timeout_event ti_event;
+ cfs_time_t ti_timeout;
+ timeout_cb_t ti_cb;
+ void *ti_cb_data;
+ struct list_head ti_obd_list;
+ struct list_head ti_chain;
};
#define OBD_MAX_RIF_DEFAULT 8
NUM_SYNC_ON_CANCEL_STATES
};
+/*
+ * Limit reply buffer size for striping data to one x86_64 page. This
+ * value is chosen to fit the striping data for common use cases while
+ * staying well below the limit at which the buffer must be backed by
+ * vmalloc(). Excessive use of vmalloc() may cause spinlock contention
+ * on the MDS.
+ */
+#define OBD_MAX_DEFAULT_EA_SIZE 4096
+#define OBD_MAX_DEFAULT_COOKIE_SIZE 4096
+
struct mdc_rpc_lock;
struct obd_import;
struct client_obd {
- struct rw_semaphore cl_sem;
- struct obd_uuid cl_target_uuid;
- struct obd_import *cl_import; /* ptlrpc connection state */
- int cl_conn_count;
- /* max_mds_easize is purely a performance thing so we don't have to
- * call obd_size_diskmd() all the time. */
- int cl_default_mds_easize;
- int cl_max_mds_easize;
- int cl_default_mds_cookiesize;
- int cl_max_mds_cookiesize;
-
- enum lustre_sec_part cl_sp_me;
- enum lustre_sec_part cl_sp_to;
- struct sptlrpc_flavor cl_flvr_mgc; /* fixed flavor of mgc->mgs */
-
- /* the grant values are protected by loi_list_lock below */
- long cl_dirty; /* all _dirty_ in bytes */
- long cl_dirty_max; /* allowed w/o rpc */
- long cl_dirty_transit; /* dirty synchronous */
- long cl_avail_grant; /* bytes of credit for ost */
- long cl_lost_grant; /* lost credits (trunc) */
+ struct rw_semaphore cl_sem;
+ struct obd_uuid cl_target_uuid;
+ struct obd_import *cl_import; /* ptlrpc connection state */
+ size_t cl_conn_count;
+
+ /* Cache maximum and default values for easize and cookiesize. This is
+ * strictly a performance optimization to minimize calls to
+ * obd_size_diskmd(). The default values are used to calculate the
+ * initial size of a request buffer. The ptlrpc layer will resize the
+ * buffer as needed to accommodate a larger reply from the
+ * server. The default values should be small enough to avoid wasted
+ * memory and excessive use of vmalloc(), yet large enough to avoid
+ * reallocating the buffer in the common use case. */
+
+ /* Default EA size for striping attributes. It is initialized at
+ * mount-time based on the default stripe width of the filesystem,
+ * then it tracks the largest observed EA size advertised by
+ * the MDT, up to a maximum value of OBD_MAX_DEFAULT_EA_SIZE. */
+ __u32 cl_default_mds_easize;
+
+ /* Maximum possible EA size computed at mount-time based on
+ * the number of OSTs in the filesystem. May be increased at
+ * run-time if a larger observed size is advertised by the MDT. */
+ __u32 cl_max_mds_easize;
+
+ /* Default cookie size for llog cookies (see struct llog_cookie). It is
+ * initialized to zero at mount-time, then it tracks the largest
+ * observed cookie size advertised by the MDT, up to a maximum value of
+ * OBD_MAX_DEFAULT_COOKIE_SIZE. Note that llog_cookies are not
+ * used by clients communicating with MDS versions 2.4.0 and later.*/
+ __u32 cl_default_mds_cookiesize;
+
+ /* Maximum possible cookie size computed at mount-time based on
+ * the number of OSTs in the filesystem. May be increased at
+ * run-time if a larger observed size is advertised by the MDT. */
+ __u32 cl_max_mds_cookiesize;
+
+ enum lustre_sec_part cl_sp_me;
+ enum lustre_sec_part cl_sp_to;
+ struct sptlrpc_flavor cl_flvr_mgc; /* fixed flavor of mgc->mgs */
+
+ /* the grant values are protected by loi_list_lock below */
+ unsigned long cl_dirty_pages; /* all _dirty_ in pages */
+ unsigned long cl_dirty_max_pages; /* allowed w/o rpc */
+ unsigned long cl_dirty_transit; /* dirty synchronous */
+ unsigned long cl_avail_grant; /* bytes of credit for ost */
+ unsigned long cl_lost_grant; /* lost credits (trunc) */
/* since we allocate grant by blocks, we don't know how many grant will
* be used to add a page into cache. As a solution, we reserve maximum
* grant before trying to dirty a page and unreserve the rest.
* See osc_{reserve|unreserve}_grant for details. */
- long cl_reserved_grant;
- cfs_list_t cl_cache_waiters; /* waiting for cache/grant */
- cfs_time_t cl_next_shrink_grant; /* jiffies */
- cfs_list_t cl_grant_shrink_list; /* Timeout event list */
- int cl_grant_shrink_interval; /* seconds */
+ long cl_reserved_grant;
+ struct list_head cl_cache_waiters; /* waiting for cache/grant */
+ cfs_time_t cl_next_shrink_grant; /* jiffies */
+ struct list_head cl_grant_shrink_list; /* Timeout event list */
+ int cl_grant_shrink_interval; /* seconds */
/* A chunk is an optimal size used by osc_extent to determine
* the extent size. A chunk is max(PAGE_CACHE_SIZE, OST block size) */
- int cl_chunkbits;
- int cl_chunk;
- int cl_extent_tax; /* extent overhead, by bytes */
+ int cl_chunkbits;
+ unsigned int cl_extent_tax; /* extent overhead, by bytes */
/* keep track of objects that have lois that contain pages which
* have been queued for async brw. this lock also protects the
* blocking everywhere, but we don't want to slow down fast-path of
* our main platform.)
*
- * Exact type of ->cl_loi_list_lock is defined in arch/obd.h together
- * with client_obd_list_{un,}lock() and
- * client_obd_list_lock_{init,done}() functions.
- *
* NB by Jinshan: though field names are still _loi_, but actually
* osc_object{}s are in the list.
*/
- client_obd_lock_t cl_loi_list_lock;
- cfs_list_t cl_loi_ready_list;
- cfs_list_t cl_loi_hp_ready_list;
- cfs_list_t cl_loi_write_list;
- cfs_list_t cl_loi_read_list;
- int cl_r_in_flight;
- int cl_w_in_flight;
+ spinlock_t cl_loi_list_lock;
+ struct list_head cl_loi_ready_list;
+ struct list_head cl_loi_hp_ready_list;
+ struct list_head cl_loi_write_list;
+ struct list_head cl_loi_read_list;
+ __u32 cl_r_in_flight;
+ __u32 cl_w_in_flight;
/* just a sum of the loi/lop pending numbers to be exported by /proc */
- atomic_t cl_pending_w_pages;
- atomic_t cl_pending_r_pages;
- __u32 cl_max_pages_per_rpc;
- int cl_max_rpcs_in_flight;
- struct obd_histogram cl_read_rpc_hist;
- struct obd_histogram cl_write_rpc_hist;
- struct obd_histogram cl_read_page_hist;
- struct obd_histogram cl_write_page_hist;
- struct obd_histogram cl_read_offset_hist;
- struct obd_histogram cl_write_offset_hist;
+ atomic_t cl_pending_w_pages;
+ atomic_t cl_pending_r_pages;
+ __u32 cl_max_pages_per_rpc;
+ __u32 cl_max_rpcs_in_flight;
+ struct obd_histogram cl_read_rpc_hist;
+ struct obd_histogram cl_write_rpc_hist;
+ struct obd_histogram cl_read_page_hist;
+ struct obd_histogram cl_write_page_hist;
+ struct obd_histogram cl_read_offset_hist;
+ struct obd_histogram cl_write_offset_hist;
/* lru for osc caching pages */
struct cl_client_cache *cl_cache;
- cfs_list_t cl_lru_osc; /* member of cl_cache->ccc_lru */
- atomic_t *cl_lru_left;
- atomic_t cl_lru_busy;
+ struct list_head cl_lru_osc; /* member of cl_cache->ccc_lru */
+ atomic_long_t *cl_lru_left;
+ atomic_long_t cl_lru_busy;
+ atomic_long_t cl_lru_in_list;
+ atomic_long_t cl_unstable_count;
+ struct list_head cl_lru_list; /* lru page list */
+ spinlock_t cl_lru_list_lock; /* page list protector */
atomic_t cl_lru_shrinkers;
- atomic_t cl_lru_in_list;
- cfs_list_t cl_lru_list; /* lru page list */
- client_obd_lock_t cl_lru_list_lock; /* page list protector */
- atomic_t cl_unstable_count;
/* number of in flight destroy rpcs is limited to max_rpcs_in_flight */
- atomic_t cl_destroy_in_flight;
- wait_queue_head_t cl_destroy_waitq;
+ atomic_t cl_destroy_in_flight;
+ wait_queue_head_t cl_destroy_waitq;
struct mdc_rpc_lock *cl_rpc_lock;
struct mdc_rpc_lock *cl_close_lock;
/* mgc datastruct */
- struct mutex cl_mgc_mutex;
+ struct mutex cl_mgc_mutex;
struct local_oid_storage *cl_mgc_los;
- struct dt_object *cl_mgc_configs_dir;
- atomic_t cl_mgc_refcount;
- struct obd_export *cl_mgc_mgsexp;
+ struct dt_object *cl_mgc_configs_dir;
+ atomic_t cl_mgc_refcount;
+ struct obd_export *cl_mgc_mgsexp;
/* checksumming for data sent over the network */
unsigned int cl_checksum:1; /* 0 = disabled, 1 = enabled */
/* used by quotacheck when the servers are older than 2.4 */
int cl_qchk_stat; /* quotacheck stat of the peer */
#define CL_NOT_QUOTACHECKED 1 /* client->cl_qchk_stat init value */
-#if LUSTRE_VERSION_CODE >= OBD_OCD_VERSION(2, 7, 50, 0)
+#if LUSTRE_VERSION_CODE >= OBD_OCD_VERSION(2, 7, 53, 0)
#warning "please consider removing quotacheck compatibility code"
#endif
#define obd2cli_tgt(obd) ((char *)(obd)->u.cli.cl_target_uuid.uuid)
struct obd_id_info {
- __u32 idx;
- obd_id *data;
+ u32 idx;
+ u64 *data;
};
struct echo_client_obd {
- struct obd_export *ec_exp; /* the local connection to osc/lov */
+ struct obd_export *ec_exp; /* the local connection to osc/lov */
spinlock_t ec_lock;
- cfs_list_t ec_objects;
- cfs_list_t ec_locks;
- int ec_nstripes;
- __u64 ec_unique;
+ struct list_head ec_objects;
+ struct list_head ec_locks;
+ __u64 ec_unique;
};
/* Generic subset of OSTs */
#define OBD_STATFS_CACHE_SECONDS 1
struct lov_tgt_desc {
- cfs_list_t ltd_kill;
+ struct list_head ltd_kill;
struct obd_uuid ltd_uuid;
struct obd_device *ltd_obd;
struct obd_export *ltd_exp;
enum lustre_sec_part lov_sp_me;
/* Cached LRU and unstable data from upper layer */
- void *lov_cache;
+ struct cl_client_cache *lov_cache;
struct rw_semaphore lov_notify_lock;
};
struct lmv_tgt_desc {
struct obd_uuid ltd_uuid;
struct obd_export *ltd_exp;
- int ltd_idx;
+ __u32 ltd_idx;
struct mutex ltd_fid_mutex;
unsigned long ltd_active:1; /* target up for requests */
};
struct obd_export *exp;
struct proc_dir_entry *targets_proc_entry;
- struct mutex init_mutex;
+ struct mutex lmv_init_mutex;
int connected;
int max_easize;
int max_def_easize;
int max_cookiesize;
int max_def_cookiesize;
- int server_timeout;
- int tgts_size; /* size of tgts array */
+ __u32 tgts_size; /* size of tgts array */
struct lmv_tgt_desc **tgts;
struct obd_connect_data conn_data;
struct niobuf_local {
__u64 lnb_file_offset;
__u32 lnb_page_offset;
- __u32 len;
- __u32 flags;
- struct page *page;
- struct dentry *dentry;
- int lnb_grant_used;
- int rc;
+ __u32 lnb_len;
+ __u32 lnb_flags;
+ struct page *lnb_page;
+ void *lnb_data;
+ int lnb_rc;
};
#define LUSTRE_FLD_NAME "fld"
#define LUSTRE_MGS_OBDNAME "MGS"
#define LUSTRE_MGC_OBDNAME "MGC"
-static inline int is_osp_on_mdt(char *name)
-{
- char *ptr;
-
- ptr = strrchr(name, '-');
- if (ptr == NULL) {
- CERROR("%s is not a obdname\n", name);
- return 0;
- }
-
- /* 1.8 OSC/OSP name on MDT is fsname-OSTxxxx-osc */
- if (strncmp(ptr + 1, "osc", 3) == 0)
- return 1;
-
- if (strncmp(ptr + 1, "MDT", 3) != 0)
- return 0;
-
- while (*(--ptr) != '-' && ptr != name);
-
- if (ptr == name)
- return 0;
-
- if (strncmp(ptr + 1, LUSTRE_OSP_NAME, strlen(LUSTRE_OSP_NAME)) != 0 &&
- strncmp(ptr + 1, LUSTRE_OSC_NAME, strlen(LUSTRE_OSC_NAME)) != 0)
- return 0;
-
- return 1;
-}
-
-/* Don't conflict with on-wire flags OBD_BRW_WRITE, etc */
-#define N_LOCAL_TEMP_PAGE 0x10000000
-
struct obd_trans_info {
- __u64 oti_transno;
- __u64 oti_xid;
- /* Only used on the server side for tracking acks. */
- struct oti_req_ack_lock {
- struct lustre_handle lock;
- __u32 mode;
- } oti_ack_locks[4];
- void *oti_handle;
- struct llog_cookie oti_onecookie;
- struct llog_cookie *oti_logcookies;
- int oti_numcookies;
- /** synchronous write is needed */
- unsigned long oti_sync_write:1;
-
- /* initial thread handling transaction */
- struct ptlrpc_thread * oti_thread;
- __u32 oti_conn_cnt;
- /** VBR: versions */
- __u64 oti_pre_version;
- /** JobID */
- char *oti_jobid;
-
- struct obd_uuid *oti_ost_uuid;
+ __u64 oti_xid;
+ /* Only used on the server side for tracking acks. */
+ struct oti_req_ack_lock {
+ struct lustre_handle lock;
+ __u32 mode;
+ } oti_ack_locks[4];
+ void *oti_handle;
+ struct llog_cookie oti_onecookie;
+ struct llog_cookie *oti_logcookies;
+
+ /** VBR: versions */
+ __u64 oti_pre_version;
};
-static inline void oti_init(struct obd_trans_info *oti,
- struct ptlrpc_request *req)
-{
- if (oti == NULL)
- return;
- memset(oti, 0, sizeof(*oti));
-
- if (req == NULL)
- return;
-
- oti->oti_xid = req->rq_xid;
- /** VBR: take versions from request */
- if (req->rq_reqmsg != NULL &&
- lustre_msg_get_flags(req->rq_reqmsg) & MSG_REPLAY) {
- __u64 *pre_version = lustre_msg_get_versions(req->rq_reqmsg);
- oti->oti_pre_version = pre_version ? pre_version[0] : 0;
- oti->oti_transno = lustre_msg_get_transno(req->rq_reqmsg);
- }
-
- /** called from mds_create_objects */
- if (req->rq_repmsg != NULL)
- oti->oti_transno = lustre_msg_get_transno(req->rq_repmsg);
- oti->oti_thread = req->rq_svc_thread;
- if (req->rq_reqmsg != NULL)
- oti->oti_conn_cnt = lustre_msg_get_conn_cnt(req->rq_reqmsg);
-}
-
-static inline void oti_alloc_cookies(struct obd_trans_info *oti,int num_cookies)
-{
- if (!oti)
- return;
-
- if (num_cookies == 1)
- oti->oti_logcookies = &oti->oti_onecookie;
- else
- OBD_ALLOC_LARGE(oti->oti_logcookies,
- num_cookies * sizeof(oti->oti_onecookie));
-
- oti->oti_numcookies = num_cookies;
-}
-
-static inline void oti_free_cookies(struct obd_trans_info *oti)
-{
- if (!oti || !oti->oti_logcookies)
- return;
-
- if (oti->oti_logcookies == &oti->oti_onecookie)
- LASSERT(oti->oti_numcookies == 1);
- else
- OBD_FREE_LARGE(oti->oti_logcookies,
- oti->oti_numcookies*sizeof(oti->oti_onecookie));
- oti->oti_logcookies = NULL;
- oti->oti_numcookies = 0;
-}
-
/*
* Events signalled through obd_notify() upcall-chain.
*/
};
struct obd_llog_group {
- int olg_seq;
struct llog_ctxt *olg_ctxts[LLOG_MAX_CTXTS];
wait_queue_head_t olg_waitq;
spinlock_t olg_lock;
obd_starting:1, /* started setup */
obd_force:1, /* cleanup with > 0 obd refcount */
obd_fail:1, /* cleanup with failover */
- obd_async_recov:1, /* allow asynchronous orphan cleanup */
obd_no_conn:1, /* deny new connections */
obd_inactive:1, /* device active/inactive
* (for /proc/status only!!) */
cfs_hash_t *obd_nid_hash;
/* nid stats body */
cfs_hash_t *obd_nid_stats_hash;
- cfs_list_t obd_nid_stats;
- atomic_t obd_refcount;
- cfs_list_t obd_exports;
- cfs_list_t obd_unlinked_exports;
- cfs_list_t obd_delayed_exports;
+ struct list_head obd_nid_stats;
+ atomic_t obd_refcount;
+ struct list_head obd_exports;
+ struct list_head obd_unlinked_exports;
+ struct list_head obd_delayed_exports;
struct list_head obd_lwp_list;
int obd_num_exports;
spinlock_t obd_nid_lock;
struct obd_notify_upcall obd_upcall;
struct obd_export *obd_self_export;
struct obd_export *obd_lwp_export;
- /* list of exports in LRU order, for ping evictor, with obd_dev_lock */
- cfs_list_t obd_exports_timed;
- time_t obd_eviction_timer; /* for ping evictor */
+ /* list of exports in LRU order, for ping evictor, with obd_dev_lock */
+ struct list_head obd_exports_timed;
+ time_t obd_eviction_timer; /* for ping evictor */
int obd_max_recoverable_clients;
atomic_t obd_connected_clients;
int obd_stale_clients;
- int obd_delayed_clients;
/* this lock protects all recovery list_heads, timer and
* obd_next_recovery_transno value */
- spinlock_t obd_recovery_task_lock;
- __u64 obd_next_recovery_transno;
- int obd_replayed_requests;
- int obd_requests_queued_for_recovery;
- wait_queue_head_t obd_next_transno_waitq;
+ spinlock_t obd_recovery_task_lock;
+ __u64 obd_next_recovery_transno;
+ int obd_replayed_requests;
+ int obd_requests_queued_for_recovery;
+ wait_queue_head_t obd_next_transno_waitq;
/* protected by obd_recovery_task_lock */
- struct timer_list obd_recovery_timer;
- time_t obd_recovery_start; /* seconds */
- time_t obd_recovery_end; /* seconds, for lprocfs_status */
- int obd_recovery_time_hard;
- int obd_recovery_timeout;
- int obd_recovery_ir_factor;
+ struct timer_list obd_recovery_timer;
+ /* seconds */
+ time_t obd_recovery_start;
+ /* seconds, for lprocfs_status */
+ time_t obd_recovery_end;
+ int obd_recovery_time_hard;
+ int obd_recovery_timeout;
+ int obd_recovery_ir_factor;
/* new recovery stuff from CMD2 */
- struct target_recovery_data obd_recovery_data;
- int obd_replayed_locks;
- atomic_t obd_req_replay_clients;
- atomic_t obd_lock_replay_clients;
+ struct target_recovery_data obd_recovery_data;
+ int obd_replayed_locks;
+ atomic_t obd_req_replay_clients;
+ atomic_t obd_lock_replay_clients;
/* all lists are protected by obd_recovery_task_lock */
- cfs_list_t obd_req_replay_queue;
- cfs_list_t obd_lock_replay_queue;
- cfs_list_t obd_final_req_queue;
+ struct list_head obd_req_replay_queue;
+ struct list_head obd_lock_replay_queue;
+ struct list_head obd_final_req_queue;
union {
#ifdef HAVE_SERVER_SUPPORT
struct proc_dir_entry *obd_proc_exports_entry;
struct proc_dir_entry *obd_svc_procroot;
struct lprocfs_stats *obd_svc_stats;
- struct lprocfs_seq_vars *obd_vars;
+ struct lprocfs_vars *obd_vars;
atomic_t obd_evict_inprogress;
wait_queue_head_t obd_evict_inprogress_waitq;
- cfs_list_t obd_evict_list; /* protected with pet_lock */
+ struct list_head obd_evict_list; /* protected with pet_lock */
/**
* Ldlm pool part. Save last calculated SLV and Limit.
*/
rwlock_t obd_pool_lock;
- int obd_pool_limit;
- __u64 obd_pool_slv;
+ int obd_pool_limit;
+ __u64 obd_pool_slv;
/**
* A list of outstanding class_incref()'s against this obd. For
int obd_conn_inprogress;
};
-#define OBD_LLOG_FL_SENDNOW 0x0001
-#define OBD_LLOG_FL_EXIT 0x0002
-
enum obd_cleanup_stage {
/* Special case hack for MDS LOVs */
OBD_CLEANUP_EARLY,
/* get/set_info keys */
#define KEY_ASYNC "async"
-#define KEY_BLOCKSIZE_BITS "blocksize_bits"
-#define KEY_BLOCKSIZE "blocksize"
#define KEY_CAPA_KEY "capa_key"
#define KEY_CHANGELOG_CLEAR "changelog_clear"
#define KEY_FID2PATH "fid2path"
#define KEY_GRANT_SHRINK "grant_shrink"
#define KEY_HSM_COPYTOOL_SEND "hsm_send"
#define KEY_INIT_RECOV_BACKUP "init_recov_bk"
-#define KEY_INIT_RECOV "initial_recov"
#define KEY_INTERMDS "inter_mds"
#define KEY_LAST_ID "last_id"
#define KEY_LAST_FID "last_fid"
-#define KEY_LOCK_TO_STRIPE "lock_to_stripe"
#define KEY_LOVDESC "lovdesc"
-#define KEY_LOV_IDX "lov_idx"
#define KEY_MAX_EASIZE "max_easize"
#define KEY_DEFAULT_EASIZE "default_easize"
#define KEY_MAX_COOKIESIZE "max_cookiesize"
#define KEY_DEFAULT_COOKIESIZE "default_cookiesize"
-#define KEY_MDS_CONN "mds_conn"
#define KEY_MGSSEC "mgssec"
-#define KEY_NEXT_ID "next_id"
#define KEY_READ_ONLY "read-only"
#define KEY_REGISTER_TARGET "register_target"
#define KEY_SET_FS "set_fs"
#define KEY_TGT_COUNT "tgt_count"
/* KEY_SET_INFO in lustre_idl.h */
#define KEY_SPTLRPC_CONF "sptlrpc_conf"
-#define KEY_CONNECT_FLAG "connect_flags"
-#define KEY_SYNC_LOCK_CANCEL "sync_lock_cancel"
#define KEY_CACHE_SET "cache_set"
#define KEY_CACHE_LRU_SHRINK "cache_lru_shrink"
-#define KEY_CHANGELOG_INDEX "changelog_index"
#define KEY_OSP_CONNECTED "osp_connected"
struct lu_context;
return -EINVAL;
}
+enum md_op_flags {
+ MF_MDC_CANCEL_FID1 = 1 << 0,
+ MF_MDC_CANCEL_FID2 = 1 << 1,
+ MF_MDC_CANCEL_FID3 = 1 << 2,
+ MF_MDC_CANCEL_FID4 = 1 << 3,
+ MF_GET_MDT_IDX = 1 << 4,
+};
+
+enum md_cli_flags {
+ CLI_SET_MEA = 1 << 0,
+ CLI_RM_ENTRY = 1 << 1,
+ CLI_HASH64 = 1 << 2,
+ CLI_API32 = 1 << 3,
+ CLI_MIGRATE = 1 << 4,
+};
+
struct md_op_data {
struct lu_fid op_fid1; /* operation fid1 (usualy parent) */
struct lu_fid op_fid2; /* operation fid2 (usualy child) */
struct lu_fid op_fid3; /* 2 extra fids to find conflicting */
struct lu_fid op_fid4; /* to the operation locks. */
- mdsno_t op_mds; /* what mds server open will go to */
- struct lustre_handle op_handle;
- obd_time op_mod_time;
+ u32 op_mds; /* what mds server open will go to */
+ struct lustre_handle op_handle;
+ s64 op_mod_time;
const char *op_name;
- int op_namelen;
+ size_t op_namelen;
__u32 op_mode;
struct lmv_stripe_md *op_mea1;
struct lmv_stripe_md *op_mea2;
size_t op_data_size;
/* iattr fields and blocks. */
- struct iattr op_attr;
-#ifdef __KERNEL__
- unsigned int op_attr_flags;
-#endif
- __u64 op_valid;
+ struct iattr op_attr;
loff_t op_attr_blocks;
+ unsigned int op_attr_flags; /* LUSTRE_{SYNC,..}_FL */
+ __u64 op_valid; /* OBD_MD_* */
- /* Size-on-MDS epoch and flags. */
- __u64 op_ioepoch;
- __u32 op_flags;
+ enum md_op_flags op_flags;
/* Capa fields */
struct obd_capa *op_capa1;
enum mds_op_bias op_bias;
/* Used by readdir */
- __u64 op_hash_offset;
-
- /* Used by readdir */
- __u32 op_npages;
+ unsigned int op_max_pages;
/* used to transfer info between the stacks of MD client
* see enum op_cli_flags */
- __u32 op_cli_flags;
+ enum md_cli_flags op_cli_flags;
/* File object data version for HSM release, on client */
__u64 op_data_version;
struct lustre_handle op_lease_handle;
-};
-#define op_stripe_offset op_ioepoch
-#define op_max_pages op_valid
+ /* default stripe offset */
+ __u32 op_default_stripe_offset;
+};
struct md_callback {
int (*md_blocking_ast)(struct ldlm_lock *lock,
int rc);
struct md_enqueue_info {
- struct md_op_data mi_data;
- struct lookup_intent mi_it;
- struct lustre_handle mi_lockh;
- struct inode *mi_dir;
- md_enqueue_cb_t mi_cb;
- __u64 mi_cbdata;
- unsigned int mi_generation;
+ struct md_op_data mi_data;
+ struct lookup_intent mi_it;
+ struct lustre_handle mi_lockh;
+ struct inode *mi_dir;
+ md_enqueue_cb_t mi_cb;
+ void *mi_cbdata;
};
struct obd_ops {
struct module *o_owner;
int (*o_iocontrol)(unsigned int cmd, struct obd_export *exp, int len,
- void *karg, void *uarg);
- int (*o_get_info)(const struct lu_env *env, struct obd_export *,
- __u32 keylen, void *key, __u32 *vallen, void *val,
- struct lov_stripe_md *lsm);
- int (*o_set_info_async)(const struct lu_env *, struct obd_export *,
- __u32 keylen, void *key,
- __u32 vallen, void *val,
- struct ptlrpc_request_set *set);
- int (*o_attach)(struct obd_device *dev, obd_count len, void *data);
- int (*o_detach)(struct obd_device *dev);
+ void *karg, void __user *uarg);
+ int (*o_get_info)(const struct lu_env *env, struct obd_export *,
+ __u32 keylen, void *key, __u32 *vallen, void *val);
+ int (*o_set_info_async)(const struct lu_env *, struct obd_export *,
+ __u32 keylen, void *key,
+ __u32 vallen, void *val,
+ struct ptlrpc_request_set *set);
int (*o_setup) (struct obd_device *dev, struct lustre_cfg *cfg);
int (*o_precleanup)(struct obd_device *dev,
enum obd_cleanup_stage cleanup_stage);
int (*o_cleanup)(struct obd_device *dev);
- int (*o_process_config)(struct obd_device *dev, obd_count len,
- void *data);
+ int (*o_process_config)(struct obd_device *dev, size_t len, void *data);
int (*o_postrecov)(struct obd_device *dev);
int (*o_add_conn)(struct obd_import *imp, struct obd_uuid *uuid,
int priority);
struct lov_stripe_md *mem_src);
int (*o_unpackmd)(struct obd_export *exp,struct lov_stripe_md **mem_tgt,
struct lov_mds_md *disk_src, int disk_len);
- int (*o_create)(const struct lu_env *env, struct obd_export *exp,
- struct obdo *oa, struct lov_stripe_md **ea,
- struct obd_trans_info *oti);
- int (*o_create_async)(struct obd_export *exp, struct obd_info *oinfo,
- struct lov_stripe_md **ea,
- struct obd_trans_info *oti);
- int (*o_destroy)(const struct lu_env *env, struct obd_export *exp,
- struct obdo *oa, struct lov_stripe_md *ea,
- struct obd_trans_info *oti, struct obd_export *md_exp,
- void *capa);
+ int (*o_create)(const struct lu_env *env, struct obd_export *exp,
+ struct obdo *oa, struct obd_trans_info *oti);
+ int (*o_destroy)(const struct lu_env *env, struct obd_export *exp,
+ struct obdo *oa, struct obd_trans_info *oti);
int (*o_setattr)(const struct lu_env *, struct obd_export *exp,
struct obd_info *oinfo, struct obd_trans_info *oti);
- int (*o_setattr_async)(struct obd_export *exp, struct obd_info *oinfo,
- struct obd_trans_info *oti,
- struct ptlrpc_request_set *rqset);
int (*o_getattr)(const struct lu_env *env, struct obd_export *exp,
struct obd_info *oinfo);
- int (*o_getattr_async)(struct obd_export *exp, struct obd_info *oinfo,
- struct ptlrpc_request_set *set);
int (*o_preprw)(const struct lu_env *env, int cmd,
struct obd_export *exp, struct obdo *oa, int objcount,
struct obd_ioobj *obj, struct niobuf_remote *remote,
- int *nr_pages, struct niobuf_local *local,
- struct obd_trans_info *oti, struct lustre_capa *capa);
+ int *nr_pages, struct niobuf_local *local,
+ struct obd_trans_info *oti);
int (*o_commitrw)(const struct lu_env *env, int cmd,
struct obd_export *exp, struct obdo *oa,
int objcount, struct obd_ioobj *obj,
struct niobuf_remote *remote, int pages,
struct niobuf_local *local,
struct obd_trans_info *oti, int rc);
- int (*o_change_cbdata)(struct obd_export *, struct lov_stripe_md *,
- ldlm_iterator_t it, void *data);
- int (*o_find_cbdata)(struct obd_export *, struct lov_stripe_md *,
- ldlm_iterator_t it, void *data);
int (*o_init_export)(struct obd_export *exp);
int (*o_destroy_export)(struct obd_export *exp);
- /* llog related obd_methods */
- int (*o_llog_init)(struct obd_device *obd, struct obd_llog_group *grp,
- struct obd_device *disk_obd, int *idx);
- int (*o_llog_finish)(struct obd_device *obd, int count);
- int (*o_llog_connect)(struct obd_export *, struct llogd_conn_body *);
-
- /* metadata-only methods */
- int (*o_pin)(struct obd_export *, const struct lu_fid *fid,
- struct obd_capa *, struct obd_client_handle *, int flag);
- int (*o_unpin)(struct obd_export *, struct obd_client_handle *, int);
-
int (*o_import_event)(struct obd_device *, struct obd_import *,
enum obd_import_event);
struct mdt_remote_perm *remote_perm;
struct obd_capa *mds_capa;
struct obd_capa *oss_capa;
- __u64 lm_flags;
};
struct md_open_data {
bool mod_is_create;
};
+struct obd_client_handle {
+ struct lustre_handle och_fh;
+ struct lu_fid och_fid;
+ struct md_open_data *och_mod;
+ struct lustre_handle och_lease_handle; /* open lock for lease */
+ __u32 och_magic;
+ int och_flags;
+};
+
+#define OBD_CLIENT_HANDLE_MAGIC 0xd15ea5ed
+
struct lookup_intent;
struct cl_attr;
struct md_open_data *, struct ptlrpc_request **);
int (*m_create)(struct obd_export *, struct md_op_data *,
- const void *, int, int, __u32, __u32, cfs_cap_t,
- __u64, struct ptlrpc_request **);
+ const void *, size_t, umode_t, uid_t, gid_t,
+ cfs_cap_t, __u64, struct ptlrpc_request **);
int (*m_enqueue)(struct obd_export *, struct ldlm_enqueue_info *,
const union ldlm_policy_data *,
struct ptlrpc_request **);
int (*m_rename)(struct obd_export *, struct md_op_data *,
- const char *, int, const char *, int,
+ const char *, size_t, const char *, size_t,
struct ptlrpc_request **);
int (*m_setattr)(struct obd_export *, struct md_op_data *, void *,
- int , void *, int, struct ptlrpc_request **,
- struct md_open_data **mod);
+ size_t , struct ptlrpc_request **);
int (*m_fsync)(struct obd_export *, const struct lu_fid *,
struct obd_capa *, struct ptlrpc_request **);
- int (*m_readpage)(struct obd_export *, struct md_op_data *,
- struct page **, struct ptlrpc_request **);
-
- int (*m_read_entry)(struct obd_export *, struct md_op_data *,
- struct md_callback *cb_op, struct lu_dirent **ld,
- struct page **ppage);
+ int (*m_read_page)(struct obd_export *, struct md_op_data *,
+ struct md_callback *cb_op, __u64 hash_offset,
+ struct page **ppage);
int (*m_unlink)(struct obd_export *, struct md_op_data *,
struct ptlrpc_request **);
int (*m_setxattr)(struct obd_export *, const struct lu_fid *,
- struct obd_capa *, obd_valid, const char *,
+ struct obd_capa *, u64, const char *,
const char *, int, int, int, __u32,
struct ptlrpc_request **);
int (*m_getxattr)(struct obd_export *, const struct lu_fid *,
- struct obd_capa *, obd_valid, const char *,
+ struct obd_capa *, u64, const char *,
const char *, int, int, int,
struct ptlrpc_request **);
int (*m_revalidate_lock)(struct obd_export *, struct lookup_intent *,
struct lu_fid *, __u64 *bits);
+
#define MD_STATS_LAST_OP m_revalidate_lock
int (*m_getstatus)(struct obd_export *, struct lu_fid *,
int (*m_find_cbdata)(struct obd_export *, const struct lu_fid *,
ldlm_iterator_t, void *);
- int (*m_done_writing)(struct obd_export *, struct md_op_data *,
- struct md_open_data *);
-
int (*m_getattr_name)(struct obd_export *, struct md_op_data *,
struct ptlrpc_request **);
- int (*m_is_subdir)(struct obd_export *, const struct lu_fid *,
- const struct lu_fid *,
- struct ptlrpc_request **);
-
- int (*m_init_ea_size)(struct obd_export *, int, int, int, int);
+ int (*m_init_ea_size)(struct obd_export *, __u32, __u32, __u32, __u32);
int (*m_get_lustre_md)(struct obd_export *, struct ptlrpc_request *,
struct obd_export *, struct obd_export *,
int (*m_merge_attr)(struct obd_export *,
const struct lmv_stripe_md *lsm,
- struct cl_attr *attr);
-
- int (*m_update_lsm_md)(struct obd_export *, struct lmv_stripe_md *lsm,
- struct mdt_body *, ldlm_blocking_callback);
+ struct cl_attr *attr, ldlm_blocking_callback);
int (*m_set_open_replay_data)(struct obd_export *,
struct obd_client_handle *,
int (*m_get_remote_perm)(struct obd_export *, const struct lu_fid *,
struct obd_capa *, __u32,
struct ptlrpc_request **);
+
+ int (*m_get_fid_from_lsm)(struct obd_export *,
+ const struct lmv_stripe_md *,
+ const char *name, int namelen,
+ struct lu_fid *fid);
};
struct lsm_operations {
- void (*lsm_free)(struct lov_stripe_md *);
- int (*lsm_destroy)(struct lov_stripe_md *, struct obdo *oa,
- struct obd_export *md_exp);
- void (*lsm_stripe_by_index)(struct lov_stripe_md *, int *, obd_off *,
- obd_off *);
- void (*lsm_stripe_by_offset)(struct lov_stripe_md *, int *, obd_off *,
- obd_off *);
+ void (*lsm_free)(struct lov_stripe_md *);
+ void (*lsm_stripe_by_index)(struct lov_stripe_md *, int *,
+ loff_t *, loff_t *);
+ void (*lsm_stripe_by_offset)(struct lov_stripe_md *, int *,
+ loff_t *, loff_t *);
int (*lsm_lmm_verify) (struct lov_mds_md *lmm, int lmm_bytes,
__u16 *stripe_count);
int (*lsm_unpackmd) (struct lov_obd *lov, struct lov_stripe_md *lsm,
extern const struct lsm_operations lsm_v1_ops;
extern const struct lsm_operations lsm_v3_ops;
-static inline const struct lsm_operations *lsm_op_find(int magic)
+static inline const struct lsm_operations *lsm_op_find(u32 magic)
{
switch(magic) {
case LOV_MAGIC_V1:
}
}
-/* Requests for obd_extent_calc() */
-#define OBD_CALC_STRIPE_START 1
-#define OBD_CALC_STRIPE_END 2
-
-static inline struct lustre_capa *oinfo_capa(struct obd_info *oinfo)
-{
- return oinfo->oi_capa;
-}
-
static inline struct md_open_data *obd_mod_alloc(void)
{
struct md_open_data *mod;
} \
})
-void obdo_from_inode(struct obdo *dst, struct inode *src, obd_flag valid);
+void obdo_from_inode(struct obdo *dst, struct inode *src, u64 valid);
void obdo_set_parent_fid(struct obdo *dst, const struct lu_fid *parent);
/* return 1 if client should be resend request */
return lu_dev->ld_obd->obd_name;
}
-static inline bool filename_is_volatile(const char *name, int namelen, int *idx)
+static inline bool filename_is_volatile(const char *name, size_t namelen,
+ int *idx)
{
const char *start;
char *end;
}
/* we have an idx, read it */
start = name + LUSTRE_VOLATILE_HDR_LEN + 1;
- *idx = strtoul(start, &end, 16);
+ *idx = simple_strtoul(start, &end, 16);
/* error cases:
* no digit, no trailing :, negative value
*/
return obd->u.cli.cl_max_pages_per_rpc << PAGE_CACHE_SHIFT;
}
+/* when RPC size or the max RPCs in flight is increased, the max dirty pages
+ * of the client should be increased accordingly to avoid sending fragmented
+ * RPCs over the network when the client runs out of the maximum dirty space
+ * when so many RPCs are being generated.
+ */
+static inline void client_adjust_max_dirty(struct client_obd *cli)
+{
+ /* initializing */
+ if (cli->cl_dirty_max_pages <= 0)
+ cli->cl_dirty_max_pages = (OSC_MAX_DIRTY_DEFAULT * 1024 * 1024)
+ >> PAGE_CACHE_SHIFT;
+ else {
+ unsigned long dirty_max = cli->cl_max_rpcs_in_flight *
+ cli->cl_max_pages_per_rpc;
+
+ if (dirty_max > cli->cl_dirty_max_pages)
+ cli->cl_dirty_max_pages = dirty_max;
+ }
+
+ if (cli->cl_dirty_max_pages > totalram_pages / 8)
+ cli->cl_dirty_max_pages = totalram_pages / 8;
+}
+
#endif /* __OBD_H */