* Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
- * Copyright (c) 2011, 2016, Intel Corporation.
+ * Copyright (c) 2011, 2017, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
struct md_ops *typ_md_ops;
struct proc_dir_entry *typ_procroot;
struct proc_dir_entry *typ_procsym;
- __u32 typ_sym_filter;
+ struct dentry *typ_debugfs_entry;
+#ifdef HAVE_SERVER_SUPPORT
+ bool typ_sym_filter;
+#endif
char *typ_name;
int typ_refcnt;
struct lu_device_type *typ_lu;
spinlock_t obd_type_lock;
- struct kobject typ_kobj;
- struct completion typ_kobj_unregister;
+ struct kobject *typ_kobj;
};
struct brw_page {
#define OBD_MAX_RIF_DEFAULT 8
#define OBD_MAX_RIF_MAX 512
#define OSC_MAX_RIF_MAX 256
-#define OSC_MAX_DIRTY_DEFAULT (OBD_MAX_RIF_DEFAULT * 4)
+#define OSC_MAX_DIRTY_DEFAULT 2000 /* Arbitrary large value */
#define OSC_MAX_DIRTY_MB_MAX 2048 /* arbitrary, but < MAX_LONG bytes */
#define OSC_DEFAULT_RESENDS 10
* run-time if a larger observed size is advertised by the MDT. */
__u32 cl_max_mds_easize;
+ /* Data-on-MDT specific value to set larger reply buffer for possible
+ * data read along with open/stat requests. By default it tries to use
+ * unused space in reply buffer.
+ * This value is used to ensure that reply buffer has at least as
+ * much free space as value indicates. That free space is gained from
+ * LOV EA buffer which is small for DoM files and on big systems can
+ * provide up to 32KB of extra space in reply buffer.
+ * Default value is 8K now.
+ */
+ __u32 cl_dom_min_inline_repsize;
+
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 */
long cl_reserved_grant;
struct list_head cl_cache_waiters; /* waiting for cache/grant */
time64_t cl_next_shrink_grant; /* seconds */
- struct list_head cl_grant_shrink_list; /* Timeout event list */
+ struct list_head cl_grant_chain;
time64_t cl_grant_shrink_interval; /* seconds */
/* A chunk is an optimal size used by osc_extent to determine
/* 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;
- __u32 cl_max_rpcs_in_flight;
- __u32 cl_short_io_bytes;
+ u32 cl_max_pages_per_rpc;
+ u32 cl_max_rpcs_in_flight;
+ u32 cl_max_short_io_bytes;
struct obd_histogram cl_read_rpc_hist;
struct obd_histogram cl_write_rpc_hist;
struct obd_histogram cl_read_page_hist;
atomic_t cl_destroy_in_flight;
wait_queue_head_t cl_destroy_waitq;
- struct mdc_rpc_lock *cl_rpc_lock;
-
/* modify rpcs in flight
* currently used for metadata only */
spinlock_t cl_mod_rpcs_lock;
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;
+ atomic_t cl_mgc_refcount;
+ /* in-flight control list and total RPCs counter */
+ struct list_head cl_flight_waiters;
+ __u32 cl_rpcs_in_flight;
/* checksumming for data sent over the network */
unsigned int cl_checksum:1, /* 0 = disabled, 1 = enabled */
/* ptlrpc work for writeback in ptlrpcd context */
void *cl_writeback_work;
void *cl_lru_work;
+ struct mutex cl_quota_mutex;
/* hash tables for osc_quota_info */
struct cfs_hash *cl_quota_hash[LL_MAXQUOTAS];
+ /* the xid of the request updating the hash tables */
+ __u64 cl_quota_last_xid;
/* Links to the global list of registered changelog devices */
struct list_head cl_chg_dev_linkage;
};
ltd_reap:1; /* should this target be deleted */
};
+struct lov_md_tgt_desc {
+ struct obd_device *lmtd_mdc;
+ __u32 lmtd_index;
+};
+
struct lov_obd {
struct lov_desc desc;
struct lov_tgt_desc **lov_tgts; /* sparse array */
struct cl_client_cache *lov_cache;
struct rw_semaphore lov_notify_lock;
+ /* Data-on-MDT: MDC array */
+ struct lov_md_tgt_desc *lov_mdc_tgts;
+
+ struct kobject *lov_tgts_kobj;
};
struct lmv_tgt_desc {
struct obd_uuid ltd_uuid;
+ struct obd_device *ltd_obd;
struct obd_export *ltd_exp;
__u32 ltd_idx;
struct mutex ltd_fid_mutex;
struct lu_client_fld lmv_fld;
spinlock_t lmv_lock;
struct lmv_desc desc;
- struct obd_uuid cluuid;
- struct proc_dir_entry *targets_proc_entry;
struct mutex lmv_init_mutex;
int connected;
int max_easize;
int max_def_easize;
+ u32 lmv_statfs_start;
- __u32 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 kobject *lmv_tgts_kobj;
+ void *lmv_cache;
};
+/* Minimum sector size is 512 */
+#define MAX_GUARD_NUMBER (PAGE_SIZE / 512)
+
struct niobuf_local {
__u64 lnb_file_offset;
__u32 lnb_page_offset;
int lnb_rc;
struct page *lnb_page;
void *lnb_data;
+ __u16 lnb_guards[MAX_GUARD_NUMBER];
+ __u16 lnb_guard_rpc:1;
+ __u16 lnb_guard_disk:1;
};
struct tgt_thread_big_cache {
* (for /proc/status only!!) */
obd_no_ir:1, /* no imperative recovery. */
obd_process_conf:1, /* device is processing mgs config */
- obd_uses_nid_stats:1, /* maintain per-client OBD stats */
obd_checksum_dump:1; /* dump pages upon cksum error */
/* use separate field as it is set in interrupt to don't mess with
spinlock_t obd_dev_lock; /* protect OBD bitfield above */
spinlock_t obd_osfs_lock;
struct obd_statfs obd_osfs; /* locked by obd_osfs_lock */
- __u64 obd_osfs_age;
+ time64_t obd_osfs_age;
__u64 obd_last_committed;
struct mutex obd_dev_mutex;
struct lvfs_run_ctxt obd_lvfs_ctxt;
/* Fields used by LProcFS */
struct lprocfs_stats *obd_stats;
- unsigned int obd_cntr_base;
- unsigned int obd_md_cntr_base;
struct lprocfs_stats *obd_md_stats;
+ struct dentry *obd_debugfs_entry;
struct proc_dir_entry *obd_proc_entry;
struct proc_dir_entry *obd_proc_exports_entry;
- struct proc_dir_entry *obd_svc_procroot;
+ struct dentry *obd_svc_debugfs_entry;
struct lprocfs_stats *obd_svc_stats;
- struct attribute_group obd_attrs_group;
- struct attribute **obd_attrs;
+ const struct attribute **obd_attrs;
struct lprocfs_vars *obd_vars;
atomic_t obd_evict_inprogress;
wait_queue_head_t obd_evict_inprogress_waitq;
/* KEY_SET_INFO in lustre_idl.h */
#define KEY_SPTLRPC_CONF "sptlrpc_conf"
-#define KEY_CACHE_SET "cache_set"
#define KEY_CACHE_LRU_SHRINK "cache_lru_shrink"
#define KEY_OSP_CONNECTED "osp_connected"
+/* Flags for op_xvalid */
+enum op_xvalid {
+ OP_XVALID_CTIME_SET = BIT(0), /* 0x0001 */
+ OP_XVALID_BLOCKS = BIT(1), /* 0x0002 */
+ OP_XVALID_OWNEROVERRIDE = BIT(2), /* 0x0004 */
+ OP_XVALID_FLAGS = BIT(3), /* 0x0008 */
+ OP_XVALID_PROJID = BIT(4), /* 0x0010 */
+ OP_XVALID_LAZYSIZE = BIT(5), /* 0x0020 */
+ OP_XVALID_LAZYBLOCKS = BIT(6), /* 0x0040 */
+};
+
struct lu_context;
static inline int it_to_lock_mode(struct lookup_intent *it)
/* CREAT needs to be tested before open (both could be set) */
if (it->it_op & IT_CREAT)
return LCK_CW;
- else if (it->it_op & (IT_GETATTR | IT_OPEN | IT_LOOKUP |
- IT_LAYOUT))
+ else if (it->it_op & (IT_GETATTR | IT_OPEN | IT_LOOKUP))
return LCK_CR;
+ else if (it->it_op & IT_LAYOUT)
+ return (it->it_flags & FMODE_WRITE) ? LCK_EX : LCK_CR;
else if (it->it_op & IT_READDIR)
return LCK_PR;
else if (it->it_op & IT_GETXATTR)
return LCK_PR;
- else if (it->it_op & IT_SETXATTR)
- return LCK_PW;
LASSERTF(0, "Invalid it_op: %d\n", it->it_op);
return -EINVAL;
*/
static inline bool it_has_reply_body(const struct lookup_intent *it)
{
- return it->it_op & (IT_OPEN | IT_UNLINK | IT_LOOKUP | IT_GETATTR);
+ return it->it_op & (IT_OPEN | IT_LOOKUP | IT_GETATTR);
}
struct md_op_data {
struct lu_fid op_fid4; /* to the operation locks. */
u32 op_mds; /* what mds server open will go to */
__u32 op_mode;
- struct lustre_handle op_handle;
+ struct lustre_handle op_open_handle;
s64 op_mod_time;
const char *op_name;
size_t op_namelen;
+ struct rw_semaphore *op_mea1_sem;
+ struct rw_semaphore *op_mea2_sem;
struct lmv_stripe_md *op_mea1;
struct lmv_stripe_md *op_mea2;
__u32 op_suppgids[2];
/* iattr fields and blocks. */
struct iattr op_attr;
+ enum op_xvalid op_xvalid; /* eXtra validity flags */
loff_t op_attr_blocks;
- __u64 op_valid; /* OBD_MD_* */
- unsigned int op_attr_flags; /* LUSTRE_{SYNC,..}_FL */
+ u64 op_valid; /* OBD_MD_* */
+ unsigned int op_attr_flags; /* LUSTRE_{SYNC,..}_FL */
enum md_op_flags op_flags;
__u64 op_data_version;
struct lustre_handle op_lease_handle;
- /* File security context, for creates. */
+ /* File security context, for creates/metadata ops */
const char *op_file_secctx_name;
+ __u32 op_file_secctx_name_size;
void *op_file_secctx;
__u32 op_file_secctx_size;
/* Used by readdir */
unsigned int op_max_pages;
+ __u16 op_mirror_id;
+
+ /*
+ * used to access migrating dir: if it's set, assume migration is
+ * finished, use the new layout to access dir, otherwise use old layout.
+ * By default it's not set, because new files are created under new
+ * layout, if we can't find file with name under both old and new
+ * layout, we are sure file with name doesn't exist, but in reverse
+ * order there may be a race with creation by others.
+ */
+ bool op_post_migrate;
+ /* used to access dir with bash hash */
+ __u32 op_stripe_index;
};
struct md_callback {
* about this.
*/
int (*o_statfs)(const struct lu_env *, struct obd_export *exp,
- struct obd_statfs *osfs, __u64 max_age, __u32 flags);
+ struct obd_statfs *osfs, time64_t max_age, __u32 flags);
int (*o_statfs_async)(struct obd_export *exp, struct obd_info *oinfo,
- __u64 max_age, struct ptlrpc_request_set *set);
+ time64_t max_age, struct ptlrpc_request_set *set);
int (*o_create)(const struct lu_env *env, struct obd_export *exp,
struct obdo *oa);
int (*o_destroy)(const struct lu_env *env, struct obd_export *exp,
int (*o_quotactl)(struct obd_device *, struct obd_export *,
struct obd_quotactl *);
- int (*o_ping)(const struct lu_env *, struct obd_export *exp);
-
/* pools methods */
int (*o_pool_new)(struct obd_device *obd, char *poolname);
int (*o_pool_del)(struct obd_device *obd, char *poolname);
char *ostname);
int (*o_pool_rem)(struct obd_device *obd, char *poolname,
char *ostname);
- void (*o_getref)(struct obd_device *obd);
- void (*o_putref)(struct obd_device *obd);
- /*
- * NOTE: If adding ops, add another LPROCFS_OBD_OP_INIT() line
- * to lprocfs_alloc_obd_stats() in obdclass/lprocfs_status.c.
- * Also, add a wrapper function in include/linux/obd_class.h. */
};
/* lmv structures */
};
struct obd_client_handle {
- struct lustre_handle och_fh;
+ struct lustre_handle och_open_handle;
struct lu_fid och_fid;
struct md_open_data *och_mod;
struct lustre_handle och_lease_handle; /* open lock for lease */
struct cl_attr;
struct md_ops {
- /* Every operation from MD_STATS_FIRST_OP up to and including
- * MD_STATS_LAST_OP will be counted by EXP_MD_OP_INCREMENT()
- * and will appear in /proc/fs/lustre/{lmv,mdc}/.../md_stats.
- * Operations after MD_STATS_LAST_OP are excluded from stats.
- * There are a few reasons for doing this: we prune the 17
- * counters which will be of minimal use in understanding
- * metadata utilization, we save memory by allocating 15
- * instead of 32 counters, we save cycles by not counting.
- *
- * MD_STATS_FIRST_OP must be the first member of md_ops.
- */
-#define MD_STATS_FIRST_OP m_close
int (*m_close)(struct obd_export *, struct md_op_data *,
struct md_open_data *, struct ptlrpc_request **);
struct ptlrpc_request **);
int (*m_setxattr)(struct obd_export *, const struct lu_fid *,
- u64, const char *, const char *, int, int, int, u32,
- struct ptlrpc_request **);
+ u64, const char *, const void *, size_t, unsigned int,
+ u32, struct ptlrpc_request **);
int (*m_getxattr)(struct obd_export *, const struct lu_fid *,
- u64, const char *, const char *, int, int, int,
- struct ptlrpc_request **);
+ u64, const char *, size_t, struct ptlrpc_request **);
int (*m_intent_getattr_async)(struct obd_export *,
struct md_enqueue_info *);
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_file_resync)(struct obd_export *, struct md_op_data *);
int (*m_get_root)(struct obd_export *, const char *, struct lu_fid *);
int (*m_null_inode)(struct obd_export *, const struct lu_fid *);
return obd->u.cli.cl_max_pages_per_rpc << PAGE_SHIFT;
}
-/* when RPC size or the max RPCs in flight is increased, the max dirty pages
+/*
+ * 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_SHIFT;
- else {
+ if (cli->cl_dirty_max_pages <= 0) {
+ cli->cl_dirty_max_pages =
+ (OSC_MAX_DIRTY_DEFAULT * 1024 * 1024) >> PAGE_SHIFT;
+ } else {
unsigned long dirty_max = cli->cl_max_rpcs_in_flight *
cli->cl_max_pages_per_rpc;
if (cli->cl_dirty_max_pages > totalram_pages / 8)
cli->cl_dirty_max_pages = totalram_pages / 8;
+
+ /* This value is exported to userspace through the max_dirty_mb
+ * parameter. So we round up the number of pages to make it a round
+ * number of MBs. */
+ cli->cl_dirty_max_pages = round_up(cli->cl_dirty_max_pages,
+ 1 << (20 - PAGE_SHIFT));
}
#endif /* __OBD_H */
git://git.whamcloud.com / fs / lustre-release.git / blobdiff