/* usual integer division */
#define div_floor(A, B) ((A)/(B))
-/* drbd_meta-data.c (still in drbd_main.c) */
-/* 4th incarnation of the disk layout. */
-#define DRBD_MD_MAGIC (DRBD_MAGIC+4)
-
extern struct ratelimit_state drbd_ratelimit_state;
extern struct idr minors; /* RCU, updates: genl_lock() */
extern struct list_head drbd_tconns; /* RCU, updates: genl_lock() */
u32 bit_map_gen[5];
} __packed;
-struct p_discard {
- u64 block_id;
- u32 seq_num;
- u32 pad;
-} __packed;
-
struct p_block_desc {
u64 sector;
u32 blksize;
struct bio *private_bio;
struct drbd_interval i;
- unsigned int epoch; /* barrier_nr */
- /* barrier_nr: used to check on "completion" whether this req was in
+ /* epoch: used to check on "completion" whether this req was in
* the current epoch, and we therefore have to close it,
- * starting a new epoch...
+ * causing a p_barrier packet to be send, starting a new epoch.
+ *
+ * This corresponds to "barrier" in struct p_barrier[_ack],
+ * and to "barrier_nr" in struct drbd_epoch (and various
+ * comments/function parameters/local variable names).
*/
+ unsigned int epoch;
struct list_head tl_requests; /* ring list in the transfer log */
struct bio *master_bio; /* master bio pointer */
- unsigned long rq_state; /* see comments above _req_mod() */
- int seq_num;
unsigned long start_time;
-};
-struct drbd_tl_epoch {
- struct drbd_work w;
- struct list_head requests; /* requests before */
- struct drbd_tl_epoch *next; /* pointer to the next barrier */
- unsigned int br_number; /* the barriers identifier. */
- int n_writes; /* number of requests attached before this barrier */
+ /* once it hits 0, we may complete the master_bio */
+ atomic_t completion_ref;
+ /* once it hits 0, we may destroy this drbd_request object */
+ struct kref kref;
+
+ unsigned rq_state; /* see comments above _req_mod() */
};
struct drbd_epoch {
+ struct drbd_tconn *tconn;
struct list_head list;
unsigned int barrier_nr;
atomic_t epoch_size; /* increased on every request added. */
/* flag bits per mdev */
enum {
- CREATE_BARRIER, /* next P_DATA is preceded by a P_BARRIER */
- UNPLUG_QUEUED, /* only relevant with kernel 2.4 */
UNPLUG_REMOTE, /* sending a "UnplugRemote" could help */
MD_DIRTY, /* current uuids and flags not yet on disk */
USE_DEGR_WFC_T, /* degr-wfc-timeout instead of wfc-timeout. */
CRASHED_PRIMARY, /* This node was a crashed primary.
* Gets cleared when the state.conn
* goes into C_CONNECTED state. */
- NO_BARRIER_SUPP, /* underlying block device doesn't implement barriers */
CONSIDER_RESYNC,
MD_NO_FUA, /* Users wants us to not use FUA/FLUSH on meta data dev */
BITMAP_IO_QUEUED, /* Started bitmap IO */
GO_DISKLESS, /* Disk is being detached, on io-error or admin request. */
WAS_IO_ERROR, /* Local disk failed returned IO error */
+ FORCE_DETACH, /* Force-detach from local disk, aborting any pending local IO */
RESYNC_AFTER_NEG, /* Resync after online grow after the attach&negotiate finished. */
RESIZE_PENDING, /* Size change detected locally, waiting for the response from
* the peer, if it changed there as well. */
AL_SUSPENDED, /* Activity logging is currently suspended. */
AHEAD_TO_SYNC_SOURCE, /* Ahead -> SyncSource queued */
B_RS_H_DONE, /* Before resync handler done (already executed) */
+ DISCARD_MY_DATA, /* discard_my_data flag per volume */
+ READ_BALANCE_RR,
};
struct drbd_bitmap; /* opaque for drbd_conf */
BM_P_VMALLOCED = 0x10000, /* internal use only, will be masked out */
/* currently locked for bulk operation */
- BM_LOCKED_MASK = 0x7,
+ BM_LOCKED_MASK = 0xf,
/* in detail, that is: */
BM_DONT_CLEAR = 0x1,
BM_DONT_SET = 0x2,
BM_DONT_TEST = 0x4,
+ /* so we can mark it locked for bulk operation,
+ * and still allow all non-bulk operations */
+ BM_IS_LOCKED = 0x8,
+
/* (test bit, count bit) allowed (common case) */
- BM_LOCKED_TEST_ALLOWED = 0x3,
+ BM_LOCKED_TEST_ALLOWED = BM_DONT_CLEAR | BM_DONT_SET | BM_IS_LOCKED,
/* testing bits, as well as setting new bits allowed, but clearing bits
* would be unexpected. Used during bitmap receive. Setting new bits
* requires sending of "out-of-sync" information, though. */
- BM_LOCKED_SET_ALLOWED = 0x1,
+ BM_LOCKED_SET_ALLOWED = BM_DONT_CLEAR | BM_IS_LOCKED,
- /* clear is not expected while bitmap is locked for bulk operation */
+ /* for drbd_bm_write_copy_pages, everything is allowed,
+ * only concurrent bulk operations are locked out. */
+ BM_LOCKED_CHANGE_ALLOWED = BM_IS_LOCKED,
};
-
-/* TODO sort members for performance
- * MAYBE group them further */
-
-/* THINK maybe we actually want to use the default "event/%s" worker threads
- * or similar in linux 2.6, which uses per cpu data and threads.
- */
struct drbd_work_queue {
struct list_head q;
- struct semaphore s; /* producers up it, worker down()s it */
spinlock_t q_lock; /* to protect the list. */
+ wait_queue_head_t q_wait;
};
struct drbd_socket {
- struct drbd_work_queue work;
struct mutex mutex;
struct socket *socket;
/* this way we get our
};
struct drbd_md_io {
- struct drbd_conf *mdev;
- struct completion event;
+ unsigned int done;
int error;
};
SEND_PING, /* whether asender should send a ping asap */
SIGNAL_ASENDER, /* whether asender wants to be interrupted */
GOT_PING_ACK, /* set when we receive a ping_ack packet, ping_wait gets woken */
+ CONN_WD_ST_CHG_REQ, /* A cluster wide state change on the connection is active */
CONN_WD_ST_CHG_OKAY,
CONN_WD_ST_CHG_FAIL,
CONN_DRY_RUN, /* Expect disconnect after resync handshake. */
+ CREATE_BARRIER, /* next P_DATA is preceded by a P_BARRIER */
+ STATE_SENT, /* Do not change state/UUIDs while this is set */
+ CALLBACK_PENDING, /* Whether we have a call_usermodehelper(, UMH_WAIT_PROC)
+ * pending, from drbd worker context.
+ * If set, bdi_write_congested() returns true,
+ * so shrink_page_list() would not recurse into,
+ * and potentially deadlock on, this drbd worker.
+ */
};
struct drbd_tconn { /* is a resource from the config file */
unsigned int ko_count;
spinlock_t req_lock;
- struct drbd_tl_epoch *unused_spare_tle; /* for pre-allocation */
- struct drbd_tl_epoch *newest_tle;
- struct drbd_tl_epoch *oldest_tle;
- struct list_head out_of_sequence_requests;
+
+ struct list_head transfer_log; /* all requests not yet fully processed */
struct crypto_hash *cram_hmac_tfm;
struct crypto_hash *integrity_tfm; /* checksums we compute, updates protected by tconn->data->mutex */
void *int_dig_in;
void *int_dig_vv;
+ /* receiver side */
+ struct drbd_epoch *current_epoch;
+ spinlock_t epoch_lock;
+ unsigned int epochs;
+ enum write_ordering_e write_ordering;
+ atomic_t current_tle_nr; /* transfer log epoch number */
+ unsigned current_tle_writes; /* writes seen within this tl epoch */
+
+ unsigned long last_reconnect_jif;
struct drbd_thread receiver;
struct drbd_thread worker;
struct drbd_thread asender;
cpumask_var_t cpu_mask;
+
+ /* sender side */
+ struct drbd_work_queue sender_work;
+
+ struct {
+ /* whether this sender thread
+ * has processed a single write yet. */
+ bool seen_any_write_yet;
+
+ /* Which barrier number to send with the next P_BARRIER */
+ int current_epoch_nr;
+
+ /* how many write requests have been sent
+ * with req->epoch == current_epoch_nr.
+ * If none, no P_BARRIER will be sent. */
+ unsigned current_epoch_writes;
+ } send;
};
struct drbd_conf {
struct block_device *this_bdev;
struct gendisk *vdisk;
+ unsigned long last_reattach_jif;
struct drbd_work resync_work,
unplug_work,
go_diskless,
/* where does the admin want us to start? (sector) */
sector_t ov_start_sector;
+ sector_t ov_stop_sector;
/* where are we now? (sector) */
sector_t ov_position;
/* Start sector of out of sync range (to merge printk reporting). */
int open_cnt;
u64 *p_uuid;
- struct drbd_epoch *current_epoch;
- spinlock_t epoch_lock;
- unsigned int epochs;
- enum write_ordering_e write_ordering;
+
struct list_head active_ee; /* IO in progress (P_DATA gets written to disk) */
struct list_head sync_ee; /* IO in progress (P_RS_DATA_REPLY gets written to disk) */
struct list_head done_ee; /* need to send P_WRITE_ACK */
atomic_t pp_in_use_by_net; /* sendpage()d, still referenced by tcp */
wait_queue_head_t ee_wait;
struct page *md_io_page; /* one page buffer for md_io */
- struct mutex md_io_mutex; /* protects the md_io_buffer */
+ struct drbd_md_io md_io;
+ atomic_t md_io_in_use; /* protects the md_io, md_io_page and md_io_tmpp */
spinlock_t al_lock;
wait_queue_head_t al_wait;
struct lru_cache *act_log; /* activity log */
extern void tl_release(struct drbd_tconn *, unsigned int barrier_nr,
unsigned int set_size);
extern void tl_clear(struct drbd_tconn *);
-extern void _tl_add_barrier(struct drbd_tconn *, struct drbd_tl_epoch *);
extern void drbd_free_sock(struct drbd_tconn *tconn);
extern int drbd_send(struct drbd_tconn *tconn, struct socket *sock,
void *buf, size_t size, unsigned msg_flags);
extern int drbd_send_uuids_skip_initial_sync(struct drbd_conf *mdev);
extern void drbd_gen_and_send_sync_uuid(struct drbd_conf *mdev);
extern int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags flags);
-extern int _conn_send_state_req(struct drbd_tconn *, int vnr, enum drbd_packet cmd,
- union drbd_state, union drbd_state);
-extern int _drbd_send_state(struct drbd_conf *mdev);
-extern int drbd_send_state(struct drbd_conf *mdev);
+extern int drbd_send_state(struct drbd_conf *mdev, union drbd_state s);
+extern int drbd_send_current_state(struct drbd_conf *mdev);
extern int drbd_send_sync_param(struct drbd_conf *mdev);
-extern void drbd_send_b_ack(struct drbd_conf *mdev, u32 barrier_nr,
+extern void drbd_send_b_ack(struct drbd_tconn *tconn, u32 barrier_nr,
u32 set_size);
extern int drbd_send_ack(struct drbd_conf *, enum drbd_packet,
struct drbd_peer_request *);
extern void drbd_bm_mark_for_writeout(struct drbd_conf *mdev, int page_nr);
extern int drbd_bm_write(struct drbd_conf *mdev) __must_hold(local);
extern int drbd_bm_write_hinted(struct drbd_conf *mdev) __must_hold(local);
-extern unsigned long drbd_bm_ALe_set_all(struct drbd_conf *mdev,
- unsigned long al_enr);
+extern int drbd_bm_write_copy_pages(struct drbd_conf *mdev) __must_hold(local);
extern size_t drbd_bm_words(struct drbd_conf *mdev);
extern unsigned long drbd_bm_bits(struct drbd_conf *mdev);
extern sector_t drbd_bm_capacity(struct drbd_conf *mdev);
extern int proc_details;
/* drbd_req */
-extern int __drbd_make_request(struct drbd_conf *, struct bio *, unsigned long);
+extern void __drbd_make_request(struct drbd_conf *, struct bio *, unsigned long);
extern int drbd_make_request(struct request_queue *q, struct bio *bio);
extern int drbd_read_remote(struct drbd_conf *mdev, struct drbd_request *req);
extern int drbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bvm, struct bio_vec *bvec);
extern void suspend_other_sg(struct drbd_conf *mdev);
extern int drbd_resync_finished(struct drbd_conf *mdev);
/* maybe rather drbd_main.c ? */
+extern void *drbd_md_get_buffer(struct drbd_conf *mdev);
+extern void drbd_md_put_buffer(struct drbd_conf *mdev);
extern int drbd_md_sync_page_io(struct drbd_conf *mdev,
struct drbd_backing_dev *bdev, sector_t sector, int rw);
extern void drbd_ov_out_of_sync_found(struct drbd_conf *, sector_t, int);
+extern void wait_until_done_or_disk_failure(struct drbd_conf *mdev, struct drbd_backing_dev *bdev,
+ unsigned int *done);
extern void drbd_rs_controller_reset(struct drbd_conf *mdev);
static inline void ov_out_of_sync_print(struct drbd_conf *mdev)
extern void drbd_csum_ee(struct drbd_conf *, struct crypto_hash *,
struct drbd_peer_request *, void *);
/* worker callbacks */
-extern int w_read_retry_remote(struct drbd_work *, int);
extern int w_e_end_data_req(struct drbd_work *, int);
extern int w_e_end_rsdata_req(struct drbd_work *, int);
extern int w_e_end_csum_rs_req(struct drbd_work *, int);
extern int w_send_write_hint(struct drbd_work *, int);
extern int w_make_resync_request(struct drbd_work *, int);
extern int w_send_dblock(struct drbd_work *, int);
-extern int w_send_barrier(struct drbd_work *, int);
extern int w_send_read_req(struct drbd_work *, int);
extern int w_prev_work_done(struct drbd_work *, int);
extern int w_e_reissue(struct drbd_work *, int);
(char*)&val, sizeof(val));
}
-void drbd_bump_write_ordering(struct drbd_conf *mdev, enum write_ordering_e wo);
+void drbd_bump_write_ordering(struct drbd_tconn *tconn, enum write_ordering_e wo);
/* drbd_proc.c */
extern struct proc_dir_entry *drbd_proc;
extern int drbd_rs_del_all(struct drbd_conf *mdev);
extern void drbd_rs_failed_io(struct drbd_conf *mdev,
sector_t sector, int size);
-extern int drbd_al_read_log(struct drbd_conf *mdev, struct drbd_backing_dev *);
extern void drbd_advance_rs_marks(struct drbd_conf *mdev, unsigned long still_to_go);
extern void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector,
int size, const char *file, const unsigned int line);
int size, const char *file, const unsigned int line);
#define drbd_set_out_of_sync(mdev, sector, size) \
__drbd_set_out_of_sync(mdev, sector, size, __FILE__, __LINE__)
-extern void drbd_al_apply_to_bm(struct drbd_conf *mdev);
extern void drbd_al_shrink(struct drbd_conf *mdev);
/* drbd_nl.c */
#define page_chain_for_each_safe(page, n) \
for (; page && ({ n = page_chain_next(page); 1; }); page = n)
-static inline int drbd_bio_has_active_page(struct bio *bio)
-{
- struct bio_vec *bvec;
- int i;
-
- __bio_for_each_segment(bvec, bio, i, 0) {
- if (page_count(bvec->bv_page) > 1)
- return 1;
- }
-
- return 0;
-}
static inline int drbd_peer_req_has_active_page(struct drbd_peer_request *peer_req)
{
return rv;
}
+enum drbd_force_detach_flags {
+ DRBD_IO_ERROR,
+ DRBD_META_IO_ERROR,
+ DRBD_FORCE_DETACH,
+};
+
#define __drbd_chk_io_error(m,f) __drbd_chk_io_error_(m,f, __func__)
-static inline void __drbd_chk_io_error_(struct drbd_conf *mdev, int forcedetach, const char *where)
+static inline void __drbd_chk_io_error_(struct drbd_conf *mdev,
+ enum drbd_force_detach_flags forcedetach,
+ const char *where)
{
enum drbd_io_error_p ep;
rcu_read_unlock();
switch (ep) {
case EP_PASS_ON: /* FIXME would this be better named "Ignore"? */
- if (!forcedetach) {
+ if (forcedetach == DRBD_IO_ERROR) {
if (__ratelimit(&drbd_ratelimit_state))
dev_err(DEV, "Local IO failed in %s.\n", where);
if (mdev->state.disk > D_INCONSISTENT)
case EP_DETACH:
case EP_CALL_HELPER:
set_bit(WAS_IO_ERROR, &mdev->flags);
+ if (forcedetach == DRBD_FORCE_DETACH)
+ set_bit(FORCE_DETACH, &mdev->flags);
if (mdev->state.disk > D_FAILED) {
_drbd_set_state(_NS(mdev, disk, D_FAILED), CS_HARD, NULL);
dev_err(DEV,
*/
#define drbd_chk_io_error(m,e,f) drbd_chk_io_error_(m,e,f, __func__)
static inline void drbd_chk_io_error_(struct drbd_conf *mdev,
- int error, int forcedetach, const char *where)
+ int error, enum drbd_force_detach_flags forcedetach, const char *where)
{
if (error) {
unsigned long flags;
unsigned long flags;
spin_lock_irqsave(&q->q_lock, flags);
list_add(&w->list, &q->q);
- up(&q->s); /* within the spinlock,
- see comment near end of drbd_worker() */
spin_unlock_irqrestore(&q->q_lock, flags);
+ wake_up(&q->q_wait);
}
static inline void
unsigned long flags;
spin_lock_irqsave(&q->q_lock, flags);
list_add_tail(&w->list, &q->q);
- up(&q->s); /* within the spinlock,
- see comment near end of drbd_worker() */
spin_unlock_irqrestore(&q->q_lock, flags);
+ wake_up(&q->q_wait);
}
static inline void wake_asender(struct drbd_tconn *tconn)
case D_OUTDATED:
case D_CONSISTENT:
case D_UP_TO_DATE:
+ case D_FAILED:
/* disk state is stable as well. */
break;
/* no new io accepted during transitional states */
case D_ATTACHING:
- case D_FAILED:
case D_NEGOTIATING:
case D_UNKNOWN:
case D_MASK:
int ap_bio = atomic_dec_return(&mdev->ap_bio_cnt);
D_ASSERT(ap_bio >= 0);
+
+ if (ap_bio == 0 && test_bit(BITMAP_IO, &mdev->flags)) {
+ if (!test_and_set_bit(BITMAP_IO_QUEUED, &mdev->flags))
+ drbd_queue_work(&mdev->tconn->sender_work, &mdev->bm_io_work.w);
+ }
+
/* this currently does wake_up for every dec_ap_bio!
* maybe rather introduce some type of hysteresis?
* e.g. (ap_bio == mxb/2 || ap_bio == 0) ? */
if (ap_bio < mxb)
wake_up(&mdev->misc_wait);
- if (ap_bio == 0 && test_bit(BITMAP_IO, &mdev->flags)) {
- if (!test_and_set_bit(BITMAP_IO_QUEUED, &mdev->flags))
- drbd_queue_work(&mdev->tconn->data.work, &mdev->bm_io_work.w);
- }
+}
+
+static inline bool verify_can_do_stop_sector(struct drbd_conf *mdev)
+{
+ return mdev->tconn->agreed_pro_version >= 97 &&
+ mdev->tconn->agreed_pro_version != 100;
}
static inline int drbd_set_ed_uuid(struct drbd_conf *mdev, u64 val)