*/
struct idr minors;
struct list_head drbd_tconns; /* list of struct drbd_tconn */
-DEFINE_MUTEX(drbd_cfg_mutex);
struct kmem_cache *drbd_request_cache;
struct kmem_cache *drbd_ee_cache; /* peer requests */
tconn->oldest_tle = b;
tconn->newest_tle = b;
INIT_LIST_HEAD(&tconn->out_of_sequence_requests);
+ INIT_LIST_HEAD(&tconn->barrier_acked_requests);
return 1;
}
These have been list_move'd to the out_of_sequence_requests list in
_req_mod(, BARRIER_ACKED) above.
*/
- list_del_init(&b->requests);
+ list_splice_init(&b->requests, &tconn->barrier_acked_requests);
mdev = b->w.mdev;
nob = b->next;
- if (test_and_clear_bit(CREATE_BARRIER, &mdev->flags)) {
+ if (test_and_clear_bit(CREATE_BARRIER, &tconn->flags)) {
_tl_add_barrier(tconn, b);
if (nob)
tconn->oldest_tle = nob;
req = list_entry(le, struct drbd_request, tl_requests);
rv = _req_mod(req, what);
- n_writes += (rv & MR_WRITE) >> MR_WRITE_SHIFT;
- n_reads += (rv & MR_READ) >> MR_READ_SHIFT;
+ if (rv & MR_WRITE)
+ n_writes++;
+ if (rv & MR_READ)
+ n_reads++;
}
tmp = b->next;
if (b->w.cb == NULL) {
b->w.cb = w_send_barrier;
inc_ap_pending(b->w.mdev);
- set_bit(CREATE_BARRIER, &b->w.mdev->flags);
+ set_bit(CREATE_BARRIER, &tconn->flags);
}
drbd_queue_work(&tconn->data.work, &b->w);
b = tmp;
list_splice(&carry_reads, &b->requests);
}
-}
+ /* Actions operating on the disk state, also want to work on
+ requests that got barrier acked. */
+ switch (what) {
+ case FAIL_FROZEN_DISK_IO:
+ case RESTART_FROZEN_DISK_IO:
+ list_for_each_safe(le, tle, &tconn->barrier_acked_requests) {
+ req = list_entry(le, struct drbd_request, tl_requests);
+ _req_mod(req, what);
+ }
+ case CONNECTION_LOST_WHILE_PENDING:
+ case RESEND:
+ break;
+ default:
+ conn_err(tconn, "what = %d in _tl_restart()\n", what);
+ }
+}
/**
* tl_clear() - Clears all requests and &struct drbd_tl_epoch objects out of the TL
*/
void tl_clear(struct drbd_tconn *tconn)
{
- struct drbd_conf *mdev;
struct list_head *le, *tle;
struct drbd_request *r;
- int vnr;
spin_lock_irq(&tconn->req_lock);
}
/* ensure bit indicating barrier is required is clear */
- idr_for_each_entry(&tconn->volumes, mdev, vnr)
- clear_bit(CREATE_BARRIER, &mdev->flags);
+ clear_bit(CREATE_BARRIER, &tconn->flags);
spin_unlock_irq(&tconn->req_lock);
}
spin_unlock_irq(&tconn->req_lock);
}
+/**
+ * tl_abort_disk_io() - Abort disk I/O for all requests for a certain mdev in the TL
+ * @mdev: DRBD device.
+ */
+void tl_abort_disk_io(struct drbd_conf *mdev)
+{
+ struct drbd_tconn *tconn = mdev->tconn;
+ struct drbd_tl_epoch *b;
+ struct list_head *le, *tle;
+ struct drbd_request *req;
+
+ spin_lock_irq(&tconn->req_lock);
+ b = tconn->oldest_tle;
+ while (b) {
+ list_for_each_safe(le, tle, &b->requests) {
+ req = list_entry(le, struct drbd_request, tl_requests);
+ if (!(req->rq_state & RQ_LOCAL_PENDING))
+ continue;
+ if (req->w.mdev == mdev)
+ _req_mod(req, ABORT_DISK_IO);
+ }
+ b = b->next;
+ }
+
+ list_for_each_safe(le, tle, &tconn->barrier_acked_requests) {
+ req = list_entry(le, struct drbd_request, tl_requests);
+ if (!(req->rq_state & RQ_LOCAL_PENDING))
+ continue;
+ if (req->w.mdev == mdev)
+ _req_mod(req, ABORT_DISK_IO);
+ }
+
+ spin_unlock_irq(&tconn->req_lock);
+}
+
static int drbd_thread_setup(void *arg)
{
struct drbd_thread *thi = (struct drbd_thread *) arg;
thi->task = NULL;
thi->t_state = NONE;
smp_mb();
- complete(&thi->stop);
+ complete_all(&thi->stop);
spin_unlock_irqrestore(&thi->t_lock, flags);
conn_info(tconn, "Terminating %s\n", current->comm);
/* Release mod reference taken when thread was started */
+
+ kref_put(&tconn->kref, &conn_destroy);
module_put(THIS_MODULE);
return retval;
}
return false;
}
+ kref_get(&thi->tconn->kref);
+
init_completion(&thi->stop);
thi->reset_cpu_mask = 1;
thi->t_state = RUNNING;
if (IS_ERR(nt)) {
conn_err(tconn, "Couldn't start thread\n");
+ kref_put(&tconn->kref, &conn_destroy);
module_put(THIS_MODULE);
return false;
}
int conn_lowest_minor(struct drbd_tconn *tconn)
{
- int vnr = 0;
struct drbd_conf *mdev;
+ int vnr = 0, m;
+ rcu_read_lock();
mdev = idr_get_next(&tconn->volumes, &vnr);
- if (!mdev)
- return -1;
- return mdev_to_minor(mdev);
+ m = mdev ? mdev_to_minor(mdev) : -1;
+ rcu_read_unlock();
+
+ return m;
}
#ifdef CONFIG_SMP
return prepare_header80(buffer, cmd, size);
}
+static void *__conn_prepare_command(struct drbd_tconn *tconn,
+ struct drbd_socket *sock)
+{
+ if (!sock->socket)
+ return NULL;
+ return sock->sbuf + drbd_header_size(tconn);
+}
+
void *conn_prepare_command(struct drbd_tconn *tconn, struct drbd_socket *sock)
{
+ void *p;
+
mutex_lock(&sock->mutex);
- if (!sock->socket) {
+ p = __conn_prepare_command(tconn, sock);
+ if (!p)
mutex_unlock(&sock->mutex);
- return NULL;
- }
- return sock->sbuf + drbd_header_size(tconn);
+
+ return p;
}
void *drbd_prepare_command(struct drbd_conf *mdev, struct drbd_socket *sock)
return err;
}
+static int __conn_send_command(struct drbd_tconn *tconn, struct drbd_socket *sock,
+ enum drbd_packet cmd, unsigned int header_size,
+ void *data, unsigned int size)
+{
+ return __send_command(tconn, 0, sock, cmd, header_size, data, size);
+}
+
int conn_send_command(struct drbd_tconn *tconn, struct drbd_socket *sock,
enum drbd_packet cmd, unsigned int header_size,
void *data, unsigned int size)
{
int err;
- err = __send_command(tconn, 0, sock, cmd, header_size, data, size);
+ err = __conn_send_command(tconn, sock, cmd, header_size, data, size);
mutex_unlock(&sock->mutex);
return err;
}
int size;
const int apv = mdev->tconn->agreed_pro_version;
enum drbd_packet cmd;
+ struct net_conf *nc;
+ struct disk_conf *dc;
sock = &mdev->tconn->data;
p = drbd_prepare_command(mdev, sock);
if (!p)
return -EIO;
+ rcu_read_lock();
+ nc = rcu_dereference(mdev->tconn->net_conf);
+
size = apv <= 87 ? sizeof(struct p_rs_param)
: apv == 88 ? sizeof(struct p_rs_param)
- + strlen(mdev->tconn->net_conf->verify_alg) + 1
+ + strlen(nc->verify_alg) + 1
: apv <= 94 ? sizeof(struct p_rs_param_89)
: /* apv >= 95 */ sizeof(struct p_rs_param_95);
memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX);
if (get_ldev(mdev)) {
- p->rate = cpu_to_be32(mdev->ldev->dc.resync_rate);
- p->c_plan_ahead = cpu_to_be32(mdev->ldev->dc.c_plan_ahead);
- p->c_delay_target = cpu_to_be32(mdev->ldev->dc.c_delay_target);
- p->c_fill_target = cpu_to_be32(mdev->ldev->dc.c_fill_target);
- p->c_max_rate = cpu_to_be32(mdev->ldev->dc.c_max_rate);
+ dc = rcu_dereference(mdev->ldev->disk_conf);
+ p->resync_rate = cpu_to_be32(dc->resync_rate);
+ p->c_plan_ahead = cpu_to_be32(dc->c_plan_ahead);
+ p->c_delay_target = cpu_to_be32(dc->c_delay_target);
+ p->c_fill_target = cpu_to_be32(dc->c_fill_target);
+ p->c_max_rate = cpu_to_be32(dc->c_max_rate);
put_ldev(mdev);
} else {
- p->rate = cpu_to_be32(DRBD_RATE_DEF);
+ p->resync_rate = cpu_to_be32(DRBD_RESYNC_RATE_DEF);
p->c_plan_ahead = cpu_to_be32(DRBD_C_PLAN_AHEAD_DEF);
p->c_delay_target = cpu_to_be32(DRBD_C_DELAY_TARGET_DEF);
p->c_fill_target = cpu_to_be32(DRBD_C_FILL_TARGET_DEF);
}
if (apv >= 88)
- strcpy(p->verify_alg, mdev->tconn->net_conf->verify_alg);
+ strcpy(p->verify_alg, nc->verify_alg);
if (apv >= 89)
- strcpy(p->csums_alg, mdev->tconn->net_conf->csums_alg);
+ strcpy(p->csums_alg, nc->csums_alg);
+ rcu_read_unlock();
return drbd_send_command(mdev, sock, cmd, size, NULL, 0);
}
-int drbd_send_protocol(struct drbd_tconn *tconn)
+int __drbd_send_protocol(struct drbd_tconn *tconn, enum drbd_packet cmd)
{
struct drbd_socket *sock;
struct p_protocol *p;
+ struct net_conf *nc;
int size, cf;
- if (tconn->net_conf->dry_run && tconn->agreed_pro_version < 92) {
- conn_err(tconn, "--dry-run is not supported by peer");
- return -EOPNOTSUPP;
- }
-
sock = &tconn->data;
- p = conn_prepare_command(tconn, sock);
+ p = __conn_prepare_command(tconn, sock);
if (!p)
return -EIO;
+ rcu_read_lock();
+ nc = rcu_dereference(tconn->net_conf);
+
+ if (nc->tentative && tconn->agreed_pro_version < 92) {
+ rcu_read_unlock();
+ mutex_unlock(&sock->mutex);
+ conn_err(tconn, "--dry-run is not supported by peer");
+ return -EOPNOTSUPP;
+ }
+
size = sizeof(*p);
if (tconn->agreed_pro_version >= 87)
- size += strlen(tconn->net_conf->integrity_alg) + 1;
+ size += strlen(nc->integrity_alg) + 1;
- p->protocol = cpu_to_be32(tconn->net_conf->wire_protocol);
- p->after_sb_0p = cpu_to_be32(tconn->net_conf->after_sb_0p);
- p->after_sb_1p = cpu_to_be32(tconn->net_conf->after_sb_1p);
- p->after_sb_2p = cpu_to_be32(tconn->net_conf->after_sb_2p);
- p->two_primaries = cpu_to_be32(tconn->net_conf->two_primaries);
+ p->protocol = cpu_to_be32(nc->wire_protocol);
+ p->after_sb_0p = cpu_to_be32(nc->after_sb_0p);
+ p->after_sb_1p = cpu_to_be32(nc->after_sb_1p);
+ p->after_sb_2p = cpu_to_be32(nc->after_sb_2p);
+ p->two_primaries = cpu_to_be32(nc->two_primaries);
cf = 0;
- if (tconn->net_conf->want_lose)
- cf |= CF_WANT_LOSE;
- if (tconn->net_conf->dry_run)
+ if (nc->discard_my_data)
+ cf |= CF_DISCARD_MY_DATA;
+ if (nc->tentative)
cf |= CF_DRY_RUN;
p->conn_flags = cpu_to_be32(cf);
if (tconn->agreed_pro_version >= 87)
- strcpy(p->integrity_alg, tconn->net_conf->integrity_alg);
- return conn_send_command(tconn, sock, P_PROTOCOL, size, NULL, 0);
+ strcpy(p->integrity_alg, nc->integrity_alg);
+ rcu_read_unlock();
+
+ return __conn_send_command(tconn, sock, cmd, size, NULL, 0);
+}
+
+int drbd_send_protocol(struct drbd_tconn *tconn)
+{
+ int err;
+
+ mutex_lock(&tconn->data.mutex);
+ err = __drbd_send_protocol(tconn, P_PROTOCOL);
+ mutex_unlock(&tconn->data.mutex);
+
+ return err;
}
int _drbd_send_uuids(struct drbd_conf *mdev, u64 uuid_flags)
mdev->comm_bm_set = drbd_bm_total_weight(mdev);
p->uuid[UI_SIZE] = cpu_to_be64(mdev->comm_bm_set);
- uuid_flags |= mdev->tconn->net_conf->want_lose ? 1 : 0;
+ rcu_read_lock();
+ uuid_flags |= rcu_dereference(mdev->tconn->net_conf)->discard_my_data ? 1 : 0;
+ rcu_read_unlock();
uuid_flags |= test_bit(CRASHED_PRIMARY, &mdev->flags) ? 2 : 0;
uuid_flags |= mdev->new_state_tmp.disk == D_INCONSISTENT ? 4 : 0;
p->uuid[UI_FLAGS] = cpu_to_be64(uuid_flags);
D_ASSERT(mdev->state.disk == D_UP_TO_DATE);
- uuid = mdev->ldev->md.uuid[UI_BITMAP] + UUID_NEW_BM_OFFSET;
+ uuid = mdev->ldev->md.uuid[UI_BITMAP];
+ if (uuid && uuid != UUID_JUST_CREATED)
+ uuid = uuid + UUID_NEW_BM_OFFSET;
+ else
+ get_random_bytes(&uuid, sizeof(u64));
drbd_uuid_set(mdev, UI_BITMAP, uuid);
drbd_print_uuids(mdev, "updated sync UUID");
drbd_md_sync(mdev);
if (get_ldev_if_state(mdev, D_NEGOTIATING)) {
D_ASSERT(mdev->ldev->backing_bdev);
d_size = drbd_get_max_capacity(mdev->ldev);
- u_size = mdev->ldev->dc.disk_size;
+ rcu_read_lock();
+ u_size = rcu_dereference(mdev->ldev->disk_conf)->disk_size;
+ rcu_read_unlock();
q_order_type = drbd_queue_order_type(mdev);
max_bio_size = queue_max_hw_sectors(mdev->ldev->backing_bdev->bd_disk->queue) << 9;
max_bio_size = min_t(int, max_bio_size, DRBD_MAX_BIO_SIZE);
p = drbd_prepare_command(mdev, sock);
if (!p)
return -EIO;
+
+ if (mdev->tconn->agreed_pro_version <= 94)
+ max_bio_size = min_t(int, max_bio_size, DRBD_MAX_SIZE_H80_PACKET);
+ else if (mdev->tconn->agreed_pro_version < 100)
+ max_bio_size = min_t(int, max_bio_size, DRBD_MAX_BIO_SIZE_P95);
+
p->d_size = cpu_to_be64(d_size);
p->u_size = cpu_to_be64(u_size);
p->c_size = cpu_to_be64(trigger_reply ? 0 : drbd_get_capacity(mdev->this_bdev));
}
/**
- * drbd_send_state() - Sends the drbd state to the peer
+ * drbd_send_current_state() - Sends the drbd state to the peer
* @mdev: DRBD device.
*/
-int drbd_send_state(struct drbd_conf *mdev)
+int drbd_send_current_state(struct drbd_conf *mdev)
{
struct drbd_socket *sock;
struct p_state *p;
return drbd_send_command(mdev, sock, P_STATE, sizeof(*p), NULL, 0);
}
+/**
+ * drbd_send_state() - After a state change, sends the new state to the peer
+ * @mdev: DRBD device.
+ * @state: the state to send, not necessarily the current state.
+ *
+ * Each state change queues an "after_state_ch" work, which will eventually
+ * send the resulting new state to the peer. If more state changes happen
+ * between queuing and processing of the after_state_ch work, we still
+ * want to send each intermediary state in the order it occurred.
+ */
+int drbd_send_state(struct drbd_conf *mdev, union drbd_state state)
+{
+ struct drbd_socket *sock;
+ struct p_state *p;
+
+ sock = &mdev->tconn->data;
+ p = drbd_prepare_command(mdev, sock);
+ if (!p)
+ return -EIO;
+ p->state = cpu_to_be32(state.i); /* Within the send mutex */
+ return drbd_send_command(mdev, sock, P_STATE, sizeof(*p), NULL, 0);
+}
+
int drbd_send_state_req(struct drbd_conf *mdev, union drbd_state mask, union drbd_state val)
{
struct drbd_socket *sock;
p->mask = cpu_to_be32(mask.i);
p->val = cpu_to_be32(val.i);
return drbd_send_command(mdev, sock, P_STATE_CHG_REQ, sizeof(*p), NULL, 0);
-
}
int conn_send_state_req(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val)
unsigned long rl;
unsigned len;
unsigned toggle;
- int bits;
+ int bits, use_rle;
/* may we use this feature? */
- if ((mdev->tconn->net_conf->use_rle == 0) ||
- (mdev->tconn->agreed_pro_version < 90))
- return 0;
+ rcu_read_lock();
+ use_rle = rcu_dereference(mdev->tconn->net_conf)->use_rle;
+ rcu_read_unlock();
+ if (!use_rle || mdev->tconn->agreed_pro_version < 90)
+ return 0;
if (c->bit_offset >= c->bm_bits)
return 0; /* nothing to do. */
void drbd_send_ack_dp(struct drbd_conf *mdev, enum drbd_packet cmd,
struct p_data *dp, int data_size)
{
- data_size -= (mdev->tconn->agreed_pro_version >= 87 && mdev->tconn->integrity_r_tfm) ?
- crypto_hash_digestsize(mdev->tconn->integrity_r_tfm) : 0;
+ if (mdev->tconn->peer_integrity_tfm)
+ data_size -= crypto_hash_digestsize(mdev->tconn->peer_integrity_tfm);
_drbd_send_ack(mdev, cmd, dp->sector, cpu_to_be32(data_size),
dp->block_id);
}
struct bio_vec *bvec;
int i;
/* hint all but last page with MSG_MORE */
- __bio_for_each_segment(bvec, bio, i, 0) {
+ bio_for_each_segment(bvec, bio, i) {
int err;
err = _drbd_no_send_page(mdev, bvec->bv_page,
struct bio_vec *bvec;
int i;
/* hint all but last page with MSG_MORE */
- __bio_for_each_segment(bvec, bio, i, 0) {
+ bio_for_each_segment(bvec, bio, i) {
int err;
err = _drbd_send_page(mdev, bvec->bv_page,
int dgs;
int err;
- dgs = (mdev->tconn->agreed_pro_version >= 87 && mdev->tconn->integrity_w_tfm) ?
- crypto_hash_digestsize(mdev->tconn->integrity_w_tfm) : 0;
-
sock = &mdev->tconn->data;
p = drbd_prepare_command(mdev, sock);
+ dgs = mdev->tconn->integrity_tfm ? crypto_hash_digestsize(mdev->tconn->integrity_tfm) : 0;
+
if (!p)
return -EIO;
p->sector = cpu_to_be64(req->i.sector);
p->block_id = (unsigned long)req;
- p->seq_num = cpu_to_be32(req->seq_num = atomic_inc_return(&mdev->packet_seq));
+ p->seq_num = cpu_to_be32(atomic_inc_return(&mdev->packet_seq));
dp_flags = bio_flags_to_wire(mdev, req->master_bio->bi_rw);
if (mdev->state.conn >= C_SYNC_SOURCE &&
mdev->state.conn <= C_PAUSED_SYNC_T)
dp_flags |= DP_MAY_SET_IN_SYNC;
+ if (mdev->tconn->agreed_pro_version >= 100) {
+ if (req->rq_state & RQ_EXP_RECEIVE_ACK)
+ dp_flags |= DP_SEND_RECEIVE_ACK;
+ if (req->rq_state & RQ_EXP_WRITE_ACK)
+ dp_flags |= DP_SEND_WRITE_ACK;
+ }
p->dp_flags = cpu_to_be32(dp_flags);
if (dgs)
- drbd_csum_bio(mdev, mdev->tconn->integrity_w_tfm, req->master_bio, p + 1);
+ drbd_csum_bio(mdev, mdev->tconn->integrity_tfm, req->master_bio, p + 1);
err = __send_command(mdev->tconn, mdev->vnr, sock, P_DATA, sizeof(*p) + dgs, NULL, req->i.size);
if (!err) {
/* For protocol A, we have to memcpy the payload into
* out ok after sending on this side, but does not fit on the
* receiving side, we sure have detected corruption elsewhere.
*/
- if (mdev->tconn->net_conf->wire_protocol == DRBD_PROT_A || dgs)
+ if (!(req->rq_state & (RQ_EXP_RECEIVE_ACK | RQ_EXP_WRITE_ACK)) || dgs)
err = _drbd_send_bio(mdev, req->master_bio);
else
err = _drbd_send_zc_bio(mdev, req->master_bio);
/* 64 byte, 512 bit, is the largest digest size
* currently supported in kernel crypto. */
unsigned char digest[64];
- drbd_csum_bio(mdev, mdev->tconn->integrity_w_tfm, req->master_bio, digest);
+ drbd_csum_bio(mdev, mdev->tconn->integrity_tfm, req->master_bio, digest);
if (memcmp(p + 1, digest, dgs)) {
dev_warn(DEV,
"Digest mismatch, buffer modified by upper layers during write: %llus +%u\n",
int err;
int dgs;
- dgs = (mdev->tconn->agreed_pro_version >= 87 && mdev->tconn->integrity_w_tfm) ?
- crypto_hash_digestsize(mdev->tconn->integrity_w_tfm) : 0;
-
sock = &mdev->tconn->data;
p = drbd_prepare_command(mdev, sock);
+
+ dgs = mdev->tconn->integrity_tfm ? crypto_hash_digestsize(mdev->tconn->integrity_tfm) : 0;
+
if (!p)
return -EIO;
p->sector = cpu_to_be64(peer_req->i.sector);
p->block_id = peer_req->block_id;
p->seq_num = 0; /* unused */
+ p->dp_flags = 0;
if (dgs)
- drbd_csum_ee(mdev, mdev->tconn->integrity_w_tfm, peer_req, p + 1);
+ drbd_csum_ee(mdev, mdev->tconn->integrity_tfm, peer_req, p + 1);
err = __send_command(mdev->tconn, mdev->vnr, sock, cmd, sizeof(*p) + dgs, NULL, peer_req->i.size);
if (!err)
err = _drbd_send_zc_ee(mdev, peer_req);
msg.msg_flags = msg_flags | MSG_NOSIGNAL;
if (sock == tconn->data.socket) {
- tconn->ko_count = tconn->net_conf->ko_count;
+ rcu_read_lock();
+ tconn->ko_count = rcu_dereference(tconn->net_conf)->ko_count;
+ rcu_read_unlock();
drbd_update_congested(tconn);
}
do {
atomic_set(&mdev->rs_sect_in, 0);
atomic_set(&mdev->rs_sect_ev, 0);
atomic_set(&mdev->ap_in_flight, 0);
+ atomic_set(&mdev->md_io_in_use, 0);
- mutex_init(&mdev->md_io_mutex);
mutex_init(&mdev->own_state_mutex);
mdev->state_mutex = &mdev->own_state_mutex;
spin_lock_init(&mdev->al_lock);
spin_lock_init(&mdev->peer_seq_lock);
- spin_lock_init(&mdev->epoch_lock);
INIT_LIST_HEAD(&mdev->active_ee);
INIT_LIST_HEAD(&mdev->sync_ee);
init_waitqueue_head(&mdev->al_wait);
init_waitqueue_head(&mdev->seq_wait);
- /* mdev->tconn->agreed_pro_version gets initialized in drbd_connect() */
- mdev->write_ordering = WO_bdev_flush;
mdev->resync_wenr = LC_FREE;
mdev->peer_max_bio_size = DRBD_MAX_BIO_SIZE_SAFE;
mdev->local_max_bio_size = DRBD_MAX_BIO_SIZE_SAFE;
dev_err(DEV, "ASSERT FAILED: receiver t_state == %d expected 0.\n",
mdev->tconn->receiver.t_state);
- /* no need to lock it, I'm the only thread alive */
- if (atomic_read(&mdev->current_epoch->epoch_size) != 0)
- dev_err(DEV, "epoch_size:%d\n", atomic_read(&mdev->current_epoch->epoch_size));
mdev->al_writ_cnt =
mdev->bm_writ_cnt =
mdev->read_cnt =
drbd_bm_cleanup(mdev);
}
- drbd_free_resources(mdev);
+ drbd_free_bc(mdev->ldev);
+ mdev->ldev = NULL;
+
clear_bit(AL_SUSPENDED, &mdev->flags);
- /*
- * currently we drbd_init_ee only on module load, so
- * we may do drbd_release_ee only on module unload!
- */
D_ASSERT(list_empty(&mdev->active_ee));
D_ASSERT(list_empty(&mdev->sync_ee));
D_ASSERT(list_empty(&mdev->done_ee));
.notifier_call = drbd_notify_sys,
};
-static void drbd_release_ee_lists(struct drbd_conf *mdev)
+static void drbd_release_all_peer_reqs(struct drbd_conf *mdev)
{
int rr;
- rr = drbd_release_ee(mdev, &mdev->active_ee);
+ rr = drbd_free_peer_reqs(mdev, &mdev->active_ee);
if (rr)
dev_err(DEV, "%d EEs in active list found!\n", rr);
- rr = drbd_release_ee(mdev, &mdev->sync_ee);
+ rr = drbd_free_peer_reqs(mdev, &mdev->sync_ee);
if (rr)
dev_err(DEV, "%d EEs in sync list found!\n", rr);
- rr = drbd_release_ee(mdev, &mdev->read_ee);
+ rr = drbd_free_peer_reqs(mdev, &mdev->read_ee);
if (rr)
dev_err(DEV, "%d EEs in read list found!\n", rr);
- rr = drbd_release_ee(mdev, &mdev->done_ee);
+ rr = drbd_free_peer_reqs(mdev, &mdev->done_ee);
if (rr)
dev_err(DEV, "%d EEs in done list found!\n", rr);
- rr = drbd_release_ee(mdev, &mdev->net_ee);
+ rr = drbd_free_peer_reqs(mdev, &mdev->net_ee);
if (rr)
dev_err(DEV, "%d EEs in net list found!\n", rr);
}
/* caution. no locking. */
-void drbd_delete_device(unsigned int minor)
+void drbd_minor_destroy(struct kref *kref)
{
- struct drbd_conf *mdev = minor_to_mdev(minor);
-
- if (!mdev)
- return;
+ struct drbd_conf *mdev = container_of(kref, struct drbd_conf, kref);
+ struct drbd_tconn *tconn = mdev->tconn;
- idr_remove(&mdev->tconn->volumes, mdev->vnr);
- idr_remove(&minors, minor);
- synchronize_rcu();
+ del_timer_sync(&mdev->request_timer);
/* paranoia asserts */
D_ASSERT(mdev->open_cnt == 0);
D_ASSERT(list_empty(&mdev->tconn->data.work.q));
/* end paranoia asserts */
- del_gendisk(mdev->vdisk);
-
/* cleanup stuff that may have been allocated during
* device (re-)configuration or state changes */
if (mdev->this_bdev)
bdput(mdev->this_bdev);
- drbd_free_resources(mdev);
+ drbd_free_bc(mdev->ldev);
+ mdev->ldev = NULL;
- drbd_release_ee_lists(mdev);
+ drbd_release_all_peer_reqs(mdev);
lc_destroy(mdev->act_log);
lc_destroy(mdev->resync);
kfree(mdev->p_uuid);
/* mdev->p_uuid = NULL; */
- /* cleanup the rest that has been
- * allocated from drbd_new_device
- * and actually free the mdev itself */
- drbd_free_mdev(mdev);
+ if (mdev->bitmap) /* should no longer be there. */
+ drbd_bm_cleanup(mdev);
+ __free_page(mdev->md_io_page);
+ put_disk(mdev->vdisk);
+ blk_cleanup_queue(mdev->rq_queue);
+ kfree(mdev->rs_plan_s);
+ kfree(mdev);
+
+ kref_put(&tconn->kref, &conn_destroy);
}
static void drbd_cleanup(void)
{
unsigned int i;
struct drbd_conf *mdev;
+ struct drbd_tconn *tconn, *tmp;
unregister_reboot_notifier(&drbd_notifier);
drbd_genl_unregister();
- idr_for_each_entry(&minors, mdev, i)
- drbd_delete_device(i);
+ idr_for_each_entry(&minors, mdev, i) {
+ idr_remove(&minors, mdev_to_minor(mdev));
+ idr_remove(&mdev->tconn->volumes, mdev->vnr);
+ del_gendisk(mdev->vdisk);
+ /* synchronize_rcu(); No other threads running at this point */
+ kref_put(&mdev->kref, &drbd_minor_destroy);
+ }
+
+ /* not _rcu since, no other updater anymore. Genl already unregistered */
+ list_for_each_entry_safe(tconn, tmp, &drbd_tconns, all_tconn) {
+ list_del(&tconn->all_tconn); /* not _rcu no proc, not other threads */
+ /* synchronize_rcu(); */
+ kref_put(&tconn->kref, &conn_destroy);
+ }
+
drbd_destroy_mempools();
unregister_blkdev(DRBD_MAJOR, "drbd");
INIT_LIST_HEAD(&wq->q);
}
-struct drbd_tconn *conn_by_name(const char *name)
+struct drbd_tconn *conn_get_by_name(const char *name)
{
struct drbd_tconn *tconn;
if (!name || !name[0])
return NULL;
- mutex_lock(&drbd_cfg_mutex);
- list_for_each_entry(tconn, &drbd_tconns, all_tconn) {
- if (!strcmp(tconn->name, name))
+ rcu_read_lock();
+ list_for_each_entry_rcu(tconn, &drbd_tconns, all_tconn) {
+ if (!strcmp(tconn->name, name)) {
+ kref_get(&tconn->kref);
goto found;
+ }
}
tconn = NULL;
found:
- mutex_unlock(&drbd_cfg_mutex);
+ rcu_read_unlock();
+ return tconn;
+}
+
+struct drbd_tconn *conn_get_by_addrs(void *my_addr, int my_addr_len,
+ void *peer_addr, int peer_addr_len)
+{
+ struct drbd_tconn *tconn;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(tconn, &drbd_tconns, all_tconn) {
+ if (tconn->my_addr_len == my_addr_len &&
+ tconn->peer_addr_len == peer_addr_len &&
+ !memcmp(&tconn->my_addr, my_addr, my_addr_len) &&
+ !memcmp(&tconn->peer_addr, peer_addr, peer_addr_len)) {
+ kref_get(&tconn->kref);
+ goto found;
+ }
+ }
+ tconn = NULL;
+found:
+ rcu_read_unlock();
return tconn;
}
free_page((unsigned long) socket->rbuf);
}
-struct drbd_tconn *drbd_new_tconn(const char *name)
+void conn_free_crypto(struct drbd_tconn *tconn)
+{
+ drbd_free_sock(tconn);
+
+ crypto_free_hash(tconn->csums_tfm);
+ crypto_free_hash(tconn->verify_tfm);
+ crypto_free_hash(tconn->cram_hmac_tfm);
+ crypto_free_hash(tconn->integrity_tfm);
+ crypto_free_hash(tconn->peer_integrity_tfm);
+ kfree(tconn->int_dig_in);
+ kfree(tconn->int_dig_vv);
+
+ tconn->csums_tfm = NULL;
+ tconn->verify_tfm = NULL;
+ tconn->cram_hmac_tfm = NULL;
+ tconn->integrity_tfm = NULL;
+ tconn->peer_integrity_tfm = NULL;
+ tconn->int_dig_in = NULL;
+ tconn->int_dig_vv = NULL;
+}
+
+int set_resource_options(struct drbd_tconn *tconn, struct res_opts *res_opts)
+{
+ cpumask_var_t new_cpu_mask;
+ int err;
+
+ if (!zalloc_cpumask_var(&new_cpu_mask, GFP_KERNEL))
+ return -ENOMEM;
+ /*
+ retcode = ERR_NOMEM;
+ drbd_msg_put_info("unable to allocate cpumask");
+ */
+
+ /* silently ignore cpu mask on UP kernel */
+ if (nr_cpu_ids > 1 && res_opts->cpu_mask[0] != 0) {
+ /* FIXME: Get rid of constant 32 here */
+ err = __bitmap_parse(res_opts->cpu_mask, 32, 0,
+ cpumask_bits(new_cpu_mask), nr_cpu_ids);
+ if (err) {
+ conn_warn(tconn, "__bitmap_parse() failed with %d\n", err);
+ /* retcode = ERR_CPU_MASK_PARSE; */
+ goto fail;
+ }
+ }
+ tconn->res_opts = *res_opts;
+ if (!cpumask_equal(tconn->cpu_mask, new_cpu_mask)) {
+ cpumask_copy(tconn->cpu_mask, new_cpu_mask);
+ drbd_calc_cpu_mask(tconn);
+ tconn->receiver.reset_cpu_mask = 1;
+ tconn->asender.reset_cpu_mask = 1;
+ tconn->worker.reset_cpu_mask = 1;
+ }
+ err = 0;
+
+fail:
+ free_cpumask_var(new_cpu_mask);
+ return err;
+
+}
+
+/* caller must be under genl_lock() */
+struct drbd_tconn *conn_create(const char *name, struct res_opts *res_opts)
{
struct drbd_tconn *tconn;
if (!zalloc_cpumask_var(&tconn->cpu_mask, GFP_KERNEL))
goto fail;
+ if (set_resource_options(tconn, res_opts))
+ goto fail;
+
if (!tl_init(tconn))
goto fail;
+ tconn->current_epoch = kzalloc(sizeof(struct drbd_epoch), GFP_KERNEL);
+ if (!tconn->current_epoch)
+ goto fail;
+ INIT_LIST_HEAD(&tconn->current_epoch->list);
+ tconn->epochs = 1;
+ spin_lock_init(&tconn->epoch_lock);
+ tconn->write_ordering = WO_bdev_flush;
+
tconn->cstate = C_STANDALONE;
mutex_init(&tconn->cstate_mutex);
spin_lock_init(&tconn->req_lock);
- atomic_set(&tconn->net_cnt, 0);
- init_waitqueue_head(&tconn->net_cnt_wait);
+ mutex_init(&tconn->conf_update);
init_waitqueue_head(&tconn->ping_wait);
idr_init(&tconn->volumes);
drbd_thread_init(tconn, &tconn->worker, drbd_worker, "worker");
drbd_thread_init(tconn, &tconn->asender, drbd_asender, "asender");
- tconn->res_opts = (struct res_opts) {
- {}, 0, /* cpu_mask */
- DRBD_ON_NO_DATA_DEF, /* on_no_data */
- };
-
- mutex_lock(&drbd_cfg_mutex);
- list_add_tail(&tconn->all_tconn, &drbd_tconns);
- mutex_unlock(&drbd_cfg_mutex);
+ kref_init(&tconn->kref);
+ list_add_tail_rcu(&tconn->all_tconn, &drbd_tconns);
return tconn;
fail:
+ kfree(tconn->current_epoch);
tl_cleanup(tconn);
free_cpumask_var(tconn->cpu_mask);
drbd_free_socket(&tconn->meta);
return NULL;
}
-void drbd_free_tconn(struct drbd_tconn *tconn)
+void conn_destroy(struct kref *kref)
{
- list_del(&tconn->all_tconn);
+ struct drbd_tconn *tconn = container_of(kref, struct drbd_tconn, kref);
+
+ if (atomic_read(&tconn->current_epoch->epoch_size) != 0)
+ conn_err(tconn, "epoch_size:%d\n", atomic_read(&tconn->current_epoch->epoch_size));
+ kfree(tconn->current_epoch);
+
idr_destroy(&tconn->volumes);
free_cpumask_var(tconn->cpu_mask);
if (!mdev)
return ERR_NOMEM;
+ kref_get(&tconn->kref);
mdev->tconn = tconn;
+
mdev->minor = minor;
mdev->vnr = vnr;
mdev->read_requests = RB_ROOT;
mdev->write_requests = RB_ROOT;
- mdev->current_epoch = kzalloc(sizeof(struct drbd_epoch), GFP_KERNEL);
- if (!mdev->current_epoch)
- goto out_no_epoch;
-
- INIT_LIST_HEAD(&mdev->current_epoch->list);
- mdev->epochs = 1;
-
if (!idr_pre_get(&minors, GFP_KERNEL))
goto out_no_minor_idr;
if (idr_get_new_above(&minors, mdev, minor, &minor_got))
goto out_idr_remove_vol;
}
add_disk(disk);
+ kref_init(&mdev->kref); /* one ref for both idrs and the the add_disk */
/* inherit the connection state */
mdev->state.conn = tconn->cstate;
if (mdev->state.conn == C_WF_REPORT_PARAMS)
- drbd_connected(vnr, mdev, tconn);
+ drbd_connected(mdev);
return NO_ERROR;
idr_remove(&minors, minor_got);
synchronize_rcu();
out_no_minor_idr:
- kfree(mdev->current_epoch);
-out_no_epoch:
drbd_bm_cleanup(mdev);
out_no_bitmap:
__free_page(mdev->md_io_page);
blk_cleanup_queue(q);
out_no_q:
kfree(mdev);
+ kref_put(&tconn->kref, &conn_destroy);
return err;
}
-/* counterpart of drbd_new_device.
- * last part of drbd_delete_device. */
-void drbd_free_mdev(struct drbd_conf *mdev)
-{
- kfree(mdev->current_epoch);
- if (mdev->bitmap) /* should no longer be there. */
- drbd_bm_cleanup(mdev);
- __free_page(mdev->md_io_page);
- put_disk(mdev->vdisk);
- blk_cleanup_queue(mdev->rq_queue);
- kfree(mdev);
-}
-
-
int __init drbd_init(void)
{
int err;
#ifdef MODULE
return -EINVAL;
#else
- minor_count = 8;
+ minor_count = DRBD_MINOR_COUNT_DEF;
#endif
}
}
}
-
-void drbd_free_resources(struct drbd_conf *mdev)
-{
- crypto_free_hash(mdev->tconn->csums_tfm);
- mdev->tconn->csums_tfm = NULL;
- crypto_free_hash(mdev->tconn->verify_tfm);
- mdev->tconn->verify_tfm = NULL;
- crypto_free_hash(mdev->tconn->cram_hmac_tfm);
- mdev->tconn->cram_hmac_tfm = NULL;
- crypto_free_hash(mdev->tconn->integrity_w_tfm);
- mdev->tconn->integrity_w_tfm = NULL;
- crypto_free_hash(mdev->tconn->integrity_r_tfm);
- mdev->tconn->integrity_r_tfm = NULL;
-
- drbd_free_sock(mdev->tconn);
-
- __no_warn(local,
- drbd_free_bc(mdev->ldev);
- mdev->ldev = NULL;);
-}
-
/* meta data management */
struct meta_data_on_disk {
if (!get_ldev_if_state(mdev, D_FAILED))
return;
- mutex_lock(&mdev->md_io_mutex);
- buffer = (struct meta_data_on_disk *)page_address(mdev->md_io_page);
+ buffer = drbd_md_get_buffer(mdev);
+ if (!buffer)
+ goto out;
+
memset(buffer, 0, 512);
buffer->la_size = cpu_to_be64(drbd_get_capacity(mdev->this_bdev));
for (i = UI_CURRENT; i < UI_SIZE; i++)
buffer->uuid[i] = cpu_to_be64(mdev->ldev->md.uuid[i]);
buffer->flags = cpu_to_be32(mdev->ldev->md.flags);
- buffer->magic = cpu_to_be32(DRBD_MD_MAGIC);
+ buffer->magic = cpu_to_be32(DRBD_MD_MAGIC_84_UNCLEAN);
buffer->md_size_sect = cpu_to_be32(mdev->ldev->md.md_size_sect);
buffer->al_offset = cpu_to_be32(mdev->ldev->md.al_offset);
* since we updated it on metadata. */
mdev->ldev->md.la_size_sect = drbd_get_capacity(mdev->this_bdev);
- mutex_unlock(&mdev->md_io_mutex);
+ drbd_md_put_buffer(mdev);
+out:
put_ldev(mdev);
}
* @bdev: Device from which the meta data should be read in.
*
* Return 0 (NO_ERROR) on success, and an enum drbd_ret_code in case
- * something goes wrong. Currently only: ERR_IO_MD_DISK, ERR_MD_INVALID.
+ * something goes wrong.
*/
int drbd_md_read(struct drbd_conf *mdev, struct drbd_backing_dev *bdev)
{
struct meta_data_on_disk *buffer;
+ u32 magic, flags;
int i, rv = NO_ERROR;
if (!get_ldev_if_state(mdev, D_ATTACHING))
return ERR_IO_MD_DISK;
- mutex_lock(&mdev->md_io_mutex);
- buffer = (struct meta_data_on_disk *)page_address(mdev->md_io_page);
+ buffer = drbd_md_get_buffer(mdev);
+ if (!buffer)
+ goto out;
if (drbd_md_sync_page_io(mdev, bdev, bdev->md.md_offset, READ)) {
/* NOTE: can't do normal error processing here as this is
goto err;
}
- if (buffer->magic != cpu_to_be32(DRBD_MD_MAGIC)) {
- dev_err(DEV, "Error while reading metadata, magic not found.\n");
+ magic = be32_to_cpu(buffer->magic);
+ flags = be32_to_cpu(buffer->flags);
+ if (magic == DRBD_MD_MAGIC_84_UNCLEAN ||
+ (magic == DRBD_MD_MAGIC_08 && !(flags & MDF_AL_CLEAN))) {
+ /* btw: that's Activity Log clean, not "all" clean. */
+ dev_err(DEV, "Found unclean meta data. Did you \"drbdadm apply-al\"?\n");
+ rv = ERR_MD_UNCLEAN;
+ goto err;
+ }
+ if (magic != DRBD_MD_MAGIC_08) {
+ if (magic == DRBD_MD_MAGIC_07)
+ dev_err(DEV, "Found old (0.7) meta data magic. Did you \"drbdadm create-md\"?\n");
+ else
+ dev_err(DEV, "Meta data magic not found. Did you \"drbdadm create-md\"?\n");
rv = ERR_MD_INVALID;
goto err;
}
for (i = UI_CURRENT; i < UI_SIZE; i++)
bdev->md.uuid[i] = be64_to_cpu(buffer->uuid[i]);
bdev->md.flags = be32_to_cpu(buffer->flags);
- bdev->dc.al_extents = be32_to_cpu(buffer->al_nr_extents);
bdev->md.device_uuid = be64_to_cpu(buffer->device_uuid);
spin_lock_irq(&mdev->tconn->req_lock);
}
spin_unlock_irq(&mdev->tconn->req_lock);
- if (bdev->dc.al_extents < 7)
- bdev->dc.al_extents = 127;
-
err:
- mutex_unlock(&mdev->md_io_mutex);
+ drbd_md_put_buffer(mdev);
+ out:
put_ldev(mdev);
return rv;
[P_DELAY_PROBE] = "DelayProbe",
[P_OUT_OF_SYNC] = "OutOfSync",
[P_RETRY_WRITE] = "RetryWrite",
+ [P_RS_CANCEL] = "RSCancel",
+ [P_CONN_ST_CHG_REQ] = "conn_st_chg_req",
+ [P_CONN_ST_CHG_REPLY] = "conn_st_chg_reply",
+ [P_RETRY_WRITE] = "retry_write",
+ [P_PROTOCOL_UPDATE] = "protocol_update",
+
+ /* enum drbd_packet, but not commands - obsoleted flags:
+ * P_MAY_IGNORE
+ * P_MAX_OPT_CMD
+ */
};
+ /* too big for the array: 0xfffX */
if (cmd == P_INITIAL_META)
return "InitialMeta";
if (cmd == P_INITIAL_DATA)
*/
int drbd_wait_misc(struct drbd_conf *mdev, struct drbd_interval *i)
{
- struct net_conf *net_conf = mdev->tconn->net_conf;
+ struct net_conf *nc;
DEFINE_WAIT(wait);
long timeout;
- if (!net_conf)
+ rcu_read_lock();
+ nc = rcu_dereference(mdev->tconn->net_conf);
+ if (!nc) {
+ rcu_read_unlock();
return -ETIMEDOUT;
- timeout = MAX_SCHEDULE_TIMEOUT;
- if (net_conf->ko_count)
- timeout = net_conf->timeout * HZ / 10 * net_conf->ko_count;
+ }
+ timeout = nc->ko_count ? nc->timeout * HZ / 10 * nc->ko_count : MAX_SCHEDULE_TIMEOUT;
+ rcu_read_unlock();
/* Indicate to wake up mdev->misc_wait on progress. */
i->waiting = true;