drbd: Introduce "peer_device" object between "device" and "connection"
In a setup where a device (aka volume) can replicate to multiple peers and one
connection can be shared between multiple devices, we need separate objects to
represent devices on peer nodes and network connections.
As a first step to introduce multiple connections per device, give each
drbd_device object a single drbd_peer_device object which connects it to a
drbd_connection object.
Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 1b5b7ea..a8c9c86 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -246,10 +246,10 @@
/* some more paranoia, if the request was over-determined */
if (adm_ctx.device && adm_ctx.connection &&
- adm_ctx.device->connection != adm_ctx.connection) {
+ first_peer_device(adm_ctx.device)->connection != adm_ctx.connection) {
pr_warning("request: minor=%u, resource=%s; but that minor belongs to connection %s\n",
adm_ctx.minor, adm_ctx.resource_name,
- adm_ctx.device->connection->name);
+ first_peer_device(adm_ctx.device)->connection->name);
drbd_msg_put_info("minor exists in different resource");
return ERR_INVALID_REQUEST;
}
@@ -258,7 +258,7 @@
adm_ctx.volume != adm_ctx.device->vnr) {
pr_warning("request: minor=%u, volume=%u; but that minor is volume %u in %s\n",
adm_ctx.minor, adm_ctx.volume,
- adm_ctx.device->vnr, adm_ctx.device->connection->name);
+ adm_ctx.device->vnr, first_peer_device(adm_ctx.device)->connection->name);
drbd_msg_put_info("minor exists as different volume");
return ERR_INVALID_REQUEST;
}
@@ -323,7 +323,7 @@
NULL };
char mb[12];
char *argv[] = {usermode_helper, cmd, mb, NULL };
- struct drbd_connection *connection = device->connection;
+ struct drbd_connection *connection = first_peer_device(device)->connection;
struct sib_info sib;
int ret;
@@ -544,7 +544,7 @@
union drbd_state mask, val;
if (new_role == R_PRIMARY)
- request_ping(device->connection); /* Detect a dead peer ASAP */
+ request_ping(first_peer_device(device)->connection); /* Detect a dead peer ASAP */
mutex_lock(device->state_mutex);
@@ -575,7 +575,7 @@
device->state.disk == D_CONSISTENT && mask.pdsk == 0) {
D_ASSERT(device->state.pdsk == D_UNKNOWN);
- if (conn_try_outdate_peer(device->connection)) {
+ if (conn_try_outdate_peer(first_peer_device(device)->connection)) {
val.disk = D_UP_TO_DATE;
mask.disk = D_MASK;
}
@@ -585,7 +585,7 @@
if (rv == SS_NOTHING_TO_DO)
goto out;
if (rv == SS_PRIMARY_NOP && mask.pdsk == 0) {
- if (!conn_try_outdate_peer(device->connection) && force) {
+ if (!conn_try_outdate_peer(first_peer_device(device)->connection) && force) {
dev_warn(DEV, "Forced into split brain situation!\n");
mask.pdsk = D_MASK;
val.pdsk = D_OUTDATED;
@@ -598,7 +598,7 @@
retry at most once more in this case. */
int timeo;
rcu_read_lock();
- nc = rcu_dereference(device->connection->net_conf);
+ nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
timeo = nc ? (nc->ping_timeo + 1) * HZ / 10 : 1;
rcu_read_unlock();
schedule_timeout_interruptible(timeo);
@@ -633,11 +633,11 @@
put_ldev(device);
}
} else {
- mutex_lock(&device->connection->conf_update);
- nc = device->connection->net_conf;
+ mutex_lock(&first_peer_device(device)->connection->conf_update);
+ nc = first_peer_device(device)->connection->net_conf;
if (nc)
nc->discard_my_data = 0; /* without copy; single bit op is atomic */
- mutex_unlock(&device->connection->conf_update);
+ mutex_unlock(&first_peer_device(device)->connection->conf_update);
set_disk_ro(device->vdisk, false);
if (get_ldev(device)) {
@@ -1134,12 +1134,12 @@
Because new from 8.3.8 onwards the peer can use multiple
BIOs for a single peer_request */
if (device->state.conn >= C_WF_REPORT_PARAMS) {
- if (device->connection->agreed_pro_version < 94)
+ if (first_peer_device(device)->connection->agreed_pro_version < 94)
peer = min(device->peer_max_bio_size, DRBD_MAX_SIZE_H80_PACKET);
/* Correct old drbd (up to 8.3.7) if it believes it can do more than 32KiB */
- else if (device->connection->agreed_pro_version == 94)
+ else if (first_peer_device(device)->connection->agreed_pro_version == 94)
peer = DRBD_MAX_SIZE_H80_PACKET;
- else if (device->connection->agreed_pro_version < 100)
+ else if (first_peer_device(device)->connection->agreed_pro_version < 100)
peer = DRBD_MAX_BIO_SIZE_P95; /* drbd 8.3.8 onwards, before 8.4.0 */
else
peer = DRBD_MAX_BIO_SIZE;
@@ -1190,10 +1190,10 @@
}
drbd_al_shrink(device);
- spin_lock_irq(&device->connection->req_lock);
+ spin_lock_irq(&first_peer_device(device)->connection->req_lock);
if (device->state.conn < C_CONNECTED)
s = !test_and_set_bit(AL_SUSPENDED, &device->flags);
- spin_unlock_irq(&device->connection->req_lock);
+ spin_unlock_irq(&first_peer_device(device)->connection->req_lock);
lc_unlock(device->act_log);
if (s)
@@ -1264,7 +1264,7 @@
goto fail;
}
- mutex_lock(&device->connection->conf_update);
+ mutex_lock(&first_peer_device(device)->connection->conf_update);
old_disk_conf = device->ldev->disk_conf;
*new_disk_conf = *old_disk_conf;
if (should_set_defaults(info))
@@ -1327,7 +1327,7 @@
rcu_assign_pointer(device->rs_plan_s, new_plan);
}
- mutex_unlock(&device->connection->conf_update);
+ mutex_unlock(&first_peer_device(device)->connection->conf_update);
if (new_disk_conf->al_updates)
device->ldev->md.flags &= ~MDF_AL_DISABLED;
@@ -1339,7 +1339,7 @@
else
set_bit(MD_NO_FUA, &device->flags);
- drbd_bump_write_ordering(device->connection, WO_bdev_flush);
+ drbd_bump_write_ordering(first_peer_device(device)->connection, WO_bdev_flush);
drbd_md_sync(device);
@@ -1353,7 +1353,7 @@
goto success;
fail_unlock:
- mutex_unlock(&device->connection->conf_update);
+ mutex_unlock(&first_peer_device(device)->connection->conf_update);
fail:
kfree(new_disk_conf);
kfree(new_plan);
@@ -1388,7 +1388,7 @@
goto finish;
device = adm_ctx.device;
- conn_reconfig_start(device->connection);
+ conn_reconfig_start(first_peer_device(device)->connection);
/* if you want to reconfigure, please tear down first */
if (device->state.disk > D_DISKLESS) {
@@ -1455,7 +1455,7 @@
goto fail;
rcu_read_lock();
- nc = rcu_dereference(device->connection->net_conf);
+ nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
if (nc) {
if (new_disk_conf->fencing == FP_STONITH && nc->wire_protocol == DRBD_PROT_A) {
rcu_read_unlock();
@@ -1636,7 +1636,7 @@
new_disk_conf = NULL;
new_plan = NULL;
- drbd_bump_write_ordering(device->connection, WO_bdev_flush);
+ drbd_bump_write_ordering(first_peer_device(device)->connection, WO_bdev_flush);
if (drbd_md_test_flag(device->ldev, MDF_CRASHED_PRIMARY))
set_bit(CRASHED_PRIMARY, &device->flags);
@@ -1644,7 +1644,8 @@
clear_bit(CRASHED_PRIMARY, &device->flags);
if (drbd_md_test_flag(device->ldev, MDF_PRIMARY_IND) &&
- !(device->state.role == R_PRIMARY && device->connection->susp_nod))
+ !(device->state.role == R_PRIMARY &&
+ first_peer_device(device)->connection->susp_nod))
set_bit(CRASHED_PRIMARY, &device->flags);
device->send_cnt = 0;
@@ -1702,7 +1703,7 @@
if (_drbd_bm_total_weight(device) == drbd_bm_bits(device))
drbd_suspend_al(device); /* IO is still suspended here... */
- spin_lock_irq(&device->connection->req_lock);
+ spin_lock_irq(&first_peer_device(device)->connection->req_lock);
os = drbd_read_state(device);
ns = os;
/* If MDF_CONSISTENT is not set go into inconsistent state,
@@ -1754,7 +1755,7 @@
}
rv = _drbd_set_state(device, ns, CS_VERBOSE, NULL);
- spin_unlock_irq(&device->connection->req_lock);
+ spin_unlock_irq(&first_peer_device(device)->connection->req_lock);
if (rv < SS_SUCCESS)
goto force_diskless_dec;
@@ -1771,7 +1772,7 @@
kobject_uevent(&disk_to_dev(device->vdisk)->kobj, KOBJ_CHANGE);
put_ldev(device);
- conn_reconfig_done(device->connection);
+ conn_reconfig_done(first_peer_device(device)->connection);
drbd_adm_finish(info, retcode);
return 0;
@@ -1781,7 +1782,7 @@
drbd_force_state(device, NS(disk, D_DISKLESS));
drbd_md_sync(device);
fail:
- conn_reconfig_done(device->connection);
+ conn_reconfig_done(first_peer_device(device)->connection);
if (nbc) {
if (nbc->backing_bdev)
blkdev_put(nbc->backing_bdev,
@@ -2357,7 +2358,7 @@
if (device->state.role != device->state.peer)
iass = (device->state.role == R_PRIMARY);
else
- iass = test_bit(RESOLVE_CONFLICTS, &device->connection->flags);
+ iass = test_bit(RESOLVE_CONFLICTS, &first_peer_device(device)->connection->flags);
if (iass)
drbd_start_resync(device, C_SYNC_SOURCE);
@@ -2412,7 +2413,7 @@
goto fail_ldev;
}
- if (rs.no_resync && device->connection->agreed_pro_version < 93) {
+ if (rs.no_resync && first_peer_device(device)->connection->agreed_pro_version < 93) {
retcode = ERR_NEED_APV_93;
goto fail_ldev;
}
@@ -2454,12 +2455,12 @@
device->ldev->known_size = drbd_get_capacity(device->ldev->backing_bdev);
if (new_disk_conf) {
- mutex_lock(&device->connection->conf_update);
+ mutex_lock(&first_peer_device(device)->connection->conf_update);
old_disk_conf = device->ldev->disk_conf;
*new_disk_conf = *old_disk_conf;
new_disk_conf->disk_size = (sector_t)rs.resize_size;
rcu_assign_pointer(device->ldev->disk_conf, new_disk_conf);
- mutex_unlock(&device->connection->conf_update);
+ mutex_unlock(&first_peer_device(device)->connection->conf_update);
synchronize_rcu();
kfree(old_disk_conf);
}
@@ -2710,9 +2711,9 @@
retcode = drbd_request_state(device, NS3(susp, 0, susp_nod, 0, susp_fen, 0));
if (retcode == SS_SUCCESS) {
if (device->state.conn < C_CONNECTED)
- tl_clear(device->connection);
+ tl_clear(first_peer_device(device)->connection);
if (device->state.disk == D_DISKLESS || device->state.disk == D_FAILED)
- tl_restart(device->connection, FAIL_FROZEN_DISK_IO);
+ tl_restart(first_peer_device(device)->connection, FAIL_FROZEN_DISK_IO);
}
drbd_resume_io(device);
@@ -2778,10 +2779,10 @@
/* We need to add connection name and volume number information still.
* Minor number is in drbd_genlmsghdr. */
- if (nla_put_drbd_cfg_context(skb, device->connection, device->vnr))
+ if (nla_put_drbd_cfg_context(skb, first_peer_device(device)->connection, device->vnr))
goto nla_put_failure;
- if (res_opts_to_skb(skb, &device->connection->res_opts, exclude_sensitive))
+ if (res_opts_to_skb(skb, &first_peer_device(device)->connection->res_opts, exclude_sensitive))
goto nla_put_failure;
rcu_read_lock();
@@ -2794,7 +2795,7 @@
if (!err) {
struct net_conf *nc;
- nc = rcu_dereference(device->connection->net_conf);
+ nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
if (nc)
err = net_conf_to_skb(skb, nc, exclude_sensitive);
}
@@ -2981,7 +2982,7 @@
}
D_ASSERT(device->vnr == volume);
- D_ASSERT(device->connection == connection);
+ D_ASSERT(first_peer_device(device)->connection == connection);
dh->minor = device_to_minor(device);
dh->ret_code = NO_ERROR;
@@ -3168,7 +3169,8 @@
}
/* this is "skip initial sync", assume to be clean */
- if (device->state.conn == C_CONNECTED && device->connection->agreed_pro_version >= 90 &&
+ if (device->state.conn == C_CONNECTED &&
+ first_peer_device(device)->connection->agreed_pro_version >= 90 &&
device->ldev->md.uuid[UI_CURRENT] == UUID_JUST_CREATED && args.clear_bm) {
dev_info(DEV, "Preparing to skip initial sync\n");
skip_initial_sync = 1;
@@ -3191,10 +3193,10 @@
drbd_send_uuids_skip_initial_sync(device);
_drbd_uuid_set(device, UI_BITMAP, 0);
drbd_print_uuids(device, "cleared bitmap UUID");
- spin_lock_irq(&device->connection->req_lock);
+ spin_lock_irq(&first_peer_device(device)->connection->req_lock);
_drbd_set_state(_NS2(device, disk, D_UP_TO_DATE, pdsk, D_UP_TO_DATE),
CS_VERBOSE, NULL);
- spin_unlock_irq(&device->connection->req_lock);
+ spin_unlock_irq(&first_peer_device(device)->connection->req_lock);
}
}
@@ -3287,7 +3289,7 @@
}
/* drbd_adm_prepare made sure already
- * that device->connection and device->vnr match the request. */
+ * that first_peer_device(device)->connection and device->vnr match the request. */
if (adm_ctx.device) {
if (info->nlhdr->nlmsg_flags & NLM_F_EXCL)
retcode = ERR_MINOR_EXISTS;
@@ -3295,7 +3297,7 @@
goto out;
}
- retcode = conn_new_minor(adm_ctx.connection, dh->minor, adm_ctx.volume);
+ retcode = drbd_create_minor(adm_ctx.connection, dh->minor, adm_ctx.volume);
out:
drbd_adm_finish(info, retcode);
return 0;
@@ -3310,7 +3312,7 @@
device->state.role == R_SECONDARY) {
_drbd_request_state(device, NS(conn, C_WF_REPORT_PARAMS),
CS_VERBOSE + CS_WAIT_COMPLETE);
- idr_remove(&device->connection->volumes, device->vnr);
+ idr_remove(&first_peer_device(device)->connection->volumes, device->vnr);
idr_remove(&minors, device_to_minor(device));
destroy_workqueue(device->submit.wq);
del_gendisk(device->vdisk);
