struct page *page;
unsigned nr_pages = (data_size + PAGE_SIZE -1) >> PAGE_SHIFT;
- if (FAULT_ACTIVE(mdev, DRBD_FAULT_AL_EE))
+ if (drbd_insert_fault(mdev, DRBD_FAULT_AL_EE))
return NULL;
e = mempool_alloc(drbd_ee_mempool, gfp_mask & ~__GFP_HIGHMEM);
char tb[4];
if (!*sock)
- return FALSE;
+ return false;
rr = drbd_recv_short(mdev, *sock, tb, 4, MSG_DONTWAIT | MSG_PEEK);
if (rr > 0 || rr == -EAGAIN) {
- return TRUE;
+ return true;
} else {
sock_release(*sock);
*sock = NULL;
- return FALSE;
+ return false;
}
}
if (s || ++try >= 3)
break;
/* give the other side time to call bind() & listen() */
- __set_current_state(TASK_INTERRUPTIBLE);
- schedule_timeout(HZ / 10);
+ schedule_timeout_interruptible(HZ / 10);
}
if (s) {
}
if (sock && msock) {
- __set_current_state(TASK_INTERRUPTIBLE);
- schedule_timeout(HZ / 10);
+ schedule_timeout_interruptible(HZ / 10);
ok = drbd_socket_okay(mdev, &sock);
ok = drbd_socket_okay(mdev, &msock) && ok;
if (ok)
msock->sk->sk_rcvtimeo = mdev->net_conf->ping_int*HZ;
/* we don't want delays.
- * we use TCP_CORK where apropriate, though */
+ * we use TCP_CORK where appropriate, though */
drbd_tcp_nodelay(sock);
drbd_tcp_nodelay(msock);
put_ldev(mdev);
}
- if (!drbd_send_protocol(mdev))
+ if (drbd_send_protocol(mdev) == -1)
return -1;
drbd_send_sync_param(mdev, &mdev->sync_conf);
drbd_send_sizes(mdev, 0, 0);
drbd_send_state(mdev);
clear_bit(USE_DEGR_WFC_T, &mdev->flags);
clear_bit(RESIZE_PENDING, &mdev->flags);
+ mod_timer(&mdev->request_timer, jiffies + HZ); /* just start it here. */
return 1;
r = drbd_recv(mdev, h, sizeof(*h));
if (unlikely(r != sizeof(*h))) {
- dev_err(DEV, "short read expecting header on sock: r=%d\n", r);
- return FALSE;
+ if (!signal_pending(current))
+ dev_warn(DEV, "short read expecting header on sock: r=%d\n", r);
+ return false;
}
if (likely(h->h80.magic == BE_DRBD_MAGIC)) {
be32_to_cpu(h->h80.magic),
be16_to_cpu(h->h80.command),
be16_to_cpu(h->h80.length));
- return FALSE;
+ return false;
}
mdev->last_received = jiffies;
- return TRUE;
+ return true;
}
static void drbd_flush(struct drbd_conf *mdev)
* @mdev: DRBD device.
* @e: epoch entry
* @rw: flag field, see bio->bi_rw
+ *
+ * May spread the pages to multiple bios,
+ * depending on bio_add_page restrictions.
+ *
+ * Returns 0 if all bios have been submitted,
+ * -ENOMEM if we could not allocate enough bios,
+ * -ENOSPC (any better suggestion?) if we have not been able to bio_add_page a
+ * single page to an empty bio (which should never happen and likely indicates
+ * that the lower level IO stack is in some way broken). This has been observed
+ * on certain Xen deployments.
*/
/* TODO allocate from our own bio_set. */
int drbd_submit_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e,
unsigned ds = e->size;
unsigned n_bios = 0;
unsigned nr_pages = (ds + PAGE_SIZE -1) >> PAGE_SHIFT;
+ int err = -ENOMEM;
/* In most cases, we will only need one bio. But in case the lower
* level restrictions happen to be different at this offset on this
page_chain_for_each(page) {
unsigned len = min_t(unsigned, ds, PAGE_SIZE);
if (!bio_add_page(bio, page, len, 0)) {
- /* a single page must always be possible! */
- BUG_ON(bio->bi_vcnt == 0);
+ /* A single page must always be possible!
+ * But in case it fails anyways,
+ * we deal with it, and complain (below). */
+ if (bio->bi_vcnt == 0) {
+ dev_err(DEV,
+ "bio_add_page failed for len=%u, "
+ "bi_vcnt=0 (bi_sector=%llu)\n",
+ len, (unsigned long long)bio->bi_sector);
+ err = -ENOSPC;
+ goto fail;
+ }
goto next_bio;
}
ds -= len;
bios = bios->bi_next;
bio_put(bio);
}
- return -ENOMEM;
+ return err;
}
static int receive_Barrier(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size)
switch (mdev->write_ordering) {
case WO_none:
if (rv == FE_RECYCLED)
- return TRUE;
+ return true;
/* receiver context, in the writeout path of the other node.
* avoid potential distributed deadlock */
D_ASSERT(atomic_read(&epoch->active) == 0);
D_ASSERT(epoch->flags == 0);
- return TRUE;
+ return true;
default:
dev_err(DEV, "Strangeness in mdev->write_ordering %d\n", mdev->write_ordering);
- return FALSE;
+ return false;
}
epoch->flags = 0;
}
spin_unlock(&mdev->epoch_lock);
- return TRUE;
+ return true;
}
/* used from receive_RSDataReply (recv_resync_read)
if (dgs) {
rr = drbd_recv(mdev, dig_in, dgs);
if (rr != dgs) {
- dev_warn(DEV, "short read receiving data digest: read %d expected %d\n",
- rr, dgs);
+ if (!signal_pending(current))
+ dev_warn(DEV,
+ "short read receiving data digest: read %d expected %d\n",
+ rr, dgs);
return NULL;
}
}
data_size -= dgs;
+ ERR_IF(data_size == 0) return NULL;
ERR_IF(data_size & 0x1ff) return NULL;
ERR_IF(data_size > DRBD_MAX_BIO_SIZE) return NULL;
/* even though we trust out peer,
* we sometimes have to double check. */
if (sector + (data_size>>9) > capacity) {
- dev_err(DEV, "capacity: %llus < sector: %llus + size: %u\n",
+ dev_err(DEV, "request from peer beyond end of local disk: "
+ "capacity: %llus < sector: %llus + size: %u\n",
(unsigned long long)capacity,
(unsigned long long)sector, data_size);
return NULL;
unsigned len = min_t(int, ds, PAGE_SIZE);
data = kmap(page);
rr = drbd_recv(mdev, data, len);
- if (FAULT_ACTIVE(mdev, DRBD_FAULT_RECEIVE)) {
+ if (drbd_insert_fault(mdev, DRBD_FAULT_RECEIVE)) {
dev_err(DEV, "Fault injection: Corrupting data on receive\n");
data[0] = data[0] ^ (unsigned long)-1;
}
kunmap(page);
if (rr != len) {
drbd_free_ee(mdev, e);
- dev_warn(DEV, "short read receiving data: read %d expected %d\n",
- rr, len);
+ if (!signal_pending(current))
+ dev_warn(DEV, "short read receiving data: read %d expected %d\n",
+ rr, len);
return NULL;
}
ds -= rr;
void *data;
if (!data_size)
- return TRUE;
+ return true;
page = drbd_pp_alloc(mdev, 1, 1);
rr = drbd_recv(mdev, data, min_t(int, data_size, PAGE_SIZE));
if (rr != min_t(int, data_size, PAGE_SIZE)) {
rv = 0;
- dev_warn(DEV, "short read receiving data: read %d expected %d\n",
- rr, min_t(int, data_size, PAGE_SIZE));
+ if (!signal_pending(current))
+ dev_warn(DEV,
+ "short read receiving data: read %d expected %d\n",
+ rr, min_t(int, data_size, PAGE_SIZE));
break;
}
data_size -= rr;
if (dgs) {
rr = drbd_recv(mdev, dig_in, dgs);
if (rr != dgs) {
- dev_warn(DEV, "short read receiving data reply digest: read %d expected %d\n",
- rr, dgs);
+ if (!signal_pending(current))
+ dev_warn(DEV,
+ "short read receiving data reply digest: read %d expected %d\n",
+ rr, dgs);
return 0;
}
}
expect);
kunmap(bvec->bv_page);
if (rr != expect) {
- dev_warn(DEV, "short read receiving data reply: "
- "read %d expected %d\n",
- rr, expect);
+ if (!signal_pending(current))
+ dev_warn(DEV, "short read receiving data reply: "
+ "read %d expected %d\n",
+ rr, expect);
return 0;
}
data_size -= rr;
atomic_add(data_size >> 9, &mdev->rs_sect_ev);
if (drbd_submit_ee(mdev, e, WRITE, DRBD_FAULT_RS_WR) == 0)
- return TRUE;
+ return true;
- /* drbd_submit_ee currently fails for one reason only:
- * not being able to allocate enough bios.
- * Is dropping the connection going to help? */
+ /* don't care for the reason here */
+ dev_err(DEV, "submit failed, triggering re-connect\n");
spin_lock_irq(&mdev->req_lock);
list_del(&e->w.list);
spin_unlock_irq(&mdev->req_lock);
drbd_free_ee(mdev, e);
fail:
put_ldev(mdev);
- return FALSE;
+ return false;
}
static int receive_DataReply(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size)
spin_unlock_irq(&mdev->req_lock);
if (unlikely(!req)) {
dev_err(DEV, "Got a corrupt block_id/sector pair(1).\n");
- return FALSE;
+ return false;
}
/* hlist_del(&req->colision) is done in _req_may_be_done, to avoid
u32 dp_flags;
if (!get_ldev(mdev)) {
- if (__ratelimit(&drbd_ratelimit_state))
- dev_err(DEV, "Can not write mirrored data block "
- "to local disk.\n");
spin_lock(&mdev->peer_seq_lock);
if (mdev->peer_seq+1 == be32_to_cpu(p->seq_num))
mdev->peer_seq++;
e = read_in_block(mdev, p->block_id, sector, data_size);
if (!e) {
put_ldev(mdev);
- return FALSE;
+ return false;
}
e->w.cb = e_end_block;
put_ldev(mdev);
wake_asender(mdev);
finish_wait(&mdev->misc_wait, &wait);
- return TRUE;
+ return true;
}
if (signal_pending(current)) {
}
if (drbd_submit_ee(mdev, e, rw, DRBD_FAULT_DT_WR) == 0)
- return TRUE;
+ return true;
- /* drbd_submit_ee currently fails for one reason only:
- * not being able to allocate enough bios.
- * Is dropping the connection going to help? */
+ /* don't care for the reason here */
+ dev_err(DEV, "submit failed, triggering re-connect\n");
spin_lock_irq(&mdev->req_lock);
list_del(&e->w.list);
hlist_del_init(&e->colision);
drbd_al_complete_io(mdev, e->sector);
out_interrupted:
- /* yes, the epoch_size now is imbalanced.
- * but we drop the connection anyways, so we don't have a chance to
- * receive a barrier... atomic_inc(&mdev->epoch_size); */
+ drbd_may_finish_epoch(mdev, e->epoch, EV_PUT + EV_CLEANUP);
put_ldev(mdev);
drbd_free_ee(mdev, e);
- return FALSE;
+ return false;
}
/* We may throttle resync, if the lower device seems to be busy,
if (size <= 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_BIO_SIZE) {
dev_err(DEV, "%s:%d: sector: %llus, size: %u\n", __FILE__, __LINE__,
(unsigned long long)sector, size);
- return FALSE;
+ return false;
}
if (sector + (size>>9) > capacity) {
dev_err(DEV, "%s:%d: sector: %llus, size: %u\n", __FILE__, __LINE__,
(unsigned long long)sector, size);
- return FALSE;
+ return false;
}
if (!get_ldev_if_state(mdev, D_UP_TO_DATE)) {
e = drbd_alloc_ee(mdev, p->block_id, sector, size, GFP_NOIO);
if (!e) {
put_ldev(mdev);
- return FALSE;
+ return false;
}
switch (cmd) {
spin_unlock_irq(&mdev->req_lock);
if (drbd_submit_ee(mdev, e, READ, fault_type) == 0)
- return TRUE;
+ return true;
- /* drbd_submit_ee currently fails for one reason only:
- * not being able to allocate enough bios.
- * Is dropping the connection going to help? */
+ /* don't care for the reason here */
+ dev_err(DEV, "submit failed, triggering re-connect\n");
spin_lock_irq(&mdev->req_lock);
list_del(&e->w.list);
spin_unlock_irq(&mdev->req_lock);
out_free_e:
put_ldev(mdev);
drbd_free_ee(mdev, e);
- return FALSE;
+ return false;
}
static int drbd_asb_recover_0p(struct drbd_conf *mdev) __must_hold(local)
static int drbd_asb_recover_1p(struct drbd_conf *mdev) __must_hold(local)
{
- int self, peer, hg, rv = -100;
-
- self = mdev->ldev->md.uuid[UI_BITMAP] & 1;
- peer = mdev->p_uuid[UI_BITMAP] & 1;
+ int hg, rv = -100;
switch (mdev->net_conf->after_sb_1p) {
case ASB_DISCARD_YOUNGER_PRI:
case ASB_CALL_HELPER:
hg = drbd_asb_recover_0p(mdev);
if (hg == -1 && mdev->state.role == R_PRIMARY) {
- self = drbd_set_role(mdev, R_SECONDARY, 0);
+ enum drbd_state_rv rv2;
+
+ drbd_set_role(mdev, R_SECONDARY, 0);
/* drbd_change_state() does not sleep while in SS_IN_TRANSIENT_STATE,
* we might be here in C_WF_REPORT_PARAMS which is transient.
* we do not need to wait for the after state change work either. */
- self = drbd_change_state(mdev, CS_VERBOSE, NS(role, R_SECONDARY));
- if (self != SS_SUCCESS) {
+ rv2 = drbd_change_state(mdev, CS_VERBOSE, NS(role, R_SECONDARY));
+ if (rv2 != SS_SUCCESS) {
drbd_khelper(mdev, "pri-lost-after-sb");
} else {
dev_warn(DEV, "Successfully gave up primary role.\n");
static int drbd_asb_recover_2p(struct drbd_conf *mdev) __must_hold(local)
{
- int self, peer, hg, rv = -100;
-
- self = mdev->ldev->md.uuid[UI_BITMAP] & 1;
- peer = mdev->p_uuid[UI_BITMAP] & 1;
+ int hg, rv = -100;
switch (mdev->net_conf->after_sb_2p) {
case ASB_DISCARD_YOUNGER_PRI:
case ASB_CALL_HELPER:
hg = drbd_asb_recover_0p(mdev);
if (hg == -1) {
+ enum drbd_state_rv rv2;
+
/* drbd_change_state() does not sleep while in SS_IN_TRANSIENT_STATE,
* we might be here in C_WF_REPORT_PARAMS which is transient.
* we do not need to wait for the after state change work either. */
- self = drbd_change_state(mdev, CS_VERBOSE, NS(role, R_SECONDARY));
- if (self != SS_SUCCESS) {
+ rv2 = drbd_change_state(mdev, CS_VERBOSE, NS(role, R_SECONDARY));
+ if (rv2 != SS_SUCCESS) {
drbd_khelper(mdev, "pri-lost-after-sb");
} else {
dev_warn(DEV, "Successfully gave up primary role.\n");
-2 C_SYNC_TARGET set BitMap
-100 after split brain, disconnect
-1000 unrelated data
+-1091 requires proto 91
+-1096 requires proto 96
*/
static int drbd_uuid_compare(struct drbd_conf *mdev, int *rule_nr) __must_hold(local)
{
if (mdev->p_uuid[UI_BITMAP] == (u64)0 && mdev->ldev->md.uuid[UI_BITMAP] != (u64)0) {
if (mdev->agreed_pro_version < 91)
- return -1001;
+ return -1091;
if ((mdev->ldev->md.uuid[UI_BITMAP] & ~((u64)1)) == (mdev->p_uuid[UI_HISTORY_START] & ~((u64)1)) &&
(mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (mdev->p_uuid[UI_HISTORY_START + 1] & ~((u64)1))) {
if (mdev->ldev->md.uuid[UI_BITMAP] == (u64)0 && mdev->p_uuid[UI_BITMAP] != (u64)0) {
if (mdev->agreed_pro_version < 91)
- return -1001;
+ return -1091;
if ((mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (mdev->p_uuid[UI_BITMAP] & ~((u64)1)) &&
(mdev->ldev->md.uuid[UI_HISTORY_START + 1] & ~((u64)1)) == (mdev->p_uuid[UI_HISTORY_START] & ~((u64)1))) {
*rule_nr = 51;
peer = mdev->p_uuid[UI_HISTORY_START] & ~((u64)1);
if (self == peer) {
- self = mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1);
- peer = mdev->p_uuid[UI_HISTORY_START + 1] & ~((u64)1);
- if (self == peer) {
+ if (mdev->agreed_pro_version < 96 ?
+ (mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) ==
+ (mdev->p_uuid[UI_HISTORY_START + 1] & ~((u64)1)) :
+ peer + UUID_NEW_BM_OFFSET == (mdev->p_uuid[UI_BITMAP] & ~((u64)1))) {
/* The last P_SYNC_UUID did not get though. Undo the last start of
resync as sync source modifications of the peer's UUIDs. */
if (mdev->agreed_pro_version < 91)
- return -1001;
+ return -1091;
mdev->p_uuid[UI_BITMAP] = mdev->p_uuid[UI_HISTORY_START];
mdev->p_uuid[UI_HISTORY_START] = mdev->p_uuid[UI_HISTORY_START + 1];
+
+ dev_info(DEV, "Did not got last syncUUID packet, corrected:\n");
+ drbd_uuid_dump(mdev, "peer", mdev->p_uuid, mdev->p_uuid[UI_SIZE], mdev->p_uuid[UI_FLAGS]);
+
return -1;
}
}
*rule_nr = 71;
self = mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1);
if (self == peer) {
- self = mdev->ldev->md.uuid[UI_HISTORY_START + 1] & ~((u64)1);
- peer = mdev->p_uuid[UI_HISTORY_START] & ~((u64)1);
- if (self == peer) {
+ if (mdev->agreed_pro_version < 96 ?
+ (mdev->ldev->md.uuid[UI_HISTORY_START + 1] & ~((u64)1)) ==
+ (mdev->p_uuid[UI_HISTORY_START] & ~((u64)1)) :
+ self + UUID_NEW_BM_OFFSET == (mdev->ldev->md.uuid[UI_BITMAP] & ~((u64)1))) {
/* The last P_SYNC_UUID did not get though. Undo the last start of
resync as sync source modifications of our UUIDs. */
if (mdev->agreed_pro_version < 91)
- return -1001;
+ return -1091;
_drbd_uuid_set(mdev, UI_BITMAP, mdev->ldev->md.uuid[UI_HISTORY_START]);
_drbd_uuid_set(mdev, UI_HISTORY_START, mdev->ldev->md.uuid[UI_HISTORY_START + 1]);
- dev_info(DEV, "Undid last start of resync:\n");
-
+ dev_info(DEV, "Last syncUUID did not get through, corrected:\n");
drbd_uuid_dump(mdev, "self", mdev->ldev->md.uuid,
mdev->state.disk >= D_NEGOTIATING ? drbd_bm_total_weight(mdev) : 0, 0);
dev_alert(DEV, "Unrelated data, aborting!\n");
return C_MASK;
}
- if (hg == -1001) {
- dev_alert(DEV, "To resolve this both sides have to support at least protocol\n");
+ if (hg < -1000) {
+ dev_alert(DEV, "To resolve this both sides have to support at least protocol %d\n", -hg - 1000);
return C_MASK;
}
if (abs(hg) >= 2) {
dev_info(DEV, "Writing the whole bitmap, full sync required after drbd_sync_handshake.\n");
- if (drbd_bitmap_io(mdev, &drbd_bmio_set_n_write, "set_n_write from sync_handshake"))
+ if (drbd_bitmap_io(mdev, &drbd_bmio_set_n_write, "set_n_write from sync_handshake",
+ BM_LOCKED_SET_ALLOWED))
return C_MASK;
}
unsigned char *my_alg = mdev->net_conf->integrity_alg;
if (drbd_recv(mdev, p_integrity_alg, data_size) != data_size)
- return FALSE;
+ return false;
p_integrity_alg[SHARED_SECRET_MAX-1] = 0;
if (strcmp(p_integrity_alg, my_alg)) {
my_alg[0] ? my_alg : (unsigned char *)"<not-used>");
}
- return TRUE;
+ return true;
disconnect:
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
- return FALSE;
+ return false;
}
/* helper function
static int receive_SyncParam(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int packet_size)
{
- int ok = TRUE;
+ int ok = true;
struct p_rs_param_95 *p = &mdev->data.rbuf.rs_param_95;
unsigned int header_size, data_size, exp_max_sz;
struct crypto_hash *verify_tfm = NULL;
if (packet_size > exp_max_sz) {
dev_err(DEV, "SyncParam packet too long: received %u, expected <= %u bytes\n",
packet_size, exp_max_sz);
- return FALSE;
+ return false;
}
if (apv <= 88) {
memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX);
if (drbd_recv(mdev, &p->head.payload, header_size) != header_size)
- return FALSE;
+ return false;
mdev->sync_conf.rate = be32_to_cpu(p->rate);
dev_err(DEV, "verify-alg too long, "
"peer wants %u, accepting only %u byte\n",
data_size, SHARED_SECRET_MAX);
- return FALSE;
+ return false;
}
if (drbd_recv(mdev, p->verify_alg, data_size) != data_size)
- return FALSE;
+ return false;
/* we expect NUL terminated string */
/* but just in case someone tries to be evil */
/* but free the verify_tfm again, if csums_tfm did not work out */
crypto_free_hash(verify_tfm);
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
- return FALSE;
+ return false;
}
static void drbd_setup_order_type(struct drbd_conf *mdev, int peer)
if (p_size == 0 && mdev->state.disk == D_DISKLESS) {
dev_err(DEV, "some backing storage is needed\n");
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
- return FALSE;
+ return false;
}
/* just store the peer's disk size for now.
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
mdev->ldev->dc.disk_size = my_usize;
put_ldev(mdev);
- return FALSE;
+ return false;
}
put_ldev(mdev);
}
-#undef min_not_zero
ddsf = be16_to_cpu(p->dds_flags);
if (get_ldev(mdev)) {
dd = drbd_determin_dev_size(mdev, ddsf);
put_ldev(mdev);
if (dd == dev_size_error)
- return FALSE;
+ return false;
drbd_md_sync(mdev);
} else {
/* I am diskless, need to accept the peer's size. */
}
}
- return TRUE;
+ return true;
}
static int receive_uuids(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size)
{
struct p_uuids *p = &mdev->data.rbuf.uuids;
u64 *p_uuid;
- int i;
+ int i, updated_uuids = 0;
p_uuid = kmalloc(sizeof(u64)*UI_EXTENDED_SIZE, GFP_NOIO);
dev_err(DEV, "Can only connect to data with current UUID=%016llX\n",
(unsigned long long)mdev->ed_uuid);
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
- return FALSE;
+ return false;
}
if (get_ldev(mdev)) {
if (skip_initial_sync) {
dev_info(DEV, "Accepted new current UUID, preparing to skip initial sync\n");
drbd_bitmap_io(mdev, &drbd_bmio_clear_n_write,
- "clear_n_write from receive_uuids");
+ "clear_n_write from receive_uuids",
+ BM_LOCKED_TEST_ALLOWED);
_drbd_uuid_set(mdev, UI_CURRENT, p_uuid[UI_CURRENT]);
_drbd_uuid_set(mdev, UI_BITMAP, 0);
_drbd_set_state(_NS2(mdev, disk, D_UP_TO_DATE, pdsk, D_UP_TO_DATE),
CS_VERBOSE, NULL);
drbd_md_sync(mdev);
+ updated_uuids = 1;
}
put_ldev(mdev);
} else if (mdev->state.disk < D_INCONSISTENT &&
mdev->state.role == R_PRIMARY) {
/* I am a diskless primary, the peer just created a new current UUID
for me. */
- drbd_set_ed_uuid(mdev, p_uuid[UI_CURRENT]);
+ updated_uuids = drbd_set_ed_uuid(mdev, p_uuid[UI_CURRENT]);
}
/* Before we test for the disk state, we should wait until an eventually
new disk state... */
wait_event(mdev->misc_wait, !test_bit(CLUSTER_ST_CHANGE, &mdev->flags));
if (mdev->state.conn >= C_CONNECTED && mdev->state.disk < D_INCONSISTENT)
- drbd_set_ed_uuid(mdev, p_uuid[UI_CURRENT]);
+ updated_uuids |= drbd_set_ed_uuid(mdev, p_uuid[UI_CURRENT]);
- return TRUE;
+ if (updated_uuids)
+ drbd_print_uuids(mdev, "receiver updated UUIDs to");
+
+ return true;
}
/**
{
struct p_req_state *p = &mdev->data.rbuf.req_state;
union drbd_state mask, val;
- int rv;
+ enum drbd_state_rv rv;
mask.i = be32_to_cpu(p->mask);
val.i = be32_to_cpu(p->val);
if (test_bit(DISCARD_CONCURRENT, &mdev->flags) &&
test_bit(CLUSTER_ST_CHANGE, &mdev->flags)) {
drbd_send_sr_reply(mdev, SS_CONCURRENT_ST_CHG);
- return TRUE;
+ return true;
}
mask = convert_state(mask);
drbd_send_sr_reply(mdev, rv);
drbd_md_sync(mdev);
- return TRUE;
+ return true;
}
static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size)
peer_state.conn == C_CONNECTED) {
if (drbd_bm_total_weight(mdev) <= mdev->rs_failed)
drbd_resync_finished(mdev);
- return TRUE;
+ return true;
}
}
real_peer_disk = D_DISKLESS;
} else {
if (test_and_clear_bit(CONN_DRY_RUN, &mdev->flags))
- return FALSE;
+ return false;
D_ASSERT(os.conn == C_WF_REPORT_PARAMS);
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
- return FALSE;
+ return false;
}
}
}
drbd_uuid_new_current(mdev);
clear_bit(NEW_CUR_UUID, &mdev->flags);
drbd_force_state(mdev, NS2(conn, C_PROTOCOL_ERROR, susp, 0));
- return FALSE;
+ return false;
}
rv = _drbd_set_state(mdev, ns, cs_flags, NULL);
ns = mdev->state;
if (rv < SS_SUCCESS) {
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
- return FALSE;
+ return false;
}
if (os.conn > C_WF_REPORT_PARAMS) {
drbd_md_sync(mdev); /* update connected indicator, la_size, ... */
- return TRUE;
+ return true;
}
static int receive_sync_uuid(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size)
_drbd_uuid_set(mdev, UI_CURRENT, be64_to_cpu(p->uuid));
_drbd_uuid_set(mdev, UI_BITMAP, 0UL);
+ drbd_print_uuids(mdev, "updated sync uuid");
drbd_start_resync(mdev, C_SYNC_TARGET);
put_ldev(mdev);
} else
dev_err(DEV, "Ignoring SyncUUID packet!\n");
- return TRUE;
+ return true;
}
-enum receive_bitmap_ret { OK, DONE, FAILED };
-
-static enum receive_bitmap_ret
+/**
+ * receive_bitmap_plain
+ *
+ * Return 0 when done, 1 when another iteration is needed, and a negative error
+ * code upon failure.
+ */
+static int
receive_bitmap_plain(struct drbd_conf *mdev, unsigned int data_size,
unsigned long *buffer, struct bm_xfer_ctx *c)
{
unsigned num_words = min_t(size_t, BM_PACKET_WORDS, c->bm_words - c->word_offset);
unsigned want = num_words * sizeof(long);
+ int err;
if (want != data_size) {
dev_err(DEV, "%s:want (%u) != data_size (%u)\n", __func__, want, data_size);
- return FAILED;
+ return -EIO;
}
if (want == 0)
- return DONE;
- if (drbd_recv(mdev, buffer, want) != want)
- return FAILED;
+ return 0;
+ err = drbd_recv(mdev, buffer, want);
+ if (err != want) {
+ if (err >= 0)
+ err = -EIO;
+ return err;
+ }
drbd_bm_merge_lel(mdev, c->word_offset, num_words, buffer);
if (c->bit_offset > c->bm_bits)
c->bit_offset = c->bm_bits;
- return OK;
+ return 1;
}
-static enum receive_bitmap_ret
+/**
+ * recv_bm_rle_bits
+ *
+ * Return 0 when done, 1 when another iteration is needed, and a negative error
+ * code upon failure.
+ */
+static int
recv_bm_rle_bits(struct drbd_conf *mdev,
struct p_compressed_bm *p,
struct bm_xfer_ctx *c)
bits = bitstream_get_bits(&bs, &look_ahead, 64);
if (bits < 0)
- return FAILED;
+ return -EIO;
for (have = bits; have > 0; s += rl, toggle = !toggle) {
bits = vli_decode_bits(&rl, look_ahead);
if (bits <= 0)
- return FAILED;
+ return -EIO;
if (toggle) {
e = s + rl -1;
if (e >= c->bm_bits) {
dev_err(DEV, "bitmap overflow (e:%lu) while decoding bm RLE packet\n", e);
- return FAILED;
+ return -EIO;
}
_drbd_bm_set_bits(mdev, s, e);
}
have, bits, look_ahead,
(unsigned int)(bs.cur.b - p->code),
(unsigned int)bs.buf_len);
- return FAILED;
+ return -EIO;
}
look_ahead >>= bits;
have -= bits;
bits = bitstream_get_bits(&bs, &tmp, 64 - have);
if (bits < 0)
- return FAILED;
+ return -EIO;
look_ahead |= tmp << have;
have += bits;
}
c->bit_offset = s;
bm_xfer_ctx_bit_to_word_offset(c);
- return (s == c->bm_bits) ? DONE : OK;
+ return (s != c->bm_bits);
}
-static enum receive_bitmap_ret
+/**
+ * decode_bitmap_c
+ *
+ * Return 0 when done, 1 when another iteration is needed, and a negative error
+ * code upon failure.
+ */
+static int
decode_bitmap_c(struct drbd_conf *mdev,
struct p_compressed_bm *p,
struct bm_xfer_ctx *c)
dev_err(DEV, "receive_bitmap_c: unknown encoding %u\n", p->encoding);
drbd_force_state(mdev, NS(conn, C_PROTOCOL_ERROR));
- return FAILED;
+ return -EIO;
}
void INFO_bm_xfer_stats(struct drbd_conf *mdev,
{
struct bm_xfer_ctx c;
void *buffer;
- enum receive_bitmap_ret ret;
- int ok = FALSE;
+ int err;
+ int ok = false;
struct p_header80 *h = &mdev->data.rbuf.header.h80;
- wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt));
-
- drbd_bm_lock(mdev, "receive bitmap");
+ drbd_bm_lock(mdev, "receive bitmap", BM_LOCKED_SET_ALLOWED);
+ /* you are supposed to send additional out-of-sync information
+ * if you actually set bits during this phase */
/* maybe we should use some per thread scratch page,
* and allocate that during initial device creation? */
.bm_words = drbd_bm_words(mdev),
};
- do {
+ for(;;) {
if (cmd == P_BITMAP) {
- ret = receive_bitmap_plain(mdev, data_size, buffer, &c);
+ err = receive_bitmap_plain(mdev, data_size, buffer, &c);
} else if (cmd == P_COMPRESSED_BITMAP) {
/* MAYBE: sanity check that we speak proto >= 90,
* and the feature is enabled! */
goto out;
if (data_size <= (sizeof(*p) - sizeof(p->head))) {
dev_err(DEV, "ReportCBitmap packet too small (l:%u)\n", data_size);
- return FAILED;
+ goto out;
}
- ret = decode_bitmap_c(mdev, p, &c);
+ err = decode_bitmap_c(mdev, p, &c);
} else {
dev_warn(DEV, "receive_bitmap: cmd neither ReportBitMap nor ReportCBitMap (is 0x%x)", cmd);
goto out;
c.packets[cmd == P_BITMAP]++;
c.bytes[cmd == P_BITMAP] += sizeof(struct p_header80) + data_size;
- if (ret != OK)
+ if (err <= 0) {
+ if (err < 0)
+ goto out;
break;
-
+ }
if (!drbd_recv_header(mdev, &cmd, &data_size))
goto out;
- } while (ret == OK);
- if (ret == FAILED)
- goto out;
+ }
INFO_bm_xfer_stats(mdev, "receive", &c);
if (mdev->state.conn == C_WF_BITMAP_T) {
+ enum drbd_state_rv rv;
+
ok = !drbd_send_bitmap(mdev);
if (!ok)
goto out;
/* Omit CS_ORDERED with this state transition to avoid deadlocks. */
- ok = _drbd_request_state(mdev, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE);
- D_ASSERT(ok == SS_SUCCESS);
+ rv = _drbd_request_state(mdev, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE);
+ D_ASSERT(rv == SS_SUCCESS);
} else if (mdev->state.conn != C_WF_BITMAP_S) {
/* admin may have requested C_DISCONNECTING,
* other threads may have noticed network errors */
drbd_conn_str(mdev->state.conn));
}
- ok = TRUE;
+ ok = true;
out:
drbd_bm_unlock(mdev);
if (ok && mdev->state.conn == C_WF_BITMAP_S)
* with the data requests being unplugged */
drbd_tcp_quickack(mdev->data.socket);
- return TRUE;
+ return true;
}
static int receive_out_of_sync(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size)
{
struct p_block_desc *p = &mdev->data.rbuf.block_desc;
+ switch (mdev->state.conn) {
+ case C_WF_SYNC_UUID:
+ case C_WF_BITMAP_T:
+ case C_BEHIND:
+ break;
+ default:
+ dev_err(DEV, "ASSERT FAILED cstate = %s, expected: WFSyncUUID|WFBitMapT|Behind\n",
+ drbd_conn_str(mdev->state.conn));
+ }
+
drbd_set_out_of_sync(mdev, be64_to_cpu(p->sector), be32_to_cpu(p->blksize));
- return TRUE;
+ return true;
}
typedef int (*drbd_cmd_handler_f)(struct drbd_conf *, enum drbd_packets cmd, unsigned int to_receive);
if (shs) {
rv = drbd_recv(mdev, &header->h80.payload, shs);
if (unlikely(rv != shs)) {
- dev_err(DEV, "short read while reading sub header: rv=%d\n", rv);
+ if (!signal_pending(current))
+ dev_warn(DEV, "short read while reading sub header: rv=%d\n", rv);
goto err_out;
}
}
if (mdev->state.conn == C_STANDALONE)
return;
- if (mdev->state.conn >= C_WF_CONNECTION)
- dev_err(DEV, "ASSERT FAILED cstate = %s, expected < WFConnection\n",
- drbd_conn_str(mdev->state.conn));
/* asender does not clean up anything. it must not interfere, either */
drbd_thread_stop(&mdev->asender);
atomic_set(&mdev->rs_pending_cnt, 0);
wake_up(&mdev->misc_wait);
+ del_timer(&mdev->request_timer);
+
/* make sure syncer is stopped and w_resume_next_sg queued */
del_timer_sync(&mdev->resync_timer);
resync_timer_fn((unsigned long)mdev);
if (os.conn == C_DISCONNECTING) {
wait_event(mdev->net_cnt_wait, atomic_read(&mdev->net_cnt) == 0);
- if (!is_susp(mdev->state)) {
- /* we must not free the tl_hash
- * while application io is still on the fly */
- wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt));
- drbd_free_tl_hash(mdev);
- }
-
crypto_free_hash(mdev->cram_hmac_tfm);
mdev->cram_hmac_tfm = NULL;
drbd_request_state(mdev, NS(conn, C_STANDALONE));
}
+ /* serialize with bitmap writeout triggered by the state change,
+ * if any. */
+ wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
+
/* tcp_close and release of sendpage pages can be deferred. I don't
* want to use SO_LINGER, because apparently it can be deferred for
* more than 20 seconds (longest time I checked).
rv = drbd_recv(mdev, &p->head.payload, expect);
if (rv != expect) {
- dev_err(DEV, "short read receiving handshake packet: l=%u\n", rv);
+ if (!signal_pending(current))
+ dev_warn(DEV, "short read receiving handshake packet: l=%u\n", rv);
return 0;
}
rv = drbd_recv(mdev, peers_ch, length);
if (rv != length) {
- dev_err(DEV, "short read AuthChallenge: l=%u\n", rv);
+ if (!signal_pending(current))
+ dev_warn(DEV, "short read AuthChallenge: l=%u\n", rv);
rv = 0;
goto fail;
}
rv = drbd_recv(mdev, response , resp_size);
if (rv != resp_size) {
- dev_err(DEV, "short read receiving AuthResponse: l=%u\n", rv);
+ if (!signal_pending(current))
+ dev_warn(DEV, "short read receiving AuthResponse: l=%u\n", rv);
rv = 0;
goto fail;
}
h = drbd_connect(mdev);
if (h == 0) {
drbd_disconnect(mdev);
- __set_current_state(TASK_INTERRUPTIBLE);
- schedule_timeout(HZ);
+ schedule_timeout_interruptible(HZ);
}
if (h == -1) {
dev_warn(DEV, "Discarding network configuration.\n");
}
wake_up(&mdev->state_wait);
- return TRUE;
+ return true;
}
static int got_Ping(struct drbd_conf *mdev, struct p_header80 *h)
if (!test_and_set_bit(GOT_PING_ACK, &mdev->flags))
wake_up(&mdev->misc_wait);
- return TRUE;
+ return true;
}
static int got_IsInSync(struct drbd_conf *mdev, struct p_header80 *h)
dec_rs_pending(mdev);
atomic_add(blksize >> 9, &mdev->rs_sect_in);
- return TRUE;
+ return true;
}
/* when we receive the ACK for a write request,
return req;
}
}
- dev_err(DEV, "_ack_id_to_req: failed to find req %p, sector %llus in list\n",
- (void *)(unsigned long)id, (unsigned long long)sector);
return NULL;
}
req = validator(mdev, id, sector);
if (unlikely(!req)) {
spin_unlock_irq(&mdev->req_lock);
- dev_err(DEV, "%s: got a corrupt block_id/sector pair\n", func);
- return FALSE;
+
+ dev_err(DEV, "%s: failed to find req %p, sector %llus\n", func,
+ (void *)(unsigned long)id, (unsigned long long)sector);
+ return false;
}
__req_mod(req, what, &m);
spin_unlock_irq(&mdev->req_lock);
if (m.bio)
complete_master_bio(mdev, &m);
- return TRUE;
+ return true;
}
static int got_BlockAck(struct drbd_conf *mdev, struct p_header80 *h)
if (is_syncer_block_id(p->block_id)) {
drbd_set_in_sync(mdev, sector, blksize);
dec_rs_pending(mdev);
- return TRUE;
+ return true;
}
switch (be16_to_cpu(h->command)) {
case P_RS_WRITE_ACK:
break;
default:
D_ASSERT(0);
- return FALSE;
+ return false;
}
return validate_req_change_req_state(mdev, p->block_id, sector,
{
struct p_block_ack *p = (struct p_block_ack *)h;
sector_t sector = be64_to_cpu(p->sector);
-
- if (__ratelimit(&drbd_ratelimit_state))
- dev_warn(DEV, "Got NegAck packet. Peer is in troubles?\n");
+ int size = be32_to_cpu(p->blksize);
+ struct drbd_request *req;
+ struct bio_and_error m;
update_peer_seq(mdev, be32_to_cpu(p->seq_num));
if (is_syncer_block_id(p->block_id)) {
- int size = be32_to_cpu(p->blksize);
dec_rs_pending(mdev);
drbd_rs_failed_io(mdev, sector, size);
- return TRUE;
+ return true;
}
- return validate_req_change_req_state(mdev, p->block_id, sector,
- _ack_id_to_req, __func__ , neg_acked);
+
+ spin_lock_irq(&mdev->req_lock);
+ req = _ack_id_to_req(mdev, p->block_id, sector);
+ if (!req) {
+ spin_unlock_irq(&mdev->req_lock);
+ if (mdev->net_conf->wire_protocol == DRBD_PROT_A ||
+ mdev->net_conf->wire_protocol == DRBD_PROT_B) {
+ /* Protocol A has no P_WRITE_ACKs, but has P_NEG_ACKs.
+ The master bio might already be completed, therefore the
+ request is no longer in the collision hash.
+ => Do not try to validate block_id as request. */
+ /* In Protocol B we might already have got a P_RECV_ACK
+ but then get a P_NEG_ACK after wards. */
+ drbd_set_out_of_sync(mdev, sector, size);
+ return true;
+ } else {
+ dev_err(DEV, "%s: failed to find req %p, sector %llus\n", __func__,
+ (void *)(unsigned long)p->block_id, (unsigned long long)sector);
+ return false;
+ }
+ }
+ __req_mod(req, neg_acked, &m);
+ spin_unlock_irq(&mdev->req_lock);
+
+ if (m.bio)
+ complete_master_bio(mdev, &m);
+ return true;
}
static int got_NegDReply(struct drbd_conf *mdev, struct p_header80 *h)
if (get_ldev_if_state(mdev, D_FAILED)) {
drbd_rs_complete_io(mdev, sector);
- drbd_rs_failed_io(mdev, sector, size);
+ switch (be16_to_cpu(h->command)) {
+ case P_NEG_RS_DREPLY:
+ drbd_rs_failed_io(mdev, sector, size);
+ case P_RS_CANCEL:
+ break;
+ default:
+ D_ASSERT(0);
+ put_ldev(mdev);
+ return false;
+ }
put_ldev(mdev);
}
- return TRUE;
+ return true;
}
static int got_BarrierAck(struct drbd_conf *mdev, struct p_header80 *h)
if (mdev->state.conn == C_AHEAD &&
atomic_read(&mdev->ap_in_flight) == 0 &&
- list_empty(&mdev->start_resync_work.list)) {
- struct drbd_work *w = &mdev->start_resync_work;
- w->cb = w_start_resync;
- drbd_queue_work_front(&mdev->data.work, w);
+ !test_and_set_bit(AHEAD_TO_SYNC_SOURCE, &mdev->current_epoch->flags)) {
+ mdev->start_resync_timer.expires = jiffies + HZ;
+ add_timer(&mdev->start_resync_timer);
}
- return TRUE;
+ return true;
}
static int got_OVResult(struct drbd_conf *mdev, struct p_header80 *h)
ov_oos_print(mdev);
if (!get_ldev(mdev))
- return TRUE;
+ return true;
drbd_rs_complete_io(mdev, sector);
dec_rs_pending(mdev);
}
}
put_ldev(mdev);
- return TRUE;
+ return true;
}
static int got_skip(struct drbd_conf *mdev, struct p_header80 *h)
{
- return TRUE;
+ return true;
}
struct asender_cmd {
[P_STATE_CHG_REPLY] = { sizeof(struct p_req_state_reply), got_RqSReply },
[P_RS_IS_IN_SYNC] = { sizeof(struct p_block_ack), got_IsInSync },
[P_DELAY_PROBE] = { sizeof(struct p_delay_probe93), got_skip },
+ [P_RS_CANCEL] = { sizeof(struct p_block_ack), got_NegRSDReply},
[P_MAX_CMD] = { 0, NULL },
};
if (cmd > P_MAX_CMD || asender_tbl[cmd].process == NULL)
projects / GitHub / mt8127 / android_kernel_alcatel_ttab.git / blobdiff