return NULL;
}
-/* kick lower level device, if we have more than (arbitrary number)
- * reference counts on it, which typically are locally submitted io
- * requests. don't use unacked_cnt, so we speed up proto A and B, too. */
-static void maybe_kick_lo(struct drbd_conf *mdev)
-{
- if (atomic_read(&mdev->local_cnt) >= mdev->net_conf->unplug_watermark)
- drbd_kick_lo(mdev);
-}
-
static void reclaim_net_ee(struct drbd_conf *mdev, struct list_head *to_be_freed)
{
struct drbd_epoch_entry *e;
LIST_HEAD(reclaimed);
struct drbd_epoch_entry *e, *t;
- maybe_kick_lo(mdev);
spin_lock_irq(&mdev->req_lock);
reclaim_net_ee(mdev, &reclaimed);
spin_unlock_irq(&mdev->req_lock);
atomic_t *a = is_net ? &mdev->pp_in_use_by_net : &mdev->pp_in_use;
int i;
- if (drbd_pp_vacant > (DRBD_MAX_SEGMENT_SIZE/PAGE_SIZE)*minor_count)
+ if (drbd_pp_vacant > (DRBD_MAX_BIO_SIZE/PAGE_SIZE)*minor_count)
i = page_chain_free(page);
else {
struct page *tmp;
while (!list_empty(head)) {
prepare_to_wait(&mdev->ee_wait, &wait, TASK_UNINTERRUPTIBLE);
spin_unlock_irq(&mdev->req_lock);
- drbd_kick_lo(mdev);
- schedule();
+ io_schedule();
finish_wait(&mdev->ee_wait, &wait);
spin_lock_irq(&mdev->req_lock);
}
/* > e->sector, unless this is the first bio */
bio->bi_sector = sector;
bio->bi_bdev = mdev->ldev->backing_bdev;
- /* we special case some flags in the multi-bio case, see below
- * (REQ_UNPLUG) */
bio->bi_rw = rw;
bio->bi_private = e;
bio->bi_end_io = drbd_endio_sec;
bios = bios->bi_next;
bio->bi_next = NULL;
- /* strip off REQ_UNPLUG unless it is the last bio */
- if (bios)
- bio->bi_rw &= ~REQ_UNPLUG;
-
drbd_generic_make_request(mdev, fault_type, bio);
} while (bios);
- maybe_kick_lo(mdev);
return 0;
fail:
inc_unacked(mdev);
- if (mdev->net_conf->wire_protocol != DRBD_PROT_C)
- drbd_kick_lo(mdev);
-
mdev->current_epoch->barrier_nr = p->barrier;
rv = drbd_may_finish_epoch(mdev, mdev->current_epoch, EV_GOT_BARRIER_NR);
data_size -= dgs;
ERR_IF(data_size & 0x1ff) return NULL;
- ERR_IF(data_size > DRBD_MAX_SEGMENT_SIZE) 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 (dgs) {
drbd_csum_ee(mdev, mdev->integrity_r_tfm, e, dig_vv);
if (memcmp(dig_in, dig_vv, dgs)) {
- dev_err(DEV, "Digest integrity check FAILED.\n");
+ dev_err(DEV, "Digest integrity check FAILED: %llus +%u\n",
+ (unsigned long long)sector, data_size);
drbd_bcast_ee(mdev, "digest failed",
dgs, dig_in, dig_vv, e);
drbd_free_ee(mdev, e);
return ret;
}
-static unsigned long write_flags_to_bio(struct drbd_conf *mdev, u32 dpf)
+/* see also bio_flags_to_wire()
+ * DRBD_REQ_*, because we need to semantically map the flags to data packet
+ * flags and back. We may replicate to other kernel versions. */
+static unsigned long wire_flags_to_bio(struct drbd_conf *mdev, u32 dpf)
{
- if (mdev->agreed_pro_version >= 95)
- return (dpf & DP_RW_SYNC ? REQ_SYNC : 0) |
- (dpf & DP_UNPLUG ? REQ_UNPLUG : 0) |
- (dpf & DP_FUA ? REQ_FUA : 0) |
- (dpf & DP_FLUSH ? REQ_FUA : 0) |
- (dpf & DP_DISCARD ? REQ_DISCARD : 0);
- else
- return dpf & DP_RW_SYNC ? (REQ_SYNC | REQ_UNPLUG) : 0;
+ return (dpf & DP_RW_SYNC ? REQ_SYNC : 0) |
+ (dpf & DP_FUA ? REQ_FUA : 0) |
+ (dpf & DP_FLUSH ? REQ_FLUSH : 0) |
+ (dpf & DP_DISCARD ? REQ_DISCARD : 0);
}
/* mirrored write */
e->w.cb = e_end_block;
+ dp_flags = be32_to_cpu(p->dp_flags);
+ rw |= wire_flags_to_bio(mdev, dp_flags);
+
+ if (dp_flags & DP_MAY_SET_IN_SYNC)
+ e->flags |= EE_MAY_SET_IN_SYNC;
+
spin_lock(&mdev->epoch_lock);
e->epoch = mdev->current_epoch;
atomic_inc(&e->epoch->epoch_size);
atomic_inc(&e->epoch->active);
spin_unlock(&mdev->epoch_lock);
- dp_flags = be32_to_cpu(p->dp_flags);
- rw |= write_flags_to_bio(mdev, dp_flags);
-
- if (dp_flags & DP_MAY_SET_IN_SYNC)
- e->flags |= EE_MAY_SET_IN_SYNC;
-
/* I'm the receiver, I do hold a net_cnt reference. */
if (!mdev->net_conf->two_primaries) {
spin_lock_irq(&mdev->req_lock);
/* sync speed average over the last 2*DRBD_SYNC_MARK_STEP,
* approx. */
- i = (mdev->rs_last_mark + DRBD_SYNC_MARKS-2) % DRBD_SYNC_MARKS;
- rs_left = drbd_bm_total_weight(mdev) - mdev->rs_failed;
+ i = (mdev->rs_last_mark + DRBD_SYNC_MARKS-1) % DRBD_SYNC_MARKS;
+
+ if (mdev->state.conn == C_VERIFY_S || mdev->state.conn == C_VERIFY_T)
+ rs_left = mdev->ov_left;
+ else
+ rs_left = drbd_bm_total_weight(mdev) - mdev->rs_failed;
dt = ((long)jiffies - (long)mdev->rs_mark_time[i]) / HZ;
if (!dt)
sector = be64_to_cpu(p->sector);
size = be32_to_cpu(p->blksize);
- if (size <= 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_SEGMENT_SIZE) {
+ 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;
case P_RS_DATA_REQUEST:
e->w.cb = w_e_end_rsdata_req;
fault_type = DRBD_FAULT_RS_RD;
+ /* used in the sector offset progress display */
+ mdev->bm_resync_fo = BM_SECT_TO_BIT(sector);
break;
case P_OV_REPLY:
if (cmd == P_CSUM_RS_REQUEST) {
D_ASSERT(mdev->agreed_pro_version >= 89);
e->w.cb = w_e_end_csum_rs_req;
+ /* used in the sector offset progress display */
+ mdev->bm_resync_fo = BM_SECT_TO_BIT(sector);
} else if (cmd == P_OV_REPLY) {
+ /* track progress, we may need to throttle */
+ atomic_add(size >> 9, &mdev->rs_sect_in);
e->w.cb = w_e_end_ov_reply;
dec_rs_pending(mdev);
/* drbd_rs_begin_io done when we sent this request,
case P_OV_REQUEST:
if (mdev->ov_start_sector == ~(sector_t)0 &&
mdev->agreed_pro_version >= 90) {
+ unsigned long now = jiffies;
+ int i;
mdev->ov_start_sector = sector;
mdev->ov_position = sector;
- mdev->ov_left = mdev->rs_total - BM_SECT_TO_BIT(sector);
+ mdev->ov_left = drbd_bm_bits(mdev) - BM_SECT_TO_BIT(sector);
+ mdev->rs_total = mdev->ov_left;
+ for (i = 0; i < DRBD_SYNC_MARKS; i++) {
+ mdev->rs_mark_left[i] = mdev->ov_left;
+ mdev->rs_mark_time[i] = now;
+ }
dev_info(DEV, "Online Verify start sector: %llu\n",
(unsigned long long)sector);
}
{
struct p_sizes *p = &mdev->data.rbuf.sizes;
enum determine_dev_size dd = unchanged;
- unsigned int max_seg_s;
+ unsigned int max_bio_size;
sector_t p_size, p_usize, my_usize;
int ldsc = 0; /* local disk size changed */
enum dds_flags ddsf;
}
if (mdev->agreed_pro_version < 94)
- max_seg_s = be32_to_cpu(p->max_segment_size);
+ max_bio_size = be32_to_cpu(p->max_bio_size);
else if (mdev->agreed_pro_version == 94)
- max_seg_s = DRBD_MAX_SIZE_H80_PACKET;
+ max_bio_size = DRBD_MAX_SIZE_H80_PACKET;
else /* drbd 8.3.8 onwards */
- max_seg_s = DRBD_MAX_SEGMENT_SIZE;
+ max_bio_size = DRBD_MAX_BIO_SIZE;
- if (max_seg_s != queue_max_segment_size(mdev->rq_queue))
- drbd_setup_queue_param(mdev, max_seg_s);
+ if (max_bio_size != queue_max_hw_sectors(mdev->rq_queue) << 9)
+ drbd_setup_queue_param(mdev, max_bio_size);
drbd_setup_order_type(mdev, be16_to_cpu(p->queue_order_type));
put_ldev(mdev);
static int receive_UnplugRemote(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size)
{
- if (mdev->state.disk >= D_INCONSISTENT)
- drbd_kick_lo(mdev);
-
/* Make sure we've acked all the TCP data associated
* with the data requests being unplugged */
drbd_tcp_quickack(mdev->data.socket);
drbd_rs_complete_io(mdev, sector);
dec_rs_pending(mdev);
- if (--mdev->ov_left == 0) {
+ --mdev->ov_left;
+
+ /* let's advance progress step marks only for every other megabyte */
+ if ((mdev->ov_left & 0x200) == 0x200)
+ drbd_advance_rs_marks(mdev, mdev->ov_left);
+
+ if (mdev->ov_left == 0) {
w = kmalloc(sizeof(*w), GFP_NOIO);
if (w) {
w->cb = w_ov_finished;
projects / mirror_ubuntu-artful-kernel.git / blobdiff