diff options
| author | Philipp Reisner <philipp.reisner@linbit.com> | 2012-02-19 01:27:53 +0100 |
|---|---|---|
| committer | Philipp Reisner <philipp.reisner@linbit.com> | 2012-05-09 15:16:38 +0200 |
| commit | b6a370ba0786b5eb09c479bffeffe7baba484ab0 (patch) | |
| tree | f9c77b675c02ade44902532d9f62ea4e484de5cb /drivers/block/drbd | |
| parent | fc28845bc005995b41ae8c83c7922d088f0ad228 (diff) | |
drbd: Fix a potential write ordering issue on SyncTarget nodes
If a SyncTarget node gets a P_RS_DATA_REPLY before a P_DATA packet
for the same sector, it simply submits these two IO requests.
This is be possible because on the SyncSource node, the data of the
P_RS_DATA_REPLY packet was read from disk. Immediately after that a
write request from upper layers came in.
The disk scheduler or even the "hardware" queues on the disk drive might
reorder these writes.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Diffstat (limited to 'drivers/block/drbd')
| -rw-r--r-- | drivers/block/drbd/drbd_receiver.c | 21 |
1 files changed, 21 insertions, 0 deletions
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index 6b0505a3c4fc..d601501c336a 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -1585,6 +1585,24 @@ static int e_send_discard_ack(struct drbd_conf *mdev, struct drbd_work *w, int u return ok; } +static bool overlapping_resync_write(struct drbd_conf *mdev, struct drbd_epoch_entry *data_e) +{ + + struct drbd_epoch_entry *rs_e; + bool rv = 0; + + spin_lock_irq(&mdev->req_lock); + list_for_each_entry(rs_e, &mdev->sync_ee, w.list) { + if (overlaps(data_e->sector, data_e->size, rs_e->sector, rs_e->size)) { + rv = 1; + break; + } + } + spin_unlock_irq(&mdev->req_lock); + + return rv; +} + /* Called from receive_Data. * Synchronize packets on sock with packets on msock. * @@ -1828,6 +1846,9 @@ static int receive_Data(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned list_add(&e->w.list, &mdev->active_ee); spin_unlock_irq(&mdev->req_lock); + if (mdev->state.conn == C_SYNC_TARGET) + wait_event(mdev->ee_wait, !overlapping_resync_write(mdev, e)); + switch (mdev->net_conf->wire_protocol) { case DRBD_PROT_C: inc_unacked(mdev); |