leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

drbd: queue bitmap writeout more intelligently

Browse Source 
The "lazy writeout" of cleared bitmap pages happens during resync, and
should happen again once the resync finishes cleanly, or is aborted.

If resync finished cleanly, or was aborted because of peer disk
failure, we trigger the writeout from worker context in the after
state change work.

If resync was aborted because of connection failure, we should not
immediately trigger bitmap writeout, but rather postpone the
writeout to after the connection cleanup happened.  We now do it
in the receiver context from drbd_disconnect().

If resync was aborted because of local disk failure, well, there
is nothing to write to anymore.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
This commit is contained in:
Lars Ellenberg 2011-01-20 10:32:05 +01:00 committed by Philipp Reisner
parent 54b956abef
commit 79a30d2d71
2 changed files with 12 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
drivers/block/drbd/drbd_main.c
View File

@ -1593,8 +1593,18 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
if (os.disk < D_UP_TO_DATE && os.conn >= C_SYNC_SOURCE && ns.conn == C_CONNECTED)
drbd_send_state(mdev);
if (os.conn > C_CONNECTED && ns.conn <= C_CONNECTED)
/* This triggers bitmap writeout of potentially still unwritten pages
* if the resync finished cleanly, or aborted because of peer disk
* failure. Resync aborted because of connection failure does bitmap
* writeout from drbd_disconnect.
* For resync aborted because of local disk failure, we cannot do
* any bitmap writeout anymore.
*/
if (os.conn > C_CONNECTED && ns.conn == C_CONNECTED &&
mdev->state.conn == C_CONNECTED && get_ldev(mdev)) {
drbd_queue_bitmap_io(mdev, &drbd_bm_write, NULL, "write from resync_finished");
put_ldev(mdev);
}
/* free tl_hash if we Got thawed and are C_STANDALONE */
if (ns.conn == C_STANDALONE && !is_susp(ns) && mdev->tl_hash)

1
drivers/block/drbd/drbd_receiver.c
View File

@ -3811,6 +3811,7 @@ static void drbd_disconnect(struct drbd_conf *mdev)
fp = FP_DONT_CARE;
if (get_ldev(mdev)) {
drbd_bitmap_io(mdev, &drbd_bm_write, "write from disconnect");
fp = mdev->ldev->dc.fencing;
put_ldev(mdev);
}