From: Philipp Reisner <philipp.reisner@linbit.com>
To: linux-kernel@vger.kernel.org, Jens Axboe <axboe@kernel.dk>
Cc: drbd-dev@lists.linbit.com,
Lars Ellenberg <lars.ellenberg@linbit.com>,
Philipp Reisner <philipp.reisner@linbit.com>
Subject: [PATCH 13/17] drbd: fix drbd epoch write count for ahead/behind mode
Date: Wed, 27 Mar 2013 14:08:45 +0100 [thread overview]
Message-ID: <1364389729-17559-14-git-send-email-philipp.reisner@linbit.com> (raw)
In-Reply-To: <1364389729-17559-1-git-send-email-philipp.reisner@linbit.com>
From: Lars Ellenberg <lars.ellenberg@linbit.com>
The sanity check when receiving P_BARRIER_ACK does expect all write
requests with a given req->epoch to have been either all replicated,
or all not replicated.
Because req->epoch was assigned before calling maybe_pull_ahead(),
this expectation was not met, leading to an off-by-one in the sanity
check, and further to a "Protocol Error".
Fix: move the call to maybe_pull_ahead() a few lines up,
and assign req->epoch only after that.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
drivers/block/drbd/drbd_req.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index beefe65..c24379f 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -865,8 +865,10 @@ static void maybe_pull_ahead(struct drbd_conf *mdev)
bool congested = false;
enum drbd_on_congestion on_congestion;
+ rcu_read_lock();
nc = rcu_dereference(tconn->net_conf);
on_congestion = nc ? nc->on_congestion : OC_BLOCK;
+ rcu_read_unlock();
if (on_congestion == OC_BLOCK ||
tconn->agreed_pro_version < 96)
return;
@@ -960,14 +962,8 @@ static int drbd_process_write_request(struct drbd_request *req)
struct drbd_conf *mdev = req->w.mdev;
int remote, send_oos;
- rcu_read_lock();
remote = drbd_should_do_remote(mdev->state);
- if (remote) {
- maybe_pull_ahead(mdev);
- remote = drbd_should_do_remote(mdev->state);
- }
send_oos = drbd_should_send_out_of_sync(mdev->state);
- rcu_read_unlock();
/* Need to replicate writes. Unless it is an empty flush,
* which is better mapped to a DRBD P_BARRIER packet,
@@ -1087,9 +1083,13 @@ static void drbd_send_and_submit(struct drbd_conf *mdev, struct drbd_request *re
* but will re-aquire it before it returns here.
* Needs to be before the check on drbd_suspended() */
complete_conflicting_writes(req);
+ /* no more giving up req_lock from now on! */
+
+ /* check for congestion, and potentially stop sending
+ * full data updates, but start sending "dirty bits" only. */
+ maybe_pull_ahead(mdev);
}
- /* no more giving up req_lock from now on! */
if (drbd_suspended(mdev)) {
/* push back and retry: */
--
1.7.9.5
next prev parent reply other threads:[~2013-03-27 13:10 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-27 13:08 [PATCH 00/17] RFC: Pile of DRBD fixes Philipp Reisner
2013-03-27 13:08 ` [PATCH 01/17] idr: document exit conditions on idr_for_each_entry better Philipp Reisner
2013-03-27 13:08 ` [PATCH 02/17] drbd: reset ap_in_flight counter for new connections Philipp Reisner
2013-03-27 13:08 ` [PATCH 03/17] drbd: abort start of resync early, if it raced with connection breakage Philipp Reisner
2013-03-27 13:08 ` [PATCH 04/17] drbd: move invalidating the whole bitmap out of after_state ch() Philipp Reisner
2013-03-27 13:08 ` [PATCH 05/17] drbd: fix effective error returned when refusing an invalidate Philipp Reisner
2013-03-27 13:08 ` [PATCH 06/17] drbd: drop now useless duplicate state request from invalidate Philipp Reisner
2013-03-27 13:08 ` [PATCH 07/17] drbd: fix spurious warning about bitmap being locked from detach Philipp Reisner
2013-03-27 13:08 ` [PATCH 08/17] drbd: Fix disconnect to keep the peer disk state if connection breaks during operation Philipp Reisner
2013-03-27 13:08 ` [PATCH 09/17] drbd: only fail empty flushes if no good data is reachable Philipp Reisner
2013-03-27 13:08 ` [PATCH 10/17] drbd: fix memory leak Philipp Reisner
2013-03-27 13:08 ` [PATCH 11/17] drbd: validate resync_after dependency on attach already Philipp Reisner
2013-03-27 13:08 ` [PATCH 12/17] drbd: Fix build error when CONFIG_CRYPTO_HMAC is not set Philipp Reisner
2013-03-27 13:08 ` Philipp Reisner [this message]
2013-03-27 13:08 ` [PATCH 14/17] drbd: add module_put() on error path in drbd_proc_open() Philipp Reisner
2013-03-27 13:08 ` [PATCH 15/17] drbd: fix for deadlock when using automatic split-brain-recovery Philipp Reisner
2013-03-27 13:08 ` [PATCH 16/17] drbd: use sched_setscheduler() Philipp Reisner
2013-03-27 13:08 ` [PATCH 17/17] drbd: fix if(); found by kbuild test robot Philipp Reisner
2013-03-28 16:11 ` [PATCH 00/17] RFC: Pile of DRBD fixes Jens Axboe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1364389729-17559-14-git-send-email-philipp.reisner@linbit.com \
--to=philipp.reisner@linbit.com \
--cc=axboe@kernel.dk \
--cc=drbd-dev@lists.linbit.com \
--cc=lars.ellenberg@linbit.com \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome