Commit Graph

1909 Commits

Author SHA1 Message Date
Philipp Reisner
6a35c45f89 drbd: Ensure that an epoch contains only requests of one kind
The assert in drbd_req.c:755 forces us to have only requests of
one kind in an epoch. The two kinds we distinguish here are:
local-only or mirrored.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:42 +01:00
Philipp Reisner
2deb8336d0 drbd: Fixed P_NEG_ACK processing for protocol A and 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.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:40 +01:00
Philipp Reisner
94f2b05f03 drbd: Killed an assert that is no longer valid
The point is that drbd_disconnect() can be called with a cstate of
WFConnection.

That happens if the user issues "drbdsetup disconnect" while the
drbd_connect() function executes. Then drbdd_init() will call
drbdd(), which in turn will return without receiving any
packets. Then drbdd_init() will end up calling drbd_disconnect()
with a cstate of WFConnection.

Bottom line: This assertion is wrong as it is, and we do not
see value in fixing it. => Removing it.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:39 +01:00
Philipp Reisner
148efa165e drbd: Do not drop net config if sending in drbd_send_protocol() fails
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:37 +01:00
Philipp Reisner
370a43e798 drbd: Work on the Ahead -> SyncSource transition
The test if rs_pending_cnt == 0 was too weak. Using Test for
unacked_cnt == 0 instead. Moved that into the worker.

Since unacked_cnt gets already increased when an P_RS_DATA_REQ
comes in.

Also using a timer to make Ahead -> SyncSource -> Ahead cycles
slower...

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:36 +01:00
Philipp Reisner
71c78cfba2 drbd: Nothing should stop SyncSource -> Ahead transitions
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:34 +01:00
Philipp Reisner
4a23f26496 drbd: Do not full sync if a P_SYNC_UUID packet gets lost
See also commit from 2009-08-15
"drbd_uuid_compare(): Do not full sync in case a P_SYNC_UUID packet gets lost."

We saw cases where the History UUIDs where not as expected. So the
detection of the special case did not trigger. With the sync UUID
no longer being a random number, but deducible from the previous
bitmap UUID, the detection of this special case becomes more
reliable.

The SyncUUID now is the previous bitmap UUID + 0x1000000000000.

Rule 5a:
Cs = H1p & H1p + Offset = Bp
  Connection was lost before SyncUUID Packet came through.
  Corrent (peer) UUIDs:
   Bp = H1p
   H1p = H2p
   H2p = 0
  Become Sync target.

Rule 7a:
Cp = H1s & H1s + Offset = Bs
  Connection was lost before SyncUUID Packet came through.
  Correct (own) UUIDs:
   Bs = H1s
   H1s = H2s
   H2s = 0
  Become Sync source.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:32 +01:00
Philipp Reisner
2b8a90b555 drbd: Corrected off-by-one error in DRBD_MINOR_COUNT_MAX
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:31 +01:00
Andreas Gruenbacher
110a204a35 drbd: Remove useless / wrong comments
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:29 +01:00
Philipp Reisner
794abb753e drbd: Cleaned up the resync timer logic
Besides removed a few lines of code, this moves the inspection
of the state from before the queuing process to after the queuing.
I.e. more closely to the actual invocation of the work.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:28 +01:00
Philipp Reisner
da0a78161d drbd: Be more careful with SyncSource -> Ahead transitions
We may not get from SyncSource to Ahead if we have sent some
P_RS_DATA_REPLY packets to the peer and are waiting for
P_WRITE_ACK.

Again, this is not relevant for proper tuned systems, but makes
sure that the not-tuned system does not get diverging bitmaps.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:26 +01:00
Philipp Reisner
d612d309e4 drbd: No longer answer P_RS_DATA_REQUEST packets when in C_AHEAD mode
When the sync source node replies to a P_RS_DATA_REQUEST packet
when it is already in ahead mode. I.e. those two packets
crossed each other on the wire, that may lead to diverging
bitmaps.

  This never happens in a well-tuned-system. In a well-tuned-
  system the resync controller has reduced the resync speed
  to zero long before we got into ahead-mode.

But we have to be prepared for the not-well-tuned-system
of course as well.
Because -> diverging bitmaps = non terminating resync.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:25 +01:00
Philipp Reisner
617049aa7d drbd: Fixed an issue with AHEAD -> SYNC_SOURCE transitions
Create a new barrier when leaving the AHEAD mode.

  Otherwise we trigger the assertion in req_mod(, barrier_acked)
  D_ASSERT(req->rq_state & RQ_NET_SENT);

The new barrier is created by recycling the newest existing one.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:23 +01:00
Lars Ellenberg
0719427278 drbd: ratelimit io error messages
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:21 +01:00
Philipp Reisner
3f98688afc drbd: There might be a resync after unfreezing IO due to no disk [Bugz 332]
When on-no-data-accessible is set to suspend-io, also consider that
a Primary, SyncTarget node losses its connection.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:20 +01:00
Lars Ellenberg
725a97e43e drbd: fix potential access of on-stack wait_queue_head_t after return
I run into something declaring itself as "spinlock deadlock",
 BUG: spinlock lockup on CPU#1, kjournald/27816, ffff88000ad6bca0
 Pid: 27816, comm: kjournald Tainted: G        W 2.6.34.6 #2
 Call Trace:
  <IRQ>  [<ffffffff811ba0aa>] do_raw_spin_lock+0x11e/0x14d
  [<ffffffff81340fde>] _raw_spin_lock_irqsave+0x6a/0x81
  [<ffffffff8103b694>] ? __wake_up+0x22/0x50
  [<ffffffff8103b694>] __wake_up+0x22/0x50
  [<ffffffffa07ff661>] bm_async_io_complete+0x258/0x299 [drbd]
but the call traces do not fit at all,
all other cpus are cpu_idle.

I think it may be this race:

drbd_bm_write_page
 wait_queue_head_t io_wait;
 atomic_t in_flight;
 bm_async_io
  submit_bio
					bm_async_io_complete
					  if (atomic_dec_and_test(in_flight))
 wait_event(io_wait,
	atomic_read(in_flight) == 0)
 return
					    wake_up(io_wait)

The wake_up now accesses the wait_queue_head_t spinlock, which is no
longer valid, since the stack frame of drbd_bm_write_page has been
clobbered now.

Fix this by using struct completion, which does both the condition test
as well as the wake_up inside its spinlock, so this race cannot happen.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:45:08 +01:00
Lars Ellenberg
06d33e968d drbd: improve on bitmap write out timing
Even though we now track the need for bitmap writeout per bitmap page,
there is no need to trigger the writeout while a resync is going on.

Once the resync is finished (or aborted),
we trigger bitmap writeout anyways.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:40 +01:00
Lars Ellenberg
418e0a927d drbd: spelling fix in log message
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:38 +01:00
Lars Ellenberg
7648cdfe52 drbd: be less noisy with some log messages
We expect changes to a bitmap page in drbd_bm_write_page,
that's why we submit a copy page.

If a page changes during global writeout, that would be unexpected,
and reason to warn, though.

Also, often page writeout can be skipped (on activity log transactions
during normal operation, for example), no need to log that everytime.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:37 +01:00
Lars Ellenberg
5a22db8968 drbd: serialize sending of resync uuid with pending w_send_oos
To improve the latency of IO requests during bitmap exchange,
we recently allowed writes while waiting for the bitmap, sending "set
out-of-sync" information packets for any newly dirtied bits.

We have to make sure that the new resync-uuid does not overtake
these "set oos" packets. Once the resync-uuid is received, the
sync target starts the resync process, and expects the bitmap to
only be cleared, not re-set.

If we use this protocol extension, we queue the generation and sending
of the resync-uuid on the worker, which naturally serializes with all
previously queued packets.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:35 +01:00
Lars Ellenberg
f735e36354 drbd: add debugging assert to make sure the protocol is clean
We expect to only receive the recently introduced "set out of sync"
packets in specific states. If we receive them in different states, that
may confuse the resync process to the point where it won't terminate, or
think it made negative progress.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:34 +01:00
Philipp Reisner
c88d65e223 drbd: Documenting drbd_should_do_remote() and drbd_should_send_oos()
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:32 +01:00
Lars Ellenberg
2265b473ae drbd: fix potential dereference of NULL pointer
If drbd used to have crypto digest algorithms configured, then is being
unconfigured (but not unloaded), it frees the algorithms, but does not
reset the config.  If it then is reconfigured to use the very same
algorithm, it "forgot" to re-allocate the algorithms, thinking that the
config has not changed in that aspect.
It will then Oops on the first attempt to actually use those algorithms.

Fix this by resetting the config to defaults after cleanup.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:30 +01:00
Lars Ellenberg
02851e9f00 drbd: move bitmap write from resync_finished to after_state_change
We must not call it directly from resync_finished,
as we may be in either receiver or worker context there.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:29 +01:00
Lars Ellenberg
84e7c0f7d1 drbd: Removed a reference to debug macros removed long time ago
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:27 +01:00
Lars Ellenberg
6850c44214 drbd: get rid of unused debug code
Long time ago, we had paranoia code in the bitmap that allocated one
extra word, assigned a magic value, and checked on every occasion that
the magic value was still unchanged.

That debug code is unused, the extra long word complicates code a bit.
Get rid of it.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:26 +01:00
Lars Ellenberg
4b0715f096 drbd: allow petabyte storage on 64bit arch
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:24 +01:00
Lars Ellenberg
19f843aa08 drbd: bitmap keep track of changes vs on-disk bitmap
When we set or clear bits in a bitmap page,
also set a flag in the page->private pointer.

This allows us to skip writes of unchanged pages.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:43:19 +01:00
Lars Ellenberg
95a0f10cdd drbd: store in-core bitmap little endian, regardless of architecture
Our on-disk bitmap is a little endian bitstream.
Up to now, we have stored the in-core copy of that in
native endian, applying byte order conversion when necessary.

Instead, keep the bitmap pages little endian, as they are read from disk,
and use the generic_*_le_bit family of functions.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:40 +01:00
Lars Ellenberg
7777a8ba1f drbd: bitmap: don't count unused bits (fix non-terminating resync)
We trusted the on-disk bitmap to have unused bits cleared.
In case that is not true for whatever reason,
and we take a code path where the unused bits don't get cleared
elsewhere (bm_clear_surplus is not called), we may miscount the bits,
and get confused during resync, waiting for bits to get cleared that we
don't even use: the resync process would not terminate.

Fix this by masking out unused bits in __bm_count_bits.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:38 +01:00
Andreas Gruenbacher
1b881ef775 drbd: Rename __inc_ap_bio_cond to may_inc_ap_bio
The old name is confusing: the function does not increment anything.
Also rename _inc_ap_bio_cond to inc_ap_bio_cond: there is no need for
an underscore.
Finally, make it clear that these functions return boolean values.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:37 +01:00
Andreas Gruenbacher
24dccabb39 drbd: Fix: drbd_bitmap_io does not return an enum determine_dev_size
I guess bitmap I/O errors are supposed to cause drbd_determin_dev_size
to return dev_size_error.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:35 +01:00
Andreas Gruenbacher
2c46407d24 drbd: receive_bitmap_plain: Get rid of ugly and useless enum
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:34 +01:00
Andreas Gruenbacher
f70af118e3 drbd: send_bitmap_rle_or_plain: Get rid of ugly and useless enum
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:32 +01:00
Andreas Gruenbacher
78fcbdae22 drbd: receive_bitmap: Missing free_page() on error path
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:30 +01:00
Andreas Gruenbacher
de1f8e4a0a drbd: receive_bitmap: Avoid casting enum drbd_state_rv to int
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:29 +01:00
Andreas Gruenbacher
4114be815f drbd: receive_bitmap: Fix the wrong return value
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:27 +01:00
Andreas Gruenbacher
f2024e7ce2 drbd: drbd_nl_disk_conf: Avoid a compiler warning
Warning: comparison between ‘enum drbd_ret_code’ and ‘enum drbd_state_rv’

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:26 +01:00
Andreas Gruenbacher
81e84650c2 drbd: Use the standard bool, true, and false keywords
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:24 +01:00
Andreas Gruenbacher
6184ea2145 drbd: This code is dead now
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:22 +01:00
Andreas Gruenbacher
bb4379464e drbd: Another small enum drbd_state_rv cleanup
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:21 +01:00
Andreas Gruenbacher
bf885f8a67 drbd: Be more explicit about functions that return an enum drbd_state_rv
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:19 +01:00
Andreas Gruenbacher
c8b325632f drbd: Rename enum drbd_state_ret_codes to enum drbd_state_rv
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:18 +01:00
Andreas Gruenbacher
116676ca62 drbd: Rename enum drbd_ret_codes to enum drbd_ret_code
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:16 +01:00
Andreas Gruenbacher
0cf9d27e38 drbd: Get rid of unnecessary macros (2)
The FAULT_ACTIVE macro just wraps the drbd_insert_fault macro for no
apparent reason.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:15 +01:00
Andreas Gruenbacher
662d91a23a drbd: Get rid of unnecessary macros (1)
This macro doesn't save much code, but makes things a lot harder to read.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:13 +01:00
Andreas Gruenbacher
2f58dcfc85 drbd: Rename drbd_make_request_26 to drbd_make_request
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:11 +01:00
Andreas Gruenbacher
96756784a6 drbd: Remove left-over prototype
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:10 +01:00
Andreas Gruenbacher
cab2f74b45 drbd: Make sure that drbd_send() has sent the right number of bytes
Reviewed-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
2011-03-10 11:36:08 +01:00
Lars Ellenberg
220df4d006 drbd: fix incomplete error message
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:36:02 +01:00
Andreas Gruenbacher
7e458c32da drbd: Removed an unnecessary #undef
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:22 +01:00
Lars Ellenberg
8a3c104438 drbd: fix regression, we need to close drbd epochs during normal operation
commit e2041475e6ddb081734d161f6421977323f5a9b9
drbd: Starting with protocol 96 we can allow app-IO while receiving the bitmap

Contained a bad chunk that tried to optimize away drbd barriers during
bitmap exchange, but accidentally dropped them for normal mode as well.

Impact: depending on activity log size and access pattern, activity log
extents may not be recycled in time, causeing IO to block indefinetely.

Fix: skip drbd barriers only if there is no connection to send them on,
or the request being completed has not been on the network at all.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:20 +01:00
Philipp Reisner
09b9e79793 drbd: Implemented the before-resync-source handler
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:18 +01:00
Philipp Reisner
2561b9c1f1 drbd: --force option for disconnect
As the network connection can be lost at any time, a --force option
for disconnect is just a matter of completeness.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:17 +01:00
Lars Ellenberg
42ff269d10 drbd: add packet_type 27 (return_code_only) to netlink api
In case we ever should add an other packet type,
we must not reuse 27, as that currently used for
"empty" return code only replies.
Document it as such.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:15 +01:00
Lars Ellenberg
3e3a7766c2 drbd: use kzalloc and memset(,0,) to start with clean buffers in drbd_nl
Make sure we start with clean buffers to not accidentally send garbage
back to userspace. Note: has not been observed; but just in case.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:14 +01:00
Lars Ellenberg
17a93f3007 drbd: remove /proc/drbd before unregistering from netlink
There still exists a (theoretical) race on module unload, where
/proc/drbd may still exist, but the netlink callback has been
unregistered already, allowing drbdsetup to shout without listeners,
and get no reply.

Reorder remove_proc_entry and unregister of netlink callback.
drbdsetup first checks for existence of the proc entry,
and if that is missing, won't even try to contact the module.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:12 +01:00
Lars Ellenberg
3da127fa88 drbd: increase module count on /proc/drbd access
If someone holds /proc/drbd open, previously rmmod would
"succeed" in starting the unload, but then block on remove_proc_entry,
leading to a situation where the lsmod does not show drbd anymore,
but /proc/drbd being still there (but no longer accessible).

I'd rather have rmmod fail up front in this case.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:11 +01:00
Philipp Reisner
c507f46f26 drbd: Removed 20 seconds upper bound for side-stepping
Given low-enough network bandwidth combined with a IO
pattern that hammers onto a single RS-extent, side-stepping
might be necessary for much longer times.

Changed the code to print a single informal message after
20 seconds, but it keeps on stepping aside forever.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:09 +01:00
Philipp Reisner
1fc80cf378 drbd: Becoming sync target may not happen out of < C_WF_REPORT_PARAMS
This patch is acutally a necessary addendum to the patch
"fix for spurious full sync (becoming sync target looked like invalidate)"

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:07 +01:00
Philipp Reisner
3719094ec2 drbd: Starting with protocol 96 we can allow app-IO while receiving the bitmap
* C_STARTING_SYNC_S, C_STARTING_SYNC_T In these states the bitmap gets
  written to disk. Locking out of app-IO is done by using the
  drbd_queue_bitmap_io() and drbd_bitmap_io() functions these days.
  It is no longer necessary to lock out app-IO based on the connection
  state.
  App-IO that may come in after the BITMAP_IO flag got cleared before the
  state transition to C_SYNC_(SOURCE|TARGET) does not get mirrored, sets
  a bit in the local bitmap, that is already set, therefore changes nothing.

* C_WF_BITMAP_S In this state we send updates (P_OUT_OF_SYNC packets).
  With that we make sure they have the same number of bits when going
  into the C_SYNC_(SOURCE|TARGET) connection state.

* C_UNCONNECTED: The receiver starts, no need to lock out IO.

* C_DISCONNECTING: in drbd_disconnect() we had a wait_event()
  to wait until ap_bio_cnt reaches 0. Removed that.

* C_TIMEOUT, C_BROKEN_PIPE, C_NETWORK_FAILURE
  C_PROTOCOL_ERROR, C_TEAR_DOWN: Same as C_DISCONNECTING

* C_WF_REPORT_PARAMS: IO still possible since that is still
  like C_WF_CONNECTION.

And we do not need to send barriers in C_WF_BITMAP_S connection state.

Allow concurrent accesses to the bitmap when receiving the bitmap.
Everything gets ORed anyways.

A drbd_free_tl_hash() is in after_state_chg_work(). At that point
all the work items of the last connections must have been processed.

Introduced a call to drbd_free_tl_hash() into drbd_free_mdev()
for paranoia reasons.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:06 +01:00
Philipp Reisner
ab17b68f45 drbd: Improvements in sanitize_state()
The relevant change is that the state change to C_FW_BITMAP_S should
implicitly change pdsk to C_CONSISTENT. (Think of it as C_OUTDATED, only
without the guarantee that the peer has the outdated written to its
meta data)

At that opportunity I restructured the switch statement so that it
gets evaluated every time. (Has declarative character)

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:04 +01:00
Philipp Reisner
22afd7ee94 drbd: Fixed race condition in drbd_queue_bitmap_io
May only test for ap_bio_cnt == 0 under req_lock. It can increase
only under req_lock.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:03 +01:00
Philipp Reisner
8869d683b7 drbd: Fixed inc_ap_bio()
The condition must be checked after perpare_to_wait(). The old
implementaion could loose wakeup events. Never observed in real
life.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:35:01 +01:00
Philipp Reisner
127b317844 drbd: use test_and_set_bit() to decide if bm_io_work should be queued
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:59 +01:00
Philipp Reisner
aeda1cd6a5 drbd: Begin to account BIO processing time before inc_ap_bio()
Since inc_ap_bio() might sleep already

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:57 +01:00
Philipp Reisner
f91ab6282d drbd: Implemented side-stepping in drbd_res_begin_io()
Before:
  drbd_rs_begin_io() locked app-IO out of an RS extent, and
  waited then until all previous app-IO in that area finished.
  (But not only until the disk-IO was finished but until the
   barrier/epoch ack came in for that == round trip time latency ++)

After:
  As soon as a new app-IO waits wants to start new IO on that
  RS extent, drbd_rs_begin_io() steps aside (clearing the
  BME_NO_WRITES flag again). It retries after 100ms.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:56 +01:00
Philipp Reisner
9d77a5fee9 drbd: Make some functions static
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:54 +01:00
Philipp Reisner
e3555d8545 drbd: Implemented priority inheritance for resync requests
We only issue resync requests if there is no significant application IO
going on. = Application IO has higher priority than resnyc IO.

If application IO can not be started because the resync process locked
an resync_lru entry, start the IO operations necessary to release the
lock ASAP.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:53 +01:00
Philipp Reisner
59817f4fab drbd: Do not cleanup resync LRU for the Ahead/Behind SyncSource/SyncTarget transitions
This one should be replaced with moving this cleanup to the
'right' position.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:51 +01:00
Philipp Reisner
c4752ef128 drbd: When proxy's buffer drained off go into regular resync mode
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:49 +01:00
Philipp Reisner
73a01a18b9 drbd: New packet for Ahead/Behind mode: P_OUT_OF_SYNC
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:48 +01:00
Philipp Reisner
67531718d8 drbd: Implemented two new connection states Ahead/Behind
In this connection mode, the ahead node no longer replicates
application IO. The behind's disk becomes out dated.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:46 +01:00
Philipp Reisner
422028b1ca drbd: New configuration parameters for dealing with network congestion
net {
    on_congestion {block|pull-ahead|disconnect};
    congestion-fill {sectors};
    congestion-extents {al-extents};
}

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:45 +01:00
Philipp Reisner
759fbdfba6 drbd: Track the numbers of sectors in flight
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:43 +01:00
Lars Ellenberg
688593c5a8 drbd: Renamed write_flags_to_bio() to wire_flags_to_bio()
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:34:32 +01:00
Lars Ellenberg
4896e8c1b8 drbd: restore compatibility with 32bit kernels
With commit
drbd: further converge progress display of resync and online-verify
accidentally an u64/u64 div was introduced, causing an unresolvable
symbol __udivdi3 to be reference. Actually for that division, 32bit are
still suficient for now, so we can revert to unsigned long instead.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:13 +01:00
Lars Ellenberg
1816a2b47a drbd: properly use max_hw_sectors to limit the our bio size
To ease tracking of bios in some hash tables, we want it to
not cross certain boundaries (128k, used to be 32k).
We limit the maximum bio size using queue parameters.

Historically some defines and variables we use there have been named
max_segment_size, which was misguided. Rename them to max_bio_size,
and use [blk_]queue_max_hw_sectors where appropriate.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:11 +01:00
Lars Ellenberg
3129b1b9ae drbd: debug: limit nelink-broadcast of request on digest mismatch to 32k
We used to be limited to 32k requests,
but have increased that limit to 128k now.

This part of the code can only deal with 32k,
it would scramble arbitrary pages for larger requests.

As it is used for debugging only anyways,
it is ok to simply truncate the dumped data here.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:09 +01:00
Lars Ellenberg
470be44ab1 drbd: detect modification of in-flight buffers
With data-integrity digest enabled, double-check on the sending side
for modifications by upper layers of buffers under write back,
so we can tell it appart from corruption on the "wire".

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:08 +01:00
Lars Ellenberg
5f9915bbb8 drbd: further converge progress display of resync and online-verify
Show progressbar and ETA always, with proc_details >= 1 also show the
current sector position for both resync and online-verify on both nodes.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:06 +01:00
Lars Ellenberg
18edc0b9d7 drbd: fix potential wrap of 32bit oos:%lu display in /proc/drbd
When converting bits (4k resolution, still) to kB, we shift left.  If it
was a large number of bits on a 32bit box (>= 4 TiB storage), we may
wrap the 32bit unsigned long base type, resulting in incorrect display.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:04 +01:00
Lars Ellenberg
2649f0809f drbd: use the resync controller for online-verify requests as well
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:03 +01:00
Lars Ellenberg
e65f440d47 drbd: factor out drbd_rs_number_requests
Preparation patch to be able to use the auto-throttling resync controller
for online-verify requests as well.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:19:01 +01:00
Lars Ellenberg
9bd28d3c90 drbd: factor out drbd_rs_controller_reset
Preparation patch to be able to use the auto-throttling resync controller
for online-verify requests as well.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:18:59 +01:00
Lars Ellenberg
439d595379 drbd: show progress bar and ETA for online-verify
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:18:58 +01:00
Lars Ellenberg
ea5442aff6 drbd: advance progress step marks for online-verify
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:18:56 +01:00
Lars Ellenberg
c6ea14dfa3 drbd: factor out advancement of resync marks for progress reporting
This is in preparation to unify progress reporting of
online-verify and resync requests.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:18:54 +01:00
Lars Ellenberg
de228bba67 drbd: initialize online-verify progress tracking on verify target
For partial (resumed) online verify, initialize the resync step marks
once we know what the online verify start sector is.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:18:53 +01:00
Lars Ellenberg
30b743a2d5 drbd: improve online-verify progress tracking
For a partial (resumed) online-verify, initialize rs_total not to total
bits, but to number of bits to check in this run, to match the meaning
rs_total has for actual resync.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:18:51 +01:00
Lars Ellenberg
2652561886 drbd: only reset online-verify start sector if verify completed
For network hickups during online-verify, on the next verify
triggered, we by default want to resume where it left off.

After any replication link interruption, there will be a (possibly
empty) resync.  Do not reset online-verify start sector if some resync
completed, that would defeats the purpose.

Only reset the start sector once a verify run is completed.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
2011-03-10 11:18:49 +01:00
Jens Axboe
4c63f5646e Merge branch 'for-2.6.39/stack-plug' into for-2.6.39/core
Conflicts:
	block/blk-core.c
	block/blk-flush.c
	drivers/md/raid1.c
	drivers/md/raid10.c
	drivers/md/raid5.c
	fs/nilfs2/btnode.c
	fs/nilfs2/mdt.c

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2011-03-10 08:58:35 +01:00
Jens Axboe
721a9602e6 block: kill off REQ_UNPLUG
With the plugging now being explicitly controlled by the
submitter, callers need not pass down unplugging hints
to the block layer. If they want to unplug, it's because they
manually plugged on their own - in which case, they should just
unplug at will.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2011-03-10 08:52:27 +01:00
Jens Axboe
7eaceaccab block: remove per-queue plugging
Code has been converted over to the new explicit on-stack plugging,
and delay users have been converted to use the new API for that.
So lets kill off the old plugging along with aops->sync_page().

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2011-03-10 08:52:07 +01:00
Tejun Heo
3c0d206092 pktcdvd: Convert to bdops->check_events()
Convert from ->media_changed() to ->check_events().

pktcdvd needs to forward all event related operations to the
underlying device.  Forward ->check_events() instead of
->media_changed() and inherit disk->[async_]events.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Peter Osterlund <petero2@telia.com>
2011-03-09 19:54:28 +01:00
Tejun Heo
6fac80e3aa umem: Drop dummy ->media_changed()
umem doesn't implement media changed detection and there's no need to
implement dummy callback anymore.  Remove it.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
2011-03-09 19:54:28 +01:00
Tejun Heo
3a200911ad xsysace: Convert to bdops->check_events()
Convert from ->media_changed() to ->check_events().

xsysace buffers media changed state and clears it on revalidation.  It
will behave correctly with kernel event polling.

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
2011-03-09 19:54:28 +01:00
Tejun Heo
aaa7c01546 ub: Convert to bdops->check_events()
Convert from ->media_changed() to ->check_events().

ub buffers media changed state and clears it on revalidation.  It will
behave correctly with kernel event polling.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Pete Zaitcev <zaitcev@redhat.com>
2011-03-09 19:54:28 +01:00
Tejun Heo
4bbde77787 swim[3]: Convert to bdops->check_events()
Convert from ->media_changed() to ->check_events().

Both swim and swim3 buffer media changed state and clear it on
revalidation.  They will behave correctly with kernel event polling.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Laurent Vivier <laurent@lvivier.info>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2011-03-09 19:54:28 +01:00
Tejun Heo
507daea227 dac960: Convert to bdops->check_events()
Convert from ->media_changed() to ->check_events().

DAC960 media change notification seems to be one way (once set, never
cleared) and will generate spurious events when polled once the
condition triggers.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
2011-03-09 19:54:28 +01:00