pandora-kernel.git
13 years agodrbd: move bitmap write from resync_finished to after_state_change
Lars Ellenberg [Thu, 16 Dec 2010 13:47:39 +0000 (14:47 +0100)]
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>
13 years agodrbd: Removed a reference to debug macros removed long time ago
Lars Ellenberg [Wed, 15 Dec 2010 23:37:57 +0000 (00:37 +0100)]
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>
13 years agodrbd: get rid of unused debug code
Lars Ellenberg [Wed, 15 Dec 2010 23:32:38 +0000 (00:32 +0100)]
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>
13 years agodrbd: allow petabyte storage on 64bit arch
Lars Ellenberg [Tue, 14 Dec 2010 14:13:04 +0000 (15:13 +0100)]
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>
13 years agodrbd: bitmap keep track of changes vs on-disk bitmap
Lars Ellenberg [Wed, 15 Dec 2010 07:59:11 +0000 (08:59 +0100)]
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>
13 years agodrbd: store in-core bitmap little endian, regardless of architecture
Lars Ellenberg [Wed, 15 Dec 2010 07:59:09 +0000 (08:59 +0100)]
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>
13 years agodrbd: bitmap: don't count unused bits (fix non-terminating resync)
Lars Ellenberg [Wed, 15 Dec 2010 22:21:39 +0000 (23:21 +0100)]
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>
13 years agodrbd: Rename __inc_ap_bio_cond to may_inc_ap_bio
Andreas Gruenbacher [Mon, 13 Dec 2010 17:03:38 +0000 (18:03 +0100)]
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>
13 years agodrbd: Fix: drbd_bitmap_io does not return an enum determine_dev_size
Andreas Gruenbacher [Sun, 12 Dec 2010 16:45:41 +0000 (17:45 +0100)]
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>
13 years agodrbd: receive_bitmap_plain: Get rid of ugly and useless enum
Andreas Gruenbacher [Sat, 11 Dec 2010 20:53:12 +0000 (21:53 +0100)]
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>
13 years agodrbd: send_bitmap_rle_or_plain: Get rid of ugly and useless enum
Andreas Gruenbacher [Sat, 11 Dec 2010 17:51:50 +0000 (18:51 +0100)]
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>
13 years agodrbd: receive_bitmap: Missing free_page() on error path
Andreas Gruenbacher [Fri, 10 Dec 2010 21:18:27 +0000 (22:18 +0100)]
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>
13 years agodrbd: receive_bitmap: Avoid casting enum drbd_state_rv to int
Andreas Gruenbacher [Fri, 10 Dec 2010 20:04:00 +0000 (21:04 +0100)]
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>
13 years agodrbd: receive_bitmap: Fix the wrong return value
Andreas Gruenbacher [Fri, 10 Dec 2010 16:09:10 +0000 (17:09 +0100)]
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>
13 years agodrbd: drbd_nl_disk_conf: Avoid a compiler warning
Andreas Gruenbacher [Fri, 10 Dec 2010 12:44:05 +0000 (13:44 +0100)]
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>
13 years agodrbd: Use the standard bool, true, and false keywords
Andreas Gruenbacher [Thu, 9 Dec 2010 14:03:57 +0000 (15:03 +0100)]
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>
13 years agodrbd: This code is dead now
Andreas Gruenbacher [Thu, 9 Dec 2010 13:23:27 +0000 (14:23 +0100)]
drbd: This code is dead now

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
13 years agodrbd: Another small enum drbd_state_rv cleanup
Andreas Gruenbacher [Thu, 9 Dec 2010 13:02:35 +0000 (14:02 +0100)]
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>
13 years agodrbd: Be more explicit about functions that return an enum drbd_state_rv
Andreas Gruenbacher [Tue, 7 Dec 2010 23:39:32 +0000 (00:39 +0100)]
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>
13 years agodrbd: Rename enum drbd_state_ret_codes to enum drbd_state_rv
Andreas Gruenbacher [Wed, 8 Dec 2010 00:06:16 +0000 (01:06 +0100)]
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>
13 years agodrbd: Rename enum drbd_ret_codes to enum drbd_ret_code
Andreas Gruenbacher [Wed, 8 Dec 2010 12:33:11 +0000 (13:33 +0100)]
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>
13 years agodrbd: Get rid of unnecessary macros (2)
Andreas Gruenbacher [Tue, 7 Dec 2010 09:43:29 +0000 (10:43 +0100)]
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>
13 years agodrbd: Get rid of unnecessary macros (1)
Andreas Gruenbacher [Tue, 7 Dec 2010 02:01:41 +0000 (03:01 +0100)]
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>
13 years agodrbd: Rename drbd_make_request_26 to drbd_make_request
Andreas Gruenbacher [Mon, 13 Dec 2010 16:48:19 +0000 (17:48 +0100)]
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>
13 years agodrbd: Remove left-over prototype
Andreas Gruenbacher [Thu, 9 Dec 2010 15:23:43 +0000 (16:23 +0100)]
drbd: Remove left-over prototype

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
13 years agodrbd: Make sure that drbd_send() has sent the right number of bytes
Andreas Gruenbacher [Thu, 9 Dec 2010 15:08:46 +0000 (16:08 +0100)]
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>
13 years agodrbd: fix incomplete error message
Lars Ellenberg [Thu, 9 Dec 2010 14:21:02 +0000 (15:21 +0100)]
drbd: fix incomplete error message

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
13 years agodrbd: Removed an unnecessary #undef
Andreas Gruenbacher [Wed, 8 Dec 2010 18:02:09 +0000 (19:02 +0100)]
drbd: Removed an unnecessary #undef

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
13 years agodrbd: fix regression, we need to close drbd epochs during normal operation
Lars Ellenberg [Sun, 5 Dec 2010 13:11:14 +0000 (14:11 +0100)]
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>
13 years agodrbd: Implemented the before-resync-source handler
Philipp Reisner [Fri, 3 Dec 2010 15:04:24 +0000 (16:04 +0100)]
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>
13 years agodrbd: --force option for disconnect
Philipp Reisner [Fri, 3 Dec 2010 14:22:48 +0000 (15:22 +0100)]
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>
13 years agodrbd: add packet_type 27 (return_code_only) to netlink api
Lars Ellenberg [Wed, 24 Nov 2010 09:11:14 +0000 (10:11 +0100)]
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>
13 years agodrbd: use kzalloc and memset(,0,) to start with clean buffers in drbd_nl
Lars Ellenberg [Wed, 24 Nov 2010 09:41:45 +0000 (10:41 +0100)]
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>
13 years agodrbd: remove /proc/drbd before unregistering from netlink
Lars Ellenberg [Wed, 24 Nov 2010 09:37:35 +0000 (10:37 +0100)]
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>
13 years agodrbd: increase module count on /proc/drbd access
Lars Ellenberg [Wed, 24 Nov 2010 09:33:02 +0000 (10:33 +0100)]
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>
13 years agodrbd: Removed 20 seconds upper bound for side-stepping
Philipp Reisner [Mon, 22 Nov 2010 14:49:17 +0000 (15:49 +0100)]
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>
13 years agodrbd: Becoming sync target may not happen out of < C_WF_REPORT_PARAMS
Philipp Reisner [Mon, 22 Nov 2010 13:18:47 +0000 (14:18 +0100)]
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>
13 years agodrbd: Starting with protocol 96 we can allow app-IO while receiving the bitmap
Philipp Reisner [Wed, 10 Nov 2010 11:08:37 +0000 (12:08 +0100)]
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>
13 years agodrbd: Improvements in sanitize_state()
Philipp Reisner [Wed, 17 Nov 2010 15:54:36 +0000 (16:54 +0100)]
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>
13 years agodrbd: Fixed race condition in drbd_queue_bitmap_io
Philipp Reisner [Tue, 16 Nov 2010 14:30:44 +0000 (15:30 +0100)]
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>
13 years agodrbd: Fixed inc_ap_bio()
Philipp Reisner [Wed, 17 Nov 2010 17:24:19 +0000 (18:24 +0100)]
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>
13 years agodrbd: use test_and_set_bit() to decide if bm_io_work should be queued
Philipp Reisner [Tue, 16 Nov 2010 09:07:53 +0000 (10:07 +0100)]
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>
13 years agodrbd: Begin to account BIO processing time before inc_ap_bio()
Philipp Reisner [Tue, 9 Nov 2010 16:45:06 +0000 (17:45 +0100)]
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>
13 years agodrbd: Implemented side-stepping in drbd_res_begin_io()
Philipp Reisner [Tue, 9 Nov 2010 12:59:41 +0000 (13:59 +0100)]
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>
13 years agodrbd: Make some functions static
Philipp Reisner [Sun, 7 Nov 2010 17:02:56 +0000 (18:02 +0100)]
drbd: Make some functions static

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
13 years agodrbd: Implemented priority inheritance for resync requests
Philipp Reisner [Sun, 7 Nov 2010 14:56:29 +0000 (15:56 +0100)]
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>
13 years agodrbd: Do not cleanup resync LRU for the Ahead/Behind SyncSource/SyncTarget transitions
Philipp Reisner [Fri, 29 Oct 2010 10:44:20 +0000 (12:44 +0200)]
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>
13 years agodrbd: When proxy's buffer drained off go into regular resync mode
Philipp Reisner [Wed, 27 Oct 2010 15:32:36 +0000 (17:32 +0200)]
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>
13 years agodrbd: New packet for Ahead/Behind mode: P_OUT_OF_SYNC
Philipp Reisner [Wed, 27 Oct 2010 12:33:00 +0000 (14:33 +0200)]
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>
13 years agodrbd: Implemented two new connection states Ahead/Behind
Philipp Reisner [Wed, 27 Oct 2010 10:21:30 +0000 (12:21 +0200)]
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>
13 years agodrbd: New configuration parameters for dealing with network congestion
Philipp Reisner [Wed, 27 Oct 2010 09:12:07 +0000 (11:12 +0200)]
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>
13 years agodrbd: Track the numbers of sectors in flight
Philipp Reisner [Tue, 26 Oct 2010 14:02:27 +0000 (16:02 +0200)]
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>
13 years agodrbd: Renamed write_flags_to_bio() to wire_flags_to_bio()
Lars Ellenberg [Wed, 17 Nov 2010 21:25:03 +0000 (22:25 +0100)]
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>
13 years agodrbd: restore compatibility with 32bit kernels
Lars Ellenberg [Thu, 11 Nov 2010 21:41:04 +0000 (22:41 +0100)]
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>
13 years agodrbd: properly use max_hw_sectors to limit the our bio size
Lars Ellenberg [Thu, 11 Nov 2010 14:19:07 +0000 (15:19 +0100)]
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>
13 years agodrbd: debug: limit nelink-broadcast of request on digest mismatch to 32k
Lars Ellenberg [Thu, 11 Nov 2010 09:47:05 +0000 (10:47 +0100)]
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>
13 years agodrbd: detect modification of in-flight buffers
Lars Ellenberg [Wed, 10 Nov 2010 09:36:52 +0000 (10:36 +0100)]
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>
13 years agodrbd: further converge progress display of resync and online-verify
Lars Ellenberg [Tue, 9 Nov 2010 13:15:24 +0000 (14:15 +0100)]
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>
13 years agodrbd: fix potential wrap of 32bit oos:%lu display in /proc/drbd
Lars Ellenberg [Tue, 9 Nov 2010 13:12:10 +0000 (14:12 +0100)]
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>
13 years agodrbd: use the resync controller for online-verify requests as well
Lars Ellenberg [Fri, 5 Nov 2010 09:05:47 +0000 (10:05 +0100)]
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>
13 years agodrbd: factor out drbd_rs_number_requests
Lars Ellenberg [Fri, 5 Nov 2010 09:04:07 +0000 (10:04 +0100)]
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>
13 years agodrbd: factor out drbd_rs_controller_reset
Lars Ellenberg [Fri, 5 Nov 2010 08:55:18 +0000 (09:55 +0100)]
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>
13 years agodrbd: show progress bar and ETA for online-verify
Lars Ellenberg [Fri, 5 Nov 2010 08:52:46 +0000 (09:52 +0100)]
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>
13 years agodrbd: advance progress step marks for online-verify
Lars Ellenberg [Fri, 5 Nov 2010 08:48:01 +0000 (09:48 +0100)]
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>
13 years agodrbd: factor out advancement of resync marks for progress reporting
Lars Ellenberg [Fri, 5 Nov 2010 08:23:37 +0000 (09:23 +0100)]
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>
13 years agodrbd: initialize online-verify progress tracking on verify target
Lars Ellenberg [Fri, 5 Nov 2010 08:43:15 +0000 (09:43 +0100)]
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>
13 years agodrbd: improve online-verify progress tracking
Lars Ellenberg [Fri, 5 Nov 2010 08:39:06 +0000 (09:39 +0100)]
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>
13 years agodrbd: only reset online-verify start sector if verify completed
Lars Ellenberg [Fri, 5 Nov 2010 08:56:33 +0000 (09:56 +0100)]
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>
13 years agoMerge branch 'for-2.6.39/stack-plug' into for-2.6.39/core
Jens Axboe [Thu, 10 Mar 2011 07:58:35 +0000 (08:58 +0100)]
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>
13 years agoblk-throttle: Use blk_plug in throttle dispatch
Vivek Goyal [Wed, 9 Mar 2011 07:27:37 +0000 (08:27 +0100)]
blk-throttle: Use blk_plug in throttle dispatch

Use plug in throttle dispatch also as we are dispatching a bunch of
bios in throttle context and some of them might merge.

Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoblock: kill off REQ_UNPLUG
Jens Axboe [Wed, 9 Mar 2011 10:56:30 +0000 (11:56 +0100)]
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>
13 years agoaio: remove request submission batching
Jens Axboe [Thu, 3 Mar 2011 01:12:18 +0000 (20:12 -0500)]
aio: remove request submission batching

This should be useless now that we have on-stack plugging. So lets just
kill it.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agofs: make aio plug
Shaohua Li [Thu, 1 Jul 2010 05:55:01 +0000 (07:55 +0200)]
fs: make aio plug

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agofs: make mpage read/write_pages() plug
Jens Axboe [Tue, 22 Jun 2010 10:52:14 +0000 (12:52 +0200)]
fs: make mpage read/write_pages() plug

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoread-ahead: use plugging
Jens Axboe [Mon, 19 Apr 2010 08:04:38 +0000 (10:04 +0200)]
read-ahead: use plugging

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agofs: make generic file read/write functions plug
Jens Axboe [Thu, 24 Jun 2010 13:05:37 +0000 (15:05 +0200)]
fs: make generic file read/write functions plug

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoblock: remove per-queue plugging
Jens Axboe [Thu, 10 Mar 2011 07:52:07 +0000 (08:52 +0100)]
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>
13 years agoblock: initial patch for on-stack per-task plugging
Jens Axboe [Tue, 8 Mar 2011 12:19:51 +0000 (13:19 +0100)]
block: initial patch for on-stack per-task plugging

This patch adds support for creating a queuing context outside
of the queue itself. This enables us to batch up pieces of IO
before grabbing the block device queue lock and submitting them to
the IO scheduler.

The context is created on the stack of the process and assigned in
the task structure, so that we can auto-unplug it if we hit a schedule
event.

The current queue plugging happens implicitly if IO is submitted to
an empty device, yet callers have to remember to unplug that IO when
they are going to wait for it. This is an ugly API and has caused bugs
in the past. Additionally, it requires hacks in the vm (->sync_page()
callback) to handle that logic. By switching to an explicit plugging
scheme we make the API a lot nicer and can get rid of the ->sync_page()
hack in the vm.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoscsi: convert to blk_delay_queue()
Jens Axboe [Fri, 16 Apr 2010 19:13:15 +0000 (21:13 +0200)]
scsi: convert to blk_delay_queue()

It was always abuse to reuse the plugging infrastructure for this,
convert it to the (new) real API for delaying queueing a bit. A
default delay of 3 msec is defined, to match the previous
behaviour.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoide-cd: convert to blk_delay_queue() for a short pause
Jens Axboe [Fri, 16 Apr 2010 19:11:21 +0000 (21:11 +0200)]
ide-cd: convert to blk_delay_queue() for a short pause

It was always abuse to reuse the plugging infrastructure for this,
convert it to the (new) real API for delaying queueing a bit.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Acked-by: David S. Miller <davem@davemloft.net>
13 years agoblock: add API for delaying work/request_fn a little bit
Jens Axboe [Wed, 2 Mar 2011 16:08:00 +0000 (11:08 -0500)]
block: add API for delaying work/request_fn a little bit

Currently we use plugging for that, but as plugging is going away,
we need an alternative mechanism.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agostaging: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:29 +0000 (19:54 +0100)]
staging: Convert to bdops->check_events()

Convert two staging drivers - blkvsc_drv and cyasblkdev_block - from
->media_changed() to ->check_events().  The former always indicated
media changed while the latter always indicated media not changed.
Not sure what the drivers are trying to achieve but keep the original
behavior.

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
13 years agopktcdvd: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
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>
13 years agoumem: Drop dummy ->media_changed()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
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>
13 years agos390/tape_block: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
s390/tape_block: Convert to bdops->check_events()

Convert from ->media_changed() to ->check_events().

s390/tape_block 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: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
13 years agoi2o_block: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
i2o_block: Convert to bdops->check_events()

Convert from ->media_changed() to ->check_events().

i2o_block buffers media changed state and clears it after reporting.
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: Markus Lidel <Markus.Lidel@shadowconnect.com>
13 years agoxsysace: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
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>
13 years agoub: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
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>
13 years agoswim[3]: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
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>
13 years agodac960: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
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>
13 years agoparide: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
paride: Convert to bdops->check_events()

Convert paride drivers from ->media_changed() to ->check_events().

pcd and pd buffer and clear events after reporting; however, pf
unconditionally reports MEDIA_CHANGE and will generate spurious events
when polled.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Tim Waugh <tim@cyberelk.net>
13 years agogdrom,viocd: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:28 +0000 (19:54 +0100)]
gdrom,viocd: Convert to bdops->check_events()

Convert gdrom and viocd from ->media_changed() to ->check_events().

It's unclear how the conditions are cleared and it's possible that it
may generate spurious events when polled.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
13 years agofloppy,{ami|ata}flop: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:27 +0000 (19:54 +0100)]
floppy,{ami|ata}flop: Convert to bdops->check_events()

Convert the floppy drivers from ->media_changed() to ->check_events().
Both floppy and ataflop buffer media changed state bit and clear them
on revalidation and will behave correctly with kernel event polling.

I can't tell how amiflop clears its event and it's possible that it
may generate spurious events when polled.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
13 years agoide: Convert to bdops->check_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:27 +0000 (19:54 +0100)]
ide: Convert to bdops->check_events()

Convert ->media_changed() to the new ->check_events() method.  The
conversion is mostly mechanical.  The only notable change is that
cdrom now doesn't generate any event if @slot_nr isn't CDSL_CURRENT.
It used to return -EINVAL which would be treated as media changed.  As
media changer isn't supported anyway, this doesn't make any
difference.

This makes ide emit the standard disk events and allows kernel event
polling.  Currently, only MEDIA_CHANGE event is implemented.  Adding
support for EJECT_REQUEST shouldn't be difficult; however, given that
ide driver is already deprecated, it probably is best to leave it
alone.

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: linux-ide@vger.kernel.org
13 years agoblock: Don't check events while open is in progress
Tejun Heo [Wed, 9 Mar 2011 18:54:27 +0000 (19:54 +0100)]
block: Don't check events while open is in progress

Not all block drivers clear events immediately after reporting.  Some
do so in ->revalidate_disk() or other steps during ->open().  There is
a slim chance event poll may happen between the clearing event check
from check_disk_change() and the actual clearing of the events which
would result in spurious events.

Block event checks while block device open is in progress.  There is
no need to kick explicit event check afterwards as events are always
checked during open.

-v2: The original patch could have called disk_unblock_events() with
     an already released or %NULL @disk causing oops.  Fixed by making
     sure references are put after disk_unblock_events() is called.
     It also makes the error path of __blkdev_get() a bit simpler.
     This problem was reported by Jens.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
13 years agoblock: Don't check events on close unless it was blocked
Tejun Heo [Wed, 9 Mar 2011 18:54:27 +0000 (19:54 +0100)]
block: Don't check events on close unless it was blocked

The block event mechanism currently always checks events when the
device is being closed regardless of the open mode.  The intention was
to allow detection of EJECT_REQUEST when a device is closed whether
disk event polling is enabled or not.

This is unnecessary as, for devices of interest, events are checked
from either userland or kernel and in the former case ->check_events()
is performed on open of each poll attempt anyway.  Furthermore, this
unconditional event check on close makes the code susceptible to event
loop if the block driver doesn't clear reported events correctly - an
event triggers userland to open and close the device which in turn
causes another event, rinse and repeat.

Check events on close only if it was blocked by excl write open.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
13 years agoblock: Don't implicitly trigger event check on disk_unblock_events()
Tejun Heo [Wed, 9 Mar 2011 18:54:27 +0000 (19:54 +0100)]
block: Don't implicitly trigger event check on disk_unblock_events()

Currently, disk_unblock_events() implicitly kick event check if the
block count reaches zero.  This behavior is not described in the
comment and hinders with future changes.  Make the unblocker
explicitly check events by calling disk_check_events() as necessary.

This patch doesn't cause any behavior difference.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
13 years agoblk-cgroup: Lower minimum weight from 100 to 10.
Justin TerAvest [Tue, 8 Mar 2011 18:45:00 +0000 (19:45 +0100)]
blk-cgroup: Lower minimum weight from 100 to 10.

We've found that we still get good, useful isolation at weights this
low. I'd like to adjust the minimum so that any other changes can take
these values into account.

Signed-off-by: Justin TerAvest <teravest@google.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoblock: biovec_slab vs. CONFIG_BLK_DEV_INTEGRITY
Martin K. Petersen [Tue, 8 Mar 2011 07:28:01 +0000 (08:28 +0100)]
block: biovec_slab vs. CONFIG_BLK_DEV_INTEGRITY

The block integrity subsystem no longer uses the bio_vec slabs so this
code can safely be compiled in.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoblk-throttle: Some cleanups and race fixes in limit update code
Vivek Goyal [Mon, 7 Mar 2011 20:09:32 +0000 (21:09 +0100)]
blk-throttle: Some cleanups and race fixes in limit  update code

When throttle group limits are updated through cgroups, a thread is
woken up to process these updates. While reviewing that code, oleg noted
couple of race conditions existed in the code and he also suggested that
code can be simplified.

This patch fixes the races simplifies the code based on Oleg's suggestions:

- Use xchg().
- Introduced a common function throtl_update_blkio_group_common()
          which is shared now by all iops/bps update functions.

Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Fixed a merge issue, throtl_schedule_delayed_work() takes throtl_data
as the argument now, not the queue.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>