The typedefs are just annoying. 'mdk' probably refers to 'md_k.h'
which used to be an include file that defined this thing.
Signed-off-by: NNeilBrown <neilb@suse.de>

Two related problems:

1/ some error paths call "md_unregister_thread(mddev->thread)"
   without subsequently clearing ->thread.  A subsequent call
   to mddev_unlock will try to wake the thread, and crash.

2/ Most calls to md_wakeup_thread are protected against the thread
   disappeared either by:
      - holding the ->mutex
      - having an active request, so something else must be keeping
        the array active.
   However mddev_unlock calls md_wakeup_thread after dropping the
   mutex and without any certainty of an active request, so the
   ->thread could theoretically disappear.
   So we need a spinlock to provide some protections.

So change md_unregister_thread to take a pointer to the thread
pointer, and ensure that it always does the required locking, and
clears the pointer properly.
Reported-by: N"Moshe Melnikov" <moshe@zadarastorage.com>
Signed-off-by: NNeilBrown <neilb@suse.de>
cc: stable@kernel.org

A single request to RAID1 or RAID10 might result in multiple
requests if there are known bad blocks that need to be avoided.

To detect if we need to submit another write request we test:
 	if (sectors_handled < (bio->bi_size >> 9)) {

However this is after we call **_write_done() so the 'bio' no longer
belongs to us - the writes could have completed and the bio freed.

So move the **_write_done call until after the test against
bio->bi_size.

This addresses https://bugzilla.kernel.org/show_bug.cgi?id=41862Reported-by: NBruno Wolff III <bruno@wolff.to>
Tested-by: NBruno Wolff III <bruno@wolff.to>
Signed-off-by: NNeilBrown <neilb@suse.de>

A write can complete at two different places:
1/ when the last member-device write completes, through
   raid10_end_write_request
2/ in make_request() when we remove the initial bias from ->remaining.

These two should do exactly the same thing and the comment says they
do, but they don't.

So factor the correct code out into a function and call it in both
places.  This makes the code much more similar to RAID1.

The difference is only significant if there is an error, and they
usually take a while, so it is unlikely that there will be an error
already when make_request is completing, so this is unlikely to cause
real problems.
Signed-off-by: NNeilBrown <neilb@suse.de>

If we find more read/write errors we should record a bad block before
failing the device.
Signed-off-by: NNeilBrown <neilb@suse.de>

Currently when we get a read error during recovery, we simply abort
the recovery.

Instead, repeat the read in page-sized blocks.
On successful reads, write to the target.
On read errors, record a bad block on the destination,
and only if that fails do we abort the recovery.

As we now retry reads we need to know where we read from.  This was in
bi_sector but that can be changed during a read attempt.
So store the correct from_addr and to_addr in the r10_bio for later
access.


Signed-off-by: NeilBrown<neilb@suse.de>

If a read error is detected during recovery the code currently
fails the read device.
This isn't really necessary.  recovery_request_write will signal
a write error to end_sync_write and it will record a write
error on the destination device which will record a bad block
there or kick it from the array.

So just remove this call to do md_error.
Signed-off-by: NNeilBrown <neilb@suse.de>

If we get a write error during resync/recovery don't fail the device
but instead record a bad block.  If that fails we can then fail the
device.
Signed-off-by: NNeilBrown <neilb@suse.de>

We already attempt to fix read errors found during normal IO
and a 'repair' process.
It is best to try to repair them at any time they are found,
so move a test so that during sync and check a read error will
be corrected by over-writing with good data.

If both (all) devices have known bad blocks in the sync section we
won't try to fix even though the bad blocks might not overlap.  That
should be considered later.

Also if we hit a read error during recovery we don't try to fix it.
It would only be possible to fix if there were at least three copies
of data, which is not very common with RAID10.  But it should still
be considered later.
Signed-off-by: NNeilBrown <neilb@suse.de>

When we get a write error (in the data area, not in metadata),
update the badblock log rather than failing the whole device.

As the write may well be many blocks, we trying writing each
block individually and only log the ones which fail.
Signed-off-by: NNeilBrown <neilb@suse.de>

If we succeed in writing to a block that was recorded as
being bad, we clear the bad-block record.

This requires some delayed handling as the bad-block-list update has
to happen in process-context.
Signed-off-by: NNeilBrown <neilb@suse.de>

Writing to known bad blocks on drives that have seen a write error
is asking for trouble.  So try to avoid these blocks.
Signed-off-by: NNeilBrown <neilb@suse.de>

When recovering one or more devices, if all the good devices have
bad blocks we should record a bad block on the device being rebuilt.

If this fails, we need to abort the recovery.

To ensure we don't think that we aborted later than we actually did,
we need to move the check for MD_RECOVERY_INTR earlier in md_do_sync,
in particular before mddev->curr_resync is updated.
Signed-off-by: NNeilBrown <neilb@suse.de>

During resync/recovery limit the size of the request to avoid
reading into a bad block that does not start at-or-before the current
read address.

Similarly if there is a bad block at this address, don't allow the
current request to extend beyond the end of that bad block.

Now that we don't ever read from known bad blocks, it is safe to allow
devices with those blocks into the array.
Signed-off-by: NNeilBrown <neilb@suse.de>

When attempting to repair a read error, don't read from
devices with a known bad block.

As we are only reading PAGE_SIZE blocks, we don't try to
narrow down to smaller regions in the hope that only part of this
page is bad - it isn't worth the effort.
Signed-off-by: NNeilBrown <neilb@suse.de>

When redirecting a read error to a different device, we must
again avoid bad blocks and possibly split the request.

Spin_lock typo fixed thanks to Dan Carpenter <error27@gmail.com>
Signed-off-by: NNeilBrown <neilb@suse.de>

This patch just covers the basic read path:
 1/ read_balance needs to check for badblocks, and return not only
    the chosen slot, but also how many good blocks are available
    there.
 2/ read submission must be ready to issue multiple reads to
    different devices as different bad blocks on different devices
    could mean that a single large read cannot be served by any one
    device, but can still be served by the array.
    This requires keeping count of the number of outstanding requests
    per bio.  This count is stored in 'bi_phys_segments'

On read error we currently just fail the request if another target
cannot handle the whole request.  Next patch refines that a bit.
Signed-off-by: NNeilBrown <neilb@suse.de>

raid10d() is too big and is about to get bigger, so split
handle_read_error() out as a separate function.
Signed-off-by: NNeilBrown <neilb@suse.de>

When a loop ends with a large if, it can be neater to change the
if to invert the condition and just 'continue'.
Then the body of the if can be indented to a lower level.
Signed-off-by: NNeilBrown <neilb@suse.de>

It is only safe to choose not to write to a bad block if that bad
block is safely recorded in metadata - i.e. if it has been
'acknowledged'.

If it hasn't we need to wait for the acknowledgement.

We support that using rdev->blocked wait and
md_wait_for_blocked_rdev by introducing a new device flag
'BlockedBadBlock'.

This flag is only advisory.
It is cleared whenever we acknowledge a bad block, so that a waiter
can re-check the particular bad blocks that it is interested it.

It should be set by a caller when they find they need to wait.
This (set after test) is inherently racy, but as
md_wait_for_blocked_rdev already has a timeout, losing the race will
have minimal impact.

When we clear "Blocked" was also clear "BlockedBadBlocks" incase it
was set incorrectly (see above race).

We also modify the way we manage 'Blocked' to fit better with the new
handling of 'BlockedBadBlocks' and to make it consistent between
externally managed and internally managed metadata.   This requires
that each raidXd loop checks if the metadata needs to be written and
triggers a write (md_check_recovery) if needed.  Otherwise a queued
write request might cause raidXd to wait for the metadata to write,
and only that thread can write it.

Before writing metadata, we set FaultRecorded for all devices that
are Faulty, then after writing the metadata we clear Blocked for any
device for which the Fault was certainly Recorded.

The 'faulty' device flag now appears in sysfs if the device is faulty
*or* it has unacknowledged bad blocks.  So user-space which does not
understand bad blocks can continue to function correctly.
User space which does, should not assume a device is faulty until it
sees the 'faulty' flag, and then sees the list of unacknowledged bad
blocks is empty.
Signed-off-by: NNeilBrown <neilb@suse.de>

As no personality understand bad block lists yet, we must
reject any device that is known to contain bad blocks.
As the personalities get taught, these tests can be removed.

This only applies to raid1/raid5/raid10.
For linear/raid0/multipath/faulty the whole concept of bad blocks
doesn't mean anything so there is no point adding the checks.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

Read errors are considered to corrected if write-back and re-read
cycle is finished without further problems. Thus moving the rdev->
corrected_errors counting after the re-reading looks more reasonable
IMHO.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Signed-off-by: NNeilBrown <neilb@suse.de>

Normally we would fail a device with a READ error.  However if doing
so causes the array to fail, it is better to leave the device
in place and just return the read error to the caller.

The current test for decide if the array will fail is overly
simplistic.
We have a function 'enough' which can tell if the array is failed or
not, so use it to guide the decision.
Signed-off-by: NNeilBrown <neilb@suse.de>

When we get a read error during recovery, RAID10 previously
arranged for the recovering device to appear to fail so that
the recovery stops and doesn't restart.  This is misleading and wrong.

Instead, make use of the new recovery_disabled handling and mark
the target device and having recovery disabled.

Add appropriate checks in add_disk and remove_disk so that devices
are removed and not re-added when recovery is disabled.
Signed-off-by: NNeilBrown <neilb@suse.de>

As per printk_ratelimit comment, it should not be used.
Signed-off-by: NChristian Dietrich <christian.dietrich@informatik.uni-erlangen.de>
Signed-off-by: NNeilBrown <neilb@suse.de>

When performing a recovery, only first 2 slots in r10_bio are in use,
for read and write respectively. However all of pages in the write bio
are never used and just replaced to read bio's when the read completes.

Get rid of those unused pages and share read pages properly.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Signed-off-by: NNeilBrown <neilb@suse.de>

When normal-write and sync-read/write bio completes, we should
find out the disk number the bio belongs to. Factor those common
code out to a separate function.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Signed-off-by: NNeilBrown <neilb@suse.de>

Variable 'first' is initialized to zero and updated to @rdev->raid_disk
only if it is greater than 0. Thus condition '>= first' always implies
'>= 0' so the latter is not needed.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Signed-off-by: NNeilBrown <neilb@suse.de>

When a loop ends with an 'if' with a large body, it is neater
to make the if 'continue' on the inverse condition, and then
the body is indented less.

Apply this pattern 3 times, and wrap some other long lines.
Signed-off-by: NNeilBrown <neilb@suse.de>

This variable 'disk' is never used - how odd.
Signed-off-by: NNeilBrown <neilb@suse.de>

Now that we have a 'slot' variable, make better use of it to simplify
some code a little.
Signed-off-by: NNeilBrown <neilb@suse.de>

Currently the rdev on which a read error happened could be removed
before we perform the fix_error handling.  This requires extra tests
for NULL.

So delay the rdev_dec_pending call until after the call to
fix_read_error so that we can be sure that the rdev still exists.

This allows an 'if' clause to be removed so the body gets re-indented
back one level.
Signed-off-by: NNeilBrown <neilb@suse.de>

raid10 read balance has two different loop for looking through
possible devices to chose the best.
Collapse those into one loop and generally make the code more
readable.
Signed-off-by: NNeilBrown <neilb@suse.de>

We just need to make sure that an unplug event wakes up the md
thread, which is exactly what mddev_check_plugged does.

Also remove some plug-related code that is no longer needed.
Signed-off-by: NNeilBrown <neilb@suse.de>

md/raid submits a lot of IO from the various raid threads.
So adding start/finish plug calls to those so that some
plugging happens.
Signed-off-by: NNeilBrown <neilb@suse.de>

Fixes generated by 'codespell' and manually reviewed.
Signed-off-by: NLucas De Marchi <lucas.demarchi@profusion.mobi>

MD and DM create a new bio_set for every metadevice. Each bio_set has an
integrity mempool attached regardless of whether the metadevice is
capable of passing integrity metadata. This is a waste of memory.

Instead we defer the allocation decision to MD and DM since we know at
metadevice creation time whether integrity passthrough is needed or not.

Automatic integrity mempool allocation can then be removed from
bioset_create() and we make an explicit integrity allocation for the
fs_bio_set.
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>
Reported-by: NZdenek Kabelac <zkabelac@redhat.com>
Acked-by: NMike Snitzer <snizer@redhat.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

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: NJens Axboe <jaxboe@fusionio.com>

blk_throtl_exit assumes that ->queue_lock still exists,
so make sure that it does.
To do this, we stop redirecting ->queue_lock to conf->device_lock
and leave it pointing where it is initialised - __queue_lock.

As the blk_plug functions check the ->queue_lock is held, we now
take that spin_lock explicitly around the plug functions.  We don't
need the locking, just the warning removal.

This is needed for any kernel with the blk_throtl code, which is
which is 2.6.37 and later.

Cc: stable@kernel.org
Signed-off-by: NNeilBrown <neilb@suse.de>

Following symptoms were observed:
1. After raid0->raid10 takeover operation we have array with 2
missing disks.
When we add disk for rebuild, recovery process starts as expected
but it does not finish- it stops at about 90%, md126_resync process
hangs in "D" state.
2. Similar behavior is when we have mounted raid0 array and we
execute takeover to raid10. After this when we try to unmount array-
it causes process umount hangs in "D"

In scenarios above processes hang at the same function- wait_barrier
in raid10.c.
Process waits in macro "wait_event_lock_irq" until the
"!conf->barrier" condition will be true.
In scenarios above it never happens.

Reason was that at the end of level_store, after calling pers->run,
we call mddev_resume. This calls pers->quiesce(mddev, 0) with
RAID10, that calls lower_barrier.
However raise_barrier hadn't been called on that 'conf' yet,
so conf->barrier becomes negative, which is bad.

This patch introduces setting conf->barrier=1 after takeover
operation. It prevents to become barrier negative after call
lower_barrier().
Signed-off-by: NKrzysztof Wojcik <krzysztof.wojcik@intel.com>
Signed-off-by: NNeilBrown <neilb@suse.de>