i.e. one or more drives can be added and the array will re-stripe
while on-line.

Most of the interesting work was already done for raid5.  This just extends it
to raid6.

mdadm newer than 2.6 is needed for complete safety, however any version of
mdadm which support raid5 reshape will do a good enough job in almost all
cases (an 'echo repair > /sys/block/mdX/md/sync_action' is recommended after a
reshape that was aborted and had to be restarted with an such a version of
mdadm).
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

An error always aborts any resync/recovery/reshape on the understanding that
it will immediately be restarted if that still makes sense.  However a reshape
currently doesn't get restarted.  With this patch it does.

To avoid restarting when it is not possible to do work, we call into the
personality to check that a reshape is ok, and strengthen raid5_check_reshape
to fail if there are too many failed devices.

We also break some code out into a separate function: remove_and_add_spares as
the indent level for that code was getting crazy.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It is possible for raid5 to be sent a bio that is too big for an underlying
device.  So if it is a READ that we pass stright down to a device, it will
fail and confuse RAID5.

So in 'chunk_aligned_read' we check that the bio fits within the parameters
for the target device and if it doesn't fit, fall back on reading through
the stripe cache and making lots of one-page requests.

Note that this is the earliest time we can check against the device because
earlier we don't have a lock on the device, so it could change underneath
us.

Also, the code for handling a retry through the cache when a read fails has
not been tested and was badly broken.  This patch fixes that code.
Signed-off-by: NNeil Brown <neilb@suse.de>
Cc: "Kai" <epimetreus@fastmail.fm>
Cc: <stable@suse.de>
Cc: <org@suse.de>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

raid5_mergeable_bvec tries to ensure that raid5 never sees a read request
that does not fit within just one chunk.  However as we must always accept
a single-page read, that is not always possible.

So when "in_chunk_boundary" fails, it might be unusual, but it is not a
problem and printing a message every time is a bad idea.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If a GFP_KERNEL allocation is attempted in md while the mddev_lock is held,
it is possible for a deadlock to eventuate.

This happens if the array was marked 'clean', and the memalloc triggers a
write-out to the md device.

For the writeout to succeed, the array must be marked 'dirty', and that
requires getting the mddev_lock.

So, before attempting a GFP_KERNEL allocation while holding the lock, make
sure the array is marked 'dirty' (unless it is currently read-only).
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Thanks Jens for alerting me to this.

Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: <raziebe@gmail.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Currently raid5 depends on clearing the BIO_UPTODATE flag to signal an error
to higher levels.  While this should be sufficient, it is safer to explicitly
set the error code as well - less room for confusion.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

There are some vestiges of old code that was used for bypassing the stripe
cache on reads in raid5.c.  This was never updated after the change from
buffer_heads to bios, but was left as a reminder.

That functionality has nowe been implemented in a completely different way, so
the old code can go.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

stripe_to_pdidx finds the index of the parity disk for a given stripe.  It
assumes raid5 in that it uses "disks-1" to determine the number of data disks.

This is incorrect for raid6 but fortunately the two usages cancel each other
out.  The only way that 'data_disks' affects the calculation of pd_idx in
raid5_compute_sector is when it is divided into the sector number.  But as
that sector number is calculated by multiplying in the wrong value of
'data_disks' the division produces the right value.

So it is innocuous but needs to be fixed.

Also change the calculation of raid_disks in compute_blocknr to make it
more obviously correct (it seems at first to always use disks-1 too).
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Call the chunk_aligned_read where appropriate.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

If a bypass-the-cache read fails, we simply try again through the cache.  If
it fails again it will trigger normal recovery precedures.

update 1:

From: NeilBrown <neilb@suse.de>

1/
  chunk_aligned_read and retry_aligned_read assume that
      data_disks == raid_disks - 1
  which is not true for raid6.
  So when an aligned read request bypasses the cache, we can get the wrong data.

2/ The cloned bio is being used-after-free in raid5_align_endio
   (to test BIO_UPTODATE).

3/ We forgot to add rdev->data_offset when submitting
   a bio for aligned-read

4/ clone_bio calls blk_recount_segments and then we change bi_bdev,
   so we need to invalidate the segment counts.

5/ We don't de-reference the rdev when the read completes.
   This means we need to record the rdev to so it is still
   available in the end_io routine.  Fortunately
   bi_next in the original bio is unused at this point so
   we can stuff it in there.

6/ We leak a cloned bio if the target rdev is not usable.

From: NeilBrown <neilb@suse.de>

update 2:

1/ When aligned requests fail (read error) they need to be retried
   via the normal method (stripe cache).  As we cannot be sure that
   we can process a single read in one go (we may not be able to
   allocate all the stripes needed) we store a bio-being-retried
   and a list of bioes-that-still-need-to-be-retried.
   When find a bio that needs to be retried, we should add it to
   the list, not to single-bio...

2/ We were never incrementing 'scnt' when resubmitting failed
   aligned requests.

[akpm@osdl.org: build fix]
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This will encourage read request to be on only one device, so we will often be
able to bypass the cache for read requests.
Signed-off-by: NNeil Brown <neilb@suse.de>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Replace all uses of kmem_cache_t with struct kmem_cache.

The patch was generated using the following script:

	#!/bin/sh
	#
	# Replace one string by another in all the kernel sources.
	#

	set -e

	for file in `find * -name "*.c" -o -name "*.h"|xargs grep -l $1`; do
		quilt add $file
		sed -e "1,\$s/$1/$2/g" $file >/tmp/$$
		mv /tmp/$$ $file
		quilt refresh
	done

The script was run like this

	sh replace kmem_cache_t "struct kmem_cache"
Signed-off-by: NChristoph Lameter <clameter@sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I forgot to has the size-in-blocks to (loff_t) before shifting up to a
size-in-bytes.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This changes two if() BUG(); usages to BUG_ON(); so people
can disable it safely.
Signed-off-by: NEric Sesterhenn <snakebyte@gmx.de>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

This is very different from other raid levels and all requests go through a
'stripe cache', and it has congestion management already.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The error handling routines don't use proper locking, and so two concurrent
errors could trigger a problem.

So:
  - use test-and-set and test-and-clear to synchonise
    the In_sync bits with the ->degraded count
  - use the spinlock to protect updates to the
    degraded count (could use an atomic_t but that
    would be a bigger change in code, and isn't
    really justified)
  - remove un-necessary locking in raid5
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NCoywolf Qi Hunt <qiyong@freeforge.net>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

They are not needed.  conf->failed_disks is the same as mddev->degraded and
conf->working_disks is conf->raid_disks - mddev->degraded.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Instead of magic numbers (0,1,2,3) in sb_dirty, we have
some flags instead:
MD_CHANGE_DEVS
   Some device state has changed requiring superblock update
   on all devices.
MD_CHANGE_CLEAN
   The array has transitions from 'clean' to 'dirty' or back,
   requiring a superblock update on active devices, but possibly
   not on spares
MD_CHANGE_PENDING
   A superblock update is underway.

We wait for an update to complete by waiting for all flags to be clear.  A
flag can be set at any time, even during an update, without risk that the
change will be lost.

Stop exporting md_update_sb - isn't needed.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This is generally useful, but particularly helps see if it is the same sector
that always needs correcting, or different ones.

[akpm@osdl.org: fix printk warnings]
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The comment gives more details, but I didn't quite have the sequencing write,
so there was room for races to leave bits unset in the on-disk bitmap for
short periods of time.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

When a device is unplugged, requests are moved from one or two (depending on
whether a bitmap is in use) queues to the main request queue.

So whenever requests are put on either of those queues, we should make sure
the raid5 array is 'plugged'.  However we don't.  We currently plug the raid5
queue just before putting requests on queues, so there is room for a race.  If
something unplugs the queue at just the wrong time, requests will be left on
the queue and nothing will want to unplug them.  Normally something else will
plug and unplug the queue fairly soon, but there is a risk that nothing will.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

We introduced 'io_sectors' recently so we could count the sectors that causes
io during resync separate from sectors which didn't cause IO - there can be a
difference if a bitmap is being used to accelerate resync.

However when a speed is reported, we find the number of sectors processed
recently by subtracting an oldish io_sectors count from a current
'curr_resync' count.  This is wrong because curr_resync counts all sectors,
not just io sectors.

So, add a field to mddev to store the curren io_sectors separately from
curr_resync, and use that in the calculations.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

When an array is started we start one or two threads (two if there is a
reshape or recovery that needs to be completed).

We currently start these *before* the array is completely set up and in
particular before queue->queuedata is set.  If the thread actually starts
very quickly on another CPU, we can end up dereferencing queue->queuedata
and oops.

This patch also makes sure we don't try to start a recovery if a reshape is
being restarted.
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I have reports of a problem with raid5 which turns out to be because the raid5
device gets stuck in a 'plugged' state.  This shouldn't be able to happen as
3msec after it gets plugged it should get unplugged.  However it happens
none-the-less.  This patch fixes the problem and is a reasonable thing to do,
though it might hurt performance slightly in some cases.

Until I can find the real problem, we should probably have this workaround in
place.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NJörn Engel <joern@wohnheim.fh-wedel.de>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

As data_disks is *less* than raid_disks, the current test here is obviously
wrong.  And as the difference is already available in conf->max_degraded, it
makes much more sense to use that.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

... as raid5 sync_request is WAY too big.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

For a while we have had checkpointing of resync.  The version-1 superblock
allows recovery to be checkpointed as well, and this patch implements that.

Due to early carelessness we need to add a feature flag to signal that the
recovery_offset field is in use, otherwise older kernels would assume that a
partially recovered array is in fact fully recovered.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

There is a lot of commonality between raid5.c and raid6main.c.  This patches
merges both into one module called raid456.  This saves a lot of code, and
paves the way for online raid5->raid6 migrations.

There is still duplication, e.g.  between handle_stripe5 and handle_stripe6.
This will probably be cleaned up later.

Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The largest chunk size the code can support without substantial surgery is
2^30 bytes, so make that the limit instead of an arbitrary 4Meg.  Some day,
the 'chunksize' should change to a sector-shift instead of a byte-count.  Then
no limit would be needed.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

this changes if() BUG(); constructs to BUG_ON() which is
cleaner and can better optimized away
Signed-off-by: NEric Sesterhenn <snakebyte@gmx.de>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

raid5 overloads bi_phys_segments to count the number of blocks that the
request was broken in to so that it knows when the bio is completely handled.

Accessing this must always be done under a spinlock.  In one case we also call
bi_end_io under that spinlock, which probably isn't ideal as bi_end_io could
be expensive (even though it isn't allowed to sleep).

So we reducde the range of the spinlock to just accessing bi_phys_segments.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

wait_event_lock_irq puts a ';' after its usage of the 4th arg, so we don't
need to.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This allows user-space to access data safely.  This is needed for raid5
reshape as user-space needs to take a backup of the first few stripes before
allowing reshape to commence.

It will also be useful in cluster-aware raid1 configurations so that all
cluster members can leave a section of the array untouched while a
resync/recovery happens.

A 'start' and 'end' of the suspended range are written to 2 sysfs attributes.
Note that only one range can be suspended at a time.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

check_reshape checks validity and does things that can be done instantly -
like adding devices to raid1.  start_reshape initiates a restriping process to
convert the whole array.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>