hot-replace is a feature being added to md which will allow a
device to be replaced without removing it from the array first.
With hot-replace a spare can be activated and recovery can start while
the original device is still in place, thus allowing a transition from
an unreliable device to a reliable device without leaving the array
degraded during the transition. It can also be use when the original
device is still reliable but it not wanted for some reason.
This will eventually be supported in RAID4/5/6 and RAID10.
This patch adds a super-block flag to distinguish the replacement
device. If an old kernel sees this flag it will reject the device.
It also adds two per-device flags which are viewable and settable via
sysfs.
"want_replacement" can be set to request that a device be replaced.
"replacement" is set to show that this device is replacing another
device.
The "rd%d" links in /sys/block/mdXx/md only apply to the original
device, not the replacement. We currently don't make links for the
replacement - there doesn't seem to be a need.
Signed-off-by: NeilBrown <neilb@suse.de>
While using etags to find free_pages(), I stumbled across this debug
definition of free_pages() that is to be used while debugging some raid
code in userspace. The __get_free_pages() allocates the correct size,
but the free_pages() does not match. free_pages(), like
__get_free_pages(), takes an order and not a size.
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: NeilBrown <neilb@suse.de>
Space must have been allocated when array was created.
A feature flag is set when the badblock list is non-empty, to
ensure old kernels don't load and trust the whole device.
We only update the on-disk badblocklist when it has changed.
If the badblocklist (or other metadata) is stored on a bad block, we
don't cope very well.
If metadata has no room for bad block, flag bad-blocks as disabled,
and do the same for 0.90 metadata.
Signed-off-by: NeilBrown <neilb@suse.de>
1/ Raid5 has learned to take over also raid4 and raid6 arrays.
2/ new_chunk in mdp_superblock_1 is in sectors, not bytes.
Signed-off-by: NeilBrown <neilb@suse.de>
Move the raid6 data processing routines into a standalone module
(raid6_pq) to prepare them to be called from async_tx wrappers and other
non-md drivers/modules. This precludes a circular dependency of raid456
needing the async modules for data processing while those modules in
turn depend on raid456 for the base level synchronous raid6 routines.
To support this move:
1/ The exportable definitions in raid6.h move to include/linux/raid/pq.h
2/ The raid6_call, recovery calls, and table symbols are exported
3/ Extra #ifdef __KERNEL__ statements to enable the userspace raid6test to
compile
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>
This makes the includes more explicit, and is preparation for moving
md_k.h to drivers/md/md.h
Remove include/raid/md.h as its only remaining use was to #include
other files.
Signed-off-by: NeilBrown <neilb@suse.de>
The extern function definitions are kernel-internal definitions, so
they belong in md_k.h
The MD_*_VERSION values could reasonably go in a number of places,
but md_u.h seems most reasonable.
This leaves almost nothing in md.h. It will go soon.
Signed-off-by: NeilBrown <neilb@suse.de>
.. as they are part of the user-space interface.
Also move MdpMinorShift into there so we can remove duplication.
Lastly move mdp_major in. It is less obviously part of the user-space
interface, but do_mounts_md.c uses it, and it is acting a bit like
user-space.
Signed-off-by: NeilBrown <neilb@suse.de>
Move the headers with the local structures for the disciplines and
bitmap.h into drivers/md/ so that they are more easily grepable for
hacking and not far away. md.h is left where it is for now as there
are some uses from the outside.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: NeilBrown <neilb@suse.de>
There are two problems with is_mddev_idle.
1/ sync_io is 'atomic_t' and hence 'int'. curr_events and all the
rest are 'long'.
So if sync_io were to wrap on a 64bit host, the value of
curr_events would go very negative suddenly, and take a very
long time to return to positive.
So do all calculations as 'int'. That gives us plenty of precision
for what we need.
2/ To initialise rdev->last_events we simply call is_mddev_idle, on
the assumption that it will make sure that last_events is in a
suitable range. It used to do this, but now it does not.
So now we need to be more explicit about initialisation.
Signed-off-by: NeilBrown <neilb@suse.de>
fix the following 'make headers_check' warning:
usr/include/linux/raid/md_p.h:85: found __[us]{8,16,32,64} type without #include <linux/types.h>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
If a raid1 has only one working drive and it has a sector which
gives an error on read, then an attempt to recover onto a spare will
fail, but as the single remaining drive is not removed from the
array, the recovery will be immediately re-attempted, resulting
in an infinite recovery loop.
So detect this situation and don't retry recovery once an error
on the lone remaining drive is detected.
Allow recovery to be retried once every time a spare is added
in case the problem wasn't actually a media error.
Signed-off-by: NeilBrown <neilb@suse.de>
Using sequential numbers to identify md devices is somewhat artificial.
Using names can be a lot more user-friendly.
Also, creating md devices by opening the device special file is a bit
awkward.
So this patch provides a new option for creating and naming devices.
Writing a name such as "md_home" to
/sys/modules/md_mod/parameters/new_array
will cause an array with that name to be created. It will appear in
/sys/block/ /proc/partitions and /proc/mdstat as 'md_home'.
It will have an arbitrary minor number allocated.
md devices that a created by an open are destroyed on the last
close when the device is inactive.
For named md devices, they will not be destroyed until the array
is explicitly stopped, either with the STOP_ARRAY ioctl or by
writing 'clear' to /sys/block/md_XXXX/md/array_state.
The name of the array must start 'md_' to avoid conflict with
other devices.
Signed-off-by: NeilBrown <neilb@suse.de>
Currently md devices, once created, never disappear until the module
is unloaded. This is essentially because the gendisk holds a
reference to the mddev, and the mddev holds a reference to the
gendisk, this a circular reference.
If we drop the reference from mddev to gendisk, then we need to ensure
that the mddev is destroyed when the gendisk is destroyed. However it
is not possible to hook into the gendisk destruction process to enable
this.
So we drop the reference from the gendisk to the mddev and destroy the
gendisk when the mddev gets destroyed. However this has a
complication.
Between the call
__blkdev_get->get_gendisk->kobj_lookup->md_probe
and the call
__blkdev_get->md_open
there is no obvious way to hold a reference on the mddev any more, so
unless something is done, it will disappear and gendisk will be
destroyed prematurely.
Also, once we decide to destroy the mddev, there will be an unlockable
moment before the gendisk is unlinked (blk_unregister_region) during
which a new reference to the gendisk can be created. We need to
ensure that this reference can not be used. i.e. the ->open must
fail.
So:
1/ in md_probe we set a flag in the mddev (hold_active) which
indicates that the array should be treated as active, even
though there are no references, and no appearance of activity.
This is cleared by md_release when the device is closed if it
is no longer needed.
This ensures that the gendisk will survive between md_probe and
md_open.
2/ In md_open we check if the mddev we expect to open matches
the gendisk that we did open.
If there is a mismatch we return -ERESTARTSYS and modify
__blkdev_get to retry from the top in that case.
In the -ERESTARTSYS sys case we make sure to wait until
the old gendisk (that we succeeded in opening) is really gone so
we loop at most once.
Some udev configurations will always open an md device when it first
appears. If we allow an md device that was just created by an open
to disappear on an immediate close, then this can race with such udev
configurations and result in an infinite loop the device being opened
and closed, then re-open due to the 'ADD' even from the first open,
and then close and so on.
So we make sure an md device, once created by an open, remains active
at least until some md 'ioctl' has been made on it. This means that
all normal usage of md devices will allow them to disappear promptly
when not needed, but the worst that an incorrect usage will do it
cause an inactive md device to be left in existence (it can easily be
removed).
As an array can be stopped by writing to a sysfs attribute
echo clear > /sys/block/mdXXX/md/array_state
we need to use scheduled work for deleting the gendisk and other
kobjects. This allows us to wait for any pending gendisk deletion to
complete by simply calling flush_scheduled_work().
Signed-off-by: NeilBrown <neilb@suse.de>
The rdev_for_each macro defined in <linux/raid/md_k.h> is identical to
list_for_each_entry_safe, from <linux/list.h>, it should be defined to
use list_for_each_entry_safe, instead of reinventing the wheel.
But some calls to each_entry_safe don't really need a safe version,
just a direct list_for_each_entry is enough, this could save a temp
variable (tmp) in every function that used rdev_for_each.
In this patch, most rdev_for_each loops are replaced by list_for_each_entry,
totally save many tmp vars; and only in the other situations that will call
list_del to delete an entry, the safe version is used.
Signed-off-by: Cheng Renquan <crquan@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
This patch renames the hash_spacing and preshift members of struct
raid0_private_data to spacing and sector_shift respectively and
changes the semantics as follows:
We always have spacing = 2 * hash_spacing. In case
sizeof(sector_t) > sizeof(u32) we also have sector_shift = preshift + 1
while sector_shift = preshift = 0 otherwise.
Note that the values of nb_zone and zone are unaffected by these changes
because in the sector_div() preceeding the assignement of these two
variables both arguments double.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
This completes the block -> sector conversion of struct strip_zone.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
For the same reason as in the previous patch, rename it from zone_offset
to zone_start.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
Rename zone->dev_offset to zone->dev_start to make sure all users
have been converted.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
There is no compelling need for this, but sysfs_notify_dirent is a
nicer interface and the change is good for consistency.
Signed-off-by: NeilBrown <neilb@suse.de>
The 'state' file for a device reports, for example, when the device
has failed. Changes should be reported to userspace ASAP without
the possibility of blocking on low-memory. sysfs_notify does
have that possibility (as it takes a mutex which can be held
across a kmalloc) so use sysfs_notify_dirent instead.
Signed-off-by: NeilBrown <neilb@suse.de>
Now that we have sysfs_notify_dirent, use it to notify changes
to md/array_state.
As sysfs_notify_dirent can be called in atomic context, we can
remove the delayed notify and the MD_NOTIFY_ARRAY_STATE flag.
Signed-off-by: NeilBrown <neilb@suse.de>
Having
function (args)
instead of
function(args)
make is harder to search for calls of particular functions.
So remove all those spaces.
Signed-off-by: NeilBrown <neilb@suse.de>
This patch renames hash_spacing and preshift to spacing and
sector_shift respectively with the following change of semantics:
Case 1: (sizeof(sector_t) <= sizeof(u32)).
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In this case, we have sector_shift = preshift = 0 and spacing =
2 * hash_spacing.
Hence, the index for the hash table which is computed by the new code
in which_dev() as sector / spacing equals the old value which was
(sector/2) / hash_spacing.
Note also that the value of nb_zone stays the same because both sz
and base double.
Case 2: (sizeof(sector_t) > sizeof(u32)).
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
(aka the shifting dance case). Here we have sector_shift = preshift +
1 and
spacing = 2 * hash_spacing
during the computation of nb_zone and curr_sector, but
spacing = hash_spacing
in which_dev() because in the last hunk of the patch for linear.c we
shift down conf->spacing (= 2 * hash_spacing) by one more bit than
in the old code.
Hence in the computation of nb_zone, sz and base have the same value
as before, so nb_zone is not affected. Also curr_sector in the next
hunk stays the same.
In which_dev() the hash table index is computed as
(sector >> sector_shift) / spacing
In view of sector_shift = preshift + 1 and spacing = hash_spacing,
this equals
((sector/2) >> preshift) / hash_spacing
which is the value computed by the old code.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
Rename them to num_sectors and start_sector which is more descriptive.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
All modifications and most access to the mddev->disks list are made
under the reconfig_mutex lock. However there are three places where
the list is walked without any locking. If a reconfig happens at this
time, havoc (and oops) can ensue.
So use RCU to protect these accesses:
- wrap them in rcu_read_{,un}lock()
- use list_for_each_entry_rcu
- add to the list with list_add_rcu
- delete from the list with list_del_rcu
- delay the 'free' with call_rcu rather than schedule_work
Note that export_rdev did a list_del_init on this list. In almost all
cases the entry was not in the list anymore so it was a no-op and so
safe. It is no longer safe as after list_del_rcu we may not touch
the list_head.
An audit shows that export_rdev is called:
- after unbind_rdev_from_array, in which case the delete has
already been done,
- after bind_rdev_to_array fails, in which case the delete isn't needed.
- before the device has been put on a list at all (e.g. in
add_new_disk where reading the superblock fails).
- and in autorun devices after a failure when the device is on a
different list.
So remove the list_del_init call from export_rdev, and add it back
immediately before the called to export_rdev for that last case.
Note also that ->same_set is sometimes used for lists other than
mddev->list (e.g. candidates). In these cases rcu is not needed.
Signed-off-by: NeilBrown <neilb@suse.de>
Open isn't the only thing that increments ->active. e.g. reading
/proc/mdstat will increment it briefly. So to avoid false positives
in testing for concurrent access, introduce a new counter that counts
just the number of times the md device it open.
Signed-off-by: NeilBrown <neilb@suse.de>
This patch renames the array_size field of struct mddev_s to array_sectors
and converts all instances to use units of 512 byte sectors instead of 1k
blocks.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
Rename it to sb_start to make sure all users have been converted.
Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: Neil Brown <neilb@suse.de>
md_allow_write() marks the metadata dirty while holding mddev->lock and then
waits for the write to complete. For externally managed metadata this causes a
deadlock as userspace needs to take the lock to communicate that the metadata
update has completed.
Change md_allow_write() in the 'external' case to start the 'mark active'
operation and then return -EAGAIN. The expected side effects while waiting for
userspace to write 'active' to 'array_state' are holding off reshape (code
currently handles -ENOMEM), cause some 'stripe_cache_size' change requests to
fail, cause some GET_BITMAP_FILE ioctl requests to fall back to GFP_NOIO, and
cause updates to 'raid_disks' to fail. Except for 'stripe_cache_size' changes
these failures can be mitigated by coordinating with mdmon.
md_write_start() still prevents writes from occurring until the metadata
handler has had a chance to take action as it unconditionally waits for
MD_CHANGE_CLEAN to be cleared.
[neilb@suse.de: return -EAGAIN, try GFP_NOIO]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
From: Dan Williams <dan.j.williams@intel.com>
Currently ops_run_biodrain and other locations have extra logic to determine
which blocks are processed in the prexor and non-prexor cases. This can be
eliminated if handle_write_operations5 flags the blocks to be processed in all
cases via R5_Wantdrain. The presence of the prexor operation is tracked in
sh->reconstruct_state.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
From: Dan Williams <dan.j.williams@intel.com>
Track the state of reconstruct operations (recalculating the parity block
usually due to incoming writes, or as part of array expansion) Reduces the
scope of the STRIPE_OP_{BIODRAIN,PREXOR,POSTXOR} flags to only tracking whether
a reconstruct operation has been requested via the ops_request field of struct
stripe_head_state.
This is the final step in the removal of ops.{pending,ack,complete,count}, i.e.
the STRIPE_OP_{BIODRAIN,PREXOR,POSTXOR} flags only request an operation and do
not track the state of the operation.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
From: Dan Williams <dan.j.williams@intel.com>
The STRIPE_OP_* flags record the state of stripe operations which are
performed outside the stripe lock. Their use in indicating which
operations need to be run is straightforward; however, interpolating what
the next state of the stripe should be based on a given combination of
these flags is not straightforward, and has led to bugs. An easier to read
implementation with minimal degrees of freedom is needed.
Towards this goal, this patch introduces explicit states to replace what was
previously interpolated from the STRIPE_OP_* flags. For now this only converts
the handle_parity_checks5 path, removing a user of the
ops.{pending,ack,complete,count} fields of struct stripe_operations.
This conversion also found a remaining issue with the current code. There is
a small window for a drive to fail between when we schedule a repair and when
the parity calculation for that repair completes. When this happens we will
writeback to 'failed_num' when we really want to write back to 'pd_idx'.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
From: Dan Williams <dan.j.williams@intel.com>
The R5_Want{Read,Write} flags already gate i/o. So, this flag is
superfluous and we can unconditionally call ops_run_io().
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
From: Dan Williams <dan.j.williams@intel.com>
This micro-optimization allowed the raid code to skip a re-read of the
parity block after checking parity. It took advantage of the fact that
xor-offload-engines have their own internal result buffer and can check
parity without writing to memory. Remove it for the following reasons:
1/ It is a layering violation for MD to need to manage the DMA and
non-DMA paths within async_xor_zero_sum
2/ Bad precedent to toggle the 'ops' flags outside the lock
3/ Hard to realize a performance gain as reads will not need an updated
parity block and writes will dirty it anyways.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
The important state change happens during an interrupt
in md_error. So just set a flag there and call sysfs_notify
later in process context.
Signed-off-by: Neil Brown <neilb@suse.de>
When the 'resync' thread starts or stops, when we explicitly
set sync_action, or when we determine that there is definitely nothing
to do, we notify sync_action.
To stop "sync_action" from occasionally showing the wrong value,
we introduce a new flags - MD_RECOVERY_RECOVER - to say that a
recovery is probably needed or happening, and we make sure
that we set MD_RECOVERY_RUNNING before clearing MD_RECOVERY_NEEDED.
Signed-off-by: Neil Brown <neilb@suse.de>
This makes it possible to just resync a small part of an array.
e.g. if a drive reports that it has questionable sectors,
a 'repair' of just the region covering those sectors will
cause them to be read and, if there is an error, re-written
with correct data.
Signed-off-by: Neil Brown <neilb@suse.de>
When an array is degraded, bits in the write-intent bitmap are not
cleared, so that if the missing device is re-added, it can be synced
by only updated those parts of the device that have changed since
it was removed.
The enable this a 'events_cleared' value is stored. It is the event
counter for the array the last time that any bits were cleared.
Sometimes - if a device disappears from an array while it is 'clean' -
the events_cleared value gets updated incorrectly (there are subtle
ordering issues between updateing events in the main metadata and the
bitmap metadata) resulting in the missing device appearing to require
a full resync when it is re-added.
With this patch, we update events_cleared precisely when we are about
to clear a bit in the bitmap. We record events_cleared when we clear
the bit internally, and copy that to the superblock which is written
out before the bit on storage. This makes it more "obviously correct".
We also need to update events_cleared when the event_count is going
backwards (as happens on a dirty->clean transition of a non-degraded
array).
Thanks to Mike Snitzer for identifying this problem and testing early
"fixes".
Cc: "Mike Snitzer" <snitzer@gmail.com>
Signed-off-by: Neil Brown <neilb@suse.de>
When we get any IO error during a recovery (rebuilding a spare), we abort
the recovery and restart it.
For RAID6 (and multi-drive RAID1) it may not be best to restart at the
beginning: when multiple failures can be tolerated, the recovery may be
able to continue and re-doing all that has already been done doesn't make
sense.
We already have the infrastructure to record where a recovery is up to
and restart from there, but it is not being used properly.
This is because:
- We sometimes abort with MD_RECOVERY_ERR rather than just MD_RECOVERY_INTR,
which causes the recovery not be be checkpointed.
- We remove spares and then re-added them which loses important state
information.
The distinction between MD_RECOVERY_ERR and MD_RECOVERY_INTR really isn't
needed. If there is an error, the relevant drive will be marked as
Faulty, and that is enough to ensure correct handling of the error. So we
first remove MD_RECOVERY_ERR, changing some of the uses of it to
MD_RECOVERY_INTR.
Then we cause the attempt to remove a non-faulty device from an array to
fail (unless recovery is impossible as the array is too degraded). Then
when remove_and_add_spares attempts to remove the devices on which
recovery can continue, it will fail, they will remain in place, and
recovery will continue on them as desired.
Issue: If we are halfway through rebuilding a spare and another drive
fails, and a new spare is immediately available, do we want to:
1/ complete the current rebuild, then go back and rebuild the new spare or
2/ restart the rebuild from the start and rebuild both devices in
parallel.
Both options can be argued for. The code currently takes option 2 as
a/ this requires least code change
b/ this results in a minimally-degraded array in minimal time.
Cc: "Eivind Sarto" <ivan@kasenna.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
majianpeng