Check return value of kthread_run() in dm_old_init_request_queue().
Reported-by: Minfei Huang <mnghuan@gmail.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
We have assigned sb->block_size before the switch,
so remove the redundant one.
Reviewed-by: Coly Li <colyli@suse.de>
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
Acked-by: Eric Wheeler <bcache@lists.ewheeler.net>
Signed-off-by: Jens Axboe <axboe@fb.com>
There is no return in continue_at(), update the documentation.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
Acked-by: Coly Li <colyli@suse.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Cache_sb is not used in cache_alloc, and we have copied
sb info to cache->sb already, remove it.
Reviewed-by: Coly Li <colyli@suse.de>
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
do_div was replaced with div64_u64 at some point, causing a bug with
block calculation due to incompatible semantics of the two functions.
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Fixes: a739ff3f54 ("dm verity: add support for forward error correction")
Cc: stable@vger.kernel.org # v4.5+
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Merge the two DM_PARAMS_[KV]MALLOC flags into a single flag.
Doing so avoids the crashes seen with previous attempts to consolidate
buffer management to use kvfree() without first flagging that memory had
actually been allocated.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Avoid that sparse complains about assigning a __le64 value to a u64
variable. Remove the (u64) casts since these are superfluous. This
patch does not change the behavior of the source code.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
A newly introduced function has 'const int' as the return type,
but as "make W=1" reports, that has no meaning:
drivers/md/dm-raid.c:510:18: error: type qualifiers ignored on function return type [-Werror=ignored-qualifiers]
This changes the return type to plain 'int'.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 33e53f0685 ("dm raid: introduce extended superblock and new raid types to support takeover/reshaping")
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Superblock updates where bogus causing some takovers/reshapes to fail.
Introduce new runtime flag (RT_FLAG_KEEP_RS_FROZEN) to keep a raid set
frozen when a layout change was requested. Userpace will immediately
reload the table w/o the flags requesting such change once they made it
to the superblocks and any change of recovery/reshape offsets has to be
avoided until after read.
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add bool functions rs_is_recovering and rs_is_reshaping()
to test for ongoing recovery/reshaping respectively in order
to reject respective requests on ongoing ones.
Remove ctr array size check, because ti->len and array
sectors will differ during disk addition/removal reshape.
Use __is_raid10_near() rather than type string compare.
Introduce rs_check_reshape() and rs_start_reshape(),
use the former in the ctr to reject bogus rehsape requests
and the latter in preresume to actually start a reshape.
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add rs_is_reshapable(), rs_data_stripes(), rs_reshape_requested(),
rs_set_dev_and_array_sectors() and rs_adjust_data_offsets()
Remove superfluous check for reshape message
Correct runtime bit definitions to be incremental
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
It is more intuitive to manage each raid level's features in terms of
what is supported rather than what isn't supported.
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Renamed functions and variables with leading single underscore to have a
double underscore. Renamed some functions to have better names. Folded
functions that were split out without reason.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Also update module description to "raid0/1/10/4/5/6 target"
Reported by Alasdair G Kergon <agk@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
No idea what Heinz was doing with the versioning but upstream commit
4c9971ca6a ("dm raid: make sure no feature flags are set in metadata")
bumped to 1.8.0 already.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
There ti_error_* wrappers added very little. No other DM target has
ever gone to such lengths to wrap setting ti->error.
Also fixes some NULL derefences via rs->ti->error.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The target's status interface has to provide the new 'data_offset' value
to allow userspace to retrieve the kernels offset to the data on each
raid device of a raid set. This is the base for out-of-place reshaping
required to not write over any data during reshaping (e.g. change
raid6_zr -> raid6_nc):
- add rs_set_cur() to be able to start up existing array in case of no
takeover; use in ctr on takeover check
- enhance raid_status()
- add supporting functions to get resync/reshape progress and raid
device status chars
- fixup rebuild table line output race, which does miss to emit
'rebuild N' on fully synced/rebuild devices, because it is relying on
the transient 'In_sync' raid device flag
- add new status line output for 'data_offset', which'll later be used
for out-of-place reshaping
- fixup takeover not working for all levels
- fixup raid0 message interface oops caused by missing checks
for the md threads, which don't exist in case of raid0
- remove ALL_FREEZE_FLAGS not needed for takeover
- adjust comments
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add raid level takeover support allowing arbitrary takeovers between
raid levels supported by md personalities (i.e. raid0, raid1/10 and
raid4/5/6):
- add rs_config_{backup|restore} function to allow for temporary
storing ctr requested layout changes and restore them for takeover
conersion decision after the superblocks got loaded and analyzed
- add members to store layout to 'struct raid_set' (not mandatory
for takeover but needed for reshape in later patch)
- add rebuild_disks bitfield to 'struct raid_set' and set bits in ctr
to use in setting up takeover (base to address a 'rebuild' related
raid_status() table line bug and needed as well for reshape in future
patch)
- add runtime flags and respective manipulation functions to be able to
control e.g. wrting of superlocks to the preresume function on
takeover and (later) reshape
- add functions to detect takeover, check it's valid (mandatory here to
avoid failing on md_run()), setup for it and use in the ctr; those
will be likely moved out once reshaping gets added to simplify the
ctr
- start raid set readonly in ctr and switch to readwrite, optionally
updating superblocks, in preresume in order to allow suspend to
quiesce any active table before (which involves superblock updates);
this ensures the proper sequence of writing the current and any new
takeover(/reshape) metadata
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add transferring the new takeover/reshape related superblock
members introduced to the super_sync() function:
- add/move supporting functions
- add failed devices bitfield transfer functions to retrieve the
bitfield from superblock format or update it in the superblock
- add code to transfer all new members
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Support the follwoing arguments in the ctr parameter parser:
- add 'delta_disks', 'data_offset' taking int and sector respectively
- 'raid10_use_near_sets' bool argument to optionally select
near sets with supporting raid10 mappings
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add new members to the dm-raid superblock and new raid types to support
takeover/reshape.
Add all necessary members needed to support takeover and reshape in one
go -- aiming to limit the amount of changes to the superblock layout.
This is a larger patch due to the new superblock members, their related
flags, validation of both and involved API additions/changes:
- add additional members to keep track of:
- state about forward/backward reshaping
- reshape position
- new level, layout, stripe size and delta disks
- data offset to current and new data for out-of-place reshapes
- failed devices bitfield extensions to keep track of max raid devices
- adjust super_validate() to cope with new superblock members
- adjust super_init_validation() to cope with new superblock members
- add definitions for ctr flags supporting delta disks etc.
- add new raid types (raid6_n_6 etc.)
- add new raid10 supporting function API (_is_raid10_*())
- adjust to changed raid10 supporting function API
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Every time a device is removed with ->hot_remove_disk() a synchronize_rcu() call is made
which can delay several milliseconds in some case.
If lots of devices fail at once - as could happen with a large RAID10 where one set
of devices are removed all at once - these delays can add up to be very inconcenient.
As failure is not reversible we can check for that first, setting a
separate flag if it is found, and then all synchronize_rcu() once for
all the flagged devices. Then ->hot_remove_disk() function can skip the
synchronize_rcu() step if the flag is set.
fix build error(Shaohua)
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
It is important that we never increment rdev->nr_pending on a Faulty
device as ->hot_remove_disk() assumes that once the Faulty flag is visible
no code will take a new reference.
Some places take a new reference after only check In_sync. This should
be safe as the two are changed together. However to make the code more
obviously safe, add checks for 'Faulty' as well.
Note: the actual rule is:
Never increment nr_pending if Faulty is set and Blocked is clear,
never clear Faulty, and never set Blocked without holding a reference
through nr_pending.
fix build error (Shaohua)
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Being in the middle of resync is no longer protection against failed
rdevs disappearing. So add rcu protection.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The rdev could be freed while handle_failed_sync is running, so
rcu protection is needed.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Since remove_and_add_spares() was added to hot_remove_disk() it has
been possible for an rdev to be hot-removed while fix_read_error()
was running, so we need to be more careful, and take a reference to
the rdev while performing IO.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
'mirror' is only used to find 'rdev', several times.
So just find 'rdev' once, and use it instead.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Both functions use conf->mirrors[mirror].rdev several times, so
improve readability by storing this in a local variable.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
'tmp' is only ever used to extract 'tmp->rdev', so just use 'rdev' directly.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
rdev already holds conf->mirrors[d].rdev, so no need to load it again.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
mirrors[].rdev can become NULL at any point unless:
- a counted reference is held
- ->reconfig_mutex is held, or
- rcu_read_lock() is held
Reshape isn't always suitably careful as in the past rdev couldn't be
removed during reshape. It can now, so add protection.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
mirrors[].rdev can become NULL at any point unless:
- a counted reference is held
- ->reconfig_mutex is held, or
- rcu_read_lock() is held
Previously they could not become NULL during a resync/recovery/reshape either.
However when remove_and_add_spares() was added to hot_remove_disk(), that
changed.
So raid10_sync_request didn't previously need to protect rdev access,
but now it does.
Fix missed check(Shaohua)
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
mirrors[].rdev can become NULL at any point unless:
- a counted reference is held
- ->reconfig_mutex is held, or
- rcu_read_lock() is held
raid10_status holds none of these. So add rcu_read_lock()
protection.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
If you have a raid10 with a replacement device that is resyncing -
e.g. after a crash before the replacement was complete - the write to
the replacement will increment nr_pending on the wrong device, which
will lead to strangeness.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Re-checking the faulty flag here brings no value.
The comment about "risk" refers to the risk that the device could
be in the process of being removed by ->hot_remove_disk().
However providing that the ->nr_pending count is incremented inside
an rcu_read_locked() region, there is no risk of that happening.
This is because the rdev pointer (in the personalities array) is set
to NULL before synchronize_rcu(), and ->nr_pending is tested
afterwards. If the rcu_read_locked region happens before the
synchronize_rcu(), the test will see that nr_pending has been incremented.
If it happens afterwards, the rdev pointer will be NULL so there is nothing
to increment.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
When the HOT_REMOVE_DISK ioctl is used to remove a device, we
call remove_and_add_spares() which will remove it from the personality
if possible. This improves the chances that the removal will succeed.
When writing "remove" to dev-XX/state, we don't. So that can fail more easily.
So add the remove_and_add_spares() into "remove" handling.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
A performance drop of mkfs has been observed on RAID10 during resync
since commit 09314799e4 ("md: remove 'go_faster' option from
->sync_request()"). Resync sends so many IOs it slows down non-resync
IOs significantly (few times). Add a short delay to a resync. The
previous long sleep (1s) has proven unnecessary, even very short delay
brings performance right.
The change also applied to raid1. The problem has not been observed on
raid1, however it shares barriers code with raid10 so it might be an
issue for some setup too.
Suggested-by: NeilBrown <neilb@suse.com>
Link: http://lkml.kernel.org/r/20160609134555.GA9104@proton.igk.intel.com
Signed-off-by: Tomasz Majchrzak <tomasz.majchrzak@intel.com>
Signed-off-by: Shaohua Li <shli@fb.com>
This is a simple check before updating the superblock. It should update
the superblock when update_size return 0.
Signed-off-by: Xiao Ni <xni@redhat.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Make use if raid type rt_is_*() bool functions for simplification and
consistency reasons.
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
- add _test_flags() function
- use it to simplify rs_check_for_invalid_flags()
- use _test_flag() throughout
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Reject invalid flag combinations to avoid potential data corruption or
failing raid set construction:
- add definitions for constructor flag combinations and invalid flags
per level
- add bool test functions for the various raid types
(also will be used by future reshaping enhancements)
- introduce rs_check_for_invalid_flags() and _invalid_flags()
to perform the validity checks
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Provide necessary infrastructure to handle ctr flags and their names
and cleanup setting ti->error:
- comment constructor flags
- introduce constructor flag manipulation
- introduce ti_error_*() functions to simplify
setting the error message (use in other targets?)
- introduce array to hold ctr flag <-> flag name mapping
- introduce argument name by flag functions for that array
- use those functions throughout the ctr call path
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
- use dm_arg_set API in ctr and its callees parse_raid_params() and dev_parms()
- introduce _in_range() function to check a value is in a [ min, max ] range;
this is to support more callers in parsing parameters etc. in the future
- correct comment on MAX_RAID_DEVICES
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
alloc_workqueue replaces deprecated create_workqueue().
Dedicated workqueues have been used since bcache_wq and moving_gc_wq
are workqueues for writes and are being used on a memory reclaim path.
WQ_MEM_RECLAIM has been set to ensure forward progress under memory
pressure.
Since there are only a fixed number of work items, explicit concurrency
limit is unnecessary here.
Signed-off-by: Bhaktipriya Shridhar <bhaktipriya96@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
Allow a user to specify an optional feature 'queue_mode <mode>' where
<mode> may be "bio", "rq" or "mq" -- which corresponds to bio-based,
request_fn rq-based, and blk-mq rq-based respectively.
If the queue_mode feature isn't specified the default for the
"multipath" target is still "rq" but if dm_mod.use_blk_mq is set to Y
it'll default to mode "mq".
This new queue_mode feature introduces the ability for each multipath
device to have its own queue_mode (whereas before this feature all
multipath devices effectively had to have the same queue_mode).
This commit also goes a long way to eliminate the awkward (ab)use of
DM_TYPE_*, the associated filter_md_type() and other relatively fragile
and difficult to maintain code.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add "multipath-bio" target that offers a bio-based multipath target as
an alternative to the request-based "multipath" target -- but in a
following commit "multipath-bio" will immediately be replaced by a new
"queue_mode" feature for the "multipath" target which will allow
bio-based mode to be selected.
When DM multipath was originally converted from bio-based to
request-based the motivation for the change was better dynamic load
balancing (by leveraging block core's request-based IO schedulers, for
merging and sorting, _before_ DM multipath would make the decision on
where to steer the IO -- based on path load and/or availability).
More background is available in this "Request-based Device-mapper
multipath and Dynamic load balancing" paper:
https://www.kernel.org/doc/ols/2007/ols2007v2-pages-235-244.pdf
But we've now come full circle where significantly faster storage
devices no longer need IOs to be made larger to drive optimal IO
performance. And even if they do there have been changes to the block
and filesystem layers that help ensure upper layers are constructing
larger IOs. In addition, SCSI's differentiated IO errors will propagate
through to bio-based IO completion hooks -- so that eliminates another
historic justiciation for request-based DM multipath. Lastly, the block
layer's immutable biovec changes have made bio cloning cheaper than it
has ever been; whereas request cloning is still relatively expensive
(both on a CPU usage and memory footprint level).
As such, bio-based DM multipath offers the promise of a more efficient
IO path for high IOPs devices that are, or will be, emerging.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add some seperation between bio-based and request-based DM core code.
'struct mapped_device' and other DM core only structures and functions
have been moved to dm-core.h and all relevant DM core .c files have been
updated to include dm-core.h rather than dm.h
DM targets should _never_ include dm-core.h!
[block core merge conflict resolution from Stephen Rothwell]
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Instead of overloading the discard support with the REQ_SECURE flag.
Use the opportunity to rename the queue flag as well, and remove the
dead checks for this flag in the RAID 1 and RAID 10 drivers that don't
claim support for secure erase.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
To avoid confusion between REQ_OP_FLUSH, which is handled by
request_fn drivers, and upper layers requesting the block layer
perform a flush sequence along with possibly a WRITE, this patch
renames REQ_FLUSH to REQ_PREFLUSH.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This adds a REQ_OP_FLUSH operation that is sent to request_fn
based drivers by the block layer's flush code, instead of
sending requests with the request->cmd_flags REQ_FLUSH bit set.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
We don't need bi_rw to be so large on 64 bit archs, so
reduce it to unsigned int.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The req operation REQ_OP is separated from the rq_flag_bits
definition. This converts the block layer drivers to
use req_op to get the op from the request struct.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Separate the op from the rq_flag_bits and have md
set/get the bio using bio_set_op_attrs/bio_op.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Separate the op from the rq_flag_bits and have bcache
set/get the bio using bio_set_op_attrs/bio_op.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Separate the op from the rq_flag_bits and have dm
set/get the bio using bio_set_op_attrs/bio_op.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
It looks like dm stats cares about the data direction
(READ vs WRITE) and does not need the bio/request flags.
Commands like REQ_FLUSH, REQ_DISCARD and REQ_WRITE_SAME
are currently always set with REQ_WRITE, so the extra check for
REQ_DISCARD in dm_stats_account_io is not needed.
This patch has it use the bio and request data_dir helpers
instead of accessing the bi_rw/cmd_flags directly. This makes
the next patches that remove the operation from the cmd_flags
and bi_rw easier, because we will no longer have the REQ_WRITE
bit set for operations like discards.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This converts the block issue discard helper and users to use
the bio_set_op_attrs accessor and only pass in the operation flags
like REQ_SEQURE.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
We currently set REQ_WRITE/WRITE for all non READ IOs
like discard, flush, writesame, etc. In the next patches where we
no longer set up the op as a bitmap, we will not be able to
detect a operation direction like writesame by testing if REQ_WRITE is
set.
This has bcache use the op_is_write helper which will do the right
thing.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
We currently set REQ_WRITE/WRITE for all non READ IOs
like discard, flush, writesame, etc. In the next patches where we
no longer set up the op as a bitmap, we will not be able to
detect a operation direction like writesame by testing if REQ_WRITE is
set.
This has dm use the op_is_write helper which will do the right
thing.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This has submit_bh users pass in the operation and flags separately,
so submit_bh_wbc can setup the bio op and bi_rw flags on the bio that
is submitted.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This has callers of submit_bio/submit_bio_wait set the bio->bi_rw
instead of passing it in. This makes that use the same as
generic_make_request and how we set the other bio fields.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Fixed up fs/ext4/crypto.c
Signed-off-by: Jens Axboe <axboe@fb.com>
Add a disk to an array which is performing recovery
is a little complicated, we need to do both reap the
sync thread and perform add disk for the case, then
it caused deadlock as follows.
linux44:~ # ps aux|grep md|grep D
root 1822 0.0 0.0 0 0 ? D 16:50 0:00 [md127_resync]
root 1848 0.0 0.0 19860 952 pts/0 D+ 16:50 0:00 mdadm --manage /dev/md127 --re-add /dev/vdb
linux44:~ # cat /proc/1848/stack
[<ffffffff8107afde>] kthread_stop+0x6e/0x120
[<ffffffffa051ddb0>] md_unregister_thread+0x40/0x80 [md_mod]
[<ffffffffa0526e45>] md_reap_sync_thread+0x15/0x150 [md_mod]
[<ffffffffa05271e0>] action_store+0x260/0x270 [md_mod]
[<ffffffffa05206b4>] md_attr_store+0xb4/0x100 [md_mod]
[<ffffffff81214a7e>] sysfs_write_file+0xbe/0x140
[<ffffffff811a6b98>] vfs_write+0xb8/0x1e0
[<ffffffff811a75b8>] SyS_write+0x48/0xa0
[<ffffffff8152a5c9>] system_call_fastpath+0x16/0x1b
[<00007f068ea1ed30>] 0x7f068ea1ed30
linux44:~ # cat /proc/1822/stack
[<ffffffffa05251a6>] md_do_sync+0x846/0xf40 [md_mod]
[<ffffffffa052402d>] md_thread+0x16d/0x180 [md_mod]
[<ffffffff8107ad94>] kthread+0xb4/0xc0
[<ffffffff8152a518>] ret_from_fork+0x58/0x90
Task1848 Task1822
md_attr_store (held reconfig_mutex by call mddev_lock())
action_store
md_reap_sync_thread
md_unregister_thread
kthread_stop md_wakeup_thread(mddev->thread);
wait_event(mddev->sb_wait, !test_bit(MD_CHANGE_PENDING))
md_check_recovery is triggered by wakeup mddev->thread,
but it can't clear MD_CHANGE_PENDING flag since it can't
get lock which was held by md_attr_store already.
To solve the deadlock problem, we move "->resync_finish()"
from md_do_sync to md_reap_sync_thread (after md_update_sb),
also MD_HELD_RESYNC_LOCK is introduced since it is possible
that node can't get resync lock in md_do_sync.
Then we do not need to wait for MD_CHANGE_PENDING is cleared
or not since metadata should be updated after md_update_sb,
so just call resync_finish if MD_HELD_RESYNC_LOCK is set.
We also unified the code after skip label, since set PENDING
for non-clustered case should be harmless.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Pull block fixes from Jens Axboe:
"A set of fixes that wasn't included in the first merge window pull
request. This pull request contains:
- A set of NVMe fixes from Keith, and one from Nic for the integrity
side of it.
- Fix from Ming, clearing ->mq_ops if we don't successfully setup a
queue for multiqueue.
- A set of stability fixes for bcache from Jiri, and also marking
bcache as orphaned as it's no longer actively maintained (in
mainline, at least)"
* 'for-linus' of git://git.kernel.dk/linux-block:
blk-mq: clear q->mq_ops if init fail
MAINTAINERS: mark bcache as orphan
bcache: bch_gc_thread() is not freezable
bcache: bch_allocator_thread() is not freezable
bcache: bch_writeback_thread() is not freezable
nvme/host: Add missing blk_integrity tag_size + flags assignments
NVMe: Add device ID's with stripe quirk
NVMe: Short-cut removal on surprise hot-unplug
NVMe: Allow user initiated rescan
NVMe: Reduce driver log spamming
NVMe: Unbind driver on failure
NVMe: Delete only created queues
NVMe: Allocate queues only for online cpus
In current handle_stripe_dirtying, the code prefers rmw with
PARITY_ENABLE_RMW; while prefers rcw with PARITY_PREFER_RMW.
This patch reverses this behavior.
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Shaohua Li <shli@fb.com>
bch_gc_thread() doesn't mark itself freezable, so calling try_to_freeze()
in its context is just an expensive no-op.
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
bch_allocator_thread() is calling try_to_freeze(), but that's just an
expensive no-op given the fact that the thread is not marked freezable.
Bucket allocator has to be up and running to the very last stages of the
suspend, as the bcache I/O that's in flight (think of writing an
hibernation image to a swap device served by bcache).
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
bch_writeback_thread() is calling try_to_freeze(), but that's just an
expensive no-op given the fact that the thread is not marked freezable.
I/O helper kthreads, exactly such as the bcache writeback thread, actually
shouldn't be freezable, because they are potentially necessary for
finalizing the image write-out.
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
Pull MD updates from Shaohua Li:
"Several patches from Guoqing fixing md-cluster bugs and several
patches from Heinz fixing dm-raid bugs"
* tag 'md/4.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md:
md-cluster: check the return value of process_recvd_msg
md-cluster: gather resync infos and enable recv_thread after bitmap is ready
md: set MD_CHANGE_PENDING in a atomic region
md: raid5: add prerequisite to run underneath dm-raid
md: raid10: add prerequisite to run underneath dm-raid
md: md.c: fix oops in mddev_suspend for raid0
md-cluster: fix ifnullfree.cocci warnings
md-cluster/bitmap: unplug bitmap to sync dirty pages to disk
md-cluster/bitmap: fix wrong page num in bitmap_file_clear_bit and bitmap_file_set_bit
md-cluster/bitmap: fix wrong calcuation of offset
md-cluster: sync bitmap when node received RESYNCING msg
md-cluster: always setup in-memory bitmap
md-cluster: wakeup thread if activated a spare disk
md-cluster: change array_sectors and update size are not supported
md-cluster: fix locking when node joins cluster during message broadcast
md-cluster: unregister thread if err happened
md-cluster: wake up thread to continue recovery
md-cluser: make resync_finish only called after pers->sync_request
md-cluster: change resync lock from asynchronous to synchronous
bio_inc_remaining() and the block core's new async
__blkdev_issue_discard() interface
- make DM multipath's fast code-paths lockless, using lockless_deference,
to significantly improve large NUMA performance when using blk-mq. The
m->lock spinlock contention was a serious bottleneck.
- a few other small code cleanups and Documentation fixes
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJXNdGVAAoJEMUj8QotnQNaYYgH/Rf2am46A78kcR5b9nN2I+Tb
+MkqQyf8mXUzNHOu3v93CVugT+tBZuJcpHPJgCSc/1GXtgsjHLvbkO2Mc+Ioe45S
PlUA3HdRzxHSJ365SdYvT+bY+QQlGiySelSBrJHlikXC88kz3wqyQ146BT1Rw/w+
t0mi1liNJtZHsuH+3uO9uxe5+H7476lB84i79Kz0x8Ygv5+urgaSvDBRO5EH/hkJ
LN2WJWHDQLT4MtHKCuiMiLpu/1HGvISN2QrMPsFjC1d1DbbZvRWAxYDwGaP/C277
IflPo7sA/nds5T2vqb0fRTPuxBnzXdFMMvf+VQX7pjCnxlhfaxBkvNtnFpxW+oA=
=iCyS
-----END PGP SIGNATURE-----
Merge tag 'dm-4.7-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
Pull device mapper updates from Mike Snitzer:
- based on Jens' 'for-4.7/core' to have DM thinp's discard support use
bio_inc_remaining() and the block core's new async __blkdev_issue_discard()
interface
- make DM multipath's fast code-paths lockless, using lockless_deference,
to significantly improve large NUMA performance when using blk-mq.
The m->lock spinlock contention was a serious bottleneck.
- a few other small code cleanups and Documentation fixes
* tag 'dm-4.7-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm thin: unroll issue_discard() to create longer discard bio chains
dm thin: use __blkdev_issue_discard for async discard support
dm thin: remove __bio_inc_remaining() and switch to using bio_inc_remaining()
dm raid: make sure no feature flags are set in metadata
dm ioctl: drop use of __GFP_REPEAT in copy_params()'s __vmalloc() call
dm stats: fix spelling mistake in Documentation
dm cache: update cache-policies.txt now that mq is an alias for smq
dm mpath: eliminate use of spinlock in IO fast-paths
dm mpath: move trigger_event member to the end of 'struct multipath'
dm mpath: use atomic_t for counting members of 'struct multipath'
dm mpath: switch to using bitops for state flags
dm thin: Remove return statement from void function
dm: remove unused mapped_device argument from free_tio()
Pull block driver updates from Jens Axboe:
"On top of the core pull request, this is the drivers pull request for
this merge window. This contains:
- Switch drivers to the new write back cache API, and kill off the
flush flags. From me.
- Kill the discard support for the STEC pci-e flash driver. It's
trivially broken, and apparently unmaintained, so it's safer to
just remove it. From Jeff Moyer.
- A set of lightnvm updates from the usual suspects (Matias/Javier,
and Simon), and fixes from Arnd, Jeff Mahoney, Sagi, and Wenwei
Tao.
- A set of updates for NVMe:
- Turn the controller state management into a proper state
machine. From Christoph.
- Shuffling of code in preparation for NVMe-over-fabrics, also
from Christoph.
- Cleanup of the command prep part from Ming Lin.
- Rewrite of the discard support from Ming Lin.
- Deadlock fix for namespace removal from Ming Lin.
- Use the now exported blk-mq tag helper for IO termination.
From Sagi.
- Various little fixes from Christoph, Guilherme, Keith, Ming
Lin, Wang Sheng-Hui.
- Convert mtip32xx to use the now exported blk-mq tag iter function,
from Keith"
* 'for-4.7/drivers' of git://git.kernel.dk/linux-block: (74 commits)
lightnvm: reserved space calculation incorrect
lightnvm: rename nr_pages to nr_ppas on nvm_rq
lightnvm: add is_cached entry to struct ppa_addr
lightnvm: expose gennvm_mark_blk to targets
lightnvm: remove mgt targets on mgt removal
lightnvm: pass dma address to hardware rather than pointer
lightnvm: do not assume sequential lun alloc.
nvme/lightnvm: Log using the ctrl named device
lightnvm: rename dma helper functions
lightnvm: enable metadata to be sent to device
lightnvm: do not free unused metadata on rrpc
lightnvm: fix out of bound ppa lun id on bb tbl
lightnvm: refactor set_bb_tbl for accepting ppa list
lightnvm: move responsibility for bad blk mgmt to target
lightnvm: make nvm_set_rqd_ppalist() aware of vblks
lightnvm: remove struct factory_blks
lightnvm: refactor device ops->get_bb_tbl()
lightnvm: introduce nvm_for_each_lun_ppa() macro
lightnvm: refactor dev->online_target to global nvm_targets
lightnvm: rename nvm_targets to nvm_tgt_type
...
There is little benefit to doing this but it does structure DM thinp's
code to more cleanly use the __blkdev_issue_discard() interface --
particularly in passdown_double_checking_shared_status().
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
With commit 38f2525533 ("block: add __blkdev_issue_discard") DM thinp
no longer needs to carry its own async discard method.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Joe Thornber <ejt@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
DM thinp's use of bio_inc_remaining() is critical to ensure the original
parent discard bio isn't completed before sub-discards have. DM thinp
needs this due to the extra quiescing that occurs, via multiple DM thinp
mappings, while processing large discards. As such DM thinp must build
the async discard bio chain after some delay -- so bio_inc_remaining()
is used to enable DM thinp to take a reference on the original parent
discard bio for each mapping. This allows the immediate use of
bio_endio() on that discard bio; but with the understanding that the
actual completion won't occur until each of the sub-discards'
per-mapping references are dropped.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Joe Thornber <ejt@redhat.com>
Given we don't yet support any feature flags in the dm-raid ondisk
metadata (see: 'features' member of 'struct dm_raid_superblock'),
add a check to ensure no flags are actually set, if any features are
set reject the activation of the RAID mapping.
This is to prevent possible data corruption in case of a kernel
downgrade when there'll potentially be feature flags set by a future
dm-raid target.
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
We don't need to run the full path of recv_daemon
if process_recvd_msg doesn't return 0.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The in-memory bitmap is not ready when node joins cluster,
so it doesn't make sense to make gather_all_resync_info()
called so earlier, we need to call it after the node's
bitmap is setup. Also, recv_thread could be wake up after
node joins cluster, but it could cause problem if node
receives RESYNCING message without persionality since
mddev->pers->quiesce is called in process_suspend_info.
This commit introduces a new cluster interface load_bitmaps
to fix above problems, load_bitmaps is called in bitmap_load
where bitmap and persionality are ready, and load_bitmaps
does the following tasks:
1. call gather_all_resync_info to load all the node's
bitmap info.
2. set MD_CLUSTER_ALREADY_IN_CLUSTER bit to recv_thread
could be wake up, and wake up recv_thread if there is
pending recv event.
Then ack_bast only wakes up recv_thread after IN_CLUSTER
bit is ready otherwise MD_CLUSTER_PENDING_RESYNC_EVENT is
set.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Some code waits for a metadata update by:
1. flagging that it is needed (MD_CHANGE_DEVS or MD_CHANGE_CLEAN)
2. setting MD_CHANGE_PENDING and waking the management thread
3. waiting for MD_CHANGE_PENDING to be cleared
If the first two are done without locking, the code in md_update_sb()
which checks if it needs to repeat might test if an update is needed
before step 1, then clear MD_CHANGE_PENDING after step 2, resulting
in the wait returning early.
So make sure all places that set MD_CHANGE_PENDING are atomicial, and
bit_clear_unless (suggested by Neil) is introduced for the purpose.
Cc: Martin Kepplinger <martink@posteo.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: <linux-kernel@vger.kernel.org>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
In case md runs underneath the dm-raid target, the mddev does not have
a request queue or gendisk, thus avoid accesses.
This patch adds a missing conditional to the raid5 personality.
Signed-of-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Shaohua Li <shli@fb.com>
In case md runs underneath the dm-raid target, the mddev does not have
a request queue or gendisk, thus avoid accesses to it.
This patch adds two missing conditionals to the raid10 personality.
Signed-of-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Introduced by upstream commit 70d9798b95
The raid0 personality does not create mddev->thread as oposed to
other personalities leading to its unconditional access in
mddev_suspend() causing an oops.
Patch checks for mddev->thread in order to keep the
intention of aforementioned commit.
Fixes: 70d9798b95 ("MD: warn for potential deadlock")
Cc: stable@vger.kernel.org (4.5+)
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Shaohua Li <shli@fb.com>
copy_params()'s use of __GFP_REPEAT for the __vmalloc() call doesn't make much
sense because vmalloc doesn't rely on costly high order allocations.
Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The primary motivation of this commit is to improve the scalability of
DM multipath on large NUMA systems where m->lock spinlock contention has
been proven to be a serious bottleneck on really fast storage.
The ability to atomically read a pointer, using lockless_dereference(),
is leveraged in this commit. But all pointer writes are still protected
by the m->lock spinlock (which is fine since these all now occur in the
slow-path).
The following functions no longer require the m->lock spinlock in their
fast-path: multipath_busy(), __multipath_map(), and do_end_io()
And choose_pgpath() is modified to _not_ update m->current_pgpath unless
it also switches the path-group. This is done to avoid needing to take
the m->lock everytime __multipath_map() calls choose_pgpath().
But m->current_pgpath will be reset if it is failed via fail_path().
Suggested-by: Jeff Moyer <jmoyer@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Tested-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Allows the 'work_mutex' member to no longer cross a cacheline.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Tested-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The use of atomic_t for nr_valid_paths, pg_init_in_progress and
pg_init_count will allow relaxing the use of the m->lock spinlock.
Suggested-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Tested-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mechanical change that doesn't make any real effort to reduce the use of
m->lock; that will come later (once atomics are used for counters, etc).
Suggested-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Tested-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Return statement at the end of a void function is useless.
The Coccinelle semantic patch used to make this change is as follows:
//<smpl>
@@
identifier f;
expression e;
@@
void f(...) {
<...
- return
e;
...>
}
//</smpl>
Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
drivers/md/bitmap.c:2049:6-11: WARNING: NULL check before freeing functions like kfree, debugfs_remove, debugfs_remove_recursive or usb_free_urb is not needed. Maybe consider reorganizing relevant code to avoid passing NULL values.
NULL check before some freeing functions is not needed.
Based on checkpatch warning
"kfree(NULL) is safe this check is probably not required"
and kfreeaddr.cocci by Julia Lawall.
Generated by: scripts/coccinelle/free/ifnullfree.cocci
Acked-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Shaohua Li <shli@fb.com>
This patch is doing two distinct but related things.
1. It adds bitmap_unplug() for the main bitmap (mddev->bitmap). As bit
have been set, BITMAP_PAGE_DIRTY is set so bitmap_deamon_work() will
not write those pages out in its regular scans, only bitmap_unplug()
will. If there are no writes to the array, bitmap_unplug() won't be
called, so we need to call it explicitly here.
2. bitmap_write_all() is a bit of a confusing interface as it doesn't
actually write anything. The current code for writing "bitmap" works
but this change makes it a bit clearer.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The pnum passed to set_page_attr and test_page_attr should from
0 to storage.file_pages - 1, but bitmap_file_set_bit and
bitmap_file_clear_bit call set_page_attr and test_page_attr with
page->index parameter while page->index has already added node_offset
before.
So we need to minus node_offset in both bitmap_file_clear_bit
and bitmap_file_set_bit.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The offset is wrong in bitmap_storage_alloc, we should
set it like below in bitmap_init_from_disk().
node_offset = bitmap->cluster_slot * (DIV_ROUND_UP(store->bytes, PAGE_SIZE));
Because 'offset' is only assigned to 'page->index' and
that is usually over-written by read_sb_page. So it does
not cause problem in general, but it still need to be fixed.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
If the node received RESYNCING message which means
another node will perform resync with the area, then
we don't want to do it again in another node.
Let's set RESYNC_MASK and clear NEEDED_MASK for the
region from old-low to new-low which has finished
syncing, and the region from old-hi to new-hi is about
to syncing, bitmap_sync_with_cluste is introduced for
the purpose.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The in-memory bitmap for raid is allocated on demand,
then for cluster scenario, it is possible that slave
node which received RESYNCING message doesn't have the
in-memory bitmap when master node is perform resyncing,
so we can't make bitmap is match up well among each
nodes.
So for cluster scenario, we need always preserve the
bitmap, and ensure the page will not be freed. And a
no_hijack flag is introduced to both bitmap_checkpage
and bitmap_get_counter, which makes cluster raid returns
fail once allocate failed.
And the next patch is relied on this change since it
keeps sync bitmap among each nodes during resyncing
stage.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
When a device is re-added, it will ultimately need
to be activated and that happens in md_check_recovery,
so we need to set MD_RECOVERY_NEEDED right after
remove_and_add_spares.
A specifical issue without the change is that when
one node perform fail/remove/readd on a disk, but
slave nodes could not add the disk back to array as
expected (added as missed instead of in sync). So
give slave nodes a chance to do resync.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Currently, some features are not supported yet,
such as change array_sectors and update size, so
return EINVAL for them and listed it in document.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
If a node joins the cluster while a message broadcast
is under way, a lock issue could happen as follows.
For a cluster which included two nodes, if node A is
calling __sendmsg before up-convert CR to EX on ack,
and node B released CR on ack. But if a new node C
joins the cluster and it doesn't receive the message
which A sent before, so it could hold CR on ack before
A up-convert CR to EX on ack.
So a node joining the cluster should get an EX lock on
the "token" first to ensure no broadcast is ongoing,
then release it after held CR on ack.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The two threads need to be unregistered if a node
can't join cluster successfully.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
In recovery case, we need to set MD_RECOVERY_NEEDED
and wake up thread only if recover is not finished.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
It is not reasonable that cluster raid to release resync
lock before the last pers->sync_request has finished.
As the metadata will be changed when node performs resync,
we need to inform other nodes to update metadata, so the
MD_CHANGE_PENDING flag is set before finish resync.
Then metadata_update_finish is move ahead to ensure that
METADATA_UPDATED msg is sent before finish resync, and
metadata_update_start need to be run after "repeat:" label
accordingly.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
If multiple nodes choose to attempt do resync at the same time
they need to be serialized so they don't duplicate effort. This
serialization is done by locking the 'resync' DLM lock.
Currently if a node cannot get the lock immediately it doesn't
request notification when the lock becomes available (i.e.
DLM_LKF_NOQUEUE is set), so it may not reliably find out when it
is safe to try again.
Rather than trying to arrange an async wake-up when the lock
becomes available, switch to using synchronous locking - this is
a lot easier to think about. As it is not permitted to block in
the 'raid1d' thread, move the locking to the resync thread. So
the rsync thread is forked immediately, but it blocks until the
resync lock is available. Once the lock is locked it checks again
if any resync action is needed.
A particular symptom of the current problem is that a node can
get stuck with "resync=pending" indefinitely.
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Pull MD fixes from Shaohua Li:
"This update includes several trival fixes. The only important one is
to fix MD bio merge, which has big performance impact"
* tag 'md/4.6-rc6-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md:
raid5: delete unnecessary warnning
MD: make bio mergeable
md/raid0: remove empty line printk from dump_zones
md/raid0: fix uninitialized variable bug
If device has R5_LOCKED set, it's legit device has R5_SkipCopy set and page !=
orig_page. After R5_LOCKED is clear, handle_stripe_clean_event will clear the
SkipCopy flag and set page to orig_page. So the warning is unnecessary.
Reported-by: Joey Liao <joeyliao@qnap.com>
Signed-off-by: Shaohua Li <shli@fb.com>
blk_queue_split marks bio unmergeable, which makes sense for normal bio.
But if dispatching the bio to underlayer disk, the blk_queue_split
checks are invalid, hence it's possible the bio becomes mergeable.
In the reported bug, this bug causes trim against raid0 performance slash
https://bugzilla.kernel.org/show_bug.cgi?id=117051
Reported-and-tested-by: Park Ju Hyung <qkrwngud825@gmail.com>
Fixes: 6ac45aeb6bca(block: avoid to merge splitted bio)
Cc: stable@vger.kernel.org (v4.3+)
Cc: Ming Lei <ming.lei@canonical.com>
Cc: Neil Brown <neilb@suse.de>
Reviewed-by: Jens Axboe <axboe@fb.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Remove the final printk. All preceding output is already properly
newline-terminated and the printk isn't even KERN_CONT to begin with,
so it only adds one empty line to the log.
Signed-off-by: Michal Pecio <michal.pecio@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Commit 9567366fef ("dm cache metadata: fix READ_LOCK macros and
cleanup WRITE_LOCK macros") uses down_write() instead of down_read() in
cmd_read_lock(), yet up_read() is used to release the lock in
READ_UNLOCK(). Fix it.
Fixes: 9567366fef ("dm cache metadata: fix READ_LOCK macros and cleanup WRITE_LOCK macros")
Cc: stable@vger.kernel.org
Signed-off-by: Ahmed Samy <f.fallen45@gmail.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The READ_LOCK macro was incorrectly returning -EINVAL if
dm_bm_is_read_only() was true -- it will always be true once the cache
metadata transitions to read-only by dm_cache_metadata_set_read_only().
Wrap READ_LOCK and WRITE_LOCK multi-statement macros in do {} while(0).
Also, all accesses of the 'cmd' argument passed to these related macros
are now encapsulated in parenthesis.
A follow-up patch can be developed to eliminate the use of macros in
favor of pure C code. Avoiding that now given that this needs to apply
to stable@.
Reported-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Fixes: d14fcf3dd7 ("dm cache: make sure every metadata function checks fail_io")
Cc: stable@vger.kernel.org
If this function fails the callers expect that *private_conf is set to
an ERR_PTR() but that isn't true for the first error path where we can't
allocate "conf". It leads to some uninitialized variable bugs.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Now that we converted everything to the newer block write cache
interface, kill off the queue flush_flags and queueable flush
entries.
Signed-off-by: Jens Axboe <axboe@fb.com>
Commit c80914e81e ("dm: return error if bio_integrity_clone() fails
in clone_bio()") changed clone_bio() such that if it does return error
then the alloc_tio() created resources (both the bio that was allocated
to be a clone and the containing dm_target_io struct) will leak.
Fix this by calling free_tio() in __clone_and_map_data_bio()'s
clone_bio() error path.
Fixes: c80914e81e ("dm: return error if bio_integrity_clone() fails in clone_bio()")
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Pull MD fixes from Shaohua Li:
"This update mainly fixes bugs:
- fix error handling (Guoqing)
- fix a crash when a disk is hotremoved (me)
- fix a dead loop (Wei Fang)"
* tag 'md/4.6-rc2-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md:
md/bitmap: clear bitmap if bitmap_create failed
MD: add rdev reference for super write
md: fix a trivial typo in comments
md:raid1: fix a dead loop when read from a WriteMostly disk
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.
This promise never materialized. And unlikely will.
We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE. And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.
Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.
Let's stop pretending that pages in page cache are special. They are
not.
The changes are pretty straight-forward:
- <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};
- page_cache_get() -> get_page();
- page_cache_release() -> put_page();
This patch contains automated changes generated with coccinelle using
script below. For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.
The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.
There are few places in the code where coccinelle didn't reach. I'll
fix them manually in a separate patch. Comments and documentation also
will be addressed with the separate patch.
virtual patch
@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT
@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE
@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK
@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)
@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)
@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If bitmap_create returns an error, we need to call
either bitmap_destroy or bitmap_free to do clean up,
and the selection is based on mddev->bitmap is set
or not.
And the sysfs_put(bitmap->sysfs_can_clear) is moved
from bitmap_destroy to bitmap_free, and the comment
of bitmap_create is changed as well.
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
If first_bad == this_sector when we get the WriteMostly disk
in read_balance(), valid disk will be returned with zero
max_sectors. It'll lead to a dead loop in make_request(), and
OOM will happen because of endless allocation of struct bio.
Since we can't get data from this disk in this case, so
continue for another disk.
Signed-off-by: Wei Fang <fangwei1@huawei.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Pull MD updates from Shaohua Li:
"This update mainly fixes bugs.
- a raid5 discard related fix from Jes
- a MD multipath bio clone fix from Ming
- raid1 error handling deadlock fix from Nate and corresponding
raid10 fix from myself
- a raid5 stripe batch fix from Neil
- a patch from Sebastian to avoid unnecessary uevent
- several cleanup/debug patches"
* tag 'md/4.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md:
md/raid5: Cleanup cpu hotplug notifier
raid10: include bio_end_io_list in nr_queued to prevent freeze_array hang
raid1: include bio_end_io_list in nr_queued to prevent freeze_array hang
md: fix typos for stipe
md/bitmap: remove redundant return in bitmap_checkpage
md/raid1: remove unnecessary BUG_ON
md: multipath: don't hardcopy bio in .make_request path
md/raid5: output stripe state for debug
md/raid5: preserve STRIPE_PREREAD_ACTIVE in break_stripe_batch_list
Update MD git tree URL
md/bitmap: remove redundant check
MD: warn for potential deadlock
md: Drop sending a change uevent when stopping
RAID5: revert e9e4c377e2 to fix a livelock
RAID5: check_reshape() shouldn't call mddev_suspend
md/raid5: Compare apples to apples (or sectors to sectors)
Pull block driver updates from Jens Axboe:
"This is the block driver pull request for this merge window. It sits
on top of for-4.6/core, that was just sent out.
This contains:
- A set of fixes for lightnvm. One from Alan, fixing an overflow,
and the rest from the usual suspects, Javier and Matias.
- A set of fixes for nbd from Markus and Dan, and a fixup from Arnd
for correct usage of the signed 64-bit divider.
- A set of bug fixes for the Micron mtip32xx, from Asai.
- A fix for the brd discard handling from Bart.
- Update the maintainers entry for cciss, since that hardware has
transferred ownership.
- Three bug fixes for bcache from Eric Wheeler.
- Set of fixes for xen-blk{back,front} from Jan and Konrad.
- Removal of the cpqarray driver. It has been disabled in Kconfig
since 2013, and we were initially scheduled to remove it in 3.15.
- Various updates and fixes for NVMe, with the most important being:
- Removal of the per-device NVMe thread, replacing that with a
watchdog timer instead. From Christoph.
- Exposing the namespace WWID through sysfs, from Keith.
- Set of cleanups from Ming Lin.
- Logging the controller device name instead of the underlying
PCI device name, from Sagi.
- And a bunch of fixes and optimizations from the usual suspects
in this area"
* 'for-4.6/drivers' of git://git.kernel.dk/linux-block: (49 commits)
NVMe: Expose ns wwid through single sysfs entry
drivers:block: cpqarray clean up
brd: Fix discard request processing
cpqarray: remove it from the kernel
cciss: update MAINTAINERS
NVMe: Remove unused sq_head read in completion path
bcache: fix cache_set_flush() NULL pointer dereference on OOM
bcache: cleaned up error handling around register_cache()
bcache: fix race of writeback thread starting before complete initialization
NVMe: Create discard zero quirk white list
nbd: use correct div_s64 helper
mtip32xx: remove unneeded variable in mtip_cmd_timeout()
lightnvm: generalize rrpc ppa calculations
lightnvm: remove struct nvm_dev->total_blocks
lightnvm: rename ->nr_pages to ->nr_sects
lightnvm: update closed list outside of intr context
xen/blback: Fit the important information of the thread in 17 characters
lightnvm: fold get bb tbl when using dual/quad plane mode
lightnvm: fix up nonsensical configure overrun checking
xen-blkback: advertise indirect segment support earlier
...
The raid456_cpu_notify() hotplug callback lacks handling of the
CPU_UP_CANCELED case. That means if CPU_UP_PREPARE fails, the scratch
buffer is leaked.
Add handling for CPU_UP_CANCELED[_FROZEN] hotplug notifier transitions
to free the scratch buffer.
CC: Shaohua Li <shli@kernel.org>
CC: linux-raid@vger.kernel.org
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
Signed-off-by: Shaohua Li <shli@fb.com>
This is the raid10 counterpart of the bug fixed by Nate
(raid1: include bio_end_io_list in nr_queued to prevent freeze_array hang)
Fixes: 95af587e95(md/raid10: ensure device failure recorded before write request returns)
Cc: stable@vger.kernel.org (V4.3+)
Cc: Nate Dailey <nate.dailey@stratus.com>
Signed-off-by: Shaohua Li <shli@fb.com>
If raid1d is handling a mix of read and write errors, handle_read_error's
call to freeze_array can get stuck.
This can happen because, though the bio_end_io_list is initially drained,
writes can be added to it via handle_write_finished as the retry_list
is processed. These writes contribute to nr_pending but are not included
in nr_queued.
If a later entry on the retry_list triggers a call to handle_read_error,
freeze array hangs waiting for nr_pending == nr_queued+extra. The writes
on the bio_end_io_list aren't included in nr_queued so the condition will
never be satisfied.
To prevent the hang, include bio_end_io_list writes in nr_queued.
There's probably a better way to handle decrementing nr_queued, but this
seemed like the safest way to avoid breaking surrounding code.
I'm happy to supply the script I used to repro this hang.
Fixes: 55ce74d4bfe1b(md/raid1: ensure device failure recorded before write request returns.)
Cc: stable@vger.kernel.org (v4.3+)
Signed-off-by: Nate Dailey <nate.dailey@stratus.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Pull crypto update from Herbert Xu:
"Here is the crypto update for 4.6:
API:
- Convert remaining crypto_hash users to shash or ahash, also convert
blkcipher/ablkcipher users to skcipher.
- Remove crypto_hash interface.
- Remove crypto_pcomp interface.
- Add crypto engine for async cipher drivers.
- Add akcipher documentation.
- Add skcipher documentation.
Algorithms:
- Rename crypto/crc32 to avoid name clash with lib/crc32.
- Fix bug in keywrap where we zero the wrong pointer.
Drivers:
- Support T5/M5, T7/M7 SPARC CPUs in n2 hwrng driver.
- Add PIC32 hwrng driver.
- Support BCM6368 in bcm63xx hwrng driver.
- Pack structs for 32-bit compat users in qat.
- Use crypto engine in omap-aes.
- Add support for sama5d2x SoCs in atmel-sha.
- Make atmel-sha available again.
- Make sahara hashing available again.
- Make ccp hashing available again.
- Make sha1-mb available again.
- Add support for multiple devices in ccp.
- Improve DMA performance in caam.
- Add hashing support to rockchip"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (116 commits)
crypto: qat - remove redundant arbiter configuration
crypto: ux500 - fix checks of error code returned by devm_ioremap_resource()
crypto: atmel - fix checks of error code returned by devm_ioremap_resource()
crypto: qat - Change the definition of icp_qat_uof_regtype
hwrng: exynos - use __maybe_unused to hide pm functions
crypto: ccp - Add abstraction for device-specific calls
crypto: ccp - CCP versioning support
crypto: ccp - Support for multiple CCPs
crypto: ccp - Remove check for x86 family and model
crypto: ccp - memset request context to zero during import
lib/mpi: use "static inline" instead of "extern inline"
lib/mpi: avoid assembler warning
hwrng: bcm63xx - fix non device tree compatibility
crypto: testmgr - allow rfc3686 aes-ctr variants in fips mode.
crypto: qat - The AE id should be less than the maximal AE number
lib/mpi: Endianness fix
crypto: rockchip - add hash support for crypto engine in rk3288
crypto: xts - fix compile errors
crypto: doc - add skcipher API documentation
crypto: doc - update AEAD AD handling
...
An "old" (.request_fn) DM 'struct request' stores a pointer to the
associated 'struct dm_rq_target_io' in rq->special.
dm_requeue_original_request(), previously named
dm_requeue_unmapped_original_request(), called dm_unprep_request() to
reset rq->special to NULL. But rq_end_stats() would go on to hit a NULL
pointer deference because its call to tio_from_request() returned NULL.
Fix this by calling rq_end_stats() _before_ dm_unprep_request()
Signed-off-by: Bryn M. Reeves <bmr@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Fixes: e262f34741 ("dm stats: add support for request-based DM devices")
Cc: stable@vger.kernel.org # 4.2+
The "return 0" is not needed since bitmap_checkpage
will finally return 0 for the case.
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Since bitmap_start_sync will not return until
sync_blocks is not less than PAGE_SIZE>>9, so
the BUG_ON is not needed anymore.
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Inside multipath_make_request(), multipath maps the incoming
bio into low level device's bio, but it is totally wrong to
copy the bio into mapped bio via '*mapped_bio = *bio'. For
example, .__bi_remaining is kept in the copy, especially if
the incoming bio is chained to via bio splitting, so .bi_end_io
can't be called for the mapped bio at all in the completing path
in this kind of situation.
This patch fixes the issue by using clone style.
Cc: stable@vger.kernel.org (v3.14+)
Reported-and-tested-by: Andrea Righi <righi.andrea@gmail.com>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Commit 0a927c2f02 ("dm thin: return -ENOSPC when erroring retry list due
to out of data space") was a step in the right direction but didn't go
far enough.
Add a new 'out_of_data_space' flag to 'struct pool' and set it if/when
the pool runs of of data space. This fixes cell_error() and
error_retry_list() to not blindly return -EIO.
We cannot rely on the 'error_if_no_space' feature flag since it is
transient (in that it can be reset once space is added, plus it only
controls whether errors are issued, it doesn't reflect whether the
pool is actually out of space).
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Otherwise operations may be attempted that will only ever go on to crash
(since the metadata device is either missing or unreliable if 'fail_io'
is set).
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
clone_bio() now checks if bio_integrity_clone() returned an error rather
than just drop it on the floor.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
If a transaction abort has failed then we can no longer use the metadata
device. Typically this happens if the superblock is unreadable.
This fix addresses a crash seen during metadata device failure testing.
Fixes: 8a01a6af75 ("dm thin: prefetch missing metadata pages")
Cc: stable@vger.kernel.org # 3.19+
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
smq seems to be performing better than the old mq policy in all
situations, as well as using a quarter of the memory.
Make 'mq' an alias for 'smq' when choosing a cache policy. The tunables
that were present for the old mq are faked, and have no effect. mq
should be considered deprecated now.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
md->queue and q are the same thing in dm_old_init_request_queue() and
dm_mq_init_request_queue().
Also drop the temporary 'struct request_queue *q' in
dm_old_init_request_queue().
Signed-off-by: Bob Liu <bob.liu@oracle.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Saves 16 bytes by eliminating 4 4byte holes but more importantly:
numerous members that crossed cachelines were fixed.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Change the map pointer in 'struct mapped_device' from 'struct dm_table
__rcu *' to 'void __rcu *' to avoid the need for the dummy definition.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Allows user to control which NUMA node the memory for DM device
structures (e.g. mapped_device, request_queue, gendisk, blk_mq_tag_set)
is allocated from.
Defaults to NUMA_NO_NODE (-1). Allowable range is from -1 until the
last online NUMA node id.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
fail_path() will print a "Failing path ..." message but reinstate_path()
doesn't print a "Reinstating path ...". Add that message to
reinstate_path() to add symmetry and aid system debugging.
Remove reinstate_path()'s check for the path_selector providing
.reinstate_path hook. All path selectors provide this and any future
ones must too.
activate_path() calls pg_init_done() with SCSI_DH_DEV_OFFLINED but
pg_init_done() doesn't expicitly handle it in its swicth statement. Add
SCSI_DH_DEV_OFFLINED to the default case.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Neil recently fixed an obscure race in break_stripe_batch_list. Debug would be
quite convenient if we know the stripe state. This is what this patch does.
Signed-off-by: Shaohua Li <shli@fb.com>
break_stripe_batch_list breaks up a batch and copies some flags from
the batch head to the members, preserving others.
It doesn't preserve or copy STRIPE_PREREAD_ACTIVE. This is not
normally a problem as STRIPE_PREREAD_ACTIVE is cleared when a
stripe_head is added to a batch, and is not set on stripe_heads
already in a batch.
However there is no locking to ensure one thread doesn't set the flag
after it has just been cleared in another. This does occasionally happen.
md/raid5 maintains a count of the number of stripe_heads with
STRIPE_PREREAD_ACTIVE set: conf->preread_active_stripes. When
break_stripe_batch_list clears STRIPE_PREREAD_ACTIVE inadvertently
this could becomes incorrect and will never again return to zero.
md/raid5 delays the handling of some stripe_heads until
preread_active_stripes becomes zero. So when the above mention race
happens, those stripe_heads become blocked and never progress,
resulting is write to the array handing.
So: change break_stripe_batch_list to preserve STRIPE_PREREAD_ACTIVE
in the members of a batch.
URL: https://bugzilla.kernel.org/show_bug.cgi?id=108741
URL: https://bugzilla.redhat.com/show_bug.cgi?id=1258153
URL: http://thread.gmane.org/5649C0E9.2030204@zoner.cz
Reported-by: Martin Svec <martin.svec@zoner.cz> (and others)
Tested-by: Tom Weber <linux@junkyard.4t2.com>
Fixes: 1b956f7a8f ("md/raid5: be more selective about distributing flags across batch.")
Cc: stable@vger.kernel.org (v4.1 and later)
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
When bch_cache_set_alloc() fails to kzalloc the cache_set, the
asyncronous closure handling tries to dereference a cache_set that
hadn't yet been allocated inside of cache_set_flush() which is called
by __cache_set_unregister() during cleanup. This appears to happen only
during an OOM condition on bcache_register.
Signed-off-by: Eric Wheeler <bcache@linux.ewheeler.net>
Cc: stable@vger.kernel.org
The bch_writeback_thread might BUG_ON in read_dirty() if
dc->sb==BDEV_STATE_DIRTY and bch_sectors_dirty_init has not yet completed
its related initialization. This patch downs the dc->writeback_lock until
after initialization is complete, thus preventing bch_writeback_thread
from proceeding prematurely.
See this thread:
http://thread.gmane.org/gmane.linux.kernel.bcache.devel/3453
Signed-off-by: Eric Wheeler <bcache@linux.ewheeler.net>
Tested-by: Marc MERLIN <marc@merlins.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
daemon_sleep is an unsigned, so testing if it's 0 or less than 1 does
the same thing.
Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The personality thread shouldn't call mddev_suspend(). Because
mddev_suspend() will for all IO finish, but IO is handled in personality
thread, so this could cause deadlock. To trigger this early, add a
warning if mddev_suspend() is called from personality thread.
Suggested-by: NeilBrown <neilb@suse.com>
Cc: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Signed-off-by: Shaohua Li <shli@fb.com>
When stopping an MD device, then its device node /dev/mdX may still
exist afterwards or it is recreated by udev. The next open() call
can lead to creation of an inoperable MD device. The reason for
this is that a change event (KOBJ_CHANGE) is sent to udev which
races against the remove event (KOBJ_REMOVE) from md_free().
So drop sending the change event.
A change is likely also required in mdadm as many versions send the
change event to udev as well.
Neil mentioned the change event is a workaround for old kernel
Commit: 934d9c23b4 ("md: destroy partitions and notify udev when md array is stopped.")
new mdadm can handle device remove now, so this isn't required any more.
Cc: NeilBrown <neilb@suse.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Sebastian Parschauer <sebastian.riemer@profitbricks.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Revert commit
e9e4c377e2f563(md/raid5: per hash value and exclusive wait_for_stripe)
The problem is raid5_get_active_stripe waits on
conf->wait_for_stripe[hash]. Assume hash is 0. My test release stripes
in this order:
- release all stripes with hash 0
- raid5_get_active_stripe still sleeps since active_stripes >
max_nr_stripes * 3 / 4
- release all stripes with hash other than 0. active_stripes becomes 0
- raid5_get_active_stripe still sleeps, since nobody wakes up
wait_for_stripe[0]
The system live locks. The problem is active_stripes isn't a per-hash
count. Revert the patch makes the live lock go away.
Cc: stable@vger.kernel.org (v4.2+)
Cc: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Cc: NeilBrown <neilb@suse.de>
Signed-off-by: Shaohua Li <shli@fb.com>
check_reshape() is called from raid5d thread. raid5d thread shouldn't
call mddev_suspend(), because mddev_suspend() waits for all IO finish
but IO is handled in raid5d thread, we could easily deadlock here.
This issue is introduced by
738a273 ("md/raid5: fix allocation of 'scribble' array.")
Cc: stable@vger.kernel.org (v4.1+)
Reported-and-tested-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
'max_discard_sectors' is in sectors, while 'stripe' is in bytes.
This fixes the problem where DISCARD would get disabled on some larger
RAID5 configurations (6 or more drives in my testing), while it worked
as expected with smaller configurations.
Fixes: 620125f2bf ("MD: raid5 trim support")
Cc: stable@vger.kernel.org v3.7+
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Now that dm-mpath core is lockless in the per-IO fast path it is
critical, for performance, to have the .select_path hook
(rr_select_path) also be as lockless as possible.
The new percpu members of 'struct selector' allow for lockless support
of 'repeat_count' governed repeat use of a previously selected path. If
a path fails while it is 'current_path' the worst case is concurrent IO
might be mapped to the failed path until the .fail_path hook
(rr_fail_path) is called.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
If a path selector has any use for a repeat_count it should be handled
locally and not depend on the dm-mpath core to be concerned with it.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Proper locking of the lists used by the path selectors should be handled
within the selectors (relying on dm-mpath.c code's use of the m->lock
spinlock was reckless).
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Preparation for making __multipath_map() avoid taking the m->lock
spinlock -- in favor of using RCU locking.
repeat_count was primarily for bio-based DM multipath's benefit. There
is really no need for it anymore now that DM multipath is request-based.
As such, repeat_count > 1 is no longer honored and a warning is
displayed if the user attempts to use a value > 1. This is a temporary
change for the round-robin path-selector (as a later commit will restore
its support for repeat_count > 1).
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
There isn't any need to support both old .request_fn and blk-mq paths
in the blk-mq specific portion of __multipath_map(). Call
blk_mq_alloc_request() directly rather than use blk_get_request().
Similarly, call blk_mq_free_request(), rather than blk_put_request(), in
multipath_release_clone().
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Allow the multipath target to avoid making small allocations for each
'struct dm_mpath_io' that is needed for each request.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
This will allow DM multipath to use a portion of the blk-mq pdu space
for target data (e.g. struct dm_mpath_io).
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Rename various methods to have either a "dm_old" or "dm_mq" prefix.
Improve code comments to assist with understanding the duality of code
that handles both "dm_old" and "dm_mq" cases.
It is no much easier to quickly look at the code and _know_ that a given
method is either 1) "dm_old" only 2) "dm_mq" only 3) common to both.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Remove all fiddley code that propped up this support for a blk-mq
request-queue ontop of all .request_fn devices.
Testing has proven this niche request-based dm-mq mode to be buggy, when
testing fault tolerance with DM multipath, and there is no point trying
to preserve it.
Should help improve efficiency of pure dm-mq code and make code
maintenance less delicate.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
old_stop_queue() was checking blk_queue_stopped() without holding the
q->queue_lock.
dm_requeue_original_request() needed to check blk_queue_stopped(), with
q->queue_lock held, before calling blk_mq_kick_requeue_list(). And a
side-effect of that change is start_queue() must also call
blk_mq_kick_requeue_list().
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The blk_mq_tag_set is only needed for dm-mq support. There is point
wasting space in 'struct mapped_device' for non-dm-mq devices.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> # check kzalloc return
Allow user to change these values via module params or sysfs.
'dm_mq_nr_hw_queues' defaults to 1 (max 32).
'dm_mq_queue_depth' defaults to 2048 (up from 64, which proved far too
small under moderate sized workloads -- the dm-multipath device would
continuously block waiting for tags (requests) to become available).
The maximum is BLK_MQ_MAX_DEPTH (currently 10240).
Keep in mind the total number of pre-allocated requests per
request-based dm-mq device is 'dm_mq_nr_hw_queues' * 'dm_mq_queue_depth'
(currently 2048).
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
DM multipath is the only request-based DM target -- which only supports
tables with a single target that is immutable. Leverage this fact in
dm_request_fn().
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
DM multipath is the only dm-mq target. But that aside, request-based DM
only supports tables with a single target that is immutable. Leverage
this fact in dm_mq_queue_rq() by using the 'immutable_target' stored in
the mapped_device when the table was made active. This saves the need
to even take the read-side of the SRCU via dm_{get,put}_live_table.
If the active DM table does not have an immutable target (e.g. "error"
target was swapped in) then fallback to the slow-path where the target
is looked up from the live table.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The DM_TARGET_WILDCARD feature indicates that the "error" target may
replace any target; even immutable targets. This feature will be useful
to preserve the ability to replace the "multipath" target even once it
is formally converted over to having the DM_TARGET_IMMUTABLE feature.
Also, implicit in the DM_TARGET_WILDCARD feature flag being set is that
.map, .map_rq, .clone_and_map_rq and .release_clone_rq are all defined
in the target_type.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The request-based DM support for checking queue congestion doesn't
require access to the live DM table.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Request-based DM's blk-mq support (dm-mq) was reported to be 50% slower
than if an underlying null_blk device were used directly. One of the
reasons for this drop in performance is that blk_insert_clone_request()
was calling blk_mq_insert_request() with @async=true. This forced the
use of kblockd_schedule_delayed_work_on() to run the blk-mq hw queues
which ushered in ping-ponging between process context (fio in this case)
and kblockd's kworker to submit the cloned request. The ftrace
function_graph tracer showed:
kworker-2013 => fio-12190
fio-12190 => kworker-2013
...
kworker-2013 => fio-12190
fio-12190 => kworker-2013
...
Fixing blk_insert_clone_request()'s blk_mq_insert_request() call to
_not_ use kblockd to submit the cloned requests isn't enough to
eliminate the observed context switches.
In addition to this dm-mq specific blk-core fix, there are 2 DM core
fixes to dm-mq that (when paired with the blk-core fix) completely
eliminate the observed context switching:
1) don't blk_mq_run_hw_queues in blk-mq request completion
Motivated by desire to reduce overhead of dm-mq, punting to kblockd
just increases context switches.
In my testing against a really fast null_blk device there was no benefit
to running blk_mq_run_hw_queues() on completion (and no other blk-mq
driver does this). So hopefully this change doesn't induce the need for
yet another revert like commit 621739b00e !
2) use blk_mq_complete_request() in dm_complete_request()
blk_complete_request() doesn't offer the traditional q->mq_ops vs
.request_fn branching pattern that other historic block interfaces
do (e.g. blk_get_request). Using blk_mq_complete_request() for
blk-mq requests is important for performance. It should be noted
that, like blk_complete_request(), blk_mq_complete_request() doesn't
natively handle partial completions -- but the request-based
DM-multipath target does provide the required partial completion
support by dm.c:end_clone_bio() triggering requeueing of the request
via dm-mpath.c:multipath_end_io()'s return of DM_ENDIO_REQUEUE.
dm-mq fix#2 is _much_ more important than #1 for eliminating the
context switches.
Before: cpu : usr=15.10%, sys=59.39%, ctx=7905181, majf=0, minf=475
After: cpu : usr=20.60%, sys=79.35%, ctx=2008, majf=0, minf=472
With these changes multithreaded async read IOPs improved from ~950K
to ~1350K for this dm-mq stacked on null_blk test-case. The raw read
IOPs of the underlying null_blk device for the same workload is ~1950K.
Fixes: 7fb4898e0 ("block: add blk-mq support to blk_insert_cloned_request()")
Fixes: bfebd1cdb ("dm: add full blk-mq support to request-based DM")
Cc: stable@vger.kernel.org # 4.1+
Reported-by: Sagi Grimberg <sagig@dev.mellanox.co.il>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Rename dm_get_live_table_for_ioctl to dm_grab_bdev_for_ioctl and have it
do the dm_{get,put}_live_table() rather than split those operations.
The dm_grab_bdev_for_ioctl() callers only care about the block_device
associated with a singleton DM device so there isn't any need to retain
a reference to the live DM table. It is sufficient to:
1) dm_get_live_table()
2) bdgrab() the bdev associated with the singleton table's target
3) dm_put_live_table()
4) bdput() the bdev
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
None of the callers actually used the returned target.
Also, just reuse bdev pointer passed to dm_blk_ioctl().
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
This patch replaces uses of ablkcipher with skcipher, and the long
obsolete hash interface with ahash.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
There are several places we allocate dlm_lock_resource, but not free it.
leave() need free a lock resource too (from Guoqing)
Cc: Goldwyn Rodrigues <rgoldwyn@suse.com>
Cc: Guoqing Jiang <gqjiang@suse.com>
Cc: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Heinz Mauelshagen