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: NBryn M. Reeves <bmr@redhat.com>
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Fixes: e262f347 ("dm stats: add support for request-based DM devices")
Cc: stable@vger.kernel.org # 4.2+

Commit 0a927c2f ("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: NMike Snitzer <snitzer@redhat.com>

Signed-off-by: NMike 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: NJoe Thornber <ejt@redhat.com>
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org

Signed-off-by: NMike Snitzer <snitzer@redhat.com>

Signed-off-by: NMike Snitzer <snitzer@redhat.com>

clone_bio() now checks if bio_integrity_clone() returned an error rather
than just drop it on the floor.
Signed-off-by: NMike 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: 8a01a6af ("dm thin: prefetch missing metadata pages")
Cc: stable@vger.kernel.org # 3.19+
Signed-off-by: NJoe Thornber <ejt@redhat.com>
Signed-off-by: NMike Snitzer <snitzer@redhat.com>

Otherwise loading a "snapshot" table using the same device for the
origin and COW devices, e.g.:

echo "0 20971520 snapshot 253:3 253:3 P 8" | dmsetup create snap

will trigger:

BUG: unable to handle kernel NULL pointer dereference at 0000000000000098
[ 1958.979934] IP: [<ffffffffa040efba>] dm_exception_store_set_chunk_size+0x7a/0x110 [dm_snapshot]
[ 1958.989655] PGD 0
[ 1958.991903] Oops: 0000 [#1] SMP
...
[ 1959.059647] CPU: 9 PID: 3556 Comm: dmsetup Tainted: G          IO    4.5.0-rc5.snitm+ #150
...
[ 1959.083517] task: ffff8800b9660c80 ti: ffff88032a954000 task.ti: ffff88032a954000
[ 1959.091865] RIP: 0010:[<ffffffffa040efba>]  [<ffffffffa040efba>] dm_exception_store_set_chunk_size+0x7a/0x110 [dm_snapshot]
[ 1959.104295] RSP: 0018:ffff88032a957b30  EFLAGS: 00010246
[ 1959.110219] RAX: 0000000000000000 RBX: 0000000000000008 RCX: 0000000000000001
[ 1959.118180] RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffff880329334a00
[ 1959.126141] RBP: ffff88032a957b50 R08: 0000000000000000 R09: 0000000000000001
[ 1959.134102] R10: 000000000000000a R11: f000000000000000 R12: ffff880330884d80
[ 1959.142061] R13: 0000000000000008 R14: ffffc90001c13088 R15: ffff880330884d80
[ 1959.150021] FS:  00007f8926ba3840(0000) GS:ffff880333440000(0000) knlGS:0000000000000000
[ 1959.159047] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1959.165456] CR2: 0000000000000098 CR3: 000000032f48b000 CR4: 00000000000006e0
[ 1959.173415] Stack:
[ 1959.175656]  ffffc90001c13040 ffff880329334a00 ffff880330884ed0 ffff88032a957bdc
[ 1959.183946]  ffff88032a957bb8 ffffffffa040f225 ffff880329334a30 ffff880300000000
[ 1959.192233]  ffffffffa04133e0 ffff880329334b30 0000000830884d58 00000000569c58cf
[ 1959.200521] Call Trace:
[ 1959.203248]  [<ffffffffa040f225>] dm_exception_store_create+0x1d5/0x240 [dm_snapshot]
[ 1959.211986]  [<ffffffffa040d310>] snapshot_ctr+0x140/0x630 [dm_snapshot]
[ 1959.219469]  [<ffffffffa0005c44>] ? dm_split_args+0x64/0x150 [dm_mod]
[ 1959.226656]  [<ffffffffa0005ea7>] dm_table_add_target+0x177/0x440 [dm_mod]
[ 1959.234328]  [<ffffffffa0009203>] table_load+0x143/0x370 [dm_mod]
[ 1959.241129]  [<ffffffffa00090c0>] ? retrieve_status+0x1b0/0x1b0 [dm_mod]
[ 1959.248607]  [<ffffffffa0009e35>] ctl_ioctl+0x255/0x4d0 [dm_mod]
[ 1959.255307]  [<ffffffff813304e2>] ? memzero_explicit+0x12/0x20
[ 1959.261816]  [<ffffffffa000a0c3>] dm_ctl_ioctl+0x13/0x20 [dm_mod]
[ 1959.268615]  [<ffffffff81215eb6>] do_vfs_ioctl+0xa6/0x5c0
[ 1959.274637]  [<ffffffff81120d2f>] ? __audit_syscall_entry+0xaf/0x100
[ 1959.281726]  [<ffffffff81003176>] ? do_audit_syscall_entry+0x66/0x70
[ 1959.288814]  [<ffffffff81216449>] SyS_ioctl+0x79/0x90
[ 1959.294450]  [<ffffffff8167e4ae>] entry_SYSCALL_64_fastpath+0x12/0x71
...
[ 1959.323277] RIP  [<ffffffffa040efba>] dm_exception_store_set_chunk_size+0x7a/0x110 [dm_snapshot]
[ 1959.333090]  RSP <ffff88032a957b30>
[ 1959.336978] CR2: 0000000000000098
[ 1959.344121] ---[ end trace b049991ccad1169e ]---

Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1195899
Cc: stable@vger.kernel.org
Signed-off-by: NDing Xiang <dingxiang@huawei.com>
Signed-off-by: NMike 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: NJoe Thornber <ejt@redhat.com>
Signed-off-by: NMike 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: NBob Liu <bob.liu@oracle.com>
Signed-off-by: NMike 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: NMike 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: NMike 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: NMike Snitzer <snitzer@redhat.com>

Signed-off-by: NMike 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: NMike Snitzer <snitzer@redhat.com>

Signed-off-by: NMike Snitzer <snitzer@redhat.com>

Signed-off-by: NMike Snitzer <snitzer@redhat.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: NMike 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: NMike 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: NMike 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: NMike Snitzer <snitzer@redhat.com>

Signed-off-by: NMike 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: NMike Snitzer <snitzer@redhat.com>

Refactor and rename existing interfaces to be more specific and
self-documenting.
Signed-off-by: NMike 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: NMike 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: NMike Snitzer <snitzer@redhat.com>

Request-based DM will also make use of per_bio_data_size.
Signed-off-by: NMike 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: NMike 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: NMike 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: NMike 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: NMike 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: NMike 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: NMike 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: NMike 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: NMike 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: NMike Snitzer <snitzer@redhat.com>

Signed-off-by: NMike 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 621739b0 !

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: 7fb4898e ("block: add blk-mq support to blk_insert_cloned_request()")
Fixes: bfebd1cd ("dm: add full blk-mq support to request-based DM")
Cc: stable@vger.kernel.org # 4.1+
Reported-by: NSagi Grimberg <sagig@dev.mellanox.co.il>
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Acked-by: NJens 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: NMike Snitzer <snitzer@redhat.com>