Some systems benefit from completions always being steered to the strict
requester cpu rather than the looser "per-socket" steering that
blk_cpu_to_group() attempts by default. This is because the first
CPU in the group mask ends up being completely overloaded with work,
while the others (including the original submitter) has power left
to spare.

Allow the strict mode to be set by writing '2' to the sysfs control
file. This is identical to the scheme used for the nomerges file,
where '2' is a more aggressive setting than just being turned on.

echo 2 > /sys/block/<bdev>/queue/rq_affinity

Cc: Christoph Hellwig <hch@infradead.org>
Cc: Roland Dreier <roland@purestorage.com>
Tested-by: NDave Jiang <dave.jiang@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Reorder request_queue to remove 16 bytes of alignment padding in 64 bit
builds.

On my config this shrinks the size of this structure from 1608 to 1592
bytes and therefore needs one fewer cachelines.

Also trivially move the open bracket { to be on the same line as the
structure name to make it easier to grep.
Signed-off-by: NRichard Kennedy <richard@rsk.demon.co.uk>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

I'm often confused why not disable preempt when changing blk_plug list. It
would be better to add comments here in case others have the similar concerns.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

When I test fio script with big I/O depth, I found the total throughput drops
compared to some relative small I/O depth. The reason is the thread accumulates
big requests in its plug list and causes some delays (surely this depends
on CPU speed).
I thought we'd better have a threshold for requests. When a threshold reaches,
this means there is no request merge and queue lock contention isn't severe
when pushing per-task requests to queue, so the main advantages of blk plug
don't exist. We can force a plug list flush in this case.
With this, my test throughput actually increases and almost equals to small
I/O depth. Another side effect is irq off time decreases in blk_flush_plug_list()
for big I/O depth.
The BLK_MAX_REQUEST_COUNT is choosen arbitarily, but 16 is efficiently to
reduce lock contention to me. But I'm open here, 32 is ok in my test too.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

There is not a function rq_init but blk_rq_init in block/blk-core.c.
Signed-off-by: NWanlong Gao <wanlong.gao@gmail.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Since those defined functions require additional semicolon
from the caller, they could cause potential syntax errors
when used in if-else statements.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Acked-by: NMartin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

In some cases we would end up stacking discard_zeroes_data incorrectly.
Fix this by enabling the feature by default for stacking drivers and
clearing it for low-level drivers. Incorporating a device that does not
support dzd will then cause the feature to be disabled in the stacking
driver.

Also ensure that the maximum discard value does not overflow when
exported in sysfs and return 0 in the alignment and dzd fields for
devices that don't support discard.
Reported-by: NLukas Czerner <lczerner@redhat.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>
Acked-by: NMike Snitzer <snitzer@redhat.com>
Cc: stable@kernel.org
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

In some drives, flush requests are non-queueable. When flush request is
running, normal read/write requests can't run. If block layer dispatches
such request, driver can't handle it and requeue it.  Tejun suggested we
can hold the queue when flush is running. This can avoid unnecessary
requeue.  Also this can improve performance. For example, we have
request flush1, write1, flush 2. flush1 is dispatched, then queue is
hold, write1 isn't inserted to queue. After flush1 is finished, flush2
will be dispatched. Since disk cache is already clean, flush2 will be
finished very soon, so looks like flush2 is folded to flush1.

In my test, the queue holding completely solves a regression introduced by
commit 53d63e6b:

    block: make the flush insertion use the tail of the dispatch list

    It's not a preempt type request, in fact we have to insert it
    behind requests that do specify INSERT_FRONT.

which causes about 20% regression running a sysbench fileio
workload.

Stable: 2.6.39 only

Cc: stable@kernel.org
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

flush request isn't queueable in some drives. Add a flag to let driver
notify block layer about this. We can optimize flush performance with the
knowledge.

Stable: 2.6.39 only

Cc: stable@kernel.org
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

We are currently using this flag to check whether it's safe
to call into ->request_fn(). If it is set, we punt to kblockd.
But we get a lot of false positives and excessive punts to
kblockd, which hurts performance.

The only real abuser of this infrastructure is SCSI. So export
the async queue run and convert SCSI over to use that. There's
room for improvement in that SCSI need not always use the async
call, but this fixes our performance issue and they can fix that
up in due time.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Instead of overloading __blk_run_queue to force an offload to kblockd
add a new blk_run_queue_async helper to do it explicitly.  I've kept
the blk_queue_stopped check for now, but I suspect it's not needed
as the check we do when the workqueue items runs should be enough.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

MD can't use this since it really requires us to be able to
keep more than a single piece of state for the unplug. Commit
048c9374 added the required support for MD, so get rid of this
now unused code.

This reverts commit f7566457.

Conflicts:

	block/blk-core.c
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

md/raid requires an unplug callback, but as it does not uses
requests the current code cannot provide one.

So allow arbitrary callbacks to be attached to the blk_plug.
Signed-off-by: NNeilBrown <neilb@suse.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Linus correctly observes that the most important dispatch cases
are now done from kblockd, this isn't ideal for latency reasons.
The original reason for switching dispatches out-of-line was to
avoid too deep a stack, so by _only_ letting the "accidental"
flush directly in schedule() be guarded by offload to kblockd,
we should be able to get the best of both worlds.

So add a blk_schedule_flush_plug() that offloads to kblockd,
and only use that from the schedule() path.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

For the explicit unplugging, we'd prefer to kick things off
immediately and not pay the penalty of the latency to switch
to kblockd. So let blk_finish_plug() do the run inline, while
the implicit-on-schedule-out unplug will punt to kblockd.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

It's a bit of a mess currently. task->plug is being cleared
and reset in __blk_finish_plug(), and blk_finish_plug() is
testing for a NULL plug which cannot happen even from schedule()
anymore since it uses blk_needs_flush_plug() to determine
whether to call into this function at all.

So get rid of some of the cruft.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

MD would like to know when a queue is unplugged, so it can flush
it's bitmap writes. Add such a callback.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

The current block integrity (DIF/DIX) support in DM is verifying that
all devices' integrity profiles match during DM device resume (which
is past the point of no return).  To some degree that is unavoidable
(stacked DM devices force this late checking).  But for most DM
devices (which aren't stacking on other DM devices) the ideal time to
verify all integrity profiles match is during table load.

Introduce the notion of an "initialized" integrity profile: a profile
that was blk_integrity_register()'d with a non-NULL 'blk_integrity'
template.  Add blk_integrity_is_initialized() to allow checking if a
profile was initialized.

Update DM integrity support to:
- check all devices with _initialized_ integrity profiles match
  during table load; uninitialized profiles (e.g. for underlying DM
  device(s) of a stacked DM device) are ignored.
- disallow a table load that would result in an integrity profile that
  conflicts with a DM device's existing (in-use) integrity profile
- avoid clearing an existing integrity profile
- validate all integrity profiles match during resume; but if they
  don't all we can do is report the mismatch (during resume we're past
  the point of no return)
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

They used an older prototype, fix it up.
Reported-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Code has been converted over to the new explicit on-stack plugging,
and delay users have been converted to use the new API for that.
So lets kill off the old plugging along with aops->sync_page().
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

This patch adds support for creating a queuing context outside
of the queue itself. This enables us to batch up pieces of IO
before grabbing the block device queue lock and submitting them to
the IO scheduler.

The context is created on the stack of the process and assigned in
the task structure, so that we can auto-unplug it if we hit a schedule
event.

The current queue plugging happens implicitly if IO is submitted to
an empty device, yet callers have to remember to unplug that IO when
they are going to wait for it. This is an ugly API and has caused bugs
in the past. Additionally, it requires hacks in the vm (->sync_page()
callback) to handle that logic. By switching to an explicit plugging
scheme we make the API a lot nicer and can get rid of the ->sync_page()
hack in the vm.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Currently we use plugging for that, but as plugging is going away,
we need an alternative mechanism.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Move blk_throtl_exit() in blk_cleanup_queue() as blk_throtl_exit() is
written in such a way that it needs queue lock. In blk_release_queue()
there is no gurantee that ->queue_lock is still around.

Initially blk_throtl_exit() was in blk_cleanup_queue() but Ingo reported
one problem.

  https://lkml.org/lkml/2010/10/23/86

  And a quick fix moved blk_throtl_exit() to blk_release_queue().

        commit 7ad58c02
        Author: Jens Axboe <jaxboe@fusionio.com>
        Date:   Sat Oct 23 20:40:26 2010 +0200

        block: fix use-after-free bug in blk throttle code

This patch reverts above change and does not try to shutdown the
throtl work in blk_sync_queue(). By avoiding call to
throtl_shutdown_timer_wq() from blk_sync_queue(), we should also avoid
the problem reported by Ingo.

blk_sync_queue() seems to be used only by md driver and it seems to be
using it to make sure q->unplug_fn is not called as md registers its
own unplug functions and it is about to free up the data structures
used by unplug_fn(). Block throttle does not call back into unplug_fn()
or into md. So there is no need to cancel blk throttle work.

In fact I think cancelling block throttle work is bad because it might
happen that some bios are throttled and scheduled to be dispatched later
with the help of pending work and if work is cancelled, these bios might
never be dispatched.

Block layer also uses blk_sync_queue() during blk_cleanup_queue() and
blk_release_queue() time. That should be safe as we are also calling
blk_throtl_exit() which should make sure all the throttling related
data structures are cleaned up.
Signed-off-by: NVivek Goyal <vgoyal@redhat.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

__blk_run_queue() automatically either calls q->request_fn() directly
or schedules kblockd depending on whether the function is recursed.
blk-flush implementation needs to be able to explicitly choose
kblockd.  Add @force_kblockd.

All the current users are converted to specify %false for the
parameter and this patch doesn't introduce any behavior change.

stable: This is prerequisite for fixing ide oops caused by the new
        blk-flush implementation.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Jan Beulich <JBeulich@novell.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: stable@kernel.org
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

o Dominik Klein reported a system hang issue while doing some blkio
  throttling testing.

  https://lkml.org/lkml/2011/2/24/173

o Some tracing revealed that CFQ was not dispatching any more jobs as
  queue unplug was not happening. And queue unplug was not happening
  because unplug work was not being called as there was one throttling
  work on same cpu which as not finished yet. And throttling work had not
  finished as it was tyring to dispatch a bio to CFQ but all the request
  descriptors were consume to it was put to sleep.

o So basically it is a cyclic dependecny between CFQ unplug work and
  throtl dispatch work. Tejun suggested that use separate workqueue for
  such cases.

o This patch uses a separate workqueue for throttle related work and
  does not rely on kblockd workqueue anymore.

Cc: stable@kernel.org
Reported-by: NDominik Klein <dk@in-telegence.net>
Signed-off-by: NVivek Goyal <vgoyal@redhat.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Flush requests are never put on the IO scheduler.  Convert request
structure's elevator_private* into an array and have the flush fields
share a union with it.

Reclaim the space lost in 'struct request' by moving 'completion_data'
back in the union with 'rb_node'.
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Acked-by: NVivek Goyal <vgoyal@redhat.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

The current FLUSH/FUA support has evolved from the implementation
which had to perform queue draining.  As such, sequencing is done
queue-wide one flush request after another.  However, with the
draining requirement gone, there's no reason to keep the queue-wide
sequential approach.

This patch reimplements FLUSH/FUA support such that each FLUSH/FUA
request is sequenced individually.  The actual FLUSH execution is
double buffered and whenever a request wants to execute one for either
PRE or POSTFLUSH, it queues on the pending queue.  Once certain
conditions are met, a flush request is issued and on its completion
all pending requests proceed to the next sequence.

This allows arbitrary merging of different type of flushes.  How they
are merged can be primarily controlled and tuned by adjusting the
above said 'conditions' used to determine when to issue the next
flush.

This is inspired by Darrick's patches to merge multiple zero-data
flushes which helps workloads with highly concurrent fsync requests.

* As flush requests are never put on the IO scheduler, request fields
  used for flush share space with rq->rb_node.  rq->completion_data is
  moved out of the union.  This increases the request size by one
  pointer.

  As rq->elevator_private* are used only by the iosched too, it is
  possible to reduce the request size further.  However, to do that,
  we need to modify request allocation path such that iosched data is
  not allocated for flush requests.

* FLUSH/FUA processing happens on insertion now instead of dispatch.

- Comments updated as per Vivek and Mike.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: "Darrick J. Wong" <djwong@us.ibm.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

/proc/diskstats would display a strange output as follows.

$ cat /proc/diskstats |grep sda
   8       0 sda 90524 7579 102154 20464 0 0 0 0 0 14096 20089
   8       1 sda1 19085 1352 21841 4209 0 0 0 0 4294967064 15689 4293424691
                                                ~~~~~~~~~~
   8       2 sda2 71252 3624 74891 15950 0 0 0 0 232 23995 1562390
   8       3 sda3 54 487 2188 92 0 0 0 0 0 88 92
   8       4 sda4 4 0 8 0 0 0 0 0 0 0 0
   8       5 sda5 81 2027 2130 138 0 0 0 0 0 87 137

Its reason is the wrong way of accounting hd_struct->in_flight. When a bio is
merged into a request belongs to different partition by ELEVATOR_FRONT_MERGE.

The detailed root cause is as follows.

Assuming that there are two partition, sda1 and sda2.

1. A request for sda2 is in request_queue. Hence sda1's hd_struct->in_flight
   is 0 and sda2's one is 1.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

2. A bio belongs to sda1 is issued and is merged into the request mentioned on
   step1 by ELEVATOR_BACK_MERGE. The first sector of the request is changed
   from sda2 region to sda1 region. However the two partition's
   hd_struct->in_flight are not changed.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

3. The request is finished and blk_account_io_done() is called. In this case,
   sda2's hd_struct->in_flight, not a sda1's one, is decremented.

        | hd_struct->in_flight
   ---------------------------
   sda1 |         -1
   sda2 |          1
   ---------------------------

The patch fixes the problem by caching the partition lookup
inside the request structure, hence making sure that the increment
and decrement will always happen on the same partition struct. This
also speeds up IO with accounting enabled, since it cuts down on
the number of lookups we have to do.

Also add a refcount to struct hd_struct to keep the partition in
memory as long as users exist. We use kref_test_and_get() to ensure
we don't add a reference to a partition which is going away.
Signed-off-by: NJerome Marchand <jmarchan@redhat.com>
Signed-off-by: NYasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: stable@kernel.org
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Implement blk_limits_max_hw_sectors() and make
blk_queue_max_hw_sectors() a wrapper around it.

DM needs this to avoid setting queue_limits' max_hw_sectors and
max_sectors directly.  dm_set_device_limits() now leverages
blk_limits_max_hw_sectors() logic to establish the appropriate
max_hw_sectors minimum (PAGE_SIZE).  Fixes issue where DM was
incorrectly setting max_sectors rather than max_hw_sectors (which
caused dm_merge_bvec()'s max_hw_sectors check to be ineffective).
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Cc: stable@kernel.org
Acked-by: NMartin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

When stacking devices, a request_queue is not always available. This
forced us to have a no_cluster flag in the queue_limits that could be
used as a carrier until the request_queue had been set up for a
metadevice.

There were several problems with that approach. First of all it was up
to the stacking device to remember to set queue flag after stacking had
completed. Also, the queue flag and the queue limits had to be kept in
sync at all times. We got that wrong, which could lead to us issuing
commands that went beyond the max scatterlist limit set by the driver.

The proper fix is to avoid having two flags for tracking the same thing.
We deprecate QUEUE_FLAG_CLUSTER and use the queue limit directly in the
block layer merging functions. The queue_limit 'no_cluster' is turned
into 'cluster' to avoid double negatives and to ease stacking.
Clustering defaults to being enabled as before. The queue flag logic is
removed from the stacking function, and explicitly setting the cluster
flag is no longer necessary in DM and MD.
Reported-by: NEd Lin <ed.lin@promise.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>
Acked-by: NMike Snitzer <snitzer@redhat.com>
Cc: stable@kernel.org
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Currently, media presence polling for removeable block devices is done
from userland.  There are several issues with this.

* Polling is done by periodically opening the device.  For SCSI
  devices, the command sequence generated by such action involves a
  few different commands including TEST_UNIT_READY.  This behavior,
  while perfectly legal, is different from Windows which only issues
  single command, GET_EVENT_STATUS_NOTIFICATION.  Unfortunately, some
  ATAPI devices lock up after being periodically queried such command
  sequences.

* There is no reliable and unintrusive way for a userland program to
  tell whether the target device is safe for media presence polling.
  For example, polling for media presence during an on-going burning
  session can make it fail.  The polling program can avoid this by
  opening the device with O_EXCL but then it risks making a valid
  exclusive user of the device fail w/ -EBUSY.

* Userland polling is unnecessarily heavy and in-kernel implementation
  is lighter and better coordinated (workqueue, timer slack).

This patch implements framework for in-kernel disk event handling,
which includes media presence polling.

* bdops->check_events() is added, which supercedes ->media_changed().
  It should check whether there's any pending event and return if so.
  Currently, two events are defined - DISK_EVENT_MEDIA_CHANGE and
  DISK_EVENT_EJECT_REQUEST.  ->check_events() is guaranteed not to be
  called parallelly.

* gendisk->events and ->async_events are added.  These should be
  initialized by block driver before passing the device to add_disk().
  The former contains the mask of all supported events and the latter
  the mask of all events which the device can report without polling.
  /sys/block/*/events[_async] export these to userland.

* Kernel parameter block.events_dfl_poll_msecs controls the system
  polling interval (default is 0 which means disable) and
  /sys/block/*/events_poll_msecs control polling intervals for
  individual devices (default is -1 meaning use system setting).  Note
  that if a device can report all supported events asynchronously and
  its polling interval isn't explicitly set, the device won't be
  polled regardless of the system polling interval.

* If a device is opened exclusively with write access, event checking
  is automatically disabled until all write exclusive accesses are
  released.

* There are event 'clearing' events.  For example, both of currently
  defined events are cleared after the device has been successfully
  opened.  This information is passed to ->check_events() callback
  using @clearing argument as a hint.

* Event checking is always performed from system_nrt_wq and timer
  slack is set to 25% for polling.

* Nothing changes for drivers which implement ->media_changed() but
  not ->check_events().  Going forward, all drivers will be converted
  to ->check_events() and ->media_change() will be dropped.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

There's no reason for register_disk() and del_gendisk() to be in
fs/partitions/check.c.  Move both to genhd.c.  While at it, collapse
unlink_gendisk(), which was artificially in a separate function due to
genhd.c / check.c split, into del_gendisk().
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

REQ_HARDBARRIER is dead now, so remove the leftovers.  What's left
at this point is:

 - various checks inside the block layer.
 - sanity checks in bio based drivers.
 - now unused bio_empty_barrier helper.
 - Xen blockfront use of BLKIF_OP_WRITE_BARRIER - it's dead for a while,
   but Xen really needs to sort out it's barrier situaton.
 - setting of ordered tags in uas - dead code copied from old scsi
   drivers.
 - scsi different retry for barriers - it's dead and should have been
   removed when flushes were converted to FS requests.
 - blktrace handling of barriers - removed.  Someone who knows blktrace
   better should add support for REQ_FLUSH and REQ_FUA, though.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

This is done the same way as helper sb_issue_discard for
blkdev_issue_discard.
Signed-off-by: NLukas Czerner <lczerner@redhat.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

This reverts commit 7681bfee.

Conflicts:

	include/linux/genhd.h

It has numerous issues with the cleanup path and non-elevator
devices. Revert it for now so we can come up with a clean
version without rushing things.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

/proc/diskstats would display a strange output as follows.

$ cat /proc/diskstats |grep sda
   8       0 sda 90524 7579 102154 20464 0 0 0 0 0 14096 20089
   8       1 sda1 19085 1352 21841 4209 0 0 0 0 4294967064 15689 4293424691
                                                ~~~~~~~~~~
   8       2 sda2 71252 3624 74891 15950 0 0 0 0 232 23995 1562390
   8       3 sda3 54 487 2188 92 0 0 0 0 0 88 92
   8       4 sda4 4 0 8 0 0 0 0 0 0 0 0
   8       5 sda5 81 2027 2130 138 0 0 0 0 0 87 137

Its reason is the wrong way of accounting hd_struct->in_flight. When a bio is
merged into a request belongs to different partition by ELEVATOR_FRONT_MERGE.

The detailed root cause is as follows.

Assuming that there are two partition, sda1 and sda2.

1. A request for sda2 is in request_queue. Hence sda1's hd_struct->in_flight
   is 0 and sda2's one is 1.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

2. A bio belongs to sda1 is issued and is merged into the request mentioned on
   step1 by ELEVATOR_BACK_MERGE. The first sector of the request is changed
   from sda2 region to sda1 region. However the two partition's
   hd_struct->in_flight are not changed.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

3. The request is finished and blk_account_io_done() is called. In this case,
   sda2's hd_struct->in_flight, not a sda1's one, is decremented.

        | hd_struct->in_flight
   ---------------------------
   sda1 |         -1
   sda2 |          1
   ---------------------------

The patch fixes the problem by caching the partition lookup
inside the request structure, hence making sure that the increment
and decrement will always happen on the same partition struct. This
also speeds up IO with accounting enabled, since it cuts down on
the number of lookups we have to do.

When reloading partition tables, quiesce IO to ensure that no
request references to the partition struct exists. When it is safe
to free the partition table, the IO for that device is restarted
again.
Signed-off-by: NYasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: stable@kernel.org
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Physical block size was declared unsigned int to accomodate the maximum
size reported by READ CAPACITY(16).  Make sure we use the right type in
the related functions.
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>
Acked-by: NMike Snitzer <snitzer@redhat.com>
Cc: stable@kernel.org
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

All the blkdev_issue_* helpers can only sanely be used for synchronous
caller.  To issue cache flushes or barriers asynchronously the caller needs
to set up a bio by itself with a completion callback to move the asynchronous
state machine ahead.  So drop the BLKDEV_IFL_WAIT flag that is always
specified when calling blkdev_issue_* and also remove the now unused flags
argument to blkdev_issue_flush and blkdev_issue_zeroout.  For
blkdev_issue_discard we need to keep it for the secure discard flag, which
gains a more descriptive name and loses the bitops vs flag confusion.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

o Actual implementation of throttling policy in block layer. Currently it
  implements READ and WRITE bytes per second throttling logic. IOPS throttling
  comes in later patches.
Signed-off-by: NVivek Goyal <vgoyal@redhat.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Change type of 2nd parameter of blk_rq_aligned() into unsigned long
and remove unnecessary casting. Now we can call it with 'uaddr'
instead of 'ubuf' in __blk_rq_map_user() so that it can remove
following warnings from sparse:

 block/blk-map.c:57:31: warning: incorrect type in argument 2 (different address spaces)
 block/blk-map.c:57:31:    expected void *addr
 block/blk-map.c:57:31:    got void [noderef] <asn:1>*ubuf

However blk_rq_map_kern() needs one more local variable to handle it.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>