/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>

This includes both the number of bios merged into requests belonging to this
cgroup as well as the number of requests merged together.
In the past, we've observed different merging behavior across upstream kernels,
some by design some actual bugs. This stat helps a lot in debugging such
problems when applications report decreased throughput with a new kernel
version.

This needed adding an extra elevator function to capture bios being merged as I
did not want to pollute elevator code with blkiocg knowledge and hence needed
the accounting invocation to come from CFQ.

Signed-off-by: Divyesh Shah<dpshah@google.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Till now block layer allowed two separate modes of request execution.
A request is always acquired from the request queue via
elv_next_request().  After that, drivers are free to either dequeue it
or process it without dequeueing.  Dequeue allows elv_next_request()
to return the next request so that multiple requests can be in flight.

Executing requests without dequeueing has its merits mostly in
allowing drivers for simpler devices which can't do sg to deal with
segments only without considering request boundary.  However, the
benefit this brings is dubious and declining while the cost of the API
ambiguity is increasing.  Segment based drivers are usually for very
old or limited devices and as converting to dequeueing model isn't
difficult, it doesn't justify the API overhead it puts on block layer
and its more modern users.

Previous patches converted all block low level drivers to dequeueing
model.  This patch completes the API transition by...

* renaming elv_next_request() to blk_peek_request()

* renaming blkdev_dequeue_request() to blk_start_request()

* adding blk_fetch_request() which is combination of peek and start

* disallowing completion of queued (not started) requests

* applying new API to all LLDs

Renamings are for consistency and to break out of tree code so that
it's apparent that out of tree drivers need updating.

[ Impact: block request issue API cleanup, no functional change ]
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Mike Miller <mike.miller@hp.com>
Cc: unsik Kim <donari75@gmail.com>
Cc: Paul Clements <paul.clements@steeleye.com>
Cc: Tim Waugh <tim@cyberelk.net>
Cc: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Laurent Vivier <Laurent@lvivier.info>
Cc: Jeff Garzik <jgarzik@pobox.com>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: Adrian McMenamin <adrian@mcmen.demon.co.uk>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Borislav Petkov <petkovbb@googlemail.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alex Dubov <oakad@yahoo.com>
Cc: Pierre Ossman <drzeus@drzeus.cx>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Markus Lidel <Markus.Lidel@shadowconnect.com>
Cc: Stefan Weinhuber <wein@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Pete Zaitcev <zaitcev@redhat.com>
Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

struct request has had a few different ways to represent some
properties of a request.  ->hard_* represent block layer's view of the
request progress (completion cursor) and the ones without the prefix
are supposed to represent the issue cursor and allowed to be updated
as necessary by the low level drivers.  The thing is that as block
layer supports partial completion, the two cursors really aren't
necessary and only cause confusion.  In addition, manual management of
request detail from low level drivers is cumbersome and error-prone at
the very least.

Another interesting duplicate fields are rq->[hard_]nr_sectors and
rq->{hard_cur|current}_nr_sectors against rq->data_len and
rq->bio->bi_size.  This is more convoluted than the hard_ case.

rq->[hard_]nr_sectors are initialized for requests with bio but
blk_rq_bytes() uses it only for !pc requests.  rq->data_len is
initialized for all request but blk_rq_bytes() uses it only for pc
requests.  This causes good amount of confusion throughout block layer
and its drivers and determining the request length has been a bit of
black magic which may or may not work depending on circumstances and
what the specific LLD is actually doing.

rq->{hard_cur|current}_nr_sectors represent the number of sectors in
the contiguous data area at the front.  This is mainly used by drivers
which transfers data by walking request segment-by-segment.  This
value always equals rq->bio->bi_size >> 9.  However, data length for
pc requests may not be multiple of 512 bytes and using this field
becomes a bit confusing.

In general, having multiple fields to represent the same property
leads only to confusion and subtle bugs.  With recent block low level
driver cleanups, no driver is accessing or manipulating these
duplicate fields directly.  Drop all the duplicates.  Now rq->sector
means the current sector, rq->data_len the current total length and
rq->bio->bi_size the current segment length.  Everything else is
defined in terms of these three and available only through accessors.

* blk_recalc_rq_sectors() is collapsed into blk_update_request() and
  now handles pc and fs requests equally other than rq->sector update.
  This means that now pc requests can use partial completion too (no
  in-kernel user yet tho).

* bio_cur_sectors() is replaced with bio_cur_bytes() as block layer
  now uses byte count as the primary data length.

* blk_rq_pos() is now guranteed to be always correct.  In-block users
  converted.

* blk_rq_bytes() is now guaranteed to be always valid as is
  blk_rq_sectors().  In-block users converted.

* blk_rq_sectors() is now guaranteed to equal blk_rq_bytes() >> 9.
  More convenient one is used.

* blk_rq_bytes() and blk_rq_cur_bytes() are now inlined and take const
  pointer to request.

[ Impact: API cleanup, single way to represent one property of a request ]
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

This forces in_flight to be zero when turning off or on the I/O stat
accounting and stops updating I/O stats in attempt_merge() when
accounting is turned off.
Signed-off-by: NJerome Marchand <jmarchan@redhat.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Just use struct elevator_queue everywhere instead.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Signed-off-by: NMike Anderson <andmike@linux.vnet.ibm.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

This patch adds support for controlling the IO completion CPU of
either all requests on a queue, or on a per-request basis. We export
a sysfs variable (rq_affinity) which, if set, migrates completions
of requests to the CPU that originally submitted it. A bio helper
(bio_set_completion_cpu()) is also added, so that queuers can ask
for completion on that specific CPU.

In testing, this has been show to cut the system time by as much
as 20-40% on synthetic workloads where CPU affinity is desired.

This requires a little help from the architecture, so it'll only
work as designed for archs that are using the new generic smp
helper infrastructure.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

elv_register() always returns 0, and there isn't anything it does where
it should return an error (the only error condition is so grave that
it's handled with a BUG_ON).
Signed-off-by: NAdrian Bunk <bunk@kernel.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Some of the code has been gradually transitioned to using the proper
struct request_queue, but there's lots left. So do a full sweet of
the kernel and get rid of this typedef and replace its uses with
the proper type.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Currently we allow any merge, even if the io originates from different
processes. This can cause really bad starvation and unfairness, if those
ios happen to be synchronous (reads or direct writes).

So add a allow_merge hook to the io scheduler ops, so an io scheduler can
help decide whether a bio/process combination may be merged with an
existing request.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

- ->init_queue() does not need the elevator passed in
- ->put_request() is a hot path and need not have the queue passed in
- cfq_update_io_seektime() does not need cfqd passed in
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

elevator_type field in elevator_type structure is useless:
it isn't used anywhere in kernel sources.
Signed-off-by: NVasily Tarasov <vtaras@openvz.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Make it possible to disable the block layer.  Not all embedded devices require
it, some can make do with just JFFS2, NFS, ramfs, etc - none of which require
the block layer to be present.

This patch does the following:

 (*) Introduces CONFIG_BLOCK to disable the block layer, buffering and blockdev
     support.

 (*) Adds dependencies on CONFIG_BLOCK to any configuration item that controls
     an item that uses the block layer.  This includes:

     (*) Block I/O tracing.

     (*) Disk partition code.

     (*) All filesystems that are block based, eg: Ext3, ReiserFS, ISOFS.

     (*) The SCSI layer.  As far as I can tell, even SCSI chardevs use the
     	 block layer to do scheduling.  Some drivers that use SCSI facilities -
     	 such as USB storage - end up disabled indirectly from this.

     (*) Various block-based device drivers, such as IDE and the old CDROM
     	 drivers.

     (*) MTD blockdev handling and FTL.

     (*) JFFS - which uses set_bdev_super(), something it could avoid doing by
     	 taking a leaf out of JFFS2's book.

 (*) Makes most of the contents of linux/blkdev.h, linux/buffer_head.h and
     linux/elevator.h contingent on CONFIG_BLOCK being set.  sector_div() is,
     however, still used in places, and so is still available.

 (*) Also made contingent are the contents of linux/mpage.h, linux/genhd.h and
     parts of linux/fs.h.

 (*) Makes a number of files in fs/ contingent on CONFIG_BLOCK.

 (*) Makes mm/bounce.c (bounce buffering) contingent on CONFIG_BLOCK.

 (*) set_page_dirty() doesn't call __set_page_dirty_buffers() if CONFIG_BLOCK
     is not enabled.

 (*) fs/no-block.c is created to hold out-of-line stubs and things that are
     required when CONFIG_BLOCK is not set:

     (*) Default blockdev file operations (to give error ENODEV on opening).

 (*) Makes some /proc changes:

     (*) /proc/devices does not list any blockdevs.

     (*) /proc/diskstats and /proc/partitions are contingent on CONFIG_BLOCK.

 (*) Makes some compat ioctl handling contingent on CONFIG_BLOCK.

 (*) If CONFIG_BLOCK is not defined, makes sys_quotactl() return -ENODEV if
     given command other than Q_SYNC or if a special device is specified.

 (*) In init/do_mounts.c, no reference is made to the blockdev routines if
     CONFIG_BLOCK is not defined.  This does not prohibit NFS roots or JFFS2.

 (*) The bdflush, ioprio_set and ioprio_get syscalls can now be absent (return
     error ENOSYS by way of cond_syscall if so).

 (*) The seclvl_bd_claim() and seclvl_bd_release() security calls do nothing if
     CONFIG_BLOCK is not set, since they can't then happen.
Signed-Off-By: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

None of the in-kernel primitives for handling "atomic" counting seem
to be a good fit. We need something that is essentially free for
incrementing/decrementing, while the read side may be more expensive
as we only ever need to do that when a device is removed from the
kernel.

Use a per-cpu variable for maintaining a per-cpu ioc count and define
a reading mechanism that just sums up the values.
Signed-off-by: NJens Axboe <axboe@suse.de>

It's not needed for anything, so kill the bio passing.
Signed-off-by: NJens Axboe <axboe@suse.de>

The io schedulers can use this instead of having to allocate space for
it themselves.
Signed-off-by: NJens Axboe <axboe@suse.de>

The rbtree sort/lookup/reposition logic is mostly duplicated in
cfq/deadline/as, so move it to the elevator core. The io schedulers
still provide the actual rb root, as we don't want to impose any sort
of specific handling on the schedulers.

Introduce the helpers and rb_node in struct request to help migrate the
IO schedulers.
Signed-off-by: NJens Axboe <axboe@suse.de>

Right now, every IO scheduler implements its own backmerging (except for
noop, which does no merging). That results in duplicated code for
essentially the same operation, which is never a good thing. This patch
moves the backmerging out of the io schedulers and into the elevator
core. We save 1.6kb of text and as a bonus get backmerging for noop as
well. Win-win!
Signed-off-by: NJens Axboe <axboe@suse.de>

There's a race between shutting down one io scheduler and firing up the
next, in which a new io could enter and cause the io scheduler to be
invoked with bad or NULL data.

To fix this, we need to maintain the queue lock for a bit longer.
Unfortunately we cannot do that, since the elevator init requires to be
run without the lock held.  This isn't easily fixable, without also
changing the mempool API.  So split the initialization into two parts,
and alloc-init operation and an attach operation.  Then we can
preallocate the io scheduler and related structures, and run the attach
inside the lock after we detach the old one.

This patch has survived 30 minutes of 1 second io scheduler switching
with a very busy io load.
Signed-off-by: NJens Axboe <axboe@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

q->ordcolor must only be flipped on initial queueing of a hardbarrier
request.

Constructing ordered sequence and requeueing used to pass through
__elv_add_request() which flips q->ordcolor when it sees a barrier
request.

This patch separates out elv_insert() from __elv_add_request() and uses
elv_insert() when constructing ordered sequence and requeueing.
elv_insert() inserts the given request at the specified position and
does nothing else.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Acked-by: NJens Axboe <axboe@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NAdrian Bunk <bunk@stusta.de>

elv_try_last_merge().
Signed-off-by: NCoywolf Qi Hunt <qiyong@fc-cn.com>
Signed-off-by: NJens Axboe <axboe@suse.de>

Reimplement handling of barrier requests.

* Flexible handling to deal with various capabilities of
  target devices.
* Retry support for falling back.
* Tagged queues which don't support ordered tag can do ordered.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJens Axboe <axboe@suse.de>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

- Split elv_dispatch_insert() into two functions
- Rename rq_last_sector() to rq_end_sector()
Signed-off-by: NJens Axboe <axboe@suse.de>

Implements generic dispatch queue which can replace all
dispatch queues implemented by each iosched.  This reduces
code duplication, eases enforcing semantics over dispatch
queue, and simplifies specific ioscheds.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJens Axboe <axboe@suse.de>

This updates the CFQ io scheduler to the new time sliced design (cfq
v3).  It provides full process fairness, while giving excellent
aggregate system throughput even for many competing processes.  It
supports io priorities, either inherited from the cpu nice value or set
directly with the ioprio_get/set syscalls.  The latter closely mimic
set/getpriority.

This import is based on my latest from -mm.
Signed-off-by: NJens Axboe <axboe@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!