We already do a non-blocking allocation in blk_mq_map_request, no need
to repeat it.  Just call __blk_mq_alloc_request to wait directly.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

The current logic for blocking tag allocation is rather confusing, as we
first allocated and then free again a tag in blk_mq_wait_for_tags, just
to attempt a non-blocking allocation and then repeat if someone else
managed to grab the tag before us.

Instead change blk_mq_alloc_request_pinned to simply do a blocking tag
allocation itself and use the request we get back from it.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Both callers if __blk_mq_alloc_request want to initialize the request, so
lift it into the common path.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Instead of having two almost identical copies of the same code just let
the callers pass in the reserved flag directly.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Both the cache flush state machine and the SCSI midlayer want to submit
requests from irq context, and the current per-request requeue_work
unfortunately causes corruption due to sharing with the csd field for
flushes.  Replace them with a per-request_queue list of requests to
be requeued.

Based on an earlier test by Ming Lei.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reported-by: NMing Lei <tom.leiming@gmail.com>
Tested-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Right now we export two ways of completing a request:

1) blk_mq_complete_request(). This uses an IPI (if needed) and
   completes through q->softirq_done_fn(). It also works with
   timeouts.

2) blk_mq_end_io(). This completes inline, and ignores any timeout
   state of the request.

Let blk_mq_complete_request() handle non-softirq_done_fn completions
as well, by just completing inline. If a driver has enough completion
ports to place completions correctly, it need not define a
mq_ops->complete() and we can avoid an indirect function call by
doing the completion inline.
Signed-off-by: NJens Axboe <axboe@fb.com>

Drivers currently have to figure this out on their own, and they
are missing information to do it properly. The ones that did
attempt to do it, do it wrong.

So just pass in the suggested node directly to the alloc
function.
Signed-off-by: NJens Axboe <axboe@fb.com>

The percpu counter is only used for blk-mq, so move
its allocation and free inside blk-mq, and don't
allocate it for legacy queue device.
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

blk_mq_exit_hw_queues() and blk_mq_free_hw_queues()
are introduced to avoid code duplication.
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

hctx->ctx_map should have been freed inside blk_mq_free_queue().
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Without this we can leak the active_queues reference if a command is
freed while it is considered active.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Currently blk-mq uses the queue timeout for all requests. But
for some commands, drivers may want to set a specific timeout
for special requests. Allow this to be passed in through
request->timeout, and use it if set.
Signed-off-by: NJens Axboe <axboe@fb.com>

We want slightly different behavior from them:

- On single queue devices, we currently use the per-process plug
  for deferred IO and for merging.

- On multi queue devices, we don't use the per-process plug, but
  we want to go straight to hardware for SYNC IO.

Split blk_mq_make_request() into a blk_sq_make_request() for single
queue devices, and retain blk_mq_make_request() for multi queue
devices. Then we don't need multiple checks for q->nr_hw_queues
in the request mapping.
Signed-off-by: NJens Axboe <axboe@fb.com>

Depending on the topology of the machine and the number of queues
exposed by a device, we can end up in a situation where some of
the hardware queues are unused (as in, they don't map to any
software queues). For this case, free up the memory used by the
request map, as we will not use it. This can be a substantial
amount of memory, depending on the number of queues vs CPUs and
the queue depth of the device.
Signed-off-by: NJens Axboe <axboe@fb.com>

Prepare this for the next patch which adds more smarts in the
plugging logic, so that we can save some memory.
Signed-off-by: NJens Axboe <axboe@fb.com>

In blk_mq_make_request(), do the blk_queue_nomerges() check
outside the call to blk_attempt_plug_merge() to eliminate
function call overhead when nomerges=2 (disabled)
Signed-off-by: NRobert Elliott <elliott@hp.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

blk_queue_make_requests() overwrites our set value for q->nr_requests,
turning it into the default of 128. Set this appropriately after
initializing queue values in blk_queue_make_request().
Signed-off-by: NJens Axboe <axboe@fb.com>

For request_fn based devices, the block layer exports a 'nr_requests'
file through sysfs to allow adjusting of queue depth on the fly.
Currently this returns -EINVAL for blk-mq, since it's not wired up.
Wire this up for blk-mq, so that it now also always dynamic
adjustments of the allowed queue depth for any given block device
managed by blk-mq.
Signed-off-by: NJens Axboe <axboe@fb.com>

Each hardware queue has a bitmap of software queues with pending
requests. When new IO is queued on a software queue, the bit is
set, and when IO is pruned on a hardware queue run, the bit is
cleared. This causes a lot of traffic. Switch this from the regular
BITS_PER_LONG bitmap to a sparser layout, similarly to what was
done for blk-mq tagging.

20% performance increase was observed for single threaded IO, and
about 15% performanc increase on multiple threads driving the
same device.
Signed-off-by: NJens Axboe <axboe@fb.com>

This adds support for active queue tracking, meaning that the
blk-mq tagging maintains a count of active users of a tag set.
This allows us to maintain a notion of fairness between users,
so that we can distribute the tag depth evenly without starving
some users while allowing others to try unfair deep queues.

If sharing of a tag set is detected, each hardware queue will
track the depth of its own queue. And if this exceeds the total
depth divided by the number of active queues, the user is actively
throttled down.

The active queue count is done lazily to avoid bouncing that data
between submitter and completer. Each hardware queue gets marked
active when it allocates its first tag, and gets marked inactive
when 1) the last tag is cleared, and 2) the queue timeout grace
period has passed.
Signed-off-by: NJens Axboe <axboe@fb.com>

Commit c6d600c6 opened up a small race where we could attempt to
account IO completion on a request, racing with IO start accounting.
Fix this up by ensuring that we've accounted for IO start before
inserting the request.
Signed-off-by: NJens Axboe <axboe@fb.com>

blk-mq currently uses percpu_ida for tag allocation. But that only
works well if the ratio between tag space and number of CPUs is
sufficiently high. For most devices and systems, that is not the
case. The end result if that we either only utilize the tag space
partially, or we end up attempting to fully exhaust it and run
into lots of lock contention with stealing between CPUs. This is
not optimal.

This new tagging scheme is a hybrid bitmap allocator. It uses
two tricks to both be SMP friendly and allow full exhaustion
of the space:

1) We cache the last allocated (or freed) tag on a per blk-mq
   software context basis. This allows us to limit the space
   we have to search. The key element here is not caching it
   in the shared tag structure, otherwise we end up dirtying
   more shared cache lines on each allocate/free operation.

2) The tag space is split into cache line sized groups, and
   each context will start off randomly in that space. Even up
   to full utilization of the space, this divides the tag users
   efficiently into cache line groups, avoiding dirtying the same
   one both between allocators and between allocator and freeer.

This scheme shows drastically better behaviour, both on small
tag spaces but on large ones as well. It has been tested extensively
to show better performance for all the cases blk-mq cares about.
Signed-off-by: NJens Axboe <axboe@fb.com>

This allows us to avoid a non-atomic memset over ->atomic_flags as well
as killing lots of duplicate initializations.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Signed-off-by: NJens Axboe <axboe@fb.com>

Right now we just pick the first CPU in the mask, but that can
easily overload that one. Add some basic batching and round-robin
all the entries in the mask instead.
Signed-off-by: NJens Axboe <axboe@fb.com>

We already issue a blktrace requeue event in
__blk_mq_requeue_request(), don't do it from the original caller
as well.
Signed-off-by: NJens Axboe <axboe@fb.com>

Refactor the logic around adding a new bio to a software queue,
so we nest the ctx->lock where we really need it (merge and
insertion) and don't hold it when we don't (init and IO start
accounting).
Signed-off-by: NJens Axboe <axboe@fb.com>

It's never been of any use, lets get rid of it.
Signed-off-by: NJens Axboe <axboe@fb.com>

blk_mq_wait_for_tags() is only able to wait for "normal" tags,
not reserved tags. Pass in which one we should attempt to get
a tag for, so that waiting for reserved tags will work.

Reserved tags are used for internal commands, which are usually
serialized. Hence no waiting generally takes place, but we should
ensure that it actually works if users need that functionality.
Signed-off-by: NJens Axboe <axboe@fb.com>

The blk-mq code is using it's own version of the I/O completion affinity
tunables, which causes a few issues:

 - the rq_affinity sysfs file doesn't work for blk-mq devices, even if it
   still is present, thus breaking existing tuning setups.
 - the rq_affinity = 1 mode, which is the defauly for legacy request based
   drivers isn't implemented at all.
 - blk-mq drivers don't implement any completion affinity with the default
   flag settings.

This patches removes the blk-mq ipi_redirect flag and sysfs file, as well
as the internal BLK_MQ_F_SHOULD_IPI flag and replaces it with code that
respects the queue-wide rq_affinity flags and also implements the
rq_affinity = 1 mode.

This means I/O completion affinity can now only be tuned block-queue wide
instead of per context, which seems more sensible to me anyway.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

If a requeue event races with a timeout, we can get into the
situation where we attempt to complete a request from the
timeout handler when it's not start anymore. This causes a crash.
So have the timeout handler check that REQ_ATOM_STARTED is still
set on the request - if not, we ignore the event. If this happens,
the request has now been marked as complete. As a consequence, we
need to ensure to clear REQ_ATOM_COMPLETE in blk_mq_start_request(),
as to maintain proper request state.
Signed-off-by: NJens Axboe <axboe@fb.com>

This reverts commit 6a3c8a3a.

We need selective clearing of the request to make the init-at-free
time completely safe. Otherwise we end up stomping on
rq->atomic_flags, which we don't want to do.

set->tags should be freed in blk_mq_free_tag_set().
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

The patch basically reverts the patch of(blk-mq:
initialize request on allocation) in Jens's tree(already
in -next), and only initialize req->q in allocation
for two reasons:

	- presumed cache hotness on completion
	- blk_rq_tagged(rq) depends on reset of req->mq_ctx
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Cc: Jörg-Volker Peetz <jvpeetz@web.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

type of set->tags is struct blk_mq_tags **.
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Avoid memory leak in the failure path.
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

This allows to requeue a request that has been accepted by ->queue_rq
earlier.  This is needed by the SCSI layer in various error conditions.

The existing internal blk_mq_requeue_request is renamed to
__blk_mq_requeue_request as it is a lower level building block for this
funtionality.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Add a helper to unconditionally kick contexts of a queue.  This will
be needed by the SCSI layer to provide fair queueing between multiple
devices on a single host.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Add a blk-mq equivalent to blk_delay_queue so that the scsi layer can ask
to be kicked again after a delay.
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Modified by me to kill the unnecessary preempt disable/enable
in the delayed workqueue handler.
Signed-off-by: NJens Axboe <axboe@fb.com>