blk-crypto: update inline encryption documentation

Rework most of inline-encryption.rst to be easier to follow, to correct
some information, to add some important details and remove some
unimportant details, and to take into account the renaming from
blk_keyslot_manager to blk_crypto_profile.

Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20211018180453.40441-5-ebiggers@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-crypto: rename blk_keyslot_manager to blk_crypto_profile

blk_keyslot_manager is misnamed because it doesn't necessarily manage
keyslots.  It actually does several different things:

  - Contains the crypto capabilities of the device.

  - Provides functions to control the inline encryption hardware.
    Originally these were just for programming/evicting keyslots;
    however, new functionality (hardware-wrapped keys) will require new
    functions here which are unrelated to keyslots.  Moreover,
    device-mapper devices already (ab)use "keyslot_evict" to pass key
    eviction requests to their underlying devices even though
    device-mapper devices don't have any keyslots themselves (so it
    really should be "evict_key", not "keyslot_evict").

  - Sometimes (but not always!) it manages keyslots.  Originally it
    always did, but device-mapper devices don't have keyslots
    themselves, so they use a "passthrough keyslot manager" which
    doesn't actually manage keyslots.  This hack works, but the
    terminology is unnatural.  Also, some hardware doesn't have keyslots
    and thus also uses a "passthrough keyslot manager" (support for such
    hardware is yet to be upstreamed, but it will happen eventually).

Let's stop having keyslot managers which don't actually manage keyslots.
Instead, rename blk_keyslot_manager to blk_crypto_profile.

This is a fairly big change, since for consistency it also has to update
keyslot manager-related function names, variable names, and comments --
not just the actual struct name.  However it's still a fairly
straightforward change, as it doesn't change any actual functionality.

Acked-by: Ulf Hansson <ulf.hansson@linaro.org> # For MMC
Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20211018180453.40441-4-ebiggers@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-crypto: rename keyslot-manager files to blk-crypto-profile

In preparation for renaming struct blk_keyslot_manager to struct
blk_crypto_profile, rename the keyslot-manager.h and keyslot-manager.c
source files.  Renaming these files separately before making a lot of
changes to their contents makes it easier for git to understand that
they were renamed.

Acked-by: Ulf Hansson <ulf.hansson@linaro.org> # For MMC
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20211018180453.40441-3-ebiggers@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-crypto-fallback: properly prefix function and struct names

For clarity, avoid using just the "blk_crypto_" prefix for functions and
structs that are specific to blk-crypto-fallback.  Instead, use
"blk_crypto_fallback_".  Some places already did this, but others
didn't.

This is also a prerequisite for using "struct blk_crypto_keyslot" to
mean a generic blk-crypto keyslot (which is what it sounds like).
Rename the fallback one to "struct blk_crypto_fallback_keyslot".

No change in behavior.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20211018180453.40441-2-ebiggers@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nbd: Use invalidate_disk() helper on disconnect

When a nbd device encounters a writeback error, that error will
get propagated to the bd_inode's wb_err field. Then if this nbd
device's backend is disconnected and another is attached, we will
get back the previous writeback error on fsync, which is unexpected.

To fix it, let's use invalidate_disk() helper to invalidate the
disk on disconnect instead of just setting disk's capacity to zero.

Signed-off-by: Xie Yongji <xieyongji@bytedance.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210922123711.187-5-xieyongji@bytedance.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

loop: Remove the unnecessary bdev checks and unused bdev variable

The lo->lo_device can't be null if the lo->lo_backing_file is set.
So let's remove the unnecessary bdev checks and the entire bdev
variable in __loop_clr_fd() since the lo->lo_backing_file is already
checked before.

Signed-off-by: Xie Yongji <xieyongji@bytedance.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210922123711.187-4-xieyongji@bytedance.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

loop: Use invalidate_disk() helper to invalidate gendisk

Use invalidate_disk() helper to simplify the code for gendisk
invalidation.

Signed-off-by: Xie Yongji <xieyongji@bytedance.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210922123711.187-3-xieyongji@bytedance.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: Add invalidate_disk() helper to invalidate the gendisk

To hide internal implementation and simplify some driver code,
this adds a helper to invalidate the gendisk. It will clean the
gendisk's associated buffer/page caches and reset its internal
states.

Signed-off-by: Xie Yongji <xieyongji@bytedance.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210922123711.187-2-xieyongji@bytedance.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: kill extra rcu lock/unlock in queue enter

blk_try_enter_queue() already takes rcu_read_lock/unlock, so we can
avoid the second pair in percpu_ref_tryget_live(), use a newly added
percpu_ref_tryget_live_rcu().

As rcu_read_lock/unlock imply barrier()s, it's pretty noticeable,
especially for for !CONFIG_PREEMPT_RCU (default for some distributions),
where __rcu_read_lock/unlock() are not inlined.

3.20%  io_uring  [kernel.vmlinux]  [k] __rcu_read_unlock
3.05%  io_uring  [kernel.vmlinux]  [k] __rcu_read_lock

2.52%  io_uring  [kernel.vmlinux]  [k] __rcu_read_unlock
2.28%  io_uring  [kernel.vmlinux]  [k] __rcu_read_lock

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/6b11c67ea495ed9d44f067622d852de4a510ce65.1634822969.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

percpu_ref: percpu_ref_tryget_live() version holding RCU

Add percpu_ref_tryget_live_rcu(), which is a version of
percpu_ref_tryget_live() but the user is responsible for enclosing it in
a RCU read lock section.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Acked-by: Dennis Zhou <dennis@kernel.org>
Link: https://lore.kernel.org/r/3066500d7a6eb3e03f10adf98b87fdb3b1c49db8.1634822969.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: convert fops.c magic constants to SHIFT_SECTOR

Don't use shifting by a magic number 9 but replace with a more
descriptive SHIFT_SECTOR.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/068782b9f7e97569fb59a99529b23bb17ea4c5e2.1634755800.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: clean up blk_mq_submit_bio() merging

Combine blk_mq_sched_bio_merge() and blk_attempt_plug_merge() under a
common if, so we don't check it twice.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/daedc90d4029a5d1d73344771632b1faca3aaf81.1634755800.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: optimise boundary blkdev_read_iter's checks

Combine pos and len checks and mark unlikely. Also, don't reexpand if
it's not truncated.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/fff34e613aeaae1ad12977dc4592cb1a1f5d3190.1634755800.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

fs: bdev: fix conflicting comment from lookup_bdev

We switched to directly use dev_t to get block device, lookup changed the
meaning of use, now we fix this conflicting comment.

Fixes: 4e7b5671c6a8 ("block: remove i_bdev")
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Jackie Liu <liuyun01@kylinos.cn>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211021071344.1600362-1-liu.yun@linux.dev
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: Fix blk_mq_tagset_busy_iter() for shared tags

Since it is now possible for a tagset to share a single set of tags, the
iter function should not re-iter the tags for the count of #hw queues in
that case. Rather it should just iter once.

Fixes: e155b0c238b2 ("blk-mq: Use shared tags for shared sbitmap support")
Reported-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Kashyap Desai <kashyap.desai@broadcom.com>
Link: https://lore.kernel.org/r/1634550083-202815-1-git-send-email-john.garry@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: cleanup the flush plug helpers

Consolidate the various helpers into a single blk_flush_plug helper that
takes a plk_plug and the from_scheduler bool and switch all callsites to
call it directly.  Checks that the plug is non-NULL must be performed by
the caller, something that most already do anyway.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211020144119.142582-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: optimise blk_flush_plug_list

Don't call flush_plug_callbacks if there are no plug callbacks.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
[hch: split from a larger patch]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211020144119.142582-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: move blk_mq_flush_plug_list to block/blk-mq.h

This helper is internal to the block layer.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211020144119.142582-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: only flush requests from the plug in blk_mq_submit_bio

Replace the call to blk_flush_plug_list in blk_mq_submit_bio with a
direct call to blk_mq_flush_plug_list.  This means we do not flush
plug callback from stackable devices, which doesn't really help with
the accumulated requests anyway, and it also means the cached requests
aren't freed here as they can still be used later on.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211020144119.142582-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove inaccurate requeue check

This check is meant to catch cases where a requeue is attempted on a
request that is still inserted. It's never really been useful to catch any
misuse, and now it's actively wrong. Outside of that, this should not be a
BUG_ON() to begin with.

Remove the check as it's now causing active harm, as requeue off the plug
path will trigger it even though the request state is just fine.

Reported-by: Yi Zhang <yi.zhang@redhat.com>
Link: https://lore.kernel.org/linux-block/CAHj4cs80zAUc2grnCZ015-2Rvd-=gXRfB_dFKy=RTm+wRo09HQ@mail.gmail.com/
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: inline a part of bio_release_pages()

Inline BIO_NO_PAGE_REF check of bio_release_pages() to avoid function
call.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't bloat enter_queue with percpu_ref

percpu_ref_put() are inlined for performance and bloat the binary, we
don't care about the fail case of blk_try_enter_queue(), so we can
replace it with a call to blk_queue_exit().

Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: optimise req_bio_endio()

First, get rid of an extra branch and chain error checks. Also reshuffle
it with bio_advance(), so it goes closer to the final check, with that
the compiler loads rq->rq_flags only once, and also doesn't reload
bio->bi_iter.bi_size if bio_advance() didn't actually advanced the iter.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: convert leftovers to bdev_get_queue

Convert bdev->bd_disk->queue to bdev_get_queue(), which is faster.
Apparently, there are a few such spots in block that got lost during
rebases.

Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: turn macro helpers into inline functions

Replace bio_set_dev() with an identical inline helper and move it
further to fix a dependency problem with bio_associate_blkg(). Do the
same for bio_copy_dev().

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: support concurrent queue quiesce/unquiesce

blk_mq_quiesce_queue() has been used a bit wide now, so far we don't support
concurrent/nested quiesce. One biggest issue is that unquiesce can happen
unexpectedly in case that quiesce/unquiesce are run concurrently from
more than one context.

This patch introduces q->mq_quiesce_depth to deal concurrent quiesce,
and we only unquiesce queue when it is the last/outer-most one of all
contexts.

Several kernel panic issue has been reported[1][2][3] when running stress
quiesce test. And this patch has been verified in these reports.

[1] https://lore.kernel.org/linux-block/9b21c797-e505-3821-4f5b-df7bf9380328@huawei.com/T/#m1fc52431fad7f33b1ffc3f12c4450e4238540787
[2] https://lore.kernel.org/linux-block/9b21c797-e505-3821-4f5b-df7bf9380328@huawei.com/T/#m10ad90afeb9c8cc318334190a7c24c8b5c5e0722
[3] https://listman.redhat.com/archives/dm-devel/2021-September/msg00189.html

Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211014081710.1871747-7-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme: loop: clear NVME_CTRL_ADMIN_Q_STOPPED after admin queue is reallocated

The nvme-loop's admin queue may be freed and reallocated, and we have to
reset the flag of NVME_CTRL_ADMIN_Q_STOPPED so that the flag can match
with the quiesce state of the admin queue.

nvme-loop is the only driver to reallocate request queue, and not see
such usage in other nvme drivers.

Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211014081710.1871747-6-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme: paring quiesce/unquiesce

The current blk_mq_quiesce_queue() and blk_mq_unquiesce_queue() always
stops and starts the queue unconditionally. And there can be concurrent
quiesce/unquiesce coming from different unrelated code paths, so
unquiesce may come unexpectedly and start queue too early.

Prepare for supporting concurrent quiesce/unquiesce from multiple
contexts, so that we can address the above issue.

NVMe has very complicated quiesce/unquiesce use pattern, add one atomic
bit for makeiing sure that blk-mq quiece/unquiesce is always called in
pair.

Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211014081710.1871747-5-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme: prepare for pairing quiescing and unquiescing

Add two helpers so that we can prepare for pairing quiescing and
unquiescing which will be done in next patch.

Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211014081710.1871747-4-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme: apply nvme API to quiesce/unquiesce admin queue

Apply the added two APIs to quiesce/unquiesce admin queue.

Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211014081710.1871747-3-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme: add APIs for stopping/starting admin queue

Add two APIs for stopping and starting admin queue.

Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211014081710.1871747-2-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block, bfq: fix UAF problem in bfqg_stats_init()

In bfq_pd_alloc(), the function bfqg_stats_init() init bfqg. If
blkg_rwstat_init() init bfqg_stats->bytes successful and init
bfqg_stats->ios failed, bfqg_stats_init() return failed, bfqg will
be freed. But blkg_rwstat->cpu_cnt is not deleted from the list of
percpu_counters. If we traverse the list of percpu_counters, It will
have UAF problem.

we should use blkg_rwstat_exit() to cleanup bfqg_stats bytes in the
above scenario.

Fixes: commit fd41e60331b ("bfq-iosched: stop using blkg->stat_bytes and ->stat_ios")
Signed-off-by: Zheng Liang <zhengliang6@huawei.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20211018024225.1493938-1-zhengliang6@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: inline fast path of driver tag allocation

If we don't use an IO scheduler or have shared tags, then we don't need
to call into this external function at all. This saves ~2% for such
a setup.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: don't handle non-flush requests in blk_insert_flush

Return to the normal blk_mq_submit_bio flow if the bio did not end up
actually being a flush because the device didn't support it.  Note that
this is basically impossible to hit without special instrumentation given
that submit_bio_checks already clears these flags usually, so we'd need a
tight race to actually hit this code path.

With this the call to blk_mq_run_hw_queue for the flush requests can be
removed given that the actual flush requests are always issued via the
requeue workqueue which runs the queue unconditionally.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211019122553.2467817-1-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: attempt direct issue of plug list

If we have just one queue type in the plug list, then we can extend our
direct issue to cover a full plug list as well. This allows sending a
batch of requests for direct issue, which is more efficient than doing
one-at-a-time kind of issue.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: change plugging to use a singly linked list

Use a singly linked list for the blk_plug. This saves 8 bytes in the
blk_plug struct, and makes for faster list manipulations than doubly
linked lists. As we don't use the doubly linked lists for anything,
singly linked is just fine.

This yields a bump in default (merging enabled) performance from 7.0
to 7.1M IOPS, and ~7.5M IOPS with merging disabled.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-wbt: prevent NULL pointer dereference in wb_timer_fn

The timer callback used to evaluate if the latency is exceeded can be
executed after the corresponding disk has been released, causing the
following NULL pointer dereference:

[ 119.987108] BUG: kernel NULL pointer dereference, address: 0000000000000098
[ 119.987617] #PF: supervisor read access in kernel mode
[ 119.987971] #PF: error_code(0x0000) - not-present page
[ 119.988325] PGD 7c4a4067 P4D 7c4a4067 PUD 7bf63067 PMD 0
[ 119.988697] Oops: 0000 [#1] SMP NOPTI
[ 119.988959] CPU: 1 PID: 9353 Comm: cloud-init Not tainted 5.15-rc5+arighi #rc5+arighi
[ 119.989520] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
[ 119.990055] RIP: 0010:wb_timer_fn+0x44/0x3c0
[ 119.990376] Code: 41 8b 9c 24 98 00 00 00 41 8b 94 24 b8 00 00 00 41 8b 84 24 d8 00 00 00 4d 8b 74 24 28 01 d3 01 c3 49 8b 44 24 60 48 8b 40 78 <4c> 8b b8 98 00 00 00 4d 85 f6 0f 84 c4 00 00 00 49 83 7c 24 30 00
[ 119.991578] RSP: 0000:ffffb5f580957da8 EFLAGS: 00010246
[ 119.991937] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000004
[ 119.992412] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88f476d7f780
[ 119.992895] RBP: ffffb5f580957dd0 R08: 0000000000000000 R09: 0000000000000000
[ 119.993371] R10: 0000000000000004 R11: 0000000000000002 R12: ffff88f476c84500
[ 119.993847] R13: ffff88f4434390c0 R14: 0000000000000000 R15: ffff88f4bdc98c00
[ 119.994323] FS: 00007fb90bcd9c00(0000) GS:ffff88f4bdc80000(0000) knlGS:0000000000000000
[ 119.994952] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 119.995380] CR2: 0000000000000098 CR3: 000000007c0d6000 CR4: 00000000000006e0
[ 119.995906] Call Trace:
[ 119.996130] ? blk_stat_free_callback_rcu+0x30/0x30
[ 119.996505] blk_stat_timer_fn+0x138/0x140
[ 119.996830] call_timer_fn+0x2b/0x100
[ 119.997136] __run_timers.part.0+0x1d1/0x240
[ 119.997470] ? kvm_clock_get_cycles+0x11/0x20
[ 119.997826] ? ktime_get+0x3e/0xa0
[ 119.998110] ? native_apic_msr_write+0x2c/0x30
[ 119.998456] ? lapic_next_event+0x20/0x30
[ 119.998779] ? clockevents_program_event+0x94/0xf0
[ 119.999150] run_timer_softirq+0x2a/0x50
[ 119.999465] __do_softirq+0xcb/0x26f
[ 119.999764] irq_exit_rcu+0x8c/0xb0
[ 120.000057] sysvec_apic_timer_interrupt+0x43/0x90
[ 120.000429] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 120.000836] asm_sysvec_apic_timer_interrupt+0x12/0x20

In this case simply return from the timer callback (no action
required) to prevent the NULL pointer dereference.

BugLink: https://bugs.launchpad.net/bugs/1947557
Link: https://lore.kernel.org/linux-mm/YWRNVTk9N8K0RMst@arighi-desktop/
Fixes: 34dbad5d26e2 ("blk-stat: convert to callback-based statistics reporting")
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
Link: https://lore.kernel.org/r/YW6N2qXpBU3oc50q@arighi-desktop
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: align blkdev_dio inlined bio to a cacheline

We get all sorts of unreliable and funky results since the bio is
designed to align on a cacheline, which it does not when inlined like
this.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_mq_tag_to_rq() inline

This is in the fast path of driver issue or completion, and it's a single
array index operation. Move it inline to avoid a function call for it.

This does mean making struct blk_mq_tags block layer public, but there's
not really much in there.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: get rid of plug list sorting

Even if we have multiple queues in the plug list, chances that they
are very interspersed is minimal. Don't bother spending CPU cycles
sorting the list.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: return whether or not to unplug through boolean

Instead of returning the same queue request through a request pointer,
use a boolean to accomplish the same.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't call blk_status_to_errno in blk_update_request

We only need to call it to resolve the blk_status_t -> errno mapping for
tracing, so move the conversion into the tracepoints that are not called
at all when tracing isn't enabled.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move bdev_read_only() into the header

This is called for every write in the fast path, move it inline next
to get_disk_ro() which is called internally.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: fix too broad elevator check in blk_mq_free_request()

We added RQF_ELV to tell whether there's an IO scheduler attached, and
RQF_ELVPRIV tells us whether there's an IO scheduler with private data
attached. Don't check RQF_ELV in blk_mq_free_request(), what we care
about here is just if we have scheduler private data attached.

This fixes a boot crash

Fixes: 2ff0682da6e0 ("block: store elevator state in request")
Reported-by: Yi Zhang <yi.zhang@redhat.com>
Reported-by: syzbot+eb8104072aeab6cc1195@syzkaller.appspotmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme: wire up completion batching for the IRQ path

Trivial to do now, just need our own io_comp_batch on the stack and pass
that in to the usual command completion handling.

I pondered making this dependent on how many entries we had to process,
but even for a single entry there's no discernable difference in
performance or latency. Running a sync workload over io_uring:

t/io_uring -b512 -d1 -s1 -c1 -p0 -F1 -B1 -n2 /dev/nvme1n1 /dev/nvme2n1

yields the below performance before the patch:

IOPS=254820, BW=124MiB/s, IOS/call=1/1, inflight=(1 1)
IOPS=251174, BW=122MiB/s, IOS/call=1/1, inflight=(1 1)
IOPS=250806, BW=122MiB/s, IOS/call=1/1, inflight=(1 1)

and the following after:

IOPS=255972, BW=124MiB/s, IOS/call=1/1, inflight=(1 1)
IOPS=251920, BW=123MiB/s, IOS/call=1/1, inflight=(1 1)
IOPS=251794, BW=122MiB/s, IOS/call=1/1, inflight=(1 1)

which definitely isn't slower, about the same if you factor in a bit of
variance. For peak performance workloads, benchmarking shows a 2%
improvement.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_uring: utilize the io batching infrastructure for more efficient polled IO

Wire up using an io_comp_batch for f_op->iopoll(). If the lower stack
supports it, we can handle high rates of polled IO more efficiently.

This raises the single core efficiency on my system from ~6.1M IOPS to
~6.6M IOPS running a random read workload at depth 128 on two gen2
Optane drives.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme: add support for batched completion of polled IO

Take advantage of struct io_comp_batch, if passed in to the nvme poll
handler. If it's set, rather than complete each request individually
inline, store them in the io_comp_batch list. We only do so for requests
that will complete successfully, anything else will be completed inline as
before.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: add support for blk_mq_end_request_batch()

Instead of calling blk_mq_end_request() on a single request, add a helper
that takes the new struct io_comp_batch and completes any request stored
in there.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

sbitmap: add helper to clear a batch of tags

sbitmap currently only supports clearing tags one-by-one, add a helper
that allows the caller to pass in an array of tags to clear.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: add a struct io_comp_batch argument to fops->iopoll()

struct io_comp_batch contains a list head and a completion handler, which
will allow completions to more effciently completed batches of IO.

For now, no functional changes in this patch, we just define the
io_comp_batch structure and add the argument to the file_operations iopoll
handler.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: provide helpers for rq_list manipulation

Instead of open-coding the list additions, traversal, and removal,
provide a basic set of helpers.

Suggested-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove some blk_mq_hw_ctx debugfs entries

Just like the blk_mq_ctx counterparts, we've got a bunch of counters
in here that are only for debugfs and are of questionnable value. They
are:

- dispatched, index of how many requests were dispatched in one go

- poll_{considered,invoked,success}, which track poll sucess rates. We're
  confident in the iopoll implementation at this point, don't bother
  tracking these.

As a bonus, this shrinks each hardware queue from 576 bytes to 512 bytes,
dropping a whole cacheline.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove debugfs blk_mq_ctx dispatched/merged/completed attributes

These were added as part of early days debugging for blk-mq, and they
are not really useful anymore. Rather than spend cycles updating them,
just get rid of them.

As a bonus, this shrinks the per-cpu software queue size from 256b
to 192b. That's a whole cacheline less.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: cache rq_flags inside blk_mq_rq_ctx_init()

Add a local variable for rq_flags, it helps to compile out some of
rq_flags reloads.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: blk_mq_rq_ctx_init cache ctx/q/hctx

We should have enough of registers in blk_mq_rq_ctx_init(), store them
in local vars, so we don't keep reloading them.

note: keeping q->elevator may look unnecessary, but it's also used
inside inlined blk_mq_tags_from_data().

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: skip elevator fields init for non-elv queue

Don't init rq->hash and rq->rb_node in blk_mq_rq_ctx_init() if there is
no elevator. Also, move some other initialisers that imply barriers to
the end, so the compiler is free to rearrange and optimise other the
rest of them.

note: fold in a change from Jens leaving queue_list unconditional, as
it might lead to problems otherwise.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: store elevator state in request

Add an rq private RQF_ELV flag, which tells the block layer that this
request was initialized on a queue that has an IO scheduler attached.
This allows for faster checking in the fast path, rather than having to
deference rq->q later on.

Elevator switching does full quiesce of the queue before detaching an
IO scheduler, so it's safe to cache this in the request itself.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: only mark bio as tracked if it really is tracked

We set BIO_TRACKED unconditionally when rq_qos_throttle() is called, even
though we may not even have an rq_qos handler. Only mark it as TRACKED if
it really is potentially tracked.

This saves considerable time for the case where the bio isn't tracked:

     2.64%     -1.65%  [kernel.vmlinux]  [k] bio_endio

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: improve layout of struct request

It's been a while since this was analyzed, move some members around to
better flow with the use case. Initial state up top, and queued state
after that. This improves my peak case by about 1.5%, from 7750K to
7900K IOPS.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move update request helpers into blk-mq.c

For some reason we still have them in blk-core, with the rest of the
request completion being in blk-mq. That causes and out-of-line call
for each completion.

Move them into blk-mq.c instead, where they belong.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove useless caller argument to print_req_error()

We have exactly one caller of this, just get rid of adding the useless
function name to the output.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't bother iter advancing a fully done bio

If we're completing nbytes and nbytes is the size of the bio, don't bother
with calling into the iterator increment helpers. Just clear the bio
size and we're done.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: convert the rest of block to bdev_get_queue

Convert bdev->bd_disk->queue to bdev_get_queue(), it's uses a cached
queue pointer and so is faster.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/addf6ea988c04213697ba3684c853e4ed7642a39.1634219547.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: use bdev_get_queue() in blk-core.c

Convert bdev->bd_disk->queue to bdev_get_queue(), it's uses a cached
queue pointer and so is faster.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/efc41f880262517c8dc32f932f1b23112f21b255.1634219547.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: use bdev_get_queue() in bio.c

Convert bdev->bd_disk->queue to bdev_get_queue(), it's uses a cached
queue pointer and so is faster.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/85c36ea784d285a5075baa10049e6b59e15fb484.1634219547.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: use bdev_get_queue() in bdev.c

Convert bdev->bd_disk->queue to bdev_get_queue(), it's uses a cached
queue pointer and so is faster.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/a352936ce5d9ac719645b1e29b173d931ebcdc02.1634219547.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: cache request queue in bdev

There are tons of places where we need to get a request_queue only
having bdev, which turns into bdev->bd_disk->queue. There are probably a
hundred of such places considering inline helpers, and enough of them
are in hot paths.

Cache queue pointer in struct block_device and make use of it in
bdev_get_queue().

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/a3bfaecdd28956f03629d0ca5c63ebc096e1c809.1634219547.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: handle fast path of bio splitting inline

The fast path is no splitting needed. Separate the handling into a
check part we can inline, and an out-of-line handling path if we do
need to split.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: use flags instead of bit fields for blkdev_dio

This generates a lot better code for me, and bumps performance from
7650K IOPS to 7750K IOPS. Looking at profiles for the run and running
perf diff, it confirms that we're now sending a lot less time there:

     6.38%     -2.80%  [kernel.vmlinux]  [k] blkdev_direct_IO

Taking it from the 2nd most cycle consumer to only the 9th most at
3.35% of the CPU time.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: cache bdev in struct file for raw bdev IO

bdev = &BDEV_I(file->f_mapping->host)->bdev

Getting struct block_device from a file requires 2 memory dereferences
as illustrated above, that takes a toll on performance, so cache it in
yet unused file->private_data. That gives a noticeable peak performance
improvement.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/8415f9fe12e544b9da89593dfbca8de2b52efe03.1634115360.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

nvme-multipath: enable polled I/O

Set the poll queue flag to enable polling, given that the multipath
node just dispatches the bios to a lower queue.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-17-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't allow writing to the poll queue attribute

The poll attribute is a historic artefact from before when we had
explicit poll queues that require driver specific configuration.
Just print a warning when writing to the attribute.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-16-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: switch polling to be bio based

Replace the blk_poll interface that requires the caller to keep a queue
and cookie from the submissions with polling based on the bio.

Polling for the bio itself leads to a few advantages:

 - the cookie construction can made entirely private in blk-mq.c
 - the caller does not need to remember the request_queue and cookie
   separately and thus sidesteps their lifetime issues
 - keeping the device and the cookie inside the bio allows to trivially
   support polling BIOs remapping by stacking drivers
 - a lot of code to propagate the cookie back up the submission path can
   be removed entirely.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-15-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: define 'struct bvec_iter' as packed

'struct bvec_iter' is embedded into 'struct bio', define it as packed
so that we can get one extra 4bytes for other uses without expanding
bio.

'struct bvec_iter' is often allocated on stack, so making it packed
doesn't affect performance. Also I have run io_uring on both
nvme/null_blk, and not observe performance effect in this way.

Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-14-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: use SLAB_TYPESAFE_BY_RCU for the bio slab

This flags ensures that the pages will not be reused for non-bio
allocations before the end of an RCU grace period.  With that we can
safely use a RCU lookup for bio polling as long as we are fine with
occasionally polling the wrong device.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-13-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: rename REQ_HIPRI to REQ_POLLED

Unlike the RWF_HIPRI userspace ABI which is intentionally kept vague,
the bio flag is specific to the polling implementation, so rename and
document it properly.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-12-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_uring: don't sleep when polling for I/O

There is no point in sleeping for the expected I/O completion timeout
in the io_uring async polling model as we never poll for a specific
I/O.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-11-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: replace the spin argument to blk_iopoll with a flags argument

Switch the boolean spin argument to blk_poll to passing a set of flags
instead.  This will allow to control polling behavior in a more fine
grained way.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-10-hch@lst.de
[axboe: adapt to changed io_uring iopoll]
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: remove blk_qc_t_valid

Move the trivial check into the only caller.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: remove blk_qc_t_to_tag and blk_qc_t_is_internal

Merge both functions into their only caller to keep the blk-mq tag to
blk_qc_t mapping as private as possible in blk-mq.c.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: factor out a "classic" poll helper

Factor the code to do the classic full metal polling out of blk_poll into
a separate blk_mq_poll_classic helper.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: factor out a blk_qc_to_hctx helper

Add a helper to get the hctx from a request_queue and cookie, and fold
the blk_qc_t_to_queue_num helper into it as no other callers are left.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_uring: fix a layering violation in io_iopoll_req_issued

syscall-level code can't just poke into the details of the poll cookie,
which is private information of the block layer.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012111226.760968-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

iomap: don't try to poll multi-bio I/Os in __iomap_dio_rw

If an iocb is split into multiple bios we can't poll for both.  So don't
bother to even try to poll in that case.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't try to poll multi-bio I/Os in __blkdev_direct_IO

If an iocb is split into multiple bios we can't poll for both.  So don't
even bother to try to poll in that case.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012111226.760968-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

direct-io: remove blk_poll support

The polling support in the legacy direct-io support is a little crufty.
It already doesn't support the asynchronous polling needed for io_uring
polling, and is hard to adopt to upcoming changes in the polling
interfaces.  Given that all the major file systems already use the iomap
direct I/O code, just drop the polling support.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: only check previous entry for plug merge attempt

Currently we scan the entire plug list, which is potentially very
expensive. In an IOPS bound workload, we can drive about 5.6M IOPS with
merging enabled, and profiling shows that the plug merge check is the
(by far) most expensive thing we're doing:

  Overhead  Command   Shared Object     Symbol
  +   20.89%  io_uring  [kernel.vmlinux]  [k] blk_attempt_plug_merge
  +    4.98%  io_uring  [kernel.vmlinux]  [k] io_submit_sqes
  +    4.78%  io_uring  [kernel.vmlinux]  [k] blkdev_direct_IO
  +    4.61%  io_uring  [kernel.vmlinux]  [k] blk_mq_submit_bio

Instead of browsing the whole list, just check the previously inserted
entry. That is enough for a naive merge check and will catch most cases,
and for devices that need full merging, the IO scheduler attached to
such devices will do that anyway. The plug merge is meant to be an
inexpensive check to avoid getting a request, but if we repeatedly
scan the list for every single insert, it is very much not a cheap
check.

With this patch, the workload instead runs at ~7.0M IOPS, providing
a 25% improvement. Disabling merging entirely yields another 5%
improvement.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move CONFIG_BLOCK guard to top Makefile

Every object under block/ depends on CONFIG_BLOCK.

Move the guard to the top Makefile since there is no point to
descend into block/ if CONFIG_BLOCK=n.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210927140000.866249-5-masahiroy@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move menu "Partition type" to block/partitions/Kconfig

Move the menu to the relevant place.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210927140000.866249-4-masahiroy@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: simplify Kconfig files

Everything under block/ depends on BLOCK. BLOCK_HOLDER_DEPRECATED is
selected from drivers/md/Kconfig, which is entirely dependent on BLOCK.

Extend the 'if BLOCK' ... 'endif' so it covers the whole block/Kconfig.

Also, clean up the definition of BLOCK_COMPAT and BLK_MQ_PCI because
COMPAT and PCI are boolean.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210927140000.866249-3-masahiroy@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove redundant =y from BLK_CGROUP dependency

CONFIG_BLK_CGROUP is a boolean option, that is, its value is 'y' or 'n'.
The comparison to 'y' is redundant.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210927140000.866249-2-masahiroy@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: improve batched tag allocation

Add a blk_mq_get_tags() helper, which uses the new sbitmap API for
allocating a batch of tags all at once. This both simplifies the block
code for batched allocation, and it is also more efficient than just
doing repeated calls into __sbitmap_queue_get().

This reduces the sbitmap overhead in peak runs from ~3% to ~1% and
yields a performanc increase from 6.6M IOPS to 6.8M IOPS for a single
CPU core.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

sbitmap: add __sbitmap_queue_get_batch()

The block layer tag allocation batching still calls into sbitmap to get
each tag, but we can improve on that. Add __sbitmap_queue_get_batch(),
which returns a mask of tags all at once, along with an offset for
those tags.

An example return would be 0xff, where bits 0..7 are set, with
tag_offset == 128. The valid tags in this case would be 128..135.

A batch is specific to an individual sbitmap_map, hence it cannot be
larger than that. The requested number of tags is automatically reduced
to the max that can be satisfied with a single map.

On failure, 0 is returned. Caller should fall back to single tag
allocation at that point/

Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: optimise *end_request non-stat path

We already have a blk_mq_need_time_stamp() check in
__blk_mq_end_request() to get a timestamp, hide all the statistics
accounting under it. It cuts some cycles for requests that don't need
stats, and is free otherwise.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/e0f2ea812e93a8adcd07101212e7d7e70ca304e7.1634115360.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: mark bio_truncate static

bio_truncate is only used in bio.c, so mark it static.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012161804.991559-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move bio_get_{first,last}_bvec out of bio.h

bio_get_first_bvec and bio_get_last_bvec are only used in blk-merge.c,
so move them there.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012161804.991559-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: mark __bio_try_merge_page static

Mark __bio_try_merge_page static and move it up a bit to avoid the need
for a forward declaration.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012161804.991559-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move bio_full out of bio.h

bio_full is only used in bio.c, so move it there.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012161804.991559-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: fold bio_cur_bytes into blk_rq_cur_bytes

Fold bio_cur_bytes into the only caller.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012161804.991559-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move bio_mergeable out of bio.h

bio_mergeable is only needed by I/O schedulers, so move it to
blk-mq-sched.h.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012161804.991559-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>