block: null_blk: only set set->nr_maps as 3 if active poll_queues is > 0

It isn't correct to set set->nr_maps as 3 if g_poll_queues is > 0 since
we can change it via configfs for null_blk device created there, so only
set it as 3 if active poll_queues is > 0.

Fixes divide zero exception reported by Shinichiro.

Fixes: 2bfdbe8b7ebd ("null_blk: allow zero poll queues")
Reported-by: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Link: https://lore.kernel.org/r/20211224010831.1521805-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove the rsxx driver

This driver was for rare and shortlived high end enterprise hardware
and hasn't been maintained since 2014, which also means it never got
converted to use blk-mq.

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

rsxx: Drop PCI legacy power management

The rsxx driver doesn't support device suspend, so remove
rsxx_pci_suspend(), the legacy PCI .suspend() method, completely.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20211208192449.146076-5-helgaas@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

mtip32xx: convert to generic power management

Convert mtip32xx from legacy PCI power management to the generic power
management framework.

Previously, mtip32xx used legacy PCI power management, where
mtip_pci_suspend() and mtip_pci_resume() were responsible for both
device-specific things and generic PCI things:

  mtip_pci_suspend
    mtip_block_suspend(dd)              <-- device-specific
    pci_save_state(pdev)                <-- generic PCI
    pci_set_power_state(pdev, pci_choose_state(pdev, state))

  mtip_pci_resume
    pci_set_power_state(PCI_D0)         <-- generic PCI
    pci_restore_state(pdev)             <-- generic PCI
    pcim_enable_device(pdev)            <-- generic PCI
    pci_set_master(pdev)                <-- generic PCI
    mtip_block_resume(dd)               <-- device-specific

With generic power management, the PCI bus PM methods do the generic PCI
things, and the driver needs only the device-specific part, i.e.,

  suspend_devices_and_enter
    dpm_suspend_start(PMSG_SUSPEND)
      pci_pm_suspend                    # PCI bus .suspend() method
        mtip_pci_suspend                # dev->driver->pm->suspend
          mtip_block_suspend            <-- device-specific
    suspend_enter
      dpm_suspend_noirq(PMSG_SUSPEND)
        pci_pm_suspend_noirq            # PCI bus .suspend_noirq() method
          pci_save_state                <-- generic PCI
          pci_prepare_to_sleep          <-- generic PCI
            pci_set_power_state
    ...
    dpm_resume_end(PMSG_RESUME)
      pci_pm_resume                     # PCI bus .resume() method
        pci_restore_standard_config
          pci_set_power_state(PCI_D0)   <-- generic PCI
          pci_restore_state             <-- generic PCI
        mtip_pci_resume                 # dev->driver->pm->resume
          mtip_block_resume             <-- device-specific

[bhelgaas: commit log]

Link: https://lore.kernel.org/r/20210114115423.52414-2-vaibhavgupta40@gmail.com
Signed-off-by: Vaibhav Gupta <vaibhavgupta40@gmail.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20211208192449.146076-4-helgaas@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

mtip32xx: remove pointless drvdata lookups

Previously we passed a struct pci_dev * to mtip_check_surprise_removal(),
which immediately looked up the driver_data.  But all callers already have
the driver_data pointer, so just pass it directly and skip the extra
lookup.  No functional change intended.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20211208192449.146076-3-helgaas@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

mtip32xx: remove pointless drvdata checking

The .suspend() and .resume() methods are only called after the .probe()
method (mtip_pci_probe()) has set the drvdata and returned success.

Therefore, if we get to mtip_pci_suspend() or mtip_pci_resume(), the
drvdata must be valid.  Drop the unnecessary checking.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20211208192449.146076-2-helgaas@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

drbd: Use struct_group() to zero algs

In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memset(), avoid intentionally writing across
neighboring fields.

Add a struct_group() for the algs so that memset() can correctly reason
about the size.

Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20211118203712.1288866-1-keescook@chromium.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

loop: make autoclear operation asynchronous

syzbot is reporting circular locking problem at __loop_clr_fd() [1], for
commit 87579e9b7d8dc36e ("loop: use worker per cgroup instead of kworker")
is calling destroy_workqueue() with disk->open_mutex held.

This circular dependency cannot be broken unless we call __loop_clr_fd()
without holding disk->open_mutex. Therefore, defer __loop_clr_fd() from
lo_release() to a WQ context.

Link: https://syzkaller.appspot.com/bug?extid=643e4ce4b6ad1347d372
Reported-by: syzbot <syzbot+643e4ce4b6ad1347d372@syzkaller.appspotmail.com>
Suggested-by: Christoph Hellwig <hch@infradead.org>
Cc: Jan Kara <jack@suse.cz>
Tested-by: syzbot+643e4ce4b6ad1347d372@syzkaller.appspotmail.com
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/1ed7df28-ebd6-71fb-70e5-1c2972e05ddb@i-love.sakura.ne.jp
Signed-off-by: Jens Axboe <axboe@kernel.dk>

null_blk: cast command status to integer

kernel test robot reports that sparse now triggers a warning on null_blk:

>> drivers/block/null_blk/main.c:1577:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected int ioerror @@     got restricted blk_status_t [usertype] error @@
   drivers/block/null_blk/main.c:1577:55: sparse:     expected int ioerror
   drivers/block/null_blk/main.c:1577:55: sparse:     got restricted blk_status_t [usertype] error

because blk_mq_add_to_batch() takes an integer instead of a blk_status_t.
Just cast this to an integer to silence it, null_blk is the odd one out
here since the command status is the "right" type. If we change the
function type, then we'll have do that for other callers too (existing and
future ones).

Fixes: 2385ebf38f94 ("block: null_blk: batched complete poll requests")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

pktdvd: stop using bdi congestion framework.

The bdi congestion framework isn't widely used and should be
deprecated.

pktdvd makes use of it to track congestion, but this can be done
entirely internally to pktdvd, so it doesn't need to use the framework.

So introduce a "congested" flag.  When waiting for bio_queue_size to
drop, set this flag and a var_waitqueue() to wait for it.  When
bio_queue_size does drop and this flag is set, clear the flag and call
wake_up_var().

We don't use a wait_var_event macro for the waiting as we need to set
the flag and drop the spinlock before calling schedule() and while that
is possible with __wait_var_event(), result is not easy to read.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: NeilBrown <neilb@suse.de>
Link: https://lore.kernel.org/r/163910843527.9928.857338663717630212@noble.neil.brown.name
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: null_blk: batched complete poll requests

Complete poll requests via blk_mq_add_to_batch() and
blk_mq_end_request_batch(), so that we can cover batched complete
code path by running null_blk test.

Meantime this way shows ~14% IOPS boost on 't/io_uring /dev/nullb0'
in my test.

Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20211203081703.3506020-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

floppy: Add max size check for user space request

We need to check the max request size that is from user space before
allocating pages. If the request size exceeds the limit, return -EINVAL.
This check can avoid the warning below from page allocator.

WARNING: CPU: 3 PID: 16525 at mm/page_alloc.c:5344 current_gfp_context include/linux/sched/mm.h:195 [inline]
WARNING: CPU: 3 PID: 16525 at mm/page_alloc.c:5344 __alloc_pages+0x45d/0x500 mm/page_alloc.c:5356
Modules linked in:
CPU: 3 PID: 16525 Comm: syz-executor.3 Not tainted 5.15.0-syzkaller #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
RIP: 0010:__alloc_pages+0x45d/0x500 mm/page_alloc.c:5344
Code: be c9 00 00 00 48 c7 c7 20 4a 97 89 c6 05 62 32 a7 0b 01 e8 74 9a 42 07 e9 6a ff ff ff 0f 0b e9 a0 fd ff ff 40 80 e5 3f eb 88 <0f> 0b e9 18 ff ff ff 4c 89 ef 44 89 e6 45 31 ed e8 1e 76 ff ff e9
RSP: 0018:ffffc90023b87850 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 1ffff92004770f0b RCX: dffffc0000000000
RDX: 0000000000000000 RSI: 0000000000000033 RDI: 0000000000010cc1
RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000001
R10: ffffffff81bb4686 R11: 0000000000000001 R12: ffffffff902c1960
R13: 0000000000000033 R14: 0000000000000000 R15: ffff88804cf64a30
FS:  0000000000000000(0000) GS:ffff88802cd00000(0063) knlGS:00000000f44b4b40
CS:  0010 DS: 002b ES: 002b CR0: 0000000080050033
CR2: 000000002c921000 CR3: 000000004f507000 CR4: 0000000000150ee0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 alloc_pages+0x1a7/0x300 mm/mempolicy.c:2191
 __get_free_pages+0x8/0x40 mm/page_alloc.c:5418
 raw_cmd_copyin drivers/block/floppy.c:3113 [inline]
 raw_cmd_ioctl drivers/block/floppy.c:3160 [inline]
 fd_locked_ioctl+0x12e5/0x2820 drivers/block/floppy.c:3528
 fd_ioctl drivers/block/floppy.c:3555 [inline]
 fd_compat_ioctl+0x891/0x1b60 drivers/block/floppy.c:3869
 compat_blkdev_ioctl+0x3b8/0x810 block/ioctl.c:662
 __do_compat_sys_ioctl+0x1c7/0x290 fs/ioctl.c:972
 do_syscall_32_irqs_on arch/x86/entry/common.c:112 [inline]
 __do_fast_syscall_32+0x65/0xf0 arch/x86/entry/common.c:178
 do_fast_syscall_32+0x2f/0x70 arch/x86/entry/common.c:203
 entry_SYSENTER_compat_after_hwframe+0x4d/0x5c

Reported-by: syzbot+23a02c7df2cf2bc93fa2@syzkaller.appspotmail.com
Link: https://lore.kernel.org/r/20211116131033.27685-1-sxwjean@me.com
Signed-off-by: Xiongwei Song <sxwjean@gmail.com>
Signed-off-by: Denis Efremov <efremov@linux.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

floppy: Fix hang in watchdog when disk is ejected

When the watchdog detects a disk change, it calls cancel_activity(),
which in turn tries to cancel the fd_timer delayed work.

In the above scenario, fd_timer_fn is set to fd_watchdog(), meaning
it is trying to cancel its own work.
This results in a hang as cancel_delayed_work_sync() is waiting for the
watchdog (itself) to return, which never happens.

This can be reproduced relatively consistently by attempting to read a
broken floppy, and ejecting it while IO is being attempted and retried.

To resolve this, this patch calls cancel_delayed_work() instead, which
cancels the work without waiting for the watchdog to return and finish.

Before this regression was introduced, the code in this section used
del_timer(), and not del_timer_sync() to delete the watchdog timer.

Link: https://lore.kernel.org/r/399e486c-6540-db27-76aa-7a271b061f76@tasossah.com
Fixes: 070ad7e793dc ("floppy: convert to delayed work and single-thread wq")
Signed-off-by: Tasos Sahanidis <tasos@tasossah.com>
Signed-off-by: Denis Efremov <efremov@linux.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

null_blk: allow zero poll queues

There isn't any reason to not allow zero poll queues from user
viewpoint.

Also sometimes we need to compare io poll between poll mode and irq
mode, so not allowing poll queues is bad.

Fixes: 15dfc662ef31 ("null_blk: Fix handling of submit_queues and poll_queues attributes")
Cc: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20211203023935.3424042-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

loop: don't hold lo_mutex during __loop_clr_fd()

syzbot is reporting circular locking problem at __loop_clr_fd() [1], for
commit 87579e9b7d8dc36e ("loop: use worker per cgroup instead of kworker")
is calling destroy_workqueue() with lo->lo_mutex held.

Since all functions where lo->lo_state matters are already checking
lo->lo_state with lo->lo_mutex held (in order to avoid racing with e.g.
ioctl(LOOP_CTL_REMOVE)), and __loop_clr_fd() can be called from either
ioctl(LOOP_CLR_FD) xor close(), lo->lo_state == Lo_rundown is considered
as an exclusive lock for __loop_clr_fd(). Therefore, hold lo->lo_mutex
inside __loop_clr_fd() only when asserting/updating lo->lo_state.

Since ioctl(LOOP_CLR_FD) depends on lo->lo_state == Lo_bound, a valid
lo->lo_backing_file must have been assigned by ioctl(LOOP_SET_FD) or
ioctl(LOOP_CONFIGURE). Thus, we can remove lo->lo_backing_file test,
and convert __loop_clr_fd() into a void function.

Link: https://syzkaller.appspot.com/bug?extid=63614029dfb79abd4383
Reported-by: syzbot <syzbot+63614029dfb79abd4383@syzkaller.appspotmail.com>
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/8ebe3b2e-8975-7f26-0620-7144a3b8b8cd@i-love.sakura.ne.jp
Signed-off-by: Jens Axboe <axboe@kernel.dk>

scsi: remove the gendisk argument to scsi_ioctl

Now that blk_execute_rq does not take a gendisk argument there is no need
to pass it through the scsi_ioctl callchain either.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove the gendisk argument to blk_execute_rq

Remove the gendisk aregument to blk_execute_rq and blk_execute_rq_nowait
given that it is unused now.  Also convert the boolean at_head parameter
to actually use the bool type while touching the prototype.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove the ->rq_disk field in struct request

Just use the disk attached to the request_queue instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't check ->rq_disk in merges

There is a 1:1 relationship between request_queues and gendisks now, so
no need for these extra checks.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

mtd_blkdevs: remove the sector out of range check in do_blktrans_request

The block layer already performs this check, no need to duplicate it in
the driver.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: Remove redundant initialization of variable ret

The variable ret is being initialized with a value that is never
read, it is being updated later on. The assignment is redundant and
can be removed.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20211126230652.1175636-1-colin.i.king@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: simplify ioc_lookup_icq

Remove the ioc argument as it always points to current->io_context.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-15-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: simplify ioc_create_icq

Remove the ioc and gfp_mask argument, which are hard coded by the caller.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-14-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: return the io_context from create_task_io_context

Grab a reference to the newly allocated or existing io_context in
create_task_io_context and return it.  This simplifies the callers and
removes the need for double lookups.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-13-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: use alloc_io_context in __copy_io

In __copy_io we know that the newly allocate task_struct does not have
an I/O context yet and is not exiting.  So just allocate the I/O context
struct and install it directly.  There is no need to lock the task
either as it is just being created.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-12-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: factor out a alloc_io_context helper

Factor out a helper that just allocate an I/O context.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-11-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove get_io_context_active

Fold it into it's only caller, and remove a lof of the debug checks
that are not needed.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-10-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move the remaining elv.icq handling to the I/O scheduler

After the prepare side has been moved to the only I/O scheduler that
cares, do the same for the cleanup and the NULL initialization.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_mq_sched_assign_ioc to blk-ioc.c

Move blk_mq_sched_assign_ioc so that many interfaces from the file can
be marked static.  Rename the function to ioc_find_get_icq as well and
return the icq to simplify the interface.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: mark put_io_context_active static

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Revert "block: Provide blk_mq_sched_get_icq()"

This reverts commit 4896c4e64ba5d5d5acdbcf68c5910dd4f6d8fa62.

The helper is not needed any more.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: use bfq_bic_lookup in bfq_limit_depth

No need to create a new I/O context if there is none present yet in
->limit_depth.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: simplify bfq_bic_lookup

Remove the unused bfqd argument, and hardcode ioc to current->io_context.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

fork: move copy_io to block/blk-ioc.c

Move the copying of the I/O context to the block layer as that is where
we can use the proper low-level interfaces.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

RDMA/qib: rename copy_io to qib_copy_io

Add the proper module prefix to avoid conflicts with a function
in the scheduler.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: use bio->bi_opf after bio is checked

bio->bi_opf isn't finalized before checking the bio, so use it after
submit_bio_checks() returns.

Fixes: 5b13bc8a3fd5 ("blk-mq: cleanup request allocation")
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: Do not let waker requests skip proper accounting

Commit 7cc4ffc55564 ("block, bfq: put reqs of waker and woken in
dispatch list") added a condition to bfq_insert_request() which added
waker's requests directly to dispatch list. The rationale was that
completing waker's IO is needed to get more IO for the current queue.
Although this rationale is valid, there is a hole in it. The waker does
not necessarily serve the IO only for the current queue and maybe it's
current IO is not needed for current queue to make progress. Furthermore
injecting IO like this completely bypasses any service accounting within
bfq and thus we do not properly track how much service is waker's queue
getting or that the waker is actually doing any IO. Depending on the
conditions this can result in the waker getting too much or too few
service.

Consider for example the following job file:

[global]
directory=/mnt/repro/
rw=write
size=8g
time_based
runtime=30
ramp_time=10
blocksize=1m
direct=0
ioengine=sync

[slowwriter]
numjobs=1
prioclass=2
prio=7
fsync=200

[fastwriter]
numjobs=1
prioclass=2
prio=0
fsync=200

Despite processes have very different IO priorities, they get the same
about of service. The reason is that bfq identifies these processes as
having waker-wakee relationship and once that happens, IO from
fastwriter gets injected during slowwriter's time slice. As a result bfq
is not aware that fastwriter has any IO to do and constantly schedules
only slowwriter's queue. Thus fastwriter is forced to compete with
slowwriter's IO all the time instead of getting its share of time based
on IO priority.

Drop the special injection condition from bfq_insert_request(). As a
result, requests will be tracked and queued in a normal way and on next
dispatch bfq_select_queue() can decide whether the waker's inserted
requests should be injected during the current queue's timeslice or not.

Fixes: 7cc4ffc55564 ("block, bfq: put reqs of waker and woken in dispatch list")
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-8-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: Log waker detections

Waker - wakee relationships are important in deciding whether one queue
can preempt the other one. Print information about detected waker-wakee
relationships so that scheduling decisions can be better understood from
block traces.

Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-7-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: Provide helper to generate bfqq name

Instead of having helper formating bfqq pid, provide a helper to
generate full bfqq name as used in the traces. It saves some code
duplication and will save more in the coming tracepoints.

Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-6-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: Limit waker detection in time

Currently, when process A starts issuing requests shortly after process
B has completed some IO three times in a row, we decide that B is a
"waker" of A meaning that completing IO of B is needed for A to make
progress and generally stop separating A's and B's IO much. This logic
is useful to avoid unnecessary idling and thus throughput loss for cases
where workload needs to switch e.g. between the process and the
journaling thread doing IO. However the detection heuristic tends to
frequently give false positives when A and B are fighting IO bandwidth
and other processes aren't doing much IO as we are basically deemed to
eventually accumulate three occurences of a situation where one process
starts issuing requests after the other has completed some IO. To reduce
these false positives, cancel the waker detection also if we didn't
accumulate three detected wakeups within given timeout. The rationale is
that if wakeups are really rare, the pointless idling doesn't hurt
throughput that much anyway.

This significantly reduces false waker detection for workload like:

[global]
directory=/mnt/repro/
rw=write
size=8g
time_based
runtime=30
ramp_time=10
blocksize=1m
direct=0
ioengine=sync

[slowwriter]
numjobs=1
fsync=200

[fastwriter]
numjobs=1
fsync=200

Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-5-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: Limit number of requests consumed by each cgroup

When cgroup IO scheduling is used with BFQ it does not really provide
service differentiation if the cgroup drives a big IO depth. That for
example happens with writeback which asynchronously submits lots of IO
but it can happen with AIO as well. The problem is that if we have two
cgroups that submit IO with different weights, the cgroup with higher
weight properly gets more IO time and is able to dispatch more IO.
However this causes lower weight cgroup to accumulate more requests
inside BFQ and eventually lower weight cgroup consumes most of IO
scheduler tags. At that point higher weight cgroup stops getting better
service as it is mostly blocked waiting for a scheduler tag while its
queues inside BFQ are empty and thus lower weight cgroup gets served.

Check how many requests submitting cgroup has allocated in
bfq_limit_depth() and if it consumes more requests than what would
correspond to its weight limit available depth to 1 so that the cgroup
cannot consume many more requests. With this limitation the higher
weight cgroup gets proper service even with writeback.

Reviewed-by: Michal Koutný <mkoutny@suse.com>
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-4-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: Store full bitmap depth in bfq_data

Store bitmap depth shift inside bfq_data so that we can use it in
bfq_limit_depth() for proportioning when limiting number of available
request tags for a cgroup.

Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-3-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

bfq: Track number of allocated requests in bfq_entity

When we want to limit number of requests used by each bfqq and also
cgroup, we need to track also number of requests used by each cgroup.
So track number of allocated requests for each bfq_entity.

Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-2-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: Provide blk_mq_sched_get_icq()

Currently we lookup ICQ only after the request is allocated. However BFQ
will want to decide how many scheduler tags it allows a given bfq queue
(effectively a process) to consume based on cgroup weight. So provide a
function blk_mq_sched_get_icq() so that BFQ can lookup ICQ earlier.

Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-1-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>

mmc: core: Use blk_mq_complete_request_direct().

The completion callback for the sdhci-pci device is invoked from a
kworker.
I couldn't identify in which context is mmc_blk_mq_req_done() invoke but
the remaining caller are from invoked from preemptible context. Here it
would make sense to complete the request directly instead scheduling
ksoftirqd for its completion.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/20211025070658.1565848-3-bigeasy@linutronix.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: Add blk_mq_complete_request_direct()

Add blk_mq_complete_request_direct() which completes the block request
directly instead deferring it to softirq for single queue devices.

This is useful for devices which complete the requests in preemptible
context and raising softirq from means scheduling ksoftirqd.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211025070658.1565848-2-bigeasy@linutronix.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-crypto: remove blk_crypto_unregister()

This function is trivial and is only used in one place.  Having this
function is misleading because it implies that blk_crypto_register()
needs to be paired with blk_crypto_unregister(), which is not the case.
Just set disk->queue->crypto_profile to NULL directly.

Signed-off-by: Eric Biggers <ebiggers@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211124013733.347612-1-ebiggers@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: cleanup request allocation

Refactor the request alloction so that blk_mq_get_cached_request tries
to find a cached request first, and the entirely separate and now
self contained blk_mq_get_new_requests allocates one or more requests
if that is not possible.

There is a small change in behavior as submit_bio_checks is called
twice now if a cached request is present but can't be used, but that
is a small price to pay for unwinding this code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211124062856.1444266-1-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't include <linux/part_stat.h> in blk.h

Not needed, shift it into the source files that need it instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't include <linux/idr.h> in blk.h

Not needed.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't include <linux/blk-mq.h> in blk.h

Not needed.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't include blk-mq.h in blk.h

No needed, shift a blk-stat.h include into the source file that needs it
instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't include blk-mq-sched.h in blk.h

No needed, shift it into the source files that need it instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove the e argument to elevator_exit

All callers pass q->elevator.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove elevator_exit

Open code elevator_exit in it's only caller, and rename __elevator_exit to
elevator_exit.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_get_flush_queue to blk-flush.c

blk_get_flush_queue is only used in blk-flush.c, so move it there.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk_mq: remove repeated includes

Remove a repeated "#include<linux/sched/sysctl.h>".

Signed-off-by: Guo Zhengkui <guozhengkui@vivo.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20211123063340.25882-1-guozhengkui@vivo.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move io_context creation into where it's needed

The only user of the io_context for IO is BFQ, yet we put the checking
and logic of it into the normal IO path.

Put the creation into blk_mq_sched_assign_ioc(), and have BFQ use that
helper.

Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: only allocate poll_stats if there's a user of them

This is essentially never used, yet it's about 1/3rd of the total
queue size. Allocate it when needed, and don't embed it in the queue.

Kill the queue flag for this while at it, since we can just check the
assigned pointer now.

Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-ioprio: don't set bio priority if not needed

We don't need to write to the bio if:

1) No ioprio value has ever been assigned to the blkcg
2) We wouldn't anyway, depending on bio and blkcg IO priority

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

blk-mq: move more plug handling from blk_mq_submit_bio into blk_add_rq_to_plug

Keep all the functionality for adding a request to a plug in a single place.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123160443.1315598-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: simplify the plug handling in blk_mq_submit_bio

blk_mq_submit_bio has two different plug cases, one that uses full
plugging and a limited plugging one.

The limited plugging case is only used for a corner case that does
not matter in real life:

 - no ->commit_rqs (so not NVMe)
 - no shared tags (so not SCSI)
 - not rotational (so no old disk or floppy driver)
 - must have multiple queues (so no eMMC)

Remove the limited merging case and all the related junk to simplify
blk_mq_submit_bio and the functions called from it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123160443.1315598-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

sr: set GENHD_FL_REMOVABLE earlier

Set up GENHD_FL_REMOVABLE together with the rest of the gendisk fields.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-15-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: cleanup the GENHD_FL_* definitions

Switch to an enum and tidy up the documentation.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-14-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't set GENHD_FL_NO_PART for hidden gendisks

Hidden gendisks can't be opened using blkdev_get_*, so we can't really
reach any of the partition scanning paths or partitioning ioctls except
for the initial partition scan from add_disk.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-13-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove GENHD_FL_EXT_DEVT

All modern drivers can support extra partitions using the extended
dev_t.  In fact except for the ioctl method drivers never even see
partitions in normal operation.

So remove the GENHD_FL_EXT_DEVT and allow extra partitions for all
block devices that do support partitions, and require those that
do not support partitions to explicit disallow them using
GENHD_FL_NO_PART.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-12-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove GENHD_FL_SUPPRESS_PARTITION_INFO

This flag is not set directly anywhere and only inherited from
GENHD_FL_HIDDEN.  Just check for GENHD_FL_HIDDEN instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-11-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

mmc: don't set GENHD_FL_SUPPRESS_PARTITION_INFO

This manually reverts 07b652cdbec3 ("mmc: card: Don't show eMMC RPMB and
BOOT areas in /proc/partitions").  Based on the commit description that
change was purely cosmetic.  mmc is the last driver that sets this
flag and thus prevents it from being removed.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Link: https://lore.kernel.org/r/20211122130625.1136848-10-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

null_blk: don't suppress partitioning information

This manually reverts commit 27290b469051 ("null_blk: suppress invalid
partition info").  The message in that commit log can't appearch as
the flag is never checked during probing, and there is no good reason
to treat null_blk special in /proc/partitions.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove the GENHD_FL_HIDDEN check in blkdev_get_no_open

Hidden gendisks never hash the block device inode, so this can't happen.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: rename GENHD_FL_NO_PART_SCAN to GENHD_FL_NO_PART

The GENHD_FL_NO_PART_SCAN controls more than just partitions canning,
so rename it to GENHD_FL_NO_PART.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Link: https://lore.kernel.org/r/20211122130625.1136848-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: merge disk_scan_partitions and blkdev_reread_part

Unify the functionality that implements a partition rescan for a
gendisk.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove a dead check in show_partition

disk_max_parts never returns 0 given that ->minors for devices not using
the extended dev_t must be non-zero, and disk_max_parts always returns
DISK_MAX_PARTS for the latter.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove GENHD_FL_CD

GENHD_FL_CD marks a gendisk as a vaguely CD-ROM like device.
Besides being used internally inside of sunvdc.c an xen-blkfront it
is used by xen-blkback as a hint to claim a device exported to a
guest is a CD-ROM like device.  Just check for disk->cdi instead
which is the right indicator for "real" CD-ROM or DVD drivers.  This
will miss the paravirtualized guest drivers, but those make little
sense to report anyway.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE to disk->event_flags

GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE is all about the event reporting
mechanism, so move it to the event_flags field.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move GENHD_FL_NATIVE_CAPACITY to disk->state

The flag to indicate an unlocked native capacity is dynamic state,
not a driver capability flag, so move it to disk->state.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211122130625.1136848-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: don't include blk-mq headers in blk-core.c

All request based code is in the blk-mq files now.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-12-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_print_req_error to blk-mq.c

This function is only used by the request completion path.  Factor out
a blk_status_to_str to keep blk_errors private in blk-core.c.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-11-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_dump_rq_flags to blk-mq.c

blk_dump_rq_flags deals with a request, so move it to blk-mq.c.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-10-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_account_io_{start,done} to blk-mq.c

These are only used for request based I/O, so move them where they are
used.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_steal_bios to blk-mq.c

Keep all the request based code together.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_rq_init to blk-mq.c

blk_rq_init deals with a request structure, so move it to blk-mq.c

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move request based cloning helpers to blk-mq.c

Keep all the request based code together.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

blk-mq: move blk_mq_flush_plug_list

Move blk_mq_flush_plug_list and blk_mq_plug_issue_direct down in blk-mq.c
to prepare for marking blk_mq_request_issue_directly static without the
need of a forward declaration.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove blk-exec.c

All this code is tightly coupled to the blk-mq core, so move it
there.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-4-hch@lst.de
[axboe: remove doc generation for blk-exec.c]
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: remove rq_flush_dcache_pages

This function is trivial, and flush_dcache_page is always defined, so
just open code it in the 2.5 callers.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: move blk_rq_err_bytes to scsi

blk_rq_err_bytes is only used by the scsi midlayer, so move it there.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20211117061404.331732-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Linux 5.16-rc3

Merge tag 'for_linus' of git://git./linux/kernel/git/mst/vhost

Pull vhost,virtio,vdpa bugfixes from Michael Tsirkin:
 "Misc fixes all over the place.

  Revert of virtio used length validation series: the approach taken
  does not seem to work, breaking too many guests in the process. We'll
  need to do length validation using some other approach"

[ This merge also ends up reverting commit f7a36b03a732 ("vsock/virtio:
  suppress used length validation"), which came in through the
  networking tree in the meantime, and was part of that whole used
  length validation series   - Linus ]

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  vdpa_sim: avoid putting an uninitialized iova_domain
  vhost-vdpa: clean irqs before reseting vdpa device
  virtio-blk: modify the value type of num in virtio_queue_rq()
  vhost/vsock: cleanup removing `len` variable
  vhost/vsock: fix incorrect used length reported to the guest
  Revert "virtio_ring: validate used buffer length"
  Revert "virtio-net: don't let virtio core to validate used length"
  Revert "virtio-blk: don't let virtio core to validate used length"
  Revert "virtio-scsi: don't let virtio core to validate used buffer length"

Merge tag 'x86-urgent-2021-11-28' of git://git./linux/kernel/git/tip/tip

Pull x86 build fix from Thomas Gleixner:
 "A single fix for a missing __init annotation of prepare_command_line()"

* tag 'x86-urgent-2021-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/boot: Mark prepare_command_line() __init

Merge tag 'sched-urgent-2021-11-28' of git://git./linux/kernel/git/tip/tip

Pull scheduler fix from Thomas Gleixner:
 "A single scheduler fix to ensure that there is no stale KASAN shadow
  state left on the idle task's stack when a CPU is brought up after it
  was brought down before"

* tag 'sched-urgent-2021-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/scs: Reset task stack state in bringup_cpu()

Merge tag 'perf-urgent-2021-11-28' of git://git./linux/kernel/git/tip/tip

Pull perf fix from Thomas Gleixner:
 "A single fix for perf to prevent it from sending SIGTRAP to another
  task from a trace point event as it's not possible to deliver a
  synchronous signal to a different task from there"

* tag 'perf-urgent-2021-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf: Ignore sigtrap for tracepoints destined for other tasks

Merge tag 'locking-urgent-2021-11-28' of git://git./linux/kernel/git/tip/tip

Pull locking fixes from Thomas Gleixner:
 "Two regression fixes for reader writer semaphores:

   - Plug a race in the lock handoff which is caused by inconsistency of
     the reader and writer path and can lead to corruption of the
     underlying counter.

   - down_read_trylock() is suboptimal when the lock is contended and
     multiple readers trylock concurrently. That's due to the initial
     value being read non-atomically which results in at least two
     compare exchange loops. Making the initial readout atomic reduces
     this significantly. Whith 40 readers by 11% in a benchmark which
     enforces contention on mmap_sem"

* tag 'locking-urgent-2021-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/rwsem: Optimize down_read_trylock() under highly contended case
  locking/rwsem: Make handoff bit handling more consistent

Merge tag 'trace-v5.16-rc2-3' of git://git./linux/kernel/git/rostedt/linux-trace

Pull another tracing fix from Steven Rostedt:
 "Fix the fix of pid filtering

  The setting of the pid filtering flag tested the "trace only this pid"
  case twice, and ignored the "trace everything but this pid" case.

  The 5.15 kernel does things a little differently due to the new sparse
  pid mask introduced in 5.16, and as the bug was discovered running the
  5.15 kernel, and the first fix was initially done for that kernel,
  that fix handled both cases (only pid and all but pid), but the
  forward port to 5.16 created this bug"

* tag 'trace-v5.16-rc2-3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Test the 'Do not trace this pid' case in create event

Merge tag 'iommu-fixes-v5.16-rc2' of git://git./linux/kernel/git/joro/iommu

Pull iommu fixes from Joerg Roedel:

 - Intel VT-d fixes:
     - Remove unused PASID_DISABLED
     - Fix RCU locking
     - Fix for the unmap_pages call-back

 - Rockchip RK3568 address mask fix

 - AMD IOMMUv2 log message clarification

* tag 'iommu-fixes-v5.16-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/vt-d: Fix unmap_pages support
  iommu/vt-d: Fix an unbalanced rcu_read_lock/rcu_read_unlock()
  iommu/rockchip: Fix PAGE_DESC_HI_MASKs for RK3568
  iommu/amd: Clarify AMD IOMMUv2 initialization messages
  iommu/vt-d: Remove unused PASID_DISABLED

Merge tag '5.16-rc2-ksmbd-fixes' of git://git.samba.org/ksmbd

Pull ksmbd fixes from Steve French:
 "Five ksmbd server fixes, four of them for stable:

   - memleak fix

   - fix for default data stream on filesystems that don't support xattr

   - error logging fix

   - session setup fix

   - minor doc cleanup"

* tag '5.16-rc2-ksmbd-fixes' of git://git.samba.org/ksmbd:
  ksmbd: fix memleak in get_file_stream_info()
  ksmbd: contain default data stream even if xattr is empty
  ksmbd: downgrade addition info error msg to debug in smb2_get_info_sec()
  docs: filesystem: cifs: ksmbd: Fix small layout issues
  ksmbd: Fix an error handling path in 'smb2_sess_setup()'

vmxnet3: Use generic Kconfig option for page size limit

Use the architecture independent Kconfig option PAGE_SIZE_LESS_THAN_64KB
to indicate that VMXNET3 requires a page size smaller than 64kB.

Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

fs: ntfs: Limit NTFS_RW to page sizes smaller than 64k

NTFS_RW code allocates page size dependent arrays on the stack. This
results in build failures if the page size is 64k or larger.

  fs/ntfs/aops.c: In function 'ntfs_write_mst_block':
  fs/ntfs/aops.c:1311:1: error:
the frame size of 2240 bytes is larger than 2048 bytes

Since commit f22969a66041 ("powerpc/64s: Default to 64K pages for 64 bit
book3s") this affects ppc:allmodconfig builds, but other architectures
supporting page sizes of 64k or larger are also affected.

Increasing the maximum frame size for affected architectures just to
silence this error does not really help.  The frame size would have to
be set to a really large value for 256k pages.  Also, a large frame size
could potentially result in stack overruns in this code and elsewhere
and is therefore not desirable.  Make NTFS_RW dependent on page sizes
smaller than 64k instead.

Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Cc: Anton Altaparmakov <anton@tuxera.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

arch: Add generic Kconfig option indicating page size smaller than 64k

NTFS_RW and VMXNET3 require a page size smaller than 64kB.  Add generic
Kconfig option for use outside architecture code to avoid architecture
specific Kconfig options in that code.

Suggested-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Cc: Anton Altaparmakov <anton@tuxera.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

tracing: Test the 'Do not trace this pid' case in create event

When creating a new event (via a module, kprobe, eprobe, etc), the
descriptors that are created must add flags for pid filtering if an
instance has pid filtering enabled, as the flags are used at the time the
event is executed to know if pid filtering should be done or not.

The "Only trace this pid" case was added, but a cut and paste error made
that case checked twice, instead of checking the "Trace all but this pid"
case.

Link: https://lore.kernel.org/all/202111280401.qC0z99JB-lkp@intel.com/
Fixes: 6cb206508b62 ("tracing: Check pid filtering when creating events")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>