When multiple threads perform IO to a thin device, the underlying
dm_bufio object can become a bottleneck; slowing down access to btree
nodes that store the thin metadata. Prior to this commit, each bufio
instance had a single mutex that was taken for every bufio operation.

This commit concentrates on improving the common case where: a user of
dm_bufio wishes to access, but not modify, a buffer which is already
within the dm_bufio cache.

Implementation::

  The code has been refactored; pulling out an 'lru' abstraction and a
  'buffer cache' abstraction (see 2 previous commits). This commit
  updates higher level bufio code (that performs allocation of buffers,
  IO and eviction/cache sizing) to leverage both abstractions. It also
  deals with the delicate locking requirements of both abstractions to
  provide finer grained locking. The result is significantly better
  concurrent IO performance.

  Before this commit, bufio has a global lru list it used to evict the
  oldest, clean buffers from _all_ clients. With the new locking we
  don’t want different ways to access the same buffer, so instead
  do_global_cleanup() loops around the clients asking them to free
  buffers older than a certain time.

  This commit also converts many old BUG_ONs to WARN_ON_ONCE, see the
  lru_evict and cache_evict code in particular.  They will return
  ER_DONT_EVICT if a given buffer somehow meets the invariants that
  should _never_ happen. [Aside from revising this commit's header and
  fixing coding style and whitespace nits: this switching to
  WARN_ON_ONCE is Mike Snitzer's lone contribution to this commit]

Testing::

  Some of the low level functions have been unit tested using dm-unit:
    https://github.com/jthornber/dm-unit/blob/main/src/tests/bufio.rs

  Higher level concurrency and IO is tested via a test only target
  found here:
    https://github.com/jthornber/linux/blob/2023-03-24-thin-concurrency-9/drivers/md/dm-bufio-test.c

  The associated userland side of these tests is here:
    https://github.com/jthornber/dmtest-python/blob/main/src/dmtest/bufio/bufio_tests.py



  In addition the full dmtest suite of tests (dm-thin, dm-cache, etc)
  has been run (~450 tests).

Performance::

  Most bufio operations have unchanged performance. But if multiple
  threads are attempting to get buffers concurrently, and these
  buffers are already in the cache then there's a big speed up. Eg,
  one test has 16 'hotspot' threads simulating btree lookups while
  another thread dirties the whole device. In this case the hotspot
  threads acquire the buffers about 25 times faster.

Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>

The buffer cache is responsible for managing the holder count,
tracking clean/dirty state, and choosing buffers via predicates.
Higher level code is responsible for allocation of buffers, IO and
eviction/cache sizing.

The buffer cache has thread safe methods for acquiring a reference
to an existing buffer. All other methods in buffer cache are _not_
threadsafe, and only contain enough locking to guarantee the safe
methods.

Rather than a single mutex, sharded rw_semaphores are used to allow
concurrent threads to 'get' buffers. Each rw_semaphore protects its
own rbtree of buffer entries.

Code that uses this new dm_buffer_cache abstraction will be introduced
in a following commit.

This commit moves the dm_buffer struct in preparation for finer grained
dm_buffer changes, in the next commit, to be more easily seen. It also
introduces temporary dm_buffer struct members to allow compilation of
this intermediate commit (they will be elided in the next commit).

This commit will cause "defined but not used" compiler warnings that
will be resolved by the next commit.

Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>

A CLOCK algorithm is used in this LRU abstraction.  This avoids
relinking list nodes, which would require a write lock protecting it.

None of the LRU methods are threadsafe; locking must be done at a
higher level.

Code that uses this new LRU will be introduced in the next 2 commits.

As such, this commit will cause "defined but not used" compiler warnings
that will be resolved by the next 2 commits.

Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>

Reasonable to relax to WARN_ON because these are easily avoided but do
offer some assurance future coding mistakes won't occur (if changes
tested properly).

Signed-off-by: Mike Snitzer <snitzer@kernel.org>

All these instances are entirely avoidable given that they speak to
coding mistakes that result in inappropriate use. Proper testing during
development will catch any such coding bug so its best to relax all of
these from BUG_ON to WARN_ON_ONCE.

Signed-off-by: Mike Snitzer <snitzer@kernel.org>

Using BUG_ON when tearing down is excessive. Relax these to WARN_ONs.

Signed-off-by: Mike Snitzer <snitzer@kernel.org>

Was used by multi-snapshot DM target that never went upstream.

Signed-off-by: Joe Thornber <ejt@redhat.com>
Acked-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>

Commit 7dd76d1f ("dm: improve bio splitting and associated IO
accounting") only called setup_split_accounting() from
__send_duplicate_bios() if a single bio were being issued. But the case
where duplicate bios are issued must call it too.

Otherwise the bio won't be split and resubmitted (via recursion through
block core back to DM) to submit the later portions of a bio (which may
map to an entirely different target).

For example, when discarding an entire DM striped device with the
following DM table:
 vg-lvol0: 0 159744 striped 2 128 7:0 2048 7:1 2048
 vg-lvol0: 159744 45056 striped 2 128 7:2 2048 7:3 2048

Before (broken, discards the first striped target's devices twice):
 device-mapper: striped: target_stripe=0, bdev=7:0, start=2048 len=79872
 device-mapper: striped: target_stripe=1, bdev=7:1, start=2048 len=79872
 device-mapper: striped: target_stripe=0, bdev=7:0, start=2049 len=22528
 device-mapper: striped: target_stripe=1, bdev=7:1, start=2048 len=22528

After (works as expected):
 device-mapper: striped: target_stripe=0, bdev=7:0, start=2048 len=79872
 device-mapper: striped: target_stripe=1, bdev=7:1, start=2048 len=79872
 device-mapper: striped: target_stripe=0, bdev=7:2, start=2048 len=22528
 device-mapper: striped: target_stripe=1, bdev=7:3, start=2048 len=22528

Fixes: 7dd76d1f

 ("dm: improve bio splitting and associated IO accounting")
Cc: stable@vger.kernel.org
Reported-by: Orange Kao <orange@aiven.io>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>

"Abnormal" bios include discards, write zeroes and secure erase. By no
longer passing the calculated 'len' pointer, commit 7dd06a25 ("dm:
allow dm_accept_partial_bio() for dm_io without duplicate bios") took a
senseless approach to disallowing dm_accept_partial_bio() from working
for duplicate bios processed using __send_duplicate_bios().

It inadvertently and incorrectly stopped the use of 'len' when
initializing a target's io (in alloc_tio). As such the resulting tio
could address more area of a device than it should.

For example, when discarding an entire DM striped device with the
following DM table:
 vg-lvol0: 0 159744 striped 2 128 7:0 2048 7:1 2048
 vg-lvol0: 159744 45056 striped 2 128 7:2 2048 7:3 2048

Before this fix:

 device-mapper: striped: target_stripe=0, bdev=7:0, start=2048 len=102400
 blkdiscard: attempt to access beyond end of device
 loop0: rw=2051, sector=2048, nr_sectors = 102400 limit=81920

 device-mapper: striped: target_stripe=1, bdev=7:1, start=2048 len=102400
 blkdiscard: attempt to access beyond end of device
 loop1: rw=2051, sector=2048, nr_sectors = 102400 limit=81920

After this fix;

 device-mapper: striped: target_stripe=0, bdev=7:0, start=2048 len=79872
 device-mapper: striped: target_stripe=1, bdev=7:1, start=2048 len=79872

Fixes: 7dd06a25

 ("dm: allow dm_accept_partial_bio() for dm_io without duplicate bios")
Cc: stable@vger.kernel.org
Reported-by: Orange Kao <orange@aiven.io>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>

Pull USB / Thunderbolt driver fixes from Greg KH:
 "Here are a small set of USB and Thunderbolt driver fixes for reported
  problems and a documentation update, for 6.3-rc4.

  Included in here are:

   - documentation update for uvc gadget driver

   - small thunderbolt driver fixes

   - cdns3 driver fixes

   - dwc3 driver fixes

   - dwc2 driver fixes

   - chipidea driver fixes

   - typec driver fixes

   - onboard_usb_hub device id updates

   - quirk updates

  All of these have been in linux-next with no reported problems"

* tag 'usb-6.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (30 commits)
  usb: dwc2: fix a race, don't power off/on phy for dual-role mode
  usb: dwc2: fix a devres leak in hw_enable upon suspend resume
  usb: chipidea: core: fix possible concurrent when switch role
  usb: chipdea: core: fix return -EINVAL if request role is the same with current role
  thunderbolt: Rename shadowed variables bit to interrupt_bit and auto_clear_bit
  thunderbolt: Disable interrupt auto clear for rings
  thunderbolt: Use const qualifier for `ring_interrupt_index`
  usb: gadget: Use correct endianness of the wLength field for WebUSB
  uas: Add US_FL_NO_REPORT_OPCODES for JMicron JMS583Gen 2
  usb: cdnsp: changes PCI Device ID to fix conflict with CNDS3 driver
  usb: cdns3: Fix issue with using incorrect PCI device function
  usb: cdnsp: Fixes issue with redundant Status Stage
  MAINTAINERS: make me a reviewer of USB/IP
  thunderbolt: Use scale field when allocating USB3 bandwidth
  thunderbolt: Limit USB3 bandwidth of certain Intel USB4 host routers
  thunderbolt: Call tb_check_quirks() after initializing adapters
  thunderbolt: Add missing UNSET_INBOUND_SBTX for retimer access
  thunderbolt: Fix memory leak in margining
  usb: dwc2: drd: fix inconsistent mode if role-switch-default-mode="host"
  docs: usb: Add documentation for the UVC Gadget
  ...

Pull scheduler fix from Borislav Petkov:

 - Fix a corner case where vruntime of a task is not being sanitized

* tag 'sched_urgent_for_v6.3_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/fair: Sanitize vruntime of entity being migrated

Pull perf fix from Borislav Petkov:

 - Properly clear perf event status tracking in the AMD perf event
   overflow handler

* tag 'perf_urgent_for_v6.3_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86/amd/core: Always clear status for idx

Pull core fixes from Borislav Petkov:

 - Do the delayed RCU wakeup for kthreads in the proper order so that
   former doesn't get ignored

 - A noinstr warning fix

* tag 'core_urgent_for_v6.3_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  entry/rcu: Check TIF_RESCHED _after_ delayed RCU wake-up
  entry: Fix noinstr warning in __enter_from_user_mode()

Pull x86 fixes from Borislav Petkov:

 - Add a AMX ptrace self test

 - Prevent a false-positive warning when retrieving the (invalid)
   address of dynamic FPU features in their init state which are not
   saved in init_fpstate at all

 - Randomize per-CPU entry areas only when KASLR is enabled

* tag 'x86_urgent_for_v6.3_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  selftests/x86/amx: Add a ptrace test
  x86/fpu/xstate: Prevent false-positive warning in __copy_xstate_uabi_buf()
  x86/mm: Do not shuffle CPU entry areas without KASLR

Pull cifs client fixes from Steve French:
 "Twelve cifs/smb3 client fixes (most also for stable)

   - forced umount fix

   - fix for two perf regressions

   - reconnect fixes

   - small debugging improvements

   - multichannel fixes"

* tag 'smb3-client-fixes-6.3-rc3' of git://git.samba.org/sfrench/cifs-2.6:
  smb3: fix unusable share after force unmount failure
  cifs: fix dentry lookups in directory handle cache
  smb3: lower default deferred close timeout to address perf regression
  cifs: fix missing unload_nls() in smb2_reconnect()
  cifs: avoid race conditions with parallel reconnects
  cifs: append path to open_enter trace event
  cifs: print session id while listing open files
  cifs: dump pending mids for all channels in DebugData
  cifs: empty interface list when server doesn't support query interfaces
  cifs: do not poll server interfaces too regularly
  cifs: lock chan_lock outside match_session
  cifs: check on...

Pull nfsd fix from Chuck Lever:

 - Fix a crash when using NFS with krb5p

* tag 'nfsd-6.3-4' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux:
  SUNRPC: Fix a crash in gss_krb5_checksum()

Pull yet more xfs bug fixes from Darrick Wong:
 "The first bugfix addresses a longstanding problem where we use the
  wrong file mapping cursors when trying to compute the speculative
  preallocation quantity. This has been causing sporadic crashes when
  alwayscow mode is engaged.

  The other two fixes correct minor problems in more recent changes.

   - Fix the new allocator tracepoints because git am mismerged the
     changes such that the trace_XXX got rebased to be in function YYY
     instead of XXX

   - Ensure that the perag AGFL_RESET state is consistent with whatever
     we've just read off the disk

   - Fix a bug where we used the wrong iext cursor during a write begin"

* tag 'xfs-6.3-fixes-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: fix mismerged tracepoints
  xfs: clear incore AGFL_RESET state if it's not needed
  xfs: pass the correct cursor to xfs_iomap_prealloc_size

Pull xfs percpu counter fixes from Darrick Wong:
 "We discovered a filesystem summary counter corruption problem that was
  traced to cpu hot-remove racing with the call to percpu_counter_sum
  that sets the free block count in the superblock when writing it to
  disk. The root cause is that percpu_counter_sum doesn't cull from
  dying cpus and hence misses those counter values if the cpu shutdown
  hooks have not yet run to merge the values.

  I'm hoping this is a fairly painless fix to the problem, since the
  dying cpu mask should generally be empty. It's been in for-next for a
  week without any complaints from the bots.

   - Fix a race in the percpu counters summation code where the
     summation failed to add in the values for any CPUs that were dying
     but not yet dead. This fixes some minor discrepancies and incorrect
     assertions when running generic/650"

* tag 'xfs-6.3-fixes-4' of git://git.kernel.org/pub/scm/...

Pull xfs fixes from Darrick Wong:
 "This batch started with some debugging enhancements to the new
  allocator refactoring that we put in 6.3-rc1 to assist developers in
  rebasing their dev branches.

  As for more serious code changes -- there's a bug fix to make the
  lockless allocator scan the whole filesystem before resorting to the
  locking allocator. We're also adding a selftest for the venerable
  directory/xattr hash function to make sure that it produces consistent
  results so that we can address any fallout as soon as possible.

   - Add a few debugging assertions so that people (me) trying to port
     code to the new allocator functions don't mess up the caller
     requirements

   - Relax some overly cautious lock ordering enforcement in the new
     allocator code, which means that file allocations will locklessly
     scan for the best space they can get before backing off to the
     traditional lock-and-really-get-it behavior

   - Add tracepoints to make it easier to trace the xfs allocator
     behavior

   - Actually test the dir/xattr hash algorithm to make sure it produces
     consistent results across all the platforms XFS supports"

* tag 'xfs-6.3-fixes-3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: test dir/attr hash when loading module
  xfs: add tracepoints for each of the externally visible allocators
  xfs: walk all AGs if TRYLOCK passed to xfs_alloc_vextent_iterate_ags
  xfs: try to idiot-proof the allocators

Merge tag 'hwmon-for-v6.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging

Pull hwmon fixes from Guenter Roeck:

 - it87: Fix voltage scaling for chips with 10.9mV ADCs

 - xgene: Fix ioremap and memremap leak

 - peci/cputemp: Fix miscalculated DTS temperature for SKX

 - hwmon core: fix potential sensor registration failure with thermal
   subsystem if of_node is missing

* tag 'hwmon-for-v6.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon (it87): Fix voltage scaling for chips with 10.9mV  ADCs
  hwmon: (xgene) Fix ioremap and memremap leak
  hwmon: fix potential sensor registration fail if of_node is missing
  hwmon: (peci/cputemp) Fix miscalculated DTS for SKX

Merge tag 'mm-hotfixes-stable-2023-03-24-17-09' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
 "21 hotfixes, 8 of which are cc:stable. 11 are for MM, the remainder
  are for other subsystems"

* tag 'mm-hotfixes-stable-2023-03-24-17-09' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (21 commits)
  mm: mmap: remove newline at the end of the trace
  mailmap: add entries for Richard Leitner
  kcsan: avoid passing -g for test
  kfence: avoid passing -g for test
  mm: kfence: fix using kfence_metadata without initialization in show_object()
  lib: dhry: fix unstable smp_processor_id(_) usage
  mailmap: add entry for Enric Balletbo i Serra
  mailmap: map Sai Prakash Ranjan's old address to his current one
  mailmap: map Rajendra Nayak's old address to his current one
  Revert "kasan: drop skip_kasan_poison variable in free_pages_prepare"
  mailmap: add entry for Tobias Klauser
  kasan, powerpc: don't rename memintrinsics if compiler adds prefixes
  mm/ks...

Pull ksmbd server fixes from Steve French:

 - return less confusing messages on unsupported dialects
   (STATUS_NOT_SUPPORTED instead of I/O error)

 - fix for overly frequent inactive session termination

 - fix refcount leak

 - fix bounds check problems found by static checkers

 - fix to advertise named stream support correctly

 - Fix AES256 signing bug when connected to from MacOS

* tag '6.3-rc3-ksmbd-smb3-server-fixes' of git://git.samba.org/ksmbd:
  ksmbd: return unsupported error on smb1 mount
  ksmbd: return STATUS_NOT_SUPPORTED on unsupported smb2.0 dialect
  ksmbd: don't terminate inactive sessions after a few seconds
  ksmbd: fix possible refcount leak in smb2_open()
  ksmbd: add low bound validation to FSCTL_QUERY_ALLOCATED_RANGES
  ksmbd: add low bound validation to FSCTL_SET_ZERO_DATA
  ksmbd: set FILE_NAMED_STREAMS attribute in FS_ATTRIBUTE_INFORMATION
  ksmbd: fix wrong signingkey creation when encryption is AES256

Pull ARM SoC fixes from Arnd Bergmann:
 "As usual, most of the bug fixes address issues in the devicetree
  files, and out of these, most are for the Qualcomm and NXP platforms,
  including:

   - A missing 'reserved-memory' property on LG G Watch R that is needed
     to prevent clashing with firmware

   - Annotations for cache coherency on multiple machines

   - Corrections for pinctrl, regulator, clock, iommu and power domain
     properties for i.MX and Qualcomm to correctly reflect the hardware
     settings

   - Firmware file names on multiple machines SA8540P Ride board

   - An incompatible change to the qcom vadc driver requires adding
     individual labels

   - Fix EQoS PHY reset GPIO by dropping the deprecated/wrong property
     and switch to the new bindings.

   - A fix for PCI bus address translation Tegra194 and Tegra234.

  There are also a couple of device driver fixes, addressing:

   - A race cond...

Pull power supply fixes from Sebastian Reichel:

 - rk817: Fix compiler warning

 - cros_usbpd-charger: Fix excessive error printing

 - axp288_fuel_gauge: handle platform_get_irq error

 - bq24190 and da9150: Fix race condition in remove path

* tag 'for-v6.3-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply:
  power: supply: da9150: Fix use after free bug in da9150_charger_remove due to race condition
  power: supply: bq24190: Fix use after free bug in bq24190_remove due to race condition
  power: supply: axp288_fuel_gauge: Added check for negative values
  power: supply: cros_usbpd: reclassify "default case!" as debug
  power: supply: rk817: Fix unsigned comparison with less than zero

Pull drm fixes from Daniel Vetter:

 - usual pile of fixes for amdgpu & i915

 - probe error handling fixes for meson, lt8912b bridge

 - the host1x patch from Arnd

 - panel-orientation fix for Lenovo Book X90F

* tag 'drm-fixes-2023-03-24' of git://anongit.freedesktop.org/drm/drm: (23 commits)
  gpu: host1x: fix uninitialized variable use
  drm/amd/display: Set dcn32 caps.seamless_odm
  drm/amd/display: fix wrong index used in dccg32_set_dpstreamclk
  drm/amdgpu/nv: Apply ASPM quirk on Intel ADL + AMD Navi
  drm/amd/display: remove outdated 8bpc comments
  drm/amdgpu/gfx: set cg flags to enter/exit safe mode
  drm/amdgpu: Force signal hw_fences that are embedded in non-sched jobs
  drm/amdgpu: add mes resume when do gfx post soft reset
  drm/amdgpu: skip ASIC reset for APUs when go to S4
  drm/amdgpu: reposition the gpu reset checking for reuse
  drm/bridge: lt8912b: return EPROBE_DEFER if bridge is not found
  drm/meson: fix missing component unbind on bind errors
  drm: panel-orientation-quirks: Add quirk for Lenovo Yoga Book X90F
  Revert "drm/i915/hwmon: Enable PL1 power limit"
  drm/i915: Update vblank timestamping stuff on seamless M/N change
  drm/i915: Fix format for perf_limit_reasons
  drm/i915/gt: perform uc late init after probe error injection
  drm/i915/active: Fix missing debug object activation
  drm/i915/guc: Fix missing ecodes
  drm/i915/mtl: Disable MC6 for MTL A step
  ...

Merge tag 'for-6.3/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mike Snitzer:

 - Fix DM thin to work as a swap device by using 'limit_swap_bios' DM
   target flag (initially added to allow swap to dm-crypt) to throttle
   the amount of outstanding swap bios.

 - Fix DM crypt soft lockup warnings by calling cond_resched() from the
   cpu intensive loop in dmcrypt_write().

 - Fix DM crypt to not access an uninitialized tasklet. This fix allows
   for consistent handling of IO completion, by _not_ needlessly punting
   to a workqueue when tasklets are not needed.

 - Fix DM core's alloc_dev() initialization for DM stats to check for
   and propagate alloc_percpu() failure.

* tag 'for-6.3/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm stats: check for and propagate alloc_percpu failure
  dm crypt: avoid accessing uninitialized tasklet
  dm crypt: add cond_resched() to dmcrypt_write()
  dm thin: fix deadlock when swapping to thin device

Pull block fixes from Jens Axboe:

 - NVMe pull request via Christoph:
     - Send Identify with CNS 06h only to I/O controllers (Martin
       George)
     - Fix nvme_tcp_term_pdu to match spec (Caleb Sander)

 - Pass in issue_flags for uring_cmd, so the end_io handlers don't need
   to assume what the right context is (me)

 - Fix for ublk, marking it as LIVE before adding it to avoid races on
   the initial IO (Ming)

* tag 'block-6.3-2023-03-24' of git://git.kernel.dk/linux:
  nvme-tcp: fix nvme_tcp_term_pdu to match spec
  nvme: send Identify with CNS 06h only to I/O controllers
  block/io_uring: pass in issue_flags for uring_cmd task_work handling
  block: ublk_drv: mark device as LIVE before adding disk

Pull io_uring fixes from Jens Axboe:

 - Fix an issue with repeated -ECONNREFUSED on a socket (me)

 - Fix a NULL pointer deference due to a stale lookup cache for
   allocating direct descriptors (Savino)

* tag 'io_uring-6.3-2023-03-24' of git://git.kernel.dk/linux:
  io_uring/rsrc: fix null-ptr-deref in io_file_bitmap_get()
  io_uring/net: avoid sending -ECONNABORTED on repeated connection requests

Pull thermal control fixes from Rafael Wysocki:
 "These address two recent regressions related to thermal control.

  Specifics:

   - Restore the thermal core behavior regarding zero-temperature trip
     points to avoid a driver regression (Ido Schimmel)

   - Fix a recent regression in the ACPI processor driver preventing it
     from changing the number of CPU cooling device states exposed via
     sysfs after the given CPU cooling device has been registered
     (Rafael Wysocki)"

* tag 'thermal-6.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  thermal: core: Restore behavior regarding invalid trip points
  ACPI: processor: thermal: Update CPU cooling devices on cpufreq policy changes
  thermal: core: Introduce thermal_cooling_device_update()
  thermal: core: Introduce thermal_cooling_device_present()
  ACPI: processor: Reorder acpi_processor_driver_init()

Pull ACPI fixes from Rafael Wysocki:
 "These add new ACPI IRQ override and backlight detection quirks.

  Specifics:

   - Add backlight=native DMI quirk for Acer Aspire 3830TG to the ACPI
     backlight driver (Hans de Goede)

   - Add an ACPI IRQ override quirk for Medion S17413 (Aymeric Wibo)"

* tag 'acpi-6.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: resource: Add Medion S17413 to IRQ override quirk
  ACPI: video: Add backlight=native DMI quirk for Acer Aspire 3830TG

At some point in between sending this patch to the list and merging it
into for-next, the tracepoints got all mixed up because I've
over-reliant on automated tools not sucking.  The end result is that the
tracepoints are all wrong, so fix them.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>

If user does forced unmount ("umount -f") while files are still open
on the share (as was seen in a Kubernetes example running on SMB3.1.1
mount) then we were marking the share as "TID_EXITING" in umount_begin()
which caused all subsequent operations (except write) to fail ... but
unfortunately when umount_begin() is called we do not know yet that
there are open files or active references on the share that would prevent
unmount from succeeding.  Kubernetes had example when they were doing
umount -f when files were open which caused the share to become
unusable until the files were closed (and the umount retried).

Fix this so that TID_EXITING is not set until we are about to send
the tree disconnect (not at the beginning of forced umounts in
umount_begin) so that if "umount -f" fails (due to open files or
references) the mount is still usable.

Cc: stable@vger.kernel.org
Reviewed-by: Shyam Prasad N <sprasad@microsoft.com>
Reviewed-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>

Get rid of any prefix paths in @path before lookup_positive_unlocked()
as it will call ->lookup() which already adds those prefix paths
through build_path_from_dentry().

This has caused a performance regression when mounting shares with a
prefix path where readdir(2) would end up retrying several times to
open bad directory names that contained duplicate prefix paths.

Fix this by skipping any prefix paths in @path before calling
lookup_positive_unlocked().

Fixes: e4029e07

 ("cifs: find and use the dentry for cached non-root directories also")
Cc: stable@vger.kernel.org # 6.1+
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>

Performance tests with large number of threads noted that the change
of the default closetimeo (deferred close timeout between when
close is done by application and when client has to send the close
to the server), to 5 seconds from 1 second, significantly degraded
perf in some cases like this (in the filebench example reported,
the stats show close requests on the wire taking twice as long,
and 50% regression in filebench perf). This is stil configurable
via mount parm closetimeo, but to be safe, decrease default back
to its previous value of 1 second.

Reported-by: Yin Fengwei <fengwei.yin@intel.com>
Reported-by: kernel test robot <yujie.liu@intel.com>
Link: https://lore.kernel.org/lkml/997614df-10d4-af53-9571-edec36b0e2f3@intel.com/
Fixes: 5efdd912

 ("smb3: allow deferred close timeout to be configurable")
Cc: stable@vger.kernel.org # 6.0+
Tested-by: Yin Fengwei <fengwei.yin@intel.com>
Reviewed-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Reviewed-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>

Make sure to unload_nls() @nls_codepage if we no longer need it.

Fixes: bc962159

 ("cifs: avoid race conditions with parallel reconnects")
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Cc: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>

Pull slab fix from Vlastimil Babka:
 "A single build fix for a corner case configuration that is apparently
  possible to achieve on some arches, from Geert"

* tag 'slab-fix-for-6.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab:
  mm/slab: Fix undefined init_cache_node_node() for NUMA and !SMP

Pull EFI fixes from Ard Biesheuvel:

 - Set the NX compat flag for arm64 and zboot, to ensure compatibility
   with EFI firmware that complies with tightening requirements imposed
   across the ecosystem.

 - Improve identification of Ampere Altra systems based on SMBIOS data.

 - Fix some issues related to the EFI framebuffer that were introduced
   as a result from some refactoring related to zboot and the merge with
   sysfb.

 - Makefile tweak to avoid rebuilding vmlinuz unnecessarily.

 - Fix efi_random_alloc() return value on out of memory condition.

* tag 'efi-fixes-for-v6.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/efi/efi:
  efi/libstub: randomalloc: Return EFI_OUT_OF_RESOURCES on failure
  efi/libstub: Use relocated version of kernel's struct screen_info
  efi/libstub: zboot: Add compressed image to make targets
  efi: sysfb_efi: Add quirk for Lenovo Yoga Book X91F/L
  efi: sysfb_efi: Fix DMI quirks not working for simpledrm
  efi/libstub: smbios: Drop unused 'recsize' parameter
  arm64: efi: Use SMBIOS processor version to key off Ampere quirk
  efi/libstub: smbios: Use length member instead of record struct size
  efi: earlycon: Reprobe after parsing config tables
  arm64: efi: Set NX compat flag in PE/COFF header
  efi/libstub: arm64: Remap relocated image with strict permissions
  efi/libstub: zboot: Mark zboot EFI application as NX compatible

Merge tag 'qcom-driver-fixes-for-6.3' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux into soc/fixes

Qualcomm driver fixes for v6.3

Support for the secure world interrupting the SCM driver drive the wait
queue mechanism was recently introduced, but most platforms doesn't have
this mechanism and an error should not be printed in the log.

The rmtfs_mem driver recently gained support for assigning the region to
multiple VMIDs, but accidentally removed the support for running without
assignment. A couple of changes are introducd to correct this.

The SC8280XP LLCC slice configuration is wrong, reslting in incorrect
configuration of the hardware. The table is corrected, based on the
datasheet.

* tag 'qcom-driver-fixes-for-6.3' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux:
  firmware: qcom: scm: fix bogus irq error at probe
  soc: qcom: rmtfs: handle optional qcom,vmid correctly
  soc: qcom: rmtfs: fix error handling reading qcom,vmid
  soc: qcom: llcc: Fix slice configuration values for SC8280XP

Link: https://lore.kernel.org/r/20230323142505.1086072-1-andersson@kernel.org


Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Merge tag 'qcom-dts-fixes-for-6.3' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux into soc/fixes

Qualcomm ARM32 Devicetree fixes for v6.3

This introduces missing reserved-memory ranges on LG G Watch R,
resolving stability issues caused by Linux reusing memory used by
firmware.

* tag 'qcom-dts-fixes-for-6.3' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux:
  ARM: dts: qcom: apq8026-lg-lenok: add missing reserved memory

Link: https://lore.kernel.org/r/20230323141922.1085875-1-andersson@kernel.org


Signed-off-by: Arnd Bergmann <arnd@arndb.de>