Pull io_uring updates from Jens Axboe:

 - cancellation cleanups (Hao, Pavel)

 - io-wq accounting cleanup (Hao)

 - io_uring submit locking fix (Hao)

 - io_uring link handling fixes (Hao)

 - fixed file improvements (wangyangbo, Pavel)

 - allow updates of linked timeouts like regular timeouts (Pavel)

 - IOPOLL fix (Pavel)

 - remove batched file get optimization (Pavel)

 - improve reference handling (Pavel)

 - IRQ task_work batching (Pavel)

 - allow pure fixed file, and add support for open/accept (Pavel)

 - GFP_ATOMIC RT kernel fix

 - multiple CQ ring waiter improvement

 - funnel IRQ completions through task_work

 - add support for limiting async workers explicitly

 - add different clocksource support for timeouts

 - io-wq wakeup race fix

 - lots of cleanups and improvement (Pavel et al)

* tag 'for-5.15/io_uring-2021-08-30' of git://git.kernel.dk/linux-block: (87 commits)
  io-wq: fix wakeup race when adding new work
  io-wq: wqe and worker locks no longer need to be IRQ safe
  io-wq: check max_worker limits if a worker transitions bound state
  io_uring: allow updating linked timeouts
  io_uring: keep ltimeouts in a list
  io_uring: support CLOCK_BOOTTIME/REALTIME for timeouts
  io-wq: provide a way to limit max number of workers
  io_uring: add build check for buf_index overflows
  io_uring: clarify io_req_task_cancel() locking
  io_uring: add task-refs-get helper
  io_uring: fix failed linkchain code logic
  io_uring: remove redundant req_set_fail()
  io_uring: don't free request to slab
  io_uring: accept directly into fixed file table
  io_uring: hand code io_accept() fd installing
  io_uring: openat directly into fixed fd table
  net: add accept helper not installing fd
  io_uring: fix io_try_cancel_userdata race for iowq
  io_uring: IRQ rw completion batching
  io_uring: batch task work locking
  ...

Pull libata updates from Jens Axboe:
 "libata changes for the 5.15 release:

   - NCQ priority improvements (Damien, Niklas)

   - coccinelle warning fix (Jing)

   - dwc_460ex phy fix (Andy)"

* tag 'for-5.15/libata-2021-08-30' of git://git.kernel.dk/linux-block:
  include:libata: fix boolreturn.cocci warnings
  docs: sysfs-block-device: document ncq_prio_supported
  docs: sysfs-block-device: improve ncq_prio_enable documentation
  libata: Introduce ncq_prio_supported sysfs sttribute
  libata: print feature list on device scan
  libata: fix ata_read_log_page() warning
  libata: cleanup NCQ priority handling
  libata: cleanup ata_dev_configure()
  libata: cleanup device sleep capability detection
  libata: simplify ata_scsi_rbuf_fill()
  libata: fix ata_host_start()
  ata: sata_dwc_460ex: No need to call phy_exit() befre phy_init()

Pull block driver updates from Jens Axboe:
 "Sitting on top of the core block changes, here are the driver changes
  for the 5.15 merge window:

   - NVMe updates via Christoph:
       - suspend improvements for devices with an HMB (Keith Busch)
       - handle double completions more gacefull (Sagi Grimberg)
       - cleanup the selects for the nvme core code a bit (Sagi Grimberg)
       - don't update queue count when failing to set io queues (Ruozhu Li)
       - various nvmet connect fixes (Amit Engel)
       - cleanup lightnvm leftovers (Keith Busch, me)
       - small cleanups (Colin Ian King, Hou Pu)
       - add tracing for the Set Features command (Hou Pu)
       - CMB sysfs cleanups (Keith Busch)
       - add a mutex_destroy call (Keith Busch)

   - remove lightnvm subsystem. It's served its purpose and ultimately
     led to zoned nvme support, we no longer need it (Christoph)

   - revert floppy O_NDELAY fix (Denis)

   - nbd fixes (Hou, Pavel, Baokun)

   - nbd locking fixes (Tetsuo)

   - nbd device removal fixes (Christoph)

   - raid10 rcu warning fix (Xiao)

   - raid1 write behind fix (Guoqing)

   - rnbd fixes (Gioh, Md Haris)

   - misc fixes (Colin)"

* tag 'for-5.15/drivers-2021-08-30' of git://git.kernel.dk/linux-block: (42 commits)
  Revert "floppy: reintroduce O_NDELAY fix"
  raid1: ensure write behind bio has less than BIO_MAX_VECS sectors
  md/raid10: Remove unnecessary rcu_dereference in raid10_handle_discard
  nbd: remove nbd->destroy_complete
  nbd: only return usable devices from nbd_find_unused
  nbd: set nbd->index before releasing nbd_index_mutex
  nbd: prevent IDR lookups from finding partially initialized devices
  nbd: reset NBD to NULL when restarting in nbd_genl_connect
  nbd: add missing locking to the nbd_dev_add error path
  nvme: remove the unused NVME_NS_* enum
  nvme: remove nvm_ndev from ns
  nvme: Have NVME_FABRICS select NVME_CORE instead of transport drivers
  block: nbd: add sanity check for first_minor
  nvmet: check that host sqsize does not exceed ctrl MQES
  nvmet: avoid duplicate qid in connect cmd
  nvmet: pass back cntlid on successful completion
  nvme-rdma: don't update queue count when failing to set io queues
  nvme-tcp: don't update queue count when failing to set io queues
  nvme-tcp: pair send_mutex init with destroy
  nvme: allow user toggling hmb usage
  ...

Pull block updates from Jens Axboe:
 "Nothing major in here - lots of good cleanups and tech debt handling,
  which is also evident in the diffstats. In particular:

   - Add disk sequence numbers (Matteo)

   - Discard merge fix (Ming)

   - Relax disk zoned reporting restrictions (Niklas)

   - Bio error handling zoned leak fix (Pavel)

   - Start of proper add_disk() error handling (Luis, Christoph)

   - blk crypto fix (Eric)

   - Non-standard GPT location support (Dmitry)

   - IO priority improvements and cleanups (Damien)o

   - blk-throtl improvements (Chunguang)

   - diskstats_show() stack reduction (Abd-Alrhman)

   - Loop scheduler selection (Bart)

   - Switch block layer to use kmap_local_page() (Christoph)

   - Remove obsolete disk_name helper (Christoph)

   - block_device refcounting improvements (Christoph)

   - Ensure gendisk always has a request queue reference (Christoph)

   - Misc fixes/cleanups (Shaokun, Oliver, Guoqing)"

* tag 'for-5.15/block-2021-08-30' of git://git.kernel.dk/linux-block: (129 commits)
  sg: pass the device name to blk_trace_setup
  block, bfq: cleanup the repeated declaration
  blk-crypto: fix check for too-large dun_bytes
  blk-zoned: allow BLKREPORTZONE without CAP_SYS_ADMIN
  blk-zoned: allow zone management send operations without CAP_SYS_ADMIN
  block: mark blkdev_fsync static
  block: refine the disk_live check in del_gendisk
  mmc: sdhci-tegra: Enable MMC_CAP2_ALT_GPT_TEGRA
  mmc: block: Support alternative_gpt_sector() operation
  partitions/efi: Support non-standard GPT location
  block: Add alternative_gpt_sector() operation
  bio: fix page leak bio_add_hw_page failure
  block: remove CONFIG_DEBUG_BLOCK_EXT_DEVT
  block: remove a pointless call to MINOR() in device_add_disk
  null_blk: add error handling support for add_disk()
  virtio_blk: add error handling support for add_disk()
  block: add error handling for device_add_disk / add_disk
  block: return errors from disk_alloc_events
  block: return errors from blk_integrity_add
  block: call blk_register_queue earlier in device_add_disk
  ...

Pull timer updates from Thomas Gleixner:
 "Updates for timekeeping, timers and related drivers:

  Core code:

   - Cure a couple of correctness issues in the posix CPU timer code to
     prevent that the tick dependency for NOHZ full is kept alive for no
     reason.

   - Avoid expensive double reprogramming of the clockevent device in
     hrtimer_start_range_ns().

   - Avoid pointless SMP function calls when the clock was set to avoid
     disturbing CPUs which do not have any affected timers queued.

   - Make the clocksource watchdog test work correctly when CONFIG_HZ is
     less than 100.

  Drivers:

   - Prefer the ARM architected timer over the Exynos timer which is way
     more expensive to access.

   - Add device tree bindings for new Ingenic SoCs

   - The usual improvements and cleanups all over the place"

* tag 'timers-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (29 commits)
  clocksource: Make clocksource watchdog test safe for slow-HZ systems
  dt-bindings: timer: Add ABIs for new Ingenic SoCs
  clocksource/drivers/fttmr010: Pass around less pointers
  clocksource/drivers/mediatek: Optimize systimer irq clear flow on shutdown
  clocksource/drivers/ingenic: Use bitfield macro helpers
  clocksource/drivers/sh_cmt: Fix wrong setting if don't request IRQ for clock source channel
  dt-bindings: timer: convert rockchip,rk-timer.txt to YAML
  clocksource/drivers/exynos_mct: Mark MCT device as CLOCK_EVT_FEAT_PERCPU
  clocksource/drivers/exynos_mct: Prioritise Arm arch timer on arm64
  hrtimer: Unbreak hrtimer_force_reprogram()
  hrtimer: Use raw_cpu_ptr() in clock_was_set()
  hrtimer: Avoid more SMP function calls in clock_was_set()
  hrtimer: Avoid unnecessary SMP function calls in clock_was_set()
  hrtimer: Add bases argument to clock_was_set()
  time/timekeeping: Avoid invoking clock_was_set() twice
  timekeeping: Distangle resume and clock-was-set events
  timerfd: Provide timerfd_resume()
  hrtimer: Force clock_was_set() handling for the HIGHRES=n, NOHZ=y case
  hrtimer: Ensure timerfd notification for HIGHRES=n
  hrtimer: Consolidate reprogramming code
  ...

Pull misc x86 updates from Thomas Gleixner:
 "A set of updates for the x86 reboot code:

   - Limit the Dell Optiplex 990 quirk to early BIOS versions to avoid
     the full 'power cycle' alike reboot which is required for the buggy
     BIOSes.

   - Update documentation for the reboot=pci command line option and
     document how DMI platform quirks can be overridden"

* tag 'x86-misc-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/reboot: Limit Dell Optiplex 990 quirk to early BIOS versions
  x86/reboot: Document how to override DMI platform quirks
  x86/reboot: Document the "reboot=pci" option

Pull x86 PIRQ updates from Thomas Gleixner:
 "A set of updates to support port 0x22/0x23 based PCI configuration
  space which can be found on various ALi chipsets and is also available
  on older Intel systems which expose a PIRQ router.

  While the Intel support is more or less nostalgia, the ALi chips are
  still in use on popular embedded boards used for routers"

* tag 'x86-irq-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86: Fix typo s/ECLR/ELCR/ for the PIC register
  x86: Avoid magic number with ELCR register accesses
  x86/PCI: Add support for the Intel 82426EX PIRQ router
  x86/PCI: Add support for the Intel 82374EB/82374SB (ESC) PIRQ router
  x86/PCI: Add support for the ALi M1487 (IBC) PIRQ router
  x86: Add support for 0x22/0x23 port I/O configuration space

Pull x86 cache flush updates from Thomas Gleixner:
 "A reworked version of the opt-in L1D flush mechanism.

  This is a stop gap for potential future speculation related hardware
  vulnerabilities and a mechanism for truly security paranoid
  applications.

  It allows a task to request that the L1D cache is flushed when the
  kernel switches to a different mm. This can be requested via prctl().

  Changes vs the previous versions:

   - Get rid of the software flush fallback

   - Make the handling consistent with other mitigations

   - Kill the task when it ends up on a SMT enabled core which defeats
     the purpose of L1D flushing obviously"

* tag 'x86-cpu-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  Documentation: Add L1D flushing Documentation
  x86, prctl: Hook L1D flushing in via prctl
  x86/mm: Prepare for opt-in based L1D flush in switch_mm()
  x86/process: Make room for TIF_SPEC_L1D_FLUSH
  sched: Add task_work callback for paranoid L1D flush
  x86/mm: Refactor cond_ibpb() to support other use cases
  x86/smp: Add a per-cpu view of SMT state

Pull irq updates from Thomas Gleixner:
 "Updates to the interrupt core and driver subsystems:

  Core changes:

   - The usual set of small fixes and improvements all over the place,
     but nothing stands out

  MSI changes:

   - Further consolidation of the PCI/MSI interrupt chip code

   - Make MSI sysfs code independent of PCI/MSI and expose the MSI
     interrupts of platform devices in the same way as PCI exposes them.

  Driver changes:

   - Support for ARM GICv3 EPPI partitions

   - Treewide conversion to generic_handle_domain_irq() for all chained
     interrupt controllers

   - Conversion to bitmap_zalloc() throughout the irq chip drivers

   - The usual set of small fixes and improvements"

* tag 'irq-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (57 commits)
  platform-msi: Add ABI to show msi_irqs of platform devices
  genirq/msi: Move MSI sysfs handling from PCI to MSI core
  genirq/cpuhotplug: Demote debug printk to KERN_DEBUG
  irqchip/qcom-pdc: Trim unused levels of the interrupt hierarchy
  irqdomain: Export irq_domain_disconnect_hierarchy()
  irqchip/gic-v3: Fix priority comparison when non-secure priorities are used
  irqchip/apple-aic: Fix irq_disable from within irq handlers
  pinctrl/rockchip: drop the gpio related codes
  gpio/rockchip: drop irq_gc_lock/irq_gc_unlock for irq set type
  gpio/rockchip: support next version gpio controller
  gpio/rockchip: use struct rockchip_gpio_regs for gpio controller
  gpio/rockchip: add driver for rockchip gpio
  dt-bindings: gpio: change items restriction of clock for rockchip,gpio-bank
  pinctrl/rockchip: add pinctrl device to gpio bank struct
  pinctrl/rockchip: separate struct rockchip_pin_bank to a head file
  pinctrl/rockchip: always enable clock for gpio controller
  genirq: Fix kernel doc indentation
  EDAC/altera: Convert to generic_handle_domain_irq()
  powerpc: Bulk conversion to generic_handle_domain_irq()
  nios2: Bulk conversion to generic_handle_domain_irq()
  ...

Pull locking and atomics updates from Thomas Gleixner:
 "The regular pile:

   - A few improvements to the mutex code

   - Documentation updates for atomics to clarify the difference between
     cmpxchg() and try_cmpxchg() and to explain the forward progress
     expectations.

   - Simplification of the atomics fallback generator

   - The addition of arch_atomic_long*() variants and generic arch_*()
     bitops based on them.

   - Add the missing might_sleep() invocations to the down*() operations
     of semaphores.

  The PREEMPT_RT locking core:

   - Scheduler updates to support the state preserving mechanism for
     'sleeping' spin- and rwlocks on RT.

     This mechanism is carefully preserving the state of the task when
     blocking on a 'sleeping' spin- or rwlock and takes regular wake-ups
     targeted at the same task into account. The preserved or updated
     (via a regular wakeup) state is restored when the lock has been
     acquired.

   - Restructuring of the rtmutex code so it can be utilized and
     extended for the RT specific lock variants.

   - Restructuring of the ww_mutex code to allow sharing of the ww_mutex
     specific functionality for rtmutex based ww_mutexes.

   - Header file disentangling to allow substitution of the regular lock
     implementations with the PREEMPT_RT variants without creating an
     unmaintainable #ifdef mess.

   - Shared base code for the PREEMPT_RT specific rw_semaphore and
     rwlock implementations.

     Contrary to the regular rw_semaphores and rwlocks the PREEMPT_RT
     implementation is writer unfair because it is infeasible to do
     priority inheritance on multiple readers. Experience over the years
     has shown that real-time workloads are not the typical workloads
     which are sensitive to writer starvation.

     The alternative solution would be to allow only a single reader
     which has been tried and discarded as it is a major bottleneck
     especially for mmap_sem. Aside of that many of the writer
     starvation critical usage sites have been converted to a writer
     side mutex/spinlock and RCU read side protections in the past
     decade so that the issue is less prominent than it used to be.

   - The actual rtmutex based lock substitutions for PREEMPT_RT enabled
     kernels which affect mutex, ww_mutex, rw_semaphore, spinlock_t and
     rwlock_t. The spin/rw_lock*() functions disable migration across
     the critical section to preserve the existing semantics vs per-CPU
     variables.

   - Rework of the futex REQUEUE_PI mechanism to handle the case of
     early wake-ups which interleave with a re-queue operation to
     prevent the situation that a task would be blocked on both the
     rtmutex associated to the outer futex and the rtmutex based hash
     bucket spinlock.

     While this situation cannot happen on !RT enabled kernels the
     changes make the underlying concurrency problems easier to
     understand in general. As a result the difference between !RT and
     RT kernels is reduced to the handling of waiting for the critical
     section. !RT kernels simply spin-wait as before and RT kernels
     utilize rcu_wait().

   - The substitution of local_lock for PREEMPT_RT with a spinlock which
     protects the critical section while staying preemptible. The CPU
     locality is established by disabling migration.

  The underlying concepts of this code have been in use in PREEMPT_RT for
  way more than a decade. The code has been refactored several times over
  the years and this final incarnation has been optimized once again to be
  as non-intrusive as possible, i.e. the RT specific parts are mostly
  isolated.

  It has been extensively tested in the 5.14-rt patch series and it has
  been verified that !RT kernels are not affected by these changes"

* tag 'locking-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (92 commits)
  locking/rtmutex: Return success on deadlock for ww_mutex waiters
  locking/rtmutex: Prevent spurious EDEADLK return caused by ww_mutexes
  locking/rtmutex: Dequeue waiter on ww_mutex deadlock
  locking/rtmutex: Dont dereference waiter lockless
  locking/semaphore: Add might_sleep() to down_*() family
  locking/ww_mutex: Initialize waiter.ww_ctx properly
  static_call: Update API documentation
  locking/local_lock: Add PREEMPT_RT support
  locking/spinlock/rt: Prepare for RT local_lock
  locking/rtmutex: Add adaptive spinwait mechanism
  locking/rtmutex: Implement equal priority lock stealing
  preempt: Adjust PREEMPT_LOCK_OFFSET for RT
  locking/rtmutex: Prevent lockdep false positive with PI futexes
  futex: Prevent requeue_pi() lock nesting issue on RT
  futex: Simplify handle_early_requeue_pi_wakeup()
  futex: Reorder sanity checks in futex_requeue()
  futex: Clarify comment in futex_requeue()
  futex: Restructure futex_requeue()
  futex: Correct the number of requeued waiters for PI
  futex: Remove bogus condition for requeue PI
  ...

Pull SMP core updates from Thomas Gleixner:

 - Replace get/put_online_cpus() in various places. The final removal
   will happen shortly before v5.15-rc1 when the rest of the patches
   have been merged.

 - Add debug code to help the analysis of CPU hotplug failures

 - A set of kernel doc updates

* tag 'smp-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  mm: Replace deprecated CPU-hotplug functions.
  md/raid5: Replace deprecated CPU-hotplug functions.
  Documentation: Replace deprecated CPU-hotplug functions.
  smp: Fix all kernel-doc warnings
  cpu/hotplug: Add debug printks for hotplug callback failures
  cpu/hotplug: Use DEVICE_ATTR_*() macro
  cpu/hotplug: Eliminate all kernel-doc warnings
  cpu/hotplug: Fix kernel doc warnings for __cpuhp_setup_state_cpuslocked()
  cpu/hotplug: Fix comment typo
  smpboot: Replace deprecated CPU-hotplug functions.

Pull debugobjects update from Thomas Gleixner:
 "A single commit for debugobjects to make them work on PREEMPT_RT by
  preventing object pool refill in atomic contexts"

* tag 'core-debugobjects-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  debugobjects: Make them PREEMPT_RT aware

Pull EFI updates from Ingo Molnar:
 "A handful of EFI changes for this cycle:

   - EFI CPER parsing improvements

   - Don't take the address of efi_guid_t internal fields"

* tag 'efi-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi: cper: check section header more appropriately
  efi: Don't use knowledge about efi_guid_t internals
  efi: cper: fix scnprintf() use in cper_mem_err_location()

Pull x86 perf event updates from Ingo Molnar:

 - Add support for Intel Sapphire Rapids server CPU uncore events

 - Allow the AMD uncore driver to be built as a module

 - Misc cleanups and fixes

* tag 'perf-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits)
  perf/x86/amd/ibs: Add bitfield definitions in new <asm/amd-ibs.h> header
  perf/amd/uncore: Allow the driver to be built as a module
  x86/cpu: Add get_llc_id() helper function
  perf/amd/uncore: Clean up header use, use <linux/ include paths instead of <asm/
  perf/amd/uncore: Simplify code, use free_percpu()'s built-in check for NULL
  perf/hw_breakpoint: Replace deprecated CPU-hotplug functions
  perf/x86/intel: Replace deprecated CPU-hotplug functions
  perf/x86: Remove unused assignment to pointer 'e'
  perf/x86/intel/uncore: Fix IIO cleanup mapping procedure for SNR/ICX
  perf/x86/intel/uncore: Support IMC free-running counters on Sapphire Rapids server
  perf/x86/intel/uncore: Support IIO free-running counters on Sapphire Rapids server
  perf/x86/intel/uncore: Factor out snr_uncore_mmio_map()
  perf/x86/intel/uncore: Add alias PMU name
  perf/x86/intel/uncore: Add Sapphire Rapids server MDF support
  perf/x86/intel/uncore: Add Sapphire Rapids server M3UPI support
  perf/x86/intel/uncore: Add Sapphire Rapids server UPI support
  perf/x86/intel/uncore: Add Sapphire Rapids server M2M support
  perf/x86/intel/uncore: Add Sapphire Rapids server IMC support
  perf/x86/intel/uncore: Add Sapphire Rapids server PCU support
  perf/x86/intel/uncore: Add Sapphire Rapids server M2PCIe support
  ...

Pull scheduler updates from Ingo Molnar:

 - The biggest change in this cycle is scheduler support for asymmetric
   scheduling affinity, to support the execution of legacy 32-bit tasks
   on AArch32 systems that also have 64-bit-only CPUs.

   Architectures can fill in this functionality by defining their own
   task_cpu_possible_mask(p). When this is done, the scheduler will make
   sure the task will only be scheduled on CPUs that support it.

   (The actual arm64 specific changes are not part of this tree.)

   For other architectures there will be no change in functionality.

 - Add cgroup SCHED_IDLE support

 - Increase node-distance flexibility & delay determining it until a CPU
   is brought online. (This enables platforms where node distance isn't
   final until the CPU is only.)

 - Deadline scheduler enhancements & fixes

 - Misc fixes & cleanups.

* tag 'sched-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (27 commits)
  eventfd: Make signal recursion protection a task bit
  sched/fair: Mark tg_is_idle() an inline in the !CONFIG_FAIR_GROUP_SCHED case
  sched: Introduce dl_task_check_affinity() to check proposed affinity
  sched: Allow task CPU affinity to be restricted on asymmetric systems
  sched: Split the guts of sched_setaffinity() into a helper function
  sched: Introduce task_struct::user_cpus_ptr to track requested affinity
  sched: Reject CPU affinity changes based on task_cpu_possible_mask()
  cpuset: Cleanup cpuset_cpus_allowed_fallback() use in select_fallback_rq()
  cpuset: Honour task_cpu_possible_mask() in guarantee_online_cpus()
  cpuset: Don't use the cpu_possible_mask as a last resort for cgroup v1
  sched: Introduce task_cpu_possible_mask() to limit fallback rq selection
  sched: Cgroup SCHED_IDLE support
  sched/topology: Skip updating masks for non-online nodes
  sched: Replace deprecated CPU-hotplug functions.
  sched: Skip priority checks with SCHED_FLAG_KEEP_PARAMS
  sched: Fix UCLAMP_FLAG_IDLE setting
  sched/deadline: Fix missing clock update in migrate_task_rq_dl()
  sched/fair: Avoid a second scan of target in select_idle_cpu
  sched/fair: Use prev instead of new target as recent_used_cpu
  sched: Don't report SCHED_FLAG_SUGOV in sched_getattr()
  ...

Pull x86 cleanups from Borislav Petkov:
 "The usual round of minor cleanups and fixes"

* tag 'x86_cleanups_for_v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/kaslr: Have process_mem_region() return a boolean
  x86/power: Fix kernel-doc warnings in cpu.c
  x86/mce/inject: Replace deprecated CPU-hotplug functions.
  x86/microcode: Replace deprecated CPU-hotplug functions.
  x86/mtrr: Replace deprecated CPU-hotplug functions.
  x86/mmiotrace: Replace deprecated CPU-hotplug functions.

Pull x86 resource control updates from Borislav Petkov:
 "A first round of changes towards splitting the arch-specific bits from
  the filesystem bits of resctrl, the ultimate goal being to support
  ARM's equivalent technology MPAM, with the same fs interface (James
  Morse)"

* tag 'x86_cache_for_v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (25 commits)
  x86/resctrl: Make resctrl_arch_get_config() return its value
  x86/resctrl: Merge the CDP resources
  x86/resctrl: Expand resctrl_arch_update_domains()'s msr_param range
  x86/resctrl: Remove rdt_cdp_peer_get()
  x86/resctrl: Merge the ctrl_val arrays
  x86/resctrl: Calculate the index from the configuration type
  x86/resctrl: Apply offset correction when config is staged
  x86/resctrl: Make ctrlval arrays the same size
  x86/resctrl: Pass configuration type to resctrl_arch_get_config()
  x86/resctrl: Add a helper to read a closid's configuration
  x86/resctrl: Rename update_domains() to resctrl_arch_update_domains()
  x86/resctrl: Allow different CODE/DATA configurations to be staged
  x86/resctrl: Group staged configuration into a separate struct
  x86/resctrl: Move the schemata names into struct resctrl_schema
  x86/resctrl: Add a helper to read/set the CDP configuration
  x86/resctrl: Swizzle rdt_resource and resctrl_schema in pseudo_lock_region
  x86/resctrl: Pass the schema to resctrl filesystem functions
  x86/resctrl: Add resctrl_arch_get_num_closid()
  x86/resctrl: Store the effective num_closid in the schema
  x86/resctrl: Walk the resctrl schema list instead of an arch list
  ...

Pull x86 build updates from Borislav Petkov:

 - Remove cc-option checks which are old and already supported by the
   minimal compiler version the kernel uses and thus avoid the need to
   invoke the compiler unnecessarily.

 - Cleanups

* tag 'x86_build_for_v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/build: Move the install rule to arch/x86/Makefile
  x86/build: Remove the left-over bzlilo target
  x86/tools/relocs: Mark die() with the printf function attr format
  x86/build: Remove stale cc-option checks

Pull RAS update from Borislav Petkov:
 "A single RAS change for 5.15:

   - Do not start processing MCEs logged early because the decoding
     chain is not up yet - delay that processing until everything is
     ready"

* tag 'ras_core_for_v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mce: Defer processing of early errors

Pull EDAC updates from Borislav Petkov:
 "The usual EDAC stuff which managed to trickle in for 5.15:

   - Add new HBM2 (High Bandwidth Memory Gen 2) type and add support for
     it to the Intel SKx drivers

   - Print additional useful per-channel error information on i10nm,
     like on SKL

   - Don't load the AMD EDAC decoder in virtual images

   - The usual round of fixes and cleanups"

* tag 'edac_updates_for_v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
  EDAC/i10nm: Retrieve and print retry_rd_err_log registers
  EDAC/i10nm: Fix NVDIMM detection
  EDAC/skx_common: Set the memory type correctly for HBM memory
  EDAC/altera: Skip defining unused structures for specific configs
  EDAC/mce_amd: Do not load edac_mce_amd module on guests
  EDAC/mc: Add new HBM2 memory type
  EDAC/amd64: Use DEVICE_ATTR helper macros

Pull s390 updates from Heiko Carstens:

 - Improve ftrace code patching so that stop_machine is not required
   anymore. This requires a small common code patch acked by Steven
   Rostedt:

     https://lore.kernel.org/linux-s390/20210730220741.4da6fdf6@oasis.local.home/

 - Enable KCSAN for s390. This comes with a small common code change to
   fix a compile warning. Acked by Marco Elver:

     https://lore.kernel.org/r/20210729142811.1309391-1-hca@linux.ibm.com

 - Add KFENCE support for s390. This also comes with a minimal x86 patch
   from Marco Elver who said also this can be carried via the s390 tree:

     https://lore.kernel.org/linux-s390/YQJdarx6XSUQ1tFZ@elver.google.com/

 - More changes to prepare the decompressor for relocation.

 - Enable DAT also for CPU restart path.

 - Final set of register asm removal patches; leaving only three
   locations where needed and sane.

 - Add NNPA, Vector-Packed-Decimal-Enhancement Facility 2, PCI MIO
   support to hwcaps flags.

 - Cleanup hwcaps implementation.

 - Add new instructions to in-kernel disassembler.

 - Various QDIO cleanups.

 - Add SCLP debug feature.

 - Various other cleanups and improvements all over the place.

* tag 's390-5.15-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (105 commits)
  s390: remove SCHED_CORE from defconfigs
  s390/smp: do not use nodat_stack for secondary CPU start
  s390/smp: enable DAT before CPU restart callback is called
  s390: update defconfigs
  s390/ap: fix state machine hang after failure to enable irq
  KVM: s390: generate kvm hypercall functions
  s390/sclp: add tracing of SCLP interactions
  s390/debug: add early tracing support
  s390/debug: fix debug area life cycle
  s390/debug: keep debug data on resize
  s390/diag: make restart_part2 a local label
  s390/mm,pageattr: fix walk_pte_level() early exit
  s390: fix typo in linker script
  s390: remove do_signal() prototype and do_notify_resume() function
  s390/crypto: fix all kernel-doc warnings in vfio_ap_ops.c
  s390/pci: improve DMA translation init and exit
  s390/pci: simplify CLP List PCI handling
  s390/pci: handle FH state mismatch only on disable
  s390/pci: fix misleading rc in clp_set_pci_fn()
  s390/boot: factor out offset_vmlinux_info() function
  ...

Pull m68k updates from Geert Uytterhoeven:

 - miscellaneous fixes

 - defconfig updates

* tag 'm68k-for-v5.15-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k:
  m68k: Fix asm register constraints for atomic ops
  m68k: Fix invalid RMW_INSNS on CPUs that lack CAS
  m68k: defconfig: Update defconfigs for v5.14-rc1
  m68k: emu: Fix invalid free in nfeth_cleanup()

Pull crypto updates from Herbert Xu:
 "Algorithms:

   - Add AES-NI/AVX/x86_64 implementation of SM4.

  Drivers:

   - Add Arm SMCCC TRNG based driver"

[ And obviously a lot of random fixes and updates  - Linus]

* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (84 commits)
  crypto: sha512 - remove imaginary and mystifying clearing of variables
  crypto: aesni - xts_crypt() return if walk.nbytes is 0
  padata: Remove repeated verbose license text
  crypto: ccp - Add support for new CCP/PSP device ID
  crypto: x86/sm4 - add AES-NI/AVX2/x86_64 implementation
  crypto: x86/sm4 - export reusable AESNI/AVX functions
  crypto: rmd320 - remove rmd320 in Makefile
  crypto: skcipher - in_irq() cleanup
  crypto: hisilicon - check _PS0 and _PR0 method
  crypto: hisilicon - change parameter passing of debugfs function
  crypto: hisilicon - support runtime PM for accelerator device
  crypto: hisilicon - add runtime PM ops
  crypto: hisilicon - using 'debugfs_create_file' instead of 'debugfs_create_regset32'
  crypto: tcrypt - add GCM/CCM mode test for SM4 algorithm
  crypto: testmgr - Add GCM/CCM mode test of SM4 algorithm
  crypto: tcrypt - Fix missing return value check
  crypto: hisilicon/sec - modify the hardware endian configuration
  crypto: hisilicon/sec - fix the abnormal exiting process
  crypto: qat - store vf.compatible flag
  crypto: qat - do not export adf_iov_putmsg()
  ...

Merge branch 'core-rcu.2021.08.28a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu

Pull RCU updates from Paul McKenney:
 "RCU changes for this cycle were:

   - Documentation updates

   - Miscellaneous fixes

   - Offloaded-callbacks updates

   - Updates to the nolibc library

   - Tasks-RCU updates

   - In-kernel torture-test updates

   - Torture-test scripting, perhaps most notably the pinning of
     torture-test guest OSes so as to force differences in memory
     latency. For example, in a two-socket system, a four-CPU guest OS
     will have one pair of its CPUs pinned to threads in a single core
     on one socket and the other pair pinned to threads in a single core
     on the other socket. This approach proved able to force race
     conditions that earlier testing missed. Some of these race
     conditions are still being tracked down"

* 'core-rcu.2021.08.28a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu: (61 commits)
  torture: Replace deprecated CPU-hotplug functions.
  rcu: Replace deprecated CPU-hotplug functions
  rcu: Print human-readable message for schedule() in RCU reader
  rcu: Explain why rcu_all_qs() is a stub in preemptible TREE RCU
  rcu: Use per_cpu_ptr to get the pointer of per_cpu variable
  rcu: Remove useless "ret" update in rcu_gp_fqs_loop()
  rcu: Mark accesses in tree_stall.h
  rcu: Make rcu_gp_init() and rcu_gp_fqs_loop noinline to conserve stack
  rcu: Mark lockless ->qsmask read in rcu_check_boost_fail()
  srcutiny: Mark read-side data races
  rcu: Start timing stall repetitions after warning complete
  rcu: Do not disable GP stall detection in rcu_cpu_stall_reset()
  rcu/tree: Handle VM stoppage in stall detection
  rculist: Unify documentation about missing list_empty_rcu()
  rcu: Mark accesses to ->rcu_read_lock_nesting
  rcu: Weaken ->dynticks accesses and updates
  rcu: Remove special bit at the bottom of the ->dynticks counter
  rcu: Fix stall-warning deadlock due to non-release of rcu_node ->lock
  rcu: Fix to include first blocked task in stall warning
  torture: Make kvm-test-1-run-qemu.sh check for reboot loops
  ...

Pull file locking updates from Jeff Layton:
 "This starts with a couple of fixes for potential deadlocks in the
  fowner/fasync handling.

  The next patch removes the old mandatory locking code from the kernel
  altogether.

  The last patch cleans up rw_verify_area a bit more after the mandatory
  locking removal"

* tag 'locks-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux:
  fs: clean up after mandatory file locking support removal
  fs: remove mandatory file locking support
  fcntl: fix potential deadlock for &fasync_struct.fa_lock
  fcntl: fix potential deadlocks for &fown_struct.lock

Pull tpm driver updates from Jarkko Sakkinen:
 "The highlights are:

   - Support for signing LKM's with ECDSA keys

   - An integer overflow bug fix in pkey"

* tag 'tpmdd-next-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd:
  crypto: public_key: fix overflow during implicit conversion
  tpm: ibmvtpm: Avoid error message when process gets signal while waiting
  certs: Add support for using elliptic curve keys for signing modules
  certs: Trigger creation of RSA module signing key if it's not an RSA key
  char: tpm: cr50_i2c: convert to new probe interface
  char: tpm: Kconfig: remove bad i2c cr50 select

Pull power supply and reset updates from Sebastian Reichel:
 "Battery/charger related:
   - cros-peripheral-charger: new driver
   - mt6360-charger: new driver
   - simple-battery: support reading chemistry info
   - max17042-battery: add max77849 support
   - sbs-battery: add time_to_empty_now support
   - smb347-charger: prepare USB OTG support
   - rn5t618: add voltage_now support
   - axp288: cleanup & optimizations
   - max17042_battery: cleanups
   - ab8500: cleanups
   - misc minor cleanups and DT binding fixes

  reset related:
   - tps65086-restart: new driver
   - linkstation-poweroff: support NETGEAR ReadyNAS Duo v2"

* tag 'for-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply: (51 commits)
  power: supply: core: Fix parsing of battery chemistry/technology
  power: supply: max17042_battery: log SOC threshold using debug log level
  power: supply: max17042_battery: more robust chip type checks
  power: supply: max17042_battery: fix typo in MAx17042_TOFF
  power: supply: max17042_battery: clean up MAX17055_V_empty
  power: supply: smb347-charger: Implement USB VBUS regulator
  power: supply: smb347-charger: Add missing pin control activation
  power: supply: smb347-charger: Utilize generic regmap caching
  power: supply: smb347-charger: Make smb347_set_writable() IRQ-safe
  dt-bindings: power: supply: smb347-charger: Document USB VBUS regulator
  power: reset: Add TPS65086 restart driver
  dt-bindings: power: supply: max17042: describe interrupt
  power: supply: max17042: remove duplicated STATUS bit defines
  power: supply: max17042: handle fails of reading status register
  power: supply: core: Parse battery chemistry/technology
  dt-bindings: power: Extend battery bindings with chemistry
  power: reset: linkstation-poweroff: add new device
  power: reset: linkstation-poweroff: prepare for new devices
  power: supply: bq24735: reorganize ChargeOption command macros
  power: supply: rn5t618: Add voltage_now property
  ...

Pull spi updates from Mark Brown:
 "A quiet release for SPI, some fixes and a couple of new drivers plus
  one small refactoring:

   - Move the chip select timing configuration from the controller to
     the device to allow a bit more flexibility

   - New drivers for Rockchip SFC and Spreadtrum ADI"

* tag 'spi-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (47 commits)
  spi: spi-zynq-qspi: use wait_for_completion_timeout to make zynq_qspi_exec_mem_op not interruptible
  spi: add sprd ADI for sc9863 and ums512
  spi: Convert sprd ADI bindings to yaml
  spi: sprd: Add ADI r3 support
  spi: sprd: Fix the wrong WDG_LOAD_VAL
  spi: davinci: invoke chipselect callback
  spi: sprd: fill offset only to RD_CMD register for reading from slave device
  spi: sprd: Make sure offset not equal to slave address size
  spi: sprd: Pass offset instead of physical address to adi_read/_write()
  spi: rockchip-sfc: Fix assigned but never used return error codes
  spi: rockchip-sfc: Remove redundant IO operations
  spi: stm32: fix excluded_middle.cocci warnings
  spi: coldfire-qspi: Use clk_disable_unprepare in the remove function
  spi: tegra20-slink: remove spi_master_put() in tegra_slink_remove()
  spi: rockchip-sfc: add rockchip serial flash controller
  spi: rockchip-sfc: Bindings for Rockchip serial flash controller
  spi: orion: Prevent incorrect chip select behaviour
  spi: mxic: add missing braces
  spi: spi-pic32: Fix issue with uninitialized dma_slave_config
  spi: spi-fsl-dspi: Fix issue with uninitialized dma_slave_config
  ...

Pull regulator updates from Mark Brown:
 "A very quiet releases, some fixes and cleanups but not really that
  many of them. There were a couple of new driver specific pieces:

   - Support for controlling the over/under voltage protection on
     BD718xx devices

   - New drivers for Richtek RTQ2134, and RTQ6752"

* tag 'regulator-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator: (37 commits)
  regulator: vctrl: Avoid lockdep warning in enable/disable ops
  regulator: vctrl: Use locked regulator_get_voltage in probe path
  regulator: Documentation fix for regulator error notification helper
  regulator: Minor regulator documentation fixes.
  regulator: sy7636a: Use the regmap directly
  regulator: sy7636a: Store the epd-pwr-good GPIO locally
  regulator: sy7636a: Use the parent driver data
  regulator: sy7636a: Remove the poll_enable_time
  regulator: sy8827n: Enable REGCACHE_FLAT
  regulator: sy8824x: Enable REGCACHE_FLAT
  regulator: rtq2134: Fix coding style
  regulator: hi6421v600: rename voltage range arrays
  regulator: hi6421v600: use lowercase for ldo
  regulator: fixed: use dev_err_probe for register
  regulator: rtq2134: Add support for Richtek RTQ2134 SubPMIC
  regulator: rtq2134: Add binding document for Richtek RTQ2134 SubPMIC
  regulator: Fix a couple of spelling mistakes in Kconfig
  regulator: rtq6752: fix reg reset behavior
  regulator: da9063: Add support for full-current mode.
  regulator: rt6245: make a const array func_base static, makes object smaller
  ...

Pull regmap updates from Mark Brown:
 "A few small fixes for regmaps this time, plus support for allowing
  drivers to select raw spinlocks for the locks in order to allow usage
  in interrutpt controllers"

* tag 'regmap-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:
  regmap: teach regmap to use raw spinlocks if requested in the config
  regmap: allow const array for {devm_,}regmap_field_bulk_alloc reg_fields
  regmap: Prefer unsigned int to bare use of unsigned
  regmap: fix the offset of register error log

Pull fs hole punching vs cache filling race fixes from Jan Kara:
 "Fix races leading to possible data corruption or stale data exposure
  in multiple filesystems when hole punching races with operations such
  as readahead.

  This is the series I was sending for the last merge window but with
  your objection fixed - now filemap_fault() has been modified to take
  invalidate_lock only when we need to create new page in the page cache
  and / or bring it uptodate"

* tag 'hole_punch_for_v5.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  filesystems/locking: fix Malformed table warning
  cifs: Fix race between hole punch and page fault
  ceph: Fix race between hole punch and page fault
  fuse: Convert to using invalidate_lock
  f2fs: Convert to using invalidate_lock
  zonefs: Convert to using invalidate_lock
  xfs: Convert double locking of MMAPLOCK to use VFS helpers
  xfs: Convert to use invalidate_lock
  xfs: Refactor xfs_isilocked()
  ext2: Convert to using invalidate_lock
  ext4: Convert to use mapping->invalidate_lock
  mm: Add functions to lock invalidate_lock for two mappings
  mm: Protect operations adding pages to page cache with invalidate_lock
  documentation: Sync file_operations members with reality
  mm: Fix comments mentioning i_mutex

Pull UDF and isofs updates from Jan Kara:
 "Several smaller fixes and cleanups in UDF and isofs"

* tag 'fs_for_v5.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  udf_get_extendedattr() had no boundary checks.
  isofs: joliet: Fix iocharset=utf8 mount option
  udf: Fix iocharset=utf8 mount option
  udf: Get rid of 0-length arrays in struct fileIdentDesc
  udf: Get rid of 0-length arrays
  udf: Remove unused declaration
  udf: Check LVID earlier

Pull FIEMAP cleanups from Jan Kara:
 "FIEMAP cleanups from Christoph transitioning all remaining filesystems
  supporting FIEMAP (ext2, hpfs) to iomap API and removing the old
  helper"

* tag 'fiemap_for_v5.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  fs: remove generic_block_fiemap
  hpfs: use iomap_fiemap to implement ->fiemap
  ext2: use iomap_fiemap to implement ->fiemap
  ext2: make ext2_iomap_ops available unconditionally

Pull fsnotify updates from Jan Kara:
 "fsnotify speedups when notification actually isn't used and support
  for identifying processes which caused fanotify events through pidfd
  instead of normal pid"

* tag 'fsnotify_for_v5.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  fsnotify: optimize the case of no marks of any type
  fsnotify: count all objects with attached connectors
  fsnotify: count s_fsnotify_inode_refs for attached connectors
  fsnotify: replace igrab() with ihold() on attach connector
  fanotify: add pidfd support to the fanotify API
  fanotify: introduce a generic info record copying helper
  fanotify: minor cosmetic adjustments to fid labels
  kernel/pid.c: implement additional checks upon pidfd_create() parameters
  kernel/pid.c: remove static qualifier from pidfd_create()

As per the long-suffering comment.
Reported-by: Minh Yuan <yuanmingbuaa@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jirislaby@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

The Linux console's VT102 implementation already consumes OSC
("Operating System Command") sequences, probably because that's how
palette changes are transmitted.

In addition to OSC, there are three other major clases of ANSI control
strings: APC ("Application Program Command"), PM ("Privacy Message"),
and DCS ("Device Control String").  They are handled similarly to OSC in
terms of termination.

Source: vt100.net

Add three new enumerated states, one for each of these types.  All three
are handled the same way right now--they simply consume input until
terminated.  I hope to expand upon this firmament in the future.  Add
new predicate ansi_control_string(), returning true for any of these
states.  Replace explicit checks against ESosc with calls to this
function.  Transition to these states appropriately from the escape
initiation (ESesc) state.

This was motivated by the following Notcurses bugs:

 https://github.com/dankamongmen/notcurses/issues/2050
 https://github.com/dankamongmen/notcurses/issues/1828
 https://github.com/dankamongmen/notcurses/issues/2069

where standard VT sequences are not consumed by the Linux console.  It's
not necessary that the Linux console *support* these sequences, but it
ought *consume* these well-specified classes of sequences.

Tested by sending a variety of escape sequences to the console, and
verifying that they still worked, or were now properly consumed.
Verified that the escapes were properly terminated at a generic level.
Verified that the Notcurses tools continued to show expected output on
the Linux console, except now without escape bleedthrough.

Link: https://lore.kernel.org/lkml/YSydL0q8iaUfkphg@schwarzgerat.orthanc/Signed-off-by: nick black <dankamongmen@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jirislaby@kernel.org>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

This just does the "if the architecture does efficient unaligned
handling, start the memcmp using 'unsigned long' accesses", since
Nikolay Borisov found a load that cares.

This is basically the minimal patch, and limited to architectures that
are known to not have slow unaligned handling.  We've had the stupid
byte-at-a-time version forever, and nobody has ever even noticed before,
so let's keep the fix minimal.

A potential further improvement would be to align one of the sources in
order to at least minimize unaligned cases, but the only real case of
bigger memcmp() users seems to be the FIDEDUPERANGE ioctl().  As David
Sterba says, the dedupe ioctl is typically called on ranges spanning
many pages so the common case will all be page-aligned anyway.

All the relevant architectures select HAVE_EFFICIENT_UNALIGNED_ACCESS,
so I'm not going to worry about the combination of a very rare use-case
and a rare architecture until somebody actually hits it.  Particularly
since Nikolay also tested the more complex patch with extra alignment
handling code, and it only added overhead.

Link: https://lore.kernel.org/lkml/20210721135926.602840-1-nborisov@suse.com/Reported-by: Nikolay Borisov <nborisov@suse.com>
Cc: David Sterba <dsterba@suse.cz>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

When new work is added, io_wqe_enqueue() checks if we need to wake or
create a new worker. But that check is done outside the lock that
otherwise synchronizes us with a worker going to sleep, so we can end
up in the following situation:

CPU0				CPU1
lock
insert work
unlock
atomic_read(nr_running) != 0
				lock
				atomic_dec(nr_running)
no wakeup needed

Hold the wqe lock around the "need to wakeup" check. Then we can also get
rid of the temporary work_flags variable, as we know the work will remain
valid as long as we hold the lock.

Cc: stable@vger.kernel.org
Reported-by: Andres Freund <andres@anarazel.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_uring no longer queues async work off completion handlers that run in
hard or soft interrupt context, and that use case was the only reason that
io-wq had to use IRQ safe locks for wqe and worker locks.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

For the two places where new workers are created, we diligently check if
we are allowed to create a new worker. If we're currently at the limit
of how many workers of a given type we can have, then we don't create
any new ones.

If you have a mixed workload with various types of bound and unbounded
work, then it can happen that a worker finishes one type of work and
is then transitioned to the other type. For this case, we don't check
if we are actually allowed to do so. This can cause io-wq to temporarily
exceed the allowed number of workers for a given type.

When retrieving work, check that the types match. If they don't, check
if we are allowed to transition to the other type. If not, then don't
handle the new work.

Cc: stable@vger.kernel.org
Reported-by: Johannes Lundberg <johalun0@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>