Problem statement:

Once the user has disabled turbo frequency by

# echo 1 > /sys/devices/system/cpu/intel_pstate/no_turbo

the cfs_rq's util_avg becomes quite small when compared with
CPU capacity.

Step to reproduce:

# echo 1 > /sys/devices/system/cpu/intel_pstate/no_turbo

# ./x86_cpuload --count 1 --start 3 --timeout 100 --busy 99

would launch 1 thread and bind it to CPU3, lasting for 100 seconds,
with a CPU utilization of 99%. [1]

top result:
%Cpu3  : 98.4 us,  0.0 sy,  0.0 ni,  1.6 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st

check util_avg:
cat /sys/kernel/debug/sched/debug | grep "cfs_rq\[3\]" -A 20 | grep util_avg
  .util_avg                      : 611

So the util_avg/cpu capacity is 611/1024, which is much smaller than
98.4% shown in the top result.

This might impact some logic in the scheduler. For example,
group_is_overloaded() would compare the group_capacity and group_util
in the sched group, to check if this sched group is overloaded or not.
With this gap, even when there is a nearly 100% workload, the sched
group will not be regarded as overloaded. Besides group_is_overloaded(),
there are also other victims. There is a ongoing work that aims to
optimize the task wakeup in a LLC domain. The main idea is to stop
searching idle CPUs if the sched domain is overloaded[2]. This proposal
also relies on the util_avg/CPU capacity to decide whether the LLC
domain is overloaded.

Analysis:

CPU frequency invariance has caused this difference. In summary,
the util_sum of cfs rq would decay quite fast when the CPU is in
idle, when the CPU frequency invariance is enabled.

The detail is as followed:

As depicted in update_rq_clock_pelt(), when the frequency invariance
is enabled, there would be two clock variables on each rq, clock_task
and clock_pelt:

   The clock_pelt scales the time to reflect the effective amount of
   computation done during the running delta time but then syncs back to
   clock_task when rq is idle.

   absolute time    | 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11|12|13|14|15|16
   @ max frequency  ------******---------------******---------------
   @ half frequency ------************---------************---------
   clock pelt       | 1| 2|    3|    4| 7| 8| 9|   10|   11|14|15|16

The fast decay of util_sum during idle is due to:

 1. rq->clock_pelt is always behind rq->clock_task
 2. rq->last_update is updated to rq->clock_pelt' after invoking
    ___update_load_sum()
 3. Then the CPU becomes idle, the rq->clock_pelt' would be suddenly
    increased a lot to rq->clock_task
 4. Enters ___update_load_sum() again, the idle period is calculated by
    rq->clock_task - rq->last_update, AKA, rq->clock_task - rq->clock_pelt'.
    The lower the CPU frequency is, the larger the delta =
    rq->clock_task - rq->clock_pelt' will be. Since the idle period will be
    used to decay the util_sum only, the util_sum drops significantly during
    idle period.

Proposal:

This symptom is not only caused by disabling turbo frequency, but it
would also appear if the user limits the max frequency at runtime.

Because, if the frequency is always lower than the max frequency,
CPU frequency invariance would decay the util_sum quite fast during
idle.

As some end users would disable turbo after boot up, this patch aims to
present this symptom and deals with turbo scenarios for now.

It might be ideal if CPU frequency invariance is aware of the max CPU
frequency (user specified) at runtime in the future.

Link: https://github.com/yu-chen-surf/x86_cpuload.git #1
Link: https://lore.kernel.org/lkml/20220310005228.11737-1-yu.c.chen@intel.com/ #2
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Giovanni Gherdovich <ggherdovich@suse.cz>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

powerpc's asm/prom.h brings some headers that it doesn't
need itself.

In order to clean it up, first add missing headers in
users of asm/prom.h
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

The struct dbs_data embeds a struct gov_attr_set and
the struct gov_attr_set embeds a kobject. Since every kobject must have
a release() method and we can't use kfree() to free it directly,
so introduce cpufreq_dbs_data_release() to release the dbs_data via
the kobject::release() method. This fixes the calltrace like below:

  ODEBUG: free active (active state 0) object type: timer_list hint: delayed_work_timer_fn+0x0/0x34
  WARNING: CPU: 12 PID: 810 at lib/debugobjects.c:505 debug_print_object+0xb8/0x100
  Modules linked in:
  CPU: 12 PID: 810 Comm: sh Not tainted 5.16.0-next-20220120-yocto-standard+ #536
  Hardware name: Marvell OcteonTX CN96XX board (DT)
  pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
  pc : debug_print_object+0xb8/0x100
  lr : debug_print_object+0xb8/0x100
  sp : ffff80001dfcf9a0
  x29: ffff80001dfcf9a0 x28: 0000000000000001 x27: ffff0001464f0000
  x26: 0000000000000000 x25: ffff8000090e3f00 x24: ffff80000af60210
  x23: ffff8000094dfb78 x22: ffff8000090e3f00 x21: ffff0001080b7118
  x20: ffff80000aeb2430 x19: ffff800009e8f5e0 x18: 0000000000000000
  x17: 0000000000000002 x16: 00004d62e58be040 x15: 013590470523aff8
  x14: ffff8000090e1828 x13: 0000000001359047 x12: 00000000f5257d14
  x11: 0000000000040591 x10: 0000000066c1ffea x9 : ffff8000080d15e0
  x8 : ffff80000a1765a8 x7 : 0000000000000000 x6 : 0000000000000001
  x5 : ffff800009e8c000 x4 : ffff800009e8c760 x3 : 0000000000000000
  x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0001474ed040
  Call trace:
   debug_print_object+0xb8/0x100
   __debug_check_no_obj_freed+0x1d0/0x25c
   debug_check_no_obj_freed+0x24/0xa0
   kfree+0x11c/0x440
   cpufreq_dbs_governor_exit+0xa8/0xac
   cpufreq_exit_governor+0x44/0x90
   cpufreq_set_policy+0x29c/0x570
   store_scaling_governor+0x110/0x154
   store+0xb0/0xe0
   sysfs_kf_write+0x58/0x84
   kernfs_fop_write_iter+0x12c/0x1c0
   new_sync_write+0xf0/0x18c
   vfs_write+0x1cc/0x220
   ksys_write+0x74/0x100
   __arm64_sys_write+0x28/0x3c
   invoke_syscall.constprop.0+0x58/0xf0
   do_el0_svc+0x70/0x170
   el0_svc+0x54/0x190
   el0t_64_sync_handler+0xa4/0x130
   el0t_64_sync+0x1a0/0x1a4
  irq event stamp: 189006
  hardirqs last  enabled at (189005): [<ffff8000080849d0>] finish_task_switch.isra.0+0xe0/0x2c0
  hardirqs last disabled at (189006): [<ffff8000090667a4>] el1_dbg+0x24/0xa0
  softirqs last  enabled at (188966): [<ffff8000080106d0>] __do_softirq+0x4b0/0x6a0
  softirqs last disabled at (188957): [<ffff80000804a618>] __irq_exit_rcu+0x108/0x1a4

[ rjw: Because can be freed by the gov_attr_set_put() in
  cpufreq_dbs_governor_exit() now, it is also necessary to put the
  invocation of the governor ->exit() callback into the new
  cpufreq_dbs_data_release() function. ]

Fixes: c4435630 ("cpufreq: governor: New sysfs show/store callbacks for governor tunables")
Signed-off-by: Kevin Hao <haokexin@gmail.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Pull serial driver fix from Greg KH:
 "This is a single serial driver fix for a build issue that showed up
  due to changes that came in through the tty tree in 5.18-rc1 that were
  missed previously. It resolves a build error with the mpc52xx_uart
  driver.

  It has been in linux-next this week with no reported problems"

* tag 'tty-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  tty: serial: mpc52xx_uart: make rx/tx hooks return unsigned, part II.

Pull staging driver fix from Greg KH:
 "Here is a single staging driver fix for 5.18-rc2 that resolves an
  endian issue for the r8188eu driver. It has been in linux-next all
  this week with no reported problems"

* tag 'staging-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: r8188eu: Fix PPPoE tag insertion on little endian systems

Merge tag 'driver-core-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core

Pull driver core updates from Greg KH:
 "Here are two small driver core changes for 5.18-rc2.

  They are the final bits in the removal of the default_attrs field in
  struct kobj_type. I had to wait until after 5.18-rc1 for all of the
  changes to do this came in through different development trees, and
  then one new user snuck in. So this series has two changes:

   - removal of the default_attrs field in the powerpc/pseries/vas code.

     The change has been acked by the PPC maintainers to come through
     this tree

   - removal of default_attrs from struct kobj_type now that all
     in-kernel users are removed.

     This cleans up the kobject code a little bit and removes some
     duplicated functionality that confused people (now there is only
     one way to do default groups)

  Both of these have been in linux-next for all of this week with no
  reported problems"

* tag 'driver-core-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  kobject: kobj_type: remove default_attrs
  powerpc/pseries/vas: use default_groups in kobj_type

Pull char/misc driver fix from Greg KH:
 "A single driver fix. It resolves the build warning issue on 32bit
  systems in the habannalabs driver that came in during the 5.18-rc1
  merge cycle.

  It has been in linux-next for all this week with no reported problems"

* tag 'char-misc-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  habanalabs: Fix test build failures

Pull powerpc fixes from Michael Ellerman:

 - Fix KVM "lost kick" race, where an attempt to pull a vcpu out of the
   guest could be lost (or delayed until the next guest exit).

 - Disable SCV (system call vectored) when PR KVM guests could be run.

 - Fix KVM PR guests using SCV, by disallowing AIL != 0 for KVM PR
   guests.

 - Add a new KVM CAP to indicate if AIL == 3 is supported.

 - Fix a regression when hotplugging a CPU to a memoryless/cpuless node.

 - Make virt_addr_valid() stricter for 64-bit Book3E & 32-bit, which
   fixes crashes seen due to hardened usercopy.

 - Revert a change to max_mapnr which broke HIGHMEM.

Thanks to Christophe Leroy, Fabiano Rosas, Kefeng Wang, Nicholas Piggin,
and Srikar Dronamraju.

* tag 'powerpc-5.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  Revert "powerpc: Set max_mapnr correctly"
  powerpc: Fix virt_addr_valid() for 64-bit Book3E & 32-bit
  KVM: PPC: Move kvmhv_on_pseries() into kvm_ppc.h
  powerpc/numa: Handle partially initialized numa nodes
  powerpc/64: Fix build failure with allyesconfig in book3s_64_entry.S
  KVM: PPC: Use KVM_CAP_PPC_AIL_MODE_3
  KVM: PPC: Book3S PR: Disallow AIL != 0
  KVM: PPC: Book3S PR: Disable SCV when AIL could be disabled
  KVM: PPC: Book3S HV P9: Fix "lost kick" race

Pull irq fixes from Thomas Gleixner:
 "A set of interrupt chip driver fixes:

   - A fix for a long standing bug in the ARM GICv3 redistributor
     polling which uses the wrong bit number to test.

   - Prevent translation of bogus ACPI table entries which map device
     interrupts into the IPI space on ARM GICs.

   - Don't write into the pending register of ARM GICV4 before the scan
     in hardware has completed.

   - A set of build and correctness fixes for the Qualcomm MPM driver"

* tag 'irq-urgent-2022-04-10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/gic, gic-v3: Prevent GSI to SGI translations
  irqchip/gic-v3: Fix GICR_CTLR.RWP polling
  irqchip/gic-v4: Wait for GICR_VPENDBASER.Dirty to clear before descheduling
  irqchip/irq-qcom-mpm: fix return value check in qcom_mpm_init()
  irq/qcom-mpm: Fix build error without MAILBOX

Pull x86 fixes from Borislav Petkov:

 - Fix the MSI message data struct definition

 - Use local labels in the exception table macros to avoid symbol
   conflicts with clang LTO builds

 - A couple of fixes to objtool checking of the relatively newly added
   SLS and IBT code

 - Rename a local var in the WARN* macro machinery to prevent shadowing

* tag 'x86_urgent_for_v5.18_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/msi: Fix msi message data shadow struct
  x86/extable: Prefer local labels in .set directives
  x86,bpf: Avoid IBT objtool warning
  objtool: Fix SLS validation for kcov tail-call replacement
  objtool: Fix IBT tail-call detection
  x86/bug: Prevent shadowing in __WARN_FLAGS
  x86/mm/tlb: Revert retpoline avoidance approach

Pull perf fixes from Borislav Petkov:

 - A couple of fixes to cgroup-related handling of perf events

 - A couple of fixes to event encoding on Sapphire Rapids

 - Pass event caps of inherited events so that perf doesn't fail wrongly
   at fork()

 - Add support for a new Raptor Lake CPU

* tag 'perf_urgent_for_v5.18_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/core: Always set cpuctx cgrp when enable cgroup event
  perf/core: Fix perf_cgroup_switch()
  perf/core: Use perf_cgroup_info->active to check if cgroup is active
  perf/core: Don't pass task around when ctx sched in
  perf/x86/intel: Update the FRONTEND MSR mask on Sapphire Rapids
  perf/x86/intel: Don't extend the pseudo-encoding to GP counters
  perf/core: Inherit event_caps
  perf/x86/uncore: Add Raptor Lake uncore support
  perf/x86/msr: Add Raptor Lake CPU support
  perf/x86/cstate: Add Raptor Lake support
  perf/x86: Add Intel Raptor Lake support

Pull locking fixes from Borislav Petkov:

 - Allow the compiler to optimize away unused percpu accesses and change
   the local_lock_* macros back to inline functions

 - A couple of fixes to static call insn patching

* tag 'locking_urgent_for_v5.18_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  Revert "mm/page_alloc: mark pagesets as __maybe_unused"
  Revert "locking/local_lock: Make the empty local_lock_*() function a macro."
  x86/percpu: Remove volatile from arch_raw_cpu_ptr().
  static_call: Remove __DEFINE_STATIC_CALL macro
  static_call: Properly initialise DEFINE_STATIC_CALL_RET0()
  static_call: Don't make __static_call_return0 static
  x86,static_call: Fix __static_call_return0 for i386

Pull scheduler fixes from Borislav Petkov:

 - Use the correct static key checking primitive on the IRQ exit path

 - Two fixes for the new forceidle balancer

* tag 'sched_urgent_for_v5.18_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  entry: Fix compile error in dynamic_irqentry_exit_cond_resched()
  sched: Teach the forced-newidle balancer about CPU affinity limitation.
  sched/core: Fix forceidle balancing

Merge tag 'perf-tools-fixes-for-v5.18-2022-04-09' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull perf tools fixes from Arnaldo Carvalho de Melo:

 - Fix the clang command line option probing and remove some options to
   filter out, fixing the build with the latest clang versions

 - Fix 'perf bench' futex and epoll benchmarks to deal with machines
   with more than 1K CPUs

 - Fix 'perf test tsc' error message when not supported

 - Remap perf ring buffer if there is no space for event, fixing perf
   usage in 32-bit ChromeOS

 - Drop objdump stderr to avoid getting stuck waiting for stdout output
   in 'perf annotate'

 - Fix up garbled output by now showing unwind error messages when
   augmenting frame in best effort mode

 - Fix perf's libperf_print callback, use the va_args eprintf() variant

 - Sync vhost and arm64 cputype headers with the kernel sources

 - Fix 'perf report --mem-mode' with ARM SPE

 - Add missing external commands ('iiostat', etc) to 'perf --list-cmds'

* tag 'perf-tools-fixes-for-v5.18-2022-04-09' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  perf annotate: Drop objdump stderr to avoid getting stuck waiting for stdout output
  perf tools: Add external commands to list-cmds
  perf docs: Add perf-iostat link to manpages
  perf session: Remap buf if there is no space for event
  perf bench: Fix epoll bench to correct usage of affinity for machines with #CPUs > 1K
  perf bench: Fix futex bench to correct usage of affinity for machines with #CPUs > 1K
  perf tools: Fix perf's libperf_print callback
  perf: arm-spe: Fix perf report --mem-mode
  perf unwind: Don't show unwind error messages when augmenting frame pointer stack
  tools headers arm64: Sync arm64's cputype.h with the kernel sources
  perf test tsc: Fix error message when not supported
  perf build: Don't use -ffat-lto-objects in the python feature test when building with clang-13
  perf python: Fix probing for some clang command line options
  tools build: Filter out options and warnings not supported by clang
  tools build: Use $(shell ) instead of `` to get embedded libperl's ccopts
  tools include UAPI: Sync linux/vhost.h with the kernel sources

Pull cxl and nvdimm fixes from Dan Williams:

 - Fix a compile error in the nvdimm unit tests

 - Fix a shadowed variable warning in the CXL PCI driver

* tag 'cxl+nvdimm-for-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  cxl/pci: Drop shadowed variable
  tools/testing/nvdimm: Fix security_init() symbol collision

Pull gpio fix from Bartosz Golaszewski:

 - fix a race condition with consumers accessing the fields of GPIO IRQ
   chips before they're fully initialized

* tag 'gpio-fixes-for-v5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpio: Restrict usage of GPIO chip irq members before initialization

Merge tag 'irqchip-fixes-5.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms into irq/urgent

Pull irqchip fixes from Marc Zyngier:

 - Fix GICv3 polling for RWP in redistributors

 - Reject ACPI attempts to use SGIs on GIC/GICv3

 - Fix unpredictible behaviour when making a VPE non-resident
   with GICv4

 - A couple of fixes for the newly merged qcom-mpm driver

Link: https://lore.kernel.org/lkml/20220409094229.267649-1-maz@kernel.org

If objdump writes to stderr it can block waiting for it to be read. As
perf doesn't read stderr then progress stops with perf waiting for
stdout output.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexandre Truong <alexandre.truong@arm.com>
Cc: Dave Marchevsky <davemarchevsky@fb.com>
Cc: Denis Nikitin <denik@chromium.org>
Cc: German Gomez <german.gomez@arm.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Lexi Shao <shaolexi@huawei.com>
Cc: Li Huafei <lihuafei1@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin Liška <mliska@suse.cz>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Remi Bernon <rbernon@codeweavers.com>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Cc: Will Deacon <will@kernel.org>
Cc: William Cohen <wcohen@redhat.com>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lore.kernel.org/lkml/20220407230503.1265036-2-irogers@google.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

The `perf --list-cmds` output prints only internal commands, although
there is no reason for that from users' perspective.

Adding the external commands to commands array with NULL function
pointer allows printing all perf commands while not changing the logic
of command handler selection.
Signed-off-by: Michael Petlan <mpetlan@redhat.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20220404221541.30312-2-mpetlan@redhat.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Signed-off-by: Michael Petlan <mpetlan@redhat.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20220404221541.30312-1-mpetlan@redhat.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

If a perf event doesn't fit into remaining buffer space return NULL to
remap buf and fetch the event again.

Keep the logic to error out on inadequate input from fuzzing.

This fixes perf failing on ChromeOS (with 32b userspace):

  $ perf report -v -i perf.data
  ...
  prefetch_event: head=0x1fffff8 event->header_size=0x30, mmap_size=0x2000000: fuzzed or compressed perf.data?
  Error:
  failed to process sample

Fixes: 57fc032a ("perf session: Avoid infinite loop when seeing invalid header.size")
Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: Denis Nikitin <denik@chromium.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey Budankov <alexey.budankov@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220330031130.2152327-1-denik@chromium.orgSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Pull SCSI fixes from James Bottomley:

 - add support for new devices (ufs, mvsas)

 - a major set of fixes in lpfc

 - get rid of a driver specific ioctl in pcmraid

 - a major rework of aha152x to get rid of the scsi_pointer.

 - minor fixes and obvious changes including several spelling updates.

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (36 commits)
  scsi: megaraid_sas: Target with invalid LUN ID is deleted during scan
  scsi: ufs: ufshpb: Fix a NULL check on list iterator
  scsi: sd: Clean up gendisk if device_add_disk() failed
  scsi: message: fusion: Remove redundant variable dmp
  scsi: mvsas: Add PCI ID of RocketRaid 2640
  scsi: sd: sd_read_cpr() requires VPD pages
  scsi: mpt3sas: Fail reset operation if config request timed out
  scsi: sym53c500_cs: Stop using struct scsi_pointer
  scsi: ufs: ufs-pci: Add support for Intel MTL
  scsi: mpt3sas: Fix mpt3sas_check_same_4gb_region() kdoc comment
  scsi: scsi_debug: Fix sdebug_blk_mq_poll() in_use_bm bitmap use
  scsi: bnx2i: Fix spelling mistake "mis-match" -> "mismatch"
  scsi: bnx2fc: Fix spelling mistake "mis-match" -> "mismatch"
  scsi: zorro7xx: Fix a resource leak in zorro7xx_remove_one()
  scsi: aic7xxx: Use standard PCI subsystem, subdevice defines
  scsi: ufs: qcom: Drop custom Android boot parameters
  scsi: core: sysfs: Remove comments that conflict with the actual logic
  scsi: hisi_sas: Remove stray fallthrough annotation
  scsi: virtio-scsi: Eliminate anonymous module_init & module_exit
  scsi: isci: Fix spelling mistake "doesnt" -> "doesn't"
  ...

The 'perf bench epoll' testcase fails on systems with more than 1K CPUs.

Testcase: perf bench epoll all

Result snippet:
<<>>
Run summary [PID 106497]: 1399 threads monitoring on 64 file-descriptors for 8 secs.

perf: pthread_create: No such file or directory
<<>>

In epoll benchmarks (ctl, wait) pthread_create is invoked in do_threads
from respective bench_epoll_*  function. Though the logs shows direct
failure from pthread_create, the actual failure is from
"sched_setaffinity" returning EINVAL (invalid argument).

This happens because the default mask size in glibc is 1024. To overcome
this 1024 CPUs mask size limitation of cpu_set_t, change the mask size
using the CPU_*_S macros.

Patch addresses this by fixing all the epoll benchmarks to use CPU_ALLOC
to allocate cpumask, CPU_ALLOC_SIZE for size, and CPU_SET_S to set the
mask.
Reported-by: Disha Goel <disgoel@linux.vnet.ibm.com>
Signed-off-by: Athira Jajeev <atrajeev@linux.vnet.ibm.com>
Tested-by: Disha Goel <disgoel@linux.vnet.ibm.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nageswara R Sastry <rnsastry@linux.ibm.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: linuxppc-dev@lists.ozlabs.org
Link: https://lore.kernel.org/r/20220406175113.87881-3-atrajeev@linux.vnet.ibm.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

The 'perf bench futex' testcase fails on systems with more than 1K CPUs.

Testcase: perf bench futex all

Failure snippet:
<<>>Running futex/hash benchmark...

perf: pthread_create: No such file or directory
<<>>

All the futex benchmarks (ie hash, lock-api, requeue, wake,
wake-parallel), pthread_create is invoked in respective bench_futex_*
function. Though the logs shows direct failure from pthread_create,
strace logs showed that actual failure is from  "sched_setaffinity"
returning EINVAL (invalid argument).

This happens because the default mask size in glibc is 1024. To overcome
this 1024 CPUs mask size limitation of cpu_set_t, change the mask size
using the CPU_*_S macros.

Patch addresses this by fixing all the futex benchmarks to use CPU_ALLOC
to allocate cpumask, CPU_ALLOC_SIZE for size, and CPU_SET_S to set the
mask.
Reported-by: Disha Goel <disgoel@linux.vnet.ibm.com>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Athira Jajeev <atrajeev@linux.vnet.ibm.com>
Tested-by: Disha Goel <disgoel@linux.vnet.ibm.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nageswara R Sastry <rnsastry@linux.ibm.com>
Cc: linuxppc-dev@lists.ozlabs.org
Link: https://lore.kernel.org/r/20220406175113.87881-2-atrajeev@linux.vnet.ibm.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

eprintf() does not expect va_list as the type of the 4th parameter.

Use veprintf() because it does.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Fixes: 428dab81 ("libperf: Merge libperf_set_print() into libperf_init()")
Cc: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20220408132625.2451452-1-adrian.hunter@intel.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Since commit bb30acae ("perf report: Bail out --mem-mode if mem
info is not available") "perf mem report" and "perf report --mem-mode"
don't allow opening the file unless one of the events has
PERF_SAMPLE_DATA_SRC set.

SPE doesn't have this set even though synthetic memory data is generated
after it is decoded. Fix this issue by setting DATA_SRC on SPE events.
This has no effect on the data collected because the SPE driver doesn't
do anything with that flag and doesn't generate samples.

Fixes: bb30acae ("perf report: Bail out --mem-mode if mem info is not available")
Signed-off-by: James Clark <james.clark@arm.com>
Tested-by: Leo Yan <leo.yan@linaro.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: German Gomez <german.gomez@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20220408144056.1955535-1-james.clark@arm.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Commit Fixes: b9f6fbb3 ("perf arm64: Inject missing frames when
using 'perf record --call-graph=fp'") intended to add a 'best effort'
DWARF unwind that improved the frame pointer stack in most scenarios.

It's expected that the unwind will fail sometimes, but this shouldn't be
reported as an error. It only works when the return address can be
determined from the contents of the link register alone.

Fix the error shown when the unwinder requires extra registers by adding
a new flag that suppresses error messages. This flag is not set in the
normal --call-graph=dwarf unwind mode so that behavior is not changed.

Fixes: b9f6fbb3 ("perf arm64: Inject missing frames when using 'perf record --call-graph=fp'")
Reported-by: John Garry <john.garry@huawei.com>
Signed-off-by: James Clark <james.clark@arm.com>
Tested-by: John Garry <john.garry@huawei.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexandre Truong <alexandre.truong@arm.com>
Cc: German Gomez <german.gomez@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220406145651.1392529-1-james.clark@arm.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

To get the changes in:

  83bea32a ("arm64: Add part number for Arm Cortex-A78AE")

That addresses this perf build warning:

  Warning: Kernel ABI header at 'tools/arch/arm64/include/asm/cputype.h' differs from latest version at 'arch/arm64/include/asm/cputype.h'
  diff -u tools/arch/arm64/include/asm/cputype.h arch/arm64/include/asm/cputype.h

Cc: Ali Saidi <alisaidi@amazon.com>
Cc: Andrew Kilroy <andrew.kilroy@arm.com>
Cc: Chanho Park <chanho61.park@samsung.com>
Cc: German Gomez <german.gomez@arm.com>
Cc: James Clark <james.clark@arm.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Will Deacon <will@kernel.org>
Link: http://lore.kernel.org/lkml/Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

By default `perf test tsc` does not return the error message when the
child process detected kernel does not support it. Instead, the child
process prints an error message to stderr, unfortunately stderr is
redirected to /dev/null when verbose <= 0.

This patch does:

- return TEST_SKIP to the parent process instead of TEST_OK when
  perf_read_tsc_conversion() is not supported.

- Add a new subtest of testing if TSC is supported on current
  architecture by moving exist code to a separate function.
  It avoids two places in test__perf_time_to_tsc() that return
  TEST_SKIP by doing this.

- Extend the test suite definition to contain above two subtests.
  Current test_suite and test_case structs do not support printing skip
  reason when the number of subtest less than 1. To print skip reason, it
  is necessary to extend current test suite definition.
Reviewed-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chengdong Li <chengdongli@tencent.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: likexu@tencent.com
Link: https://lore.kernel.org/r/20220408084748.43707-1-chengdongli@tencent.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Using -ffat-lto-objects in the python feature test when building with
clang-13 results in:

  clang-13: error: optimization flag '-ffat-lto-objects' is not supported [-Werror,-Wignored-optimization-argument]
  error: command '/usr/sbin/clang' failed with exit code 1
  cp: cannot stat '/tmp/build/perf/python_ext_build/lib/perf*.so': No such file or directory
  make[2]: *** [Makefile.perf:639: /tmp/build/perf/python/perf.so] Error 1

Noticed when building on a docker.io/library/archlinux:base container.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Fangrui Song <maskray@google.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Keeping <john@metanate.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

The clang compiler complains about some options even without a source
file being available, while others require one, so use the simple
tools/build/feature/test-hello.c file.

Then check for the "is not supported" string in its output, in addition
to the "unknown argument" already being looked for.

This was noticed when building with clang-13 where -ffat-lto-objects
isn't supported and since we were looking just for "unknown argument"
and not providing a source code to clang, was mistakenly assumed as
being available and not being filtered to set of command line options
provided to clang, leading to a build failure.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Fangrui Song <maskray@google.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Keeping <john@metanate.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Link: http://lore.kernel.org/lkml/Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

These make the feature check fail when using clang, so remove them just
like is done in tools/perf/Makefile.config to build perf itself.

Adding -Wno-compound-token-split-by-macro to tools/perf/Makefile.config
when building with clang is also necessary to avoid these warnings
turned into errors (-Werror):

    CC      /tmp/build/perf/util/scripting-engines/trace-event-perl.o
  In file included from util/scripting-engines/trace-event-perl.c:35:
  In file included from /usr/lib64/perl5/CORE/perl.h:4085:
  In file included from /usr/lib64/perl5/CORE/hv.h:659:
  In file included from /usr/lib64/perl5/CORE/hv_func.h:34:
  In file included from /usr/lib64/perl5/CORE/sbox32_hash.h:4:
  /usr/lib64/perl5/CORE/zaphod32_hash.h:150:5: error: '(' and '{' tokens introducing statement expression appear in different macro expansion contexts [-Werror,-Wcompound-token-split-by-macro]
      ZAPHOD32_SCRAMBLE32(state[0],0x9fade23b);
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  /usr/lib64/perl5/CORE/zaphod32_hash.h:80:38: note: expanded from macro 'ZAPHOD32_SCRAMBLE32'
  #define ZAPHOD32_SCRAMBLE32(v,prime) STMT_START {  \
                                       ^~~~~~~~~~
  /usr/lib64/perl5/CORE/perl.h:737:29: note: expanded from macro 'STMT_START'
  #   define STMT_START   (void)( /* gcc supports "({ STATEMENTS; })" */
                                ^
  /usr/lib64/perl5/CORE/zaphod32_hash.h:150:5: note: '{' token is here
      ZAPHOD32_SCRAMBLE32(state[0],0x9fade23b);
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  /usr/lib64/perl5/CORE/zaphod32_hash.h:80:49: note: expanded from macro 'ZAPHOD32_SCRAMBLE32'
  #define ZAPHOD32_SCRAMBLE32(v,prime) STMT_START {  \
                                                  ^
  /usr/lib64/perl5/CORE/zaphod32_hash.h:150:5: error: '}' and ')' tokens terminating statement expression appear in different macro expansion contexts [-Werror,-Wcompound-token-split-by-macro]
      ZAPHOD32_SCRAMBLE32(state[0],0x9fade23b);
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  /usr/lib64/perl5/CORE/zaphod32_hash.h:87:41: note: expanded from macro 'ZAPHOD32_SCRAMBLE32'
      v ^= (v>>23);                       \
                                          ^
  /usr/lib64/perl5/CORE/zaphod32_hash.h:150:5: note: ')' token is here
      ZAPHOD32_SCRAMBLE32(state[0],0x9fade23b);
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  /usr/lib64/perl5/CORE/zaphod32_hash.h:88:3: note: expanded from macro 'ZAPHOD32_SCRAMBLE32'
  } STMT_END
    ^~~~~~~~
  /usr/lib64/perl5/CORE/perl.h:738:21: note: expanded from macro 'STMT_END'
  #   define STMT_END     )
                          ^

Please refer to the discussion on the Link: tag below, where Nathan
clarifies the situation:

<quote>
acme> And then get to the problems at the end of this message, which seem
acme> similar to the problem described here:
acme>
acme> From  Nathan Chancellor <>
acme> Subject	[PATCH] mwifiex: Remove unnecessary braces from HostCmd_SET_SEQ_NO_BSS_INFO
acme>
acme> https://lkml.org/lkml/2020/9/1/135
acme>
acme> So perhaps in this case its better to disable that
acme> -Werror,-Wcompound-token-split-by-macro when building with clang?

Yes, I think that is probably the best solution. As far as I can tell,
at least in this file and context, the warning appears harmless, as the
"create a GNU C statement expression from two different macros" is very
much intentional, based on the presence of PERL_USE_GCC_BRACE_GROUPS.
The warning is fixed in upstream Perl by just avoiding creating GNU C
statement expressions using STMT_START and STMT_END:

  https://github.com/Perl/perl5/issues/18780
  https://github.com/Perl/perl5/pull/18984

If I am reading the source code correctly, an alternative to disabling
the warning would be specifying -DPERL_GCC_BRACE_GROUPS_FORBIDDEN but it
seems like that might end up impacting more than just this site,
according to the issue discussion above.
</quote>
Based-on-a-patch-by: Sedat Dilek <sedat.dilek@gmail.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com> # Debian/Selfmade LLVM-14 (x86-64)
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Fangrui Song <maskray@google.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Keeping <john@metanate.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Link: http://lore.kernel.org/lkml/YkxWcYzph5pC1EK8@kernel.orgSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Just like its done for ldopts and for both in tools/perf/Makefile.config.

Using `` to initialize PERL_EMBED_CCOPTS somehow precludes using:

  $(filter-out SOMETHING_TO_FILTER,$(PERL_EMBED_CCOPTS))

And we need to do it to allow for building with versions of clang where
some gcc options selected by distros are not available.

Tested-by: Sedat Dilek <sedat.dilek@gmail.com> # Debian/Selfmade LLVM-14 (x86-64)
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Fangrui Song <maskray@google.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Keeping <john@metanate.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Link: http://lore.kernel.org/lkml/YktYX2OnLtyobRYD@kernel.orgSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

To get the changes in:

  b04d910a ("vdpa: support exposing the count of vqs to userspace")
  a61280dd ("vdpa: support exposing the config size to userspace")

Silencing this perf build warning:

  Warning: Kernel ABI header at 'tools/include/uapi/linux/vhost.h' differs from latest version at 'include/uapi/linux/vhost.h'
  diff -u tools/include/uapi/linux/vhost.h include/uapi/linux/vhost.h

  $ diff -u tools/include/uapi/linux/vhost.h include/uapi/linux/vhost.h
  --- tools/include/uapi/linux/vhost.h	2021-07-15 16:17:01.840818309 -0300
  +++ include/uapi/linux/vhost.h	2022-04-02 18:55:05.702522387 -0300
  @@ -150,4 +150,11 @@
   /* Get the valid iova range */
   #define VHOST_VDPA_GET_IOVA_RANGE	_IOR(VHOST_VIRTIO, 0x78, \
   					     struct vhost_vdpa_iova_range)
  +
  +/* Get the config size */
  +#define VHOST_VDPA_GET_CONFIG_SIZE	_IOR(VHOST_VIRTIO, 0x79, __u32)
  +
  +/* Get the count of all virtqueues */
  +#define VHOST_VDPA_GET_VQS_COUNT	_IOR(VHOST_VIRTIO, 0x80, __u32)
  +
   #endif
  $ tools/perf/trace/beauty/vhost_virtio_ioctl.sh > before
  $ cp include/uapi/linux/vhost.h tools/include/uapi/linux/vhost.h
  $ tools/perf/trace/beauty/vhost_virtio_ioctl.sh > after
  $ diff -u before after
  --- before	2022-04-04 14:52:25.036375145 -0300
  +++ after	2022-04-04 14:52:31.906549976 -0300
  @@ -38,4 +38,6 @@
   	[0x73] = "VDPA_GET_CONFIG",
   	[0x76] = "VDPA_GET_VRING_NUM",
   	[0x78] = "VDPA_GET_IOVA_RANGE",
  +	[0x79] = "VDPA_GET_CONFIG_SIZE",
  +	[0x80] = "VDPA_GET_VQS_COUNT",
   };
  $

Cc: Longpeng <longpeng2@huawei.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Link: https://lore.kernel.org/lkml/YksxoFcOARk%2Fldev@kernel.orgSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Pull block fixes from Jens Axboe:
 "Nothing major in here, just a few small fixes:

   - Small series of neglected drbd patches (Christoph, Lv, Xiaomeng)

   - Remove dead variable in cdrom (Enze)"

* tag 'block-5.18-2022-04-08' of git://git.kernel.dk/linux-block:
  drbd: set QUEUE_FLAG_STABLE_WRITES
  drbd: fix an invalid memory access caused by incorrect use of list iterator
  drbd: Fix five use after free bugs in get_initial_state
  cdrom: remove unused variable

Pull io_uring fixes from Jens Axboe:
 "A bit bigger than usual post merge window, largely due to a revert and
  a fix of at what point files are assigned for requests.

  The latter fixing a linked request use case where a dependent link can
  rely on what file is assigned consistently.

  Summary:

   - 32-bit compat fix for IORING_REGISTER_IOWQ_AFF (Eugene)

   - File assignment fixes (me)

   - Revert of the NAPI poll addition from this merge window. The author
     isn't available right now to engage on this, so let's revert it and
     we can retry for the 5.19 release (me, Jakub)

   - Fix a timeout removal race (me)

   - File update and SCM fixes (Pavel)"

* tag 'io_uring-5.18-2022-04-08' of git://git.kernel.dk/linux-block:
  io_uring: fix race between timeout flush and removal
  io_uring: use nospec annotation for more indexes
  io_uring: zero tag on rsrc removal
  io_uring: don't touch scm_fp_list after queueing skb
  io_uring: nospec index for tags on files update
  io_uring: implement compat handling for IORING_REGISTER_IOWQ_AFF
  Revert "io_uring: Add support for napi_busy_poll"
  io_uring: drop the old style inflight file tracking
  io_uring: defer file assignment
  io_uring: propagate issue_flags state down to file assignment
  io_uring: move read/write file prep state into actual opcode handler
  io_uring: defer splice/tee file validity check until command issue
  io_uring: don't check req->file in io_fsync_prep()

Pull rdma fixes from Jason Gunthorpe:
 "Several bug fixes for old bugs:

   - Welcome Leon as co-maintainer for RDMA so we are back to having two
     people

   - Some corner cases are fixed in mlx5's MR code

   - Long standing CM bug where a DREQ at the wrong time can result in a
     long timeout

   - Missing locking and refcounting in hf1"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
  RDMA/hfi1: Fix use-after-free bug for mm struct
  IB/rdmavt: add lock to call to rvt_error_qp to prevent a race condition
  IB/cm: Cancel mad on the DREQ event when the state is MRA_REP_RCVD
  RDMA/mlx5: Add a missing update of cache->last_add
  RDMA/mlx5: Don't remove cache MRs when a delay is needed
  MAINTAINERS: Update qib and hfi1 related drivers
  MAINTAINERS: Add Leon Romanovsky to RDMA maintainers

Pull ACPI updates from Rafael Wysocki:
 "These revert a problematic commit from the 5.17 development cycle and
  finalize the elimination of acpi_bus_get_device() that mostly took
  place during the recent merge window.

  Specifics:

   - Revert an ACPI processor driver change related to cache
     invalidation in acpi_idle_play_dead() that clearly was a mistake
     and introduced user-visible regressions (Akihiko Odaki).

   - Replace the last instance of acpi_bus_get_device() added during the
     recent merge window and drop the function to prevent more users of
     it from being added (Rafael Wysocki)"

* tag 'acpi-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: bus: Eliminate acpi_bus_get_device()
  Revert "ACPI: processor: idle: Only flush cache on entering C3"

Merge tag 'linux-kselftest-kunit-fixes-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest

Pull KUnit fix from Shuah Khan:
 "A single documentation fix to incorrect and outdated usage
  information"

* tag 'linux-kselftest-kunit-fixes-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
  Documentation: kunit: fix path to .kunitconfig in start.rst