The capable() hook returns an error number. -EPERM is actually the same as
-1, so this doesn't make a difference in behavior.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

Someone might write a ruleset like the following, expecting that it
securely constrains UID 1 to UIDs 1, 2 and 3:

    1:2
    1:3

However, because no constraints are applied to UIDs 2 and 3, an attacker
with UID 1 can simply first switch to UID 2, then switch to any UID from
there. The secure way to write this ruleset would be:

    1:2
    1:3
    2:2
    3:3

, which uses "transition to self" as a way to inhibit the default-allow
policy without allowing anything specific.

This is somewhat unintuitive. To make sure that policy authors don't
accidentally write insecure policies because of this, let the kernel verify
that a new ruleset does not contain any entries that are constrained, but
transitively unconstrained.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

For debugging a running system, it is very helpful to be able to see what
policy the system is using. Add a read handler that can dump out a copy of
the loaded policy.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

The current API of the SafeSetID LSM uses one write() per rule, and applies
each written rule instantly. This has several downsides:

 - While a policy is being loaded, once a single parent-child pair has been
   loaded, the parent is restricted to that specific child, even if
   subsequent rules would allow transitions to other child UIDs. This means
   that during policy loading, set*uid() can randomly fail.
 - To replace the policy without rebooting, it is necessary to first flush
   all old rules. This creates a time window in which no constraints are
   placed on the use of CAP_SETUID.
 - If we want to perform sanity checks on the final policy, this requires
   that the policy isn't constructed in a piecemeal fashion without telling
   the kernel when it's done.

Other kernel APIs - including things like the userns code and netfilter -
avoid this problem by performing updates atomically. Luckily, SafeSetID
hasn't landed in a stable (upstream) release yet, so maybe it's not too
late to completely change the API.

The new API for SafeSetID is: If you want to change the policy, open
"safesetid/whitelist_policy" and write the entire policy,
newline-delimited, in there.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

Looking at current_cred() in write handlers is bad form, stop doing that.

Also, let's just require that the write is coming from the initial user
namespace. Especially SAFESETID_WHITELIST_FLUSH requires privilege over all
namespaces, and SAFESETID_WHITELIST_ADD should probably require it as well.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

In preparation for changing the policy parsing logic, refactor the line
parsing logic to be less verbose and move it into a separate function.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

At the moment, safesetid_security_capable() has two nested conditional
blocks, and one big comment for all the logic. Chop it up and reduce the
amount of indentation.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

parent_kuid and child_kuid are kuids, there is no reason to make them
uint64_t. (And anyway, in the kernel, the normal name for that would be
u64, not uint64_t.)

check_setuid_policy_hashtable_key() and
check_setuid_policy_hashtable_key_value() are basically the same thing,
merge them.

Also fix the comment that claimed that (1<<8)==128.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

With the old code, when a process with the (real,effective,saved) UID set
(1,1,1) calls setresuid(2,3,4), safesetid_task_fix_setuid() only checks
whether the transition 1->2 is permitted; the transitions 1->3 and 1->4 are
not checked. Fix this.

This is also a good opportunity to refactor safesetid_task_fix_setuid() to
be less verbose - having one branch per set*uid() syscall is unnecessary.

Note that this slightly changes semantics: The UID transition check for
UIDs that were not in the old cred struct is now always performed against
the policy of the RUID. I think that's more consistent anyway, since the
RUID is also the one that decides whether any policy is enforced at all.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

Fix the pr_warn() calls in the SafeSetID LSM to have newlines at the end.
Without this, denial messages will be buffered as incomplete lines in
log_output(), and will then only show up once something else prints into
dmesg.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>

Pull HMM updates from Jason Gunthorpe:
 "Improvements and bug fixes for the hmm interface in the kernel:

   - Improve clarity, locking and APIs related to the 'hmm mirror'
     feature merged last cycle. In linux-next we now see AMDGPU and
     nouveau to be using this API.

   - Remove old or transitional hmm APIs. These are hold overs from the
     past with no users, or APIs that existed only to manage cross tree
     conflicts. There are still a few more of these cleanups that didn't
     make the merge window cut off.

   - Improve some core mm APIs:
       - export alloc_pages_vma() for driver use
       - refactor into devm_request_free_mem_region() to manage
         DEVICE_PRIVATE resource reservations
       - refactor duplicative driver code into the core dev_pagemap
         struct

   - Remove hmm wrappers of improved core mm APIs, instead have drivers
     use the simplified API directly

   - Remove DEVICE_PUBLIC

   - Simplify the kconfig flow for the hmm users and core code"

* tag 'for-linus-hmm' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (42 commits)
  mm: don't select MIGRATE_VMA_HELPER from HMM_MIRROR
  mm: remove the HMM config option
  mm: sort out the DEVICE_PRIVATE Kconfig mess
  mm: simplify ZONE_DEVICE page private data
  mm: remove hmm_devmem_add
  mm: remove hmm_vma_alloc_locked_page
  nouveau: use devm_memremap_pages directly
  nouveau: use alloc_page_vma directly
  PCI/P2PDMA: use the dev_pagemap internal refcount
  device-dax: use the dev_pagemap internal refcount
  memremap: provide an optional internal refcount in struct dev_pagemap
  memremap: replace the altmap_valid field with a PGMAP_ALTMAP_VALID flag
  memremap: remove the data field in struct dev_pagemap
  memremap: add a migrate_to_ram method to struct dev_pagemap_ops
  memremap: lift the devmap_enable manipulation into devm_memremap_pages
  memremap: pass a struct dev_pagemap to ->kill and ->cleanup
  memremap: move dev_pagemap callbacks into a separate structure
  memremap: validate the pagemap type passed to devm_memremap_pages
  mm: factor out a devm_request_free_mem_region helper
  mm: export alloc_pages_vma
  ...

Merge tag 'ecryptfs-5.3-rc1-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tyhicks/ecryptfs

Pull eCryptfs updates from Tyler Hicks:

 - Fix error handling when ecryptfs_read_lower() encounters an error

 - Fix read-only file creation when the eCryptfs mount is configured to
   store metadata in xattrs

 - Minor code cleanups

* tag 'ecryptfs-5.3-rc1-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tyhicks/ecryptfs:
  ecryptfs: Change return type of ecryptfs_process_flags
  ecryptfs: Make ecryptfs_xattr_handler static
  ecryptfs: remove unnessesary null check in ecryptfs_keyring_auth_tok_for_sig
  ecryptfs: use print_hex_dump_bytes for hexdump
  eCryptfs: fix permission denied with ecryptfs_xattr mount option when create readonly file
  ecryptfs: re-order a condition for static checkers
  eCryptfs: fix a couple type promotion bugs

Pull UBIFS updates from Richard Weinberger:

 - Support for zstd compression

 - Support for offline signed filesystems

 - Various fixes for regressions

* tag 'upstream-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/ubifs:
  ubifs: Don't leak orphans on memory during commit
  ubifs: Check link count of inodes when killing orphans.
  ubifs: Add support for zstd compression.
  ubifs: support offline signed images
  ubifs: remove unnecessary check in ubifs_log_start_commit
  ubifs: Fix typo of output in get_cs_sqnum
  ubifs: Simplify redundant code
  ubifs: Correctly use tnc_next() in search_dh_cookie()

Pull UML updates from Richard Weinberger:

 - A new timer mode, time travel, for testing with UML

 - Many bugixes/improvements for the serial line driver

 - Various bugfixes

* tag 'for-linus-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml:
  um: fix build without CONFIG_UML_TIME_TRAVEL_SUPPORT
  um: Fix kcov crash during startup
  um: configs: Remove useless UEVENT_HELPER_PATH
  um: Support time travel mode
  um: Pass nsecs to os timer functions
  um: Remove drivers/ssl.h
  um: Don't garbage collect in deactivate_all_fds()
  um: Silence lockdep complaint about mmap_sem
  um: Remove locking in deactivate_all_fds()
  um: Timer code cleanup
  um: fix os_timer_one_shot()
  um: Fix IRQ controller regression on console read

Pull stream_open() updates from Kirill Smelkov:
 "This time on stream_open front it is only two small changes:

   - the first one converts stream_open.cocci to treat all functions
     that start with wait_.* as blocking. Previously it was only
     wait_event_.* functions that were considered as blocking, but this
     was falsely reporting several deadlock cases as only warning.

     This was picked by linux-kbuild and entered mainline as commit
     0c4ab18f ("coccinelle: api/stream_open: treat all wait_.*()
     calls as blocking"), and already merged earlier.

   - the second one teaches stream_open.cocci to consider files as being
     stream-like even if they use noop_llseek. It results in two more
     drivers being converted to stream_open() (mousedev.c and
     hid-sensor-custom.c)"

* tag 'stream_open-5.3' of https://lab.nexedi.com/kirr/linux:
  *: convert stream-like files -> stream_open, even if they use noop_llseek

Pull x86 platform driver updates from Andy Shevchenko:
 "Gathered a bunch of x86 platform driver changes. It's rather big,
  since includes two big refactors and completely new driver:

   - ASUS WMI driver got a big refactoring in order to support the TUF
     Gaming laptops. Besides that, the regression with backlight being
     permanently off on various EeePC laptops has been fixed.

   - Accelerometer on HP ProBook 450 G0 shows wrong measurements due to
     X axis being inverted. This has been fixed.

   - Intel PMC core driver has been extended to be ACPI enumerated if
     the DSDT provides device with _HID "INT33A1". This allows to
     convert the driver to be pure platform and support new hardware
     purely based on ACPI DSDT.

   - From now on the Intel Speed Select Technology is supported thru a
     corresponding driver. This driver provides an access to the
     features of the ISST, such as Performance Profile, Core Power, Base
     frequency and Turbo Frequency.

   - Mellanox platform drivers has been refactored and now extended to
     support more systems, including new coming ones.

   - The OLPC XO-1.75 platform is now supported.

   - CB4063 Beckhoff Automation board is using PMC clocks, provided via
     pmc_atom driver, for ethernet controllers in a way that they can't
     be managed by the clock driver. The quirk has been extended to
     cover this case.

   - Touchscreen on Chuwi Hi10 Plus tablet has been enabled. Meanwhile
     the information of Chuwi Hi10 Air has been fixed to cover more
     models based on the same platform.

   - Xiaomi notebooks have WMI interface enabled. Thus, the driver to
     support it has been provided. It required some extension of the
     generic WMI library, which allows to propagate opaque context to
     the ->probe() of the individual drivers.

  This release includes debugfs clean up from Greg KH for several
  drivers that drop return code check and make debugfs absence or
  failure non-fatal.

  Also miscellaneous fixes here and there, mostly for Acer WMI and
  various Intel drivers"

* tag 'platform-drivers-x86-v5.3-1' of git://git.infradead.org/linux-platform-drivers-x86: (74 commits)
  platform/x86: Fix PCENGINES_APU2 Kconfig warning
  tools/power/x86/intel-speed-select: Add .gitignore file
  platform/x86: mlx-platform: Fix error handling in mlxplat_init()
  platform/x86: intel_pmc_core: Attach using APCI HID "INT33A1"
  platform/x86: intel_pmc_core: transform Pkg C-state residency from TSC ticks into microseconds
  platform/x86: asus-wmi: Use dev_get_drvdata()
  Documentation/ABI: Add new attribute for mlxreg-io sysfs interfaces
  platform/x86: mlx-platform: Add more reset cause attributes
  platform/x86: mlx-platform: Modify DMI matching order
  platform/x86: mlx-platform: Add regmap structure for the next generation systems
  platform/x86: mlx-platform: Change API for i2c-mlxcpld driver activation
  platform/x86: mlx-platform: Move regmap initialization before all drivers activation
  MAINTAINERS: Update for Intel Speed Select Technology
  tools/power/x86: A tool to validate Intel Speed Select commands
  platform/x86: ISST: Restore state on resume
  platform/x86: ISST: Add Intel Speed Select PUNIT MSR interface
  platform/x86: ISST: Add Intel Speed Select mailbox interface via MSRs
  platform/x86: ISST: Add Intel Speed Select mailbox interface via PCI
  platform/x86: ISST: Add Intel Speed Select mmio interface
  platform/x86: ISST: Add IOCTL to Translate Linux logical CPU to PUNIT CPU number
  ...

Pull mailbox updates from Jassi Brar:

 - stm32: race fix by adding a spinlock

 - mhu: trim included headers

 - omap: add support for K3 SoCs

 - imx: Irq disable fix

 - bcm: tidy up extracting driver data

 - tegra: make resume 'noirq'

 - api: fix error handling

* tag 'mailbox-v5.3' of git://git.linaro.org/landing-teams/working/fujitsu/integration:
  mailbox: handle failed named mailbox channel request
  mailbox: tegra: avoid resume NULL mailboxes
  mailbox: tegra: hsp: add noirq resume
  mailbox: bcm-flexrm-mailbox: using dev_get_drvdata directly
  mailbox: imx: Clear GIEn bit at shutdown
  mailbox: omap: Add support for TI K3 SoCs
  dt-bindings: mailbox: omap: Update bindings for TI K3 SoCs
  mailbox: arm_mhu: reorder header inclusion and drop unneeded ones
  mailbox: stm32_ipcc: add spinlock to fix channels concurrent access

Pull percpu updates from Dennis Zhou:
 "This includes changes to let percpu_ref release the backing percpu
  memory earlier after it has been switched to atomic in cases where the
  percpu ref is not revived.

  This will help recycle percpu memory earlier in cases where the
  refcounts are pinned for prolonged periods of time"

* 'for-5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu:
  percpu_ref: release percpu memory early without PERCPU_REF_ALLOW_REINIT
  md: initialize percpu refcounters using PERCU_REF_ALLOW_REINIT
  io_uring: initialize percpu refcounters using PERCU_REF_ALLOW_REINIT
  percpu_ref: introduce PERCPU_REF_ALLOW_REINIT flag

Pull perf fixes from Ingo Molnar:
 "A number of PMU driver corner case fixes, a race fix, an event
  grouping fix, plus a bunch of tooling fixes/updates"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (30 commits)
  perf/x86/intel: Fix spurious NMI on fixed counter
  perf/core: Fix exclusive events' grouping
  perf/x86/amd/uncore: Set the thread mask for F17h L3 PMCs
  perf/x86/amd/uncore: Do not set 'ThreadMask' and 'SliceMask' for non-L3 PMCs
  perf/core: Fix race between close() and fork()
  perf intel-pt: Fix potential NULL pointer dereference found by the smatch tool
  perf intel-bts: Fix potential NULL pointer dereference found by the smatch tool
  perf script: Assume native_arch for pipe mode
  perf scripts python: export-to-sqlite.py: Fix DROP VIEW power_events_view
  perf scripts python: export-to-postgresql.py: Fix DROP VIEW power_events_view
  perf hists browser: Fix potential NULL pointer dereference found by the smatch tool
  perf cs-etm: Fix potential NULL pointer dereference found by the smatch tool
  perf parse-events: Remove unused variable: error
  perf parse-events: Remove unused variable 'i'
  perf metricgroup: Add missing list_del_init() when flushing egroups list
  perf tools: Use list_del_init() more thorougly
  perf tools: Use zfree() where applicable
  tools lib: Adopt zalloc()/zfree() from tools/perf
  perf tools: Move get_current_dir_name() cond prototype out of util.h
  perf namespaces: Move the conditional setns() prototype to namespaces.h
  ...

Pull locking fix from Ingo Molnar:
 "A single fix for a locking statistics bug"

* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/lockdep: Fix lock used or unused stats error

Pull x86 fix from Ingo Molnar:
 "A single build system bugfix"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/vdso: Fix flip/flop vdso build bug

Pull scheduler fix from Ingo Molnar:
 "Fix a sched statistics related bug that would trigger a kernel warning
  on certain configs"

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/core: Fix preempt warning in ttwu

This patch continues 10dce8af (fs: stream_open - opener for
stream-like files so that read and write can run simultaneously without
deadlock) and c5bf68fe (*: convert stream-like files from
nonseekable_open -> stream_open) and teaches steam_open.cocci to
consider files as being stream-like not only if they have
.llseek=no_llseek, but also if they have .llseek=noop_llseek.

This is safe to do: the comment about noop_llseek says

	This is an implementation of ->llseek useable for the rare special case when
	userspace expects the seek to succeed but the (device) file is actually not
	able to perform the seek. In this case you use noop_llseek() instead of
	falling back to the default implementation of ->llseek.

and in general noop_llseek was massively added to drivers in 6038f373
(llseek: automatically add .llseek fop) when changing default for NULL .llseek
from NOP to no_llseek with the idea to avoid breaking compatibility, if
maybe some user-space program was using lseek on a device without caring
about the result, but caring if it was an error or not.

Amended semantic patch produces two changes when applied tree-wide:

        drivers/hid/hid-sensor-custom.c:690:8-24: WARNING: hid_sensor_custom_fops: .read() has stream semantic; safe to change nonseekable_open -> stream_open.
        drivers/input/mousedev.c:564:1-17: ERROR: mousedev_fops: .read() can deadlock .write(); change nonseekable_open -> stream_open to fix.

Cc: Julia Lawall <Julia.Lawall@lip6.fr>
Cc: Jan Blunck <jblunck@suse.de>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Jiri Kosina <jikos@kernel.org>
Cc: Jonathan Cameron <jic23@kernel.org>
Cc: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Cc: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Kirill Smelkov <kirr@nexedi.com>

Pull powerpc updates from Michael Ellerman:
 "Notable changes:

   - Removal of the NPU DMA code, used by the out-of-tree Nvidia driver,
     as well as some other functions only used by drivers that haven't
     (yet?) made it upstream.

   - A fix for a bug in our handling of hardware watchpoints (eg. perf
     record -e mem: ...) which could lead to register corruption and
     kernel crashes.

   - Enable HAVE_ARCH_HUGE_VMAP, which allows us to use large pages for
     vmalloc when using the Radix MMU.

   - A large but incremental rewrite of our exception handling code to
     use gas macros rather than multiple levels of nested CPP macros.

  And the usual small fixes, cleanups and improvements.

  Thanks to: Alastair D'Silva, Alexey Kardashevskiy, Andreas Schwab,
  Aneesh Kumar K.V, Anju T Sudhakar, Anton Blanchard, Arnd Bergmann,
  Athira Rajeev, Cédric Le Goater, Christian Lamparter, Christophe
  Leroy, Christophe Lombard, Christoph Hellwig, Daniel Axtens, Denis
  Efremov, Enrico Weigelt, Frederic Barrat, Gautham R. Shenoy, Geert
  Uytterhoeven, Geliang Tang, Gen Zhang, Greg Kroah-Hartman, Greg Kurz,
  Gustavo Romero, Krzysztof Kozlowski, Madhavan Srinivasan, Masahiro
  Yamada, Mathieu Malaterre, Michael Neuling, Nathan Lynch, Naveen N.
  Rao, Nicholas Piggin, Nishad Kamdar, Oliver O'Halloran, Qian Cai, Ravi
  Bangoria, Sachin Sant, Sam Bobroff, Satheesh Rajendran, Segher
  Boessenkool, Shaokun Zhang, Shawn Anastasio, Stewart Smith, Suraj
  Jitindar Singh, Thiago Jung Bauermann, YueHaibing"

* tag 'powerpc-5.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (163 commits)
  powerpc/powernv/idle: Fix restore of SPRN_LDBAR for POWER9 stop state.
  powerpc/eeh: Handle hugepages in ioremap space
  ocxl: Update for AFU descriptor template version 1.1
  powerpc/boot: pass CONFIG options in a simpler and more robust way
  powerpc/boot: add {get, put}_unaligned_be32 to xz_config.h
  powerpc/irq: Don't WARN continuously in arch_local_irq_restore()
  powerpc/module64: Use symbolic instructions names.
  powerpc/module32: Use symbolic instructions names.
  powerpc: Move PPC_HA() PPC_HI() and PPC_LO() to ppc-opcode.h
  powerpc/module64: Fix comment in R_PPC64_ENTRY handling
  powerpc/boot: Add lzo support for uImage
  powerpc/boot: Add lzma support for uImage
  powerpc/boot: don't force gzipped uImage
  powerpc/8xx: Add microcode patch to move SMC parameter RAM.
  powerpc/8xx: Use IO accessors in microcode programming.
  powerpc/8xx: replace #ifdefs by IS_ENABLED() in microcode.c
  powerpc/8xx: refactor programming of microcode CPM params.
  powerpc/8xx: refactor printing of microcode patch name.
  powerpc/8xx: Refactor microcode write
  powerpc/8xx: refactor writing of CPM microcode arrays
  ...

Pull sparc updates from David Miller:
 "Just a few small changes:

   - Fix console naming inconsistency with hypervisor consoles, from
     John Paul Adrian Glaubitz

   - Fix userland compilation due to use of u_int, from Masahiro Yamada"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
  sparc64: Add missing newline at end of file
  sparc: fix unknown type name u_int in uapi header
  sparc: configs: Remove useless UEVENT_HELPER_PATH
  sparc: Remove redundant copy of the LGPL-2.0
  sunhv: Fix device naming inconsistency between sunhv_console and sunhv_reg

Pull networking fixes from David Miller:

 1) Fix excessive stack usage in cxgb4, from Arnd Bergmann.

 2) Missing skb queue lock init in tipc, from Chris Packham.

 3) Fix some regressions in ipv6 flow label handling, from Eric Dumazet.

 4) Elide flow dissection of local packets in FIB rules, from Petar
    Penkov.

 5) Fix TLS support build failure in mlx5, from Tariq Toukab.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (36 commits)
  ppp: mppe: Revert "ppp: mppe: Add softdep to arc4"
  net: dsa: qca8k: replace legacy gpio include
  net: hisilicon: Use devm_platform_ioremap_resource
  cxgb4: reduce kernel stack usage in cudbg_collect_mem_region()
  tipc: ensure head->lock is initialised
  tc-tests: updated skbedit tests
  nfp: flower: ensure ip protocol is specified for L4 matches
  nfp: flower: fix ethernet check on match fields
  net/mlx5e: Provide cb_list pointer when setting up tc block on rep
  net: phy: make exported variables non-static
  net: sched: Fix NULL-pointer dereference in tc_indr_block_ing_cmd()
  davinci_cpdma: don't cast dma_addr_t to pointer
  net: openvswitch: do not update max_headroom if new headroom is equal to old headroom
  net/mlx5e: Convert single case statement switch statements into if statements
  net/mlx5: E-Switch, Reduce ingress acl modify metadata stack usage
  net/mlx5e: Fix unused variable warning when CONFIG_MLX5_ESWITCH is off
  net/mlx5e: Fix compilation error in TLS code
  ipv6: fix static key imbalance in fl_create()
  ipv6: fix potential crash in ip6_datagram_dst_update()
  ipv6: tcp: fix flowlabels reflection for RST packets
  ...

Pull IDE update from David Miller:
 "Small cleanup"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide:
  ide: use BIT() macro for defining bit-flags

Pull MTD updates from Miquel Raynal:
 "This contains the following changes for MTD:

  MTD core changes:
   - New Hyperbus framework
   - New _is_locked (concat) implementation
   - Various cleanups

  NAND core changes:
   - use longest matching pattern in ->exec_op() default parser
   - export NAND operation tracer
   - add flag to indicate panic_write in MTD
   - use kzalloc() instead of kmalloc() and memset()

  Raw NAND controller drivers changes:
   - brcmnand:
       - fix BCH ECC layout for large page NAND parts
       - fallback to detected ecc-strength, ecc-step-size
       - when oops in progress use pio and interrupt polling
       - code refactor code to introduce helper functions
       - add support for v7.3 controller
   - FSMC:
       - use nand_op_trace for operation tracing
   - GPMI:
       - move all driver code into single file
       - various cleanups (including dmaengine changes)
       - use runtime PM to manage clocks
       - implement exec_op
   - MTK:
       - correct low level time calculation of r/w cycle
       - improve data sampling timing for read cycle
       - add validity check for CE# pin setting
       - fix wrongly assigned OOB buffer pointer issue
       - re-license MTK NAND driver as Dual MIT/GPL
   - STM32:
       - manage the get_irq error case
       - increase DMA completion timeouts

  Raw NAND chips drivers changes:
   - Macronix: add read-retry support

  Onenand driver changes:
   - add support for 8Gb datasize chips
   - avoid fall-through warnings

  SPI-NAND changes:
   - define macros for page-read ops with three-byte addresses
   - add support for two-byte device IDs and then for GigaDevice
     GD5F1GQ4UFxxG
   - add initial support for Paragon PN26G0xA
   - handle the case where the last page read has bitflips

  SPI-NOR core changes:
   - add support for the mt25ql02g and w25q16jv flashes
   - print error in case of jedec read id fails
   - is25lp256: add post BFPT fix to correct the addr_width

  SPI NOR controller drivers changes:
   - intel-spi: Add support for Intel Elkhart Lake SPI serial flash
   - smt32: remove the driver as the driver was replaced by spi-stm32-qspi.c
   - cadence-quadspi: add reset control"

* tag 'mtd/for-5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux: (60 commits)
  mtd: concat: implement _is_locked mtd operation
  mtd: concat: refactor concat_lock/concat_unlock
  mtd: abi: do not use C++ style comments in uapi header
  mtd: afs: remove unneeded NULL check
  mtd: rawnand: stm32_fmc2: increase DMA completion timeouts
  mtd: rawnand: Use kzalloc() instead of kmalloc() and memset()
  mtd: hyperbus: Add driver for TI's HyperBus memory controller
  mtd: spinand: read returns badly if the last page has bitflips
  mtd: spinand: Add initial support for Paragon PN26G0xA
  mtd: rawnand: mtk: Re-license MTK NAND driver as Dual MIT/GPL
  mtd: rawnand: gpmi: remove double assignment to block_size
  dt-bindings: mtd: brcmnand: Add brcmnand, brcmnand-v7.3 support
  mtd: rawnand: brcmnand: Add support for v7.3 controller
  mtd: rawnand: brcmnand: Refactored code to introduce helper functions
  mtd: rawnand: brcmnand: When oops in progress use pio and interrupt polling
  mtd: Add flag to indicate panic_write
  mtd: rawnand: Add Macronix NAND read retry support
  mtd: onenand: Avoid fall-through warnings
  mtd: spinand: Add support for GigaDevice GD5F1GQ4UFxxG
  mtd: spinand: Add support for two-byte device IDs
  ...

Merge tag 'for-5.3/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper updates from Mike Snitzer:

 - Add encrypted byte-offset initialization vector (eboiv) to DM crypt.

 - Add optional discard features to DM snapshot which allow freeing
   space from a DM device whose free space was exhausted.

 - Various small improvements to use struct_size() and kzalloc().

 - Fix to check if DM thin metadata is in fail_io mode before attempting
   to update the superblock to set the needs_check flag. Otherwise the
   DM thin-pool can hang.

 - Fix DM bufio shrinker's potential for ABBA recursion deadlock with DM
   thin provisioning on loop usecase.

* tag 'for-5.3/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm bufio: fix deadlock with loop device
  dm snapshot: add optional discard support features
  dm crypt: implement eboiv - encrypted byte-offset initialization vector
  dm crypt: remove obsolete comment about plumb IV
  dm crypt: wipe private IV struct after key invalid flag is set
  dm integrity: use kzalloc() instead of kmalloc() + memset()
  dm: update stale comment in end_clone_bio()
  dm log writes: fix incorrect comment about the logged sequence example
  dm log writes: use struct_size() to calculate size of pending_block
  dm crypt: use struct_size() when allocating encryption context
  dm integrity: always set version on superblock update
  dm thin metadata: check if in fail_io mode when setting needs_check

Pull IPMI updates from Corey Minyard:
 "Some small fixes for various things, nothing huge, mostly found by
  automated tools.

  Plus add a driver that allows Linux to act as an IPMB slave device, so
  it can be a satellite MC in an IPMI network"

* tag 'for-linus-5.3' of git://github.com/cminyard/linux-ipmi:
  docs: ipmb: place it at driver-api and convert to ReST
  fix platform_no_drv_owner.cocci warnings
  ipmi: ipmb: don't allocate i2c_client on stack
  ipmi: ipmb: Fix build error while CONFIG_I2C is set to m
  Add support for IPMB driver
  drivers: ipmi: Drop device reference
  ipmi_ssif: fix unexpected driver unregister warning
  ipmi_si: use bool type for initialized variable
  ipmi_si: fix unexpected driver unregister warning

Pull pin control updates from Linus Walleij:
 "This is the bulk of pin control changes for the v5.3 kernel cycle:

  Core changes:

   - Device links can optionally be added between a pin control producer
     and its consumers. This will affect how the system power management
     is handled: a pin controller will not suspend before all of its
     consumers have been suspended.

     This was necessary for the ST Microelectronics STMFX expander and
     need to be tested on other systems as well: it makes sense to make
     this default in the long run.

     Right now it is opt-in per driver.

   - Drive strength can be specified in microamps. With decreases in
     silicon technology, milliamps isn't granular enough, let's make it
     possible to select drive strengths in microamps.

     Right now the Meson (AMlogic) driver needs this.

  New drivers:

   - New subdriver for the Tegra 194 SoC.

   - New subdriver for the Qualcomm SDM845.

   - New subdriver for the Qualcomm SM8150.

   - New subdriver for the Freescale i.MX8MN (Freescale is now a product
     line of NXP).

   - New subdriver for Marvell MV98DX1135.

  Driver improvements:

   - The Bitmain BM1880 driver now supports pin config in addition to
     muxing.

   - The Qualcomm drivers can now reserve some GPIOs as taken aside and
     not usable for users. This is used in ACPI systems to take out some
     GPIO lines used by the BIOS so that noone else (neither kernel nor
     userspace) will play with them by mistake and crash the machine.

   - A slew of refurbishing around the Aspeed drivers (board management
     controllers for servers) in preparation for the new Aspeed AST2600
     SoC.

   - A slew of improvements over the SH PFC drivers as usual.

   - Misc cleanups and fixes"

* tag 'pinctrl-v5.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl: (106 commits)
  pinctrl: aspeed: Strip moved macros and structs from private header
  pinctrl: aspeed: Fix missed include
  pinctrl: baytrail: Use GENMASK() consistently
  pinctrl: baytrail: Re-use data structures from pinctrl-intel.h
  pinctrl: baytrail: Use defined macro instead of magic in byt_get_gpio_mux()
  pinctrl: qcom: Add SM8150 pinctrl driver
  dt-bindings: pinctrl: qcom: Add SM8150 pinctrl binding
  dt-bindings: pinctrl: qcom: Document missing gpio nodes
  pinctrl: aspeed: Add implementation-related documentation
  pinctrl: aspeed: Split out pinmux from general pinctrl
  pinctrl: aspeed: Clarify comment about strapping W1C
  pinctrl: aspeed: Correct comment that is no longer true
  MAINTAINERS: Add entry for ASPEED pinctrl drivers
  dt-bindings: pinctrl: aspeed: Convert AST2500 bindings to json-schema
  dt-bindings: pinctrl: aspeed: Convert AST2400 bindings to json-schema
  dt-bindings: pinctrl: aspeed: Split bindings document in two
  pinctrl: qcom: Add irq_enable callback for msm gpio
  pinctrl: madera: Fixup SPDX headers
  pinctrl: qcom: sdm845: Fix CONFIG preprocessor guard
  pinctrl: tegra: Add bitmask support for parked bits
  ...

Pull input updates from Dmitry Torokhov:

 - an update to Elan touchpad SMBus driver to fetch device parameters
   (size, resolution) while it is still in PS/2 mode, before switching
   over to SMBus, as in that mode some devices return garbage dimensions

 - update to iforce joystick driver

 - miscellaneous driver fixes

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input: (48 commits)
  Input: gpio_keys_polled - allow specifying name of input device
  Input: edt-ft5x06 - simplify event reporting code
  Input: max77650-onkey - add MODULE_ALIAS()
  Input: atmel_mxt_ts - fix leak in mxt_update_cfg()
  Input: synaptics - enable SMBUS on T480 thinkpad trackpad
  Input: atmel_mxt_ts - fix -Wunused-const-variable
  Input: joydev - extend absolute mouse detection
  HID: quirks: Refactor ELAN 400 and 401 handling
  Input: elan_i2c - export the device id whitelist
  Input: edt-ft5x06 - use get_unaligned_be16()
  Input: iforce - add the Saitek R440 Force Wheel
  Input: iforce - use unaligned accessors, where appropriate
  Input: iforce - drop couple of temps from transport code
  Input: iforce - drop bus type from iforce structure
  Input: iforce - use DMA-safe buffores for USB transfers
  Input: iforce - allow callers supply data buffer when fetching device IDs
  Input: iforce - only call iforce_process_packet() if initialized
  Input: iforce - signal command completion from transport code
  Input: iforce - do not combine arguments for iforce_process_packet()
  Input: iforce - factor out hat handling when parsing packets
  ...

Pull io_uring updates from Jens Axboe:
 "This contains:

   - Support for recvmsg/sendmsg as first class opcodes.

     I don't envision going much further down this path, as there are
     plans in progress to support potentially any system call in an
     async fashion through io_uring. But I think it does make sense to
     have certain core ops available directly, especially those that can
     support a "try this non-blocking" flag/mode. (me)

   - Handle generic short reads automatically.

     This can happen fairly easily if parts of the buffered read is
     cached. Since the application needs to issue another request for
     the remainder, just do this internally and save kernel/user
     roundtrip while providing a nicer more robust API. (me)

   - Support for linked SQEs.

     This allows SQEs to depend on each other, enabling an application
     to eg queue a read-from-this-file,write-to-that-file pair. (me)

   - Fix race in stopping SQ thread (Jackie)"

* tag 'for-5.3/io_uring-20190711' of git://git.kernel.dk/linux-block:
  io_uring: fix io_sq_thread_stop running in front of io_sq_thread
  io_uring: add support for recvmsg()
  io_uring: add support for sendmsg()
  io_uring: add support for sqe links
  io_uring: punt short reads to async context
  uio: make import_iovec()/compat_import_iovec() return bytes on success

The stats variable nr_unused_locks is incremented every time a new lock
class is register and decremented when the lock is first used in
__lock_acquire(). And after all, it is shown and checked in lockdep_stats.

However, under configurations that either CONFIG_TRACE_IRQFLAGS or
CONFIG_PROVE_LOCKING is not defined:

The commit:

  09180651 ("locking/lockdep: Consolidate lock usage bit initialization")

missed marking the LOCK_USED flag at IRQ usage initialization because
as mark_usage() is not called. And the commit:

  886532ae ("locking/lockdep: Move mark_lock() inside CONFIG_TRACE_IRQFLAGS && CONFIG_PROVE_LOCKING")

further made mark_lock() not defined such that the LOCK_USED cannot be
marked at all when the lock is first acquired.

As a result, we fix this by not showing and checking the stats under such
configurations for lockdep_stats.
Reported-by: Qian Cai <cai@lca.pw>
Signed-off-by: Yuyang Du <duyuyang@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: arnd@arndb.de
Cc: frederic@kernel.org
Link: https://lkml.kernel.org/r/20190709101522.9117-1-duyuyang@gmail.comSigned-off-by: Ingo Molnar <mingo@kernel.org>

John reported a DEBUG_PREEMPT warning caused by commit:

  aacedf26 ("sched/core: Optimize try_to_wake_up() for local wakeups")

I overlooked that ttwu_stat() requires preemption disabled.
Reported-by: John Stultz <john.stultz@linaro.org>
Tested-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: aacedf26 ("sched/core: Optimize try_to_wake_up() for local wakeups")
Link: https://lkml.kernel.org/r/20190710105736.GK3402@hirez.programming.kicks-ass.netSigned-off-by: Ingo Molnar <mingo@kernel.org>

If a user first sample a PEBS event on a fixed counter, then sample a
non-PEBS event on the same fixed counter on Icelake, it will trigger
spurious NMI. For example:

  perf record -e 'cycles:p' -a
  perf record -e 'cycles' -a

The error message for spurious NMI:

  [June 21 15:38] Uhhuh. NMI received for unknown reason 30 on CPU 2.
  [    +0.000000] Do you have a strange power saving mode enabled?
  [    +0.000000] Dazed and confused, but trying to continue

The bug was introduced by the following commit:

  commit 6f55967a ("perf/x86/intel: Fix race in intel_pmu_disable_event()")

The commit moves the intel_pmu_pebs_disable() after intel_pmu_disable_fixed(),
which returns immediately.  The related bit of PEBS_ENABLE MSR will never be
cleared for the fixed counter. Then a non-PEBS event runs on the fixed counter,
but the bit on PEBS_ENABLE is still set, which triggers spurious NMIs.

Check and disable PEBS for fixed counters after intel_pmu_disable_fixed().
Reported-by: Yi, Ammy <ammy.yi@intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: <stable@vger.kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Fixes: 6f55967a ("perf/x86/intel: Fix race in intel_pmu_disable_event()")
Link: https://lkml.kernel.org/r/20190625142135.22112-1-kan.liang@linux.intel.comSigned-off-by: Ingo Molnar <mingo@kernel.org>

So far, we tried to disallow grouping exclusive events for the fear of
complications they would cause with moving between contexts. Specifically,
moving a software group to a hardware context would violate the exclusivity
rules if both groups contain matching exclusive events.

This attempt was, however, unsuccessful: the check that we have in the
perf_event_open() syscall is both wrong (looks at wrong PMU) and
insufficient (group leader may still be exclusive), as can be illustrated
by running:

  $ perf record -e '{intel_pt//,cycles}' uname
  $ perf record -e '{cycles,intel_pt//}' uname

ultimately successfully.

Furthermore, we are completely free to trigger the exclusivity violation
by:

   perf -e '{cycles,intel_pt//}' -e '{intel_pt//,instructions}'

even though the helpful perf record will not allow that, the ABI will.

The warning later in the perf_event_open() path will also not trigger, because
it's also wrong.

Fix all this by validating the original group before moving, getting rid
of broken safeguards and placing a useful one to perf_install_in_context().
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: mathieu.poirier@linaro.org
Cc: will.deacon@arm.com
Fixes: bed5b25a ("perf: Add a pmu capability for "exclusive" events")
Link: https://lkml.kernel.org/r/20190701110755.24646-1-alexander.shishkin@linux.intel.comSigned-off-by: Ingo Molnar <mingo@kernel.org>

Fill in the L3 performance event select register ThreadMask
bitfield, to enable per hardware thread accounting.
Signed-off-by: Kim Phillips <kim.phillips@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Gary Hook <Gary.Hook@amd.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Janakarajan Natarajan <Janakarajan.Natarajan@amd.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Martin Liska <mliska@suse.cz>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Pu Wen <puwen@hygon.cn>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Link: https://lkml.kernel.org/r/20190628215906.4276-2-kim.phillips@amd.comSigned-off-by: Ingo Molnar <mingo@kernel.org>

The following commit:

  d7cbbe49 ("perf/x86/amd/uncore: Set ThreadMask and SliceMask for L3 Cache perf events")

enables L3 PMC events for all threads and slices by writing 1's in
'ChL3PmcCfg' (L3 PMC PERF_CTL) register fields.

Those bitfields overlap with high order event select bits in the Data
Fabric PMC control register, however.

So when a user requests raw Data Fabric events (-e amd_df/event=0xYYY/),
the two highest order bits get inadvertently set, changing the counter
select to events that don't exist, and for which no counts are read.

This patch changes the logic to write the L3 masks only when dealing
with L3 PMC counters.

AMD Family 16h and below Northbridge (NB) counters were not affected.
Signed-off-by: Kim Phillips <kim.phillips@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Gary Hook <Gary.Hook@amd.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Janakarajan Natarajan <Janakarajan.Natarajan@amd.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Martin Liska <mliska@suse.cz>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Pu Wen <puwen@hygon.cn>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Fixes: d7cbbe49 ("perf/x86/amd/uncore: Set ThreadMask and SliceMask for L3 Cache perf events")
Link: https://lkml.kernel.org/r/20190628215906.4276-1-kim.phillips@amd.comSigned-off-by: Ingo Molnar <mingo@kernel.org>

Syzcaller reported the following Use-after-Free bug:

	close()						clone()

							  copy_process()
							    perf_event_init_task()
							      perf_event_init_context()
							        mutex_lock(parent_ctx->mutex)
								inherit_task_group()
								  inherit_group()
								    inherit_event()
								      mutex_lock(event->child_mutex)
								      // expose event on child list
								      list_add_tail()
								      mutex_unlock(event->child_mutex)
							        mutex_unlock(parent_ctx->mutex)

							    ...
							    goto bad_fork_*

							  bad_fork_cleanup_perf:
							    perf_event_free_task()

	  perf_release()
	    perf_event_release_kernel()
	      list_for_each_entry()
		mutex_lock(ctx->mutex)
		mutex_lock(event->child_mutex)
		// event is from the failing inherit
		// on the other CPU
		perf_remove_from_context()
		list_move()
		mutex_unlock(event->child_mutex)
		mutex_unlock(ctx->mutex)

							      mutex_lock(ctx->mutex)
							      list_for_each_entry_safe()
							        // event already stolen
							      mutex_unlock(ctx->mutex)

							    delayed_free_task()
							      free_task()

	     list_for_each_entry_safe()
	       list_del()
	       free_event()
	         _free_event()
		   // and so event->hw.target
		   // is the already freed failed clone()
		   if (event->hw.target)
		     put_task_struct(event->hw.target)
		       // WHOOPSIE, already quite dead

Which puts the lie to the the comment on perf_event_free_task():
'unexposed, unused context' not so much.

Which is a 'fun' confluence of fail; copy_process() doing an
unconditional free_task() and not respecting refcounts, and perf having
creative locking. In particular:

  82d94856 ("perf/core: Fix lock inversion between perf,trace,cpuhp")

seems to have overlooked this 'fun' parade.

Solve it by using the fact that detached events still have a reference
count on their (previous) context. With this perf_event_free_task()
can detect when events have escaped and wait for their destruction.
Debugged-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Reported-by: syzbot+a24c397a29ad22d86c98@syzkaller.appspotmail.com
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Cc: <stable@vger.kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Fixes: 82d94856 ("perf/core: Fix lock inversion between perf,trace,cpuhp")
Signed-off-by: Ingo Molnar <mingo@kernel.org>