On platforms with no numa support and with several CPUs, logs have lots
of noise for message "Fail to get numa node for CPU:.."

Change pr_info() to pr_info_once() as one print is enough to show the
issue.

Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Link: https://lore.kernel.org/r/20230808174359.50602-1-srinivas.pandruvada@linux.intel.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

The msi-ec driver fails to build for me (gcc 7.5):

  CC [M]  drivers/platform/x86/msi-ec.o
drivers/platform/x86/msi-ec.c:72:6: error: initializer element is not constant
    { SM_ECO_NAME,     0xc2 },
      ^~~~~~~~~~~
drivers/platform/x86/msi-ec.c:72:6: note: (near initialization for ‘CONF0.shift_mode.modes[0].name’)
drivers/platform/x86/msi-ec.c:73:6: error: initializer element is not constant
    { SM_COMFORT_NAME, 0xc1 },
      ^~~~~~~~~~~~~~~
drivers/platform/x86/msi-ec.c:73:6: note: (near initialization for ‘CONF0.shift_mode.modes[1].name’)
drivers/platform/x86/msi-ec.c:74:6: error: initializer element is not constant
    { SM_SPORT_NAME,   0xc0 },
      ^~~~~~~~~~~~~
drivers/platform/x86/msi-ec.c:74:6: note: (near initialization for ‘CONF0.shift_mode.modes[2].name’)
(...)

Don't try to be smart, just use defines for the constant strings. The
compiler will recognize it's the same string and will store it only
once in the data section anyway.

Signed-off-by: Jean Delvare <jdelvare@suse.de>
Fixes: 392cacf2 ("platform/x86: Add new msi-ec driver")
Cc: stable@vger.kernel.org
Cc: Nikita Kravets <teackot@gmail.com>
Cc: Hans de Goede <hdegoede@redhat.com>
Cc: Mark Gross <markgross@kernel.org>
Link: https://lore.kernel.org/r/20230805101010.54d49e91@endymion.delvare


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

The ACPI device CSC3556 is a Cirrus Logic CS35L56 mono amplifier which
is used in multiples, and can be connected either to I2C or SPI.

There will be multiple instances under the same Device() node. Add it
to ignore_serial_bus_ids and handle it in the serial-multi-instantiate
driver.

There can be a 5th I2cSerialBusV2, but this is an alias address and doesn't
represent a real device. Ignore this by having a dummy 5th entry in the
serial-multi-instantiate instance list with the name of a non-existent
driver, on the same pattern as done for bsg2150.

Signed-off-by: Simon Trimmer <simont@opensource.cirrus.com>
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Link: https://lore.kernel.org/r/20230728111345.7224-1-rf@opensource.cirrus.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

The return value from the call to amd_pmf_get_pprof_modes() is int.
However, the return value is being assigned to an unsigned char
variable 'mode', so making 'mode' an int.

silence the warning:
./drivers/platform/x86/amd/pmf/sps.c:183:5-9: WARNING: Unsigned expression compared with zero: mode < 0

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5995


Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Link: https://lore.kernel.org/r/20230727014315.51375-1-yang.lee@linux.alibaba.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Currently huawei-wmi causes a lot of spam in dmesg on my
Huawei MateBook X Pro 2022:

  ...
  [36409.328463] input input9: Unknown key pressed, code: 0x02c1
  [36411.335104] input input9: Unknown key pressed, code: 0x02c1
  [36412.338674] input input9: Unknown key pressed, code: 0x02c1
  [36414.848564] input input9: Unknown key pressed, code: 0x02c1
  [36416.858706] input input9: Unknown key pressed, code: 0x02c1
  ...

Fix that by ignoring events generated by ambient light sensor.

This issue was reported on GitHub and resolved with the following merge
request:

  https://github.com/aymanbagabas/Huawei-WMI/pull/70



I've contacted the mainter of this repo and he gave me the "go ahead" to
send this patch to the maling list.

Signed-off-by: Konstantin Shelekhin <k.shelekhin@ftml.net>
Link: https://lore.kernel.org/r/20230722155922.173856-1-k.shelekhin@ftml.net


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Only the HW rfkill state is toggled on laptops with quirks->ec_read_only
(so far only MSI Wind U90/U100). There are, however, a few issues with
the implementation:

1. The initial HW state is always unblocked, regardless of the actual
   state on boot, because msi_init_rfkill only sets the SW state,
   regardless of ec_read_only.

2. The initial SW state corresponds to the actual state on boot, but it
   can't be changed afterwards, because set_device_state returns
   -EOPNOTSUPP. It confuses the userspace, making Wi-Fi and/or Bluetooth
   unusable if it was blocked on boot, and breaking the airplane mode if
   the rfkill was unblocked on boot.

Address the above issues by properly initializing the HW state on
ec_read_only laptops and by allowing the userspace to toggle the SW
state. Don't set the SW state ourselves and let the userspace fully
control it. Toggling the SW state is a no-op, however, it allows the
userspace to properly toggle the airplane mode. The actual SW radio
disablement is handled by the corresponding rtl818x_pci and btusb
drivers that have their own rfkills.

Tested on MSI Wind U100 Plus, BIOS ver 1.0G, EC ver 130.

Fixes: 0816392b ("msi-laptop: merge quirk tables to one")
Fixes: 0de6575a ("msi-laptop: Add MSI Wind U90/U100 support")
Signed-off-by: Maxim Mikityanskiy <maxtram95@gmail.com>
Link: https://lore.kernel.org/r/20230721145423.161057-1-maxtram95@gmail.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

This patch fixes setting the cmd values to 0xb3 and 0xb4.
This is necessary on some TUF laptops in order to set the RGB mode.

Closes: https://lore.kernel.org/platform-driver-x86/443078148.491022.1677576298133@nm83.abv.bg


Signed-off-by: Kristian Angelov <kristiana2000@abv.bg>
Reviewed-by: Luke D. Jones <luke@ljones.dev>
Link: https://lore.kernel.org/r/ZLlS7o6UdTUBkyqa@wyvern.localdomain


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

key might contain private part of the key, so better use
kfree_sensitive to free it.

Signed-off-by: Wang Ming <machel@vivo.com>
Link: https://lore.kernel.org/r/20230717101114.18966-1-machel@vivo.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

HP Elite Dragonfly G2 (a convertible laptop/tablet) has a reliable VGBS
method. If VGBS is not called on boot, the firmware sends an initial
0xcd event shortly after calling the BTNL method, but only if the device
is booted in the laptop mode. However, if the device is booted in the
tablet mode and VGBS is not called, there is no initial 0xcc event, and
the input device for SW_TABLET_MODE is not registered up until the user
turns the device into the laptop mode.

Call VGBS on boot on this device to get the initial state of
SW_TABLET_MODE in a reliable way.

Tested with BIOS 1.13.1.

Signed-off-by: Maxim Mikityanskiy <maxtram95@gmail.com>
Link: https://lore.kernel.org/r/20230716183213.64173-1-maxtram95@gmail.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

On a HP Elite Dragonfly G2 the 0xcc and 0xcd events for SW_TABLET_MODE
are only send after the BTNL ACPI method has been called.

Likely more devices need this, so make the BTNL ACPI method unconditional
instead of only doing it on devices with a 5 button array.

Note this also makes the intel_button_array_enable() call in probe()
unconditional, that function does its own priv->array check. This makes
the intel_button_array_enable() call in probe() consistent with the calls
done on suspend/resume which also rely on the priv->array check inside
the function.

Reported-by: Maxim Mikityanskiy <maxtram95@gmail.com>
Closes: https://lore.kernel.org/platform-driver-x86/20230712175023.31651-1-maxtram95@gmail.com/


Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20230715181516.5173-1-hdegoede@redhat.com

APMF fn8 can notify EC about the OS slider position change. Add this
capability to the PMF driver so that it can call the APMF fn8 based on
the changes in the Platform profile events.

Co-developed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Patil Rajesh Reddy <Patil.Reddy@amd.com>
Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
Link: https://lore.kernel.org/r/20230714144435.1239776-2-Shyam-sundar.S-k@amd.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

apmf_get_system_params() failure is not a critical event, reduce its
verbosity from dev_err to dev_dbg.

Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
Link: https://lore.kernel.org/r/20230714144435.1239776-1-Shyam-sundar.S-k@amd.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

The current code assumes that the CSC3551(multiple cs35l41) always have
its interrupt pin connected to GPIO thus the IRQ can be acquired with
acpi_dev_gpio_irq_get. However on some newer laptop models this is no
longer the case as they have the CSC3551's interrupt pin connected to
APIC. This causes smi_i2c_probe to fail on these machines.

To support these machines, a new macro IRQ_RESOURCE_AUTO was introduced
for cs35l41 smi_node, and smi_get_irq function was modified so it tries
to get GPIO irq resource first and if failed, tries to get
APIC irq resource for cs35l41.

This patch affects only the cs35l41's probing and brings no negative
influence on machines that indeed have the cs35l41's interrupt pin
connected to GPIO.

Signed-off-by: David Xu <xuwd1@hotmail.com>
Link: https://lore.kernel.org/r/SY4P282MB18350CD8288687B87FFD2243E037A@SY4P282MB1835.AUSP282.PROD.OUTLOOK.COM


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Muxed (mem) regions will wait in request_mem_region_muxed() if the region
is busy (in use by another consumer) during the call.

In order to wake-up possibly waiting other consumers of the region,
it must be released by a release_mem_region() call, which will actually
wake up any waiters.

release_mem_region() also frees the resource created by
request_mem_region_muxed(), avoiding the need for the unmatched kfree().

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andy@kernel.org>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20230711095920.264308-1-hdegoede@redhat.com

Fix the axes and add home button support as suggested by Hans de Goede.

Signed-off-by: Thomas GENTY <tomlohave@gmail.com>
Link: https://lore.kernel.org/r/20230714105117.192938-1-tomlohave@gmail.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Add info for the Archos 101 Cesium Educ tablet
It was tested using gslx680_ts_acpi module
PR at https://github.com/onitake/gsl-firmware/pull/210

 for the firmware

Signed-off-by: Thomas GENTY <tomlohave@gmail.com>
Link: https://lore.kernel.org/r/20230707141425.21473-1-tomlohave@gmail.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Add missing empty line necessary for sphinx to recognize
the list. Also reword the first entry a little bit.

Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Link: https://lore.kernel.org/r/20230707010333.12954-2-W_Armin@gmx.de


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

If for some reason a external function returns -ENODEV,
no error message is being displayed because the driver
assumes that -ENODEV can only be returned internally if
no sensors, etc where found.
Fix this by explicitly returning 0 in such a case since
missing hardware is no error. Also remove the now obsolete
check for -ENODEV.

Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Link: https://lore.kernel.org/r/20230707010333.12954-1-W_Armin@gmx.de


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Add new ACPI ID AMDI0103 used by upcoming AMD platform to the PMF
supported list of devices.

Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
Link: https://lore.kernel.org/r/20230711100903.384151-1-Shyam-sundar.S-k@amd.com


Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Add new ACPI ID AMDI000A used by upcoming AMD platform to the pmc
supported list of devices

Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
Link: https://lore.kernel.org/r/20230711100344.383948-1-Shyam-sundar.S-k@amd.com


Signed-off-by: Hans de Goede <hdegoede@redhat.com>

HP 15s-eq2xxx is an older Lucienne laptop that has a problem resuming
from s2idle when IOMMU is enabled.  The symptoms very closely resemble
that of the Lenovo issues with NVME resume. Lucienne was released in
a similar timeframe as the Renoir / Cezanne Lenovo laptops and they
may have similar BIOS code.

Applying the same quirk to this system allows the system to work with
IOMMU enabled and s2idle resume to work.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2684


Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20230710183934.17315-3-mario.limonciello@amd.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

It turns out that some-non Lenovo systems can benefit from the quirk
introduced for Lenovo systems in commit 455cd867 ("platform/x86:
thinkpad_acpi: Add a s2idle resume quirk for a number of laptops").

So move this quirk into running from the amd-pmc driver instead.
No intended functional changes.

Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20230710183934.17315-2-mario.limonciello@amd.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

platform/x86: int3472/discrete: set variable skl_int3472_regulator_second_sensor storage-class-specifier to static

smatch reports
drivers/platform/x86/intel/int3472/clk_and_regulator.c:263:28: warning: symbol
  'skl_int3472_regulator_second_sensor' was not declared. Should it be static?

This variable is only used in its defining file, so it should be static.

Signed-off-by: Tom Rix <trix@redhat.com>
Link: https://lore.kernel.org/r/20230702134419.3438361-1-trix@redhat.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

cap_offset is a u16 field, so multiplying with TPMI_CAP_OFFSET_UNIT
(which is equal to 1024) to covert to bytes will cause overflow. This
will be a problem once more TPMI features are added.

This field is not used except for calculating pfs->vsec_offset. So, leave
cap_offset field unchanged and multiply with TPMI_CAP_OFFSET_UNIT while
calculating pfs->vsec_offset.

Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Link: https://lore.kernel.org/r/20230622195717.3125088-1-srinivas.pandruvada@linux.intel.com


Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

Even though we have no issues in the code, let's replace the open
coded guid_parse_and_compare().

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20230621151155.78279-2-andriy.shevchenko@linux.intel.com


Tested-by: Armin Wolf <W_Armin@gmx.de>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

The while-loop may break on one of the two conditions, either ID string
is empty or GUID matches. The second one, may never be reached if the
parsed string is not correct GUID. In such a case the loop will never
advance to check the next ID.

Break possible infinite loop by factoring out guid_parse_and_compare()
helper which may be moved to the generic header for everyone later on
and preventing from similar mistake in the future.

Interestingly that firstly it appeared when WMI was turned into a bus
driver, but later when duplicated GUIDs were checked, the while-loop
has been replaced by for-loop and hence no mistake made again.

Fixes: a48e2338 ("platform/x86: wmi: add context pointer field to struct wmi_device_id")
Fixes: 844af950 ("platform/x86: wmi: Turn WMI into a bus driver")
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20230621151155.78279-1-andriy.shevchenko@linux.intel.com


Tested-by: Armin Wolf <W_Armin@gmx.de>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>

We just sorted the entries and fields last release, so just out of a
perverse sense of curiosity, I decided to see if we can keep things
ordered for even just one release.

The answer is "No. No we cannot".

I suggest that all kernel developers will need weekly training sessions,
involving a lot of Big Bird and Sesame Street.  And at the yearly
maintainer summit, we will all sing the alphabet song together.

I doubt I will keep doing this.  At some point "perverse sense of
curiosity" turns into just a cold dark place filled with sadness and
despair.

Repeats: 80e62bc8 ("MAINTAINERS: re-sort all entries and fields")
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull dma-mapping fixes from Christoph Hellwig:

 - swiotlb area sizing fixes (Petr Tesarik)

* tag 'dma-mapping-6.5-2023-07-09' of git://git.infradead.org/users/hch/dma-mapping:
  swiotlb: reduce the number of areas to match actual memory pool size
  swiotlb: always set the number of areas before allocating the pool

Pull irq update from Borislav Petkov:

 - Optimize IRQ domain's name assignment

* tag 'irq_urgent_for_v6.5_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqdomain: Use return value of strreplace()

Pull x86 fpu fix from Borislav Petkov:

 - Do FPU AP initialization on Xen PV too which got missed by the recent
   boot reordering work

* tag 'x86_urgent_for_v6.5_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/xen: Fix secondary processors' FPU initialization

Pull x86 fix from Thomas Gleixner:
 "A single fix for the mechanism to park CPUs with an INIT IPI.

  On shutdown or kexec, the kernel tries to park the non-boot CPUs with
  an INIT IPI. But the same code path is also used by the crash utility.
  If the CPU which panics is not the boot CPU then it sends an INIT IPI
  to the boot CPU which resets the machine.

  Prevent this by validating that the CPU which runs the stop mechanism
  is the boot CPU. If not, leave the other CPUs in HLT"

* tag 'x86-core-2023-07-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/smp: Don't send INIT to boot CPU

Pull MIPS fixes from Thomas Bogendoerfer:

 - fixes for KVM

 - fix for loongson build and cpu probing

 - DT fixes

* tag 'mips_6.5_1' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
  MIPS: kvm: Fix build error with KVM_MIPS_DEBUG_COP0_COUNTERS enabled
  MIPS: dts: add missing space before {
  MIPS: Loongson: Fix build error when make modules_install
  MIPS: KVM: Fix NULL pointer dereference
  MIPS: Loongson: Fix cpu_probe_loongson() again

Pull xfs fix from Darrick Wong:
 "Nothing exciting here, just getting rid of a gcc warning that I got
  tired of seeing when I turn on gcov"

* tag 'xfs-6.5-merge-6' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: fix uninit warning in xfs_growfs_data

Pull more smb client updates from Steve French:

 - fix potential use after free in unmount

 - minor cleanup

 - add worker to cleanup stale directory leases

* tag '6.5-rc-smb3-client-fixes-part2' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: Add a laundromat thread for cached directories
  smb: client: remove redundant pointer 'server'
  cifs: fix session state transition to avoid use-after-free issue

Pull NTB updates from Jon Mason:
 "Fixes for pci_clean_master, error handling in driver inits, and
  various other issues/bugs"

* tag 'ntb-6.5' of https://github.com/jonmason/ntb:
  ntb: hw: amd: Fix debugfs_create_dir error checking
  ntb.rst: Fix copy and paste error
  ntb_netdev: Fix module_init problem
  ntb: intel: Remove redundant pci_clear_master
  ntb: epf: Remove redundant pci_clear_master
  ntb_hw_amd: Remove redundant pci_clear_master
  ntb: idt: drop redundant pci_enable_pcie_error_reporting()
  MAINTAINERS: git://github -> https://github.com for jonmason
  NTB: EPF: fix possible memory leak in pci_vntb_probe()
  NTB: ntb_tool: Add check for devm_kcalloc
  NTB: ntb_transport: fix possible memory leak while device_register() fails
  ntb: intel: Fix error handling in intel_ntb_pci_driver_init()
  NTB: amd: Fix error handling in amd_ntb_pci_driver_init()
  ntb: idt: Fix error handling in idt_pci_driver_init()

Lockdep is certainly right to complain about

  (&vma->vm_lock->lock){++++}-{3:3}, at: vma_start_write+0x2d/0x3f
                 but task is already holding lock:
  (&mapping->i_mmap_rwsem){+.+.}-{3:3}, at: mmap_region+0x4dc/0x6db

Invert those to the usual ordering.

Fixes: 33313a74 ("mm: lock newly mapped VMA which can be modified after it becomes visible")
Cc: stable@vger.kernel.org
Signed-off-by: Hugh Dickins <hughd@google.com>
Tested-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Merge tag 'mm-hotfixes-stable-2023-07-08-10-43' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Pull hotfixes from Andrew Morton:
 "16 hotfixes. Six are cc:stable and the remainder address post-6.4
  issues"

The merge undoes the disabling of the CONFIG_PER_VMA_LOCK feature, since
it was all hopefully fixed in mainline.

* tag 'mm-hotfixes-stable-2023-07-08-10-43' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  lib: dhry: fix sleeping allocations inside non-preemptable section
  kasan, slub: fix HW_TAGS zeroing with slub_debug
  kasan: fix type cast in memory_is_poisoned_n
  mailmap: add entries for Heiko Stuebner
  mailmap: update manpage link
  bootmem: remove the vmemmap pages from kmemleak in free_bootmem_page
  MAINTAINERS: add linux-next info
  mailmap: add Markus Schneider-Pargmann
  writeback: account the number of pages written back
  mm: call arch_swap_restore() from do_swap_page()
  squashfs: fix cache race with migration
  mm/hugetlb.c: fix a bug within a BUG(): inconsistent pte comparison
  docs: update ocfs2-devel mailing list address
  MAINTAINERS: update ocfs2-devel mailing list address
  mm: disable CONFIG_PER_VMA_LOCK until its fixed
  fork: lock VMAs of the parent process when forking

When forking a child process, the parent write-protects anonymous pages
and COW-shares them with the child being forked using copy_present_pte().

We must not take any concurrent page faults on the source vma's as they
are being processed, as we expect both the vma and the pte's behind it
to be stable.  For example, the anon_vma_fork() expects the parents
vma->anon_vma to not change during the vma copy.

A concurrent page fault on a page newly marked read-only by the page
copy might trigger wp_page_copy() and a anon_vma_prepare(vma) on the
source vma, defeating the anon_vma_clone() that wasn't done because the
parent vma originally didn't have an anon_vma, but we now might end up
copying a pte entry for a page that has one.

Before the per-vma lock based changes, the mmap_lock guaranteed
exclusion with concurrent page faults.  But now we need to do a
vma_start_write() to make sure no concurrent faults happen on this vma
while it is being processed.

This fix can potentially regress some fork-heavy workloads.  Kernel
build time did not show noticeable regression on a 56-core machine while
a stress test mapping 10000 VMAs and forking 5000 times in a tight loop
shows ~5% regression.  If such fork time regression is unacceptable,
disabling CONFIG_PER_VMA_LOCK should restore its performance.  Further
optimizations are possible if this regression proves to be problematic.

Suggested-by: David Hildenbrand <david@redhat.com>
Reported-by: Jiri Slaby <jirislaby@kernel.org>
Closes: https://lore.kernel.org/all/dbdef34c-3a07-5951-e1ae-e9c6e3cdf51b@kernel.org/


Reported-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Closes: https://lore.kernel.org/all/b198d649-f4bf-b971-31d0-e8433ec2a34c@applied-asynchrony.com/


Reported-by: Jacob Young <jacobly.alt@gmail.com>
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217624


Fixes: 0bff0aae ("x86/mm: try VMA lock-based page fault handling first")
Cc: stable@vger.kernel.org
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mmap_region adds a newly created VMA into VMA tree and might modify it
afterwards before dropping the mmap_lock.  This poses a problem for page
faults handled under per-VMA locks because they don't take the mmap_lock
and can stumble on this VMA while it's still being modified.  Currently
this does not pose a problem since post-addition modifications are done
only for file-backed VMAs, which are not handled under per-VMA lock.
However, once support for handling file-backed page faults with per-VMA
locks is added, this will become a race.

Fix this by write-locking the VMA before inserting it into the VMA tree.
Other places where a new VMA is added into VMA tree do not modify it
after the insertion, so do not need the same locking.

Cc: stable@vger.kernel.org
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>