WMI is logically a bus: the WMI driver binds to an ACPI node (or
more than one), and each instance of the WMI driver enumerates its
children and hopes that drivers will attach to the children that are
useful.

This patch gives WMI a driver model bus type and the ability to
match to drivers. The bus itself is a device in the new "wmi_bus"
class, and all of the individual WMI devices are slotted into the
device hierarchy correctly.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Mario Limonciello <mario_limonciello@dell.com>
Cc: Pali Rohár <pali.rohar@gmail.com>
Cc: linux-kernel@vger.kernel.org
Cc: platform-driver-x86@vger.kernel.org
Cc: linux-acpi@vger.kernel.org
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

Currently we free all devices when we detach from any ACPI node.
Instead, keep track of which node WMI devices are attached to and
free them only as needed. While we are at it, match up notifications
with the device they came from correctly.

This will make our behavior more straightforward on systems with
more than one WMI node in the ACPI tables (e.g. the Dell XPS 13
9350).

This also adds a warning when GUIDs are not unique.

NB: The guid_string parameter in guid_already_parsed was a
little-endian binary GUID, not a string.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Mario Limonciello <mario_limonciello@dell.com>
Cc: Pali Rohár <pali.rohar@gmail.com>
Cc: linux-kernel@vger.kernel.org
Cc: platform-driver-x86@vger.kernel.org
Cc: linux-acpi@vger.kernel.org
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

Rearrange acpi_wmi_add to use Linux's error handling conventions.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Mario Limonciello <mario_limonciello@dell.com>
Cc: Pali Rohár <pali.rohar@gmail.com>
Cc: linux-kernel@vger.kernel.org
Cc: platform-driver-x86@vger.kernel.org
Cc: linux-acpi@vger.kernel.org
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

We will need the device to convert to a bus architecture and bind WMI to
the platform device.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Mario Limonciello <mario_limonciello@dell.com>
Cc: Pali Rohár <pali.rohar@gmail.com>
Cc: linux-kernel@vger.kernel.org
Cc: platform-driver-x86@vger.kernel.org
Cc: linux-acpi@vger.kernel.org
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

WMI is just a driver. There is no need to announce when it is loaded.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Mario Limonciello <mario_limonciello@dell.com>
Cc: Pali Rohár <pali.rohar@gmail.com>
Cc: linux-kernel@vger.kernel.org
Cc: platform-driver-x86@vger.kernel.org
Cc: linux-acpi@vger.kernel.org
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

According to Mario at Dell, the DELLABC6 device should not be used on a
Linux system. It also conflicts with Intel-HID and its interactions with
Network Manager. Document that we are aware of the device, but that we
are intentionally ignoring it.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
[dvhart: New commit message and minor comment wording fixes]
Cc: Mario Limonciello <mario_limonciello@dell.com>
Cc: "Pali Rohár" <pali.rohar@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

This is based on Mario's explanation and observation of my laptop.
Suggested-by: "Pali Rohár" <pali.rohar@gmail.com>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Mario Limonciello <mario_limonciello@dell.com>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

The hotkey table is 0xb2, add a comment for clarity.
Suggested-by: Darren Hart <dvhart@infradead.org>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
Reviewed-by: Pali Rohár <pali.rohar@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

To avoid using module-wide data in remaining module code, employ
acpi_driver_data() and dev_get_drvdata() to fetch device-specific data
to work on in each function.  This makes the input local variables in
hotkey-related callbacks and the module-wide struct fujitsu_laptop
redundant, so remove them.  Adjust whitespace to make checkpatch happy.
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

In order to perform their duties, all LED callbacks need a pointer to
the struct acpi_device representing the FUJ02E3 ACPI device.  To limit
the use of the module-wide pointer, the same pointer should be extracted
from data that gets passed to LED callbacks as arguments.  However, LED
core does not currently support supplying driver-specific pointers to
struct led_classdev callbacks, so the latter have to be implemented a
bit differently than backlight device callbacks and platform device
attribute callbacks.  As the FUJ02E3 ACPI device is the parent device of
all LED class devices registered by fujitsu-laptop, struct acpi_device
representing the former can be extracted by following the parent link
present inside the struct device belonging to the struct led_classdev
passed as an argument to each LED callback.

To get rid of module-wide structures defining LED class devices,
allocate them dynamically using devm_kzalloc() and initialize them in
acpi_fujitsu_laptop_leds_register().
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

Prepare for not using module-wide data in call_fext_func() by explicitly
passing it a pointer to struct acpi_device while still using a
module-wide pointer in each call.

Doing this enables call_fext_func() to fetch the ACPI handle from its
argument, making the acpi_handle field of struct fujitsu_laptop useless,
so remove that field.  While we are at it, the dev field of the same
structure is assigned in acpi_fujitsu_laptop_add() but not used for
anything, so remove it as well.
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

fujitsu-laptop registers two ACPI drivers: one for ACPI device FUJ02B1
enabling backlight control and another for ACPI device FUJ02E3 which
handles various other stuff (hotkeys, LEDs, etc.)  In a perfect world,
private data used by each of these drivers would be neatly encapsulated
in a structure specific to a given driver instance.  Sadly, firmware
present on some Fujitsu laptops makes that impossible by exposing
backlight power control (which is what the FUJ02B1 ACPI device should
take care of) through the FUJ02E3 ACPI device.  This means the backlight
driver needs a way to access an ACPI device it is not bound to.  When
the backlight driver is extracted into a separate module, it will not be
able to rely on a module-wide variable any more and such access will
happen through an API exposed by fujitsu-laptop.

For all known firmwares out in the wild, it seems that whenever the
FUJ02B1 ACPI device is present, it is always accompanied by a single
instance of the FUJ02E3 ACPI device.  We could independently grab an
ACPI handle to the FUJ02E3 ACPI device from the backlight driver, but
that would require using a hardcoded absolute path to that ACPI device,
which is subject to change.  It is easier to simply store a module-wide
pointer to the last (most likely only) FUJ02E3 ACPI device found, make
the aforementioned API use it and cover our bases by warning the user if
firmware exposes multiple FUJ02E3 ACPI devices.

Introducing this pointer in advance allows us to get rid of the
acpi_handle field of struct fujitsu_bl and also enables a bit more
step-by-step migration to a device-specific implementation of
call_fext_func().
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

Only allocate memory for struct fujitsu_laptop when the FUJ02E3 ACPI
device is present.  Use devm_kzalloc() for allocating memory to simplify
cleanup.
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

To prevent using module-wide data in backlight-related code, employ
acpi_driver_data() and bl_get_data() where possible to fetch
device-specific data to work on in each function.  This makes the input
local variable in acpi_fujitsu_bl_notify() and the acpi_handle field of
struct fujitsu_bl redundant, so remove them.
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

Only allocate memory for struct fujitsu_bl when the FUJ02B1 ACPI device
is present.  Use devm_kzalloc() for allocating memory to simplify
cleanup.

Due to the fact that the power property of the backlight device created
by the backlight driver is accessed from acpi_fujitsu_laptop_add(),
pointer to the allocated memory will remain stored in a module-wide
variable until the backlight driver is extracted into a separate module.
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

In portions of the driver which use device-specific data, rename local
variables from fujitsu_bl and fujitsu_laptop to priv in order to clearly
distinguish these parts from code that uses module-wide data.
Signed-off-by: Michał Kępień <kernel@kempniu.pl>
Reviewed-by: Jonathan Woithe <jwoithe@just42.net>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

Devices with the intel_cht_int33fe ACPI device use a max17047 fuel-gauge
combined with a bq24272i charger, in order for the fuel-gauge driver to
correctly display charging / discharging status it needs to know which
charger is supplying the battery.

This commit sets the supplied-from device property to the name of the
bq24272i charger for this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

The function is currently not used, however it is part of the API and
might be used in the future. Adding the attribute fixes the following
warning when building with clang:

    drivers/platform/x86/intel_pmc_ipc.c:189:18: error: unused function
        'ipc_data_readb' [-Werror,-Wunused-function]
Signed-off-by: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

Don't simply throw this to userspace via the sparse_keymap (which does not
have a mapping for scancode 1), as this causes KEY_UNKNOWN to be emitted,
which is a nuisance and of no use at all (it is not the right way to expose
this ACPI event to userspace, anyway, and the original intention of the commit
which added this (cfee5d63) was only to suppress
an unhandled event log message).
Signed-off-by: Hao Wei Tee <angelsl@angelsl.xyz>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

A readonly sysfs property must not have a 'store' function:

drivers/platform/x86/ideapad-laptop.c:438:16: error: 'touchpad_store' defined but not used [-Werror=unused-function]

We can either comment it out or remove the function entirely,
without a good reason one or or another I picked the second option.

Fixes: 7f363145 ("platform/x86: ideapad-laptop: Switch touchpad attribute to be RO")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

The latest Topstar BIOS updates (109_931P) advertise the "TPS0001"
device id by default, preventing the topstar-laptop module from being
loaded automatically.
Signed-off-by: Guillaume Douézan-Grard <gdouezangrard@gmail.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

PEAQ is a new European OEM, I've bought one of their 2-in-1 x86 devices,
which is actually quite a nice device. Under Windows it has Dolby
software for "better" sound and you can select different equalizer
presets using a special button.

This WMI interface for this button is not really nice, as it does not do
notifies (it really does not I triple checked), but since I had already
figured out the entire WMI interface for this I decided to go the full
mile anyway and implement a WMI based input driver for this using
input_polldev since, well, we need to poll.

This commit adds support for this button making it report KEY_SOUND
input events. KEY_SOUND is already used in various places to switch
sound into theatre mode and things like that so it seems appropriate
here.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
[dvhart: minor declaration ordering and commit log typo fixes]
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

As per discussion [1] there are only few users of module_param_call() in
kernel which prevent to read module parameters back.

It thinkpad_acpi driver there is even no method do so. Thus, for now,
add just a comment to explain why 0 is used as permissions in
module_param_call().

[1]: https://patchwork.ozlabs.org/patch/713245/

Cc: Richard Weinberger <richard.weinberger@gmail.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>

There is no point to keep string literal split. It even makes slightly
harder to maintain and debug.

Join string literals back to be oneliners.

While here, print negative error without changing a sign as it is a
common pattern in the kernel.

Other than above there were no functional changes intended.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>

Add touchscreen info for the GP-electronic T701 tablet.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>

Use memdup_user_nul() helper instead of open-coding to simplify the
code.
Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

For now let's restrict touchpad attribute to be read only.
We might revisit this in the future though.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

Lenovo Yoga (many variants: Yoga, Yoga2 Pro, Yoga2 13, Yoga3 Pro, Yoga 3
14, etc) has multiple modles that are a hybrid laptop, working in laptop
mode as well as tablet mode.

Currently, there is no easy interface to determine the touchpad status,
which in case of the Yoga family of machines, can also be useful to
assume tablet mode status.

Note: The ideapad-laptop driver does not provide a SW_TABLET_MODE either.

For a detailed discussion  on why we want either of the interfaces,
please see:
https://bugs.launchpad.net/onboard/+bug/1366421/comments/43

This patch adds a sysfs interface for read/write access under:
/sys/bus/platform/devices/VPC2004\:00/touchpad_mode
Signed-off-by: Ritesh Raj Sarraf <rrs@debian.org>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

Pull some more input subsystem updates from Dmitry Torokhov:
 "An updated xpad driver with a few more recognized device IDs, and a
  new psxpad-spi driver, allowing connecting Playstation 1 and 2 joypads
  via SPI bus"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: cros_ec_keyb - remove extraneous 'const'
  Input: add support for PlayStation 1/2 joypads connected via SPI
  Input: xpad - add USB IDs for Mad Catz Brawlstick and Razer Sabertooth
  Input: xpad - sync supported devices with xboxdrv
  Input: xpad - sort supported devices by USB ID

Pull UBI/UBIFS updates from Richard Weinberger:

 - new config option CONFIG_UBIFS_FS_SECURITY

 - minor improvements

 - random fixes

* tag 'upstream-4.12-rc1' of git://git.infradead.org/linux-ubifs:
  ubi: Add debugfs file for tracking PEB state
  ubifs: Fix a typo in comment of ioctl2ubifs & ubifs2ioctl
  ubifs: Remove unnecessary assignment
  ubifs: Fix cut and paste error on sb type comparisons
  ubi: fastmap: Fix slab corruption
  ubifs: Add CONFIG_UBIFS_FS_SECURITY to disable/enable security labels
  ubi: Make mtd parameter readable
  ubi: Fix section mismatch

Pull UML fixes from Richard Weinberger:
 "No new stuff, just fixes"

* 'for-linus-4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml:
  um: Add missing NR_CPUS include
  um: Fix to call read_initrd after init_bootmem
  um: Include kbuild.h instead of duplicating its macros
  um: Fix PTRACE_POKEUSER on x86_64
  um: Set number of CPUs
  um: Fix _print_addr()

Merge misc fixes from Andrew Morton:
 "15 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  mm, docs: update memory.stat description with workingset* entries
  mm: vmscan: scan until it finds eligible pages
  mm, thp: copying user pages must schedule on collapse
  dax: fix PMD data corruption when fault races with write
  dax: fix data corruption when fault races with write
  ext4: return to starting transaction in ext4_dax_huge_fault()
  mm: fix data corruption due to stale mmap reads
  dax: prevent invalidation of mapped DAX entries
  Tigran has moved
  mm, vmalloc: fix vmalloc users tracking properly
  mm/khugepaged: add missed tracepoint for collapse_huge_page_swapin
  gcov: support GCC 7.1
  mm, vmstat: Remove spurious WARN() during zoneinfo print
  time: delete current_fs_time()
  hwpoison, memcg: forcibly uncharge LRU pages

Commit 4b4cea91691d ("mm: vmscan: fix IO/refault regression in cache
workingset transition") introduced three new entries in memory stat
file:

 - workingset_refault
 - workingset_activate
 - workingset_nodereclaim

This commit adds a corresponding description to the cgroup v2 docs.

Link: http://lkml.kernel.org/r/1494530293-31236-1-git-send-email-guro@fb.comSigned-off-by: Roman Gushchin <guro@fb.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Li Zefan <lizefan@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Although there are a ton of free swap and anonymous LRU page in elgible
zones, OOM happened.

  balloon invoked oom-killer: gfp_mask=0x17080c0(GFP_KERNEL_ACCOUNT|__GFP_ZERO|__GFP_NOTRACK), nodemask=(null),  order=0, oom_score_adj=0
  CPU: 7 PID: 1138 Comm: balloon Not tainted 4.11.0-rc6-mm1-zram-00289-ge228d67e9677-dirty #17
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014
  Call Trace:
   oom_kill_process+0x21d/0x3f0
   out_of_memory+0xd8/0x390
   __alloc_pages_slowpath+0xbc1/0xc50
   __alloc_pages_nodemask+0x1a5/0x1c0
   pte_alloc_one+0x20/0x50
   __pte_alloc+0x1e/0x110
   __handle_mm_fault+0x919/0x960
   handle_mm_fault+0x77/0x120
   __do_page_fault+0x27a/0x550
   trace_do_page_fault+0x43/0x150
   do_async_page_fault+0x2c/0x90
   async_page_fault+0x28/0x30
  Mem-Info:
  active_anon:424716 inactive_anon:65314 isolated_anon:0
   active_file:52 inactive_file:46 isolated_file:0
   unevictable:0 dirty:27 writeback:0 unstable:0
   slab_reclaimable:3967 slab_unreclaimable:4125
   mapped:133 shmem:43 pagetables:1674 bounce:0
   free:4637 free_pcp:225 free_cma:0
  Node 0 active_anon:1698864kB inactive_anon:261256kB active_file:208kB inactive_file:184kB unevictable:0kB isolated(anon):0kB isolated(file):0kB mapped:532kB dirty:108kB writeback:0kB shmem:172kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no
  DMA free:7316kB min:32kB low:44kB high:56kB active_anon:8064kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:15992kB managed:15908kB mlocked:0kB slab_reclaimable:464kB slab_unreclaimable:40kB kernel_stack:0kB pagetables:24kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB
  lowmem_reserve[]: 0 992 992 1952
  DMA32 free:9088kB min:2048kB low:3064kB high:4080kB active_anon:952176kB inactive_anon:0kB active_file:36kB inactive_file:0kB unevictable:0kB writepending:88kB present:1032192kB managed:1019388kB mlocked:0kB slab_reclaimable:13532kB slab_unreclaimable:16460kB kernel_stack:3552kB pagetables:6672kB bounce:0kB free_pcp:56kB local_pcp:24kB free_cma:0kB
  lowmem_reserve[]: 0 0 0 959
  Movable free:3644kB min:1980kB low:2960kB high:3940kB active_anon:738560kB inactive_anon:261340kB active_file:188kB inactive_file:640kB unevictable:0kB writepending:20kB present:1048444kB managed:1010816kB mlocked:0kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB bounce:0kB free_pcp:832kB local_pcp:60kB free_cma:0kB
  lowmem_reserve[]: 0 0 0 0
  DMA: 1*4kB (E) 0*8kB 18*16kB (E) 10*32kB (E) 10*64kB (E) 9*128kB (ME) 8*256kB (E) 2*512kB (E) 2*1024kB (E) 0*2048kB 0*4096kB = 7524kB
  DMA32: 417*4kB (UMEH) 181*8kB (UMEH) 68*16kB (UMEH) 48*32kB (UMEH) 14*64kB (MH) 3*128kB (M) 1*256kB (H) 1*512kB (M) 2*1024kB (M) 0*2048kB 0*4096kB = 9836kB
  Movable: 1*4kB (M) 1*8kB (M) 1*16kB (M) 1*32kB (M) 0*64kB 1*128kB (M) 2*256kB (M) 4*512kB (M) 1*1024kB (M) 0*2048kB 0*4096kB = 3772kB
  378 total pagecache pages
  17 pages in swap cache
  Swap cache stats: add 17325, delete 17302, find 0/27
  Free swap  = 978940kB
  Total swap = 1048572kB
  524157 pages RAM
  0 pages HighMem/MovableOnly
  12629 pages reserved
  0 pages cma reserved
  0 pages hwpoisoned
  [ pid ]   uid  tgid total_vm      rss nr_ptes nr_pmds swapents oom_score_adj name
  [  433]     0   433     4904        5      14       3       82             0 upstart-udev-br
  [  438]     0   438    12371        5      27       3      191         -1000 systemd-udevd

With investigation, skipping page of isolate_lru_pages makes reclaim
void because it returns zero nr_taken easily so LRU shrinking is
effectively nothing and just increases priority aggressively.  Finally,
OOM happens.

The problem is that get_scan_count determines nr_to_scan with eligible
zones so although priority drops to zero, it couldn't reclaim any pages
if the LRU contains mostly ineligible pages.

get_scan_count:

        size = lruvec_lru_size(lruvec, lru, sc->reclaim_idx);
	size = size >> sc->priority;

Assumes sc->priority is 0 and LRU list is as follows.

	N-N-N-N-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H

(Ie, small eligible pages are in the head of LRU but others are
 almost ineligible pages)

In that case, size becomes 4 so VM want to scan 4 pages but 4 pages from
tail of the LRU are not eligible pages.  If get_scan_count counts
skipped pages, it doesn't reclaim any pages remained after scanning 4
pages so it ends up OOM happening.

This patch makes isolate_lru_pages try to scan pages until it encounters
eligible zones's pages.

[akpm@linux-foundation.org: clean up mind-bending `for' statement.  Tweak comment text]
Fixes: 3db65812 ("Revert "mm, vmscan: account for skipped pages as a partial scan"")
Link: http://lkml.kernel.org/r/1494457232-27401-1-git-send-email-minchan@kernel.orgSigned-off-by: Minchan Kim <minchan@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

We have encountered need_resched warnings in __collapse_huge_page_copy()
while doing {clear,copy}_user_highpage() over HPAGE_PMD_NR source pages.

mm->mmap_sem is held for write, but the iteration is well bounded.

Reschedule as needed.

Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1705101426380.109808@chino.kir.corp.google.comSigned-off-by: David Rientjes <rientjes@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

This is based on a patch from Jan Kara that fixed the equivalent race in
the DAX PTE fault path.

Currently DAX PMD read fault can race with write(2) in the following
way:

CPU1 - write(2)                 CPU2 - read fault
                                dax_iomap_pmd_fault()
                                  ->iomap_begin() - sees hole

dax_iomap_rw()
  iomap_apply()
    ->iomap_begin - allocates blocks
    dax_iomap_actor()
      invalidate_inode_pages2_range()
        - there's nothing to invalidate

                                  grab_mapping_entry()
				  - we add huge zero page to the radix tree
				    and map it to page tables

The result is that hole page is mapped into page tables (and thus zeros
are seen in mmap) while file has data written in that place.

Fix the problem by locking exception entry before mapping blocks for the
fault.  That way we are sure invalidate_inode_pages2_range() call for
racing write will either block on entry lock waiting for the fault to
finish (and unmap stale page tables after that) or read fault will see
already allocated blocks by write(2).

Fixes: 9f141d6e ("dax: Call ->iomap_begin without entry lock during dax fault")
Link: http://lkml.kernel.org/r/20170510172700.18991-1-ross.zwisler@linux.intel.comSigned-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Currently DAX read fault can race with write(2) in the following way:

CPU1 - write(2)			CPU2 - read fault
				dax_iomap_pte_fault()
				  ->iomap_begin() - sees hole
dax_iomap_rw()
  iomap_apply()
    ->iomap_begin - allocates blocks
    dax_iomap_actor()
      invalidate_inode_pages2_range()
        - there's nothing to invalidate
				  grab_mapping_entry()
				  - we add zero page in the radix tree
				    and map it to page tables

The result is that hole page is mapped into page tables (and thus zeros
are seen in mmap) while file has data written in that place.

Fix the problem by locking exception entry before mapping blocks for the
fault.  That way we are sure invalidate_inode_pages2_range() call for
racing write will either block on entry lock waiting for the fault to
finish (and unmap stale page tables after that) or read fault will see
already allocated blocks by write(2).

Fixes: 9f141d6e
Link: http://lkml.kernel.org/r/20170510085419.27601-5-jack@suse.czSigned-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

DAX will return to locking exceptional entry before mapping blocks for a
page fault to fix possible races with concurrent writes.  To avoid lock
inversion between exceptional entry lock and transaction start, start
the transaction already in ext4_dax_huge_fault().

Fixes: 9f141d6e
Link: http://lkml.kernel.org/r/20170510085419.27601-4-jack@suse.czSigned-off-by: Jan Kara <jack@suse.cz>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Currently, we didn't invalidate page tables during invalidate_inode_pages2()
for DAX.  That could result in e.g. 2MiB zero page being mapped into
page tables while there were already underlying blocks allocated and
thus data seen through mmap were different from data seen by read(2).
The following sequence reproduces the problem:

 - open an mmap over a 2MiB hole

 - read from a 2MiB hole, faulting in a 2MiB zero page

 - write to the hole with write(3p). The write succeeds but we
   incorrectly leave the 2MiB zero page mapping intact.

 - via the mmap, read the data that was just written. Since the zero
   page mapping is still intact we read back zeroes instead of the new
   data.

Fix the problem by unconditionally calling invalidate_inode_pages2_range()
in dax_iomap_actor() for new block allocations and by properly
invalidating page tables in invalidate_inode_pages2_range() for DAX
mappings.

Fixes: c6dcf52c
Link: http://lkml.kernel.org/r/20170510085419.27601-3-jack@suse.czSigned-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>