Add SHAM crypto accelerator.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

OMAP4 has AES2 instance, so add its integration data under DT.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

"aes1_fck" and "aes2_fck" are controlled by hwmod. Drop clock
entries to avoid conflicts.
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

OMAP4 has a second aes module, so let's use proper name for
the first instance.
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Configure CPCAP battery driver.

Cc: devicetree@vger.kernel.org
Cc: Marcel Partap <mpartap@gmx.net>
Cc: Michael Scott <michael.scott@linaro.org>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The GPU has two functional clocks - GPU_CORE_GCLK and GPU_HYD_GCLK.
Both of these are mux clocks and are derived from the DPLL_CORE
H14 output clock CORE_GPU_CLK by default. These clocks can also be
be derived from DPLL_PER or DPLL_GPU.

The GPU DPLL provides the output clocks primarily for the GPU.
Configuring the GPU for different OPP clock frequencies is easier
to achieve when using the DPLL_GPU rather than the other two DPLLs
due to:
1. minimal affect on any other output clocks from these DPLLs
2. may require an impossible post-divider values on existing DPLLs
   without affecting other clocks.

So, switch the GPU functional clocks to be sourced from GPU DPLL by
default. This is done using the DT standard properties "assigned-clocks"
and "assigned-clock-parents". Newer u-boots (from 2017.01 onwards) reuse
and can update these properties to choose an appropriate one-time fixed
OPP configuration as all the required ABB/AVS setup is performed within
the bootloader. Note that there is no DVFS supported for any of the
non-MPU domains. The DPLL will automatically transition into a low-power
stop mode when the associated output clocks are not utilized or gated
automatically.

This patch also sets the initial values for the DPLL_GPU outputs.
These values are chosen based on the OPP_NOM values defined as per
recommendation from design team. The DPLL locked frequency is kept
at 1277 MHz, so that the value for the divider clock, dpll_gpu_m2_ck,
can be set to 425.67 MHz for OPP_NOM.
Signed-off-by: Subhajit Paul <subhajit_paul@ti.com>
[s-anna@ti.com: revise patch description]
Signed-off-by: Suman Anna <s-anna@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The IVA DPLL in DRA7xx provides the output clocks for only the IVAHD
subsystem in DRA7xx as compared to previous OMAP generations when it
provided the clocks for both DSP and IVAHD subsystems. This DPLL is
currently not configured by older bootloaders. Use the DT standard
properties "assigned-clocks" and "assigned-clock-rates" to set the
IVA DPLL clock rate and the rates for its derivative clocks at boot
time to properly initialize/lock this DPLL and be independent of the
bootloader version. Newer u-boots (from 2017.01 onwards) reuse and
can update these properties to choose an appropriate one-time fixed
OPP configuration. The DPLL will automatically transition into a
low-power stop mode when the associated output clocks are not
utilized or gated automatically.

The reset value of the divider M2 (that supplies the IVA_GFLCK, the
functional clock for the IVAHD subsystem) does not match a specific
OPP. So, the derived output clock from this IVA DPLL has to be
initialized as well to avoid initializing these divider outputs to an
incorrect frequencies.

The OPP_NOM clock frequencies are defined in the AM572x SR2.0 Data
Sheet vB, section 5.5.2 "Voltage And Core Clock Specifications". The
clock rates are chosen based on these OPP_NOM values and defined as per
a DRA7xx PLL spec document. The DPLL locked frequency is 2300 MHz, so
the dpll_iva_ck clock rate used is half of this value. The value for the
divider clock, dpll_iva_m2_ck, has to be set to 388.333334 MHz or more
for the divider clk logic to compute the appropriate divider value for
OPP_NOM.
Signed-off-by: Suman Anna <s-anna@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The DSP DPLL is a new DPLL compared to previous OMAP generations and
supplies the root clocks for the DSP processors, as well as a mux
input source for EVE sub-system (on applicable SoCs). This DPLL is
currently not configured by older bootloaders. Use the DT standard
properties "assigned-clocks" and "assigned-clock-rates" to set the
DSP DPLL clock rate and the rates for its derivative clocks at boot
time to properly initialize/lock this DPLL and be independent of the
bootloader version. Newer u-boots (from 2017.01 onwards) reuse and
can update these properties to choose an appropriate one-time fixed
OPP configuration. The DPLL will automatically transition into a
low-power stop mode when the associated output clocks are not
utilized or gated automatically.

The DSP DPLL provides two output clocks, DSP_GFCLK and EVE_GCLK. The
desired rate for DSP_GFCLK is 600 MHz (same as DSP DPLL CLKOUT frequency),
and is currently auto set due to the desired M2 divider value being the
same as reset value for the locked frequency of 600 MHz. The EVE_GCLK
however is required to be 400 MHz, so set the dpll_dsp_m3x2_ck's rate
explicitly so that the divider is set properly. The dpll_dsp_m2_ck rate
is also set explicitly to not rely on any implicit matching divider reset
values to the locked DPLL frequency.

The OPP_NOM clock frequencies are defined in the AM572x SR2.0 Data
Sheet vB, section 5.5.2 "Voltage And Core Clock Specifications". The
clock rates are chosen based on these OPP_NOM values and defined as per
a DRA7xx PLL spec document. The DPLL locked frequency is 1200 MHz, so
the dpll_dsp_ck clock rate used is half of this value.
Signed-off-by: Suman Anna <s-anna@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The IPU1 functional clock is actually the output of a mux clock,
ipu1_gfclk_mux. The mux clock is sourced by default from the
DPLL_ABE_X2_CLK, and this results in a rather odd clock frequency
(361 MHz) for the IPU1 functional clock on platforms where ABE_DPLL
is configured properly. Reconfigure the mux clock to be sourced from
CORE_IPU_ISS_BOOST_CLK (dpll_core_h22x2_ck), so that both the IPU1
and IPU2 are running from the same clock and clocked at the same
nominal frequency of 425 MHz.

This also ensures that IPU1 functional clock is always configured
properly and becomes independent of the state of the ABE DPLL on
all boards.
Signed-off-by: Suman Anna <s-anna@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The IVA DPLL is not an essential DPLL for the functionality of a
bootloader and is usually not configured (e.g. older u-boots configure
it only if CONFIG_SYS_CLOCKS_ENABLE_ALL is enabled and u-boots newer
than 2014.01 do not even have an option), and this results in incorrect
operating frequencies when trying to use a DSP or IVAHD, whose root
clocks are derived from this DPLL. Use the DT standard properties
"assigned-clocks" and "assigned-clock-rates" to set the IVA DPLL clock
rate and the rates for its derivative clocks at boot time to properly
initialize/lock this DPLL. The DPLL will automatically transition
into a low-power stop mode when the associated output clocks are
not utilized or gated automatically.

The reset values of the dividers H11 & H12 (functional clocks for DSP
and IVAHD respectively) are identical to each other, but are different
at each OPP. The reset values also do not match a specific OPP. So, the
derived output clocks from the IVA DPLL have to be initialized as well
to avoid initializing these divider outputs to incorrect frequencies.

The clock rates are chosen based on the OPP_NOM values as defined in
the OMAP5432 SR2.0 Data Manual Book vK, section 5.2.3.5 "DPLL_IVA
Preferred Settings". The recommended maximum DPLL locked frequency is
2330 MHz for OPP_NOM (value for DPLL_IVA_X2_CLK), so the dpll_iva_ck
clock rate used is half of this value. The value 465.92 MHz is used
instead of 465.9 MHz for dpll_iva_h11x2_ck so that proper divider
value can be calculated.
Signed-off-by: Suman Anna <s-anna@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The IVA DPLL is not an essential DPLL for the functionality of a
bootloader and is usually not configured (e.g. older u-boots configure
it only if CONFIG_SYS_CLOCKS_ENABLE_ALL is enabled and u-boots newer
than 2014.01 do not even have an option), and this results in incorrect
operating frequencies when trying to use a DSP or IVAHD, whose root
clocks are derived from this DPLL. Use the DT standard properties
"assigned-clocks" and "assigned-clock-rates" to set the IVA DPLL clock
rate and the rates for its derivative clocks at boot time to properly
initialize/lock this DPLL. The DPLL will automatically transition
into a low-power stop mode when the associated output clocks are
not utilized or gated automatically.

The reset values of the dividers M4 & M5 (functional clocks for DSP and
IVAHD respectively) are identical to each other, but are different at
each OPP. The reset values also do not match a specific OPP. So, the
derived output clocks from the IVA DPLL have to be initialized as well
to avoid initializing these divider outputs to incorrect frequencies.

The clock rates are chosen based on the OPP100 values as defined in the
OMAP4430 ES2.x Public TRM vAP, section "3.6.3.8.7 DPLL_IVA Preferred
Settings". The DPLL locked frequency is 1862.4 MHz (value for
DPLL_IVA_X2_CLK), so the dpll_iva_ck clock rate used is half of
this value.
Signed-off-by: Suman Anna <s-anna@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Motorola Droid 4 uses a WL1285C, so use proper compatible value.
To avoid regressions while support for the new compatible value
is added to the Linux kernel, the old compatible value is preserved
as fallback.
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The Droid 4 has a isl29030 to measure ambient light (e.g. for
automatically adapting display brightness) and proximity.
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Interrupt-parent property is defined in the root node as
"interrupt-parent = <&intc>". This interrupt-parent value becomes
the default for the system, so removed redundant "interrupt-parent"
property from mmc, mac, lcdc and tscadc nodes.
Signed-off-by: Suniel Mahesh <sunil.m@techveda.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

This adds the serial slave device for the WL1835 Bluetooth interface.
Signed-off-by: Robert Nelson <robertcnelson@gmail.com>
CC: Ricardo Salveti <ricardo.salveti@linaro.org>
CC: Tony Lindgren <tony@atomide.com>
CC: Jason Kridner <jkridner@beagleboard.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

dcdc3 supplies to DDR on AM43x series. When we set both
min and max values to the same value. The regulator framework
sets that particular voltage. This is bad as we are changing
the ddr voltage when executing from ddr. Hence remove the min and
max values. The ddr supply voltage shall be set from bootloader
when not executing from ddr and not while executing from kernel.

The previous discussion can be found here:

http://www.spinics.net/lists/devicetree/msg56399.htmlSigned-off-by: Keerthy <j-keerthy@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

BeagleBone Blue is robotics-oriented version of the BeagleBone Black (BBB).

This board can be indentified by the BLAx value after A335BNLT (BBB)
in the at24 eeprom:
BLAx [aa 55 33 ee 41 33 33 35  42 4e 4c 54 42 4c 41 30 |.U3.A335BNLTBLA2|]

http://beagleboard.org/blue
https://github.com/beagleboard/beaglebone-blue

firmware: https://github.com/beagleboard/beaglebone-black-wireless/tree/master/firmware
wl18xx mac address: /proc/device-tree/ocp/ethernet@4a100000/slave@4a100200/mac-address
Signed-off-by: Robert Nelson <robertcnelson@gmail.com>
CC: Jason Kridner <jkridner@beagleboard.org>
CC: Drew Fustini <drew@beagleboard.org>
Acked-by: Jason Kridner <jkridner@beagleboard.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The twl4030_charger driver expects an iio channel to detect the
presence of an AC charger by looking at VAC (madc channel 11).
This definition is missing in the device tree.
Signed-off-by: Marek Belisko <marek@goldelico.com>
Signed-off-by: Sebastian Reichel <sre@kernel.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Add the needed clock nodes for the CLKOUT2 to be usable by boards.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Looks like nobody bothered to configure USB host for 37xx-evm
when we converted things to device tree, so let's add it. This
is similar to beagleboard configuration with few extra quirks
to configure the port. And as with beagleboard, OHCI won't work
because there is no USB LS/FS PHY. A HS USB hub is needed to use
devices like keyboard and mice.
Acked-by: Roger Quadros <rogerq@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The at24 driver allows to register I2C EEPROM chips using different vendor
and devices, but the I2C subsystem does not take the vendor into account
when matching using the I2C table since it only has device entries.

But when matching using an OF table, both the vendor and device has to be
taken into account so the driver defines only a set of compatible strings
using the "atmel" vendor as a generic fallback for compatible I2C devices.

So add this generic fallback to the device node compatible string to make
the device to match the driver using the OF device ID table.
Signed-off-by: Javier Martinez Canillas <javier@dowhile0.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

As long as the kernel cmdline has "earlycon" in it, this allows
seeing debug messages earlier and does not require DEBUG_LL to
be enabled.
Acked-by: Adam Ford <aford173@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

As long as the kernel cmdline has "earlycon" in it, this allows
seeing debug messages earlier and does not require DEBUG_LL to
be enabled.
Signed-off-by: Tony Lindgren <tony@atomide.com>

This adds the serial slave device for the WL1835 Bluetooth interface.
Signed-off-by: Ricardo Salveti <ricardo.salveti@linaro.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Enable the System Mailboxes 5 and 6 and the corresponding
child sub-mailbox (IPC 3.x) nodes for the AM571x IDK board.
This is needed to enable communication with the respective
remote processors IPU1, IPU2, and DSP1 from the MPU.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Enable the System Mailboxes 5 and 6 and the corresponding
child sub-mailbox (IPC 3.x) nodes for the AM572x IDK board.
This is needed to enable communication with the respective
remote processors IPU1, IPU2, DSP1 and DSP2 from the MPU.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

With two separate &gpmc nodes the second ranges property overwrites the
first. So put nand and ethernet in a single node and merge the ranges.

While at it also fix the ethernet suffix.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Property 'enable-active-low' does not exist. Only 'enable-active-high' is
valid, and when this property is absent the gpio regulator will act as
active low by default.

So remove the unexisting 'enable-active-low' property.
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

If we have Linux installed in eMMC we can boot without
µSD card, but inserting one is not recognised.

The reason is that the card detect gpio (gpio5_152)
is not configured and attached to the mmc1 interface
driver and the mmc driver does not poll by default.

Hence we add pinmux and gpio setup for the SDCARD_NCD
signal.
Signed-off-by: H. Nikolaus Schaller <hns@goldelico.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Droid 4 has WL 1285C connected to the OMAP's UART4 port, which is
used for Bluetooth and most likely can also be used for controlling
the FM radio and GPS receivers.
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

USB1 port is micro-AB type and can function as peripheral
as well as host. Enable dual-role mode for USB1.

We don't want to use the OTG controller block on this
platform as it limits host mode to high-speed. Instead
we rely on extcon framework to give us ID events for
dual-role mode detection.
Signed-off-by: Roger Quadros <rogerq@ti.com>
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Compiling the DT file with W=1, DTC warns like follows:

Warning (unit_address_vs_reg): Node /opp_table0/opp@1000000000 has a
unit name, but no reg property

Fix this by replacing '@' with '-' as the OPP nodes will never have a
"reg" property.
Reported-by: Krzysztof Kozlowski <krzk@kernel.org>
Reported-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Suggested-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The rv4162 compatbile string is missing the vendor part, add it.
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Signed-off-by: Tony Lindgren <tony@atomide.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>