Merge tag 'omap-for-v4.3/soc-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into next/soc

SoC changes for omaps for v4.3 merge window:

- Clean-up omap4_local_timer_init to drop deal legacy code

- Provide proper IO map table for dra7

- Clean-up IOMMU layer init code as it now uses IOMMU framework

- A series of changes to fix up dm814x support that's been in a broken
  half-merged state for quite some time

- A series of PRCM and hwmod changes via Paul Walmsley <paul@pwsan.com>:

  - I/O wakeup support for AM43xx
  - register lock and unlock support to the hwmod code (needed for the RTC
    IP blocks on some chips)
  - several fixes for sparse warnings and an unnecessary null pointer test
  - a DRA7xx clockdomain configuration workaround, to deal with some hardware
    bugs

* tag 'omap-for-v4.3/soc-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap: (21 commits)
  ARM: OMAP2: Add minimal dm814x hwmod support
  ARM: OMAP2+: Prepare dm81xx hwmod code for adding minimal dm814x support
  ARM: PRM: AM437x: Enable IO wakeup feature
  ARM: OMAP4+: PRM: Add AM437x specific data
  ARM: OMAP: PRM: Remove hardcoding of IRQENABLE_MPU_2 and IRQSTATUS_MPU_2 register offsets
  ARM: dts: AM4372: Add PRCM IRQ entry
  ARM: AM43xx: Add the PRM IRQ register offsets
  ARM: OMAP4: PRM: Remove hardcoding of PRM_IO_PMCTRL_OFFSET register
  ARM: OMAP2+: Add support for initializing dm814x clocks
  ARM: OMAP2+: Add custom prwdm_operations for 81xx to support dm814x
  ARM: OMAP2+: Add minimal clockdomains for dm814x
  ARM: OMAP2+: Fix scm compatible for dm814x
  ARM: OMAP2+: Fix dm814x DT_MACHINE_START
  ARM: OMAP2+: Remove module references from IOMMU machine layer
  ARM: DRA7: Provide proper IO map table
  ARM: OMAP2+: Clean up omap4_local_timer_init
  ARM: OMAP2: Delete an unnecessary check
  ARM: OMAP2+: sparse: add missing function declarations
  ARM: OMAP2+: sparse: add missing static declaration
  ARM: OMAP2+: hwmod: add support for lock and unlock hooks
  ...
Signed-off-by: Olof Johansson <olof@lixom.net>

Merge tag 'renesas-soc-for-v4.3' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into next/soc

Renesas ARM Based SoC Updates for v4.3

* Add basic support for gose/r8a7793

* tag 'renesas-soc-for-v4.3' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:
  ARM: shmobile: gose: enable R-Car Gen2 regulator quirk
  ARM: shmobile: Basic r8a7793 SoC support
Signed-off-by: Olof Johansson <olof@lixom.net>

Merge tag 'sti-soc-for-v4.3-1' of https://git.kernel.org/pub/scm/linux/kernel/git/mcoquelin/sti into next/soc

STi SoC updates for v4.3, round 1.

Highlights:
-----------
 - Add code to release secondary cores from holding pen.
 - Remove useless call to trace_hardirqs_off() in secondary core init function.

* tag 'sti-soc-for-v4.3-1' of https://git.kernel.org/pub/scm/linux/kernel/git/mcoquelin/sti:
  ARM: STi: Remove platform call to trace_hardirqs_off()
  ARM: STi: Add code to release secondary cores from holding pen.
Signed-off-by: Olof Johansson <olof@lixom.net>

Update dm814x changes for sparse fixes to make data structures
static.

Conflicts:
	arch/arm/mach-omap2/omap_hwmod_81xx_data.c

Let's add minimal set of dm814x hwmods to have a bootable system.

Cc: Matthijs van Duin <matthijsvanduin@gmail.com>
Cc: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Let's change the defines so we can share the hwmod code better between
dm816x and dm814x, and let's add the dm814x specific defines. And let's
rename the shared ones to start with dm81xx. No functional changes.

Cc: Matthijs van Duin <matthijsvanduin@gmail.com>
Cc: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Merge tag 'for-v4.3/omap-hwmod-prcm-a' of git://git.kernel.org/pub/scm/linux/kernel/git/pjw/omap-pending into omap-for-v4.3/soc

ARM: OMAP2+: PRCM and hwmod changes for v4.3

This series adds:

- I/O wakeup support for AM43xx
- register lock and unlock support to the hwmod code (needed for the RTC
  IP blocks on some chips)
- several fixes for sparse warnings and an unnecessary null pointer test
- a DRA7xx clockdomain configuration workaround, to deal with some hardware
  bugs

Basic build, boot, and PM tests are here:

http://www.pwsan.com/omap/testlogs/hwmod-prcm-for-v4.3/20150723080012/

Since I do not have an AM43xx or DRA7xx device, I can't test on those
platforms.

Enable IO wakeup feature. This enables am437x pads to generate daisy
chained wake ups(eventually generates aprcm Interrupt) especially
when in low power modes.
Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

The register offsets for some of the PRM Registers are different
hence populating the differing fields. This is needed to support
IO wake up feature for am437x family.
Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

The register offsets of IRQENABLE_MPU_2 and IRQSTATUS_MPU_2 are hardcoded.
This makes it difficult to reuse the code for SoCs like AM437x that have
a single instance of IRQENABLE_MPU and IRQSTATUS_MPU registers.
Hence handling the case using offset of 4 to accommodate single set of IRQ*
registers generically.
Signed-off-by: Keerthy <j-keerthy@ti.com>
[paul@pwsan.com: fixed whitespace alignment problems reported by checkpatch.pl]
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

Add PRCM IRQ entry.  This is needed for I/O wakeup support.
Signed-off-by: Keerthy <j-keerthy@ti.com>
[paul@pwsan.com: added I/O wakeup note in commit description]
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

Add the PRM IRQ register offsets.  This is needed to support PRM I/O
wakeup on AM43xx.
Signed-off-by: Keerthy <j-keerthy@ti.com>
[paul@pwsan.com: improved patch description, moved the PRM_IO_PMCTRL macro
 out of the CM section]
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

PRM_IO_PMCTRL_OFFSET need not be same for all SOCs hence
remove hardcoding and use the value provided by the omap_prcm_irq_setup
structure. This is done to support IO wakeup on am437x series.
Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

Calling trace_hardirqs_off() from the platform specific
secondary startup code as not been necessary since Dec 2010
when Russell King consolidated the call into the common SMP
code.

2c0136db ARM: SMP: consolidate trace_hardirqs_off() into common SMP code
Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Signed-off-by: Maxime Coquelin <maxime.coquelin@st.com>

Most upstream devs boot STi platform via JTAG which abuses the
boot process by setting the PC of secondary cores directly. As
a consquence, booting STi platforms via u-boot results in only
the primary core being brought up as the code to manage the
holding pen is not upstream.

This patch adds the necessary code to bring the secondary cores
out of the holding pen. It uses the cpu-release-addr DT property
to get the address of the holding pen from the bootloader.

With this patch booting upstream kernels via u-boot works
correctly:

[    0.045456] CPU: Testing write buffer coherency: ok
[    0.045597] CPU0: thread -1, cpu 0, socket 0, mpidr 80000000
[    0.045734] Setting up static identity map for 0x40209000 - 0x40209098
[    0.065047] CPU1: thread -1, cpu 1, socket 0, mpidr 80000001
[    0.065081] Brought up 2 CPUs
[    0.065089] SMP: Total of 2 processors activated (5983.43 BogoMIPS).
[    0.065092] CPU: All CPU(s) started in SVC mode.
Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Acked-by: Maxime Coquelin <maxime.coquelin@st.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@st.com>

Merge tag 'v4.3-rockchip32-soc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into next/soc

Merge "Rockchip soc changes for 4.3, part1" from Heiko Stuebner:

Some suspend improvements enabling the possibility to wakeup
from usbphy events and a rework of how cpu cores are brought
up and down, as it was possible to produce lockups when
hammering the cpu hotplug functions.

* tag 'v4.3-rockchip32-soc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:
  ARM: rockchip: fix broken build
  ARM: rockchip: remove some useless macro in pm.h
  ARM: rockchip: add support holding 24Mhz osc during suspend
  ARM: rockchip: fix the SMP code style
  ARM: rockchip: ensure CPU to enter WFI/WFE state
  ARM: rockchip: fix the CPU soft reset
  ARM: rockchip: restore dapswjdp after suspend
Signed-off-by: Olof Johansson <olof@lixom.net>

Let's add a minimal clocks for dm814x to get it booted. This is
mostly a placeholder and relies on the PLLs being on from the
bootloader.

Note that the divider clocks work the same way as on dm816x and
am335x.

Cc: Matthijs van Duin <matthijsvanduin@gmail.com>
Cc: Mike Turquette <mturquette@linaro.org>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Tero Kristo <t-kristo@ti.com>
Acked-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Looking at the TI kernel tree I noticed that dm81xx need custom
ti81xx_pwrdm_operations. Let's also change dm816x over to use them
as the registers are different for dm81xx compared to others.

Cc: Matthijs van Duin <matthijsvanduin@gmail.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

For now, let's just add the ones shared with dm816x.
The dm814x specific ones can be added as they are tested.

Cc: Matthijs van Duin <matthijsvanduin@gmail.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Fix scm compatible for dm814x.

Cc: Matthijs van Duin <matthijsvanduin@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Fix dm814x DT_MACHINE_START.

Cc: Matthijs van Duin <matthijsvanduin@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The OMAP IOMMU driver has been adapted to the IOMMU framework
for a while now, and it no longer supports being built as a
module. Cleanup all the module related references both from
the code and in the build.

While at it, also relocate a comment around the initcall to
avoid a checkpatch strict warning about using a blank line
after function/struct/union/enum declarations.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

DRA7 uses OMAP5 IO table at the moment. This is purely spurious since
the OMAP5 and DRA7 register maps are different in many aspects.

AM57xx/DRA7 TRM Reference: http://www.ti.com/lit/ug/spruhz6/spruhz6.pdf

NOTE: Most of the drivers are already doing ioremap, so, there should'nt
be any functional improvement involved here, other than making the
initial iotable more accurate.

Fixes: a3a9384a ("ARM: DRA7: Reuse io tables and add a new .init_early")
Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Inspired by a patch from Felipe Balbi <balbi@ti.com>, we can
now get rid of most the code in omap4_local_timer_init.

Omap4 is now device tree only.. And we have not properly supported
omap4 ES1.0 revision for a really long time AFAIK.

Let's just remove all that code to simplify things. This assumes
we have arm,cortex-a9-twd-timer entry in the omap4.dtsi file, which
we do.
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

The of_node_put() function tests whether its argument is NULL and then
returns immediately if so.  Furthermore, the kerneldoc for
of_node_put() explicitly supports passing in a NULL pointer as its
argument.  Thus the test around the call is not needed.

This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
[paul@pwsan.com: dropped the omap_device.c and omap_hwmod.c changes for
 now, edited the commit message accordingly and to note the documented
 "contract"]
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

omap3xxx_restart() and omap44xx_restart() are global
functions declared in common.h. Include this file
in omap3-restart.c and omap4-restart.c to prevent
sparse warnings of type:

arch/arm/mach-omap2/omap4-restart.c:22:6: warning: symbol 'omap44xx_restart' was not declared. Should it be static?
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

Add missing static declaration for file local variables.
This fixes sparse warnings of type:

arch/arm/mach-omap2/omap_hwmod_81xx_data.c:491:26: warning: symbol 'dm81xx_alwon_l3_slow__gpmc' was not declared. Should it be static?
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

Some IP blocks like RTC, needs an additional setting for writing to its
registers. This is to prevent any spurious writes from changing the
register values.

This patch adds optional lock and unlock function pointers to
the IP block's hwmod data. These unlock and lock function pointers
are called by hwmod code before and after writing sysconfig registers.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
[paul@pwsan.com: fixed indentation level to conform with the rest of the
 structure members]
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

Legacy IPs like PWMSS, present under l4per2_7xx_clkdm, cannot support
smart-idle when its clock domain is in HW_AUTO on DRA7 SoCs. Hence,
program clock domain to SW_WKUP.
Signed-off-by: Vignesh R <vigneshr@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: <stable@vger.kernel.org>

This reverts commit dec4f799.

Jörg Otte reports a NULL pointder dereference due to this commit, as
'crtc_state' very much can be NULL:

        crtc_state = state->base.state ?
                intel_atomic_get_crtc_state(state->base.state, intel_crtc) : NULL;

So the change to test 'crtc_state->base.active' cannot possibly be
correct as-is.

There may be some other minimal fix (like just checking crtc_state for
NULL), but I'm just reverting it now for the rc2 release, and people
like Daniel Vetter who actually know this code will figure out what the
right solution is in the longer term.
Reported-and-bisected-by: Jörg Otte <jrg.otte@gmail.com>
Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Daniel Vetter <daniel.vetter@intel.com>
CC: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull VFS fixes from Al Viro:
 "Fixes for this cycle regression in overlayfs and a couple of
  long-standing (== all the way back to 2.6.12, at least) bugs"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  freeing unlinked file indefinitely delayed
  fix a braino in ovl_d_select_inode()
  9p: don't leave a half-initialized inode sitting around

Pull MIPS fixes from Ralf Baechle:
 "A fair number of 4.2 fixes also because Markos opened the flood gates.

   - Patch up the math used calculate the location for the page bitmap.

   - The FDC (Not what you think, FDC stands for Fast Debug Channel) IRQ
     around was causing issues on non-Malta platforms, so move the code
     to a Malta specific location.

   - A spelling fix replicated through several files.

   - Fix to the emulation of an R2 instruction for R6 cores.

   - Fix the JR emulation for R6.

   - Further patching of mindless 64 bit issues.

   - Ensure the kernel won't crash on CPUs with L2 caches with >= 8
     ways.

   - Use compat_sys_getsockopt for O32 ABI on 64 bit kernels.

   - Fix cache flushing for multithreaded cores.

   - A build fix"

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
  MIPS: O32: Use compat_sys_getsockopt.
  MIPS: c-r4k: Extend way_string array
  MIPS: Pistachio: Support CDMM & Fast Debug Channel
  MIPS: Malta: Make GIC FDC IRQ workaround Malta specific
  MIPS: c-r4k: Fix cache flushing for MT cores
  Revert "MIPS: Kconfig: Disable SMP/CPS for 64-bit"
  MIPS: cps-vec: Use macros for various arithmetics and memory operations
  MIPS: kernel: cps-vec: Replace KSEG0 with CKSEG0
  MIPS: kernel: cps-vec: Use ta0-ta3 pseudo-registers for 64-bit
  MIPS: kernel: cps-vec: Replace mips32r2 ISA level with mips64r2
  MIPS: kernel: cps-vec: Replace 'la' macro with PTR_LA
  MIPS: kernel: smp-cps: Fix 64-bit compatibility errors due to pointer casting
  MIPS: Fix erroneous JR emulation for MIPS R6
  MIPS: Fix branch emulation for BLTC and BGEC instructions
  MIPS: kernel: traps: Fix broken indentation
  MIPS: bootmem: Don't use memory holes for page bitmap
  MIPS: O32: Do not handle require 32 bytes from the stack to be readable.
  MIPS, CPUFREQ: Fix spelling of Institute.
  MIPS: Lemote 2F: Fix build caused by recent mass rename.

Pull x86 fixes from Thomas Gleixner:

 - the high latency PIT detection fix, which slipped through the cracks
   for rc1

 - a regression fix for the early printk mechanism

 - the x86 part to plug irq/vector related hotplug races

 - move the allocation of the espfix pages on cpu hotplug to non atomic
   context.  The current code triggers a might_sleep() warning.

 - a series of KASAN fixes addressing boot crashes and usability

 - a trivial typo fix for Kconfig help text

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/kconfig: Fix typo in the CONFIG_CMDLINE_BOOL help text
  x86/irq: Retrieve irq data after locking irq_desc
  x86/irq: Use proper locking in check_irq_vectors_for_cpu_disable()
  x86/irq: Plug irq vector hotplug race
  x86/earlyprintk: Allow early_printk() to use console style parameters like '115200n8'
  x86/espfix: Init espfix on the boot CPU side
  x86/espfix: Add 'cpu' parameter to init_espfix_ap()
  x86/kasan: Move KASAN_SHADOW_OFFSET to the arch Kconfig
  x86/kasan: Add message about KASAN being initialized
  x86/kasan: Fix boot crash on AMD processors
  x86/kasan: Flush TLBs after switching CR3
  x86/kasan: Fix KASAN shadow region page tables
  x86/init: Clear 'init_level4_pgt' earlier
  x86/tsc: Let high latency PIT fail fast in quick_pit_calibrate()

Pull timer fixes from Thomas Gleixner:
 "This update from the timer departement contains:

   - A series of patches which address a shortcoming in the tick
     broadcast code.

     If the broadcast device is not available or an hrtimer emulated
     broadcast device, some of the original assumptions lead to boot
     failures.  I rather plugged all of the corner cases instead of only
     addressing the issue reported, so the change got a little larger.

     Has been extensivly tested on x86 and arm.

   - Get rid of the last holdouts using do_posix_clock_monotonic_gettime()

   - A regression fix for the imx clocksource driver

   - An update to the new state callbacks mechanism for clockevents.
     This is required to simplify the conversion, which will take place
     in 4.3"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  tick/broadcast: Prevent NULL pointer dereference
  time: Get rid of do_posix_clock_monotonic_gettime
  cris: Replace do_posix_clock_monotonic_gettime()
  tick/broadcast: Unbreak CONFIG_GENERIC_CLOCKEVENTS=n build
  tick/broadcast: Handle spurious interrupts gracefully
  tick/broadcast: Check for hrtimer broadcast active early
  tick/broadcast: Return busy when IPI is pending
  tick/broadcast: Return busy if periodic mode and hrtimer broadcast
  tick/broadcast: Move the check for periodic mode inside state handling
  tick/broadcast: Prevent deep idle if no broadcast device available
  tick/broadcast: Make idle check independent from mode and config
  tick/broadcast: Sanity check the shutdown of the local clock_event
  tick/broadcast: Prevent hrtimer recursion
  clockevents: Allow set-state callbacks to be optional
  clocksource/imx: Define clocksource for mx27

Pull irq fix from Thomas Gleixner:
 "A single fix for a cpu hotplug race vs. interrupt descriptors:

  Prevent irq setup/teardown across the cpu starting/dying parts of cpu
  hotplug so that the starting/dying cpu has a stable view of the
  descriptor space.  This has been an issue for all architectures in the
  cpu dying phase, where interrupts are migrated away from the dying
  cpu.  In the starting phase its mostly a x86 issue vs the vector space
  update"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  hotplug: Prevent alloc/free of irq descriptors during cpu up/down

	Normally opening a file, unlinking it and then closing will have
the inode freed upon close() (provided that it's not otherwise busy and
has no remaining links, of course).  However, there's one case where that
does *not* happen.  Namely, if you open it by fhandle with cold dcache,
then unlink() and close().

	In normal case you get d_delete() in unlink(2) notice that dentry
is busy and unhash it; on the final dput() it will be forcibly evicted from
dcache, triggering iput() and inode removal.  In this case, though, we end
up with *two* dentries - disconnected (created by open-by-fhandle) and
regular one (used by unlink()).  The latter will have its reference to inode
dropped just fine, but the former will not - it's considered hashed (it
is on the ->s_anon list), so it will stay around until the memory pressure
will finally do it in.  As the result, we have the final iput() delayed
indefinitely.  It's trivial to reproduce -

void flush_dcache(void)
{
        system("mount -o remount,rw /");
}

static char buf[20 * 1024 * 1024];

main()
{
        int fd;
        union {
                struct file_handle f;
                char buf[MAX_HANDLE_SZ];
        } x;
        int m;

        x.f.handle_bytes = sizeof(x);
        chdir("/root");
        mkdir("foo", 0700);
        fd = open("foo/bar", O_CREAT | O_RDWR, 0600);
        close(fd);
        name_to_handle_at(AT_FDCWD, "foo/bar", &x.f, &m, 0);
        flush_dcache();
        fd = open_by_handle_at(AT_FDCWD, &x.f, O_RDWR);
        unlink("foo/bar");
        write(fd, buf, sizeof(buf));
        system("df .");			/* 20Mb eaten */
        close(fd);
        system("df .");			/* should've freed those 20Mb */
        flush_dcache();
        system("df .");			/* should be the same as #2 */
}

will spit out something like
Filesystem     1K-blocks   Used Available Use% Mounted on
/dev/root         322023 303843      1131 100% /
Filesystem     1K-blocks   Used Available Use% Mounted on
/dev/root         322023 303843      1131 100% /
Filesystem     1K-blocks   Used Available Use% Mounted on
/dev/root         322023 283282     21692  93% /
- inode gets freed only when dentry is finally evicted (here we trigger
than by remount; normally it would've happened in response to memory
pressure hell knows when).

Cc: stable@vger.kernel.org # v2.6.38+; earlier ones need s/kill_it/unhash_it/
Acked-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

when opening a directory we want the overlayfs inode, not one from
the topmost layer.
Reported-By: Andrey Jr. Melnikov <temnota.am@gmail.com>
Tested-By: Andrey Jr. Melnikov <temnota.am@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Cc: stable@vger.kernel.org # all branches
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>