input: vsense: round multipliers up

This will make read values consistent.

input: vsense: add middle click

Addition to mouse buttons mode; Down is middle click. Reorganized the
the logic a bit.

Signed-off-by: Urja Rannikko <urjaman@gmail.com>

mtd: omap2: performance hack

hides wakeup latency problem, but uses more power

mtd: omap2: use busy pin monitoring

mtd: omap2: get rid of omap_wait

omap_wait() is mostly just a copy of generic nand_wait(),
gpmc_nand_read() that it's doing is equivalent to normal data read.
After this change OMAP driver benefits from better tested generic wait
implementation and gains LED support.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>

mtd: omap2: fix resource leak in prefetch-busy path

If prefetch engine is busy, current code "forgets" to call
dma_unmap_single(), which results in a deadlock later, so add it.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>

ARM: OMAP: sram: fix BUG in dpll code for !PM case

_omap3_sram_configure_core_dpll is called when SDRC is reprogrammed,
which is done regardless of CONFIG_PM setting, so we always need it's
setup code too. Without this, we hit a BUG() on OMAP3 when kernel is
built without CONFIG_PM:

Reprogramming SDRC clock to 332000000 Hz
------------[ cut here ]------------
kernel BUG at arch/arm/plat-omap/sram.c:342!
Internal error: Oops - BUG: 0 [#1] ARM
...
[<c001c694>] (omap3_configure_core_dpll+0x68/0x6c) from [<c001b2dc>] (omap3_core_dpll_m2_set_rate+0x1)
[<c001b2dc>] (omap3_core_dpll_m2_set_rate+0x138/0x1b0) from [<c001a478>] (omap2_clk_set_rate+0x14/0x2)
[<c001a478>] (omap2_clk_set_rate+0x14/0x20) from [<c001c9dc>] (clk_set_rate+0x54/0x74)
[<c001c9dc>] (clk_set_rate+0x54/0x74) from [<c022b9c8>] (omap_sdrc_init+0x70/0x90)
[<c022b9c8>] (omap_sdrc_init+0x70/0x90) from [<c022f178>] (omap3pandora_init+0x11c/0x164)
[<c022f178>] (omap3pandora_init+0x11c/0x164) from [<c022849c>] (customize_machine+0x20/0x28)
[<c022849c>] (customize_machine+0x20/0x28) from [<c0225810>] (do_one_initcall+0xa0/0x16c)
[<c0225810>] (do_one_initcall+0xa0/0x16c) from [<c02259e0>] (kernel_init+0x104/0x1ac)
[<c02259e0>] (kernel_init+0x104/0x1ac) from [<c0009cec>] (kernel_thread_exit+0x0/0x8)

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>

ARM: omap2+: set IRQCHIP_SKIP_SET_WAKE for INTC interrupts.

Without an ->irq_set_wake() method in an irq_chip, calls to
enable_irq_wake() will fail.  This also causes these interrupts to not
be able to abort suspend (via check_wakeup_irqs() in late suspend.)

Currently, we don't implement ->irq_set_wake() for INTC interrupts
because they default to be wakeup enabled by setting the GRPSEL bits
in PM init.  Even though there is no ->irq_set_wake(), we want
enable_irq_wake() to succeed so these interrupts can abort suspend
when necessary.

To fix, set IRQCHIP_SKIP_SET_WAKE flag for all the INTC
interrupts which avoids trying to check irq_chip->irq_set_wake()
and failing when it doesn't exist.

Longer term, we need to implement ->irq_set_wake() for the INTC
which can manage the appropriate GRPSEL bits.

Signed-off-by: NeilBrown <neilb@suse.de>
[khilman@ti.com: rework changelog]
Signed-off-by: Kevin Hilman <khilman@ti.com>

ARM: OMAP3: PM: cleanup cam_pwrdm leftovers

commit e7410cf7 (02fdb03e69699f26e1370d0e51593dbc8a4e5265) moved
mangement of cam_pwrdm to CPUidle but left some remnants behind,
namely the call to clkcm_allo_idle() for the clockdomains in the MPU
pwrdm.  Remove these since they are not necessary and cause unwanted
latency in the idle path.

Acked-by: Tero Kristo <Tero Kristo <t-kristo@ti.com>
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>

ARM: OMAP3: PM: call pre/post transition per powerdomain

We only need to call the pre/post transtion methods when we know the
power state is changing.  First, split up the pre/post transition
calls to be per-powerdomain, and then make them conditional on whether
the power domain is actually changing states.

[notasas@gmail.com: backport to 3.2]
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>

ARM: OMAP2+: powerdomain: allow pre/post transtion to be per pwrdm

Iteration over all power domains in the idle path is unnecessary since
only power domains that are transitioning need to be accounted for.
Also PRCM register accesses are known to be expensive, so the
additional latency added to the idle path is signficiant.

In order allow the pre/post transitions to be isolated and called
per-pwrdm, change the API so passing in a specific power domain will
trigger the pre/post transtion accounting for only that specific power
domain.  Passing NULL means iterating over all power domains as is
current behavior.

[notasas@gmail.com: backport to 3.2]
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>

ARM: OMAP: clock: fix race in disable all clocks

clk_disable_unused is invoked when CONFIG_OMAP_RESET_CLOCKS=y.
Since clk_disable_unused is called as lateinitcall, there can
be more than a few workqueues executing off secondary CPU(s).
The current code does the following:
a) checks if clk is unused
b) holds lock
c) disables clk
d) unlocks

Between (a) and (b) being executed on CPU0, It is possible to
have a driver executing on CPU1 which could do a get_sync->clk_get
(and increase the use_count) of the clock which was just about
to be disabled by clk_disable_unused.

We ensure instead that the entire list traversal is protected by
the lock allowing for parent child clock traversal which could be
potentially be done by runtime operations to be safe as well.

Reported-by: Todd Poynor <toddpoynor@google.com>
Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>

ARM: OMAP2/3: VENC hwmods: Remove OCPIF_SWSUP_IDLE flag from VENC slave interface

The clocks for all DSS slave interfaces were recently changed to "dss_ick" on
OMAP2 and OMAP3, this clock can be autoidled by PRCM. The VENC interface
previously had "dss_54m_fck" as it's clock which couldn't be autoidled, and
hence the OCPIF_SWSUP_IDLE flag was needed.

Remove the OCPIF_SWSUP_IDLE flag from VENC interfaces as it's clock is
now "dss_ick".  This allows the PRCM hardware to autoidle the VENC
interface clocks when they are not active, rather than relying on the
software to do it, which can keep the interface clocks active
unnecessarily.

Signed-off-by: Archit Taneja <archit@ti.com>
[paul@pwsan.com: add a short description of the fix to the commit log]
Signed-off-by: Paul Walmsley <paul@pwsan.com>

ARM: OMAP2+: powerdomain: Wait for powerdomain transition in pwrdm_state_switch()

Commit b1cbdb00d ("OMAP: clockdomain: Wait for powerdomain to be ON
when using clockdomain force wakeup") was assuming that
pwrdm_state_switch() does wait for the powerdomain transition which is
not the case.  The missing wait for the powerdomain transition
violates the sequence which the hardware expects, which causes power
management failures on some devices.

Fix this API by adding the pwrdm_wait_transition().

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
[paul@pwsan.com: added some more details in the commit log]
Signed-off-by: Paul Walmsley <paul@pwsan.com>

ARM: OMAP3: clock data: fill in some missing clockdomains

Several clocks are missing clockdomains.  This can cause problems with
the hwmod and power management code.  Fill these in.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Matt Porter <mporter@ti.com>
Cc: Vaibhav Hiremath <hvaibhav@ti.com>

ARM: OMAP2+: ads7846_init: put gpio_pendown into pdata if it's provided

If platform data is provided by the caller gpio_pendown is put into
unused static ads7846_config structure and effectively has no effect.
Of course caller can set gpio_pendown field in platform data himself
but it seems natural to do this in ads7846_init to remove duplication.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

ARM: OMAP2+: PM debug: fix the use of debugfs_create_* API

Check the return code pointer value from debugfs_create_dir for error
or NULL.
Also added an additional check to prevent the creation of a 'suspend'
entry at the debugfs root in case a power domain directory cannot be
created.

Signed-off-by: Jean Pihet <j-pihet@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>

ARM: OMAP2+: PM: clean up omap_set_pwrdm_state()

Clean up a few different parts of omap_set_pwrdm_state():

- Remove a superfluous call to pwrdm_state_switch().  Not needed
  unless LOWPOWERSTATECHANGE is used, because the state switch code is
  called by either clkdm_sleep() or clkdm_allow_idle().

- Add code to wait for the power state transition in the OMAP4+ low
  power state change.  This is speculative, so I would particularly
  appreciate feedback on this part.

- Remove a superfluous call to pwrdm_read_pwrst().

- Update variable names to be more meaningful (hopefully) and precise.

- Fix an error path bug that would not place the clockdomain back into
  hardware-supervised idle or sleep mode if the power state could not
  be programmed.

The documentation for this function still needs major improvements;
that's left for a later patch.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Tested-by: Tero Kristo <t-kristo@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>

ARM: OMAP3: PM: remove superfluous calls to pwrdm_clear_all_prev_pwrst()

Remove some superfluous calls to pwrdm_clear_all_prev_pwrst().
pwrdm_pre_transition(), which appears a few lines after these calls,
invokes pwrdm_clear_all_prev_pwrst() on each powerdomain -- there's no
need to do it twice.

N.B.: some of us have observed that accesses to the previous
powerstate registers seem to be quite slow.  Although the writes
removed by this patch should be buffered by the write buffer, there is
a read to a PRM register immediately afterwards.  That will block the
OMAP3 MPU until all of those writes complete.  So this patch should
result in a minor performance improvement during idle entry.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Tero Kristo <t-kristo@ti.com>
[khilman@ti.com: removed a couple more for OMAP4]
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>

pandora: defconfig: enable android cruft

we'll see if anything good (or bad) comes out of it

export symbols for android cruft

staging: android: make binder a loadable module

don't see a reason for it to be built-in only,
except of needing export symbol hacks.

staging: android: asmem: make selectable as module

Seems to work that way, not tested extensively though.
Needs export symbol hacks.

Staging: android: add the code back to the build

It builds, so ship it!

Cc: Arve Hjønnevåg <arve@android.com>
Cc: Brian Swetland <swetland@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: fix mem leaks in __persistent_ram_init()

If, in __persistent_ram_init(), the call to
persistent_ram_buffer_init() fails or the call to
persistent_ram_init_ecc() fails then we fail to free the memory we
allocated to 'prz' with kzalloc() - thus leaking it.

To prevent the leaks I consolidated all error exits from the function
at a 'err:' label at the end and made all error cases jump to that
label where we can then make sure we always free 'prz'. This is safe
since all the situations where the code bails out happen before 'prz'
has been stored anywhere and although we'll do a redundant kfree(NULL)
call in the case of kzalloc() itself failing that's OK since kfree()
deals gracefully with NULL pointers and I felt it was more important
to keep all error exits at a single location than to avoid that one
harmless/redundant kfree() on a error path.

Signed-off-by: Jesper Juhl <jj@chaosbits.net>
Acked-by: Colin Cross <ccross@android.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: timed_gpio: Fix resource leak in timed_gpio_probe error paths

If gpio_request fails, we need to free all allocated resources.
Current code uses wrong array index to access gpio_data array.
So current code actually frees gpio_data[i].gpio by j times.

This patch moves the error handling code to err_out and thus improves
readability.

Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

android: make persistent_ram based drivers depend on HAVE_MEMBLOCK

m68k doesn't have memblock_reserve, which causes a build failure
with allmodconfig.  Make PERSISTENT_RAM and RAM_CONSOLE depend on
HAVE_MEMBLOCK.

Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reported-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

android, lowmemorykiller: remove task handoff notifier

The task handoff notifier leaks task_struct since it never gets freed
after the callback returns NOTIFY_OK, which means it is responsible for
doing so.

It turns out the lowmemorykiller actually doesn't need this notifier at
all.  It's used to prevent unnecessary killing by waiting for a thread
to exit as a result of lowmem_shrink(), however, it's possible to do
this in the same way the kernel oom killer works by setting TIF_MEMDIE
and avoid killing if we're still waiting for it to exit.

The kernel oom killer will already automatically set TIF_MEMDIE for
threads that are attempting to allocate memory that have a fatal signal.
The thread selected by lowmem_shrink() will have such a signal after the
lowmemorykiller sends it a SIGKILL, so this won't result in an
unnecessary use of memory reserves for the thread to exit.

This has the added benefit that we don't have to rely on
CONFIG_PROFILING to prevent needlessly killing tasks.

Reported-by: Werner Landgraf <w.landgraf@ru.ru>
Cc: stable@vger.kernel.org
Signed-off-by: David Rientjes <rientjes@google.com>
Acked-by: Colin Cross <ccross@android.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Staging: android: ram_console.c:

Fix build error when CONFIG_PRINTK is not selected.

Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Acked-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: binder: Fix use-after-free bug

binder_update_page_range could read freed memory if the vma of the
selected process was freed right before the check that the vma
belongs to the mm struct it just locked.

If the vm_mm pointer in that freed vma struct had also been rewritten
with a value that matched the locked mm struct, then the code would
proceed and possibly modify the freed vma.

Signed-off-by: Arve Hjønnevåg <arve@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ram_console: Fix section mismatches

WARNING: vmlinux.o(.text+0xfcf6e): Section mismatch in reference
from the function ram_console_driver_probe() to the function
.init.text:persistent_ram_init_ringbuffer()
The function ram_console_driver_probe() references
the function __init persistent_ram_init_ringbuffer().
This is often because ram_console_driver_probe lacks a __init
annotation or the annotation of persistent_ram_init_ringbuffer is
wrong.

Move this driver to platform_driver_probe() because ram console
devices aren't going to be added and removed at runtime. Also
shorten the probe function name since driver is redundant and
makes the function name long.

Cc: Android Kernel Team <kernel-team@android.com>
Cc: John Stultz <john.stultz@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console: honor dmesg_restrict

The Linux kernel has a setting called dmesg_restrict. When true,
only processes with CAP_SYSLOG can view the kernel dmesg logs. This
helps prevent leaking of kernel information into user space.

On Android, it's possible to bypass these restrictions by viewing
/proc/last_kmsg.

This change makes /proc/last_kmsg require the same permissions as
dmesg.

CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Nick Kralevich <nnk@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console: fix crash in ram_console_late_init

If the persistent ram buffer is not available, ram_console_late_init
would crash when dereferencing ram_console_zone.  Return early if
ram console was not initialized.

CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: persistent_ram: add notrace to persistent_ram_write

Add the notrace attribute to persistent_ram_write and the
non-ecc functions that it calls to allow persistent_ram to
be used for ftracing (only when ecc is disabled).

CC: Greg KH <gregkh@linuxfoundation.org>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: persistent_ram: make persistent_ram_write atomic

Allow persistent_ram_write to be called on multiple cpus at
the same time, as long as ecc is not in use.  Uses atomics
for the buffer->start and buffer->size counters.

[jstultz: Fix up some pr_info casting issues on 64bit]
CC: Greg KH <gregkh@linuxfoundation.org>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: persistent_ram: handle reserving and mapping memory

Replace the ioremapped memory passed in from the drivers with
a memblock_reserve and vmap.  Adds a new function,
persistent_ram_early_init, designed to be called from the machine
init_early callback, that calls memblock_remove and saves the
provided persistent ram area layout.

Drivers only pass in their struct device * and ecc settings.
Locating and mapping the memory is now handled entirely within
persistent_ram.

Also, convert ram_console to the new persistent_ram_init
parameters that only take a struct device * and ecc settings.

[jstultz: Fix pr_info casting issues on 64bit, folded two
patches as the build breaks if they are apart. Also replaced
phys_to_page() w/ pfn_to_page(addr>>PAGE_SHIFT), as phys_to_page
is only on a few arches.]
CC: Greg KH <gregkh@linuxfoundation.org>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: persistent_ram: refactor ecc support

Remove CONFIG_ANDROID_RAM_CONSOLE_ERROR_CORRECTION and related
 #ifdefs.  Also allow persistent ram zones without ecc enabled.
For some use cases, like the data portion of the upcoming
persistent_vars patches, or a persistent ftrace ringbuffer,
ecc on every update is too expensive.

CC: Greg KH <gregkh@linuxfoundation.org>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console: split out persistent ram

Split ram_console into two halves.

persistent_ram is a set of apis that handle a block of memory
that does not get erased across a reboot.  It provides functions
to fill it as a single buffer or a ring buffer, and to extract
the old data after a reboot.  It handles ecc on the data to
correct bit errors introduced during reboot.

ram_console is now a small wrapper around persistent_ram that
feeds console data into the ringbuffer, and exports the old
data to /proc/last_kmsg after a reboot.

[jstultz: Moved persistent_ram.h to staging dir]
CC: Greg KH <gregkh@linuxfoundation.org>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console: drop verbose ram_console support

Equivalent functionality can be obtained with loglevel=15

CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console: drop early buffer support

Early ramconsole is not very useful, an early crash will prevent
getting the logged data out on the next boot, and CONFIG_DEBUG_LL=y
will get the same information.  Drop it to simplify a future
refactoring.

CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console: move footer strings

Don't store the bootinfo string and the ecc status string with
the recovered old log data.  This will simplify refactoring the
persistent ram code out of the ram console code later.

CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console: set CON_ANYTIME console flag

We want to ensure that we get all the console messages, even ones
that occur while the printing CPU is not yet online.

[jstultz: tweaked commit subject line]
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Dima Zavin <dima@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: lowmemorykiller.c

Fix compiler warning about the type of the module parameter.

Cc: San Mehat <san@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android/lowmemorykiller: Don't unregister notifier from atomic context

The lowmemorykiller registers an atomic notifier for notfication of when
the task is freed.  From this atomic notifier callback, it removes the
atomic notifier via task_free_unregister().  This is incorrect because
atomic_notifier_chain_unregister() calls syncronize_rcu(), which can
sleep, which shouldn't be done from an atomic notifier.

Fix this by registering the notifier during init, and only unregister it
if the lowmemorykiller is unloaded.

Rebased to -next by Paul E. McKenney.
Rebased to -next again by Anton Vorontsov.

Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
Signed-off-by: Christian Bejram <christian.bejram@stericsson.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reported-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:android: Change type for binder_debug_no_lock switch to bool

GCC warns that module_param_named() indirectly returns a bool type value
which is different from 'int' type binder_debug_no_lock declared. Change
it to bool because it is a internal switch for debugging.

Signed-off-by: Zhengwang Ruan <ruan.zhengwang@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: timed_gpio: Removed spaces before tabs

Signed-off-by: Johannes Thumshirn <morbidrsa@googlemail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: logger: hold mutex while removing reader

The readers list is traversed under the log->mutex lock
(for example from fix_up_readers()), but the deletion of
elements from this list is not being done under this lock.

Cc: Brian Swetland <swetland@google.com>
Cc: Dima Zavin <dima@android.com>
Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

fs: Remove missed ->fds_bits from cessation use of fd_set structs internally

Stephen Rothwell reported that the following commit broke the
linux-next build:

  1fd36adcd98c: Replace the fd_sets in struct fdtable with an array of unsigned longs

Fix places where ->fds_bits needed to be removed as the core
kernel no longer uses fd_set internally for file descriptor
table management.  There are two places:

 (1) drivers/staging/android/binder.c

 (2) arch/mips/kernel/kspd.c

     Question: Should sp_cleanup() in the MIPS arch be using find_next_bit()
     or fls()?

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Ralf Bächle <ralf@linux-mips.org>
cc: Arve Hjønnevåg <arve@android.com>
Link: http://lkml.kernel.org/r/20120224105707.32170.11550.stgit@warthog.procyon.org.uk
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Wrap accesses to the fd_sets in struct fdtable

Wrap accesses to the fd_sets in struct fdtable (for recording open files and
close-on-exec flags) so that we can move away from using fd_sets since we
abuse the fd_set structs by not allocating the full-sized structure under
normal circumstances and by non-core code looking at the internals of the
fd_sets.

The first abuse means that use of FD_ZERO() on these fd_sets is not permitted,
since that cannot be told about their abnormal lengths.

This introduces six wrapper functions for setting, clearing and testing
close-on-exec flags and fd-is-open flags:

void __set_close_on_exec(int fd, struct fdtable *fdt);
void __clear_close_on_exec(int fd, struct fdtable *fdt);
bool close_on_exec(int fd, const struct fdtable *fdt);
void __set_open_fd(int fd, struct fdtable *fdt);
void __clear_open_fd(int fd, struct fdtable *fdt);
bool fd_is_open(int fd, const struct fdtable *fdt);

Note that I've prepended '__' to the names of the set/clear functions because
they require the caller to hold a lock to use them.

Note also that I haven't added wrappers for looking behind the scenes at the
the array.  Possibly that should exist too.

Signed-off-by: David Howells <dhowells@redhat.com>
Link: http://lkml.kernel.org/r/20120216174942.23314.1364.stgit@warthog.procyon.org.uk
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>

staging: android, lowmemorykiller: convert to use oom_score_adj

/proc/pid/oom_adj is deprecated and will be removed in August 2012
according to Documentation/feature-removal-schedule.txt.  Convert its
usage in the lowmemorykiller to use the new interface, oom_score_adj,
instead.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Support old drivers via preprocessor aliasing

Older out of tree drivers that were desgined to the Android Alarm
in-kernel API may not build due to the namespace collision fixed in
an earlier patch. Per Arve's suggestion, this patch provides
preprocessor macros that allow older drivers to build.

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Fixup minor pr_alarm warnings

This patch fixes the following warnings:

drivers/staging/android/alarm.c: In function ‘alarm_timer_triggered’:
drivers/staging/android/alarm.c:344: warning: format ‘%d’ expects type ‘int’, but argument 2 has type ‘long int’
drivers/staging/android/alarm.c:367: warning: format ‘%d’ expects type ‘int’, but argument 2 has type ‘long int’

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Fix bad index when canceling alarms[]

It was using ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP_MASK as an
index.

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
CC: JP Abgrall <jpa@google.com>
Change-Id: I919860cc71254453e382616bce9fd5455802cb3d
Signed-off-by: JP Abgrall <jpa@google.com>
[jstultz: Tweaked commit subject]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Update hrtimer if alarm at the head of the queue is reprogrammed

If an alarm was restarted with a value that moved it away from the head
of a queue, the hrtimer would not be updated. This would cause unnecessary
wakeups.

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Change-Id: If379f8dd92b0bdb3173bd8d057adfe0dc1d15259
Signed-off-by: Arve Hjønnevåg <arve@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Disable Android alarm driver by default

Do not enable Android alarm driver by default

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
CC: Praneeth Bajjuri <praneeth@ti.com>
Change-Id: Iff8f7a65c4eceecfd084074937c72824697b5e7f
Signed-off-by: Praneeth Bajjuri <praneeth@ti.com>
[jstultz: tweaked commit subject & msg]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Reenable android alarm driver

Now that it builds, re-enable android alarm driver in
the makefile and kconfig

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: HACK: wakelock workaround

Allow Android alarmtimer device to build while wakelocks are still
out of tree.

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Fix namespace collision with upstreamed alarmtimers

The upstreamed alarmtimers are similar but not quite 100% API
compatibile with the android in-kernel alarm api. To aid the
transition, prefix the the android in-kernel api with android_

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Fix include compile issues

The file asm/mach/time.h doesn't exist on all arches,
so include <linux/time.h>. Also linux/sysdev.h is gone
so kill it.

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Add needed module.h includes

Add module.h includes required to build

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
CC: Andy Green <andy.green@linaro.org>
Signed-off-by: Andy Green <andy.green@linaro.org>
[jstultz: Tweaked commit subject, folded two patches
into one]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Don't use save_time_delta.

Remove references to non-existant save_time_delta.

Change-Id: Iaefeca497de02fe36b7f5d79075912f6e349ec53
CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Arve Hjønnevåg <arve@android.com>
[Added commit message -jstultz]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android-alarm: Add android alarm driver & in-kernel alarm interface

Drivers can now create alarms that will use an hrtimer while the
system is running and the rtc to wake up from suspend.

CC: Arve Hjønnevåg <arve@android.com>
CC: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Arve Hjønnevåg <arve@android.com>
[Fold and move alarm driver and interface to staging,
 fix whitespace issue, drop kconfig & make file changes
 as it currently doesn't build  -jstultz]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android/lowmemorykiller: Do not kill kernel threads

LMK should not try killing kernel threads.

Suggested-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android/lowmemorykiller: No need for task->signal check

task->signal == NULL is not possible, so no need for these checks.

Suggested-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android/lowmemorykiller: Better mm handling

LMK should not directly check for task->mm. The reason is that the
process' threads may exit or detach its mm via use_mm(), but other
threads may still have a valid mm. To catch this we use
find_lock_task_mm(), which walks up all threads and returns an
appropriate task (with lock held).

Suggested-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android/lowmemorykiller: Don't grab tasklist_lock

Grabbing tasklist_lock has its disadvantages, i.e. it blocks
process creation and destruction. If there are lots of processes,
blocking doesn't sound as a great idea.

For LMK, it is sufficient to surround tasks list traverse with
rcu_read_{,un}lock().

>From now on using force_sig() is not safe, as it can race with an
already exiting task, so we use send_sig() now. As a downside, it
won't kill PID namespace init processes, but that's not what we
want anyway.

Suggested-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: logger: clarify non-update of w_off in do_write_log_from_user

Add comment to explain when w_off is not updated in case of failed second
fragment copy to buffer.

Signed-off-by: Tim Bird <tim.bird@am.sony.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: logger: clarify code in clock_interval

Add commentary, rename the function and make the code easier to read.

Signed-off-by: Tim Bird <tim.bird@am.sony.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: logger: reorder prepare_to_wait and mutex_lock

If mutex_lock waits, it will return in state TASK_RUNNING,
rubbing out the effect of prepare_to_wait().

Signed-off-by: Tim Bird <tim.bird@am.sony.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: logger: simplify and optimize get_entry_len

Make this code slightly easier to read, and eliminate calls
to sub-routines.  Some of these were previously optimized away
by the compiler, but one memcpy was not.

In my testing, this makes the code about 20% smaller, and
has no sub-routine calls and no branches (on ARM).

v2 of this patch is, IMHO, easier to read than v1. Compared to
that patch it uses __u8 instead of unsigned char, for
consistency with the __u16 val data type, simplifies the
conditional expression, adds a another comment, and
moves a common statement out of the if.

Signed-off-by: Tim Bird <tim.bird@am.sony.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: logger: Change logger_offset() from macro to function

Convert to function and add log as a parameter, rather than relying
on log in the context of the macro.

Signed-off-by: Tim Bird <tim.bird@am.sony.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging/android/timed_gpio.c: Stlye fixes

Just made it more neat and not bother scripts/checkpatch.pl

Signed-off-by: Tracey Dent <tdent48227@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging/android/ashmem.c: Cleanups

Minor cleanups that consist of removal of a whitespace and
make file_operations const.

Signed-off-by: Tracey Dent <tdent48227@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: ram_console: Use resource_size function

The semantic patch that makes this change is available
in scripts/coccinelle/api/resource_size.cocci.

More information about semantic patching is available at
http://coccinelle.lip6.fr/

Signed-off-by: Thomas Meyer <thomas@m3y3r.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers:staging:android Typos: fix some comments that have typos in them.

Below is a patch that fixes some typos in some comments.

Signed-off-by: Justin P. Mattock <justinmattock@gmail.com>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ram_console use kmemdup instead of kmalloc

Replace kmalloc + memcpy will kmemdup in ram_console_late_init

Signed-off-by: Greg Dietsche <Gregory.Dietsche@cuw.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android/ram_console: Don't build on arches w/o ioremap

This patch fixes UML build:

    CC      drivers/staging/android/ram_console.o
  drivers/staging/android/ram_console.c: In function
  'ram_console_driver_probe':
  drivers/staging/android/ram_console.c:358:2: error: implicit declaration
  of function 'ioremap' [-Werror=implicit-function-declaration]
  cc1: some warnings being treated as errors
  make[3]: *** [drivers/staging/android/ram_console.o] Error 1

Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: lowmemorykiller: Don't wait more than one second for a process to die

If a process forked and the child process was killed by the
lowmemorykiller, the lowmemory killer would be disabled until
the parent process reaped the child or it died itself.

Signed-off-by: Arve Hjønnevåg <arve@android.com>
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: binder: Fix crashes when sharing a binder file between processes

Opening the binder driver and sharing the file returned with
other processes (e.g. by calling fork) can crash the kernel.
Prevent these crashes with the following changes:
- Add a mutex to protect against two processes mmapping the
  same binder_proc.
- After locking mmap_sem, check that the vma we want to access
  (still) points to the same mm_struct.
- Use proc->tsk instead of current to get the files struct since
  this is where we get the rlimit from.

Signed-off-by: Arve Hjønnevåg <arve@android.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: Remove pmem driver

Addroid pmem driver is no longer used in any of the Android products.
This patch removes pmem driver from Android staging area

Reference: https://lkml.org/lkml/2012/1/23/183

Signed-off-by: Shuah Khan <shuahkhan@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: android: binder: Don't call dump_stack in binder_vma_open

If user-space partially unmaps the driver, binder_vma_open
would dump the kernel stack. This is not a kernel bug however
and will be treated as if the whole area was unmapped once
binder_vma_close gets called.

Signed-off-by: Arve Hjønnevåg <arve@android.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: android: fixed 80 characters warnings in lowmemorykiller.c

This patch fixes some 80 chatacters limit warnings in the lowmemorykiller.c file

Signed-off-by: Marco Navarra <fromenglish@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: android: fixed a space warning in binder.h

This patch fixes a simple tab-space warning in binder.h found by checkpatch tool

Signed-off-by: Marco Navarra <fromenglish@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

ashmem: Whitespace cleanups

Fixes checkpatch warnings with the ashmem.c file

CC: Brian Swetland <swetland@google.com>
CC: Colin Cross <ccross@android.com>
CC: Arve Hjønnevåg <arve@android.com>
CC: Dima Zavin <dima@android.com>
CC: Robert Love <rlove@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

ashmem: Fix arguments to ashmem_shrink

The arguments to shrink functions have changed, update
ashmem_shrink to match.

Signed-off-by: Colin Cross <ccross@android.com>
[jstultz: tweaked commit subject]
CC: Brian Swetland <swetland@google.com>
CC: Colin Cross <ccross@android.com>
CC: Arve Hjønnevåg <arve@android.com>
CC: Dima Zavin <dima@android.com>
CC: Robert Love <rlove@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

ashmem: Support lseek(2) in ashmem driver

Signed-off-by: Bjorn Bringert <bringert@android.com>
[jstultz: tweaked commit subject]
CC: Brian Swetland <swetland@google.com>
CC: Colin Cross <ccross@android.com>
CC: Arve Hjønnevåg <arve@android.com>
CC: Dima Zavin <dima@android.com>
CC: Robert Love <rlove@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

ashmem: Update arguments of shrinker for 2.6.35

Signed-off-by: Colin Cross <ccross@google.com>
CC: Brian Swetland <swetland@google.com>
CC: Colin Cross <ccross@android.com>
CC: Arve Hjønnevåg <arve@android.com>
CC: Dima Zavin <dima@android.com>
CC: Robert Love <rlove@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

ashmem: Fix ASHMEM_SET_PROT_MASK.

Signed-off-by: Arve Hjønnevåg <arve@android.com>
CC: Brian Swetland <swetland@google.com>
CC: Colin Cross <ccross@android.com>
CC: Arve Hjønnevåg <arve@android.com>
CC: Dima Zavin <dima@android.com>
CC: Robert Love <rlove@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

ashmem: Implement read(2) in ashmem driver

Signed-off-by: Bjorn Bringert <bringert@android.com>
[jstultz: Tweaked commit subject]
CC: Brian Swetland <swetland@google.com>
CC: Colin Cross <ccross@android.com>
CC: Arve Hjønnevåg <arve@android.com>
CC: Dima Zavin <dima@android.com>
CC: Robert Love <rlove@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

ashmem: Anonymous shared memory subsystem

The anonymous shared memory (ashmem) subsystem provides a
Unix-y,file-based shared memory interface to user-space. It
works like anonymous memory (e.g. mmapping fd=0) except if
you share the file descriptor via the usual means, you will
share the mapping. The shared memory can be accessed via both
mmap or file I/O. The backing store is a simple shmem file.

Additionally, ashmem introduces the concept of page pinning.
Pinned pages (the default) behave like any anonymous memory.
Unpinned pages are available to the kernel for eviction during
VM pressure. When repinning the pages, the return value
instructs user-space as to any eviction. In this manner,
user-space processes may implement caching and similar
resource management that efficiently integrates with kernel
memory management.

Signed-off-by: Robert Love <rlove@google.com>
ashmem: Don't install fault handler for private mmaps.

Ashmem is used to create named private heaps. If this heap is backed
by a tmpfs file it will allocate two pages for every page touched.
In 2.6.27, the extra page would later be freed, but 2.6.29 does not
scan anonymous pages when running without swap so the memory is not
freed while the file is referenced. This change changes the behavior
of private ashmem mmaps to match /dev/zero instead tmpfs.

Signed-off-by: Arve Hjønnevåg <arve@android.com>
ashmem: Add common prefix to name reported in /proc/pid/maps

Signed-off-by: Arve Hjønnevåg <arve@android.com>
ashmem: don't require a page aligned size

This makes ashmem more similar to shmem and mmap, by
not requiring the specified size to be page aligned,
instead rounding it internally as needed.

Signed-off-by: Marco Nelissen <marcone@android.com>
[jstultz: Improved commit subject and included patch description
from rlove. Also moved ashmem files to staging dir, and reworked
code to avoid touching mm/shmem.c while we're in staging.]
CC: Brian Swetland <swetland@google.com>
CC: Colin Cross <ccross@android.com>
CC: Arve Hjønnevåg <arve@android.com>
CC: Dima Zavin <dima@android.com>
CC: Robert Love <rlove@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: add pmem driver

This adds the Android pmem driver to the staging tree.

[At this point in time, it is dependent on the ARM platform, due to some
build issues that require it. - gregkh]

Signed-off-by: Rebecca Schultz <rschultz@google.com>
Signed-off-by: Rebecca Schultz Zavin <rebecca@android.com>
Signed-off-by: Dima Zavin <dima@android.com>
Signed-off-by: Jamie Gennis <jgennis@google.com>
Cc: Brian Swetland <swetland@google.com>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Colin Cross <ccross@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: switch: minor code formatting cleanups

This fixes a number of minor space issues in the Android switch code.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: switch: switch class and GPIO drivers.

This adds the Android switch driver code to the staging tree.

[Note, this code was located in drivers/switch/ in the Android kernel
releases, but as that api wasn't generally accepted, and the interface
is working toward changing to the newly proposed extcon inteface, this
driver was placed here until the extcon code is merged into mainline and
the Android userspace code is converted over to using it. - gregkh]

Signed-off-by: Arve Hjønnevåg <arve@android.com>
Signed-off-by: Mike Lockwood <lockwood@android.com>
Cc: MyungJoo Ham <myungjoo.ham@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Donggeun Kim <dg77.kim@samsung.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: MyungJoo Ham <myungjoo.ham@gmail.com>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: NeilBrown <neilb@suse.de>
Cc: Morten CHRISTIANSEN <morten.christiansen@stericsson.com>
Cc: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: android: fixed white spaces coding style issue in logger.c

This patch fixes some space-before-tabs warnings found by checkpatch tool on the staging android driver file logger.c

Signed-off-by: Marco Navarra <fromenglish@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: ram_console: pass in a boot info string

Allow the board file to pass a boot info string through the
platform data that is appended to the /proc/last_kmsg file.

[moved the .h file to drivers/staging/android/ to be self-contained - gregkh]

Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

android: logger: bump up the logger buffer sizes

(port from common android-2.6.39
  commit: 11430f16545205c614dd5bd58e4a7ee630fc0f9f)

events: (no change, 256)
main: 64 -> 256
radio: 64 -> 256
system: 64 -> 256

Signed-off-by: JP Abgrall <jpa@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

android: lowmemorykiller: Fix arguments to lowmem_shrink

The arguments to shrink functions have changed, update
lowmem_shrink to match.

Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: lowmemorykiller: Ignore shmem pages in page-cache

Signed-off-by: Arve Hjønnevåg <arve@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: lowmemorykiller: Update arguments of shrinker for 2.6.35

Signed-off-by: Colin Cross <ccross@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: android: lowmemorykiller: Remove bitrotted codepath

Now that we're murder-synchronous, this code path will never be
called (and if it does, it doesn't tell us anything useful other
than we killed a task that was already being killed by somebody
else but hadn't gotten its' signal yet)

Signed-off-by: San Mehat <san@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>