staging: sm750fb: lynxfb_pci_probe: remove some pr_info

Several pr_info statements in lynxfb_pci_probe seem like debug leftovers
and may be removed.

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: introduce sm750fb_frambuffer_alloc

Split framebuffer allocation and registration into a dedicated function
to simplify lynxfb_pci_probe

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: introduce sm750fb_frambuffer_release

Use a function to unregister framebuffer info and release its resources.

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: replace dual member of sm750_dev with fb_count

Will be used in futher refactoring of driver _probe and _remove methods.

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: lynxfb_pci_remove: remove unused variable

The par variable in lynxfb_pci_remove is only assigned a value and never
used afterwards. Remove it.

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: lynxfb_pci_probe: return actual errors

The lynxfb_pci_probe always returned -ENODEV in case of error. Modify it
so that actual error code will be propogated to the caller.

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: disable PCI device if lynxfb_pci_probe fails

In case of error during lynxfb_pci_probe, the function returned without
calling pci_disable_device. Fix it by adding pci_disable_device on the
error cleanup path.

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/rdma/hfi1: Fix Xmit Wait calculation

Total XMIT wait needs to sum the xmit wait values of all the VLs not just
those requested in the query.  Also, make the algorithm used for both
PortStatus and PortDataCounters the same.

Reviewed-by: Arthur Kepner <arthur.kepner@intel.com>
Reviewed-by: Breyer, Scott J <scott.j.breyer@intel.com>
Signed-off-by: Ira Weiny <iweiny@gmail.com>
Signed-off-by: Jubin John <jubin.john@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/rdma/hfi1: Clean up comments

Clean up comments by deleting numbering and terms internal to Intel.

The information on the actual bugs is not deleted.

Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Edward Mascarenhas <edward.mascarenhas@intel.com>
Signed-off-by: Jubin John <jubin.john@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rdma: hfi1: diag: constify hfi1_filter_array structure

The hfi1_filter_array structure is never modified, so declare it as const.

Done with the help of Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Acked-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/rdma/hfi1: add dd_dev_dbg

To be used in future patches add dd_dev_dbg.  dd_* functions properly decode
the hfi1_devdata structure used throughout the driver

Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/rdma/hfi1: set Gen3 half-swing for integrated devices

Correctly set half-swing for integrated devices.  A0 needs all fields set for
CcePcieCtrl.  B0 and later only need a few fields set.

Reviewed-by: Stuart Summers <john.s.summers@intel.com>
Signed-off-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: drop kbuild workaround dummy module from top dir

This is essentially a revert of 86f69fe9c069dd8608d238581eea259caa1dfc99
("Staging: workaround build system bug"), but to call it a revert would
imply it was wrong -- rather it seems more like it is just no longer
required anymore.

It doesn't list the full details of the failing use case, but the most
obvious would appear to be a "make allnoconfig" that subsequently had set
just CONFIG_STAGING=y in the .config file, to give:

  paul@builder:~/git/linux-head$ grep STAGING ../staging-allno/.config
  CONFIG_STAGING=y
  # CONFIG_STAGING_MEDIA is not set
  # CONFIG_STAGING_BOARD is not set
  paul@builder:~/git/linux-head$

After building this .config (on ARM, just to be different), we see that
built-in.o is created, and the final vmlinux links OK:

  paul@builder:~/git/linux-head$ ls -l ../staging-allno/drivers/staging/built-in.o
  -rw-rw-r-- 1 paul paul 257 Nov 16 18:06 ../staging-allno/drivers/staging/built-in.o
  paul@builder:~/git/linux-head$ file ../staging-allno/drivers/staging/built-in.o
  ../staging-allno/drivers/staging/built-in.o: ELF 32-bit LSB relocatable, ARM, version 1, not stripped
  paul@builder:~/git/linux-head$ nm ../staging-allno/drivers/staging/built-in.o
  nm: ../staging-allno/drivers/staging/built-in.o: no symbols
  paul@builder:~/git/linux-head$ ls -l ../staging-allno/vmlinux
  -rwxrwxr-x 1 paul paul 1236326 Nov 16 18:07 ../staging-allno/vmlinux
  paul@builder:~/git/linux-head$

I also tested an "allmodconfig" and did not see any problems there
either.  Switching back to x86-64 and testing several things there didn't
show any issues either.  So it appears we do not need to carry the
workaround in tree any longer.

Cc: linux-kbuild@vger.kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Michal Marek <mmarek@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

STAGING: COMEDI: Wrap line over 80 characters

This patch wraps lines over 80 characters.

Signed-off-by: Pablo G. Gallardo <pggllrd@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

STAGING: COMEDI: Remove unnecessary typecast of c90 int constant

This patch removes unnecessary typecast of c90 int constant.

Signed-off-by: Pablo G. Gallardo <pggllrd@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_mio_common: add finite regeneration to dio output

This patch continues the implementation of reinterpreting stop_arg when
stop_src == TRIG_NONE for national instruments cdio output on e/m-series
devices.  This is part of a series of patches that allow a user to have a
specific buffer repeated as-is indefinitely.  The contents of the DMA
buffer can be left static or changed by the user via mmap access to the DMA
buffer.  If the contents are changed by the user, additional munging is not
performed by the driver and only a single call to
comedi_mark_buffer_written should be done.  The original behavior is
preserved when stop_arg == 0, as would be the prior use case.

As opposed to analog output, this patch is relatively simple.  First, the
digital output capabilities are much more limited/simple as compared to the
analog output device on NI e/m-series hardware, and second, this patch
relies on changes made with the earlier patch to accomplish limiting the
DMA buffer transfer.

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_mio_common: adds finite regeneration to AO output

This patch implements for analog output the reinterpretation of stop_arg
when stop_src == TRIG_NONE to allow the user to specify the length of the
buffer that should be repeated.  The intent is to allow a user to have a
specific buffer repeated as-is indefinitely.  The contents of the DMA
buffer can be left static or changed by the user via mmap access to the DMA
buffer.  If the contents are changed by the user, additional munging is not
performed by the driver and only a single call to
comedi_mark_buffer_written should be done.

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_mio_common: Cleans up/clarifies ni_ao_cmd

This patch implements ni_ao_cmd much more closely organized like NI MHDDK
examples and DAQ-STC pseudo-code.  Adds comments with some more specific
references to the DAQ-STC.

For stop_src==TRIG_NONE (continuous output mode of entire buffer), the
count for the UC counter was corrected to represent the maximum count
possible (0xffffff).  Prior behavior for stop_src=TRIG_NONE did not
actually follow the DAQ-STC.  Furthermore, stop_src==TRIG_NONE now
correctly uses code specialized for either m-series or e-series devices.

It should be noted that stop_src==TRIG_NONE does _not_ with this patch
(or with prior behavior in ni_mio_common) actually implement true
continuous output.  Rather, the output is simply configured to operate as a
single buffer output, but where the buffer is as large as is possible with
NI-STC hardware.

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_mio_common: make more bits in ao_cmd1 reg be stateful

Bits NISTC_AO_CMD1_DAC0_UPDATE_MODE and NISTC_AO_CMD1_DAC1_UPDATE_MODE are
now saved in the local copy of the AO_CMD1 register.  This is more
appropriate than prior methods of setting these bits specifically _both_
in the ni_ao_cmd configuration function _and_ the ni_ao_inttrig trigger
function.  With this patch, the bits are only specifically called out now
in the ni_ao_cmd configuration function.  In the ni_ao_inttrig trigger
function, only the UI_ARM, UC_ARM, BC_ARM bits of the ao_cmd1 register are
specifically called out.  Each of these bits is a strobe bit, while the
DAC[0-1]_UPDATE_MODE bits are simple write bits.

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_mio_common: Cleans up/clarifies ni_ao_reset

This patch implements ni_ao_reset much more closely organized like NI MHDDK
examples and DAQ-STC pseudo-code.  Adds comments with some more specific
references to the DAQ-STC.

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: mite: enable continuous regeneration of finite samples

This change enables the mite DMA controller to only transfer the amount of
data needed by a command.  By default, the old behavior of transferring the
entire comedi DMA data buffer is still in effect.  These changes allow a
command to only transmit a limited portion of that data buffer as needed.

This patch begins to reinterprets stop_arg when stop_src == TRIG_NONE to
allow the user to specify the length of the buffer that should be repeated.
The intent is to allow a user to have a specific buffer repeated as-is
indefinitely.  The contents of the DMA buffer can be left static or changed
by the user via mmap access to the DMA buffer.  If the contents are changed
by the user, additional munging is not performed by the driver and only a
single call to comedi_mark_buffer_written should be done.

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_mio_common: use CR_CHAN more consistently

Generally, the CR_CHAN macro is/should be used to access the relevant bits
for channel identification in cmd->*_arg when the corresponding
cmd->*_src==TRIG_EXT, including cmd->convert_arg in this case.

This patch does not fix a bug per se, as NISTC_AI_MODE1_CONVERT_SRC() already
masks the value sufficiently, but using CR_CHAN() here makes the code clearer as
it avoids passing some irrelevant bits to NISTC_AI_MODE1_CONVERT_SRC() in the
first place.

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_tiocmd: change mistaken use of start_src for start_arg

This fixes a bug in function ni_tio_input_inttrig().  The trigger number
should be compared to cmd->start_arg, not cmd->start_src.

Fixes: 6a760394d7eb ("staging: comedi: ni_tiocmd: clarify the cmd->start_arg validation and use")
Cc: <stable@vger.kernel.org> # 3.17+
Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_pcimio: Adds PXI-6251 to supported boards

Signed-off-by: Spencer E. Olson <olsonse@umich.edu>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: dt2801: Prefer using the BIT macro

As suggested by checkpatch.pl, this patch replaces bit shifting on 1 with the
BIT(x) macro.

Signed-off-by: Bhaktipriya Shridhar <bhaktipriya96@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

ion:synchronize debugfs callback and ion_client_destroy

There are race condition B/T ion_client_destroy and debugfs callbacks.
Let's use a mutex to synchronize them.

Signed-off-by: Neil Zhang <neilzhang1123@hotmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/ion: Add support to get ion handle from dma buf

Currently we can only import dma buf fd's to get ion_handle.
Adding support to import dma buf handles to support kernel
specific use cases.

An example use case is in linux platforms such as Tizen, in which
DRM-GEM is used for buffer management for graphics. It has gem_handle
corresponding to a buffer and uses gem_name for sharing the buffer
with other processes. However,it also uses dma_buf fd for 3d operations.
For wayland, there are multiple calls for gem_handle to dma_buf fd
conversion. So, we store dma_buf associated with buffer. But, there is
no api for getting ion_handle from dma_buf. This patch exposes api to
retrieve the ion handle from dma_buf for similar use cases. With this
patch, we can integrate ION within DRM-GEM for buffer management and
dma_buf sharing.

Signed-off-by: Rohit kumar <rohit.kr@samsung.com>
Reviewed-by: Laura Abbott <labbott@redhat.com>
Reviewed-by: Sumit Semwal <sumit.semwal@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove len field from struct sync_fence_info

After removing driver_data struct sync_fence_info has now a fixed size,
thus it doesn't need any field to tell its size, it is already known.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove driver_data from struct sync_fence_info

It is unclear in what situations driver_data should be used thus better do
not upstream it for now. If a need arises in the future a discussion can
be started to re-add it.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: rename sync_file_info_data to sync_file_info

info_data is a bit redundant, let's keep it as only sync_file_info. It is
also smaller.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: rename sync_pt_info to sync_fence_info

As struct sync_pt doesn't exist anymore it is a good idea remove any
reference to it in the sync_framework. sync_pts were replaced directly by
fences and here we rename it to sync_fence_info to let the fence namespace
clean.

v2: rename fence_info to sync_fence_info (Maarten)

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove SYNC_WAIT ioctl

This ioctl is replicating the work of poll() syscall so let's take the
opportunity that this is still on staging tree and remove the duplication
and force new users to use the poll() standard interface.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: Android: memory allocation style change in ion_page_pool.c

This is a patch to ion_page_pool.c that changes a memory allocation
style issue as found by checkpatch.pl.

Signed-off-by: Ben Marsh <bmarsh94@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove sync_fence_create_dma()

With the removal of struct sync_pt sync_fence_create_dma() now takes
the same arguments as sync_fence_create() so let's keep only
sync_fence_create().

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove pointless sync_timeline_signal at destroy phase

All changes to timeline value come through the user via
sync_timeline_signal() calls. When sync_timeline_destroy() is called no
changes on timeline->value happens hence call sync_timeline_signal() with
no increment is pointless.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove unused var from sync_timeline_signal()

signaled_pts is not used in this function.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove struct sync_pt

struct sync_pt was just wrapping around struct fence and creating an
extra abstraction layer. The only two members of struct sync_pt, child_list
and active_list, were moved to struct fence in an earlier commit. After
removing those two members struct sync_pt is nothing more than struct
fence, so remove it all and use struct fence directly.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: rename 'sync_pt' to 'fence' in struct sync_fence_cb

'sync_pt' is actually declared as struct fence so to make the name means
its type we rename it to 'fence'.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: rename sync_fence to sync_file

sync_file has a more close meaning to what a sync_fence really, a struct
that represent a file that can be used by userspace to get information on
a fence, or wait for it to be signaled.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: move SW_SYNC_USER to a debugfs file

This remove CONFIG_SW_SYNC_USER and instead compile the sw_sync file into
debugpfs under <debugfs>/sync/sw_sync.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: create a 'sync' dir for debugfs information

Creates the 'sync' dir on debugfs root dir and move the 'sync' file
to sync/info. This is the preparation to add more debug info and control.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: remove not used sync_timeline ops

.dup and .compare are not used by the sync framework, so remove them
from sw_sync.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: sync: remove interfaces that are not used

These interfaces are not used nor have plans to be used in the near
future so remove them for a cleaner solution before de-staging the sync
framework.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/android: fix sync framework documentation

Updates comments about functions and structures.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

android: binder: Sanity check at binder ioctl

Sanity check at binder ioctl function,
Only allow the shared mm_struct to use the same binder-object
to do binder operate.

And add proc->vma_vm_mm = current->mm at the open function.
The libbinder do ioctl before mmap called.

V2: Fix compile error for error commit
V3: Change the condition to proc->vma_vm_mm

Signed-off-by: Chen Feng <puck.chen@hisilicon.com>
Signed-off-by: Wei Dong <weidong2@hisilicon.com>
Signed-off-by: Junmin Zhao <zhaojunmin@huawei.com>
Reviewed-by: Zhuangluan Su <suzhuangluan@hisilicon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ion: Set the length of the DMA sg entries in buffer

ion_buffer_create() will allocate a buffer and then create a DMA
mapping for it, but it forgot to set the length of the page entries.

Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Signed-off-by: Jon Medhurst <tixy@linaro.org>
Acked-by: Laura Abbott <labbott@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ion : Donnot wakeup kswapd in ion system alloc

Since ion alloc can be called by userspace,eg gralloc.
When it is called frequently, the efficiency of kswapd is
to low. And the reclaimed memory is too lower. In this way,
the kswapd can use to much cpu resources.

With 3.5GB DMA Zone and 0.5 Normal Zone.

pgsteal_kswapd_dma 9364140
pgsteal_kswapd_normal 7071043
pgscan_kswapd_dma 10428250
pgscan_kswapd_normal 37840094

With this change the reclaim ratio has greatly improved
18.9% -> 72.5%

Signed-off-by: Chen Feng <puck.chen@hisilicon.com>
Signed-off-by: Lu bing <albert.lubing@hisilicon.com>
Reviewed-by: Laura Abbott <labbott@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: android: Fix brace coding style warning in sync_debug.c

This is a patch to the sync_debug.c file that rectifies a brace warning
that was found with the checkpatch.pl tool

Signed-off-by: Bopamo Osaisai <bopamo@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ion: Set minimum carveout heap allocation order to PAGE_SHIFT

In carveout heap, change minimum allocation order from 12 to
PAGE_SHIFT. After this change each bit in bitmap (genalloc -
General purpose special memory pool) represents one page size
memory.

Cc: sprd-ind-kernel-group@googlegroups.com
Cc: sanjeev.yadav@spreadtrum.com
Cc: Colin Cross <ccross@android.com>
Cc: Android Kernel Team <kernel-team@android.com>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Signed-off-by: Rajmal Menariya <rajmal.menariya@spreadtrum.com>
[jstultz: Reworked commit message]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Acked-by: Laura Abbott <labbott@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lowmemorykiller: Make default lowmemorykiller debug message useful

lowmemorykiller debug messages are inscrutable and mostly useful
for debugging the lowmemorykiller, not explaining why a process
was killed.  Make the messages more useful by prefixing them
with "lowmemorykiller: " and explaining in more readable terms
what was killed, who it was killed for, and why it was killed.

The messages now look like:
[   76.997631] lowmemorykiller: Killing 'droid.gallery3d' (2172), adj 1000,
[   76.997635]    to free 27436kB on behalf of 'kswapd0' (29) because
[   76.997638]    cache 122624kB is below limit 122880kB for oom_score_adj 1000
[   76.997641]    Free memory is -53356kB above reserved

A negative number for free memory above reserved means some of the
reserved memory has been used and is being regenerated by kswapd,
which is likely what called the shrinkers.

Cc: Android Kernel Team <kernel-team@android.com>
Cc: Greg KH <gregkh@linuxfoundation.org>
Signed-off-by: Colin Cross <ccross@android.com>
[jstultz: Minor checkpatch tweaks]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ashmem: Add missing include

Include <linux/types.h> into ashmem.h to ensure referenced types
are defined

Cc: Android Kernel Team <kernel-team@android.com>
Cc: Greg KH <gregkh@linuxfoundation.org>
Signed-off-by: Rom Lemarchand <romlem@android.com>
[jstultz: Minor commit message tweaks]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ashmem: Avoid deadlock with mmap/shrink

Both ashmem_mmap and ashmem_shrink take the ashmem_lock. It may
be possible for ashmem_mmap to invoke ashmem_shrink:

-000|mutex_lock(lock = 0x0)
-001|ashmem_shrink(?, sc = 0x0) <--- try to take ashmem_mutex again
-002|shrink_slab(shrink = 0xDA5F1CC0, nr_pages_scanned = 0, lru_pages
-002|=
-002|124)
-003|try_to_free_pages(zonelist = 0x0, ?, ?, ?)
-004|__alloc_pages_nodemask(gfp_mask = 21200, order = 1, zonelist =
-004|0xC11D0940,
-005|new_slab(s = 0xE4841E80, ?, node = -1)
-006|__slab_alloc.isra.43.constprop.50(s = 0xE4841E80, gfpflags =
-006|2148925462, ad
-007|kmem_cache_alloc(s = 0xE4841E80, gfpflags = 208)
-008|shmem_alloc_inode(?)
-009|alloc_inode(sb = 0xE480E800)
-010|new_inode_pseudo(?)
-011|new_inode(?)
-012|shmem_get_inode(sb = 0xE480E800, dir = 0x0, ?, dev = 0, flags =
-012|187)
-013|shmem_file_setup(?, ?, flags = 187)
-014|ashmem_mmap(?, vma = 0xC5D64210) <---- Acquire ashmem_mutex
-015|mmap_region(file = 0xDF8E2C00, addr = 1772974080, len = 233472,
-015|flags = 57,
-016|sys_mmap_pgoff(addr = 0, len = 230400, prot = 3, flags = 1, fd =
-016|157, pgoff
-017|ret_fast_syscall(asm)
-->|exception
-018|NUR:0x40097508(asm)
---|end of frame

Avoid this deadlock by using mutex_trylock in ashmem_shrink; if the mutex
is already held, do not attempt to shrink.

Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Android Kernel Team <kernel-team@android.com>
Reported-by: Matt Wagantall <mattw@codeaurora.org>
Reported-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org>
Reported-by: Osvaldo Banuelos <osvaldob@codeaurora.org>
Reported-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Signed-off-by: Laura Abbott <lauraa@codeaurora.org>
[jstultz: Minor commit message tweaks]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: add __iomem for io_base and registers

This removes a few Sparse warnings.

Signed-off-by: Hugo Camboulive <hugo.camboulive@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: replace multiple if..else with table lookup

Replace multiple if..else if..statements with simple table lookup in two
functions.

Signed-off-by: Hari Prasath Gujulan Elango <hgujulan@visteon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: fix error comparison

device_create() returns ERR_PTR on error, it does not return NULL.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove 2nd forward declaration of struct most_aim

This patch removes the second forwared declaration of struct most_aim.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: add statistics for dropped packets

This patch adds a counter for dropped packets. It needed for statistical
analysis.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: rearrange function aim_write

This patch straightens and rearranges the code of function aim_write()
of module aim-cdev.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove stacked_mbo

This patch makes use of kfifo_peek and kfifo_skip, which renders the
variable stacked_mbo useless. It is therefore removed.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: change type of access_ref

This patch changes the type of the access reference from atomit_t to int.
It is needed, because the reference variable is secured by synchronization
locks and does not need to be atomic anymore.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: fix race conditions

This patch fixes race conditions that might emerge from functions
aim_open, aim_close, aim_read, aim_write and aim_disconnect_channel
within module cdev.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: rename variable channel

This patch renames the variable 'channel' to 'c'. This is needed to have
the code look more homogeneous and to prevent format violations.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: fix retrieval of buffer availability

This patch fixes the function channel_has_mbo that delivers the false
information in case two AIMs are using the same tx channel.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: encapsulate shared code

This patch encapsulates shared code. It therefore creates the new functions
stop_channel and destroy_cdev.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove redundant call to wake_up_interruptible

This patch prevents the cdev module from rousing the channel wait queue in
case the channel is about to be closed. It is safe to do so, because the
application can not be waiting within read or write and at the same time
be calling close.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove redundant mutexes

This patch removes the mutexes stop_task_mutex and deregister mutex,
since they can safely be left out.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove code to destroy channel

This patch removes unnecessary code to destroy channel objects. It is
needed, because function most_stop_channel, which is indirectly
triggered by function most_deregister_interface, already destroys the
channels.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove reference counter

This patch removes the unnecessary reference conter mod_ref.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove tainted flag

This patch removes the atomic tainted flag. It is needed to get rid of
logical overhead.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: move channel disconnect to function most_deregister_interface

This patch moves the code that disconnects linked channels. It is needed
to have cleaning things up done right.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: move mutex

This patch removes mutex from code that doesn't need any locking.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: move initialization of pointer

This patch makes function store_add_link initialize the pointer to an AIM
right before the channel is probed. It is needed, the AIM may already call
most_start_channel while probe_channel is still running. At this point the
pointer to the AIM must not be NULL.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: move call to disconnect_channel callback

This patch invokes AIM's disconnect_channel callback before the
corresponding pointers are re-initialized to NULL.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: add missing call to ida_simple_remove

This patch adds two missing calls to function ida_simpel_remove.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove function destroy_most_c_obj

This patch removes the function destroy_most_c_obj and executes its code
within function destroy_most_inst_obj.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: use readl and writel functions

This patch makes use of functions readl and writel instead of
the __raw_* variants.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: fix tracking of MBO offset

This patch increments mbo_offs by the number of bytes that have
been copied and resets it in case a complete mbo has been transferred
to user buffer.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: fix mbo leak

This patch fixes a potential MBO leak in case function aim_read()
exits right after the MBO has been fetched from kfifo and before
it has been saved to the variable stacked_mbo.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: use min_t

This patch replaces min with min_t.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: unify types

This patch unifies variable types to get less castings.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: simplify expression

This patch replaces the ternary ?-operator with a way simpler subtraction.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: rename variables

This patch renames some variables for better readability.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: remove unnecessary keep_mbo variable

The MBO pointer stacked_mbo and the boolean variable keep_mbo are
always changed together and therefore provide the same information.
This patch removes keep_mbo and uses stacked_mbo instead.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visornic: remove useless memset

alloc_etherdev() calls alloc_netdev_mqs(), which
already uses kzalloc/vzalloc.

This clears a sparse warning :
drivers/staging/unisys/visornic/visornic_main.c:1366:15: warning: memset with byte count of 1460112

Signed-off-by: Hugo Camboulive <hugo.camboulive@gmail.com>
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: remove some dead code

queue_delayed_work() returns bool, not negative error codes.  It returns
false if the work has already been queued or true otherwise.  Since
we don't care about that, we can just remove the test.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Acked-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: unisys: fix potential format string leak

Since "name" is always used directly, force "%s" for the kthread
format string to avoid any potential format string leaks.

Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorchipset.c fixed spacing around operator

This patch fixes checkpatch.pl warning for visorchipset.c

CHECK: spaces preferred around that '*' (ctx:VxV)
+#define MAX_CONTROLVM_PAYLOAD_BYTES (1024*128)

Signed-off-by: Bhaktipriya Shridhar <bhaktipriya96@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: Only process up to budget amount of responses

>From napi documentation you should only process the amount your
budget allows, if you go over it just wait for the next napi poll
to continue.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: fix parenthesis in toolaction_show()

Fix the only fixable parenthesis alignment issue in
visorchipset.c. The rest are unworkable because of the length
of the symbol names used.

Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix return error code of wilc_deinit

It should be returned error code as -EFAULT instead of 0 when hif_drv is
NULL.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix return type of wilc_deinit

This patch changes return type of wilc_deinit from s32 to int.
The result variable gets return value from wilc_mq_send that has return
type of int. It should be changed return type of this function as well
as data type of result variable.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename pstrCfgParamVal in wilc_hif_set_cfg

This patch renames pstrCfgParamVal to cfg_param to avoid camelcase.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix return type of wilc_hif_set_cfg

This patch changes return type of wilc_hif_set_cfg from s32 to int.
The result variable gets return value from wilc_mq_send that has data
type of int. It should be changed return type of this function as well
as data type of result variable.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove warnings line over 80 characters

This patch removes the warnings reported by checkpatch.pl
for line over 80 characters.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unnecessary log message using GENERIC_DBG tag

This patch remove unnecessary log message using GENERIC_DBG tag.
Needed log messages has changed netdev_dbg from PRINT_D.
And removes variables that were used to log message.

Signed-off-by: Chris Park <chris.park@atmel.com>
Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unused debug tag

This patch remove unused debug tag INT_DBG, LOCK_DBG, BUS_DBG,
MEM_DBG and FIRM_DBG

Signed-off-by: Chris Park <chris.park@atmel.com>
Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unused log message using the RX_DBG tag

This patch remove unused log message using the RX_DBG tag

Signed-off-by: Chris Park <chris.park@atmel.com>
Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unnecessary wilc_rx_complete function

This patch remove unnecessary wilc_rx_complete function
because this function only print unused log message.
remove unused has_packet variable nomore.

Signed-off-by: Chris Park <chris.park@atmel.com>
Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: renames u64Tsf of connect_resp_info structure

This patch renames u64Tsf variable to tsf_hi.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: renames strRssi of connect_resp_info structure

This patch renames strRssi variable to str_rssi.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>