staging: unisys: visorbus: remove controlvm_respond_chipset_init prototype

Remove the unneeded controlvm_respond_chipset_init prototype by
moving the functions controlvm_init_response and
controlvm_respond_chipset_init up in the file.

Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <Timothy.Sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: remove prototype controlvm_respond

Remove the unneeded prototype for controlvm_respond.

Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <Timothy.Sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: move parahotplug_process_list

The function parahotplug_process_list needs to be moved lower in the file
to avoid extraneous function prototypes.

Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <Timothy.Sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: move read_controlvm_event

The function read_controlvm_event needs to be moved lower in the file
to avoid extraneous function prototypes.

Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <Timothy.Sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: move handle_command

Handle_command gets moved lower in the file to avoid the need for
extraneous function prototypes.

Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <Timothy.Sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: reorder controlvm_periodic_work

Move controlvm_periodic_work lower in the file to avoid having
to create function prototypes.

Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <Timothy.Sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: move controlvm_payload_bytes_buffered

This patch moves the unsigned long controlvm_payload_bytes_buffered
from the module parameters location to a more appropriate location
below the controlvm_payload_info.

Signed-off-by: Jon Frisch <jon.frisch@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: remove unused dump_vhba_bus

This patch removes the unused u32 dump_vhba_bus.

Signed-off-by: Jon Frisch <jon.frisch@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: remove unused long long bus_count

This patch removes the unused long long bus_count.

Signed-off-by: Jon Frisch <jon.frisch@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus remove unused module parameter

This patch removes the unused module parameter
visorchipset_visorbusregwait

Signed-off-by: Jon Frisch <jon.frisch@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: use preferred kernel type u16

As suggested by checkpatch.pl:

CHECK: Prefer kernel type 'u16' over 'uint16_t'

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

staging: greybus: use preferred kernel type u8

As suggested by checkpatch.pl:

CHECK: Prefer kernel type 'u8' over 'uint8_t'

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

staging: lustre: lnet: klnds: socklnd: Remove unnecessary line continuations

Fixes checkpatch warning:

WARNING: Avoid unnecessary line continuations

Signed-off-by: Anchal Jain <anchalj109@gmail.com>
Acked-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lustre/ldlm: Fixed sparse warnings

Added __acquires / __releases sparse locking annotations
to lock_res_and_lock() and unlock_res_and_lock() functions
in l_lock.c, to fix below sparse warnings:

l_lock.c:47:22: warning: context imbalance in 'lock_res_and_lock' - wrong count at exit
l_lock.c:61:6: warning: context imbalance in 'unlock_res_and_lock' - unexpected unlock

Signed-off-by: Nayeemahmed Badebade <itachi.opsrc@gmail.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: lustre: lmv_obd: Remove redundant test.

filp != NULL is already checked, so no need to check it again.

Signed-off-by: Sandhya Bankar <bankarsandhya512@gmail.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: lustre: Convert array index from the loop bound to the loop index.

Convert array index from the loop bound to the loop index.

The structure cl_env_percpu[NR_CPUS] has been initializing for each possible
cpu (i.e 0 to i). During initialization if any error will occurred, the error
handling code should uninitialize cl_env_percpu upto i. But currently
unitialization is repeatedly done for the same cl_env_percpu[i] element.
This does not seems to be correct.

Used below semantic patch to identify this issue:

@@
expression e1,e2,ar;
@@
for(e1 = 0; e1 < e2; e1++) { <...
  ar[
- e2
+ e1
  ]
  ...> }

Signed-off-by: Sandhya Bankar <bankarsandhya512@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: MAINTAINERS: add greybus protocol drivers maintainers

Add bod to some Greybus protocol drivers maintainers, loopback and
timesync.

Signed-off-by: Bryan O'Donoghue <pure.logic@nexus-software.ie>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: hdm-usb: rename channel related spinlock

To increase comprehension this patch renames the channel related
spinlock with a more fitting name.

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

staging: most: hdm-usb: remove calls to usb_unachor_urb

This patch removes the calls to usb_unanchor_urb() from the completion
routines, since disassociation of the URBs is already handles by the
USB subsystem.

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

staging: most: hdm-usb: remove redundant spinlocks

This patch removes the spinlocks when accessing the usb_anchor list,
because the struct usb_anchor is using its own lock.

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

staging: most: hdm-usb: remove proprietary urb anchoring

This patch removes the propietary tracking of URBs. Instead the structure
usb_anchor of the USB subsystem is used.

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

Merge tag 'iio-for-4.9c' of git://git./linux/kernel/git/jic23/iio into staging-next

Jonathan writes:

Third set of new device support, functionality and cleanups for IIO in the 4.9 cycle.

Given Linus is hinting (strongly!) at an rc8 this last set is hopefully in
time for the 4.9 merge window.  The zpa2326 and si1145 drivers provide
fine illustrations that devices aren't getting any simpler!

I'm also particularly pleased Linus Walliej did such a thorough job of cleaning
up one of my old drivers.

New device support
* mCube MC3230 accelerometer
  - new fairly minimal driver.
* Murata zpa2326
  - extensive new driver supporting the rather 'novel' buffering of data this
    device provides and handling both it's own data ready trigger and other
    triggers rather elegantly.
* si1141, si1142, si1143, si1145, si1146 and si1147 proximity, UV, visible and
  IR sensors.
  - another extensive new driver supporting all the key bits of what this
    set of devices supplies including dataready triggers, buffers and all the
    various data channels.

Functionality
* kxsd9 - Linus brought this scratch driver I wrote in one afternoon years ago
  up to date adding lots of good stuff along the way.
  - SPI support after extensive rework of the driver.
  - Triggered buffer capture support.
  - Runtime PM.
  - Regulator handling.
  - Mounting matrix support.
* mma7660
  - Add MODULE_DEVICE_TABLE to support autoprobing.

Cleanups
* ad5933
  - Align some function arguements nicely.
* med_z188
  - Constify iio_info structure.
* sca3000
  - Implement IIO_CHAN_INFO_SAMP_FREQ rather than a hand rolled attr.
    There are still quite a few drivers that would benefit from similar updates.
* ssp_sensors
  - Constify iio_info structures in accel and gyro drivers.

Merge branch 'greybus' into staging-testing

This merges the greybus branch into staging-testing.  It contains the
drivers/staging/greybus/ subsystem and related drivers and has passed
the 0-day bot tests so no builds should break.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: MAINTAINERS: add greybus protocol drivers maintainers

Add me to some Greybus protocol drivers maintainers, spi, sdio, power
supply, light and gpio.

Signed-off-by: Rui Miguel Silva <rmfrfs@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: add some MAINTAINERS

Johan and Alex and I are going to maintain the greybus code, so add it
to MAINTAINERS so we get cc:ed on patches.

Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Alex Elder <elder@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: remove CONFIG_PM_RUNTIME from kernel_ver.h

The last thing remaining in kernel_ver.h was the setting of
CONFIG_PM_RUNTIME, which isn't needed in a in-tree implementation.  So
remove the setting of this value, and the .h file entirely as that was
the last thing left in it.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: Kconfig: enable possibility to select light driver

Remove BROKEN keyword to allow the light driver to be select now that
we fixed the kernel version dependencies. Also fix the module name in
the help section.

Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: makefile: fix dependency of spi to spilib

Greybus SPI driver depends on gb-spilib and we need to state that at
makefile to make it link correctly.

Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: power_supply: remove KERNEL_VERSION checks

No need to support older kernel versions in the Greybus Power Supply
driver, so remove the checks as needed, we can now rely on all of the
correct Power Supply core apis being present. Also move some properties
definitions to the power supply greybus code.

Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: sdio: remove KERNEL_VERSION checks

No need to support older kernel versions in the Greybus SDIO driver, so
remove the checks as needed, we can now rely on all of the correct SDIO
core apis being present.

Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: light: remove KERNEL_VERSION checks

No need to support older kernel versions in the Greybus Light driver, so
remove the checks as needed, we can now rely on all of the correct LED
core apis being present. And compile only if flash and v4l2 flash is
reachable.

Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: sdio: fix min() type check

The 0-day bot pointed out a type difference in one min() call, so fix it
up by being explicit about the type being compared.

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: spi: remove KERNEL_VERSION checks

No need to support older kernel versions in the Greybus SPI and spilib
driver, so remove the checks as needed, we can now rely on all of the
correct SPI core apis being present.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: gpio: it's CONFIG_GPIOLIB, not CONFIG_GPIO

The GPIO dependancy is CONFIG_GPIOLIB, not CONFIG_GPIO, no wonder it
wasn't building properly...

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: gpio: remove KERNEL_VERSION checks

No need to support older kernel versions in the Greybus GPIO driver, so
remove the checks as needed, we can now rely on all of the correct
GPIO core apis being present.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: vibrator: remove KERNEL_VERSION checks

No need to support older kernel versions in the Greybus Vibrator driver,
so remove the checks as needed, we can now rely on all of the correct
driver core apis being present.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: hid: remove KERNEL_VERSION checks

No need to support older kernel versions in the Greybus HID driver, so
remove the checks as needed, we can now rely on all of the "new" apis
being present.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: kernel_ver.h: remove lots of stuff

Now that we do not care about the kernel version we are building
against, we can strip out lots of backward compatibilty that was added
to kernel_ver.h in order to write semi-portable driver code.

To start with, remove the functions and #defines that are now in the
kernel tree, no need to have duplicate copies of them all.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: Add drivers/staging/greybus to the build

This adds a proper Kconfig file for drivers/staging/greybus and fixes up
the Makefile to work correctly within the kernel build system (modules
depend on the .config options, etc.)

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: remove old es1 endpoint description

The Toshiba ES1 chip is no longer around, so remove the USB descriptor
documentation for it as no one cares anymore.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: remove .gitignore file

Only the tools subdirectory needs a .gitignore entry, so move it there
and fix it up to only list the needed file.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: remove README and LICENSE

We don't need yet-another-copy of the GPLv2 in the tree, and the README
is now pointless, so remove both of these files.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

staging: greybus: Documentation: remove sysfs tree

We do not need an example of the sysfs tree in the kernel code itself,
so remove these files, as they are now pointless.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

greybus: remove local checkpatch.pl copy

When the greybus tree was external, it contained a copy of checkpatch.pl
to keep everyone "in line".  This is no longer needed and can now be
removed.

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Merge greybus driver tree into 4.8-rc6

This pulls the external greybus driver tree into 4.8-rc6 as it should be
part of the main kernel tree and not live outside in some lonely github
repo, never to be reunited with it's true love...

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: update version to 2.6.99

With all of the the missing patches from the
lustre 2.7 version merged upstream its time to update
the upstream clients version.

Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: o2iblnd: Put back work queue check previously removed

The previous patch, http://review.whamcloud.com/21304/, removed
a check needed until LU-5718 is properly addressed.  With
the check, LU-5718 results in an error message and a lost
RDMA operation.  Without it, we have memory corruption and
a crash (much harder to debug).

Putting the check back in case LU-5718 is not fixed soon.

Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-7650
Reviewed-on: http://review.whamcloud.com/22281
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Olaf Weber <olaf@sgi.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lnet: Enable setting per NI peer_credits

The code to allow peer_credits to be set per NI was originally
"left inactive" because there were concerns about peer_credits
interfering with the ability for IB nodes to connect to each
other when peer_credits are not the same (peer_credits controls
the queue depth for IB). With LU-3322, the values do not have
to match so it is now safe to enable this code so peer_credits
can be set per NI.

This patch enables existing code for setting per NI peer_credits.

Second this patch fixes a long standing bug in that the conf data
was not being used to set variables in the lnet_ni structure until
after lnd_startup() was called which meant LND drivers were
ignoring struct lnet_ni tunable values being set. Now we change
struct lnet_ni data fields based on conf data before calling
lnd_startup().

Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8507
Reviewed-on: http://review.whamcloud.com/21948
Reviewed-by: Olaf Weber <olaf@sgi.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lnet: Ensure routing is turned on first time

In lnet_rtrpools_enable(), a mistake was made and routing
was not being turned on when the rtrpools are being allocated
for the first time.

This patch fixes that routine so we remember to turn on
routing after allocating the rtrpools.

Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8501
Reviewed-on: http://review.whamcloud.com/21934
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Amir Shehata <amir.shehata@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lnet: check if ni is in current net namespace

Add new 'ni_net_ns' field to struct lnet_ni to hold a reference
to original net namespace in which ni is created.
In LNetDist(), check if ni was created in same net namespace as
current's one. If not, assign order above 0xffff0000, to make
this ni not a priority.

Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-7845
Reviewed-on: http://review.whamcloud.com/21884
Reviewed-by: Olaf Weber <olaf@sgi.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lnet: potential deadlock in lnet

Fixes potential deadlock in LNetMDAttach

Signed-off-by: Quentin Bouget <quentin.bouget.ocre@cea.fr>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8249
Reviewed-on: http://review.whamcloud.com/20676
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Henri Doreau <henri.doreau@cea.fr>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lmv: fix parent FID for migration

If the migrating directory is under striped directory, it needs
to set right stripe FID for its parent.

Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6263
Reviewed-on: http://review.whamcloud.com/13817
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: mdc: cl_default_mds_easize not refreshed

The client_obd::cl_default_mds_easize field should track the largest
observed EA size advertised by the MDT, subject to a reasonable upper
bound. The MDC uses cl_default_mds_easize to calculate the initial
size of request buffers.  The default value should be small enough to
avoid wasted memory and excessive use of vmalloc(), yet large enough
to accommodate the common use case.

In the current code, the default value is only updated if
client_obd::cl_max_mds_easize is strictly less than
mdt_body::mbo_max_mdsize. This condition is almost never met, because
client_obd::cl_max_mds_easize is computed at client mount-time based
on the number of OSTs in the filesystem, so the MDT won't ever observe
and advertise an EA size larger than that.

As a result, client_obd::cl_default_mds_easize indefinitely retains
its initial value, which is computed at client mount-time based on
the filesystem's default stripe width. Any getattr() requests for
widely striped files will consequently allocate a request buffer
that is too small, forcing reallocations on both the client and
server side. To avoid this, update client_obd::cl_default_mds_easize
independently of the value of client_obd::cl_max_mds_easize.

In addition, this patch includes these changes:

 - Add comments to the client_obd structure to clarify what the
   cl_{default,max}_mds_{cookie,ea}size values mean.

 - Prevent mdc_get_info() from storing uninitialized data in
   client_obd::cl_max_mds_cookiesize.

 - Use 4096 as an upper bound for the default values.  The former
   bound of PAGE_CACHE_SIZE is too large on 64k-page platforms
   (i.e. PPC), so it fails to prevent the vmalloc() spinlock
   contention described in LU-3338. The new value was chosen to
   be large enough to accommodate common use cases while staying
   well below the 16k threshold at which allocations start using
   vmalloc().

Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Kyle Blatter <kyleblatter@llnl.gov>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5549
Reviewed-on: http://review.whamcloud.com/11614
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: llite: make default_easize writeable in /sysfs

Allow default_easize to be tuned via /sysfs. A system administrator
might want this if a rare access to widely striped files drives up the
value on a filesystem where narrowly striped files are the more common
case. In practice, however, this is wanted primarily to facilitate
a test case for LU-5549.

 - Plumb the necessary interfaces through the LMV and MDC layers
   to expose write access to this value by higher layers.

 - Add block comments to modified functions.

Signed-off-by: Ned Bass <bass6@llnl.gov>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5549
Reviewed-on: http://review.whamcloud.com/13112
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: mdt: add indexing option to default dir stripe

Add indexing option to default dirstripe EA. If MDT find
out the client send the create req to the wrong MDT because
of default stripeEA, it will return -EREMOTE, then client
will retrieve default stripeEA through xattr cache, and
re-create the object.

Also merged patch for LU-6341 to resolve the following problem.
Use ll_dir_getstripe to get default stripeEA in ll_new_node(),
Because ll_getxattr_common requires admin rights for retrieving
default LMVEA (because of trusted- prefix), which might cause
mkdir (from normal user) failure.

If parent does not have default stripeEA, then child should always
be in the same MDT for mkdir. Otherwise MDT should return -EREMOTE,
then client will refresh the default stripe index, and recreate
the object.

Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5523
Reviewed-on: http://review.whamcloud.com/13360
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6341
Reviewed-on: http://review.whamcloud.com/13990
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ptlrpc: prevent request timeout grow due to recovery

Patch fixes the issue seen on the client with growing request
timeout which occurred after the server side patch landed for
LU-5079. While commit itself is correct, it reveals another
issue. If request is being processed for a long time on server
then client adaptive timeouts will adapt to that after receiving
reply and new requests will have bigger timeout. Another problem
is that server AT history is corrupted by recovery request
processing time which not pure service time but includes
also waiting time for clients to recover.

Patch prevents the AT stats update from early replies on client and
from recovering requests processing time on server.
The ptlrpc_at_recv_early_reply() still updates the current request
timeout as asked by server, but don't include this into AT stats.
The real reply will bring that data from server after all.

Signed-off-by: Mikhail Pershin <mike.pershin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6084
Reviewed-on: http://review.whamcloud.com/13520
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obd: use proper flags for call_usermodehelper

When a parameter is permanently changed on the MGS the
MGS send a changelog packet to the proper nodes that
are affected by the change. Once the nodes receive the
change they then call the userland utility lctl to
change its local value. When calling a userland
application from the kernel you specify a flag to
control the interaction with the application. Originally
by default the flag was set to 0 which is UMH_NO_WAIT
which meant lctl was being called asynchronously. In
older kernels this was fine since UHM_NO_WAIT and
UHM_WAIT_PROC had nearly the same logic. This changed
with newer kernels which broke updating our parameters.
Plus doing a UHM_NO_WAIT doesn't report back a error
if something goes wrong with lctl. The fix is to set
the flag to UHM_WAIT_PROC so kernel space waits until
lctl has finished and we get a proper error code if
something does go wrong with lctl.

Signed-off-by: James Simmons <uja.ornl@gmail.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6063
Reviewed-on: http://review.whamcloud.com/13677
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ptlrpc: remove unnecessary EXPORT_SYMBOL

A lot of symbols don't need to be exported at all because they are
only used in the module they belong to.

Signed-off-by: frank zago <fzago@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5829
Reviewed-on: http://review.whamcloud.com/12510
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: llite: lock the inode to be migrated

Because the inode and its connected dentries will be cleared
out of the cache after migration, the inode needs to be locked
during the migration.

Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4712
Reviewed-on: http://review.whamcloud.com/9689
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obdclass: remove unnecessary EXPORT_SYMBOL

A lot of symbols don't need to be exported at all because they are
only used in the module they belong to.

Signed-off-by: frank zago <fzago@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5829
Reviewed-on: http://review.whamcloud.com/13323
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: misc: remove unnecessary EXPORT_SYMBOL

A lot of symbols don't need to be exported at all because they are
only used in the module they belong to.

Signed-off-by: frank zago <fzago@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5829
Reviewed-on: http://review.whamcloud.com/13321
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: grant: quiet message on grant waiting timeout

Use at_max in osc_enter_cache() to bound how long we wait for grant
space before switching to synchronous I/Os. Do not print a message
on the console when the timeout is hit since such long wait can
be legitimate with flaky network (i.e. BRW is resent multiple times).

Signed-off-by: Johann Lombardi <johann.lombardi@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5521
Reviewed-on: http://review.whamcloud.com/12146
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lmv: Do not revalidate stripes with master lock

Do not revalidate slave stripes while holding master lock.
Otherwise if the revalidating slaves are blocked, then the
master lock can not be released in time.

Remove some unnecesary merging in ll_revalidate_slave(), and
the attributes will be stored in each stripe, only
merging them if required.

Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6088
Reviewed-on: http://review.whamcloud.com/13432
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: client: Fix mkdir -i 1 from DNE2 client to DNE1 server

After DNE phase 2 has been added to client it sends
create request to slave MDT.  DNT1-only server doesn't
expect request to slave MDT from client. It expects
only cross-mdt request from master MDT. Thus if DNE2
client tries to "mkdir -i 1" on DNE1 server, then
LBUG happened.

This patch adds OBD_CONNECT_DIR_STRIPE connection
flag check on client side. If striped directories are not
supported by server, then create requrest is sent to
master MDT.

Signed-off-by: Artem Blagodarenko <artem_blagodarenko@xyratex.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6071
Xyratex-bug-id: MRP-2319
Reviewed-on: http://review.whamcloud.com/13189
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: wang di <di.wang@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: clio: pass fid for OST setattr

Store inode's fid in cl_setattr_ost() and OSC packs this info on the
wire (via lustre_set_wire_obdo) so that OST can use.

NOTE: currently lu_fid::f_ver and obdo::o_parent_ver are not used on
OFD device, and we use obdo::o_stripe_idx as
filter_fid::ff_parent::f_ver and save it to the device.

Signed-off-by: Bobi Jam <bobijam.xu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-1154
Reviewed-on: http://review.whamcloud.com/12902
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: clio: rename coo_attr_set to coo_attr_update

coo_attr_set() is used to update object's attribute but its name
makes confusion that people intuitively think that it is used to
pass object's attribute down to server sides.

Signed-off-by: Bobi Jam <bobijam.xu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-1154
Reviewed-on: http://review.whamcloud.com/12888
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: llite: pack suppgid to MDS correctly

The ll_lookup_it() may trigger IT_OPEN RPC to open a file by name.
But at that time, the client does not know the target file's GID,
so it cannot pack the necessary supplementary group ID in the RPC.
Because of missing the supplementary group ID, the RPC maybe fail
for open permission check on the MDS. Under such case, MDS should
return the target file's GID, if the current thread on the client
in the right group (according to the file's GID), the client will
try the IT_OPEN RPC again with the right supplementary group ID.

This patch is also helpful if some other(s) changed the file's GID
after current RPC sent to the MDS with the suppgid as the original
GID by race.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5423
Reviewed-on: http://review.whamcloud.com/12476
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: remove lustre/include/linux/

Merge the contents of lustre/include/linux/lvfs.h into
lustre/include/lvfs.h. Merge lustre/include/linux/lustre_user.h into
lustre/include/lustre/lustre_user.h. Move lustre_compat25.h and
lustre_patchless_compat.h from lustre/include/linux/ to
lustre/include/ and rename lustre_compat25.h to lustre_compat.h.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675
Reviewed-on: http://review.whamcloud.com/13271
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Amir Shehata <amir.shehata@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: libcfs: check mask returned by cpumask_of_node

cpumask_of_node can return NULL if NUMA node is unavailable,
in this case cfs_node_to_cpumask will try to copy from NULL
and cause kernel panic.

Signed-off-by: Liang Zhen <liang.zhen@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5751
Reviewed-on: http://review.whamcloud.com/13207
Reviewed-by: Li Wei <wei.g.li@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obd: change type of cl_conn_count to size_t

Change type of cl_conn_count to size_t.

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577
Reviewed-on: http://review.whamcloud.com/13125
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: llite: unlock inode size in ll_lov_setstripe_ea_info()

In ll_lov_setstripe_ea_info() release the inode size lock on all
appropriate exit paths.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6059
Reviewed-on: http://review.whamcloud.com/13167
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lprocfs: cleanup stats locking code

Add comment blocks on lprocfs_stats_lock() and lprocfs_stats_unlock().
Move common NOPERCPU code out of the switch() statements to reduce
code size and complexity, since it doesn't depend on the opc at all.

Replace switch() in lprocfs_stats_unlock() with a simple if/else,
since the lock opc was already checked in lprocfs_stats_lock().

Add an enum for the lprocfs_stats_lock() operations to make it clear
what the valid values are and allow compiler checking.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5946
Reviewed-on: http://review.whamcloud.com/12872
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: osc: change cl_extent_tax and *grants to unsigned

Change the type accordant usage and remove warnings.

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577
Reviewed-on: http://review.whamcloud.com/12386
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: osc: osc_object_ast_clear() LBUG

An OSC object could be destroyed with AGL locks waiting for granted,
so we'd get rid of the osc_object_ast_clear() assertion that its
dlm locks all getting granted.

Signed-off-by: Bobi Jam <bobijam.xu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6042
Reviewed-on: http://review.whamcloud.com/13163
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: mgc: add nid iteration

mgc_apply_recover_logs use only first nid from entry,
this could be the problem for a cluster with several network
address for a one node.

Signed-off-by: Alexander Boyko <alexander.boyko@seagate.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5950
Xyratex-bug-id: MRP-2255
Reviewed-on: http://review.whamcloud.com/12829
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Mike Pershin <mike.pershin@intel.com>
Reviewed-by: Ann Koehler <amk@cray.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ptlrpc: fix race between connect vs resend

Buggy code at ptlrpc_connect_interpret()
finish:
    rc = ptlrpc_import_recovery_state_machine(imp);
    ...
    Set import connection flags
When import has FULL state ptlrpc_import_recovery_state_machine()
wakeup all waiters on import and all delayed request, which was
resented. And it could happened that request was send without
updated flags and AT is disabled. If such request is in progress
on the server, server drop the new instance, and could do early reply
for it. But this early reply confuse client, cause it wait real
reply(no AT for this request). Client try to touch buffer outside
reply and got EPROTO error.
The same bug existed for initital connect too. Import became FULL
before import connection flags was set.

Signed-off-by: Alexander Boyko <alexander_boyko@xyratex.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5528
Xyratex-bug-id: MRP-2034
Reviewed-on: http://review.whamcloud.com/11723
Reviewed-by: Li Wei <wei.g.li@intel.com>
Reviewed-by: Alexander Boyko <alexander.boyko@seagate.com>
Reviewed-by: Liang Zhen <liang.zhen@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lov: flatten struct lov_stripe_md

Flatten out the lsm_wire struct from the middle of struct
lov_stripe_md and remove the member name macros.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5814
Reviewed-on: http://review.whamcloud.com/12581
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ldlm: move LDLM_GID_ANY to lustre_dlm.h

lustre_idl.h only includes wire data; lustre_dlm.h is the
right place for LDLM_GID_ANY.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6028
Reviewed-on: http://review.whamcloud.com/13074
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ptlrpc: fix comparison between signed and unsigned

Change return type and size argiments of lustre_msg_hdr_size(),
lustre_msg_buf{len,count}() and req_capsule_*_size() to __u32.
Change type of req_format->rf_idx and req_format->rf_fields.nr
to size_t. Also return zero for incorrect message magic instead
of -EINVAL. This will be more robust because of few of them after
LASSERTF(0, "...") and will not be returned. In the rest places
it return zero size instead of huge number after implicit
unsigned conversion.

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577
Reviewed-on: http://review.whamcloud.com/12475
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: clio: add coo_getstripe interface

Use cl_object_operations::coo_getstripe() to handle
LL_IOC_LOV_GETSTRIPE ops.

Signed-off-by: Bobi Jam <bobijam.xu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5823
Reviewed-on: http://review.whamcloud.com/12452
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obdclass: change cl_fault_io->ft_nob to size_t

Change the type accordant usage.

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577
Reviewed-on: http://review.whamcloud.com/12380
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obd: change brw_page->count to unsigned

Pages count is unsigned. So, change the type accordant usage.

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577
Reviewed-on: http://review.whamcloud.com/12378
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ldlm: Recalculate interval in ldlm_pool_recalc()

Instead of rechecking a static value, recalculate to see if pool stats
need to be updated.
Add newline so message will print instead of warning about missing
newline.

Signed-off-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4536
Reviewed-on: http://review.whamcloud.com/12547
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ptlrpc: Suppress error message when imp_sec is freed

There is a race condition on client reconnect when the import
is being destroyed.  Some outstanding client bound requests
are being processed when the imp_sec has alread been freed.
Ensure to suppress the error message in import_sec_validate_get()
in that case

Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3353
Reviewed-on: http://review.whamcloud.com/10200
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obdclass: eliminate NULL error return

Always return an ERR_PTR() on errors, never return a NULL,
in lu_object_find_slice().  Also clean up callers who
no longer need special case handling of NULL returns.

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5858
Reviewed-on: http://review.whamcloud.com/12554
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obdclass: change loop indexes to unsigned

Cleanup warnings about comparison between signed and unsigned.

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577
Reviewed-on: http://review.whamcloud.com/12387
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: fiemap: set FIEMAP_EXTENT_LAST correctly

When we've collected enough extents as user requested, we'd check one
further to decide whether we've reached the last extent of the file.

Signed-off-by: Bobi Jam <bobijam.xu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5933
Reviewed-on: http://review.whamcloud.com/12781
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ldlm: evict clients returning errors on ASTs

To test proper behavior of clients returning errors on ASTs
we can induce a failure with setting OBD_FAIL_LDLM_BL_CALLBACK_NET.
Handle the new additonal case of cfs_fail_err being set as well
so that the cfs_fail_err can be sent back in a reply.

Signed-off-by: Alexey Lyashkov <alexey_lyashkov@xyratex.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5581
Xyratex-bug-id: MRP-2041
Reviewed-on: http://review.whamcloud.com/11752
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: mdc: Proper accessing struct lov_user_md

In mdc_setattr_pack() access the members of struct lov_user_md by
little endian byte order.

Signed-off-by: Yoshifumi Uemura <kogexe@gmail.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5889
Reviewed-on: http://review.whamcloud.com/12683
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obdclass: lu_htable_order() return type to long

Change the type accordant usage.

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577
Reviewed-on: http://review.whamcloud.com/12385
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: llite: fix dup flags names

The name 'xattr' is used for two different ll_flags bits.
Change the names to be distinct and different, reflecting
the names of the bits used in LL_SBI_xbitnamex #defines.

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5586
Reviewed-on: http://review.whamcloud.com/12892
Reviewed-by: Minh Diep <minh.diep@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: llog: prevent out-of-bound index

llog_process_thread() can be called from llog_cat_process_cb with an
index already out of bound, leading to the following crash:

LustreError: 3773:0:(llog.c:310:llog_process_thread())
  ASSERTION(index <= last_index + 1 ) failed:
LustreError: 3773:0:(llog.c:310:llog_process_thread()) LBUG

 #0 [ffff8801144bf900] machine_kexec at ffffffff81038f3b
 #1 [ffff8801144bf960] crash_kexec at ffffffff810c5d82
 #2 [ffff8801144bfa30] panic at ffffffff8152798a
 #3 [ffff8801144bfab0] lbug_with_loc at ffffffffa02f8eeb [libcfs]
 #4 [ffff8801144bfad0] llog_process_thread at ffffffffa0413fff [obdclass]
 #5 [ffff8801144bfb80] llog_process_or_fork at ffffffffa041585f [obdclass]
 #6 [ffff8801144bfbd0] llog_cat_process_cb at ffffffffa0418612 [obdclass]
 #7 [ffff8801144bfc30] llog_process_thread at ffffffffa0413c22 [obdclass]
 #8 [ffff8801144bfce0] llog_process_or_fork at ffffffffa041585f [obdclass]
 #9 [ffff8801144bfd30] llog_cat_process_or_fork at ffffffffa0416b9d [obdclass]

If index is too big, simply return success.

Signed-off-by: frank zago <fzago@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5635
Reviewed-on: http://review.whamcloud.com/12161
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Patrick Farrell <paf@cray.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ptlrpc: quiet errors on initial connection

It may be that a client or MDS is trying to connect to a target (OST
or peer MDT) before that target is finished setup. Rather than
spamming the console logs during initial connection, only print a
console error message if there are repeated failures trying to
connect to the target, which may indicate an error on that node.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Bobi Jam <bobijam.xu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3456
Reviewed-on: http://review.whamcloud.com/10057
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: ldlm: revert the changes for lock canceling policy

The changes for LRU lock policy was introduced by commit bfae5a4e,
where I was trying to revise the policy to pick locks for canceling.

However, this caused two problems as mentioned in LU-5727. The first
problem is that the lock can only be picked for canceling only if
the number of LRU locks is over preset LRU number AND it's aged; the
second problem is that mdc_cancel_weight() tends to not cancel OPEN
locks, therefore open locks can be kept forever and finally exhausts
memory on the MDT side.

The commit 7b2d26b0 ("revert changes to ldlm_cancel_aged_policy") fixed
the first problem. This patch will revert the rest of changes related
to LRU policy revise.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5727
Reviewed-on: http://review.whamcloud.com/12733
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: recovery: don't replay closed open

To avoid scanning the replay open list every time in the
ptlrpc_free_committed(), the fix of LU-2613 (4322e0f9) changed
the ptlrpc_free_committed() to skip the open list unless the
import generation is changed. That introduced a race which could
make a closed open being replayed:

1. Application calls ll_close_inode_openhandle()-> mdc_close(),
   to close file, rq_replay is cleared, but the open request is
   still on the imp_committed_list;

2. Before the md_clear_open_replay_data() is called for close,
   client start replay, and that closed open will be replayed
   mistakenly;

3. Open replay interpret callback (mdc_replay_open) could race
   with the mdc_clear_open_replay_data() at the end;

This patch fix the ptlrpc_free_committed() to make sure the
open list is scanned on recovery to prevent the closed open request
from being replayed.

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5507
Reviewed-on: http://review.whamcloud.com/12667
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: changelog: Proper record remapping

Fixed changelog_remap_rec() to correctly remap records emitted
with jobid_var=disabled, i.e. delivered by new servers but with
no jobid field.

Signed-off-by: Henri Doreau <henri.doreau@cea.fr>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5862
Reviewed-on: http://review.whamcloud.com/12574
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Robert Read <robert.read@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: llite: remove ll_objects_destroy()

Remove ll_objects_destroy(). This function is not needed for
interoperability with servers of version 2.4 or higher.

Remove the then unused function lov_destroy() and its supporting
functions. Remove the lsm_destroy method of struct lsm_operations.

Remove the unused struct lov_stripe_md, MD export, and capa parameters
from obd_destroy() and its implementations.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5814
Reviewed-on: http://review.whamcloud.com/12618
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: echo: replace lov_stripe_md with lov_oinfo

In echo_client replace uses of struct lov_stripe_md with struct
lov_oinfo (since the instances of the former really only contained a
single instance of the latter). Remove the then unneccessary functions
echo_alloc_memmd(), echo_free_memmd(), osc_unpackmd(), and
obd_alloc_memmd(). Remove the struct lov_stripe_md * parameter from
obd_create().

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5418
Reviewed-on: http://review.whamcloud.com/12447
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: obd: remove unused obd methods

Remove no longer used osc_packmd() and osc_getstripe().
Several ioctls cases that are no longer used are removed.
Remove no longer used adjust_kms() infrastructure.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2785
Reviewed-on: http://review.whamcloud.com/8547
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: statahead: small fixes and cleanup

small fixes:
 * when 'unplug' is set for ll_statahead(), sa_put() shouldn't kill
   the entry found, because its inflight RPC may not finish yet.
 * remove 'sai_generation', add 'lli_sa_generation' because the
   former one is not safe to access without lock.
 * revalidate_statahead_dentry() may fail to wait for statahead
   entry to become ready, in this case it should not release this
   entry, because it may be used by inflight statahead RPC.

cleanups:
 * rename ll_statahead_enter() to ll_statahead().
 * move dentry 'lld_sa_generation' update to ll_statahead() to
   simplify code and logic.
 * other small cleanups.

Signed-off-by: Lai Siyao <lai.siyao@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3270
Reviewed-on: http://review.whamcloud.com/9667
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6222
Reviewed-on: http://review.whamcloud.com/13708
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>