Don't update *inode in __follow_mount_rcu() until we'd verified that
there is mountpoint there.  Kudos to Hugh Dickins for catching that
one in the first place and eventually figuring out the solution (and
catching a braino in the earlier version of patch).
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Add missing ->i_mutex, convert to lookup_one_len() instead of
(broken) open-coded analog, cope with getting something like
a//b as relative pathname.  Simplify the hell out of it, while
we are there...
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Reviewed-by: NJeff Layton <jlayton@redhat.com>

... and it's getting it wrong, too - missing ->d_revalidate() calls when
it's dealing with filesystem (procfs) that has non-trivial ->d_revalidate()...
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

... and we want to report these failures in ->lookup() anyway.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

it's NFS-exportable, so...
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

NULL is not a possible return value for that method, TYVM...
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

... unfortunately, cifs bug got copied.  Fix is essentially the same.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

deal with d_move() races properly; rename_lock read-retry loop,
rcu_read_lock() held while walking to root, d_lock held over
subtraction from namelen and copying the component to stabilize
->d_name.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

* 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/staging:
  hwmon: (adm1275) Fix coefficients per datasheet revision B
  hwmon: (pmbus) Use long variables for register to data conversions

* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon/kms: add new NI pci ids

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
  fix loop checks in d_materialise_unique()
  Fix ->d_lock locking order in unlazy_walk()

* 'rcu/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-2.6-rcu:
  rcu: Prevent RCU callbacks from executing before scheduler initialized

Commit 3fe1698b ("sched: Deal with non-atomic min_vruntime reads
on 32bit") forgot to initialize min_vruntime_copy which could lead to
an infinite while loop in task_waking_fair() under some circumstances
(early boot, lucky timing).

[ This bug was also reported by others that blamed it on the RCU
  initialization problems ]
Reported-and-tested-by: NBruno Wolff III <bruno@wolff.to>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@kernel.org
Signed-off-by: NDave Airlie <airlied@redhat.com>

Coefficients to convert chip register values to voltage/current have been
slightly changed in revision B of the chip datasheet. Update driver coefficients
to match the coefficients in the datasheet.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJean Delvare <khali@linux-fr.org>

Both __d_unalias() and __d_materialise_dentry() need loop prevention.
Grab rename_lock in caller, check for loops there...

As a side benefit, we have dentry_lock_for_move() called only under
rename_lock, which seriously reduces deadlock potential of the
execrable "locking order" used for ->d_lock.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-fixes:
  GFS2: Resolve inode eviction and ail list interaction bug
  GFS2: Fix race during filesystem mount
  GFS2: force a log flush when invalidating the rindex glock

This patch contains a few misc fixes which resolve a recently
reported issue. This patch has been a real team effort and has
received a lot of testing.

The first issue is that the ail lock needs to be held over a few
more operations. The lock thats added into gfs2_releasepage() may
possibly be a candidate for replacing with RCU at some future
point, but at this stage we've gone for the obvious fix.

The second issue is that gfs2_write_inode() can end up calling
a glock recursively when called from gfs2_evict_inode() via the
syncing code, so it needs a guard added.

The third issue is that we either need to not truncate the metadata
pages of inodes which have zero link count, but which we cannot
deallocate due to them still being in use by other nodes, or we need
to ensure that those pages have all made it through the journal and
ail lists first. This patch takes the former approach, but the
latter has also been tested and there is nothing to choose between
them performance-wise. So again, we could revise that decision
in the future.

Also, the inode eviction process is now better documented.
Signed-off-by: NSteven Whitehouse <swhiteho@redhat.com>
Tested-by: NBob Peterson <rpeterso@redhat.com>
Tested-by: NAbhijith Das <adas@redhat.com>
Reported-by: NBarry J. Marson <bmarson@redhat.com>
Reported-by: NDavid Teigland <teigland@redhat.com>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc:
  mmc: core: Bus width testing needs to handle suspend/resume

* 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6:
  SUNRPC: Fix use of static variable in rpcb_getport_async
  NFSv4.1: update nfs4_fattr_bitmap_maxsz
  SUNRPC: Fix a race between work-queue and rpc_killall_tasks
  pnfs: write: Set mds_offset in the generic layer - it is needed by all LDs

* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
  kbuild: Do not write to builddir in modules_install

* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon/kms/evergreen: emit SQ_LDS_RESOURCE_MGMT for blits
  agp/intel: Fix typo in G4x_GMCH_SIZE_VT_2M
  drm/radeon/kms: fix typo in read_disabled vbios code
  drm/radeon/kms: use correct BUS_CNTL reg on rs600
  drm/radeon/kms: fix backend map typo on juniper
  drm/radeon/kms: fix regression in hotplug

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (21 commits)
  slip: fix wrong SLIP6 ifdef-endif placing
  natsemi: fix another dma-debug report
  sctp: ABORT if receive, reassmbly, or reodering queue is not empty while closing socket
  net: Fix default in docs for tcp_orphan_retries.
  hso: fix a use after free condition
  net/natsemi: Fix module parameter permissions
  XFRM: Fix memory leak in xfrm_state_update
  sctp: Enforce retransmission limit during shutdown
  mac80211: fix TKIP replay vulnerability
  mac80211: fix ie memory allocation for scheduled scans
  ssb: fix init regression of hostmode PCI core
  rtlwifi: rtl8192cu: Add new USB ID for Netgear WNA1000M
  ath9k: Fix tx throughput drops for AR9003 chips with AES encryption
  carl9170: add NEC WL300NU-AG usbid
  cfg80211: fix deadlock with rfkill/sched_scan by adding new mutex
  ath5k: fix incorrect use of drvdata in PCI suspend/resume code
  ath5k: fix incorrect use of drvdata in sysfs code
  Bluetooth: Fix memory leak under page timeouts
  Bluetooth: Fix regression with incoming L2CAP connections
  Bluetooth: Fix hidp disconnect deadlocks and lost wakeup
  ...

On reading the ext_csd for the first time (in 1 bit mode), save the
ext_csd information needed for bus width compare.

On every pass we make re-reading the ext_csd, compare the data
against the saved ext_csd data.

This fixes a regression introduced in 3.0-rc1 by 08ee80cc
("mmc: core: eMMC bus width may not work on all platforms"), which
incorrectly assumed we would be re-reading the ext_csd at resume-
time.
Signed-off-by: NPhilip Rakity <prakity@marvell.com>
Tested-by: NJaehoon Chung <jh80.chung@samsung.com>
Signed-off-by: NChris Ball <cjb@laptop.org>

Under some rare but real combinations of configuration parameters, RCU
callbacks are posted during early boot that use kernel facilities that
are not yet initialized.  Therefore, when these callbacks are invoked,
hard hangs and crashes ensue.  This commit therefore prevents RCU
callbacks from being invoked until after the scheduler is fully up and
running, as in after multiple tasks have been spawned.

It might well turn out that a better approach is to identify the specific
RCU callbacks that are causing this problem, but that discussion will
wait until such time as someone really needs an RCU callback to be invoked
(as opposed to merely registered) during early boot.
Reported-by: Njulie Sullivan <kernelmail.jms@gmail.com>
Reported-by: NRKK <kulkarni.ravi4@gmail.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Njulie Sullivan <kernelmail.jms@gmail.com>
Tested-by: NRKK <kulkarni.ravi4@gmail.com>

Compute drivers may change this, so make sure to emit it to
avoid errors in bo blits.

Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=39119Signed-off-by: NAlex Deucher <alexdeucher@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Konstantin Belousov found an error in the define of G4x_GMCH_SIZE_VT_2M
relative to the GMCH specs, and confirmed that indeed one of his users
with a Q45 reports 0xb not 0xc for a 2/2MiB GATT.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Konstantin Belousov <kostikbel@gmail.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Make sure that child is still a child of parent before nested locking
of child->d_lock in unlazy_walk(); otherwise we are risking a violation
of locking order and deadlocks.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
  powerpc/mm: Fix memory_block_size_bytes() for non-pseries
  mm: Move definition of MIN_MEMORY_BLOCK_SIZE to a header

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/linux-arm-soc:
  pcmcia: pxa2xx/vpac270: free gpios on exist rather than requesting
  ARM: pxa/raumfeld: fix device name for codec ak4104
  ARM: pxa/raumfeld: display initialisation fixes
  ARM: pxa/raumfeld: adapt to upcoming hardware change
  ARM: pxa: fix gpio_to_chip() clash with gpiolib namespace
  genirq: replace irq_gc_ack() with {set,clr}_bit variants (fwd)
  arm: mach-vt8500: add forgotten irq_data conversion
  ARM: pxa168: correct nand pmu setting
  ARM: pxa910: correct nand pmu setting
  ARM: pxa: fix PGSR register address calculation

* 'drm-intel-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/keithp/linux-2.6:
  drm/i915/ringbuffer: Idling requires waiting for the ring to be empty
  Revert "drm/i915: enable rc6 by default"
  drm/i915: Clean up i915_driver_load failure path
  drm/i915: Enable GPU reset on Ivybridge.
  drm/i915/dp: manage sink power state if possible
  drm/i915/dp: consolidate AUX retry code
  drm/i915/dp: remove DPMS mode tracking from DP
  drm/i915/dp: try to read receiver capabilities 3 times when detecting
  drm/i915/dp: read more receiver capability bits on hotplug
  drm/i915/dp: use DP DPCD defines when looking at DPCD values
  drm/i915/dp: retry link status read 3 times on failure

Because struct rpcbind_args *map was declared static, if two
threads entered this method at the same time, the values
assigned to map could be sent two two differen tasks.
This could cause all sorts of problems, include use-after-free
and double-free of memory.

Fix this by removing the static declaration so that the map
pointer is on the stack.
Signed-off-by: NBen Greear <greearb@candelatech.com>
Cc: stable@kernel.org
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

...which is measured by the size and not the amount of space remaining.

Waiting upon size-8, did one of two things. In the common case with more
than 8 bytes available to write into the ring, it would return
immediately. Otherwise, it would timeout given the impossible condition
of waiting for more space than is available in the ring, leading to
warnings such as:

[drm:intel_cleanup_ring_buffer] *ERROR* failed to quiesce render ring
whilst cleaning up: -16
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NKeith Packard <keithp@keithp.com>

This reverts commit a51f7a66.

We still have a few Ironlake and Sandybridge machines which fail when
RC6 is enabled. Better luck next release?
Signed-off-by: NKeith Packard <keithp@keithp.com>

i915_driver_load adds a write-combining MTRR region for the GTT
aperture to improve memory speeds through the aperture. If
i915_driver_load fails after this, it would not have cleaned up the
MTRR. This shouldn't cause any problems, except for consuming an MTRR
register. Still, it's best to clean up completely in the failure path,
which is easily done by calling mtrr_del if the mtrr was successfully
allocated.

i915_driver_load calls i915_gem_load which register
i915_gem_inactive_shrink. If i915_driver_load fails after calling
i915_gem_load, the shrinker will be left registered. When called, it
will access freed memory and crash. The fix is to unregister the shrinker in the
failure path using code duplicated from i915_driver_unload.

i915_driver_load also has some incorrect gotos in the error cleanup
paths:

 * After failing to initialize the GTT (which cannot happen, btw,
   intel_gtt_get returns a fixed (non-NULL) value), it tries to
   free the uninitialized WC IO mapping. Fixed this by changing the
   target from out_iomapfree to out_rmmap
Signed-off-by: NKeith Packard <keithp@keithp.com>
Tested-by: NLin Ming <ming.m.lin@intel.com>

Using integer variable types for register to data conversions can cause
overflows especially for power calculations, which are in microwatt.
Use long variables instead.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJean Delvare <khali@linux-fr.org>
Cc: stable@kernel.org # 2.6.39+

Let depmod.sh create a temporary directory in /tmp instead of writing to
the build directory as root. The mktemp utility should be available on
any recent system (and there is already scripts/gen_initramfs_list.sh
relying on it).
Reported-by: NChristian Kujau <lists@nerdbynature.de>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

There is a potential race during filesystem mounting which has recently
been reported. It occurs when the userland gfs_controld is able to
process requests fast enough that it tries to use the sysfs interface
before the lock module is properly initialised. This is a pretty
unusual case as normally the lock module initialisation is very quick
compared with gfs_controld.

This patch adds an interruptible completion which is used to ensure that
userland will wait for the initialisation of the lock module to
complete.

There are other potential solutions to this problem, but this is the
quickest at this stage and has been tested both with and without
mount.gfs2 present in the system.
Signed-off-by: NSteven Whitehouse <swhiteho@redhat.com>
Reported-by: NDavid Booher <dbooher@adams.net>

Right now, there is nothing that forces the log to get flushed when a node
drops its rindex glock so that another node can grow the filesystem. If the
log doesn't get flushed, GFS2 can corrupt the sd_log_le_rg list in the
following way.

A node puts an rgd on the list in rg_lo_add(), and then the rindex glock is
dropped so the other node can grow the filesystem. When the node reacquires the
rindex glock, that rgd gets deleted in clear_rgrpdi() before ever being
removed from the list by gfs2_log_flush().

This code simply forces a log flush when the rindex glock is invalidated,
solving the problem.
Signed-off-by: NBenjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: NSteven Whitehouse <swhiteho@redhat.com>