iomap_page_mkwrite_actor() calls __block_write_begin_int() with position
masked as pos & ~PAGE_MASK which is equivalent to pos & (PAGE_SIZE-1).
Thus it masks off high bits of file position. However
__block_write_begin_int() expects full file position on input. This does
not cause any visible issues because all __block_write_begin_int()
really cares about are low file position bits but still it is a bug
waiting to happen.
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Since no one uses it anymore.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Instead of doing a full extent list search for each extent that is
to be deleted using xfs_bmapi_read and then doing another one inside
of xfs_bunmapi_cow use the same scheme that xfs_bumapi uses:  look
up the last extent to be deleted and then use the extent index to
walk downward until we are outside the range to be deleted.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Rewrite xfs_reflink_cancel_cow_blocks so that we only do a search for
the first extent in the extent list and then iterate over the remaining
extents using the extent index, passing the extent we operate on
directly to xfs_bmap_del_extent_delay or xfs_bmap_del_extent_cow instead
of going through xfs_bunmapi and doing yet another extent list lookup.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Split out two helpers for deleting delayed or real extents from the COW fork.
This allows to call them directly from xfs_reflink_cow_end_io once that
function is refactored to iterate the extent tree.  It will also allow
to reuse the delalloc deletion from xfs_bunmapi in the future.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Instead of reserving space as the first thing in write_begin move it past
reading the extent in the data fork.  That way we only have to read from
the data fork once and can reuse that information for trimming the extent
to the shared/unshared boundary.  Additionally this allows to easily
limit the actual write size to said boundary, and avoid a roundtrip on the
ilock.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

There is no need to trim an extent into a shared or non-shared one, or
report any flags for plain old reads.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Delalloc extents in the extent list contain the number of reserved
indirect blocks in their startblock value and don't use the magic
DELAYSTARTBLOCK constant.  Ensure that xfs_reflink_trim_around_shared
handles them properly by checking for isnullstartblock().
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

This helpers allows to trim an extent to a subset of it's original range
while making sure the block numbers in it remain valid,

In the future xfs_trim_extent and xfs_bmapi_trim_map should probably be
merged in some form.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
[hch: split from a previous patch from Darrick, moved around and added
 support for "raw" delayed extents"]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

This allows the file system to tell a FIEMAP from a read operation, and thus
avoids the need to report flags that aren't actually used in the read path.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

There is no clear division of responsibility between those functions, so
just merge them into one to keep the code simple.  Also move
xfs_file_wait_for_io to xfs_reflink.c together with its only caller.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

filemap_write_and_wait_range operates on full pages, so there is no
need for the rounding operations.  Additionally this allows us to
micro-optimize by skipping the second inode_dio_wait for a
intra-file clone.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

We need the iolock protection to stabilizie the IS_SWAPFILE and
IS_IMMUTABLE values, as well as preventing new buffered writers
re-dirtying the file data that we just wrote out.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

The VFS i_ino is an unsigned long, while XFS inode numbers are 64-bit
wide, so checking i_ino for equality could lead to rate false positives
on 32-bit architectures.  Just compare the inode pointers themselves
to be safe.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

The VFS already does the check, and the placement of this duplicate
is in the way of the following locking rework.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

xfs_repair was not detecting that version 3 inodes are invalid for
for non-CRC filesystems. The result is specific inode corruptions go
undetected and hence aren't repaired if only the version number is
out of range.

The core of the problem is that the XFS_DINODE_GOOD_VERSION() macro
doesn't know that valid inode versions are dependent on a superblock
version number. Fix this in libxfs, and propagate the new function
out into the rest of xfsprogs to fix the issue.

[Darrick: port to kernel from xfsprogs]
Reported-by: Leslie Rhorer <lrhorer@mygrande.net>
Signed-off-by: Roger Willcocks <roger@filmlight.ltd.uk>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

The function xfs_calc_dquots_per_chunk takes a parameter in units
of basic blocks.  The kernel seems to get the units wrong, but
userspace got 'fixed' by commenting out the unnecessary conversion.
Fix both.

cc: <stable@vger.kernel.org>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

As part of the inode block map intent log item recovery process, we
had to set the IRECOVERY flag to prevent an unlinked inode from
being truncated during the first iput call.  This required us to set
MS_ACTIVE so that iput puts the inode on the lru instead of
immediately evicting the inode.

Unfortunately, if the mount fails later on, the inodes that have
been loaded (root dir and realtime) actually need to be evicted
since we're aborting the mount.  If we don't clear MS_ACTIVE in the
failure step, those inodes are not evicted and therefore leak.   The
leak was found by running xfs/130 and rmmoding xfs immediately after
the test.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>

The commit:

f65306ea xfs: map an inode's offset to an exact physical block

added a pointless error0: target; remove it.

Addresses-Coverity-Id: 1373865
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Bill O'Donnell <billodo@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

XFS historically took the iolock exclusive when invalidating pages
before direct I/O operations to protect against writeback starvations.

But this writeback starvation issues has been fixed a long time ago
in the core writeback code, and all other file systems manage to do
without the exclusive lock.  Convert XFS over to avoid the exclusive
lock in this case, and also move to range invalidations like done
by the other file systems.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

sparse reported that several variables and a function were not
forward-declared anywhere and therefore should be 'static'.

Found with sparse by running 'make C=2 CF=-D__CHECK_ENDIAN__ fs/xfs/'
Signed-off-by: Eric Biggers <ebiggers@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>

with gcc 4.1.2:

    fs/xfs/xfs_reflink.c: In function xfs_reflink_reserve_cow_range:
    fs/xfs/xfs_reflink.c:327: warning: error may be used uninitialized in this function

Indeed, if "count" is zero, the function will return an uninitialized
error value.

While "count" is unlikely to be zero, this function is called through
the public iomap API. Hence fix this by preinitializing error to zero.

Fixes: 2a06705c ("xfs: create delalloc extents in CoW fork")
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Remove redundant ifp = ifp statement, it does nothing. Found with
static analysis by CoverityScan.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

Pull befs fixes from Luis de Bethencourt:
 "I recently took maintainership of the befs file system [0]. This is
  the first time I send you a git pull request, so please let me know if
  all the below is OK.

  Salah Triki and myself have been cleaning the code and fixing a few
  small bugs.

  Sorry I couldn't send this sooner in the merge window, I was waiting
  to have my GPG key signed by kernel members at ELCE in Berlin a few
  days ago."

[0] https://lkml.org/lkml/2016/7/27/502

* tag 'befs-v4.9-rc1' of git://github.com/luisbg/linux-befs: (39 commits)
  befs: befs: fix style issues in datastream.c
  befs: improve documentation in datastream.c
  befs: fix typos in datastream.c
  befs: fix typos in btree.c
  befs: fix style issues in super.c
  befs: fix comment style
  befs: add check for ag_shift in superblock
  befs: dump inode_size superblock information
  befs: remove unnecessary initialization
  befs: fix typo in befs_sb_info
  befs: add flags field to validate superblock state
  befs: fix typo in befs_find_key
  befs: remove unused BEFS_BT_PARMATCH
  fs: befs: remove ret variable
  fs: befs: remove in vain variable assignment
  fs: befs: remove unnecessary *befs_sb variable
  fs: befs: remove useless initialization to zero
  fs: befs: remove in vain variable assignment
  fs: befs: Insert NULL inode to dentry
  fs: befs: Remove useless calls to brelse in befs_find_brun_dblindirect
  ...

Pull gcc plugins update from Kees Cook:
 "This adds a new gcc plugin named "latent_entropy". It is designed to
  extract as much possible uncertainty from a running system at boot
  time as possible, hoping to capitalize on any possible variation in
  CPU operation (due to runtime data differences, hardware differences,
  SMP ordering, thermal timing variation, cache behavior, etc).

  At the very least, this plugin is a much more comprehensive example
  for how to manipulate kernel code using the gcc plugin internals"

* tag 'gcc-plugins-v4.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  latent_entropy: Mark functions with __latent_entropy
  gcc-plugins: Add latent_entropy plugin

Pull MIPS updates from Ralf Baechle:
 "This is the main MIPS pull request for 4.9:

  MIPS core arch code:
   - traps: 64bit kernels should read CP0_EBase 64bit
   - traps: Convert ebase to KSEG0
   - c-r4k: Drop bc_wback_inv() from icache flush
   - c-r4k: Split user/kernel flush_icache_range()
   - cacheflush: Use __flush_icache_user_range()
   - uprobes: Flush icache via kernel address
   - KVM: Use __local_flush_icache_user_range()
   - c-r4k: Fix flush_icache_range() for EVA
   - Fix -mabi=64 build of vdso.lds
   - VDSO: Drop duplicated -I*/-E* aflags
   - tracing: move insn_has_delay_slot to a shared header
   - tracing: disable uprobe/kprobe on compact branch instructions
   - ptrace: Fix regs_return_value for kernel context
   - Squash lines for simple wrapper functions
   - Move identification of VP(E) into proc.c from smp-mt.c
   - Add definitions of SYNC barrierstype values
   - traps: Ensure full EBase is written
   - tlb-r4k: If there are wired entries, don't use TLBINVF
   - Sanitise coherentio semantics
   - dma-default: Don't check hw_coherentio if device is non-coherent
   - Support per-device DMA coherence
   - Adjust MIPS64 CAC_BASE to reflect Config.K0
   - Support generating Flattened Image Trees (.itb)
   - generic: Introduce generic DT-based board support
   - generic: Convert SEAD-3 to a generic board
   - Enable hardened usercopy
   - Don't specify STACKPROTECTOR in defconfigs

  Octeon:
   - Delete dead code and files across the platform.
   - Change to use all memory into use by default.
   - Rename upper case variables in setup code to lowercase.
   - Delete legacy hack for broken bootloaders.
   - Leave maintaining the link state to the actual ethernet/PHY drivers.
   - Add DTS for D-Link DSR-500N.
   - Fix PCI interrupt routing on D-Link DSR-500N.

  Pistachio:
   - Remove ANDROID_TIMED_OUTPUT from defconfig

  TX39xx:
   - Move GPIO setup from .mem_setup() to .arch_init()
   - Convert to Common Clock Framework

  TX49xx:
   - Move GPIO setup from .mem_setup() to .arch_init()
   - Convert to Common Clock Framework

  txx9wdt:
   - Add missing clock (un)prepare calls for CCF

  BMIPS:
   - Add PW, GPIO SDHCI and NAND device node names
   - Support APPENDED_DTB
   - Add missing bcm97435svmb to DT_NONE
   - Rename bcm96358nb4ser to bcm6358-neufbox4-sercom
   - Add DT examples for BCM63268, BCM3368 and BCM6362
   - Add support for BCM3368 and BCM6362

  PCI
   - Reduce stack frame usage
   - Use struct list_head lists
   - Support for CONFIG_PCI_DOMAINS_GENERIC
   - Make pcibios_set_cache_line_size an initcall
   - Inline pcibios_assign_all_busses
   - Split pci.c into pci.c & pci-legacy.c
   - Introduce CONFIG_PCI_DRIVERS_LEGACY
   - Support generic drivers

  CPC
   - Convert bare 'unsigned' to 'unsigned int'
   - Avoid lock when MIPS CM >= 3 is present

  GIC:
   - Delete unused file smp-gic.c

  mt7620:
   - Delete unnecessary assignment for the field "owner" from PCI

  BCM63xx:
   - Let clk_disable() return immediately if clk is NULL

  pm-cps:
   - Change FSB workaround to CPU blacklist
   - Update comments on barrier instructions
   - Use MIPS standard lightweight ordering barrier
   - Use MIPS standard completion barrier
   - Remove selection of sync types
   - Add MIPSr6 CPU support
   - Support CM3 changes to Coherence Enable Register

  SMP:
   - Wrap call to mips_cpc_lock_other in mips_cm_lock_other
   - Introduce mechanism for freeing and allocating IPIs

  cpuidle:
   - cpuidle-cps: Enable use with MIPSr6 CPUs.

  SEAD3:
   - Rewrite to use DT and generic kernel feature.

  USB:
   - host: ehci-sead3: Remove SEAD-3 EHCI code

  FBDEV:
   - cobalt_lcdfb: Drop SEAD3 support

  dt-bindings:
   -  Document a binding for simple ASCII LCDs

  auxdisplay:
   - img-ascii-lcd: driver for simple ASCII LCD displays

  irqchip i8259:
   - i8259: Add domain before mapping parent irq
   - i8259: Allow platforms to override poll function
   - i8259: Remove unused i8259A_irq_pending

  Malta:
   - Rewrite to use DT

  of/platform:
   - Probe "isa" busses by default

  CM:
   - Print CM error reports upon bus errors

  Module:
   - Migrate exception table users off module.h and onto extable.h
   - Make various drivers explicitly non-modular:
   - Audit and remove any unnecessary uses of module.h

  mailmap:
   - Canonicalize to Qais' current email address.

  Documentation:
   - MIPS supports HAVE_REGS_AND_STACK_ACCESS_API

  Loongson1C:
   - Add CPU support for Loongson1C
   - Add board support
   - Add defconfig
   - Add RTC support for Loongson1C board

  All this except one Documentation fix has sat in linux-next and has
  survived Imagination's automated build test system"

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus: (127 commits)
  Documentation: MIPS supports HAVE_REGS_AND_STACK_ACCESS_API
  MIPS: ptrace: Fix regs_return_value for kernel context
  MIPS: VDSO: Drop duplicated -I*/-E* aflags
  MIPS: Fix -mabi=64 build of vdso.lds
  MIPS: Enable hardened usercopy
  MIPS: generic: Convert SEAD-3 to a generic board
  MIPS: generic: Introduce generic DT-based board support
  MIPS: Support generating Flattened Image Trees (.itb)
  MIPS: Adjust MIPS64 CAC_BASE to reflect Config.K0
  MIPS: Print CM error reports upon bus errors
  MIPS: Support per-device DMA coherence
  MIPS: dma-default: Don't check hw_coherentio if device is non-coherent
  MIPS: Sanitise coherentio semantics
  MIPS: PCI: Support generic drivers
  MIPS: PCI: Introduce CONFIG_PCI_DRIVERS_LEGACY
  MIPS: PCI: Split pci.c into pci.c & pci-legacy.c
  MIPS: PCI: Inline pcibios_assign_all_busses
  MIPS: PCI: Make pcibios_set_cache_line_size an initcall
  MIPS: PCI: Support for CONFIG_PCI_DOMAINS_GENERIC
  MIPS: PCI: Use struct list_head lists
  ...

Pull sound fixes from Takashi Iwai:
 "Just a few trivial small fixes"

* tag 'sound-fix-4.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: line6: fix a crash in line6_hwdep_write()
  ALSA: seq: fix passing wrong pointer in function call of compatibility layer
  ALSA: hda - Fix a failure of micmute led when having multi adcs
  ALSA: line6: Fix POD X3 Live audio input

Pull more misc uaccess and vfs updates from Al Viro:
 "The rest of the stuff from -next (more uaccess work) + assorted fixes"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  score: traps: Add missing include file to fix build error
  fs/super.c: don't fool lockdep in freeze_super() and thaw_super() paths
  fs/super.c: fix race between freeze_super() and thaw_super()
  overlayfs: Fix setting IOP_XATTR flag
  iov_iter: kernel-doc import_iovec() and rw_copy_check_uvector()
  blackfin: no access_ok() for __copy_{to,from}_user()
  arm64: don't zero in __copy_from_user{,_inatomic}
  arm: don't zero in __copy_from_user_inatomic()/__copy_from_user()
  arc: don't leak bits of kernel stack into coredump
  alpha: get rid of tail-zeroing in __copy_user()

Pull cifs fixes from Steve French:
 "Including:

   - nine bug fixes for stable. Some of these we found at the recent two
     weeks of SMB3 test events/plugfests.

   - significant improvements in reconnection (e.g. if server or network
     crashes) especially when mounted with "persistenthandles" or to
     server which advertises Continuous Availability on the share.

   - a new mount option "idsfromsid" which improves POSIX compatibility
     in some cases (when winbind not configured e.g.) by better (and
     faster) fetching uid/gid from acl (when "cifsacl" mount option is
     enabled). NB: we are almost complete work on "cifsacl" (querying
     mode/uid/gid from ACL) for SMB3, but SMB3 support for cifsacl is
     not included in this set.

   - improved handling for SMB3 "credits" (even if server is buggy)

  Still working on two sets of changes:

   - cifsacl enablement for SMB3

   - cleanup of RFC1001 length calculation (so we can handle encryption
     and multichannel and RDMA)

  And a couple of new bugs were reported recently (unrelated to above)
  so will probably have another merge request next week"

* 'for-next' of git://git.samba.org/sfrench/cifs-2.6: (21 commits)
  CIFS: Retrieve uid and gid from special sid if enabled
  CIFS: Add new mount option to set owner uid and gid from special sids in acl
  CIFS: Reset read oplock to NONE if we have mandatory locks after reopen
  CIFS: Fix persistent handles re-opening on reconnect
  SMB2: Separate RawNTLMSSP authentication from SMB2_sess_setup
  SMB2: Separate Kerberos authentication from SMB2_sess_setup
  Expose cifs module parameters in sysfs
  Cleanup missing frees on some ioctls
  Enable previous version support
  Do not send SMB3 SET_INFO request if nothing is changing
  SMB3: Add mount parameter to allow user to override max credits
  fs/cifs: reopen persistent handles on reconnect
  Clarify locking of cifs file and tcon structures and make more granular
  Fix regression which breaks DFS mounting
  fs/cifs: keep guid when assigning fid to fileinfo
  SMB3: GUIDs should be constructed as random but valid uuids
  Set previous session id correctly on SMB3 reconnect
  cifs: Limit the overall credit acquired
  Display number of credits available
  Add way to query creation time of file via cifs xattr
  ...

Pull btrfs fixes from Chris Mason:
 "Some fixes from Omar and Dave Sterba for our new free space tree.

  This isn't heavily used yet, but as we move toward making it the new
  default we wanted to nail down an endian bug"

* 'for-linus-4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  btrfs: tests: uninline member definitions in free_space_extent
  btrfs: tests: constify free space extent specs
  Btrfs: expand free space tree sanity tests to catch endianness bug
  Btrfs: fix extent buffer bitmap tests on big-endian systems
  Btrfs: catch invalid free space trees
  Btrfs: fix mount -o clear_cache,space_cache=v2
  Btrfs: fix free space tree bitmaps on big-endian systems

score images fail to build as follows.

arch/score/kernel/traps.c: In function 'show_stack':
arch/score/kernel/traps.c:55:3: error:
	implicit declaration of function '__get_user'

__get_user() is declared in asm/uaccess.h, which was previously included
through asm/module.h.

Cc: Al Viro <viro@zeniv.linux.org.uk>
Fixes: 88dd4a74 ("score: separate extable.h, switch module.h to it")
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

sb_wait_write()->percpu_rwsem_release() fools lockdep to avoid the
false-positives. Now that xfs was fixed by Dave's commit dbad7c99
("xfs: stop holding ILOCK over filldir callbacks") we can remove it and
change freeze_super() and thaw_super() to run with s_writers.rw_sem locks
held; we add two trivial helpers for that, lockdep_sb_freeze_release()
and lockdep_sb_freeze_acquire().

xfstests-dev/check `grep -il freeze tests/*/???` does not trigger any
warning from lockdep.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Pull overlayfs updates from Miklos Szeredi:
 "This update contains fixes to the "use mounter's permission to access
  underlying layers" area, and miscellaneous other fixes and cleanups.

  No new features this time"

* 'overlayfs-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
  ovl: use vfs_get_link()
  vfs: add vfs_get_link() helper
  ovl: use generic_readlink
  ovl: explain error values when removing acl from workdir
  ovl: Fix info leak in ovl_lookup_temp()
  ovl: during copy up, switch to mounter's creds early
  ovl: lookup: do getxattr with mounter's permission
  ovl: copy_up_xattr(): use strnlen

Change thaw_super() to check frozen != SB_FREEZE_COMPLETE rather than
frozen == SB_UNFROZEN, otherwise it can race with freeze_super() which
drops sb->s_umount after SB_FREEZE_WRITE to preserve the lock ordering.

In this case thaw_super() will wrongly call s_op->unfreeze_fs() before
it was actually frozen, and call sb_freeze_unlock() which leads to the
unbalanced percpu_up_write(). Unfortunately lockdep can't detect this,
so this triggers misc BUG_ON()'s in kernel/rcu/sync.c.
Reported-and-tested-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

ovl_fill_super calls ovl_new_inode to create a root inode for the new
superblock before initializing sb->s_xattr.  This wrongly causes
IOP_XATTR to be cleared in i_opflags of the new inode, causing SELinux
to log the following message:

  SELinux: (dev overlay, type overlay) has no xattr support

Fix this by initializing sb->s_xattr and similar fields before calling
ovl_new_inode.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Both import_iovec() and rw_copy_check_uvector() take an array
(typically small and on-stack) which is used to hold an iovec array copy
from userspace. This is to avoid an expensive memory allocation in the
fast path (i.e. few iovec elements).

The caller may have to check whether these functions actually used
the provided buffer or allocated a new one -- but this differs between
the too. Let's just add a kernel doc to clarify what the semantics are
for each function.
Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Merge tag 'linux-kselftest-4.9-rc1-update' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest

Pull kselftest updates from Shuah Khan:
 "This update consists of:

   - Fixes and improvements to existing tests

   - Moving code from Documentation to selftests, samples, and tools:

     * Moves dnotify_test, prctl, ptp, vDSO, ia64, watchdog, and
       networking tests from Documentation to selftests.

     * Moves mic/mpssd, misc-devices/mei, timers, watchdog, auxdisplay,
       and blackfin examples from Documentation to samples.

     * Moves accounting, laptops/dslm, and pcmcia/crc32hash tools from
       Documentation to tools.

     * Deletes BUILD_DOCSRC and its dependencies"

* tag 'linux-kselftest-4.9-rc1-update' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest: (21 commits)
  selftests/futex: Check ANSI terminal color support
  Doc: update 00-INDEX files to reflect the runnable code move
  samples: move blackfin gptimers-example from Documentation
  tools: move pcmcia crc32hash tool from Documentation
  tools: move laptops dslm tool from Documentation
  tools: move accounting tool from Documentation
  samples: move auxdisplay example code from Documentation
  samples: move watchdog example code from Documentation
  samples: move timers example code from Documentation
  samples: move misc-devices/mei example code from Documentation
  samples: move mic/mpssd example code from Documentation
  selftests: Move networking/timestamping from Documentation
  selftests: move watchdog tests from Documentation/watchdog
  selftests: move ia64 tests from Documentation/ia64
  selftests: move vDSO tests from Documentation/vDSO
  selftests: move ptp tests from Documentation/ptp
  selftests: move prctl tests from Documentation/prctl
  selftests: move dnotify_test from Documentation/filesystems
  selftests/timers: Add missing error code assignment before test
  selftests/zram: replace ZRAM_LZ4_COMPRESS
  ...

Pull misc kbuild changes from Michal Marek:
 "Just a few patches on the kbuild.git#misc branch this time:

   - New Coccinelle patch by Nicholas Mc Guire
   - Existing patch fixes by Julia Lawall
   - Minor comment fix by Markus Elfring"

* 'misc' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
  Coccinelle: flag conditions with no effect
  scripts/coccicheck: Update reference for the corresponding documentation
  Coccinelle: pm_runtime: ensure relevance of pm_runtime reports
  Coccinelle: limit memdup_user transformation to GFP_KERNEL case