While building latest Linus git, I hit the following:

    CC [M]  drivers/staging/bcm/Qos.o
  drivers/staging/bcm/Qos.c: In function ‘PruneQueue’:
  drivers/staging/bcm/Qos.c:367: error: ‘struct netdev_queue’ has no member named ‘tx_dropped’
  drivers/staging/bcm/Qos.c: In function ‘flush_all_queues’:
  drivers/staging/bcm/Qos.c:416: error: ‘struct netdev_queue’ has no member named ‘tx_dropped’
  make[5]: *** [drivers/staging/bcm/Qos.o] Error 1
  make[4]: *** [drivers/staging/bcm] Error 2
  make[3]: *** [drivers/staging] Error 2

As well as:

    CC [M]  drivers/staging/bcm/Transmit.o
  drivers/staging/bcm/Transmit.c: In function ‘SetupNextSend’:
  drivers/staging/bcm/Transmit.c:163: error: ‘struct netdev_queue’ has no member named ‘tx_bytes’
  drivers/staging/bcm/Transmit.c:164: error: ‘struct netdev_queue’ has no member named ‘tx_packets’
  make[2]: *** [drivers/staging/bcm/Transmit.o] Error 1

tx_dropped/tx_bytes_tx_packets were removed in commit 1ac9ad13.  This patch
converts bcm to use net_device_stats instead of netdev_queue.
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Andres Salomon <dilinger@queued.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  RDMA: Update workqueue usage
  RDMA/nes: Fix incorrect SFP+ link status detection on driver init
  RDMA/nes: Fix SFP+ link down detection issue with switch port disable
  RDMA/nes: Generate IB_EVENT_PORT_ERR/PORT_ACTIVE events
  RDMA/nes: Fix bonding on iw_nes
  IB/srp: Test only once whether iu allocation succeeded
  IB/mlx4: Handle protocol field in multicast table
  RDMA: Use vzalloc() to replace vmalloc()+memset(0)
  mlx4_{core, ib, en}: Fix driver when sizeof (phys_addr_t) > sizeof (long)
  IB/mthca: Fix driver when sizeof (phys_addr_t) > sizeof (long)

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable: (25 commits)
  Btrfs: forced readonly mounts on errors
  btrfs: Require CAP_SYS_ADMIN for filesystem rebalance
  Btrfs: don't warn if we get ENOSPC in btrfs_block_rsv_check
  btrfs: Fix memory leak in btrfs_read_fs_root_no_radix()
  btrfs: check NULL or not
  btrfs: Don't pass NULL ptr to func that may deref it.
  btrfs: mount failure return value fix
  btrfs: Mem leak in btrfs_get_acl()
  btrfs: fix wrong free space information of btrfs
  btrfs: make the chunk allocator utilize the devices better
  btrfs: restructure find_free_dev_extent()
  btrfs: fix wrong calculation of stripe size
  btrfs: try to reclaim some space when chunk allocation fails
  btrfs: fix wrong data space statistics
  fs/btrfs: Fix build of ctree
  Btrfs: fix off by one while setting block groups readonly
  Btrfs: Add BTRFS_IOC_SUBVOL_GETFLAGS/SETFLAGS ioctls
  Btrfs: Add readonly snapshots support
  Btrfs: Refactor btrfs_ioctl_snap_create()
  btrfs: Extract duplicate decompress code
  ...

This reverts commit d8505dee.

Chris Mason ended up chasing down some page allocation errors and pages
stuck waiting on the IO scheduler, and was able to narrow it down to two
commits: commit 744ed144 ("mm: batch activate_page() to reduce lock
contention") and d8505dee ("mm: simplify code of swap.c").

This reverts the second one.
Reported-and-debugged-by: Chris Mason <chris.mason@oracle.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jens Axboe <jaxboe@fusionio.com>
Cc: linux-mm <linux-mm@kvack.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

This reverts commit 744ed144.

Chris Mason ended up chasing down some page allocation errors and pages
stuck waiting on the IO scheduler, and was able to narrow it down to two
commits: commit 744ed144 ("mm: batch activate_page() to reduce lock
contention") and d8505dee ("mm: simplify code of swap.c").

This reverts the first of them.
Reported-and-debugged-by: Chris Mason <chris.mason@oracle.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jens Axboe <jaxboe@fusionio.com>
Cc: linux-mm <linux-mm@kvack.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ecryptfs/ecryptfs-2.6:
  ecryptfs: remove unnecessary decrypt when extending a file
  ecryptfs: Fix ecryptfs_printk() size_t warnings
  fs/ecryptfs: Add printf format/argument verification and fix fallout
  ecryptfs: fixed testing of file descriptor flags
  ecryptfs: test lower_file pointer when lower_file_mutex is locked
  ecryptfs: missing initialization of the superblock 'magic' field
  ecryptfs: moved ECRYPTFS_SUPER_MAGIC definition to linux/magic.h
  ecryptfs: fix truncation error in ecryptfs_read_update_atime

On platforms that call panic() inside their BUG() macro (m68k/sun3, and
all platforms that don't set HAVE_ARCH_BUG), compilation fails with:

| fs/xfs/support/debug.c: In function ‘xfs_cmn_err’:
| fs/xfs/support/debug.c:92: error: called object ‘panic’ is not a function

as the local variable "panic" conflicts with the "panic()" function.
Rename the local variable to resolve this.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

This patch comes from "Forced readonly mounts on errors" ideas.

As we know, this is the first step in being more fault tolerant of disk
corruptions instead of just using BUG() statements.

The major content:
- add a framework for generating errors that should result in filesystems
  going readonly.
- keep FS state in disk super block.
- make sure that all of resource will be freed and released at umount time.
- make sure that fter FS is forced readonly on error, there will be no more
  disk change before FS is corrected. For this, we should stop write operation.

After this patch is applied, the conversion from BUG() to such a framework can
happen incrementally.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>

* 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6:
  spi/spi_sh_msiof: fix a wrong free_irq() parameter
  dt/flattree: Return virtual address from early_init_dt_alloc_memory_arch()

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  cifs: add cruid= mount option
  cifs: cFYI the entire error code in map_smb_to_linux_error

* git://git.infradead.org/mtd-2.6: (59 commits)
  mtd: mtdpart: disallow reading OOB past the end of the partition
  mtd: pxa3xx_nand: NULL dereference in pxa3xx_nand_probe
  UBI: use mtd->writebufsize to set minimal I/O unit size
  mtd: initialize writebufsize in the MTD object of a partition
  mtd: onenand: add mtd->writebufsize initialization
  mtd: nand: add mtd->writebufsize initialization
  mtd: cfi: add writebufsize initialization
  mtd: add writebufsize field to mtd_info struct
  mtd: OneNAND: OMAP2/3: prevent regulator sleeping while OneNAND is in use
  mtd: OneNAND: add enable / disable methods to onenand_chip
  mtd: m25p80: Fix JEDEC ID for AT26DF321
  mtd: txx9ndfmc: limit transfer bytes to 512 (ECC provides 6 bytes max)
  mtd: cfi_cmdset_0002: add support for Samsung K8D3x16UxC NOR chips
  mtd: cfi_cmdset_0002: add support for Samsung K8D6x16UxM NOR chips
  mtd: nand: ams-delta: drop omap_read/write, use ioremap
  mtd: m25p80: add debugging trace in sst_write
  mtd: nand: ams-delta: select for built-in by default
  mtd: OneNAND: lighten scary initial bad block messages
  mtd: OneNAND: OMAP2/3: add support for command line partitioning
  mtd: nand: rearrange ONFI revision checking, add ONFI 2.3
  ...

Fix up trivial conflict in drivers/mtd/Kconfig as per DavidW.

Removes an unecessary page decrypt from ecryptfs_begin_write when the
page is beyond the current file size. Previously, the call to
ecryptfs_decrypt_page would result in a read of 0 bytes, but still
attempt to decrypt an entire page. This patch detects that case and
merely zeros the page before marking it up-to-date.
Signed-off-by: Frank Swiderski <fes@chromium.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

Commit cb55d21f6fa19d8c6c2680d90317ce88c1f57269 revealed a number of
missing 'z' length modifiers in calls to ecryptfs_printk() when
printing variables of type size_t. This patch fixes those compiler
warnings.
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

Add __attribute__((format... to __ecryptfs_printk
Make formats and arguments match.
Add casts to (unsigned long long) for %llu.
Signed-off-by: Joe Perches <joe@perches.com>
[tyhicks: 80 columns cleanup and fixed typo]
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
  fs: fix address space warnings in ioctl_fiemap()
  aio: check return value of create_workqueue()
  hpfs_setattr error case avoids unlock_kernel
  compat: copy missing fields in compat_statfs64 to user
  compat: update comment of compat statfs syscalls
  compat: remove unnecessary assignment in compat_rw_copy_check_uvector()
  fs: FS_POSIX_ACL does not depend on BLOCK
  fs: Remove unlikely() from fget_light()
  fs: Remove unlikely() from fput_light()
  fallocate should be a file operation
  make the feature checks in ->fallocate future proof
  staging: smbfs building fix
  tidy up around finish_automount()
  don't drop newmnt on error in do_add_mount()
  Take the completion of automount into new helper

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mattst88/alpha-2.6:
  alpha: fix WARN_ON in __local_bh_enable()
  alpha: fix breakage caused by df9ee292
  alpha: add GENERIC_HARDIRQS_NO__DO_IRQ to Kconfig
  alpha/osf_sys: remove unused MAX_SELECT_SECONDS
  alpha: change to new Makefile flag variables
  alpha: kill off alpha_do_IRQ
  alpha: irq clean up
  alpha: use set_irq_chip and push down __do_IRQ to the machine types

* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon/kms: balance asic_reset functions
  drm/radeon/kms: remove duplicate card_posted() functions
  drm/radeon/kms: add module option for pcie gen2
  drm/radeon/kms: fix typo in evergreen safe reg
  drm/nouveau: fix gpu page faults triggered by plymouthd
  drm/nouveau: greatly simplify mm, killing some bugs in the process
  drm/nvc0: enable protection of system-use-only structures in vm
  drm/nv40: initialise 0x17xx on all chipsets that have it
  drm/nv40: make detection of 0x4097-ful chipsets available everywhere

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (63 commits)
  ARM: PL08x: cleanup comments
  Update CONFIG_MD_RAID6_PQ to CONFIG_RAID6_PQ in drivers/dma/iop-adma.c
  ARM: PL08x: fix a warning
  Fix dmaengine_submit() return type
  dmaengine: at_hdmac: fix race while monitoring channel status
  dmaengine: at_hdmac: flags located in first descriptor
  dmaengine: at_hdmac: use subsys_initcall instead of module_init
  dmaengine: at_hdmac: no need set ACK in new descriptor
  dmaengine: at_hdmac: trivial add precision to unmapping comment
  dmaengine: at_hdmac: use dma_address to program DMA hardware
  pch_dma: support new device ML7213 IOH
  ARM: PL08x: prevent dma_set_runtime_config() reconfiguring memcpy channels
  ARM: PL08x: allow dma_set_runtime_config() to return errors
  ARM: PL08x: fix locking between prepare function and submit function
  ARM: PL08x: introduce 'phychan_hold' to hold on to physical channels
  ARM: PL08x: put txd's on the pending list in pl08x_tx_submit()
  ARM: PL08x: rename 'desc_list' as 'pend_list'
  ARM: PL08x: implement unmapping of memcpy buffers
  ARM: PL08x: store prep_* flags in async_tx structure
  ARM: PL08x: shrink srcbus/dstbus in txd structure
  ...

This patch replaces the check (lower_file->f_flags & O_RDONLY) with
((lower_file & O_ACCMODE) == O_RDONLY).
Signed-off-by: Roberto Sassu <roberto.sassu@polito.it>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

This patch prevents the lower_file pointer in the 'ecryptfs_inode_info'
structure to be checked when the mutex 'lower_file_mutex' is not locked.
Signed-off-by: Roberto Sassu <roberto.sassu@polito.it>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

This patch initializes the 'magic' field of ecryptfs filesystems to
ECRYPTFS_SUPER_MAGIC.
Signed-off-by: Roberto Sassu <roberto.sassu@polito.it>
[tyhicks: merge with 66cb7666]
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

Without this fix reloading of the driver is impossible.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

The definition of ECRYPTFS_SUPER_MAGIC has been moved to the include
file 'linux/magic.h' to become available to other kernel subsystems.
Signed-off-by: Roberto Sassu <roberto.sassu@polito.it>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

This is similar to the bug found in direct-io not so long ago.

Fix up truncation (ssize_t->int).  This only matters with >2G
reads/writes, which the kernel doesn't permit.
Signed-off-by: Edward Shishkin <edward.shishkin@gmail.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Eric Sandeen <esandeen@redhat.com>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

This patch fixes the mtdpart bug which allows users reading OOB past the
end of the partition. This happens because 'part_read_oob()' allows reading
multiple OOB areas in one go, and mtdparts does not validate the OOB
length in the request.

Although there is such check in 'nand_do_read_oob()' in nand_base.c, but
it checks that we do not read past the flash chip, not the partition,
because in nand_base.c we work with the whole chip (e.g., mtd->size
in nand_base.c is the size of the whole chip). So this check cannot
be done correctly in nand_base.c and should be instead done in mtdparts.c.

This problem was reported by Jason Liu <r64343@freescale.com> and reproduced
with nandsim:

$ modprobe nandsim first_id_byte=0x20 second_id_byte=0xaa third_id_byte=0x00 \
                   fourth_id_byte=0x15 parts=0x400,0x400
$ modprobe nandsim mtd_oobtest.ko dev=0
$ dmesg
= snip =
mtd_oobtest: attempting to read past end of device
mtd_oobtest: an error is expected...
mtd_oobtest: error: read past end of device
= snip =
mtd_oobtest: finished with 2 errors
Reported-by: Jason Liu <liu.h.jason@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>

The fi_extents_start field of struct fiemap_extent_info is a
user pointer but was not marked as __user. This makes sparse
emit following warnings:

  CHECK   fs/ioctl.c
fs/ioctl.c:114:26: warning: incorrect type in argument 1 (different address spaces)
fs/ioctl.c:114:26:    expected void [noderef] <asn:1>*dst
fs/ioctl.c:114:26:    got struct fiemap_extent *[assigned] dest
fs/ioctl.c:202:14: warning: incorrect type in argument 1 (different address spaces)
fs/ioctl.c:202:14:    expected void const volatile [noderef] <asn:1>*<noident>
fs/ioctl.c:202:14:    got struct fiemap_extent *[assigned] fi_extents_start
fs/ioctl.c:212:27: warning: incorrect type in argument 1 (different address spaces)
fs/ioctl.c:212:27:    expected void [noderef] <asn:1>*dst
fs/ioctl.c:212:27:    got char *<noident>

Also add 'ufiemap' variable to eliminate unnecessary casts.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

This fixed a case that 'sparse' spotted where hpfs_setattr has an error return
that didn't go through it's path that unlocks.

This is against git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
version 6313e3c2.

Build tested only, I don't have an hpfs file system to test.

Dave
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

f_flags and f_spare fields were not copied to userspace when
compat_sys_[f]statfs64 called.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

The commit 7ed1ee61 ("Take statfs variants to fs/statfs.c")
separates out statfs syscalls from fs/open.c. Thus the comment
should be changed also.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Cc: Jiri Kosina <trivial@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

*@ret_pointer is initialized to @fast_pointer thus the assignment is
redundant.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

- Fix a kconfig unmet dependency warning.
- Remove the comment that identifies which filesystems use POSIX ACL
  utility routines.
- Move the FS_POSIX_ACL symbol outside of the BLOCK symbol if/endif block
  because its functions do not depend on BLOCK and some of the filesystems
  that use it do not depend on BLOCK.

warning: (GENERIC_ACL && JFFS2_FS_POSIX_ACL && NFSD_V4 && NFS_ACL_SUPPORT && 9P_FS_POSIX_ACL) selects FS_POSIX_ACL which has unmet direct dependencies (BLOCK)
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

There's an unlikely() in fget_light() that assumes the file ref count
will be 1. Running the annotate branch profiler on a desktop that is
performing daily tasks (running firefox, evolution, xchat and is also part
of a distcc farm), it shows that the ref count is not 1 that often.

 correct incorrect      %    Function                  File              Line
 ------- ---------      -    --------                  ----              ----
1035099358 6209599193  85    fget_light              file_table.c         315

Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

In fput_light(), there's an unlikely(fput_needed), which running on
my normal desktop doing firefox, xchat, evolution and part of my distcc farm,
and running the annotate branch profiler shows that the unlikely is not
very unlikely.

 correct incorrect  %        Function             File              Line
 ------- ---------  -        --------             ----              ----
       0       48 100 fput_light                file.h               26
115828710 897415279  88 fput_light              file.h               26
865271179 5286128445  85 fput_light             file.h               26
19568539  8923664  31 fput_light                file.h               26
12353677  3562279  22 fput_light                file.h               26
  267691    67062  20 fput_light                file.h               26
15014853   348172   2 fput_light                file.h               26
  209258      205   0 fput_light                file.h               26
 1364164        0   0 fput_light                file.h               26

Which gives 1032903812 times it was correct and 6203351846 times it was
incorrect, or 85% incorrect.

Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Currently all filesystems except XFS implement fallocate asynchronously,
while XFS forced a commit.  Both of these are suboptimal - in case of O_SYNC
I/O we really want our allocation on disk, especially for the !KEEP_SIZE
case where we actually grow the file with user-visible zeroes.  On the
other hand always commiting the transaction is a bad idea for fast-path
uses of fallocate like for example in recent Samba versions.   Given
that block allocation is a data plane operation anyway change it from
an inode operation to a file operation so that we have the file structure
available that lets us check for O_SYNC.

This also includes moving the code around for a few of the filesystems,
and remove the already unnedded S_ISDIR checks given that we only wire
up fallocate for regular files.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Instead of various home grown checks that might need updates for new
flags just check for any bit outside the mask of the features supported
by the filesystem.  This makes the check future proof for any newly
added flag.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Building error for smbfs:

drivers/staging/smbfs/dir.c:286: error: static declaration of 'smbfs_dentry_operations' follows non-static declaration
drivers/staging/smbfs/proto.h:42: error: previous declaration of 'smbfs_dentry_operations' was here
drivers/staging/smbfs/dir.c:294: error: static declaration of 'smbfs_dentry_operations_case' follows non-static declaration
drivers/staging/smbfs/proto.h:41: error: previous declaration of 'smbfs_dentry_operations_case' was here
make[3]: *** [drivers/staging/smbfs/dir.o] Error 1
make[2]: *** [drivers/staging/smbfs] Error 2
make[1]: *** [drivers/staging] Error 2
make[1]: *** Waiting for unfinished jobs....

Fix it by removing static keywords
Signed-off-by: Yang Ruirui <ruirui.r.yang@tieto.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

do_add_mount() and mnt_clear_expiry() are not needed outside of
namespace.c anymore, now that namei has finish_automount() to
use.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

That gets rid of the kludge in finish_automount() - we need
to keep refcount on the vfsmount as-is until we evict it from
expiry list.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

... and shift it from namei.c to namespace.c
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>