Inotify watch removals suck violently.

To kick the watch out we need (in this order) inode->inotify_mutex and
ih->mutex.  That's fine if we have a hold on inode; however, for all
other cases we need to make damn sure we don't race with umount.  We can
*NOT* just grab a reference to a watch - inotify_unmount_inodes() will
happily sail past it and we'll end with reference to inode potentially
outliving its superblock.

Ideally we just want to grab an active reference to superblock if we
can; that will make sure we won't go into inotify_umount_inodes() until
we are done.  Cleanup is just deactivate_super().

However, that leaves a messy case - what if we *are* racing with
umount() and active references to superblock can't be acquired anymore?
We can bump ->s_count, grab ->s_umount, which will almost certainly wait
until the superblock is shut down and the watch in question is pining
for fjords.  That's fine, but there is a problem - we might have hit the
window between ->s_active getting to 0 / ->s_count - below S_BIAS (i.e.
the moment when superblock is past the point of no return and is heading
for shutdown) and the moment when deactivate_super() acquires
->s_umount.

We could just do drop_super() yield() and retry, but that's rather
antisocial and this stuff is luser-triggerable.  OTOH, having grabbed
->s_umount and having found that we'd got there first (i.e.  that
->s_root is non-NULL) we know that we won't race with
inotify_umount_inodes().

So we could grab a reference to watch and do the rest as above, just
with drop_super() instead of deactivate_super(), right? Wrong.  We had
to drop ih->mutex before we could grab ->s_umount.  So the watch
could've been gone already.

That still can be dealt with - we need to save watch->wd, do idr_find()
and compare its result with our pointer.  If they match, we either have
the damn thing still alive or we'd lost not one but two races at once,
the watch had been killed and a new one got created with the same ->wd
at the same address.  That couldn't have happened in inotify_destroy(),
but inotify_rm_wd() could run into that.  Still, "new one got created"
is not a problem - we have every right to kill it or leave it alone,
whatever's more convenient.

So we can use idr_find(...) == watch && watch->inode->i_sb == sb as
"grab it and kill it" check.  If it's been our original watch, we are
fine, if it's a newcomer - nevermind, just pretend that we'd won the
race and kill the fscker anyway; we are safe since we know that its
superblock won't be going away.

And yes, this is far beyond mere "not very pretty"; so's the entire
concept of inotify to start with.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Acked-by: NGreg KH <greg@kroah.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The file(s) below do not use LINUX_VERSION_CODE nor KERNEL_VERSION.
  drivers/hwmon/lis3lv02d.c

This patch removes the said #include <version.h>.
Signed-off-by: NHuang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'sh/for-2.6.28' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6:
  serial: sh-sci: Reorder the SCxTDR write after the TDxE clear.
  sh: __copy_user function can corrupt the stack in case of exception
  sh: Fixed the TMU0 reload value on resume
  sh: Don't factor in PAGE_OFFSET for valid_phys_addr_range() check.
  sh: early printk port type fix
  i2c: fix i2c-sh_mobile rx underrun
  sh: Provide a sane valid_phys_addr_range() to prevent TLB reset with PMB.
  usb: r8a66597-hcd: fix wrong data access in SuperH on-chip USB
  fix sci type for SH7723
  serial: sh-sci: fix cannot work SH7723 SCIFA
  sh: Handle fixmap TLB eviction more coherently.

* 'doc-subdirs' of git://git.kernel.org/pub/scm/linux/kernel/git/rdunlap/linux-docs:
  Create/use more directory structure in the Documentation/ tree.

A common reason for device drivers to implement their own printk macros
is the lack of a printk prefix with the standard pr_xyz macros.
Introduce a pr_fmt() macro that is applied for every pr_xyz macro to the
format string.

The most common use of the pr_fmt macro would be to add the name of the
device driver to all pr_xyz messages in a source file.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ericvh/v9fs:
  9p: restrict RDMA usage

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6:
  V4L/DVB (9624): CVE-2008-5033: fix OOPS on tvaudio when controlling bass/treble
  V4L/DVB (9623): tvaudio: Improve debug msg by printing something more human
  V4L/DVB (9622): tvaudio: Improve comments and remove a unneeded prototype
  V4L/DVB (9621): Avoid writing outside shadow.bytes[] array
  V4L/DVB (9620): tvaudio: use a direct reference for chip description
  V4L/DVB (9619): tvaudio: update initial comments
  V4L/DVB (9618): tvaudio: add additional logic to avoid OOPS
  V4L/DVB (9617): tvtime: remove generic_checkmode callback
  V4L/DVB (9616): tvaudio: cleanup - group all callbacks together
  V4L/DVB (9615): tvaudio: instead of using a magic number, use ARRAY_SIZE
  V4L/DVB (9613): tvaudio: fix a memory leak

* 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6:
  [S390] fix s390x_newuname
  [S390] dasd: log sense for fatal errors
  [S390] cpu topology: fix locking
  [S390] cio: Fix refcount after moving devices.
  [S390] ftrace: fix kernel stack backchain walking
  [S390] ftrace: disable tracing on idle psw
  [S390] lockdep: fix compile bug
  [S390] kvm_s390: Fix oops in virtio device detection with "mem="
  [S390] sclp: emit error message if assign storage fails
  [S390] Fix range for add_active_range() in setup_memory()

* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
  [SCSI] dpt_i2o: fix transferred data length for scsi_set_resid()
  [SCSI] scsi_error regression: Fix idempotent command handling
  [SCSI] zfcp: Fix hexdump data in s390dbf traces
  [SCSI] zfcp: fix erp timeout cleanup for port open requests
  [SCSI] zfcp: Wait for port scan to complete when setting adapter online
  [SCSI] zfcp: Fix cast warning
  [SCSI] zfcp: Fix request list handling in error path
  [SCSI] zfcp: fix mempool usage for status_read requests
  [SCSI] zfcp: fix req_list_locking.
  [SCSI] zfcp: Dont clear reference from SCSI device to unit
  [SCSI] qla2xxx: Update version number to 8.02.01-k9.
  [SCSI] qla2xxx: Return a FAILED status when abort mailbox-command fails.
  [SCSI] qla2xxx: Do not honour max_vports from firmware for 2G ISPs and below.
  [SCSI] qla2xxx: Use pci_disable_rom() to manipulate PCI config space.
  [SCSI] qla2xxx: Correct Atmel flash-part handling.
  [SCSI] megaraid: fix mega_internal_command oops

This reverts commit e51af663, which was
wrongly hoovered up and submitted about a month after a better fix had
already been merged.

The better fix is commit cbda1ba8
("PCI/iommu: blacklist DMAR on Intel G31/G33 chipsets"), where we do
this blacklisting based on the DMI identification for the offending
motherboard, since sometimes this chipset (or at least a chipset with
the same PCI ID) apparently _does_ actually have an IOMMU.
Signed-off-by: NDavid Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Hugh Dickins reported show_page_path() is buggy and unsafe because

 - lack dput() against d_find_alias()
 - don't concern vma->vm_mm->owner == NULL
 - lack lock_page()

it was only for debugging, so rather than trying to fix it, just remove
it now.
Reported-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
CC: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
CC: Rik van Riel <riel@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

m68k allmodconfig:

| drivers/misc/c2port/core.c: In function 'c2port_reset':
| drivers/misc/c2port/core.c:73: error: dereferencing pointer to incomplete type
| drivers/misc/c2port/core.c: In function 'c2port_strobe_ck':
| drivers/misc/c2port/core.c:91: error: dereferencing pointer to incomplete type

Include <linux/sched.h> to fix it, as m68k's local_irq_enable() needs to know
about struct task_struct.
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

commit 69961c37 ("[PATCH] m68k/Atari:
Interrupt updates") added a BUG_ON() with an incorrect upper bound
comparison, which causes an early crash on VME boards, where IRQ_USER is
8, cnt is 192 and NR_IRQS is 200.
Reported-by: NStephen N Chivers <schivers@csc.com.au>
Tested-by: NKars de Jong <jongk@linux-m68k.org>
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  libata: improve phantom device detection

We don't want to get rid of the futexes just at exit() time, we want to
drop them when doing an execve() too, since that gets rid of the
previous VM image too.

Doing it at mm_release() time means that we automatically always do it
when we disassociate a VM map from the task.

Reported-by: pageexec@freemail.hu
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Brad Spengler <spender@grsecurity.net>
Cc: Alex Efros <powerman@powerman.name>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

dpt_i2o.c::adpt_i2o_to_scsi() reads the value at (reply+5) which
should contain the length in bytes of the transferred data. This
would be correct if reply was a u32 *. However it is a void * here,
so we need to read the value at (reply+20) instead.

The value at (reply+5) is usually 0xff0000, which is apparently
'large enough' and didn't cause any trouble until 2.6.27 where

commit 427e59f0
Author: James Bottomley <James.Bottomley@HansenPartnership.com>
Date:   Sat Mar 8 18:24:17 2008 -0600

    [SCSI] make use of the residue value

caused this to become visible through e.g. iostat -x .
Signed-off-by: NMiquel van Smoorenburg <mikevs@xs4all.net>
Cc: Stable Tree <stable@kernel.org>
Signed-off-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>

Currently libata uses four methods to detect device presence.

1. PHY status if available.
2. TF register R/W test (only promotes presence, never demotes)
3. device signature after reset
4. IDENTIFY failure detection in SFF state machine

Combination of the above works well in most cases but recently there
have been a few reports where a phantom device causes unnecessary
delay during probe.  In both cases, PHY status wasn't available.  In
one case, it passed #2 and #3 and failed IDENTIFY with ATA_ERR which
didn't qualify as #4.  The other failed #2 but as it passed #3 and #4,
it still caused failure.

In both cases, phantom device reported diagnostic failure, so these
cases can be safely worked around by considering any !ATA_DRQ IDENTIFY
failure as NODEV_HINT if diagnostic failure is set.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

linux-next:

Make 9p's RDMA option depend on INET since it uses Infiniband rdma_*
functions and that code depends on INET.  Otherwise 9p can try to
use symbols which don't exist.

ERROR: "rdma_destroy_id" [net/9p/9pnet_rdma.ko] undefined!
ERROR: "rdma_connect" [net/9p/9pnet_rdma.ko] undefined!
ERROR: "rdma_create_id" [net/9p/9pnet_rdma.ko] undefined!
ERROR: "rdma_create_qp" [net/9p/9pnet_rdma.ko] undefined!
ERROR: "rdma_resolve_route" [net/9p/9pnet_rdma.ko] undefined!
ERROR: "rdma_disconnect" [net/9p/9pnet_rdma.ko] undefined!
ERROR: "rdma_resolve_addr" [net/9p/9pnet_rdma.ko] undefined!

I used an if/endif block so that the menu items would remain
presented together.

Also correct an article adjective.
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

Create Documentation/blockdev/ sub-directory and populate it.
Populate the Documentation/serial/ sub-directory.
Move MSI-HOWTO.txt to Documentation/PCI/.
Move ioctl-number.txt to Documentation/ioctl/.
Update all relevant 00-INDEX files.
Update all relevant Kconfig files and source files.
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>

The uname system call for 64 bit compares current->personality without
masking the upper 16 bits. If e.g. READ_IMPLIES_EXEC is set the result
of a uname system call will always be s390x even if the process uses
the s390 personality.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The logging of sense data for fatal errors was accidentally removed
during Hyper PAV implementation.
Signed-off-by: NStefan Haberland <stefan.haberland@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

cpu_coregroup_map used to grab a mutex on s390 since it was only
called from process context.
Since c7c22e4d "block: add support
for IO CPU affinity" this is not true anymore.
It now also gets called from softirq context.

To prevent possible deadlocks change this in architecture code and
use a spinlock instead of a mutex.

Cc: stable@kernel.org
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

In ccw_device_move_to_orphanage(), a replacing ccw_device
is searched via get_{disc,orphaned}_ccwdev_by_dev_id()
which obtain a reference on the returned ccw_device.
This reference must be given up again after the device
has been moved to its new parent.
Signed-off-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

With CONFIG_IRQSOFF_TRACER the trace_hardirqs_off() function includes
a call to __builtin_return_address(1). But we calltrace_hardirqs_off()
from early entry code. There we have just a single stack frame.
So this results in a kernel stack backchain walk that would walk beyond
the kernel stack. Following the NULL terminated backchain this results
in a lowcore read access.

To fix this we simply call trace_hardirqs_off_caller() and pass the
current instruction pointer.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Disable tracing on idle psw. Otherwise it would give us huge
preempt off times for idle. Which is rather pointless.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

arch/s390/kernel/built-in.o: In function `cleanup_io_leave_insn':
mem_detect.c:(.text+0x10592): undefined reference to `lockdep_sys_exit'
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The current virtio model on s390 has the descriptor page above the main
memory. The guest virtio detection will oops if the mem= parameter is
used to reduce/change the memory size.
We have to use real_memory_size instead of max_pfn to detect the virtio
descriptor pages.
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>

Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

add_active_range() expects start_pfn + size as end_pfn value, i.e. not
the pfn of the last page frame but the one behind that.
We used the pfn of the last page frame so far, which can lead to a
BUG_ON in move_freepages(), when the kernelcore parameter is specified
(page_zone(start_page) != page_zone(end_page)).
Signed-off-by: NGerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

This bug were supposed to be fixed by 5ba2f67a,
where a call to NULL happens.

Not all tvaudio chips allow controlling bass/treble. So, the driver
has a table with a flag to indicate if the chip does support it.

Unfortunately, the handling of this logic were broken for a very long
time (probably since the first module version). Due to that, an OOPS
were generated for devices that don't support bass/treble.

This were the resulting OOPS message before the patch, with debug messages
enabled:

tvaudio' 1-005b: VIDIOC_S_CTRL
BUG: unable to handle kernel NULL pointer dereference at 00000000
IP: [<00000000>]
*pde = 22fda067 *pte = 00000000
Oops: 0000 [#1] SMP
Modules linked in: snd_hda_intel snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device
snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd_hwdep snd soundcore tuner_simple tuner_types tea5767 tuner
tvaudio bttv bridgebnep rfcomm l2cap bluetooth it87 hwmon_vid hwmon fuse sunrpc ipt_REJECT
nf_conntrack_ipv4 iptable_filter ip_tables ip6t_REJECT xt_tcpudp nf_conntrack_ipv6 xt_state nf_conntrack
ip6table_filter ip6_tables x_tables ipv6 dm_mirrordm_multipath dm_mod configfs videodev v4l1_compat
ir_common 8139cp compat_ioctl32 v4l2_common 8139too videobuf_dma_sg videobuf_core mii btcx_risc tveeprom
i915 button snd_page_alloc serio_raw drm pcspkr i2c_algo_bit i2c_i801 i2c_core iTCO_wdt
iTCO_vendor_support sr_mod cdrom sg ata_generic pata_acpi ata_piix libata sd_mod scsi_mod ext3 jbdmbcache
uhci_hcd ohci_hcd ehci_hcd [last unloaded: soundcore]

Pid: 15413, comm: qv4l2 Not tainted (2.6.25.14-108.fc9.i686 #1)
EIP: 0060:[<00000000>] EFLAGS: 00210246 CPU: 0
EIP is at 0x0
EAX: 00008000 EBX: ebd21600 ECX: e2fd9ec4 EDX: 00200046
ESI: f8c0f0c4 EDI: f8c0f0c4 EBP: e2fd9d50 ESP: e2fd9d2c
 DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process qv4l2 (pid: 15413, ti=e2fd9000 task=ebe44000 task.ti=e2fd9000)
Stack: f8c0c6ae e2ff2a00 00000d00 e2fd9ec4 ebc4e000 e2fd9d5c f8c0c448 00000000
       f899c12a e2fd9d5c f899c154 e2fd9d68 e2fd9d80 c0560185 e2fd9d88 f8f3e1d8
       f8f3e1dc ebc4e034 f8f3e18c e2fd9ec4 00000000 e2fd9d90 f899c286 c008561c
Call Trace:
 [<f8c0c6ae>] ? chip_command+0x266/0x4b6 [tvaudio]
 [<f8c0c448>] ? chip_command+0x0/0x4b6 [tvaudio]
 [<f899c12a>] ? i2c_cmd+0x0/0x2f [i2c_core]
 [<f899c154>] ? i2c_cmd+0x2a/0x2f [i2c_core]
 [<c0560185>] ? device_for_each_child+0x21/0x49
 [<f899c286>] ? i2c_clients_command+0x1c/0x1e [i2c_core]
 [<f8f283d8>] ? bttv_call_i2c_clients+0x14/0x16 [bttv]
 [<f8f23601>] ? bttv_s_ctrl+0x1bc/0x313 [bttv]
 [<f8f23445>] ? bttv_s_ctrl+0x0/0x313 [bttv]
 [<f8b6096d>] ? __video_do_ioctl+0x1f84/0x3726 [videodev]
 [<c05abb4e>] ? sock_aio_write+0x100/0x10d
 [<c041b23e>] ? kmap_atomic_prot+0x1dd/0x1df
 [<c043a0c9>] ? enqueue_hrtimer+0xc2/0xcd
 [<c04f4fa4>] ? copy_from_user+0x39/0x121
 [<f8b622b9>] ? __video_ioctl2+0x1aa/0x24a [videodev]
 [<c04054fd>] ? do_notify_resume+0x768/0x795
 [<c043c0f7>] ? getnstimeofday+0x34/0xd1
 [<c0437b77>] ? autoremove_wake_function+0x0/0x33
 [<f8b62368>] ? video_ioctl2+0xf/0x13 [videodev]
 [<c048c6f0>] ? vfs_ioctl+0x50/0x69
 [<c048c942>] ? do_vfs_ioctl+0x239/0x24c
 [<c048c995>] ? sys_ioctl+0x40/0x5b
 [<c0405bf2>] ? syscall_call+0x7/0xb
 [<c0620000>] ? cpuid4_cache_sysfs_exit+0x3d/0x69
 =======================
Code:  Bad EIP value.
EIP: [<00000000>] 0x0 SS:ESP 0068:e2fd9d2c
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Before the patch, the used ioctl were printed as an hexadecimal code,
hard to be understand without consulting the way _IO macros work.
Instead, use the V4L default handler for printing such errors into a way
that would be easier to understand.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Some comments are not clear enough. Improve it to allow a better
understanding of the driver behavior.

While there, remove an unneeded struct prototype.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

There were no check about the limits of shadow.bytes array. This offers
a risk of writing values outside the limits, overriding other data
areas.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Instead of storing the pointer for the proper entry at chip description
table, the driver were storing an indirect reference, by using an index.
Better to reference directly the data.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

A driver used on several bttv boards since 2000 is not experimental
anymore ;) Remove it from the comments.

While there, update copyrights addind a quick note about the "recent"
updates since 2005.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

This patch checks for volume, bass, treble, set mode and get mode
callbacks before actually enabling the code that would use them.

Instead of aborting the driver for load, this patch will allow it to
load with a reduced number of functionatities.

This prevents OOPS if some board entry is missing a needed callback.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

generic_checkmode() were called, via a callback, for some tvaudio chips.
There's just one callback code used on all those boards. So, it makes no
sense on keeping this as a callback.

Since there were some OOPS reported on tvaudio on kerneloops.org, this
patch removes this callback, adding the code at the only place were it
is called: inside chip_tread. A flag were added to indicate the need for
a kernel thread to set stereo mode on cards that needs it.

Using this more direct approach simplifies the code, making it more
robust against human errors.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Also, the default standard is the first one. So, fix the comment at the
array.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>