mm-add-extra-free-kbytes-tunable-update

All the fixes suggested by Andrew Morton.   Not much of a changelog
since the patch should probably be folded into
mm-add-extra-free-kbytes-tunable.patch

Thank you for pointing these out, Andrew.

Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: add extra free kbytes tunable

Add a userspace visible knob to tell the VM to keep an extra amount of
memory free, by increasing the gap between each zone's min and low
watermarks.

This is useful for realtime applications that call system calls and have a
bound on the number of allocations that happen in any short time period.
In this application, extra_free_kbytes would be left at an amount equal to
or larger than than the maximum number of allocations that happen in any
burst.

It may also be useful to reduce the memory use of virtual machines
(temporarily?), in a way that does not cause memory fragmentation like
ballooning does.

Testing results from Satoru Moriya:

: I ran some sample workloads and measure memory allocation latency
: (latency of __alloc_page_nodemask()).
: The test is like following:
:
:  - CPU: 1 socket, 4 core
:  - Memory: 4GB
:
:  - Background load:
:    $ dd if=3D/dev/zero of=3D/tmp/tmp1
:    $ dd if=3D/dev/zero of=3D/tmp/tmp2
:    $ dd if=3D/dev/zero of=3D/tmp/tmp3
:
:  - Main load:
:    $ mapped-file-stream 1 $((1024 * 1024 * 640))  --(*)
:
:  (*) This is made by Johannes Weiner
:      https://lkml.org/lkml/2010/8/30/226
:
:      It allocates/access 640MByte memory at a burst.
:
: The result is follwoing:
:
:                                |         |  extra   |
:                                | default |  kbytes  |
: --------------------------------------------------------------
: min_free_kbytes                |    8113 |   8113   |
: extra_free_kbytes              |       0 | 640*1024 | (KB)
: --------------------------------------------------------------
: worst latency                  | 517.762 |  20.775  | (usec)
: --------------------------------------------------------------
: vmstat result                  |         |          |
:  nr_vmscan_write               |       0 |      0   |
:  pgsteal_dma                   |       0 |      0   |
:  pgsteal_dma32                 |  143667 | 144882   |
:  pgsteal_normal                |   31486 |  27001   |
:  pgsteal_movable               |       0 |      0   |
:  pgscan_kswapd_dma             |       0 |      0   |
:  pgscan_kswapd_dma32           |  138617 | 156351   |
:  pgscan_kswapd_normal          |   30593 |  27955   |
:  pgscan_kswapd_movable         |       0 |      0   |
:  pgscan_direct_dma             |       0 |      0   |
:  pgscan_direct_dma32           |    5050 |      0   |
:  pgscan_direct_normal          |     896 |      0   |
:  pgscan_direct_movable         |       0 |      0   |
:  kswapd_steal                  |  169207 | 171883   |
:  kswapd_inodesteal             |       0 |      0   |
:  kswapd_low_wmark_hit_quickly  |      43 |     45   |
:  kswapd_high_wmark_hit_quickly |       1 |      0   |
:  allocstall                    |      32 |      0   |
:
:
: As you can see, in the default case there were 32 direct reclaim
: (allocstal= l) and its worst latency was 517.762 usecs.  This value may be
: larger if a process would sleep or issue I/O in the direct reclaim path.
: OTOH, ii the other case where I add extra free bytes, there were no direct
: reclaim and its worst latency was 20.775 usecs.
:
: In this test case, we can avoid direct reclaim and keep a latency low.

Signed-off-by: Rik van Riel<riel@redhat.com>
Acked-by: Johannes Weiner <jweiner@redhat.com>
Tested-by: Satoru Moriya <satoru.moriya@hds.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: fix page-faults detection in swap-token logic

After commit v2.6.36-5896-gd065bd8 "mm: retry page fault when blocking on
disk transfer" we usually wait in page-faults without mmap_sem held, so
all swap-token logic was broken, because it based on using
rwsem_is_locked(&mm->mmap_sem) as sign of in progress page-faults.

Add an atomic counter of in progress page-faults for mm to the mm_struct
with swap-token.

Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/memory_failure: let the compiler add the function name

These things tend to get out of sync with time so let the compiler
automatically enter the current function name using __func__.

No functional change.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
Acked-by: Andi Kleen <andi@firstfloor.org>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

brlocks/lglocks: cleanups

lglocks and brlocks are currently generated with some complicated macros
in lglock.h.  But there's no reason to not just use common utility
functions and put all the data into a common data structure.

Since there are at least two users it makes sense to share this code in a
library.  This is also easier maintainable than a macro forest.

This will also make it later possible to dynamically allocate lglocks and
also use them in modules (this would both still need some additional, but
now straightforward, code)

In general the users now look more like normal function calls with
pointers, not magic macros.

The patch is rather large because I move over all users in one go to keep
it bisectable.  This impacts the VFS somewhat in terms of lines changed.
But no actual behaviour change.

[akpm@linux-foundation.org: checkpatch fixes]
[levinsasha928@gmail.com: fix dup_mnt_ns()]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

fs: hardlink creation restrictions

On systems that have user-writable directories on the same partition as
system files, a long-standing class of security issues is the
hardlink-based time-of-check-time-of-use race, most commonly seen in
world-writable directories like /tmp.  The common method of exploitation
of this flaw is to cross privilege boundaries when following a given
hardlink (i.e.  a root process follows a hardlink created by another
user).  Additionally, an issue exists where users can "pin" a potentially
vulnerable setuid/setgid file so that an administrator will not actually
upgrade a system fully.

The solution is to permit hardlinks to only be created when the user is
already the existing file's owner, or if they already have read/write
access to the existing file.

Many Linux users are surprised when they learn they can link to files they
have no access to, so this change appears to follow the doctrine of "least
surprise".  Additionally, this change does not violate POSIX, which states
"the implementation may require that the calling process has permission to
access the existing file"[1].

This change is known to break some implementations of the "at" daemon,
though the version used by Fedora and Ubuntu has been fixed[2] for a
while.  Otherwise, the change has been undisruptive while in use in Ubuntu
for the last 1.5 years.

This patch is based on the patch in Openwall and grsecurity.  I have added
a sysctl to enable the protected behavior, documentation, and an audit
notification.

[1] http://pubs.opengroup.org/onlinepubs/9699919799/functions/linkat.html
[2] http://anonscm.debian.org/gitweb/?p=collab-maint/at.git;a=commitdiff;h=f4114656c3a6c6f6070e315ffdf940a49eda3279

[akpm@linux-foundation.org: uninline may_linkat() and audit_log_link_denied()]
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Rik van Riel <riel@redhat.com>
Cc: Federica Teodori <federica.teodori@googlemail.com>
Cc: Lucian Adrian Grijincu <lucian.grijincu@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Eric Paris <eparis@redhat.com>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Dan Rosenberg <drosenberg@vsecurity.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

fs: symlink restrictions on sticky directories

A longstanding class of security issues is the symlink-based
time-of-check-time-of-use race, most commonly seen in world-writable
directories like /tmp.  The common method of exploitation of this flaw is
to cross privilege boundaries when following a given symlink (i.e.  a root
process follows a symlink belonging to another user).  For a likely
incomplete list of hundreds of examples across the years, please see:
http://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=/tmp

The solution is to permit symlinks to only be followed when outside a
sticky world-writable directory, or when the uid of the symlink and
follower match, or when the directory owner matches the symlink's owner.

Some pointers to the history of earlier discussion that I could find:

 1996 Aug, Zygo Blaxell
  http://marc.info/?l=bugtraq&m=87602167419830&w=2
 1996 Oct, Andrew Tridgell
  http://lkml.indiana.edu/hypermail/linux/kernel/9610.2/0086.html
 1997 Dec, Albert D Cahalan
  http://lkml.org/lkml/1997/12/16/4
 2005 Feb, Lorenzo Hernández García-Hierro
  http://lkml.indiana.edu/hypermail/linux/kernel/0502.0/1896.html
 2010 May, Kees Cook
  https://lkml.org/lkml/2010/5/30/144

Past objections and rebuttals could be summarized as:

 - Violates POSIX.
   - POSIX didn't consider this situation and it's not useful to follow
     a broken specification at the cost of security.
 - Might break unknown applications that use this feature.
   - Applications that break because of the change are easy to spot and
     fix. Applications that are vulnerable to symlink ToCToU by not having
     the change aren't. Additionally, no applications have yet been found
     that rely on this behavior.
 - Applications should just use mkstemp() or O_CREATE|O_EXCL.
   - True, but applications are not perfect, and new software is written
     all the time that makes these mistakes; blocking this flaw at the
     kernel is a single solution to the entire class of vulnerability.
 - This should live in the core VFS.
   - This should live in an LSM. (https://lkml.org/lkml/2010/5/31/135)
 - This should live in an LSM.
   - This should live in the core VFS. (https://lkml.org/lkml/2010/8/2/188)

This patch is based on the patch in Openwall and grsecurity, along with
suggestions from Al Viro.  I have added a sysctl to enable the protected
behavior, documentation, and an audit notification.

[akpm@linux-foundation.org: move sysctl_protected_sticky_symlinks declaration into .h]
Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Rik van Riel <riel@redhat.com>
Cc: Federica Teodori <federica.teodori@googlemail.com>
Cc: Lucian Adrian Grijincu <lucian.grijincu@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Eric Paris <eparis@redhat.com>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Dan Rosenberg <drosenberg@vsecurity.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

vfs: increment iversion when a file is truncated

When a file is truncated with truncate()/ftruncate() and then closed,
iversion is not updated.  This patch uses ATTR_SIZE flag as an indication
to increment iversion.

Mimi said:

On fput(), i_version is used to detect and flag files that have changed
and need to be re-measured in the IMA measurement policy.  When a file
is truncated with truncate()/ftruncate() and then closed, i_version is
not updated.  As a result, although the file has changed, it will not be
re-measured and added to the IMA measurement list on subsequent access.

Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
Acked-by: Mimi Zohar <zohar@us.ibm.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

parisc: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Helge Deller <deller@gmx.de>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ocfs2: use bitmap_weight()

Use bitmap_weight() instead of reinventing the wheel.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ocfs2: use find_last_bit()

We already have find_last_bit().  So just use it as described in the
comment.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

blackfin: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

unicore32: use block_sigmask()

Use the new helper function introduced in commit 5e6292c0f28f ("signal:
add block_sigmask() for adding sigmask to current->blocked") which
centralises the code for updating current->blocked after successfully
delivering a signal and reduces the amount of duplicate code across
architectures.  In the past some architectures got this code wrong, so
using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Guan Xuetao <gxt@mprc.pku.edu.cn>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

h8300: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

score: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Chen Liqin <liqin.chen@sunplusct.com>
Cc: Lennox Wu <lennox.wu@gmail.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

score: don't mask signals if we fail to setup signal stack

If setup_rt_frame() returns -EFAULT then we must not block any signals
in the current process.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Chen Liqin <liqin.chen@sunplusct.com>
Cc: Lennox Wu <lennox.wu@gmail.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

MIPS: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

microblaze: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

microblaze: fix signal masking

There are a couple of problems with the current signal code,

1. If we failed to setup the signal stack frame then we should not be
   masking any signals.

2. ka->sa.sa_mask is only added to the current blocked signals list if
   SA_NODEFER is set in ka->sa.sa_flags.  If we successfully setup the
   signal frame and are going to run the handler then we must honour
   sa_mask.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

microblaze: no need to reset handler if SA_ONESHOT

get_signal_to_deliver() already resets the signal handler if SA_ONESHOT is
set in ka->sa.sa_flags, there's no need to do it again in handle_signal().
 Furthermore, because we were modifying ka->sa.sa_handler (which is a copy
of sighand->action[]) instead of sighand->action[] the original code
actually had no effect on signal delivery.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

microblaze: don't reimplement force_sigsegv()

Instead of open coding the sequence from force_sigsegv() just call it.
This also fixes a bug because we were modifying ka->sa.sa_handler (which
is a copy of sighand->action[]), whereas the intention of the code was to
modify sighand->action[] directly.

As the original code was working with a copy it had no effect on signal
delivery.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ia64: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

percpu-remove-percpu_xxx-functions-fix

Cc: Alex Shi <alex.shi@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

percpu: remove percpu_xxx() functions

There are no percpu_xxx callers remaining

Signed-off-by: Alex Shi <alex.shi@intel.com>
Acked-by: Christoph Lameter <cl@gentwo.org>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

net: use this_cpu_xxx replace percpu_xxx funcs

percpu_xxx funcs are duplicated with this_cpu_xxx funcs, so replace them
for further code clean up.

And in preempt safe scenario, __this_cpu_xxx funcs has a bit better
performance since __this_cpu_xxx has no redundant preempt_disable()

Signed-off-by: Alex Shi <alex.shi@intel.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: David S. Miller <davem@davemloft.net>
Cc: Patrick McHardy <kaber@trash.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86: change percpu_read_stable() to this_cpu_read_stable()

It has no function change. It's a preparation for percpu_xxx serial
function change.

Signed-off-by: Alex Shi <alex.shi@intel.com>
Acked-by: Christoph Lameter <cl@gentwo.org>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86-use-this_cpu_xxx-to-replace-percpu_xxx-funcs-fix

Cc: Alex Shi <alex.shi@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86: use this_cpu_xxx to replace percpu_xxx funcs

Since percpu_xxx() serial functions are duplicate with this_cpu_xxx().
Removing percpu_xxx() definition and replacing them by this_cpu_xxx() in
code.

And further more, as Christoph Lameter's requirement, I try to use
__this_cpu_xx to replace this_cpu_xxx if it is in preempt safe scenario.
The preempt safe scenarios include:
1, in irq/softirq/nmi handler
2, protected by preempt_disable
3, protected by spin_lock
4, if the code context imply that it is preempt safe, like the code is
follows or be followed a preempt safe code.

BTW, In fact, this_cpu_xxx are same as __this_cpu_xxx since all funcs
implement in a single instruction for x86 machine.  But it maybe other
platforms' performance.

[akpm@linux-foundation.org: fix build]
[sfr@canb.auug.org.au: arch/x86/include/asm/desc.h: fix smp_processor_id's need for this_cpu_read]
Signed-off-by: Alex Shi <alex.shi@intel.com>
Acked-by: Christoph Lameter <cl@gentwo.org>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

cris: select GENERIC_ATOMIC64

Cris doesn't implement atomic64 operations neither, should select
GENERIC_ATOMIC64.

Signed-off-by: WANG Cong <xiyou.wangcong@gmail.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

cris: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Mikael Starvik <starvik@axis.com>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mn10300: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: David Howells <dhowells@redhat.com>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

m68k: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

m32r: use set_current_blocked() and block_sigmask()

As described in e6fa16ab ("signal: sigprocmask() should do
retarget_shared_pending()") the modification of current->blocked is
incorrect as we need to check whether the signal we're about to block is
pending in the shared queue.

Also, use the new helper function introduced in commit 5e6292c0f28f
("signal: add block_sigmask() for adding sigmask to current->blocked")
which centralises the code for updating current->blocked after
successfully delivering a signal and reduces the amount of duplicate code
across architectures.  In the past some architectures got this code wrong,
so using this helper function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Kyle McMartin <kyle@redhat.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

avr32: use block_sigmask()

Use the new helper function introduced in commit 5e6292c0f28f ("signal:
add block_sigmask() for adding sigmask to current->blocked") which
centralises the code for updating current->blocked after successfully
delivering a signal and reduces the amount of duplicate code across
architectures.

In the past some architectures got this code wrong, so using this helper
function should stop that from happening again.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Cc: Havard Skinnemoen <hskinnemoen@gmail.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

avr32: use set_current_blocked() in handle_signal/sys_rt_sigreturn

It is wrong to change ->blocked directly, see e6fa16ab.  Change
handle_signal() and sys_rt_sigreturn() to use the right helper,
set_current_blocked().

Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Acked-by: Havard Skinnemoen <hskinnemoen@gmail.com>
Reviewed-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

avr32: don't mask signals in the error path

The current handle_signal() implementation is broken - it will mask
signals if we fail to setup the signal stack frame, which isn't the
desired behaviour, we should only be masking signals if we succeed in
setting up the stack frame.  It looks like this code was copied from the
old (broken) arm implementation but wasn't updated when the arm code was
fixed in commit a6c61e9dfdd0 ("[ARM] 3168/1: Update ARM signal delivery
and masking").

Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Acked-by: Havard Skinnemoen <hskinnemoen@gmail.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

arch/arm/mach-ux500/mbox-db5500.c: world-writable sysfs fifo file

Don't allow everybody to use a modem.

Signed-off-by: Vasiliy Kulikov <segoon@openwall.com>
Cc: Srinidhi Kasagar <srinidhi.kasagar@stericsson.com>
Cc: Linus Walleij <linus.walleij@stericsson.com>
Cc: Russell King <linux@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

drivers/xen/Kconfig: fix Kconfig layout

Fit it into 80 columns so that it is readable in menuconfig.

Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

intel_mid_powerbtn: mark irq as IRQF_NO_SUSPEND

So that the power button still wakes up the platform.

Signed-off-by: Pierre Tardy <pierre.tardy@intel.com>
Tested-by: Kangkai Yin <kangkai.yin@intel.com>
Tested-by: Yong Wang <yong.y.wang@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Matthew Garrett <mjg@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

arch/x86/platform/iris/iris.c: register a platform device and a platform driver

This makes the iris driver use the platform API, so it is properly exposed
in /sys.

[akpm@linux-foundation.org: remove commented-out code, add missing space to printk, clean up code layout]
Signed-off-by: Shérab <Sebastien.Hinderer@ens-lyon.org>
Cc: Len Brown <lenb@kernel.org>
Cc: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

acerhdf: lowered default temp fanon/fanoff values

Due to new supported hardware, of which the actual temperature limits of
processor, harddisk and other components are unknown, it feels safer with
lower fanon / fanoff settings.

It won't change much for most people, already using acerhdf, as they use
their own fanon/fanoff variable settings when loading the module.

Furthermore seems like kernel and userspace tools have been improved to
work more efficient and netbooks don't get so hot anymore.

Signed-off-by: Peter Feuerer <peter@piie.net>
Acked-by: Borislav Petkov <petkovbb@gmail.com>
Cc: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

acerhdf: add support for new hardware

Add support for new hardware:
Acer Aspire LT-10Q/531/751/1810/1825,
Acer Travelmate 7730,
Packard Bell ENBFT/DOTVR46

Signed-off-by: Peter Feuerer <peter@piie.net>
Acked-by: Borislav Petkov <petkovbb@gmail.com>
Cc: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

acerhdf: add support for Aspire 1410 BIOS v1.3314

Add support for Aspire 1410 BIOS v1.3314.  Fixes the following error:

acerhdf: unknown (unsupported) BIOS version Acer/Aspire 1410/v1.3314,
please report, aborting!

Signed-off-by: Clay Carpenter <claycarpenter@gmail.com>
Signed-off-by: Peter Feuerer <peter@piie.net>
Cc: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

Merge remote-tracking branch 'cpuidle-cons/cpuidle_consol_pull'

Conflicts:
arch/arm/include/asm/cpuidle.h
arch/arm/mach-at91/cpuidle.c
arch/arm/mach-davinci/cpuidle.c
arch/arm/mach-shmobile/cpuidle.c
drivers/cpuidle/cpuidle.c
include/linux/cpuidle.h

Merge remote-tracking branch 'dma-buf/for-next'

Conflicts:
include/linux/dma-buf.h

Merge remote-tracking branch 'kvmtool/master'

Merge remote-tracking branch 'arm-soc/for-next'

Merge remote-tracking branch 'gpio/gpio/next'

Merge remote-tracking branch 'irqdomain/irqdomain/next'

Conflicts:
arch/arm/Kconfig

Merge remote-tracking branch 'modem-shm/for-next'

Merge remote-tracking branch 'vhost/linux-next'

Merge remote-tracking branch 'moduleh/for-sfr'

Merge remote-tracking branch 'tegra/for-next'

Merge remote-tracking branch 'pinctrl/for-next'

Merge remote-tracking branch 'tmem/linux-next'

Merge remote-tracking branch 'staging/staging-next'

Merge remote-tracking branch 'usb/usb-next'

Merge remote-tracking branch 'tty/tty-next'

Merge remote-tracking branch 'driver-core/driver-core-next'

Merge commit 'refs/next/20120410/regmap'

Merge remote-tracking branch 'xen-two/linux-next'

Conflicts:
arch/x86/include/asm/io_apic.h
arch/x86/kernel/apic/io_apic.c

Merge remote-tracking branch 'xen/upstream/xen'

Conflicts:
arch/x86/include/asm/cmpxchg.h

Merge remote-tracking branch 'tip/auto-latest'

Conflicts:
Documentation/feature-removal-schedule.txt

Merge remote-tracking branch 'spi/spi/next'

Merge remote-tracking branch 'edac-amd/for-next'

Merge remote-tracking branch 'fsnotify/for-next'

Merge remote-tracking branch 'trivial/for-next'

Merge remote-tracking branch 'selinux/master'

Merge remote-tracking branch 'security/next'

Merge remote-tracking branch 'regulator/for-next'

Merge remote-tracking branch 'fbdev/fbdev-next'

Merge commit 'refs/next/20120410/md'

Merge remote-tracking branch 'slab/for-next'

Merge remote-tracking branch 'mmc/mmc-next'

Merge branch 'quilt/device-mapper'

Merge remote-tracking branch 'cgroup/for-next'

Merge remote-tracking branch 'input/next'

Merge branch 'quilt/rr'

Conflicts:
arch/arm/kernel/kprobes.c
drivers/virtio/virtio_balloon.c
init/main.c

Merge remote-tracking branch 'sound-asoc/for-next'

Merge remote-tracking branch 'sound/for-next'

Merge remote-tracking branch 'crypto/master'

Merge remote-tracking branch 'l2-mtd/master'

Merge remote-tracking branch 'bluetooth/master'

Merge remote-tracking branch 'wireless-next/master'

Conflicts:
drivers/net/wireless/iwlwifi/iwl-testmode.c
net/wireless/nl80211.c

Merge remote-tracking branch 'net-next/master'

Merge remote-tracking branch 'dmaengine/next'

Merge remote-tracking branch 'slave-dma/next'

Merge remote-tracking branch 'isci/all'

Merge remote-tracking branch 'target-merge/for-next-merge'

Merge remote-tracking branch 'ieee1394/for-next'

Merge remote-tracking branch 'kconfig/for-next'

Merge remote-tracking branch 'kbuild/for-next'

Merge commit 'refs/next/20120410/v4l-dvb'

Merge remote-tracking branch 'hwmon-staging/hwmon-next'

Merge branch 'quilt/jdelvare-hwmon'

Merge remote-tracking branch 'bjdooks-i2c/next-i2c'

Conflicts:
drivers/i2c/busses/i2c-omap.c

Merge branch 'quilt/i2c'

Merge remote-tracking branch 'hid/for-next'

Merge remote-tracking branch 'nfs/linux-next'

Merge remote-tracking branch 'logfs/master'