Merge remote branch 'tip/auto-latest'

Conflicts:
	kernel/kgdb.c

186 files changed, 8483 insertions, 6165 deletions

diff --git a/Documentation/RCU/00-INDEX b/Documentation/RCU/00-INDEX
index 9bb62f7b89c3..0a27ea9621fa 100644
--- a/Documentation/RCU/00-INDEX
+++ b/Documentation/RCU/00-INDEX
@@ -8,14 +8,18 @@ listRCU.txt
 	- Using RCU to Protect Read-Mostly Linked Lists
 NMI-RCU.txt
 	- Using RCU to Protect Dynamic NMI Handlers
+rcubarrier.txt
+	- RCU and Unloadable Modules
+rculist_nulls.txt
+	- RCU list primitives for use with SLAB_DESTROY_BY_RCU
 rcuref.txt
 	- Reference-count design for elements of lists/arrays protected by RCU
 rcu.txt
 	- RCU Concepts
-rcubarrier.txt
-	- Unloading modules that use RCU callbacks
 RTFP.txt
 	- List of RCU papers (bibliography) going back to 1980.
+stallwarn.txt
+	- RCU CPU stall warnings (CONFIG_RCU_CPU_STALL_DETECTOR)
 torture.txt
 	- RCU Torture Test Operation (CONFIG_RCU_TORTURE_TEST)
 trace.txt
diff --git a/Documentation/RCU/RTFP.txt b/Documentation/RCU/RTFP.txt
index d2b85237c76e..5051209e6835 100644
--- a/Documentation/RCU/RTFP.txt
+++ b/Documentation/RCU/RTFP.txt
@@ -25,10 +25,10 @@ to be referencing the data structure.  However, this mechanism was not
 optimized for modern computer systems, which is not surprising given
 that these overheads were not so expensive in the mid-80s.  Nonetheless,
 passive serialization appears to be the first deferred-destruction
-mechanism to be used in production.  Furthermore, the relevant patent has
-lapsed, so this approach may be used in non-GPL software, if desired.
-(In contrast, use of RCU is permitted only in software licensed under
-GPL.  Sorry!!!)
+mechanism to be used in production.  Furthermore, the relevant patent
+has lapsed, so this approach may be used in non-GPL software, if desired.
+(In contrast, implementation of RCU is permitted only in software licensed
+under either GPL or LGPL.  Sorry!!!)
 
 In 1990, Pugh [Pugh90] noted that explicitly tracking which threads
 were reading a given data structure permitted deferred free to operate
@@ -150,6 +150,18 @@ preemptible RCU [PaulEMcKenney2007PreemptibleRCU], and the three-part
 LWN "What is RCU?" series [PaulEMcKenney2007WhatIsRCUFundamentally,
 PaulEMcKenney2008WhatIsRCUUsage, and PaulEMcKenney2008WhatIsRCUAPI].
 
+2008 saw a journal paper on real-time RCU [DinakarGuniguntala2008IBMSysJ],
+a history of how Linux changed RCU more than RCU changed Linux
+[PaulEMcKenney2008RCUOSR], and a design overview of hierarchical RCU
+[PaulEMcKenney2008HierarchicalRCU].
+
+2009 introduced user-level RCU algorithms [PaulEMcKenney2009MaliciousURCU],
+which Mathieu Desnoyers is now maintaining [MathieuDesnoyers2009URCU]
+[MathieuDesnoyersPhD].  TINY_RCU [PaulEMcKenney2009BloatWatchRCU] made
+its appearance, as did expedited RCU [PaulEMcKenney2009expeditedRCU].
+The problem of resizeable RCU-protected hash tables may now be on a path
+to a solution [JoshTriplett2009RPHash].
+
 Bibtex Entries
 
 @article{Kung80
@@ -730,6 +742,11 @@ Revised:
 "
 }
 
+#
+#	"What is RCU?" LWN series.
+#
+########################################################################
+
 @article{DinakarGuniguntala2008IBMSysJ
 ,author="D. Guniguntala and P. E. McKenney and J. Triplett and J. Walpole"
 ,title="The read-copy-update mechanism for supporting real-time applications on shared-memory multiprocessor systems with {Linux}"
@@ -820,3 +837,36 @@ Revised:
 	Uniprocessor assumptions allow simplified RCU implementation.
 "
 }
+
+@unpublished{PaulEMcKenney2009expeditedRCU
+,Author="Paul E. McKenney"
+,Title="[{PATCH} -tip 0/3] expedited 'big hammer' {RCU} grace periods"
+,month="June"
+,day="25"
+,year="2009"
+,note="Available:
+\url{http://lkml.org/lkml/2009/6/25/306}
+[Viewed August 16, 2009]"
+,annotation="
+	First posting of expedited RCU to be accepted into -tip.
+"
+}
+
+@unpublished{JoshTriplett2009RPHash
+,Author="Josh Triplett"
+,Title="Scalable concurrent hash tables via relativistic programming"
+,month="September"
+,year="2009"
+,note="Linux Plumbers Conference presentation"
+,annotation="
+	RP fun with hash tables.
+"
+}
+
+@phdthesis{MathieuDesnoyersPhD
+, title  = "Low-impact Operating System Tracing"
+, author = "Mathieu Desnoyers"
+, school = "Ecole Polytechnique de Montr\'{e}al"
+, month  = "December"
+, year   = 2009
+}
diff --git a/Documentation/RCU/checklist.txt b/Documentation/RCU/checklist.txt
index 51525a30e8b4..767cf06a4276 100644
--- a/Documentation/RCU/checklist.txt
+++ b/Documentation/RCU/checklist.txt
@@ -8,13 +8,12 @@ would cause.  This list is based on experiences reviewing such patches
 over a rather long period of time, but improvements are always welcome!
 
 0.	Is RCU being applied to a read-mostly situation?  If the data
-	structure is updated more than about 10% of the time, then
-	you should strongly consider some other approach, unless
-	detailed performance measurements show that RCU is nonetheless
-	the right tool for the job.  Yes, you might think of RCU
-	as simply cutting overhead off of the readers and imposing it
-	on the writers.  That is exactly why normal uses of RCU will
-	do much more reading than updating.
+	structure is updated more than about 10% of the time, then you
+	should strongly consider some other approach, unless detailed
+	performance measurements show that RCU is nonetheless the right
+	tool for the job.  Yes, RCU does reduce read-side overhead by
+	increasing write-side overhead, which is exactly why normal uses
+	of RCU will do much more reading than updating.
 
 	Another exception is where performance is not an issue, and RCU
 	provides a simpler implementation.  An example of this situation
@@ -35,13 +34,13 @@ over a rather long period of time, but improvements are always welcome!
 
 	If you choose #b, be prepared to describe how you have handled
 	memory barriers on weakly ordered machines (pretty much all of
-	them -- even x86 allows reads to be reordered), and be prepared
-	to explain why this added complexity is worthwhile.  If you
-	choose #c, be prepared to explain how this single task does not
-	become a major bottleneck on big multiprocessor machines (for
-	example, if the task is updating information relating to itself
-	that other tasks can read, there by definition can be no
-	bottleneck).
+	them -- even x86 allows later loads to be reordered to precede
+	earlier stores), and be prepared to explain why this added
+	complexity is worthwhile.  If you choose #c, be prepared to
+	explain how this single task does not become a major bottleneck on
+	big multiprocessor machines (for example, if the task is updating
+	information relating to itself that other tasks can read, there
+	by definition can be no bottleneck).
 
 2.	Do the RCU read-side critical sections make proper use of
 	rcu_read_lock() and friends?  These primitives are needed
@@ -51,8 +50,10 @@ over a rather long period of time, but improvements are always welcome!
 	actuarial risk of your kernel.
 
 	As a rough rule of thumb, any dereference of an RCU-protected
-	pointer must be covered by rcu_read_lock() or rcu_read_lock_bh()
-	or by the appropriate update-side lock.
+	pointer must be covered by rcu_read_lock(), rcu_read_lock_bh(),
+	rcu_read_lock_sched(), or by the appropriate update-side lock.
+	Disabling of preemption can serve as rcu_read_lock_sched(), but
+	is less readable.
 
 3.	Does the update code tolerate concurrent accesses?
 
@@ -62,25 +63,27 @@ over a rather long period of time, but improvements are always welcome!
 	of ways to handle this concurrency, depending on the situation:
 
 	a.	Use the RCU variants of the list and hlist update
-		primitives to add, remove, and replace elements on an
-		RCU-protected list.  Alternatively, use the RCU-protected
-		trees that have been added to the Linux kernel.
+		primitives to add, remove, and replace elements on
+		an RCU-protected list.	Alternatively, use the other
+		RCU-protected data structures that have been added to
+		the Linux kernel.
 
 		This is almost always the best approach.
 
 	b.	Proceed as in (a) above, but also maintain per-element
 		locks (that are acquired by both readers and writers)
 		that guard per-element state.  Of course, fields that
-		the readers refrain from accessing can be guarded by the
-		update-side lock.
+		the readers refrain from accessing can be guarded by
+		some other lock acquired only by updaters, if desired.
 
 		This works quite well, also.
 
 	c.	Make updates appear atomic to readers.  For example,
-		pointer updates to properly aligned fields will appear
-		atomic, as will individual atomic primitives.  Operations
-		performed under a lock and sequences of multiple atomic
-		primitives will -not- appear to be atomic.
+		pointer updates to properly aligned fields will
+		appear atomic, as will individual atomic primitives.
+		Sequences of perations performed under a lock will -not-
+		appear to be atomic to RCU readers, nor will sequences
+		of multiple atomic primitives.
 
 		This can work, but is starting to get a bit tricky.
 
@@ -98,9 +101,9 @@ over a rather long period of time, but improvements are always welcome!
 		a new structure containing updated values.
 
 4.	Weakly ordered CPUs pose special challenges.  Almost all CPUs
-	are weakly ordered -- even i386 CPUs allow reads to be reordered.
-	RCU code must take all of the following measures to prevent
-	memory-corruption problems:
+	are weakly ordered -- even x86 CPUs allow later loads to be
+	reordered to precede earlier stores.  RCU code must take all of
+	the following measures to prevent memory-corruption problems:
 
 	a.	Readers must maintain proper ordering of their memory
 		accesses.  The rcu_dereference() primitive ensures that
@@ -113,14 +116,21 @@ over a rather long period of time, but improvements are always welcome!
 		The rcu_dereference() primitive is also an excellent
 		documentation aid, letting the person reading the code
 		know exactly which pointers are protected by RCU.
-
-		The rcu_dereference() primitive is used by the various
-		"_rcu()" list-traversal primitives, such as the
-		list_for_each_entry_rcu().  Note that it is perfectly
-		legal (if redundant) for update-side code to use
-		rcu_dereference() and the "_rcu()" list-traversal
-		primitives.  This is particularly useful in code
-		that is common to readers and updaters.
+		Please note that compilers can also reorder code, and
+		they are becoming increasingly aggressive about doing
+		just that.  The rcu_dereference() primitive therefore
+		also prevents destructive compiler optimizations.
+
+		The rcu_dereference() primitive is used by the
+		various "_rcu()" list-traversal primitives, such
+		as the list_for_each_entry_rcu().  Note that it is
+		perfectly legal (if redundant) for update-side code to
+		use rcu_dereference() and the "_rcu()" list-traversal
+		primitives.  This is particularly useful in code that
+		is common to readers and updaters.  However, neither
+		rcu_dereference() nor the "_rcu()" list-traversal
+		primitives can substitute for a good concurrency design
+		coordinating among multiple updaters.
 
 	b.	If the list macros are being used, the list_add_tail_rcu()
 		and list_add_rcu() primitives must be used in order
@@ -135,11 +145,14 @@ over a rather long period of time, but improvements are always welcome!
 		readers.  Similarly, if the hlist macros are being used,
 		the hlist_del_rcu() primitive is required.
 
-		The list_replace_rcu() primitive may be used to
-		replace an old structure with a new one in an
-		RCU-protected list.
+		The list_replace_rcu() and hlist_replace_rcu() primitives
+		may be used to replace an old structure with a new one
+		in their respective types of RCU-protected lists.
+
+	d.	Rules similar to (4b) and (4c) apply to the "hlist_nulls"
+		type of RCU-protected linked lists.
 
-	d.	Updates must ensure that initialization of a given
+	e.	Updates must ensure that initialization of a given
 		structure happens before pointers to that structure are
 		publicized.  Use the rcu_assign_pointer() primitive
 		when publicizing a pointer to a structure that can
@@ -151,16 +164,31 @@ over a rather long period of time, but improvements are always welcome!
 	it cannot block.
 
 6.	Since synchronize_rcu() can block, it cannot be called from
-	any sort of irq context.  Ditto for synchronize_sched() and
-	synchronize_srcu().
-
-7.	If the updater uses call_rcu(), then the corresponding readers
-	must use rcu_read_lock() and rcu_read_unlock().  If the updater
-	uses call_rcu_bh(), then the corresponding readers must use
-	rcu_read_lock_bh() and rcu_read_unlock_bh().  If the updater
-	uses call_rcu_sched(), then the corresponding readers must
-	disable preemption.  Mixing things up will result in confusion
-	and broken kernels.
+	any sort of irq context.  The same rule applies for
+	synchronize_rcu_bh(), synchronize_sched(), synchronize_srcu(),
+	synchronize_rcu_expedited(), synchronize_rcu_bh_expedited(),
+	synchronize_sched_expedite(), and synchronize_srcu_expedited().
+
+	The expedited forms of these primitives have the same semantics
+	as the non-expedited forms, but expediting is both expensive
+	and unfriendly to real-time workloads.	Use of the expedited
+	primitives should be restricted to rare configuration-change
+	operations that would not normally be undertaken while a real-time
+	workload is running.
+
+7.	If the updater uses call_rcu() or synchronize_rcu(), then the
+	corresponding readers must use rcu_read_lock() and
+	rcu_read_unlock().  If the updater uses call_rcu_bh() or
+	synchronize_rcu_bh(), then the corresponding readers must
+	use rcu_read_lock_bh() and rcu_read_unlock_bh().  If the
+	updater uses call_rcu_sched() or synchronize_sched(), then
+	the corresponding readers must disable preemption, possibly
+	by calling rcu_read_lock_sched() and rcu_read_unlock_sched().
+	If the updater uses synchronize_srcu(), the the corresponding
+	readers must use srcu_read_lock() and srcu_read_unlock(),
+	and with the same srcu_struct.	The rules for the expedited
+	primitives are the same as for their non-expedited counterparts.
+	Mixing things up will result in confusion and broken kernels.
 
 	One exception to this rule: rcu_read_lock() and rcu_read_unlock()
 	may be substituted for rcu_read_lock_bh() and rcu_read_unlock_bh()
@@ -212,6 +240,8 @@ over a rather long period of time, but improvements are always welcome!
 	e.	Periodically invoke synchronize_rcu(), permitting a limited
 		number of updates per grace period.
 
+	The same cautions apply to call_rcu_bh() and call_rcu_sched().
+
 9.	All RCU list-traversal primitives, which include
 	rcu_dereference(), list_for_each_entry_rcu(),
 	list_for_each_continue_rcu(), and list_for_each_safe_rcu(),
@@ -229,7 +259,8 @@ over a rather long period of time, but improvements are always welcome!
 10.	Conversely, if you are in an RCU read-side critical section,
 	and you don't hold the appropriate update-side lock, you -must-
 	use the "_rcu()" variants of the list macros.  Failing to do so
-	will break Alpha and confuse people reading your code.
+	will break Alpha, cause aggressive compilers to generate bad code,
+	and confuse people trying to read your code.
 
 11.	Note that synchronize_rcu() -only- guarantees to wait until
 	all currently executing rcu_read_lock()-protected RCU read-side
@@ -239,15 +270,21 @@ over a rather long period of time, but improvements are always welcome!
 	rcu_read_lock()-protected read-side critical sections, do -not-
 	use synchronize_rcu().
 
-	If you want to wait for some of these other things, you might
-	instead need to use synchronize_irq() or synchronize_sched().
+	Similarly, disabling preemption is not an acceptable substitute
+	for rcu_read_lock().  Code that attempts to use preemption
+	disabling where it should be using rcu_read_lock() will break
+	in real-time kernel builds.
+
+	If you want to wait for interrupt handlers, NMI handlers, and
+	code under the influence of preempt_disable(), you instead
+	need to use synchronize_irq() or synchronize_sched().
 
 12.	Any lock acquired by an RCU callback must be acquired elsewhere
 	with softirq disabled, e.g., via spin_lock_irqsave(),
 	spin_lock_bh(), etc.  Failing to disable irq on a given
-	acquisition of that lock will result in deadlock as soon as the
-	RCU callback happens to interrupt that acquisition's critical
-	section.
+	acquisition of that lock will result in deadlock as soon as
+	the RCU softirq handler happens to run your RCU callback while
+	interrupting that acquisition's critical section.
 
 13.	RCU callbacks can be and are executed in parallel.  In many cases,
 	the callback code simply wrappers around kfree(), so that this
@@ -265,29 +302,30 @@ over a rather long period of time, but improvements are always welcome!
 	not the case, a self-spawning RCU callback would prevent the
 	victim CPU from ever going offline.)
 
-14.	SRCU (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu())
-	may only be invoked from process context.  Unlike other forms of
-	RCU, it -is- permissible to block in an SRCU read-side critical
-	section (demarked by srcu_read_lock() and srcu_read_unlock()),
-	hence the "SRCU": "sleepable RCU".  Please note that if you
-	don't need to sleep in read-side critical sections, you should
-	be using RCU rather than SRCU, because RCU is almost always
-	faster and easier to use than is SRCU.
+14.	SRCU (srcu_read_lock(), srcu_read_unlock(), synchronize_srcu(),
+	and synchronize_srcu_expedited()) may only be invoked from
+	process context.  Unlike other forms of RCU, it -is- permissible
+	to block in an SRCU read-side critical section (demarked by
+	srcu_read_lock() and srcu_read_unlock()), hence the "SRCU":
+	"sleepable RCU".  Please note that if you don't need to sleep
+	in read-side critical sections, you should be using RCU rather
+	than SRCU, because RCU is almost always faster and easier to
+	use than is SRCU.
 
 	Also unlike other forms of RCU, explicit initialization
 	and cleanup is required via init_srcu_struct() and
 	cleanup_srcu_struct().	These are passed a "struct srcu_struct"
 	that defines the scope of a given SRCU domain.	Once initialized,
 	the srcu_struct is passed to srcu_read_lock(), srcu_read_unlock()
-	and synchronize_srcu().  A given synchronize_srcu() waits only
-	for SRCU read-side critical sections governed by srcu_read_lock()
-	and srcu_read_unlock() calls that have been passd the same
-	srcu_struct.  This property is what makes sleeping read-side
-	critical sections tolerable -- a given subsystem delays only
-	its own updates, not those of other subsystems using SRCU.
-	Therefore, SRCU is less prone to OOM the system than RCU would
-	be if RCU's read-side critical sections were permitted to
-	sleep.
+	synchronize_srcu(), and synchronize_srcu_expedited().  A given
+	synchronize_srcu() waits only for SRCU read-side critical
+	sections governed by srcu_read_lock() and srcu_read_unlock()
+	calls that have been passed the same srcu_struct.  This property
+	is what makes sleeping read-side critical sections tolerable --
+	a given subsystem delays only its own updates, not those of other
+	subsystems using SRCU.	Therefore, SRCU is less prone to OOM the
+	system than RCU would be if RCU's read-side critical sections
+	were permitted to sleep.
 
 	The ability to sleep in read-side critical sections does not
 	come for free.	First, corresponding srcu_read_lock() and
@@ -311,12 +349,12 @@ over a rather long period of time, but improvements are always welcome!
 	destructive operation, and -only- -then- invoke call_rcu(),
 	synchronize_rcu(), or friends.
 
-	Because these primitives only wait for pre-existing readers,
-	it is the caller's responsibility to guarantee safety to
-	any subsequent readers.
+	Because these primitives only wait for pre-existing readers, it
+	is the caller's responsibility to guarantee that any subsequent
+	readers will execute safely.
 
-16.	The various RCU read-side primitives do -not- contain memory
-	barriers.  The CPU (and in some cases, the compiler) is free
-	to reorder code into and out of RCU read-side critical sections.
-	It is the responsibility of the RCU update-side primitives to
-	deal with this.
+16.	The various RCU read-side primitives do -not- necessarily contain
+	memory barriers.  You should therefore plan for the CPU
+	and the compiler to freely reorder code into and out of RCU
+	read-side critical sections.  It is the responsibility of the
+	RCU update-side primitives to deal with this.
diff --git a/Documentation/RCU/rcu.txt b/Documentation/RCU/rcu.txt
index 2a23523ce471..31852705b586 100644
--- a/Documentation/RCU/rcu.txt
+++ b/Documentation/RCU/rcu.txt
@@ -75,6 +75,8 @@ o	I hear that RCU is patented?  What is with that?
 	search for the string "Patent" in RTFP.txt to find them.
 	Of these, one was allowed to lapse by the assignee, and the
 	others have been contributed to the Linux kernel under GPL.
+	There are now also LGPL implementations of user-level RCU
+	available (http://lttng.org/?q=node/18).
 
 o	I hear that RCU needs work in order to support realtime kernels?
 
@@ -91,48 +93,4 @@ o	Where can I find more information on RCU?
 
 o	What are all these files in this directory?
 
-
-	NMI-RCU.txt
-
-		Describes how to use RCU to implement dynamic
-		NMI handlers, which can be revectored on the fly,
-		without rebooting.
-
-	RTFP.txt
-
-		List of RCU-related publications and web sites.
-
-	UP.txt
-
-		Discussion of RCU usage in UP kernels.
-
-	arrayRCU.txt
-
-		Describes how to use RCU to protect arrays, with
-		resizeable arrays whose elements reference other
-		data structures being of the most interest.
-
-	checklist.txt
-
-		Lists things to check for when inspecting code that
-		uses RCU.
-
-	listRCU.txt
-
-		Describes how to use RCU to protect linked lists.
-		This is the simplest and most common use of RCU
-		in the Linux kernel.
-
-	rcu.txt
-
-		You are reading it!
-
-	rcuref.txt
-
-		Describes how to combine use of reference counts
-		with RCU.
-
-	whatisRCU.txt
-
-		Overview of how the RCU implementation works.  Along
-		the way, presents a conceptual view of RCU.
+	See 00-INDEX for the list.
diff --git a/Documentation/RCU/stallwarn.txt b/Documentation/RCU/stallwarn.txt
new file mode 100644
index 000000000000..1423d2570d78
--- /dev/null
+++ b/Documentation/RCU/stallwarn.txt
@@ -0,0 +1,58 @@
+Using RCU's CPU Stall Detector
+
+The CONFIG_RCU_CPU_STALL_DETECTOR kernel config parameter enables
+RCU's CPU stall detector, which detects conditions that unduly delay
+RCU grace periods.  The stall detector's idea of what constitutes
+"unduly delayed" is controlled by a pair of C preprocessor macros:
+
+RCU_SECONDS_TILL_STALL_CHECK
+
+	This macro defines the period of time that RCU will wait from
+	the beginning of a grace period until it issues an RCU CPU
+	stall warning.	It is normally ten seconds.
+
+RCU_SECONDS_TILL_STALL_RECHECK
+
+	This macro defines the period of time that RCU will wait after
+	issuing a stall warning until it issues another stall warning.
+	It is normally set to thirty seconds.
+
+RCU_STALL_RAT_DELAY
+
+	The CPU stall detector tries to make the offending CPU rat on itself,
+	as this often gives better-quality stack traces.  However, if
+	the offending CPU does not detect its own stall in the number
+	of jiffies specified by RCU_STALL_RAT_DELAY, then other CPUs will
+	complain.  This is normally set to two jiffies.
+
+The following problems can result in an RCU CPU stall warning:
+
+o	A CPU looping in an RCU read-side critical section.
+	
+o	A CPU looping with interrupts disabled.
+
+o	A CPU looping with preemption disabled.
+
+o	For !CONFIG_PREEMPT kernels, a CPU looping anywhere in the kernel
+	without invoking schedule().
+
+o	A bug in the RCU implementation.
+
+o	A hardware failure.  This is quite unlikely, but has occurred
+	at least once in a former life.  A CPU failed in a running system,
+	becoming unresponsive, but not causing an immediate crash.
+	This resulted in a series of RCU CPU stall warnings, eventually
+	leading the realization that the CPU had failed.
+
+The RCU, RCU-sched, and RCU-bh implementations have CPU stall warning.
+SRCU does not do so directly, but its calls to synchronize_sched() will
+result in RCU-sched detecting any CPU stalls that might be occurring.
+
+To diagnose the cause of the stall, inspect the stack traces.  The offending
+function will usually be near the top of the stack.  If you have a series
+of stall warnings from a single extended stall, comparing the stack traces
+can often help determine where the stall is occurring, which will usually
+be in the function nearest the top of the stack that stays the same from
+trace to trace.
+
+RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE.
diff --git a/Documentation/RCU/torture.txt b/Documentation/RCU/torture.txt
index 9dba3bb90e60..0e50bc2aa1e2 100644
--- a/Documentation/RCU/torture.txt
+++ b/Documentation/RCU/torture.txt
@@ -30,6 +30,18 @@ MODULE PARAMETERS
 
 This module has the following parameters:
 
+fqs_duration	Duration (in microseconds) of artificially induced bursts
+		of force_quiescent_state() invocations.  In RCU
+		implementations having force_quiescent_state(), these
+		bursts help force races between forcing a given grace
+		period and that grace period ending on its own.
+
+fqs_holdoff	Holdoff time (in microseconds) between consecutive calls
+		to force_quiescent_state() within a burst.
+
+fqs_stutter	Wait time (in seconds) between consecutive bursts
+		of calls to force_quiescent_state().
+
 irqreaders	Says to invoke RCU readers from irq level.  This is currently
 		done via timers.  Defaults to "1" for variants of RCU that
 		permit this.  (Or, more accurately, variants of RCU that do
diff --git a/Documentation/RCU/whatisRCU.txt b/Documentation/RCU/whatisRCU.txt
index d542ca243b80..469a58b2e67e 100644
--- a/Documentation/RCU/whatisRCU.txt
+++ b/Documentation/RCU/whatisRCU.txt
@@ -327,7 +327,8 @@ a.	synchronize_rcu()	rcu_read_lock() / rcu_read_unlock()
 
 b.	call_rcu_bh()		rcu_read_lock_bh() / rcu_read_unlock_bh()
 
-c.	synchronize_sched()	preempt_disable() / preempt_enable()
+c.	synchronize_sched()	rcu_read_lock_sched() / rcu_read_unlock_sched()
+				preempt_disable() / preempt_enable()
 				local_irq_save() / local_irq_restore()
 				hardirq enter / hardirq exit
 				NMI enter / NMI exit
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index a71a1c66d294..7680f92245be 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -6,21 +6,6 @@ be removed from this file.
 
 ---------------------------
 
-What:	USER_SCHED
-When:	2.6.34
-
-Why:	USER_SCHED was implemented as a proof of concept for group scheduling.
-	The effect of USER_SCHED can already be achieved from userspace with
-	the help of libcgroup. The removal of USER_SCHED will also simplify
-	the scheduler code with the removal of one major ifdef. There are also
-	issues USER_SCHED has with USER_NS. A decision was taken not to fix
-	those and instead remove USER_SCHED. Also new group scheduling
-	features will not be implemented for USER_SCHED.
-
-Who:	Dhaval Giani <dhaval@linux.vnet.ibm.com>
-
----------------------------
-
 What:	PRISM54
 When:	2.6.34
 
diff --git a/Documentation/filesystems/dentry-locking.txt b/Documentation/filesystems/dentry-locking.txt
index 4c0c575a4012..79334ed5daa7 100644
--- a/Documentation/filesystems/dentry-locking.txt
+++ b/Documentation/filesystems/dentry-locking.txt
@@ -62,7 +62,8 @@ changes are :
 2. Insertion of a dentry into the hash table is done using
    hlist_add_head_rcu() which take care of ordering the writes - the
    writes to the dentry must be visible before the dentry is
-   inserted. This works in conjunction with hlist_for_each_rcu() while
+   inserted. This works in conjunction with hlist_for_each_rcu(),
+   which has since been replaced by hlist_for_each_entry_rcu(), while
    walking the hash chain. The only requirement is that all
    initialization to the dentry must be done before
    hlist_add_head_rcu() since we don't have dcache_lock protection
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 98260d83d71d..28866f3013ad 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1748,6 +1748,9 @@ and is between 256 and 4096 characters. It is defined in the file
 	nomfgpt		[X86-32] Disable Multi-Function General Purpose
 			Timer usage (for AMD Geode machines).
 
+	nopat		[X86] Disable PAT (page attribute table extension of
+			pagetables) support.
+
 	norandmaps	Don't use address space randomization.  Equivalent to
 			echo 0 > /proc/sys/kernel/randomize_va_space
 
diff --git a/Documentation/trace/kprobetrace.txt b/Documentation/trace/kprobetrace.txt
index 47aabeebbdf6..a9100b28eb84 100644
--- a/Documentation/trace/kprobetrace.txt
+++ b/Documentation/trace/kprobetrace.txt
@@ -24,6 +24,7 @@ Synopsis of kprobe_events
 -------------------------
   p[:[GRP/]EVENT] SYMBOL[+offs]|MEMADDR [FETCHARGS]	: Set a probe
   r[:[GRP/]EVENT] SYMBOL[+0] [FETCHARGS]		: Set a return probe
+  -:[GRP/]EVENT						: Clear a probe
 
  GRP		: Group name. If omitted, use "kprobes" for it.
  EVENT		: Event name. If omitted, the event name is generated
@@ -37,15 +38,12 @@ Synopsis of kprobe_events
   @SYM[+|-offs]	: Fetch memory at SYM +|- offs (SYM should be a data symbol)
   $stackN	: Fetch Nth entry of stack (N >= 0)
   $stack	: Fetch stack address.
-  $argN		: Fetch function argument. (N >= 0)(*)
-  $retval	: Fetch return value.(**)
-  +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(***)
+  $retval	: Fetch return value.(*)
+  +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(**)
   NAME=FETCHARG: Set NAME as the argument name of FETCHARG.
 
-  (*) aN may not correct on asmlinkaged functions and at the middle of
-      function body.
-  (**) only for return probe.
-  (***) this is useful for fetching a field of data structures.
+  (*) only for return probe.
+  (**) this is useful for fetching a field of data structures.
 
 
 Per-Probe Event Filtering
@@ -82,13 +80,16 @@ Usage examples
 To add a probe as a new event, write a new definition to kprobe_events
 as below.
 
-  echo p:myprobe do_sys_open dfd=$arg0 filename=$arg1 flags=$arg2 mode=$arg3 > /sys/kernel/debug/tracing/kprobe_events
+  echo 'p:myprobe do_sys_open dfd=%ax filename=%dx flags=%cx mode=+4($stack)' > /sys/kernel/debug/tracing/kprobe_events
 
  This sets a kprobe on the top of do_sys_open() function with recording
-1st to 4th arguments as "myprobe" event. As this example shows, users can
-choose more familiar names for each arguments.
+1st to 4th arguments as "myprobe" event. Note, which register/stack entry is
+assigned to each function argument depends on arch-specific ABI. If you unsure
+the ABI, please try to use probe subcommand of perf-tools (you can find it
+under tools/perf/).
+As this example shows, users can choose more familiar names for each arguments.
 
-  echo r:myretprobe do_sys_open $retval >> /sys/kernel/debug/tracing/kprobe_events
+  echo 'r:myretprobe do_sys_open $retval' >> /sys/kernel/debug/tracing/kprobe_events
 
  This sets a kretprobe on the return point of do_sys_open() function with
 recording return value as "myretprobe" event.
@@ -97,23 +98,24 @@ recording return value as "myretprobe" event.
 
   cat /sys/kernel/debug/tracing/events/kprobes/myprobe/format
 name: myprobe
-ID: 75
+ID: 780
 format:
-	field:unsigned short common_type;	offset:0;	size:2;
-	field:unsigned char common_flags;	offset:2;	size:1;
-	field:unsigned char common_preempt_count;	offset:3;	size:1;
-	field:int common_pid;	offset:4;	size:4;
-	field:int common_tgid;	offset:8;	size:4;
+        field:unsigned short common_type;       offset:0;       size:2; signed:0;
+        field:unsigned char common_flags;       offset:2;       size:1; signed:0;
+        field:unsigned char common_preempt_count;       offset:3; size:1;signed:0;
+        field:int common_pid;   offset:4;       size:4; signed:1;
+        field:int common_lock_depth;    offset:8;       size:4; signed:1;
 
-	field: unsigned long ip;	offset:16;tsize:8;
-	field: int nargs;	offset:24;tsize:4;
-	field: unsigned long dfd;	offset:32;tsize:8;
-	field: unsigned long filename;	offset:40;tsize:8;
-	field: unsigned long flags;	offset:48;tsize:8;
-	field: unsigned long mode;	offset:56;tsize:8;
+        field:unsigned long __probe_ip; offset:12;      size:4; signed:0;
+        field:int __probe_nargs;        offset:16;      size:4; signed:1;
+        field:unsigned long dfd;        offset:20;      size:4; signed:0;
+        field:unsigned long filename;   offset:24;      size:4; signed:0;
+        field:unsigned long flags;      offset:28;      size:4; signed:0;
+        field:unsigned long mode;       offset:32;      size:4; signed:0;
 
-print fmt: "(%lx) dfd=%lx filename=%lx flags=%lx mode=%lx", REC->ip, REC->dfd, REC->filename, REC->flags, REC->mode
 
+print fmt: "(%lx) dfd=%lx filename=%lx flags=%lx mode=%lx", REC->__probe_ip,
+REC->dfd, REC->filename, REC->flags, REC->mode
 
  You can see that the event has 4 arguments as in the expressions you specified.
 
@@ -121,6 +123,12 @@ print fmt: "(%lx) dfd=%lx filename=%lx flags=%lx mode=%lx", REC->ip, REC->dfd, R
 
  This clears all probe points.
 
+ Or,
+
+  echo -:myprobe >> kprobe_events
+
+ This clears probe points selectively.
+
  Right after definition, each event is disabled by default. For tracing these
 events, you need to enable it.
 
@@ -146,4 +154,3 @@ events, you need to enable it.
 returns from SYMBOL(e.g. "sys_open+0x1b/0x1d <- do_sys_open" means kernel
 returns from do_sys_open to sys_open+0x1b).
 
-
diff --git a/MAINTAINERS b/MAINTAINERS
index 865afa09ef3b..3793ad5852fc 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4565,7 +4565,7 @@ F:	drivers/net/wireless/ray*
 RCUTORTURE MODULE
 M:	Josh Triplett <josh@freedesktop.org>
 M:	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
-S:	Maintained
+S:	Supported
 F:	Documentation/RCU/torture.txt
 F:	kernel/rcutorture.c
 
@@ -4590,11 +4590,12 @@ M:	Dipankar Sarma <dipankar@in.ibm.com>
 M:	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
 W:	http://www.rdrop.com/users/paulmck/rclock/
 S:	Supported
-F:	Documentation/RCU/rcu.txt
-F:	Documentation/RCU/rcuref.txt
-F:	include/linux/rcupdate.h
-F:	include/linux/srcu.h
-F:	kernel/rcupdate.c
+F:	Documentation/RCU/
+F:	include/linux/rcu*
+F:	include/linux/srcu*
+F:	kernel/rcu*
+F:	kernel/srcu*
+X:	kernel/rcutorture.c
 
 REAL TIME CLOCK DRIVER
 M:	Paul Gortmaker <p_gortmaker@yahoo.com>
diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c
index 3c5b7de10af5..a64e17813283 100644
--- a/arch/mips/mm/init.c
+++ b/arch/mips/mm/init.c
@@ -298,7 +298,7 @@ void __init fixrange_init(unsigned long start, unsigned long end,
 }
 
 #ifndef CONFIG_NEED_MULTIPLE_NODES
-static int __init page_is_ram(unsigned long pagenr)
+int page_is_ram(unsigned long pagenr)
 {
 	int i;
 
diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c
index a3c11cac3d71..95ad9dad298e 100644
--- a/arch/powerpc/kernel/perf_callchain.c
+++ b/arch/powerpc/kernel/perf_callchain.c
@@ -495,9 +495,6 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
 
 	entry->nr = 0;
 
-	if (current->pid == 0)		/* idle task? */
-		return entry;
-
 	if (!user_mode(regs)) {
 		perf_callchain_kernel(regs, entry);
 		if (current->mm)
diff --git a/arch/score/mm/init.c b/arch/score/mm/init.c
index dfaf458d6702..7f001bbedb00 100644
--- a/arch/score/mm/init.c
+++ b/arch/score/mm/init.c
@@ -59,7 +59,7 @@ static unsigned long setup_zero_page(void)
 }
 
 #ifndef CONFIG_NEED_MULTIPLE_NODES
-static int __init page_is_ram(unsigned long pagenr)
+int page_is_ram(unsigned long pagenr)
 {
 	if (pagenr >= min_low_pfn && pagenr < max_low_pfn)
 		return 1;
diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c
index 24ea837eac5b..a9dd3abde28e 100644
--- a/arch/sh/kernel/perf_callchain.c
+++ b/arch/sh/kernel/perf_callchain.c
@@ -68,9 +68,6 @@ perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
 
 	is_user = user_mode(regs);
 
-	if (!current || current->pid == 0)
-		return;
-
 	if (is_user && current->state != TASK_RUNNING)
 		return;
 
diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile
index 2201e9c20e4a..c1ea9eb04466 100644
--- a/arch/um/sys-x86_64/Makefile
+++ b/arch/um/sys-x86_64/Makefile
@@ -8,7 +8,8 @@ obj-y = bug.o bugs.o delay.o fault.o ldt.o mem.o ptrace.o ptrace_user.o \
 	setjmp.o signal.o stub.o stub_segv.o syscalls.o syscall_table.o \
 	sysrq.o ksyms.o tls.o
 
-subarch-obj-y = lib/csum-partial_64.o lib/memcpy_64.o lib/thunk_64.o
+subarch-obj-y = lib/csum-partial_64.o lib/memcpy_64.o lib/thunk_64.o \
+		lib/rwsem_64.o
 subarch-obj-$(CONFIG_MODULES) += kernel/module.o
 
 ldt-y = ../sys-i386/ldt.o
diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu
index f20ddf84a893..a19829374e6a 100644
--- a/arch/x86/Kconfig.cpu
+++ b/arch/x86/Kconfig.cpu
@@ -319,7 +319,7 @@ config X86_L1_CACHE_SHIFT
 
 config X86_XADD
 	def_bool y
-	depends on X86_32 && !M386
+	depends on X86_64 || !M386
 
 config X86_PPRO_FENCE
 	bool "PentiumPro memory ordering errata workaround"
diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c
index f9f472462753..9046e4af66ce 100644
--- a/arch/x86/ia32/ia32_aout.c
+++ b/arch/x86/ia32/ia32_aout.c
@@ -297,7 +297,7 @@ static int load_aout_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 	 * size limits imposed on them by creating programs with large
 	 * arrays in the data or bss.
 	 */
-	rlim = current->signal->rlim[RLIMIT_DATA].rlim_cur;
+	rlim = rlimit(RLIMIT_DATA);
 	if (rlim >= RLIM_INFINITY)
 		rlim = ~0;
 	if (ex.a_data + ex.a_bss > rlim)
diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h
index 69b74a7b877f..3b5b828767b6 100644
--- a/arch/x86/include/asm/alternative.h
+++ b/arch/x86/include/asm/alternative.h
@@ -125,11 +125,16 @@ static inline void alternatives_smp_switch(int smp) {}
 	asm volatile (ALTERNATIVE(oldinstr, newinstr, feature)		\
 		: output : "i" (0), ## input)
 
+/* Like alternative_io, but for replacing a direct call with another one. */
+#define alternative_call(oldfunc, newfunc, feature, output, input...)	\
+	asm volatile (ALTERNATIVE("call %P[old]", "call %P[new]", feature) \
+		: output : [old] "i" (oldfunc), [new] "i" (newfunc), ## input)
+
 /*
  * use this macro(s) if you need more than one output parameter
  * in alternative_io
  */
-#define ASM_OUTPUT2(a, b) a, b
+#define ASM_OUTPUT2(a...) a
 
 struct paravirt_patch_site;
 #ifdef CONFIG_PARAVIRT
diff --git a/arch/x86/include/asm/atomic.h b/arch/x86/include/asm/atomic.h
index 4e1b8873c474..8baaa719fa7f 100644
--- a/arch/x86/include/asm/atomic.h
+++ b/arch/x86/include/asm/atomic.h
@@ -1,5 +1,300 @@
+#ifndef _ASM_X86_ATOMIC_H
+#define _ASM_X86_ATOMIC_H
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+#include <asm/processor.h>
+#include <asm/alternative.h>
+#include <asm/cmpxchg.h>
+
+/*
+ * Atomic operations that C can't guarantee us.  Useful for
+ * resource counting etc..
+ */
+
+#define ATOMIC_INIT(i)	{ (i) }
+
+/**
+ * atomic_read - read atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically reads the value of @v.
+ */
+static inline int atomic_read(const atomic_t *v)
+{
+	return v->counter;
+}
+
+/**
+ * atomic_set - set atomic variable
+ * @v: pointer of type atomic_t
+ * @i: required value
+ *
+ * Atomically sets the value of @v to @i.
+ */
+static inline void atomic_set(atomic_t *v, int i)
+{
+	v->counter = i;
+}
+
+/**
+ * atomic_add - add integer to atomic variable
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v.
+ */
+static inline void atomic_add(int i, atomic_t *v)
+{
+	asm volatile(LOCK_PREFIX "addl %1,%0"
+		     : "+m" (v->counter)
+		     : "ir" (i));
+}
+
+/**
+ * atomic_sub - subtract integer from atomic variable
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically subtracts @i from @v.
+ */
+static inline void atomic_sub(int i, atomic_t *v)
+{
+	asm volatile(LOCK_PREFIX "subl %1,%0"
+		     : "+m" (v->counter)
+		     : "ir" (i));
+}
+
+/**
+ * atomic_sub_and_test - subtract value from variable and test result
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically subtracts @i from @v and returns
+ * true if the result is zero, or false for all
+ * other cases.
+ */
+static inline int atomic_sub_and_test(int i, atomic_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "subl %2,%0; sete %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : "ir" (i) : "memory");
+	return c;
+}
+
+/**
+ * atomic_inc - increment atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically increments @v by 1.
+ */
+static inline void atomic_inc(atomic_t *v)
+{
+	asm volatile(LOCK_PREFIX "incl %0"
+		     : "+m" (v->counter));
+}
+
+/**
+ * atomic_dec - decrement atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically decrements @v by 1.
+ */
+static inline void atomic_dec(atomic_t *v)
+{
+	asm volatile(LOCK_PREFIX "decl %0"
+		     : "+m" (v->counter));
+}
+
+/**
+ * atomic_dec_and_test - decrement and test
+ * @v: pointer of type atomic_t
+ *
+ * Atomically decrements @v by 1 and
+ * returns true if the result is 0, or false for all other
+ * cases.
+ */
+static inline int atomic_dec_and_test(atomic_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "decl %0; sete %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : : "memory");
+	return c != 0;
+}
+
+/**
+ * atomic_inc_and_test - increment and test
+ * @v: pointer of type atomic_t
+ *
+ * Atomically increments @v by 1
+ * and returns true if the result is zero, or false for all
+ * other cases.
+ */
+static inline int atomic_inc_and_test(atomic_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "incl %0; sete %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : : "memory");
+	return c != 0;
+}
+
+/**
+ * atomic_add_negative - add and test if negative
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v and returns true
+ * if the result is negative, or false when
+ * result is greater than or equal to zero.
+ */
+static inline int atomic_add_negative(int i, atomic_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "addl %2,%0; sets %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : "ir" (i) : "memory");
+	return c;
+}
+
+/**
+ * atomic_add_return - add integer and return
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v and returns @i + @v
+ */
+static inline int atomic_add_return(int i, atomic_t *v)
+{
+	int __i;
+#ifdef CONFIG_M386
+	unsigned long flags;
+	if (unlikely(boot_cpu_data.x86 <= 3))
+		goto no_xadd;
+#endif
+	/* Modern 486+ processor */
+	__i = i;
+	asm volatile(LOCK_PREFIX "xaddl %0, %1"
+		     : "+r" (i), "+m" (v->counter)
+		     : : "memory");
+	return i + __i;
+
+#ifdef CONFIG_M386
+no_xadd: /* Legacy 386 processor */
+	local_irq_save(flags);
+	__i = atomic_read(v);
+	atomic_set(v, i + __i);
+	local_irq_restore(flags);
+	return i + __i;
+#endif
+}
+
+/**
+ * atomic_sub_return - subtract integer and return
+ * @v: pointer of type atomic_t
+ * @i: integer value to subtract
+ *
+ * Atomically subtracts @i from @v and returns @v - @i
+ */
+static inline int atomic_sub_return(int i, atomic_t *v)
+{
+	return atomic_add_return(-i, v);
+}
+
+#define atomic_inc_return(v)  (atomic_add_return(1, v))
+#define atomic_dec_return(v)  (atomic_sub_return(1, v))
+
+static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+{
+	return cmpxchg(&v->counter, old, new);
+}
+
+static inline int atomic_xchg(atomic_t *v, int new)
+{
+	return xchg(&v->counter, new);
+}
+
+/**
+ * atomic_add_unless - add unless the number is already a given value
+ * @v: pointer of type atomic_t
+ * @a: the amount to add to v...
+ * @u: ...unless v is equal to u.
+ *
+ * Atomically adds @a to @v, so long as @v was not already @u.
+ * Returns non-zero if @v was not @u, and zero otherwise.
+ */
+static inline int atomic_add_unless(atomic_t *v, int a, int u)
+{
+	int c, old;
+	c = atomic_read(v);
+	for (;;) {
+		if (unlikely(c == (u)))
+			break;
+		old = atomic_cmpxchg((v), c, c + (a));
+		if (likely(old == c))
+			break;
+		c = old;
+	}
+	return c != (u);
+}
+
+#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
+
+/**
+ * atomic_inc_short - increment of a short integer
+ * @v: pointer to type int
+ *
+ * Atomically adds 1 to @v
+ * Returns the new value of @u
+ */
+static inline short int atomic_inc_short(short int *v)
+{
+	asm(LOCK_PREFIX "addw $1, %0" : "+m" (*v));
+	return *v;
+}
+
+#ifdef CONFIG_X86_64
+/**
+ * atomic_or_long - OR of two long integers
+ * @v1: pointer to type unsigned long
+ * @v2: pointer to type unsigned long
+ *
+ * Atomically ORs @v1 and @v2
+ * Returns the result of the OR
+ */
+static inline void atomic_or_long(unsigned long *v1, unsigned long v2)
+{
+	asm(LOCK_PREFIX "orq %1, %0" : "+m" (*v1) : "r" (v2));
+}
+#endif
+
+/* These are x86-specific, used by some header files */
+#define atomic_clear_mask(mask, addr)				\
+	asm volatile(LOCK_PREFIX "andl %0,%1"			\
+		     : : "r" (~(mask)), "m" (*(addr)) : "memory")
+
+#define atomic_set_mask(mask, addr)				\
+	asm volatile(LOCK_PREFIX "orl %0,%1"			\
+		     : : "r" ((unsigned)(mask)), "m" (*(addr))	\
+		     : "memory")
+
+/* Atomic operations are already serializing on x86 */
+#define smp_mb__before_atomic_dec()	barrier()
+#define smp_mb__after_atomic_dec()	barrier()
+#define smp_mb__before_atomic_inc()	barrier()
+#define smp_mb__after_atomic_inc()	barrier()
+
 #ifdef CONFIG_X86_32
-# include "atomic_32.h"
+# include "atomic64_32.h"
 #else
-# include "atomic_64.h"
+# include "atomic64_64.h"
 #endif
+
+#include <asm-generic/atomic-long.h>
+#endif /* _ASM_X86_ATOMIC_H */
diff --git a/arch/x86/include/asm/atomic64_32.h b/arch/x86/include/asm/atomic64_32.h
new file mode 100644
index 000000000000..03027bf28de5
--- /dev/null
+++ b/arch/x86/include/asm/atomic64_32.h
@@ -0,0 +1,160 @@
+#ifndef _ASM_X86_ATOMIC64_32_H
+#define _ASM_X86_ATOMIC64_32_H
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+#include <asm/processor.h>
+//#include <asm/cmpxchg.h>
+
+/* An 64bit atomic type */
+
+typedef struct {
+	u64 __aligned(8) counter;
+} atomic64_t;
+
+#define ATOMIC64_INIT(val)	{ (val) }
+
+extern u64 atomic64_cmpxchg(atomic64_t *ptr, u64 old_val, u64 new_val);
+
+/**
+ * atomic64_xchg - xchg atomic64 variable
+ * @ptr:      pointer to type atomic64_t
+ * @new_val:  value to assign
+ *
+ * Atomically xchgs the value of @ptr to @new_val and returns
+ * the old value.
+ */
+extern u64 atomic64_xchg(atomic64_t *ptr, u64 new_val);
+
+/**
+ * atomic64_set - set atomic64 variable
+ * @ptr:      pointer to type atomic64_t
+ * @new_val:  value to assign
+ *
+ * Atomically sets the value of @ptr to @new_val.
+ */
+extern void atomic64_set(atomic64_t *ptr, u64 new_val);
+
+/**
+ * atomic64_read - read atomic64 variable
+ * @ptr:      pointer to type atomic64_t
+ *
+ * Atomically reads the value of @ptr and returns it.
+ */
+static inline u64 atomic64_read(atomic64_t *ptr)
+{
+	u64 res;
+
+	/*
+	 * Note, we inline this atomic64_t primitive because
+	 * it only clobbers EAX/EDX and leaves the others
+	 * untouched. We also (somewhat subtly) rely on the
+	 * fact that cmpxchg8b returns the current 64-bit value
+	 * of the memory location we are touching:
+	 */
+	asm volatile(
+		"mov %%ebx, %%eax\n\t"
+		"mov %%ecx, %%edx\n\t"
+		LOCK_PREFIX "cmpxchg8b %1\n"
+			: "=&A" (res)
+			: "m" (*ptr)
+		);
+
+	return res;
+}
+
+extern u64 atomic64_read(atomic64_t *ptr);
+
+/**
+ * atomic64_add_return - add and return
+ * @delta: integer value to add
+ * @ptr:   pointer to type atomic64_t
+ *
+ * Atomically adds @delta to @ptr and returns @delta + *@ptr
+ */
+extern u64 atomic64_add_return(u64 delta, atomic64_t *ptr);
+
+/*
+ * Other variants with different arithmetic operators:
+ */
+extern u64 atomic64_sub_return(u64 delta, atomic64_t *ptr);
+extern u64 atomic64_inc_return(atomic64_t *ptr);
+extern u64 atomic64_dec_return(atomic64_t *ptr);
+
+/**
+ * atomic64_add - add integer to atomic64 variable
+ * @delta: integer value to add
+ * @ptr:   pointer to type atomic64_t
+ *
+ * Atomically adds @delta to @ptr.
+ */
+extern void atomic64_add(u64 delta, atomic64_t *ptr);
+
+/**
+ * atomic64_sub - subtract the atomic64 variable
+ * @delta: integer value to subtract
+ * @ptr:   pointer to type atomic64_t
+ *
+ * Atomically subtracts @delta from @ptr.
+ */
+extern void atomic64_sub(u64 delta, atomic64_t *ptr);
+
+/**
+ * atomic64_sub_and_test - subtract value from variable and test result
+ * @delta: integer value to subtract
+ * @ptr:   pointer to type atomic64_t
+ *
+ * Atomically subtracts @delta from @ptr and returns
+ * true if the result is zero, or false for all
+ * other cases.
+ */
+extern int atomic64_sub_and_test(u64 delta, atomic64_t *ptr);
+
+/**
+ * atomic64_inc - increment atomic64 variable
+ * @ptr: pointer to type atomic64_t
+ *
+ * Atomically increments @ptr by 1.
+ */
+extern void atomic64_inc(atomic64_t *ptr);
+
+/**
+ * atomic64_dec - decrement atomic64 variable
+ * @ptr: pointer to type atomic64_t
+ *
+ * Atomically decrements @ptr by 1.
+ */
+extern void atomic64_dec(atomic64_t *ptr);
+
+/**
+ * atomic64_dec_and_test - decrement and test
+ * @ptr: pointer to type atomic64_t
+ *
+ * Atomically decrements @ptr by 1 and
+ * returns true if the result is 0, or false for all other
+ * cases.
+ */
+extern int atomic64_dec_and_test(atomic64_t *ptr);
+
+/**
+ * atomic64_inc_and_test - increment and test
+ * @ptr: pointer to type atomic64_t
+ *
+ * Atomically increments @ptr by 1
+ * and returns true if the result is zero, or false for all
+ * other cases.
+ */
+extern int atomic64_inc_and_test(atomic64_t *ptr);
+
+/**
+ * atomic64_add_negative - add and test if negative
+ * @delta: integer value to add
+ * @ptr:   pointer to type atomic64_t
+ *
+ * Atomically adds @delta to @ptr and returns true
+ * if the result is negative, or false when
+ * result is greater than or equal to zero.
+ */
+extern int atomic64_add_negative(u64 delta, atomic64_t *ptr);
+
+#endif /* _ASM_X86_ATOMIC64_32_H */
diff --git a/arch/x86/include/asm/atomic64_64.h b/arch/x86/include/asm/atomic64_64.h
new file mode 100644
index 000000000000..51c5b4056929
--- /dev/null
+++ b/arch/x86/include/asm/atomic64_64.h
@@ -0,0 +1,224 @@
+#ifndef _ASM_X86_ATOMIC64_64_H
+#define _ASM_X86_ATOMIC64_64_H
+
+#include <linux/types.h>
+#include <asm/alternative.h>
+#include <asm/cmpxchg.h>
+
+/* The 64-bit atomic type */
+
+#define ATOMIC64_INIT(i)	{ (i) }
+
+/**
+ * atomic64_read - read atomic64 variable
+ * @v: pointer of type atomic64_t
+ *
+ * Atomically reads the value of @v.
+ * Doesn't imply a read memory barrier.
+ */
+static inline long atomic64_read(const atomic64_t *v)
+{
+	return v->counter;
+}
+
+/**
+ * atomic64_set - set atomic64 variable
+ * @v: pointer to type atomic64_t
+ * @i: required value
+ *
+ * Atomically sets the value of @v to @i.
+ */
+static inline void atomic64_set(atomic64_t *v, long i)
+{
+	v->counter = i;
+}
+
+/**
+ * atomic64_add - add integer to atomic64 variable
+ * @i: integer value to add
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically adds @i to @v.
+ */
+static inline void atomic64_add(long i, atomic64_t *v)
+{
+	asm volatile(LOCK_PREFIX "addq %1,%0"
+		     : "=m" (v->counter)
+		     : "er" (i), "m" (v->counter));
+}
+
+/**
+ * atomic64_sub - subtract the atomic64 variable
+ * @i: integer value to subtract
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically subtracts @i from @v.
+ */
+static inline void atomic64_sub(long i, atomic64_t *v)
+{
+	asm volatile(LOCK_PREFIX "subq %1,%0"
+		     : "=m" (v->counter)
+		     : "er" (i), "m" (v->counter));
+}
+
+/**
+ * atomic64_sub_and_test - subtract value from variable and test result
+ * @i: integer value to subtract
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically subtracts @i from @v and returns
+ * true if the result is zero, or false for all
+ * other cases.
+ */
+static inline int atomic64_sub_and_test(long i, atomic64_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "subq %2,%0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "er" (i), "m" (v->counter) : "memory");
+	return c;
+}
+
+/**
+ * atomic64_inc - increment atomic64 variable
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically increments @v by 1.
+ */
+static inline void atomic64_inc(atomic64_t *v)
+{
+	asm volatile(LOCK_PREFIX "incq %0"
+		     : "=m" (v->counter)
+		     : "m" (v->counter));
+}
+
+/**
+ * atomic64_dec - decrement atomic64 variable
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically decrements @v by 1.
+ */
+static inline void atomic64_dec(atomic64_t *v)
+{
+	asm volatile(LOCK_PREFIX "decq %0"
+		     : "=m" (v->counter)
+		     : "m" (v->counter));
+}
+
+/**
+ * atomic64_dec_and_test - decrement and test
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically decrements @v by 1 and
+ * returns true if the result is 0, or false for all other
+ * cases.
+ */
+static inline int atomic64_dec_and_test(atomic64_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "decq %0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "m" (v->counter) : "memory");
+	return c != 0;
+}
+
+/**
+ * atomic64_inc_and_test - increment and test
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically increments @v by 1
+ * and returns true if the result is zero, or false for all
+ * other cases.
+ */
+static inline int atomic64_inc_and_test(atomic64_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "incq %0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "m" (v->counter) : "memory");
+	return c != 0;
+}
+
+/**
+ * atomic64_add_negative - add and test if negative
+ * @i: integer value to add
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically adds @i to @v and returns true
+ * if the result is negative, or false when
+ * result is greater than or equal to zero.
+ */
+static inline int atomic64_add_negative(long i, atomic64_t *v)
+{
+	unsigned char c;
+
+	asm volatile(LOCK_PREFIX "addq %2,%0; sets %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "er" (i), "m" (v->counter) : "memory");
+	return c;
+}
+
+/**
+ * atomic64_add_return - add and return
+ * @i: integer value to add
+ * @v: pointer to type atomic64_t
+ *
+ * Atomically adds @i to @v and returns @i + @v
+ */
+static inline long atomic64_add_return(long i, atomic64_t *v)
+{
+	long __i = i;
+	asm volatile(LOCK_PREFIX "xaddq %0, %1;"
+		     : "+r" (i), "+m" (v->counter)
+		     : : "memory");
+	return i + __i;
+}
+
+static inline long atomic64_sub_return(long i, atomic64_t *v)
+{
+	return atomic64_add_return(-i, v);
+}
+
+#define atomic64_inc_return(v)  (atomic64_add_return(1, (v)))
+#define atomic64_dec_return(v)  (atomic64_sub_return(1, (v)))
+
+static inline long atomic64_cmpxchg(atomic64_t *v, long old, long new)
+{
+	return cmpxchg(&v->counter, old, new);
+}
+
+static inline long atomic64_xchg(atomic64_t *v, long new)
+{
+	return xchg(&v->counter, new);
+}
+
+/**
+ * atomic64_add_unless - add unless the number is a given value
+ * @v: pointer of type atomic64_t
+ * @a: the amount to add to v...
+ * @u: ...unless v is equal to u.
+ *
+ * Atomically adds @a to @v, so long as it was not @u.
+ * Returns non-zero if @v was not @u, and zero otherwise.
+ */
+static inline int atomic64_add_unless(atomic64_t *v, long a, long u)
+{
+	long c, old;
+	c = atomic64_read(v);
+	for (;;) {
+		if (unlikely(c == (u)))
+			break;
+		old = atomic64_cmpxchg((v), c, c + (a));
+		if (likely(old == c))
+			break;
+		c = old;
+	}
+	return c != (u);
+}
+
+#define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
+
+#endif /* _ASM_X86_ATOMIC64_64_H */
diff --git a/arch/x86/include/asm/atomic_32.h b/arch/x86/include/asm/atomic_32.h
deleted file mode 100644
index dc5a667ff791..000000000000
--- a/arch/x86/include/asm/atomic_32.h
+++ /dev/null
@@ -1,415 +0,0 @@
-#ifndef _ASM_X86_ATOMIC_32_H
-#define _ASM_X86_ATOMIC_32_H
-
-#include <linux/compiler.h>
-#include <linux/types.h>
-#include <asm/processor.h>
-#include <asm/cmpxchg.h>
-
-/*
- * Atomic operations that C can't guarantee us.  Useful for
- * resource counting etc..
- */
-
-#define ATOMIC_INIT(i)	{ (i) }
-
-/**
- * atomic_read - read atomic variable
- * @v: pointer of type atomic_t
- *
- * Atomically reads the value of @v.
- */
-static inline int atomic_read(const atomic_t *v)
-{
-	return v->counter;
-}
-
-/**
- * atomic_set - set atomic variable
- * @v: pointer of type atomic_t
- * @i: required value
- *
- * Atomically sets the value of @v to @i.
- */
-static inline void atomic_set(atomic_t *v, int i)
-{
-	v->counter = i;
-}
-
-/**
- * atomic_add - add integer to atomic variable
- * @i: integer value to add
- * @v: pointer of type atomic_t
- *
- * Atomically adds @i to @v.
- */
-static inline void atomic_add(int i, atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "addl %1,%0"
-		     : "+m" (v->counter)
-		     : "ir" (i));
-}
-
-/**
- * atomic_sub - subtract integer from atomic variable
- * @i: integer value to subtract
- * @v: pointer of type atomic_t
- *
- * Atomically subtracts @i from @v.
- */
-static inline void atomic_sub(int i, atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "subl %1,%0"
-		     : "+m" (v->counter)
-		     : "ir" (i));
-}
-
-/**
- * atomic_sub_and_test - subtract value from variable and test result
- * @i: integer value to subtract
- * @v: pointer of type atomic_t
- *
- * Atomically subtracts @i from @v and returns
- * true if the result is zero, or false for all
- * other cases.
- */
-static inline int atomic_sub_and_test(int i, atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "subl %2,%0; sete %1"
-		     : "+m" (v->counter), "=qm" (c)
-		     : "ir" (i) : "memory");
-	return c;
-}
-
-/**
- * atomic_inc - increment atomic variable
- * @v: pointer of type atomic_t
- *
- * Atomically increments @v by 1.
- */
-static inline void atomic_inc(atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "incl %0"
-		     : "+m" (v->counter));
-}
-
-/**
- * atomic_dec - decrement atomic variable
- * @v: pointer of type atomic_t
- *
- * Atomically decrements @v by 1.
- */
-static inline void atomic_dec(atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "decl %0"
-		     : "+m" (v->counter));
-}
-
-/**
- * atomic_dec_and_test - decrement and test
- * @v: pointer of type atomic_t
- *
- * Atomically decrements @v by 1 and
- * returns true if the result is 0, or false for all other
- * cases.
- */
-static inline int atomic_dec_and_test(atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "decl %0; sete %1"
-		     : "+m" (v->counter), "=qm" (c)
-		     : : "memory");
-	return c != 0;
-}
-
-/**
- * atomic_inc_and_test - increment and test
- * @v: pointer of type atomic_t
- *
- * Atomically increments @v by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-static inline int atomic_inc_and_test(atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "incl %0; sete %1"
-		     : "+m" (v->counter), "=qm" (c)
-		     : : "memory");
-	return c != 0;
-}
-
-/**
- * atomic_add_negative - add and test if negative
- * @v: pointer of type atomic_t
- * @i: integer value to add
- *
- * Atomically adds @i to @v and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
- */
-static inline int atomic_add_negative(int i, atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "addl %2,%0; sets %1"
-		     : "+m" (v->counter), "=qm" (c)
-		     : "ir" (i) : "memory");
-	return c;
-}
-
-/**
- * atomic_add_return - add integer and return
- * @v: pointer of type atomic_t
- * @i: integer value to add
- *
- * Atomically adds @i to @v and returns @i + @v
- */
-static inline int atomic_add_return(int i, atomic_t *v)
-{
-	int __i;
-#ifdef CONFIG_M386
-	unsigned long flags;
-	if (unlikely(boot_cpu_data.x86 <= 3))
-		goto no_xadd;
-#endif
-	/* Modern 486+ processor */
-	__i = i;
-	asm volatile(LOCK_PREFIX "xaddl %0, %1"
-		     : "+r" (i), "+m" (v->counter)
-		     : : "memory");
-	return i + __i;
-
-#ifdef CONFIG_M386
-no_xadd: /* Legacy 386 processor */
-	local_irq_save(flags);
-	__i = atomic_read(v);
-	atomic_set(v, i + __i);
-	local_irq_restore(flags);
-	return i + __i;
-#endif
-}
-
-/**
- * atomic_sub_return - subtract integer and return
- * @v: pointer of type atomic_t
- * @i: integer value to subtract
- *
- * Atomically subtracts @i from @v and returns @v - @i
- */
-static inline int atomic_sub_return(int i, atomic_t *v)
-{
-	return atomic_add_return(-i, v);
-}
-
-static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
-{
-	return cmpxchg(&v->counter, old, new);
-}
-
-static inline int atomic_xchg(atomic_t *v, int new)
-{
-	return xchg(&v->counter, new);
-}
-
-/**
- * atomic_add_unless - add unless the number is already a given value
- * @v: pointer of type atomic_t
- * @a: the amount to add to v...
- * @u: ...unless v is equal to u.
- *
- * Atomically adds @a to @v, so long as @v was not already @u.
- * Returns non-zero if @v was not @u, and zero otherwise.
- */
-static inline int atomic_add_unless(atomic_t *v, int a, int u)
-{
-	int c, old;
-	c = atomic_read(v);
-	for (;;) {
-		if (unlikely(c == (u)))
-			break;
-		old = atomic_cmpxchg((v), c, c + (a));
-		if (likely(old == c))
-			break;
-		c = old;
-	}
-	return c != (u);
-}
-
-#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
-
-#define atomic_inc_return(v)  (atomic_add_return(1, v))
-#define atomic_dec_return(v)  (atomic_sub_return(1, v))
-
-/* These are x86-specific, used by some header files */
-#define atomic_clear_mask(mask, addr)				\
-	asm volatile(LOCK_PREFIX "andl %0,%1"			\
-		     : : "r" (~(mask)), "m" (*(addr)) : "memory")
-
-#define atomic_set_mask(mask, addr)				\
-	asm volatile(LOCK_PREFIX "orl %0,%1"				\
-		     : : "r" (mask), "m" (*(addr)) : "memory")
-
-/* Atomic operations are already serializing on x86 */
-#define smp_mb__before_atomic_dec()	barrier()
-#define smp_mb__after_atomic_dec()	barrier()
-#define smp_mb__before_atomic_inc()	barrier()
-#define smp_mb__after_atomic_inc()	barrier()
-
-/* An 64bit atomic type */
-
-typedef struct {
-	u64 __aligned(8) counter;
-} atomic64_t;
-
-#define ATOMIC64_INIT(val)	{ (val) }
-
-extern u64 atomic64_cmpxchg(atomic64_t *ptr, u64 old_val, u64 new_val);
-
-/**
- * atomic64_xchg - xchg atomic64 variable
- * @ptr:      pointer to type atomic64_t
- * @new_val:  value to assign
- *
- * Atomically xchgs the value of @ptr to @new_val and returns
- * the old value.
- */
-extern u64 atomic64_xchg(atomic64_t *ptr, u64 new_val);
-
-/**
- * atomic64_set - set atomic64 variable
- * @ptr:      pointer to type atomic64_t
- * @new_val:  value to assign
- *
- * Atomically sets the value of @ptr to @new_val.
- */
-extern void atomic64_set(atomic64_t *ptr, u64 new_val);
-
-/**
- * atomic64_read - read atomic64 variable
- * @ptr:      pointer to type atomic64_t
- *
- * Atomically reads the value of @ptr and returns it.
- */
-static inline u64 atomic64_read(atomic64_t *ptr)
-{
-	u64 res;
-
-	/*
-	 * Note, we inline this atomic64_t primitive because
-	 * it only clobbers EAX/EDX and leaves the others
-	 * untouched. We also (somewhat subtly) rely on the
-	 * fact that cmpxchg8b returns the current 64-bit value
-	 * of the memory location we are touching:
-	 */
-	asm volatile(
-		"mov %%ebx, %%eax\n\t"
-		"mov %%ecx, %%edx\n\t"
-		LOCK_PREFIX "cmpxchg8b %1\n"
-			: "=&A" (res)
-			: "m" (*ptr)
-		);
-
-	return res;
-}
-
-extern u64 atomic64_read(atomic64_t *ptr);
-
-/**
- * atomic64_add_return - add and return
- * @delta: integer value to add
- * @ptr:   pointer to type atomic64_t
- *
- * Atomically adds @delta to @ptr and returns @delta + *@ptr
- */
-extern u64 atomic64_add_return(u64 delta, atomic64_t *ptr);
-
-/*
- * Other variants with different arithmetic operators:
- */
-extern u64 atomic64_sub_return(u64 delta, atomic64_t *ptr);
-extern u64 atomic64_inc_return(atomic64_t *ptr);
-extern u64 atomic64_dec_return(atomic64_t *ptr);
-
-/**
- * atomic64_add - add integer to atomic64 variable
- * @delta: integer value to add
- * @ptr:   pointer to type atomic64_t
- *
- * Atomically adds @delta to @ptr.
- */
-extern void atomic64_add(u64 delta, atomic64_t *ptr);
-
-/**
- * atomic64_sub - subtract the atomic64 variable
- * @delta: integer value to subtract
- * @ptr:   pointer to type atomic64_t
- *
- * Atomically subtracts @delta from @ptr.
- */
-extern void atomic64_sub(u64 delta, atomic64_t *ptr);
-
-/**
- * atomic64_sub_and_test - subtract value from variable and test result
- * @delta: integer value to subtract
- * @ptr:   pointer to type atomic64_t
- *
- * Atomically subtracts @delta from @ptr and returns
- * true if the result is zero, or false for all
- * other cases.
- */
-extern int atomic64_sub_and_test(u64 delta, atomic64_t *ptr);
-
-/**
- * atomic64_inc - increment atomic64 variable
- * @ptr: pointer to type atomic64_t
- *
- * Atomically increments @ptr by 1.
- */
-extern void atomic64_inc(atomic64_t *ptr);
-
-/**
- * atomic64_dec - decrement atomic64 variable
- * @ptr: pointer to type atomic64_t
- *
- * Atomically decrements @ptr by 1.
- */
-extern void atomic64_dec(atomic64_t *ptr);
-
-/**
- * atomic64_dec_and_test - decrement and test
- * @ptr: pointer to type atomic64_t
- *
- * Atomically decrements @ptr by 1 and
- * returns true if the result is 0, or false for all other
- * cases.
- */
-extern int atomic64_dec_and_test(atomic64_t *ptr);
-
-/**
- * atomic64_inc_and_test - increment and test
- * @ptr: pointer to type atomic64_t
- *
- * Atomically increments @ptr by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-extern int atomic64_inc_and_test(atomic64_t *ptr);
-
-/**
- * atomic64_add_negative - add and test if negative
- * @delta: integer value to add
- * @ptr:   pointer to type atomic64_t
- *
- * Atomically adds @delta to @ptr and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
- */
-extern int atomic64_add_negative(u64 delta, atomic64_t *ptr);
-
-#include <asm-generic/atomic-long.h>
-#endif /* _ASM_X86_ATOMIC_32_H */
diff --git a/arch/x86/include/asm/atomic_64.h b/arch/x86/include/asm/atomic_64.h
deleted file mode 100644
index d605dc268e79..000000000000
--- a/arch/x86/include/asm/atomic_64.h
+++ /dev/null
@@ -1,485 +0,0 @@
-#ifndef _ASM_X86_ATOMIC_64_H
-#define _ASM_X86_ATOMIC_64_H
-
-#include <linux/types.h>
-#include <asm/alternative.h>
-#include <asm/cmpxchg.h>
-
-/*
- * Atomic operations that C can't guarantee us.  Useful for
- * resource counting etc..
- */
-
-#define ATOMIC_INIT(i)	{ (i) }
-
-/**
- * atomic_read - read atomic variable
- * @v: pointer of type atomic_t
- *
- * Atomically reads the value of @v.
- */
-static inline int atomic_read(const atomic_t *v)
-{
-	return v->counter;
-}
-
-/**
- * atomic_set - set atomic variable
- * @v: pointer of type atomic_t
- * @i: required value
- *
- * Atomically sets the value of @v to @i.
- */
-static inline void atomic_set(atomic_t *v, int i)
-{
-	v->counter = i;
-}
-
-/**
- * atomic_add - add integer to atomic variable
- * @i: integer value to add
- * @v: pointer of type atomic_t
- *
- * Atomically adds @i to @v.
- */
-static inline void atomic_add(int i, atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "addl %1,%0"
-		     : "=m" (v->counter)
-		     : "ir" (i), "m" (v->counter));
-}
-
-/**
- * atomic_sub - subtract the atomic variable
- * @i: integer value to subtract
- * @v: pointer of type atomic_t
- *
- * Atomically subtracts @i from @v.
- */
-static inline void atomic_sub(int i, atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "subl %1,%0"
-		     : "=m" (v->counter)
-		     : "ir" (i), "m" (v->counter));
-}
-
-/**
- * atomic_sub_and_test - subtract value from variable and test result
- * @i: integer value to subtract
- * @v: pointer of type atomic_t
- *
- * Atomically subtracts @i from @v and returns
- * true if the result is zero, or false for all
- * other cases.
- */
-static inline int atomic_sub_and_test(int i, atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "subl %2,%0; sete %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "ir" (i), "m" (v->counter) : "memory");
-	return c;
-}
-
-/**
- * atomic_inc - increment atomic variable
- * @v: pointer of type atomic_t
- *
- * Atomically increments @v by 1.
- */
-static inline void atomic_inc(atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "incl %0"
-		     : "=m" (v->counter)
-		     : "m" (v->counter));
-}
-
-/**
- * atomic_dec - decrement atomic variable
- * @v: pointer of type atomic_t
- *
- * Atomically decrements @v by 1.
- */
-static inline void atomic_dec(atomic_t *v)
-{
-	asm volatile(LOCK_PREFIX "decl %0"
-		     : "=m" (v->counter)
-		     : "m" (v->counter));
-}
-
-/**
- * atomic_dec_and_test - decrement and test
- * @v: pointer of type atomic_t
- *
- * Atomically decrements @v by 1 and
- * returns true if the result is 0, or false for all other
- * cases.
- */
-static inline int atomic_dec_and_test(atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "decl %0; sete %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "m" (v->counter) : "memory");
-	return c != 0;
-}
-
-/**
- * atomic_inc_and_test - increment and test
- * @v: pointer of type atomic_t
- *
- * Atomically increments @v by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-static inline int atomic_inc_and_test(atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "incl %0; sete %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "m" (v->counter) : "memory");
-	return c != 0;
-}
-
-/**
- * atomic_add_negative - add and test if negative
- * @i: integer value to add
- * @v: pointer of type atomic_t
- *
- * Atomically adds @i to @v and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
- */
-static inline int atomic_add_negative(int i, atomic_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "addl %2,%0; sets %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "ir" (i), "m" (v->counter) : "memory");
-	return c;
-}
-
-/**
- * atomic_add_return - add and return
- * @i: integer value to add
- * @v: pointer of type atomic_t
- *
- * Atomically adds @i to @v and returns @i + @v
- */
-static inline int atomic_add_return(int i, atomic_t *v)
-{
-	int __i = i;
-	asm volatile(LOCK_PREFIX "xaddl %0, %1"
-		     : "+r" (i), "+m" (v->counter)
-		     : : "memory");
-	return i + __i;
-}
-
-static inline int atomic_sub_return(int i, atomic_t *v)
-{
-	return atomic_add_return(-i, v);
-}
-
-#define atomic_inc_return(v)  (atomic_add_return(1, v))
-#define atomic_dec_return(v)  (atomic_sub_return(1, v))
-
-/* The 64-bit atomic type */
-
-#define ATOMIC64_INIT(i)	{ (i) }
-
-/**
- * atomic64_read - read atomic64 variable
- * @v: pointer of type atomic64_t
- *
- * Atomically reads the value of @v.
- * Doesn't imply a read memory barrier.
- */
-static inline long atomic64_read(const atomic64_t *v)
-{
-	return v->counter;
-}
-
-/**
- * atomic64_set - set atomic64 variable
- * @v: pointer to type atomic64_t
- * @i: required value
- *
- * Atomically sets the value of @v to @i.
- */
-static inline void atomic64_set(atomic64_t *v, long i)
-{
-	v->counter = i;
-}
-
-/**
- * atomic64_add - add integer to atomic64 variable
- * @i: integer value to add
- * @v: pointer to type atomic64_t
- *
- * Atomically adds @i to @v.
- */
-static inline void atomic64_add(long i, atomic64_t *v)
-{
-	asm volatile(LOCK_PREFIX "addq %1,%0"
-		     : "=m" (v->counter)
-		     : "er" (i), "m" (v->counter));
-}
-
-/**
- * atomic64_sub - subtract the atomic64 variable
- * @i: integer value to subtract
- * @v: pointer to type atomic64_t
- *
- * Atomically subtracts @i from @v.
- */
-static inline void atomic64_sub(long i, atomic64_t *v)
-{
-	asm volatile(LOCK_PREFIX "subq %1,%0"
-		     : "=m" (v->counter)
-		     : "er" (i), "m" (v->counter));
-}
-
-/**
- * atomic64_sub_and_test - subtract value from variable and test result
- * @i: integer value to subtract
- * @v: pointer to type atomic64_t
- *
- * Atomically subtracts @i from @v and returns
- * true if the result is zero, or false for all
- * other cases.
- */
-static inline int atomic64_sub_and_test(long i, atomic64_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "subq %2,%0; sete %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "er" (i), "m" (v->counter) : "memory");
-	return c;
-}
-
-/**
- * atomic64_inc - increment atomic64 variable
- * @v: pointer to type atomic64_t
- *
- * Atomically increments @v by 1.
- */
-static inline void atomic64_inc(atomic64_t *v)
-{
-	asm volatile(LOCK_PREFIX "incq %0"
-		     : "=m" (v->counter)
-		     : "m" (v->counter));
-}
-
-/**
- * atomic64_dec - decrement atomic64 variable
- * @v: pointer to type atomic64_t
- *
- * Atomically decrements @v by 1.
- */
-static inline void atomic64_dec(atomic64_t *v)
-{
-	asm volatile(LOCK_PREFIX "decq %0"
-		     : "=m" (v->counter)
-		     : "m" (v->counter));
-}
-
-/**
- * atomic64_dec_and_test - decrement and test
- * @v: pointer to type atomic64_t
- *
- * Atomically decrements @v by 1 and
- * returns true if the result is 0, or false for all other
- * cases.
- */
-static inline int atomic64_dec_and_test(atomic64_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "decq %0; sete %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "m" (v->counter) : "memory");
-	return c != 0;
-}
-
-/**
- * atomic64_inc_and_test - increment and test
- * @v: pointer to type atomic64_t
- *
- * Atomically increments @v by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-static inline int atomic64_inc_and_test(atomic64_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "incq %0; sete %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "m" (v->counter) : "memory");
-	return c != 0;
-}
-
-/**
- * atomic64_add_negative - add and test if negative
- * @i: integer value to add
- * @v: pointer to type atomic64_t
- *
- * Atomically adds @i to @v and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
- */
-static inline int atomic64_add_negative(long i, atomic64_t *v)
-{
-	unsigned char c;
-
-	asm volatile(LOCK_PREFIX "addq %2,%0; sets %1"
-		     : "=m" (v->counter), "=qm" (c)
-		     : "er" (i), "m" (v->counter) : "memory");
-	return c;
-}
-
-/**
- * atomic64_add_return - add and return
- * @i: integer value to add
- * @v: pointer to type atomic64_t
- *
- * Atomically adds @i to @v and returns @i + @v
- */
-static inline long atomic64_add_return(long i, atomic64_t *v)
-{
-	long __i = i;
-	asm volatile(LOCK_PREFIX "xaddq %0, %1;"
-		     : "+r" (i), "+m" (v->counter)
-		     : : "memory");
-	return i + __i;
-}
-
-static inline long atomic64_sub_return(long i, atomic64_t *v)
-{
-	return atomic64_add_return(-i, v);
-}
-
-#define atomic64_inc_return(v)  (atomic64_add_return(1, (v)))
-#define atomic64_dec_return(v)  (atomic64_sub_return(1, (v)))
-
-static inline long atomic64_cmpxchg(atomic64_t *v, long old, long new)
-{
-	return cmpxchg(&v->counter, old, new);
-}
-
-static inline long atomic64_xchg(atomic64_t *v, long new)
-{
-	return xchg(&v->counter, new);
-}
-
-static inline long atomic_cmpxchg(atomic_t *v, int old, int new)
-{
-	return cmpxchg(&v->counter, old, new);
-}
-
-static inline long atomic_xchg(atomic_t *v, int new)
-{
-	return xchg(&v->counter, new);
-}
-
-/**
- * atomic_add_unless - add unless the number is a given value
- * @v: pointer of type atomic_t
- * @a: the amount to add to v...
- * @u: ...unless v is equal to u.
- *
- * Atomically adds @a to @v, so long as it was not @u.
- * Returns non-zero if @v was not @u, and zero otherwise.
- */
-static inline int atomic_add_unless(atomic_t *v, int a, int u)
-{
-	int c, old;
-	c = atomic_read(v);
-	for (;;) {
-		if (unlikely(c == (u)))
-			break;
-		old = atomic_cmpxchg((v), c, c + (a));
-		if (likely(old == c))
-			break;
-		c = old;
-	}
-	return c != (u);
-}
-
-#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
-
-/**
- * atomic64_add_unless - add unless the number is a given value
- * @v: pointer of type atomic64_t
- * @a: the amount to add to v...
- * @u: ...unless v is equal to u.
- *
- * Atomically adds @a to @v, so long as it was not @u.
- * Returns non-zero if @v was not @u, and zero otherwise.
- */
-static inline int atomic64_add_unless(atomic64_t *v, long a, long u)
-{
-	long c, old;
-	c = atomic64_read(v);
-	for (;;) {
-		if (unlikely(c == (u)))
-			break;
-		old = atomic64_cmpxchg((v), c, c + (a));
-		if (likely(old == c))
-			break;
-		c = old;
-	}
-	return c != (u);
-}
-
-/**
- * atomic_inc_short - increment of a short integer
- * @v: pointer to type int
- *
- * Atomically adds 1 to @v
- * Returns the new value of @u
- */
-static inline short int atomic_inc_short(short int *v)
-{
-	asm(LOCK_PREFIX "addw $1, %0" : "+m" (*v));
-	return *v;
-}
-
-/**
- * atomic_or_long - OR of two long integers
- * @v1: pointer to type unsigned long
- * @v2: pointer to type unsigned long
- *
- * Atomically ORs @v1 and @v2
- * Returns the result of the OR
- */
-static inline void atomic_or_long(unsigned long *v1, unsigned long v2)
-{
-	asm(LOCK_PREFIX "orq %1, %0" : "+m" (*v1) : "r" (v2));
-}
-
-#define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
-
-/* These are x86-specific, used by some header files */
-#define atomic_clear_mask(mask, addr)					\
-	asm volatile(LOCK_PREFIX "andl %0,%1"				\
-		     : : "r" (~(mask)), "m" (*(addr)) : "memory")
-
-#define atomic_set_mask(mask, addr)					\
-	asm volatile(LOCK_PREFIX "orl %0,%1"				\
-		     : : "r" ((unsigned)(mask)), "m" (*(addr))		\
-		     : "memory")
-
-/* Atomic operations are already serializing on x86 */
-#define smp_mb__before_atomic_dec()	barrier()
-#define smp_mb__after_atomic_dec()	barrier()
-#define smp_mb__before_atomic_inc()	barrier()
-#define smp_mb__after_atomic_inc()	barrier()
-
-#include <asm-generic/atomic-long.h>
-#endif /* _ASM_X86_ATOMIC_64_H */
diff --git a/arch/x86/include/asm/debugreg.h b/arch/x86/include/asm/debugreg.h
index 8240f76b531e..b81002f23614 100644
--- a/arch/x86/include/asm/debugreg.h
+++ b/arch/x86/include/asm/debugreg.h
@@ -14,6 +14,9 @@
    which debugging register was responsible for the trap.  The other bits
    are either reserved or not of interest to us. */
 
+/* Define reserved bits in DR6 which are always set to 1 */
+#define DR6_RESERVED	(0xFFFF0FF0)
+
 #define DR_TRAP0	(0x1)		/* db0 */
 #define DR_TRAP1	(0x2)		/* db1 */
 #define DR_TRAP2	(0x4)		/* db2 */
diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h
index 14f9890eb495..635f03bb4995 100644
--- a/arch/x86/include/asm/fixmap.h
+++ b/arch/x86/include/asm/fixmap.h
@@ -118,14 +118,20 @@ enum fixed_addresses {
 	 * 256 temporary boot-time mappings, used by early_ioremap(),
 	 * before ioremap() is functional.
 	 *
-	 * We round it up to the next 256 pages boundary so that we
-	 * can have a single pgd entry and a single pte table:
+	 * If necessary we round it up to the next 256 pages boundary so
+	 * that we can have a single pgd entry and a single pte table:
 	 */
 #define NR_FIX_BTMAPS		64
 #define FIX_BTMAPS_SLOTS	4
-	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
-			(__end_of_permanent_fixed_addresses & 255),
-	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
+#define TOTAL_FIX_BTMAPS	(NR_FIX_BTMAPS * FIX_BTMAPS_SLOTS)
+	FIX_BTMAP_END =
+	 (__end_of_permanent_fixed_addresses ^
+	  (__end_of_permanent_fixed_addresses + TOTAL_FIX_BTMAPS - 1)) &
+	 -PTRS_PER_PTE
+	 ? __end_of_permanent_fixed_addresses + TOTAL_FIX_BTMAPS -
+	   (__end_of_permanent_fixed_addresses & (TOTAL_FIX_BTMAPS - 1))
+	 : __end_of_permanent_fixed_addresses,
+	FIX_BTMAP_BEGIN = FIX_BTMAP_END + TOTAL_FIX_BTMAPS - 1,
 #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
 	FIX_OHCI1394_BASE,
 #endif
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index ebfb8a9e11f7..1cd5d43fee49 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -32,9 +32,11 @@ extern void __math_state_restore(void);
 extern void init_thread_xstate(void);
 extern int dump_fpu(struct pt_regs *, struct user_i387_struct *);
 
-extern user_regset_active_fn fpregs_active, xfpregs_active;
-extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;
-extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set;
+extern user_regset_active_fn fpregs_active, xfpregs_active, xstateregs_active;
+extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get,
+				xstateregs_get;
+extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set,
+				 xstateregs_set;
 
 extern struct _fpx_sw_bytes fx_sw_reserved;
 #ifdef CONFIG_IA32_EMULATION
diff --git a/arch/x86/include/asm/irq.h b/arch/x86/include/asm/irq.h
index 5458380b6ef8..262292729fc4 100644
--- a/arch/x86/include/asm/irq.h
+++ b/arch/x86/include/asm/irq.h
@@ -48,5 +48,6 @@ extern DECLARE_BITMAP(used_vectors, NR_VECTORS);
 extern int vector_used_by_percpu_irq(unsigned int vector);
 
 extern void init_ISA_irqs(void);
+extern int nr_legacy_irqs;
 
 #endif /* _ASM_X86_IRQ_H */
diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h
index 4611f085cd43..8767d99c4f64 100644
--- a/arch/x86/include/asm/irq_vectors.h
+++ b/arch/x86/include/asm/irq_vectors.h
@@ -28,28 +28,33 @@
 #define MCE_VECTOR			0x12
 
 /*
- * IDT vectors usable for external interrupt sources start
- * at 0x20:
+ * IDT vectors usable for external interrupt sources start at 0x20.
+ * (0x80 is the syscall vector, 0x30-0x3f are for ISA)
  */
 #define FIRST_EXTERNAL_VECTOR		0x20
-
-#ifdef CONFIG_X86_32
-# define SYSCALL_VECTOR			0x80
-# define IA32_SYSCALL_VECTOR		0x80
-#else
-# define IA32_SYSCALL_VECTOR		0x80
-#endif
+/*
+ * We start allocating at 0x21 to spread out vectors evenly between
+ * priority levels. (0x80 is the syscall vector)
+ */
+#define VECTOR_OFFSET_START		1
 
 /*
- * Reserve the lowest usable priority level 0x20 - 0x2f for triggering
- * cleanup after irq migration.
+ * Reserve the lowest usable vector (and hence lowest priority)  0x20 for
+ * triggering cleanup after irq migration. 0x21-0x2f will still be used
+ * for device interrupts.
  */
 #define IRQ_MOVE_CLEANUP_VECTOR		FIRST_EXTERNAL_VECTOR
 
+#define IA32_SYSCALL_VECTOR		0x80
+#ifdef CONFIG_X86_32
+# define SYSCALL_VECTOR			0x80
+#endif
+
 /*
  * Vectors 0x30-0x3f are used for ISA interrupts.
+ *   round up to the next 16-vector boundary
  */
-#define IRQ0_VECTOR			(FIRST_EXTERNAL_VECTOR + 0x10)
+#define IRQ0_VECTOR			((FIRST_EXTERNAL_VECTOR + 16) & ~15)
 
 #define IRQ1_VECTOR			(IRQ0_VECTOR +  1)
 #define IRQ2_VECTOR			(IRQ0_VECTOR +  2)
@@ -120,13 +125,6 @@
  */
 #define MCE_SELF_VECTOR			0xeb
 
-/*
- * First APIC vector available to drivers: (vectors 0x30-0xee) we
- * start at 0x31(0x41) to spread out vectors evenly between priority
- * levels. (0x80 is the syscall vector)
- */
-#define FIRST_DEVICE_VECTOR		(IRQ15_VECTOR + 2)
-
 #define NR_VECTORS			 256
 
 #define FPU_IRQ				  13
@@ -154,21 +152,21 @@ static inline int invalid_vm86_irq(int irq)
 
 #define NR_IRQS_LEGACY			  16
 
-#define CPU_VECTOR_LIMIT		(  8 * NR_CPUS      )
 #define IO_APIC_VECTOR_LIMIT		( 32 * MAX_IO_APICS )
 
 #ifdef CONFIG_X86_IO_APIC
 # ifdef CONFIG_SPARSE_IRQ
+#  define CPU_VECTOR_LIMIT		(64 * NR_CPUS)
 #  define NR_IRQS					\
 	(CPU_VECTOR_LIMIT > IO_APIC_VECTOR_LIMIT ?	\
 		(NR_VECTORS + CPU_VECTOR_LIMIT)  :	\
 		(NR_VECTORS + IO_APIC_VECTOR_LIMIT))
 # else
-#  if NR_CPUS < MAX_IO_APICS
-#   define NR_IRQS 			(NR_VECTORS + 4*CPU_VECTOR_LIMIT)
-#  else
-#   define NR_IRQS			(NR_VECTORS + IO_APIC_VECTOR_LIMIT)
-#  endif
+#  define CPU_VECTOR_LIMIT		(32 * NR_CPUS)
+#  define NR_IRQS					\
+	(CPU_VECTOR_LIMIT < IO_APIC_VECTOR_LIMIT ?	\
+		(NR_VECTORS + CPU_VECTOR_LIMIT)  :	\
+		(NR_VECTORS + IO_APIC_VECTOR_LIMIT))
 # endif
 #else /* !CONFIG_X86_IO_APIC: */
 # define NR_IRQS			NR_IRQS_LEGACY
diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h
index 139d4c1a33a7..93da9c3f3341 100644
--- a/arch/x86/include/asm/nmi.h
+++ b/arch/x86/include/asm/nmi.h
@@ -19,7 +19,6 @@ extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
 extern int check_nmi_watchdog(void);
 extern int nmi_watchdog_enabled;
 extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
-extern int avail_to_resrv_perfctr_nmi(unsigned int);
 extern int reserve_perfctr_nmi(unsigned int);
 extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
diff --git a/arch/x86/include/asm/page_types.h b/arch/x86/include/asm/page_types.h
index 642fe34b36a2..a667f24c7254 100644
--- a/arch/x86/include/asm/page_types.h
+++ b/arch/x86/include/asm/page_types.h
@@ -40,7 +40,6 @@
 
 #ifndef __ASSEMBLY__
 
-extern int page_is_ram(unsigned long pagenr);
 extern int devmem_is_allowed(unsigned long pagenr);
 
 extern unsigned long max_low_pfn_mapped;
diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h
index 1380367dabd9..befd172c82ad 100644
--- a/arch/x86/include/asm/perf_event.h
+++ b/arch/x86/include/asm/perf_event.h
@@ -27,7 +27,14 @@
 /*
  * Includes eventsel and unit mask as well:
  */
-#define ARCH_PERFMON_EVENT_MASK				    0xffff
+
+
+#define INTEL_ARCH_EVTSEL_MASK		0x000000FFULL
+#define INTEL_ARCH_UNIT_MASK		0x0000FF00ULL
+#define INTEL_ARCH_EDGE_MASK		0x00040000ULL
+#define INTEL_ARCH_INV_MASK		0x00800000ULL
+#define INTEL_ARCH_CNT_MASK		0xFF000000ULL
+#define INTEL_ARCH_EVENT_MASK	(INTEL_ARCH_UNIT_MASK|INTEL_ARCH_EVTSEL_MASK)
 
 /*
  * filter mask to validate fixed counter events.
@@ -38,7 +45,12 @@
  *  The other filters are supported by fixed counters.
  *  The any-thread option is supported starting with v3.
  */
-#define ARCH_PERFMON_EVENT_FILTER_MASK			0xff840000
+#define INTEL_ARCH_FIXED_MASK \
+	(INTEL_ARCH_CNT_MASK| \
+	 INTEL_ARCH_INV_MASK| \
+	 INTEL_ARCH_EDGE_MASK|\
+	 INTEL_ARCH_UNIT_MASK|\
+	 INTEL_ARCH_EVTSEL_MASK)
 
 #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL		      0x3c
 #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK		(0x00 << 8)
diff --git a/arch/x86/include/asm/ptrace-abi.h b/arch/x86/include/asm/ptrace-abi.h
index 86723035a515..447cfb63160e 100644
--- a/arch/x86/include/asm/ptrace-abi.h
+++ b/arch/x86/include/asm/ptrace-abi.h
@@ -80,6 +80,44 @@
 
 #define PTRACE_SINGLEBLOCK	33	/* resume execution until next branch */
 
+/*
+ * Structure layout used in PTRACE_GETXSTATEREGS/PTRACE_SETXSTATEREGS is same
+ * as the memory layout of xsave used by the processor (except for the bytes
+ * 464..511 which can be used by the software). Size of the structure that users
+ * need to use for these two interfaces can be obtained by doing:
+ *	cpuid_count(0xd, 0, &eax, &ptrace_xstateregs_struct_size, &ecx, &edx);
+ * i.e., cpuid.(eax=0xd,ecx=0).ebx will be the size that user (debuggers etc)
+ * need to use.
+ *
+ * And format of this structure will be like:
+ *	struct {
+ *		fxsave_bytes[0..463]
+ *		sw_usable_bytes[464..511]
+ *		xsave_hdr_bytes[512..575]
+ *		avx_bytes[576..831]
+ *		future_state etc
+ *	}
+ *
+ * Same memory layout will be used for the coredump NT_X86_XSTATE representing
+ * the xstate registers.
+ *
+ * For now, only first 8 bytes of the sw_usable_bytes[464..467] will be used and
+ * will be set to OS enabled xstate mask(which is same as the 64bit mask
+ * returned by the xgetbv's xCR0). Users (analyzing core dump remotely etc)
+ * can use this mask aswell as the mask saved in the xstate_hdr bytes and
+ * interpret what states the processor/OS supports and what states are in
+ * modified/initialized conditions for the particular process/thread.
+ *
+ * Also when the user modifies certain state FP/SSE/etc through this
+ * PTRACE_SETXSTATEREGS, they must ensure that the xsave_hdr.xstate_bv
+ * bytes[512..519] of the above memory layout are updated correspondingly.
+ * i.e., for example when FP state is modified to a non-init state,
+ * xsave_hdr.xstate_bv's bit 0 must be set to '1', when SSE is modified to
+ * non-init state, xsave_hdr.xstate_bv's bit 1 must to be set to '1' etc..
+ */
+#define PTRACE_GETXSTATEREGS      34
+#define PTRACE_SETXSTATEREGS      35
+
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
 
diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h
index 9d369f680321..20102808b191 100644
--- a/arch/x86/include/asm/ptrace.h
+++ b/arch/x86/include/asm/ptrace.h
@@ -274,10 +274,6 @@ static inline unsigned long regs_get_kernel_stack_nth(struct pt_regs *regs,
 		return 0;
 }
 
-/* Get Nth argument at function call */
-extern unsigned long regs_get_argument_nth(struct pt_regs *regs,
-					   unsigned int n);
-
 /*
  * These are defined as per linux/ptrace.h, which see.
  */
diff --git a/arch/x86/include/asm/rwsem.h b/arch/x86/include/asm/rwsem.h
index ca7517d33776..10204a25bf93 100644
--- a/arch/x86/include/asm/rwsem.h
+++ b/arch/x86/include/asm/rwsem.h
@@ -41,6 +41,7 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/lockdep.h>
+#include <asm/asm.h>
 
 struct rwsem_waiter;
 
@@ -55,17 +56,28 @@ extern asmregparm struct rw_semaphore *
 
 /*
  * the semaphore definition
+ *
+ * The bias values and the counter type limits the number of
+ * potential readers/writers to 32767 for 32 bits and 2147483647
+ * for 64 bits.
  */
 
-#define RWSEM_UNLOCKED_VALUE		0x00000000
-#define RWSEM_ACTIVE_BIAS		0x00000001
-#define RWSEM_ACTIVE_MASK		0x0000ffff
-#define RWSEM_WAITING_BIAS		(-0x00010000)
+#ifdef CONFIG_X86_64
+# define RWSEM_ACTIVE_MASK		0xffffffffL
+#else
+# define RWSEM_ACTIVE_MASK		0x0000ffffL
+#endif
+
+#define RWSEM_UNLOCKED_VALUE		0x00000000L
+#define RWSEM_ACTIVE_BIAS		0x00000001L
+#define RWSEM_WAITING_BIAS		(-RWSEM_ACTIVE_MASK-1)
 #define RWSEM_ACTIVE_READ_BIAS		RWSEM_ACTIVE_BIAS
 #define RWSEM_ACTIVE_WRITE_BIAS		(RWSEM_WAITING_BIAS + RWSEM_ACTIVE_BIAS)
 
+typedef signed long rwsem_count_t;
+
 struct rw_semaphore {
-	signed long		count;
+	rwsem_count_t		count;
 	spinlock_t		wait_lock;
 	struct list_head	wait_list;
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
@@ -105,7 +117,7 @@ do {								\
 static inline void __down_read(struct rw_semaphore *sem)
 {
 	asm volatile("# beginning down_read\n\t"
-		     LOCK_PREFIX "  incl      (%%eax)\n\t"
+		     LOCK_PREFIX _ASM_INC "(%1)\n\t"
 		     /* adds 0x00000001, returns the old value */
 		     "  jns        1f\n"
 		     "  call call_rwsem_down_read_failed\n"
@@ -121,14 +133,14 @@ static inline void __down_read(struct rw_semaphore *sem)
  */
 static inline int __down_read_trylock(struct rw_semaphore *sem)
 {
-	__s32 result, tmp;
+	rwsem_count_t result, tmp;
 	asm volatile("# beginning __down_read_trylock\n\t"
-		     "  movl      %0,%1\n\t"
+		     "  mov          %0,%1\n\t"
 		     "1:\n\t"
-		     "  movl	     %1,%2\n\t"
-		     "  addl      %3,%2\n\t"
+		     "  mov          %1,%2\n\t"
+		     "  add          %3,%2\n\t"
 		     "  jle	     2f\n\t"
-		     LOCK_PREFIX "  cmpxchgl  %2,%0\n\t"
+		     LOCK_PREFIX "  cmpxchg  %2,%0\n\t"
 		     "  jnz	     1b\n\t"
 		     "2:\n\t"
 		     "# ending __down_read_trylock\n\t"
@@ -143,13 +155,13 @@ static inline int __down_read_trylock(struct rw_semaphore *sem)
  */
 static inline void __down_write_nested(struct rw_semaphore *sem, int subclass)
 {
-	int tmp;
+	rwsem_count_t tmp;
 
 	tmp = RWSEM_ACTIVE_WRITE_BIAS;
 	asm volatile("# beginning down_write\n\t"
-		     LOCK_PREFIX "  xadd      %%edx,(%%eax)\n\t"
+		     LOCK_PREFIX "  xadd      %1,(%2)\n\t"
 		     /* subtract 0x0000ffff, returns the old value */
-		     "  testl     %%edx,%%edx\n\t"
+		     "  test      %1,%1\n\t"
 		     /* was the count 0 before? */
 		     "  jz        1f\n"
 		     "  call call_rwsem_down_write_failed\n"
@@ -170,9 +182,9 @@ static inline void __down_write(struct rw_semaphore *sem)
  */
 static inline int __down_write_trylock(struct rw_semaphore *sem)
 {
-	signed long ret = cmpxchg(&sem->count,
-				  RWSEM_UNLOCKED_VALUE,
-				  RWSEM_ACTIVE_WRITE_BIAS);
+	rwsem_count_t ret = cmpxchg(&sem->count,
+				    RWSEM_UNLOCKED_VALUE,
+				    RWSEM_ACTIVE_WRITE_BIAS);
 	if (ret == RWSEM_UNLOCKED_VALUE)
 		return 1;
 	return 0;
@@ -183,9 +195,9 @@ static inline int __down_write_trylock(struct rw_semaphore *sem)
  */
 static inline void __up_read(struct rw_semaphore *sem)
 {
-	__s32 tmp = -RWSEM_ACTIVE_READ_BIAS;
+	rwsem_count_t tmp = -RWSEM_ACTIVE_READ_BIAS;
 	asm volatile("# beginning __up_read\n\t"
-		     LOCK_PREFIX "  xadd      %%edx,(%%eax)\n\t"
+		     LOCK_PREFIX "  xadd      %1,(%2)\n\t"
 		     /* subtracts 1, returns the old value */
 		     "  jns        1f\n\t"
 		     "  call call_rwsem_wake\n"
@@ -201,18 +213,18 @@ static inline void __up_read(struct rw_semaphore *sem)
  */
 static inline void __up_write(struct rw_semaphore *sem)
 {
+	rwsem_count_t tmp;
 	asm volatile("# beginning __up_write\n\t"
-		     "  movl      %2,%%edx\n\t"
-		     LOCK_PREFIX "  xaddl     %%edx,(%%eax)\n\t"
+		     LOCK_PREFIX "  xadd      %1,(%2)\n\t"
 		     /* tries to transition
 			0xffff0001 -> 0x00000000 */
 		     "  jz       1f\n"
 		     "  call call_rwsem_wake\n"
 		     "1:\n\t"
 		     "# ending __up_write\n"
-		     : "+m" (sem->count)
-		     : "a" (sem), "i" (-RWSEM_ACTIVE_WRITE_BIAS)
-		     : "memory", "cc", "edx");
+		     : "+m" (sem->count), "=d" (tmp)
+		     : "a" (sem), "1" (-RWSEM_ACTIVE_WRITE_BIAS)
+		     : "memory", "cc");
 }
 
 /*
@@ -220,8 +232,25 @@ static inline void __up_write(struct rw_semaphore *sem)
  */
 static inline void __downgrade_write(struct rw_semaphore *sem)
 {
+#ifdef CONFIG_X86_64
+# if RWSEM_WAITING_BIAS != -0x100000000
+#  error "This code assumes RWSEM_WAITING_BIAS == -2^32"
+# endif
+
+	/* 64-bit immediates are special and expensive, and not needed here */
 	asm volatile("# beginning __downgrade_write\n\t"
-		     LOCK_PREFIX "  addl      %2,(%%eax)\n\t"
+		     LOCK_PREFIX "incl 4(%1)\n\t"
+		     /* transitions 0xZZZZZZZZ00000001 -> 0xYYYYYYYY00000001 */
+		     "  jns       1f\n\t"
+		     "  call call_rwsem_downgrade_wake\n"
+		     "1:\n\t"
+		     "# ending __downgrade_write\n"
+		     : "+m" (sem->count)
+		     : "a" (sem)
+		     : "memory", "cc");
+#else
+	asm volatile("# beginning __downgrade_write\n\t"
+		     LOCK_PREFIX _ASM_ADD "%2,(%1)\n\t"
 		     /* transitions 0xZZZZ0001 -> 0xYYYY0001 */
 		     "  jns       1f\n\t"
 		     "  call call_rwsem_downgrade_wake\n"
@@ -230,24 +259,27 @@ static inline void __downgrade_write(struct rw_semaphore *sem)
 		     : "+m" (sem->count)
 		     : "a" (sem), "i" (-RWSEM_WAITING_BIAS)
 		     : "memory", "cc");
+#endif
 }
 
 /*
  * implement atomic add functionality
  */
-static inline void rwsem_atomic_add(int delta, struct rw_semaphore *sem)
+static inline void rwsem_atomic_add(rwsem_count_t delta,
+				    struct rw_semaphore *sem)
 {
-	asm volatile(LOCK_PREFIX "addl %1,%0"
+	asm volatile(LOCK_PREFIX _ASM_ADD "%1,%0"
 		     : "+m" (sem->count)
-		     : "ir" (delta));
+		     : "er" (delta));
 }
 
 /*
  * implement exchange and add functionality
  */
-static inline int rwsem_atomic_update(int delta, struct rw_semaphore *sem)
+static inline rwsem_count_t rwsem_atomic_update(rwsem_count_t delta,
+						struct rw_semaphore *sem)
 {
-	int tmp = delta;
+	rwsem_count_t tmp = delta;
 
 	asm volatile(LOCK_PREFIX "xadd %0,%1"
 		     : "+r" (tmp), "+m" (sem->count)
diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
index 1e796782cd7b..4cfc90824068 100644
--- a/arch/x86/include/asm/smp.h
+++ b/arch/x86/include/asm/smp.h
@@ -135,6 +135,8 @@ int native_cpu_disable(void);
 void native_cpu_die(unsigned int cpu);
 void native_play_dead(void);
 void play_dead_common(void);
+void wbinvd_on_cpu(int cpu);
+int wbinvd_on_all_cpus(void);
 
 void native_send_call_func_ipi(const struct cpumask *mask);
 void native_send_call_func_single_ipi(int cpu);
@@ -147,6 +149,13 @@ static inline int num_booting_cpus(void)
 {
 	return cpumask_weight(cpu_callout_mask);
 }
+#else /* !CONFIG_SMP */
+#define wbinvd_on_cpu(cpu)     wbinvd()
+static inline int wbinvd_on_all_cpus(void)
+{
+	wbinvd();
+	return 0;
+}
 #endif /* CONFIG_SMP */
 
 extern unsigned disabled_cpus __cpuinitdata;
diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h
index 35e89122a42f..4dab78edbad9 100644
--- a/arch/x86/include/asm/stacktrace.h
+++ b/arch/x86/include/asm/stacktrace.h
@@ -3,8 +3,6 @@
 
 extern int kstack_depth_to_print;
 
-int x86_is_stack_id(int id, char *name);
-
 struct thread_info;
 struct stacktrace_ops;
 
diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h
index 535e421498f6..316708d5af92 100644
--- a/arch/x86/include/asm/uaccess_64.h
+++ b/arch/x86/include/asm/uaccess_64.h
@@ -8,6 +8,8 @@
 #include <linux/errno.h>
 #include <linux/prefetch.h>
 #include <linux/lockdep.h>
+#include <asm/alternative.h>
+#include <asm/cpufeature.h>
 #include <asm/page.h>
 
 /*
@@ -16,7 +18,24 @@
 
 /* Handles exceptions in both to and from, but doesn't do access_ok */
 __must_check unsigned long
-copy_user_generic(void *to, const void *from, unsigned len);
+copy_user_generic_string(void *to, const void *from, unsigned len);
+__must_check unsigned long
+copy_user_generic_unrolled(void *to, const void *from, unsigned len);
+
+static __always_inline __must_check unsigned long
+copy_user_generic(void *to, const void *from, unsigned len)
+{
+	unsigned ret;
+
+	alternative_call(copy_user_generic_unrolled,
+			 copy_user_generic_string,
+			 X86_FEATURE_REP_GOOD,
+			 ASM_OUTPUT2("=a" (ret), "=D" (to), "=S" (from),
+				     "=d" (len)),
+			 "1" (to), "2" (from), "3" (len)
+			 : "memory", "rcx", "r8", "r9", "r10", "r11");
+	return ret;
+}
 
 __must_check unsigned long
 _copy_to_user(void __user *to, const void *from, unsigned len);
diff --git a/arch/x86/include/asm/uv/bios.h b/arch/x86/include/asm/uv/bios.h
index 2751f3075d8b..3fbc1f348a7d 100644
--- a/arch/x86/include/asm/uv/bios.h
+++ b/arch/x86/include/asm/uv/bios.h
@@ -18,8 +18,8 @@
  *  along with this program; if not, write to the Free Software
  *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  *
- *  Copyright (c) 2008 Silicon Graphics, Inc.  All Rights Reserved.
- *  Copyright (c) Russ Anderson
+ *  Copyright (c) 2008-2009 Silicon Graphics, Inc.  All Rights Reserved.
+ *  Copyright (c) Russ Anderson <rja@sgi.com>
  */
 
 #include <linux/rtc.h>
@@ -89,7 +89,7 @@ extern s64 uv_bios_call(enum uv_bios_cmd, u64, u64, u64, u64, u64);
 extern s64 uv_bios_call_irqsave(enum uv_bios_cmd, u64, u64, u64, u64, u64);
 extern s64 uv_bios_call_reentrant(enum uv_bios_cmd, u64, u64, u64, u64, u64);
 
-extern s64 uv_bios_get_sn_info(int, int *, long *, long *, long *);
+extern s64 uv_bios_get_sn_info(int, int *, long *, long *, long *, long *);
 extern s64 uv_bios_freq_base(u64, u64 *);
 extern int uv_bios_mq_watchlist_alloc(unsigned long, unsigned int,
 					unsigned long *);
@@ -104,6 +104,7 @@ extern int uv_type;
 extern long sn_partition_id;
 extern long sn_coherency_id;
 extern long sn_region_size;
+extern long system_serial_number;
 #define partition_coherence_id()	(sn_coherency_id)
 
 extern struct kobject *sgi_uv_kobj;	/* /sys/firmware/sgi_uv */
diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h
index 40be813fefb1..14cc74ba5d23 100644
--- a/arch/x86/include/asm/uv/uv_hub.h
+++ b/arch/x86/include/asm/uv/uv_hub.h
@@ -329,7 +329,8 @@ static inline unsigned long uv_read_global_mmr64(int pnode, unsigned long offset
  */
 static inline unsigned long uv_global_gru_mmr_address(int pnode, unsigned long offset)
 {
-	return UV_GLOBAL_GRU_MMR_BASE | offset | (pnode << uv_hub_info->m_val);
+	return UV_GLOBAL_GRU_MMR_BASE | offset |
+		((unsigned long)pnode << uv_hub_info->m_val);
 }
 
 static inline void uv_write_global_mmr8(int pnode, unsigned long offset, unsigned char val)
diff --git a/arch/x86/include/asm/xsave.h b/arch/x86/include/asm/xsave.h
index 727acc152344..9e26171d38df 100644
--- a/arch/x86/include/asm/xsave.h
+++ b/arch/x86/include/asm/xsave.h
@@ -27,9 +27,11 @@
 extern unsigned int xstate_size;
 extern u64 pcntxt_mask;
 extern struct xsave_struct *init_xstate_buf;
+extern u64 xstate_fx_sw_bytes[6];
 
 extern void xsave_cntxt_init(void);
 extern void xsave_init(void);
+extern void update_regset_xstate_info(unsigned int size, u64 xstate_mask);
 extern int init_fpu(struct task_struct *child);
 extern int check_for_xstate(struct i387_fxsave_struct __user *buf,
 			    void __user *fpstate,
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index de7353c0ce9c..2589ea4c60ce 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -205,7 +205,7 @@ void __init_or_module apply_alternatives(struct alt_instr *start,
 					 struct alt_instr *end)
 {
 	struct alt_instr *a;
-	char insnbuf[MAX_PATCH_LEN];
+	u8 insnbuf[MAX_PATCH_LEN];
 
 	DPRINTK("%s: alt table %p -> %p\n", __func__, start, end);
 	for (a = start; a < end; a++) {
@@ -223,6 +223,8 @@ void __init_or_module apply_alternatives(struct alt_instr *start,
 		}
 #endif
 		memcpy(insnbuf, a->replacement, a->replacementlen);
+		if (*insnbuf == 0xe8 && a->replacementlen == 5)
+		    *(s32 *)(insnbuf + 1) += a->replacement - a->instr;
 		add_nops(insnbuf + a->replacementlen,
 			 a->instrlen - a->replacementlen);
 		text_poke_early(instr, insnbuf, a->instrlen);
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 53243ca7816d..9969094ce75b 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -94,8 +94,6 @@ struct mpc_intsrc mp_irqs[MAX_IRQ_SOURCES];
 /* # of MP IRQ source entries */
 int mp_irq_entries;
 
-/* Number of legacy interrupts */
-static int nr_legacy_irqs __read_mostly = NR_IRQS_LEGACY;
 /* GSI interrupts */
 static int nr_irqs_gsi = NR_IRQS_LEGACY;
 
@@ -140,27 +138,10 @@ static struct irq_pin_list *get_one_free_irq_2_pin(int node)
 
 /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */
 #ifdef CONFIG_SPARSE_IRQ
-static struct irq_cfg irq_cfgx[] = {
+static struct irq_cfg irq_cfgx[NR_IRQS_LEGACY];
 #else
-static struct irq_cfg irq_cfgx[NR_IRQS] = {
+static struct irq_cfg irq_cfgx[NR_IRQS];
 #endif
-	[0]  = { .vector = IRQ0_VECTOR,  },
-	[1]  = { .vector = IRQ1_VECTOR,  },
-	[2]  = { .vector = IRQ2_VECTOR,  },
-	[3]  = { .vector = IRQ3_VECTOR,  },
-	[4]  = { .vector = IRQ4_VECTOR,  },
-	[5]  = { .vector = IRQ5_VECTOR,  },
-	[6]  = { .vector = IRQ6_VECTOR,  },
-	[7]  = { .vector = IRQ7_VECTOR,  },
-	[8]  = { .vector = IRQ8_VECTOR,  },
-	[9]  = { .vector = IRQ9_VECTOR,  },
-	[10] = { .vector = IRQ10_VECTOR, },
-	[11] = { .vector = IRQ11_VECTOR, },
-	[12] = { .vector = IRQ12_VECTOR, },
-	[13] = { .vector = IRQ13_VECTOR, },
-	[14] = { .vector = IRQ14_VECTOR, },
-	[15] = { .vector = IRQ15_VECTOR, },
-};
 
 void __init io_apic_disable_legacy(void)
 {
@@ -185,8 +166,14 @@ int __init arch_early_irq_init(void)
 		desc->chip_data = &cfg[i];
 		zalloc_cpumask_var_node(&cfg[i].domain, GFP_NOWAIT, node);
 		zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_NOWAIT, node);
-		if (i < nr_legacy_irqs)
-			cpumask_setall(cfg[i].domain);
+		/*
+		 * For legacy IRQ's, start with assigning irq0 to irq15 to
+		 * IRQ0_VECTOR to IRQ15_VECTOR on cpu 0.
+		 */
+		if (i < nr_legacy_irqs) {
+			cfg[i].vector = IRQ0_VECTOR + i;
+			cpumask_set_cpu(0, cfg[i].domain);
+		}
 	}
 
 	return 0;
@@ -1162,7 +1149,8 @@ __assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 	 * Also, we've got to be careful not to trash gate
 	 * 0x80, because int 0x80 is hm, kind of importantish. ;)
 	 */
-	static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0;
+	static int current_vector = FIRST_EXTERNAL_VECTOR + VECTOR_OFFSET_START;
+	static int current_offset = VECTOR_OFFSET_START % 8;
 	unsigned int old_vector;
 	int cpu, err;
 	cpumask_var_t tmp_mask;
@@ -1198,7 +1186,7 @@ next:
 		if (vector >= first_system_vector) {
 			/* If out of vectors on large boxen, must share them. */
 			offset = (offset + 1) % 8;
-			vector = FIRST_DEVICE_VECTOR + offset;
+			vector = FIRST_EXTERNAL_VECTOR + offset;
 		}
 		if (unlikely(current_vector == vector))
 			continue;
@@ -1268,11 +1256,16 @@ static void __clear_irq_vector(int irq, struct irq_cfg *cfg)
 void __setup_vector_irq(int cpu)
 {
 	/* Initialize vector_irq on a new cpu */
-	/* This function must be called with vector_lock held */
 	int irq, vector;
 	struct irq_cfg *cfg;
 	struct irq_desc *desc;
 
+	/*
+	 * vector_lock will make sure that we don't run into irq vector
+	 * assignments that might be happening on another cpu in parallel,
+	 * while we setup our initial vector to irq mappings.
+	 */
+	spin_lock(&vector_lock);
 	/* Mark the inuse vectors */
 	for_each_irq_desc(irq, desc) {
 		cfg = desc->chip_data;
@@ -1291,6 +1284,7 @@ void __setup_vector_irq(int cpu)
 		if (!cpumask_test_cpu(cpu, cfg->domain))
 			per_cpu(vector_irq, cpu)[vector] = -1;
 	}
+	spin_unlock(&vector_lock);
 }
 
 static struct irq_chip ioapic_chip;
@@ -1440,6 +1434,14 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq
 
 	cfg = desc->chip_data;
 
+	/*
+	 * For legacy irqs, cfg->domain starts with cpu 0 for legacy
+	 * controllers like 8259. Now that IO-APIC can handle this irq, update
+	 * the cfg->domain.
+	 */
+	if (irq < nr_legacy_irqs && cpumask_test_cpu(0, cfg->domain))
+		apic->vector_allocation_domain(0, cfg->domain);
+
 	if (assign_irq_vector(irq, cfg, apic->target_cpus()))
 		return;
 
@@ -3847,7 +3849,7 @@ int __init arch_probe_nr_irqs(void)
 	/*
 	 * for MSI and HT dyn irq
 	 */
-	nr += nr_irqs_gsi * 16;
+	nr += nr_irqs_gsi * 64;
 #endif
 	if (nr < nr_irqs)
 		nr_irqs = nr;
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index 21db3cbea7dc..a8ae82e0a654 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -5,7 +5,7 @@
  *
  * SGI UV APIC functions (note: not an Intel compatible APIC)
  *
- * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 2007-2009 Silicon Graphics, Inc. All rights reserved.
  */
 #include <linux/cpumask.h>
 #include <linux/hardirq.h>
@@ -482,7 +482,7 @@ static void uv_heartbeat(unsigned long ignored)
 
 static void __cpuinit uv_heartbeat_enable(int cpu)
 {
-	if (!uv_cpu_hub_info(cpu)->scir.enabled) {
+	while (!uv_cpu_hub_info(cpu)->scir.enabled) {
 		struct timer_list *timer = &uv_cpu_hub_info(cpu)->scir.timer;
 
 		uv_set_cpu_scir_bits(cpu, SCIR_CPU_HEARTBEAT|SCIR_CPU_ACTIVITY);
@@ -490,11 +490,10 @@ static void __cpuinit uv_heartbeat_enable(int cpu)
 		timer->expires = jiffies + SCIR_CPU_HB_INTERVAL;
 		add_timer_on(timer, cpu);
 		uv_cpu_hub_info(cpu)->scir.enabled = 1;
-	}
 
-	/* check boot cpu */
-	if (!uv_cpu_hub_info(0)->scir.enabled)
-		uv_heartbeat_enable(0);
+		/* also ensure that boot cpu is enabled */
+		cpu = 0;
+	}
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -634,8 +633,8 @@ void __init uv_system_init(void)
 	}
 
 	uv_bios_init();
-	uv_bios_get_sn_info(0, &uv_type, &sn_partition_id,
-			    &sn_coherency_id, &sn_region_size);
+	uv_bios_get_sn_info(0, &uv_type, &sn_partition_id, &sn_coherency_id,
+			    &sn_region_size, &system_serial_number);
 	uv_rtc_init();
 
 	for_each_present_cpu(cpu) {
diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c
index b0206a211b09..c918ebab52ab 100644
--- a/arch/x86/kernel/bios_uv.c
+++ b/arch/x86/kernel/bios_uv.c
@@ -15,8 +15,8 @@
  *  along with this program; if not, write to the Free Software
  *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  *
- *  Copyright (c) 2008 Silicon Graphics, Inc.  All Rights Reserved.
- *  Copyright (c) Russ Anderson
+ *  Copyright (c) 2008-2009 Silicon Graphics, Inc.  All Rights Reserved.
+ *  Copyright (c) Russ Anderson <rja@sgi.com>
  */
 
 #include <linux/efi.h>
@@ -30,6 +30,7 @@ static struct uv_systab uv_systab;
 s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5)
 {
 	struct uv_systab *tab = &uv_systab;
+	s64 ret;
 
 	if (!tab->function)
 		/*
@@ -37,9 +38,11 @@ s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5)
 		 */
 		return BIOS_STATUS_UNIMPLEMENTED;
 
-	return efi_call6((void *)__va(tab->function),
-					(u64)which, a1, a2, a3, a4, a5);
+	ret = efi_call6((void *)__va(tab->function), (u64)which,
+			a1, a2, a3, a4, a5);
+	return ret;
 }
+EXPORT_SYMBOL_GPL(uv_bios_call);
 
 s64 uv_bios_call_irqsave(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3,
 					u64 a4, u64 a5)
@@ -73,11 +76,14 @@ long sn_coherency_id;
 EXPORT_SYMBOL_GPL(sn_coherency_id);
 long sn_region_size;
 EXPORT_SYMBOL_GPL(sn_region_size);
+long system_serial_number;
+EXPORT_SYMBOL_GPL(system_serial_number);
 int uv_type;
+EXPORT_SYMBOL_GPL(uv_type);
 
 
 s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher,
-		long *region)
+		long *region, long *ssn)
 {
 	s64 ret;
 	u64 v0, v1;
@@ -97,8 +103,11 @@ s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher,
 		*coher = part.coherence_id;
 	if (region)
 		*region = part.region_size;
+	if (ssn)
+		*ssn = v1;
 	return ret;
 }
+EXPORT_SYMBOL_GPL(uv_bios_get_sn_info);
 
 int
 uv_bios_mq_watchlist_alloc(unsigned long addr, unsigned int mq_size,
@@ -185,4 +194,3 @@ void uv_bios_init(void)
 
 void uv_bios_init(void) { }
 #endif
-
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index fc6c8ef92dcc..db1670f24c58 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -18,6 +18,7 @@
 #include <asm/processor.h>
 #include <linux/smp.h>
 #include <asm/k8.h>
+#include <asm/smp.h>
 
 #define LVL_1_INST	1
 #define LVL_1_DATA	2
@@ -31,6 +32,8 @@ struct _cache_table {
 	short size;
 };
 
+#define MB(x)	((x) * 1024)
+
 /* All the cache descriptor types we care about (no TLB or
    trace cache entries) */
 
@@ -44,9 +47,9 @@ static const struct _cache_table __cpuinitconst cache_table[] =
 	{ 0x0d, LVL_1_DATA, 16 },	/* 4-way set assoc, 64 byte line size */
 	{ 0x21, LVL_2,      256 },	/* 8-way set assoc, 64 byte line size */
 	{ 0x22, LVL_3,      512 },	/* 4-way set assoc, sectored cache, 64 byte line size */
-	{ 0x23, LVL_3,      1024 },	/* 8-way set assoc, sectored cache, 64 byte line size */
-	{ 0x25, LVL_3,      2048 },	/* 8-way set assoc, sectored cache, 64 byte line size */
-	{ 0x29, LVL_3,      4096 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x23, LVL_3,      MB(1) },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x25, LVL_3,      MB(2) },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x29, LVL_3,      MB(4) },	/* 8-way set assoc, sectored cache, 64 byte line size */
 	{ 0x2c, LVL_1_DATA, 32 },	/* 8-way set assoc, 64 byte line size */
 	{ 0x30, LVL_1_INST, 32 },	/* 8-way set assoc, 64 byte line size */
 	{ 0x39, LVL_2,      128 },	/* 4-way set assoc, sectored cache, 64 byte line size */
@@ -59,16 +62,16 @@ static const struct _cache_table __cpuinitconst cache_table[] =
 	{ 0x41, LVL_2,      128 },	/* 4-way set assoc, 32 byte line size */
 	{ 0x42, LVL_2,      256 },	/* 4-way set assoc, 32 byte line size */
 	{ 0x43, LVL_2,      512 },	/* 4-way set assoc, 32 byte line size */
-	{ 0x44, LVL_2,      1024 },	/* 4-way set assoc, 32 byte line size */
-	{ 0x45, LVL_2,      2048 },	/* 4-way set assoc, 32 byte line size */
-	{ 0x46, LVL_3,      4096 },	/* 4-way set assoc, 64 byte line size */
-	{ 0x47, LVL_3,      8192 },	/* 8-way set assoc, 64 byte line size */
-	{ 0x49, LVL_3,      4096 },	/* 16-way set assoc, 64 byte line size */
-	{ 0x4a, LVL_3,      6144 },	/* 12-way set assoc, 64 byte line size */
-	{ 0x4b, LVL_3,      8192 },	/* 16-way set assoc, 64 byte line size */
-	{ 0x4c, LVL_3,     12288 },	/* 12-way set assoc, 64 byte line size */
-	{ 0x4d, LVL_3,     16384 },	/* 16-way set assoc, 64 byte line size */
-	{ 0x4e, LVL_2,      6144 },	/* 24-way set assoc, 64 byte line size */
+	{ 0x44, LVL_2,      MB(1) },	/* 4-way set assoc, 32 byte line size */
+	{ 0x45, LVL_2,      MB(2) },	/* 4-way set assoc, 32 byte line size */
+	{ 0x46, LVL_3,      MB(4) },	/* 4-way set assoc, 64 byte line size */
+	{ 0x47, LVL_3,      MB(8) },	/* 8-way set assoc, 64 byte line size */
+	{ 0x49, LVL_3,      MB(4) },	/* 16-way set assoc, 64 byte line size */
+	{ 0x4a, LVL_3,      MB(6) },	/* 12-way set assoc, 64 byte line size */
+	{ 0x4b, LVL_3,      MB(8) },	/* 16-way set assoc, 64 byte line size */
+	{ 0x4c, LVL_3,      MB(12) },	/* 12-way set assoc, 64 byte line size */
+	{ 0x4d, LVL_3,      MB(16) },	/* 16-way set assoc, 64 byte line size */
+	{ 0x4e, LVL_2,      MB(6) },	/* 24-way set assoc, 64 byte line size */
 	{ 0x60, LVL_1_DATA, 16 },	/* 8-way set assoc, sectored cache, 64 byte line size */
 	{ 0x66, LVL_1_DATA, 8 },	/* 4-way set assoc, sectored cache, 64 byte line size */
 	{ 0x67, LVL_1_DATA, 16 },	/* 4-way set assoc, sectored cache, 64 byte line size */
@@ -77,34 +80,34 @@ static const struct _cache_table __cpuinitconst cache_table[] =
 	{ 0x71, LVL_TRACE,  16 },	/* 8-way set assoc */
 	{ 0x72, LVL_TRACE,  32 },	/* 8-way set assoc */
 	{ 0x73, LVL_TRACE,  64 },	/* 8-way set assoc */
-	{ 0x78, LVL_2,    1024 },	/* 4-way set assoc, 64 byte line size */
-	{ 0x79, LVL_2,     128 },	/* 8-way set assoc, sectored cache, 64 byte line size */
-	{ 0x7a, LVL_2,     256 },	/* 8-way set assoc, sectored cache, 64 byte line size */
-	{ 0x7b, LVL_2,     512 },	/* 8-way set assoc, sectored cache, 64 byte line size */
-	{ 0x7c, LVL_2,    1024 },	/* 8-way set assoc, sectored cache, 64 byte line size */
-	{ 0x7d, LVL_2,    2048 },	/* 8-way set assoc, 64 byte line size */
-	{ 0x7f, LVL_2,     512 },	/* 2-way set assoc, 64 byte line size */
-	{ 0x82, LVL_2,     256 },	/* 8-way set assoc, 32 byte line size */
-	{ 0x83, LVL_2,     512 },	/* 8-way set assoc, 32 byte line size */
-	{ 0x84, LVL_2,    1024 },	/* 8-way set assoc, 32 byte line size */
-	{ 0x85, LVL_2,    2048 },	/* 8-way set assoc, 32 byte line size */
-	{ 0x86, LVL_2,     512 },	/* 4-way set assoc, 64 byte line size */
-	{ 0x87, LVL_2,    1024 },	/* 8-way set assoc, 64 byte line size */
-	{ 0xd0, LVL_3,     512 },	/* 4-way set assoc, 64 byte line size */
-	{ 0xd1, LVL_3,    1024 },	/* 4-way set assoc, 64 byte line size */
-	{ 0xd2, LVL_3,    2048 },	/* 4-way set assoc, 64 byte line size */
-	{ 0xd6, LVL_3,    1024 },	/* 8-way set assoc, 64 byte line size */
-	{ 0xd7, LVL_3,    2048 },	/* 8-way set assoc, 64 byte line size */
-	{ 0xd8, LVL_3,    4096 },	/* 12-way set assoc, 64 byte line size */
-	{ 0xdc, LVL_3,    2048 },	/* 12-way set assoc, 64 byte line size */
-	{ 0xdd, LVL_3,    4096 },	/* 12-way set assoc, 64 byte line size */
-	{ 0xde, LVL_3,    8192 },	/* 12-way set assoc, 64 byte line size */
-	{ 0xe2, LVL_3,    2048 },	/* 16-way set assoc, 64 byte line size */
-	{ 0xe3, LVL_3,    4096 },	/* 16-way set assoc, 64 byte line size */
-	{ 0xe4, LVL_3,    8192 },	/* 16-way set assoc, 64 byte line size */
-	{ 0xea, LVL_3,    12288 },	/* 24-way set assoc, 64 byte line size */
-	{ 0xeb, LVL_3,    18432 },	/* 24-way set assoc, 64 byte line size */
-	{ 0xec, LVL_3,    24576 },	/* 24-way set assoc, 64 byte line size */
+	{ 0x78, LVL_2,      MB(1) },	/* 4-way set assoc, 64 byte line size */
+	{ 0x79, LVL_2,      128 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7a, LVL_2,      256 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7b, LVL_2,      512 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7c, LVL_2,      MB(1) },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7d, LVL_2,      MB(2) },	/* 8-way set assoc, 64 byte line size */
+	{ 0x7f, LVL_2,      512 },	/* 2-way set assoc, 64 byte line size */
+	{ 0x82, LVL_2,      256 },	/* 8-way set assoc, 32 byte line size */
+	{ 0x83, LVL_2,      512 },	/* 8-way set assoc, 32 byte line size */
+	{ 0x84, LVL_2,      MB(1) },	/* 8-way set assoc, 32 byte line size */
+	{ 0x85, LVL_2,      MB(2) },	/* 8-way set assoc, 32 byte line size */
+	{ 0x86, LVL_2,      512 },	/* 4-way set assoc, 64 byte line size */
+	{ 0x87, LVL_2,      MB(1) },	/* 8-way set assoc, 64 byte line size */
+	{ 0xd0, LVL_3,      512 },	/* 4-way set assoc, 64 byte line size */
+	{ 0xd1, LVL_3,      MB(1) },	/* 4-way set assoc, 64 byte line size */
+	{ 0xd2, LVL_3,      MB(2) },	/* 4-way set assoc, 64 byte line size */
+	{ 0xd6, LVL_3,      MB(1) },	/* 8-way set assoc, 64 byte line size */
+	{ 0xd7, LVL_3,      MB(2) },	/* 8-way set assoc, 64 byte line size */
+	{ 0xd8, LVL_3,      MB(4) },	/* 12-way set assoc, 64 byte line size */
+	{ 0xdc, LVL_3,      MB(2) },	/* 12-way set assoc, 64 byte line size */
+	{ 0xdd, LVL_3,      MB(4) },	/* 12-way set assoc, 64 byte line size */
+	{ 0xde, LVL_3,      MB(8) },	/* 12-way set assoc, 64 byte line size */
+	{ 0xe2, LVL_3,      MB(2) },	/* 16-way set assoc, 64 byte line size */
+	{ 0xe3, LVL_3,      MB(4) },	/* 16-way set assoc, 64 byte line size */
+	{ 0xe4, LVL_3,      MB(8) },	/* 16-way set assoc, 64 byte line size */
+	{ 0xea, LVL_3,      MB(12) },	/* 24-way set assoc, 64 byte line size */
+	{ 0xeb, LVL_3,      MB(18) },	/* 24-way set assoc, 64 byte line size */
+	{ 0xec, LVL_3,      MB(24) },	/* 24-way set assoc, 64 byte line size */
 	{ 0x00, 0, 0}
 };
 
@@ -150,7 +153,8 @@ struct _cpuid4_info {
 	union _cpuid4_leaf_ebx ebx;
 	union _cpuid4_leaf_ecx ecx;
 	unsigned long size;
-	unsigned long can_disable;
+	bool can_disable;
+	unsigned int l3_indices;
 	DECLARE_BITMAP(shared_cpu_map, NR_CPUS);
 };
 
@@ -160,7 +164,8 @@ struct _cpuid4_info_regs {
 	union _cpuid4_leaf_ebx ebx;
 	union _cpuid4_leaf_ecx ecx;
 	unsigned long size;
-	unsigned long can_disable;
+	bool can_disable;
+	unsigned int l3_indices;
 };
 
 unsigned short			num_cache_leaves;
@@ -290,6 +295,29 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
 		(ebx->split.ways_of_associativity + 1) - 1;
 }
 
+static unsigned int __cpuinit amd_calc_l3_indices(void)
+{
+	/*
+	 * We're called over smp_call_function_single() and therefore
+	 * are on the correct cpu.
+	 */
+	int cpu = smp_processor_id();
+	int node = cpu_to_node(cpu);
+	struct pci_dev *dev = node_to_k8_nb_misc(node);
+	unsigned int sc0, sc1, sc2, sc3;
+	u32 val;
+
+	pci_read_config_dword(dev, 0x1C4, &val);
+
+	/* calculate subcache sizes */
+	sc0 = !(val & BIT(0));
+	sc1 = !(val & BIT(4));
+	sc2 = !(val & BIT(8))  + !(val & BIT(9));
+	sc3 = !(val & BIT(12)) + !(val & BIT(13));
+
+	return (max(max(max(sc0, sc1), sc2), sc3) << 10) - 1;
+}
+
 static void __cpuinit
 amd_check_l3_disable(int index, struct _cpuid4_info_regs *this_leaf)
 {
@@ -299,11 +327,14 @@ amd_check_l3_disable(int index, struct _cpuid4_info_regs *this_leaf)
 	if (boot_cpu_data.x86 == 0x11)
 		return;
 
-	/* see erratum #382 */
-	if ((boot_cpu_data.x86 == 0x10) && (boot_cpu_data.x86_model < 0x8))
+	/* see errata #382 and #388 */
+	if ((boot_cpu_data.x86 == 0x10) &&
+	    ((boot_cpu_data.x86_model < 0x9) ||
+	     (boot_cpu_data.x86_mask  < 0x1)))
 		return;
 
-	this_leaf->can_disable = 1;
+	this_leaf->can_disable = true;
+	this_leaf->l3_indices  = amd_calc_l3_indices();
 }
 
 static int
@@ -726,12 +757,12 @@ static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf,
 		return -EINVAL;
 
 	pci_read_config_dword(dev, 0x1BC + index * 4, &reg);
-	return sprintf(buf, "%x\n", reg);
+	return sprintf(buf, "0x%08x\n", reg);
 }
 
 #define SHOW_CACHE_DISABLE(index)					\
 static ssize_t								\
-show_cache_disable_##index(struct _cpuid4_info *this_leaf, char *buf)  	\
+show_cache_disable_##index(struct _cpuid4_info *this_leaf, char *buf)	\
 {									\
 	return show_cache_disable(this_leaf, buf, index);		\
 }
@@ -745,7 +776,9 @@ static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf,
 	int node = cpu_to_node(cpu);
 	struct pci_dev *dev = node_to_k8_nb_misc(node);
 	unsigned long val = 0;
-	unsigned int scrubber = 0;
+
+#define SUBCACHE_MASK	(3UL << 20)
+#define SUBCACHE_INDEX	0xfff
 
 	if (!this_leaf->can_disable)
 		return -EINVAL;
@@ -759,21 +792,25 @@ static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf,
 	if (strict_strtoul(buf, 10, &val) < 0)
 		return -EINVAL;
 
-	val |= 0xc0000000;
-
-	pci_read_config_dword(dev, 0x58, &scrubber);
-	scrubber &= ~0x1f000000;
-	pci_write_config_dword(dev, 0x58, scrubber);
+	/* do not allow writes outside of allowed bits */
+	if ((val & ~(SUBCACHE_MASK | SUBCACHE_INDEX)) ||
+	    ((val & SUBCACHE_INDEX) > this_leaf->l3_indices))
+		return -EINVAL;
 
-	pci_write_config_dword(dev, 0x1BC + index * 4, val & ~0x40000000);
-	wbinvd();
+	val |= BIT(30);
 	pci_write_config_dword(dev, 0x1BC + index * 4, val);
+	/*
+	 * We need to WBINVD on a core on the node containing the L3 cache which
+	 * indices we disable therefore a simple wbinvd() is not sufficient.
+	 */
+	wbinvd_on_cpu(cpu);
+	pci_write_config_dword(dev, 0x1BC + index * 4, val | BIT(31));
 	return count;
 }
 
 #define STORE_CACHE_DISABLE(index)					\
 static ssize_t								\
-store_cache_disable_##index(struct _cpuid4_info *this_leaf,	     	\
+store_cache_disable_##index(struct _cpuid4_info *this_leaf,		\
 			    const char *buf, size_t count)		\
 {									\
 	return store_cache_disable(this_leaf, buf, count, index);	\
@@ -806,16 +843,24 @@ static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644,
 static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644,
 		show_cache_disable_1, store_cache_disable_1);
 
+#define DEFAULT_SYSFS_CACHE_ATTRS	\
+	&type.attr,			\
+	&level.attr,			\
+	&coherency_line_size.attr,	\
+	&physical_line_partition.attr,	\
+	&ways_of_associativity.attr,	\
+	&number_of_sets.attr,		\
+	&size.attr,			\
+	&shared_cpu_map.attr,		\
+	&shared_cpu_list.attr
+
 static struct attribute *default_attrs[] = {
-	&type.attr,
-	&level.attr,
-	&coherency_line_size.attr,
-	&physical_line_partition.attr,
-	&ways_of_associativity.attr,
-	&number_of_sets.attr,
-	&size.attr,
-	&shared_cpu_map.attr,
-	&shared_cpu_list.attr,
+	DEFAULT_SYSFS_CACHE_ATTRS,
+	NULL
+};
+
+static struct attribute *default_l3_attrs[] = {
+	DEFAULT_SYSFS_CACHE_ATTRS,
 	&cache_disable_0.attr,
 	&cache_disable_1.attr,
 	NULL
@@ -908,6 +953,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
 	unsigned int cpu = sys_dev->id;
 	unsigned long i, j;
 	struct _index_kobject *this_object;
+	struct _cpuid4_info   *this_leaf;
 	int retval;
 
 	retval = cpuid4_cache_sysfs_init(cpu);
@@ -926,6 +972,14 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
 		this_object = INDEX_KOBJECT_PTR(cpu, i);
 		this_object->cpu = cpu;
 		this_object->index = i;
+
+		this_leaf = CPUID4_INFO_IDX(cpu, i);
+
+		if (this_leaf->can_disable)
+			ktype_cache.default_attrs = default_l3_attrs;
+		else
+			ktype_cache.default_attrs = default_attrs;
+
 		retval = kobject_init_and_add(&(this_object->kobj),
 					      &ktype_cache,
 					      per_cpu(ici_cache_kobject, cpu),
diff --git a/arch/x86/kernel/cpu/mtrr/amd.c b/arch/x86/kernel/cpu/mtrr/amd.c
index 33af14110dfd..92ba9cd31c9a 100644
--- a/arch/x86/kernel/cpu/mtrr/amd.c
+++ b/arch/x86/kernel/cpu/mtrr/amd.c
@@ -108,7 +108,7 @@ amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
 	return 0;
 }
 
-static struct mtrr_ops amd_mtrr_ops = {
+static const struct mtrr_ops amd_mtrr_ops = {
 	.vendor            = X86_VENDOR_AMD,
 	.set               = amd_set_mtrr,
 	.get               = amd_get_mtrr,
diff --git a/arch/x86/kernel/cpu/mtrr/centaur.c b/arch/x86/kernel/cpu/mtrr/centaur.c
index de89f14eff3a..316fe3e60a97 100644
--- a/arch/x86/kernel/cpu/mtrr/centaur.c
+++ b/arch/x86/kernel/cpu/mtrr/centaur.c
@@ -110,7 +110,7 @@ centaur_validate_add_page(unsigned long base, unsigned long size, unsigned int t
 	return 0;
 }
 
-static struct mtrr_ops centaur_mtrr_ops = {
+static const struct mtrr_ops centaur_mtrr_ops = {
 	.vendor            = X86_VENDOR_CENTAUR,
 	.set               = centaur_set_mcr,
 	.get               = centaur_get_mcr,
diff --git a/arch/x86/kernel/cpu/mtrr/cyrix.c b/arch/x86/kernel/cpu/mtrr/cyrix.c
index 228d982ce09c..68a3343e5798 100644
--- a/arch/x86/kernel/cpu/mtrr/cyrix.c
+++ b/arch/x86/kernel/cpu/mtrr/cyrix.c
@@ -265,7 +265,7 @@ static void cyrix_set_all(void)
 	post_set();
 }
 
-static struct mtrr_ops cyrix_mtrr_ops = {
+static const struct mtrr_ops cyrix_mtrr_ops = {
 	.vendor            = X86_VENDOR_CYRIX,
 	.set_all	   = cyrix_set_all,
 	.set               = cyrix_set_arr,
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 55da0c5f68dd..4d755846fee6 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -752,7 +752,7 @@ int positive_have_wrcomb(void)
 /*
  * Generic structure...
  */
-struct mtrr_ops generic_mtrr_ops = {
+const struct mtrr_ops generic_mtrr_ops = {
 	.use_intel_if		= 1,
 	.set_all		= generic_set_all,
 	.get			= generic_get_mtrr,
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 84e83de54575..fe4622e8c837 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -60,14 +60,14 @@ static DEFINE_MUTEX(mtrr_mutex);
 u64 size_or_mask, size_and_mask;
 static bool mtrr_aps_delayed_init;
 
-static struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM];
+static const struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM];
 
-struct mtrr_ops *mtrr_if;
+const struct mtrr_ops *mtrr_if;
 
 static void set_mtrr(unsigned int reg, unsigned long base,
 		     unsigned long size, mtrr_type type);
 
-void set_mtrr_ops(struct mtrr_ops *ops)
+void set_mtrr_ops(const struct mtrr_ops *ops)
 {
 	if (ops->vendor && ops->vendor < X86_VENDOR_NUM)
 		mtrr_ops[ops->vendor] = ops;
diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index a501dee9a87a..df5e41f31a27 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -32,7 +32,7 @@ extern int generic_get_free_region(unsigned long base, unsigned long size,
 extern int generic_validate_add_page(unsigned long base, unsigned long size,
 				     unsigned int type);
 
-extern struct mtrr_ops generic_mtrr_ops;
+extern const struct mtrr_ops generic_mtrr_ops;
 
 extern int positive_have_wrcomb(void);
 
@@ -53,10 +53,10 @@ void fill_mtrr_var_range(unsigned int index,
 		u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi);
 void get_mtrr_state(void);
 
-extern void set_mtrr_ops(struct mtrr_ops *ops);
+extern void set_mtrr_ops(const struct mtrr_ops *ops);
 
 extern u64 size_or_mask, size_and_mask;
-extern struct mtrr_ops *mtrr_if;
+extern const struct mtrr_ops *mtrr_if;
 
 #define is_cpu(vnd)	(mtrr_if && mtrr_if->vendor == X86_VENDOR_##vnd)
 #define use_intel()	(mtrr_if && mtrr_if->use_intel_if == 1)
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 8c1c07073ccc..1846ead0576b 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -7,6 +7,7 @@
  *  Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
  *  Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
  *  Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com>
+ *  Copyright (C) 2009 Google, Inc., Stephane Eranian
  *
  *  For licencing details see kernel-base/COPYING
  */
@@ -22,6 +23,7 @@
 #include <linux/uaccess.h>
 #include <linux/highmem.h>
 #include <linux/cpu.h>
+#include <linux/bitops.h>
 
 #include <asm/apic.h>
 #include <asm/stacktrace.h>
@@ -68,26 +70,47 @@ struct debug_store {
 	u64	pebs_event_reset[MAX_PEBS_EVENTS];
 };
 
+struct event_constraint {
+	union {
+		unsigned long	idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+		u64		idxmsk64[1];
+	};
+	int	code;
+	int	cmask;
+	int	weight;
+};
+
 struct cpu_hw_events {
-	struct perf_event	*events[X86_PMC_IDX_MAX];
-	unsigned long		used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	struct perf_event	*events[X86_PMC_IDX_MAX]; /* in counter order */
 	unsigned long		active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
 	unsigned long		interrupts;
 	int			enabled;
 	struct debug_store	*ds;
-};
 
-struct event_constraint {
-	unsigned long	idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
-	int		code;
+	int			n_events;
+	int			n_added;
+	int			assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
+	struct perf_event	*event_list[X86_PMC_IDX_MAX]; /* in enabled order */
 };
 
-#define EVENT_CONSTRAINT(c, m) { .code = (c), .idxmsk[0] = (m) }
-#define EVENT_CONSTRAINT_END  { .code = 0, .idxmsk[0] = 0 }
+#define EVENT_CONSTRAINT(c, n, m) {	\
+	{ .idxmsk64[0] = (n) },		\
+	.code = (c),			\
+	.cmask = (m),			\
+	.weight = HWEIGHT64((u64)(n)),	\
+}
+
+#define INTEL_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVTSEL_MASK)
+
+#define FIXED_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, INTEL_ARCH_FIXED_MASK)
 
-#define for_each_event_constraint(e, c) \
-	for ((e) = (c); (e)->idxmsk[0]; (e)++)
+#define EVENT_CONSTRAINT_END		\
+	EVENT_CONSTRAINT(0, 0, 0)
 
+#define for_each_event_constraint(e, c)	\
+	for ((e) = (c); (e)->cmask; (e)++)
 
 /*
  * struct x86_pmu - generic x86 pmu
@@ -114,8 +137,14 @@ struct x86_pmu {
 	u64		intel_ctrl;
 	void		(*enable_bts)(u64 config);
 	void		(*disable_bts)(void);
-	int		(*get_event_idx)(struct cpu_hw_events *cpuc,
-					 struct hw_perf_event *hwc);
+
+	struct event_constraint *
+			(*get_event_constraints)(struct cpu_hw_events *cpuc,
+						 struct perf_event *event);
+
+	void		(*put_event_constraints)(struct cpu_hw_events *cpuc,
+						 struct perf_event *event);
+	struct event_constraint *event_constraints;
 };
 
 static struct x86_pmu x86_pmu __read_mostly;
@@ -124,7 +153,8 @@ static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
 	.enabled = 1,
 };
 
-static const struct event_constraint *event_constraints;
+static int x86_perf_event_set_period(struct perf_event *event,
+			     struct hw_perf_event *hwc, int idx);
 
 /*
  * Not sure about some of these
@@ -171,14 +201,14 @@ static u64 p6_pmu_raw_event(u64 hw_event)
 	return hw_event & P6_EVNTSEL_MASK;
 }
 
-static const struct event_constraint intel_p6_event_constraints[] =
+static struct event_constraint intel_p6_event_constraints[] =
 {
-	EVENT_CONSTRAINT(0xc1, 0x1),	/* FLOPS */
-	EVENT_CONSTRAINT(0x10, 0x1),	/* FP_COMP_OPS_EXE */
-	EVENT_CONSTRAINT(0x11, 0x1),	/* FP_ASSIST */
-	EVENT_CONSTRAINT(0x12, 0x2),	/* MUL */
-	EVENT_CONSTRAINT(0x13, 0x2),	/* DIV */
-	EVENT_CONSTRAINT(0x14, 0x1),	/* CYCLES_DIV_BUSY */
+	INTEL_EVENT_CONSTRAINT(0xc1, 0x1),	/* FLOPS */
+	INTEL_EVENT_CONSTRAINT(0x10, 0x1),	/* FP_COMP_OPS_EXE */
+	INTEL_EVENT_CONSTRAINT(0x11, 0x1),	/* FP_ASSIST */
+	INTEL_EVENT_CONSTRAINT(0x12, 0x2),	/* MUL */
+	INTEL_EVENT_CONSTRAINT(0x13, 0x2),	/* DIV */
+	INTEL_EVENT_CONSTRAINT(0x14, 0x1),	/* CYCLES_DIV_BUSY */
 	EVENT_CONSTRAINT_END
 };
 
@@ -196,32 +226,51 @@ static const u64 intel_perfmon_event_map[] =
   [PERF_COUNT_HW_BUS_CYCLES]		= 0x013c,
 };
 
-static const struct event_constraint intel_core_event_constraints[] =
-{
-	EVENT_CONSTRAINT(0x10, 0x1),	/* FP_COMP_OPS_EXE */
-	EVENT_CONSTRAINT(0x11, 0x2),	/* FP_ASSIST */
-	EVENT_CONSTRAINT(0x12, 0x2),	/* MUL */
-	EVENT_CONSTRAINT(0x13, 0x2),	/* DIV */
-	EVENT_CONSTRAINT(0x14, 0x1),	/* CYCLES_DIV_BUSY */
-	EVENT_CONSTRAINT(0x18, 0x1),	/* IDLE_DURING_DIV */
-	EVENT_CONSTRAINT(0x19, 0x2),	/* DELAYED_BYPASS */
-	EVENT_CONSTRAINT(0xa1, 0x1),	/* RS_UOPS_DISPATCH_CYCLES */
-	EVENT_CONSTRAINT(0xcb, 0x1),	/* MEM_LOAD_RETIRED */
+static struct event_constraint intel_core_event_constraints[] =
+{
+	FIXED_EVENT_CONSTRAINT(0xc0, (0x3|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */
+	FIXED_EVENT_CONSTRAINT(0x3c, (0x3|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */
+	INTEL_EVENT_CONSTRAINT(0x10, 0x1), /* FP_COMP_OPS_EXE */
+	INTEL_EVENT_CONSTRAINT(0x11, 0x2), /* FP_ASSIST */
+	INTEL_EVENT_CONSTRAINT(0x12, 0x2), /* MUL */
+	INTEL_EVENT_CONSTRAINT(0x13, 0x2), /* DIV */
+	INTEL_EVENT_CONSTRAINT(0x14, 0x1), /* CYCLES_DIV_BUSY */
+	INTEL_EVENT_CONSTRAINT(0x18, 0x1), /* IDLE_DURING_DIV */
+	INTEL_EVENT_CONSTRAINT(0x19, 0x2), /* DELAYED_BYPASS */
+	INTEL_EVENT_CONSTRAINT(0xa1, 0x1), /* RS_UOPS_DISPATCH_CYCLES */
+	INTEL_EVENT_CONSTRAINT(0xcb, 0x1), /* MEM_LOAD_RETIRED */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_nehalem_event_constraints[] =
+{
+	FIXED_EVENT_CONSTRAINT(0xc0, (0xf|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */
+	FIXED_EVENT_CONSTRAINT(0x3c, (0xf|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */
+	INTEL_EVENT_CONSTRAINT(0x40, 0x3), /* L1D_CACHE_LD */
+	INTEL_EVENT_CONSTRAINT(0x41, 0x3), /* L1D_CACHE_ST */
+	INTEL_EVENT_CONSTRAINT(0x42, 0x3), /* L1D_CACHE_LOCK */
+	INTEL_EVENT_CONSTRAINT(0x43, 0x3), /* L1D_ALL_REF */
+	INTEL_EVENT_CONSTRAINT(0x48, 0x3), /* L1D_PEND_MISS */
+	INTEL_EVENT_CONSTRAINT(0x4e, 0x3), /* L1D_PREFETCH */
+	INTEL_EVENT_CONSTRAINT(0x51, 0x3), /* L1D */
+	INTEL_EVENT_CONSTRAINT(0x63, 0x3), /* CACHE_LOCK_CYCLES */
 	EVENT_CONSTRAINT_END
 };
 
-static const struct event_constraint intel_nehalem_event_constraints[] =
-{
-	EVENT_CONSTRAINT(0x40, 0x3),	/* L1D_CACHE_LD */
-	EVENT_CONSTRAINT(0x41, 0x3),	/* L1D_CACHE_ST */
-	EVENT_CONSTRAINT(0x42, 0x3),	/* L1D_CACHE_LOCK */
-	EVENT_CONSTRAINT(0x43, 0x3),	/* L1D_ALL_REF */
-	EVENT_CONSTRAINT(0x4e, 0x3),	/* L1D_PREFETCH */
-	EVENT_CONSTRAINT(0x4c, 0x3),	/* LOAD_HIT_PRE */
-	EVENT_CONSTRAINT(0x51, 0x3),	/* L1D */
-	EVENT_CONSTRAINT(0x52, 0x3),	/* L1D_CACHE_PREFETCH_LOCK_FB_HIT */
-	EVENT_CONSTRAINT(0x53, 0x3),	/* L1D_CACHE_LOCK_FB_HIT */
-	EVENT_CONSTRAINT(0xc5, 0x3),	/* CACHE_LOCK_CYCLES */
+static struct event_constraint intel_westmere_event_constraints[] =
+{
+	FIXED_EVENT_CONSTRAINT(0xc0, (0xf|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */
+	FIXED_EVENT_CONSTRAINT(0x3c, (0xf|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */
+	INTEL_EVENT_CONSTRAINT(0x51, 0x3), /* L1D */
+	INTEL_EVENT_CONSTRAINT(0x60, 0x1), /* OFFCORE_REQUESTS_OUTSTANDING */
+	INTEL_EVENT_CONSTRAINT(0x63, 0x3), /* CACHE_LOCK_CYCLES */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_gen_event_constraints[] =
+{
+	FIXED_EVENT_CONSTRAINT(0xc0, (0x3|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */
+	FIXED_EVENT_CONSTRAINT(0x3c, (0x3|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */
 	EVENT_CONSTRAINT_END
 };
 
@@ -245,6 +294,97 @@ static u64 __read_mostly hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX];
 
+static __initconst u64 westmere_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x010b, /* MEM_INST_RETIRED.LOADS       */
+		[ C(RESULT_MISS)   ] = 0x0151, /* L1D.REPL                     */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x020b, /* MEM_INST_RETURED.STORES      */
+		[ C(RESULT_MISS)   ] = 0x0251, /* L1D.M_REPL                   */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x014e, /* L1D_PREFETCH.REQUESTS        */
+		[ C(RESULT_MISS)   ] = 0x024e, /* L1D_PREFETCH.MISS            */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS                    */
+		[ C(RESULT_MISS)   ] = 0x0280, /* L1I.MISSES                   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0324, /* L2_RQSTS.LOADS               */
+		[ C(RESULT_MISS)   ] = 0x0224, /* L2_RQSTS.LD_MISS             */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0c24, /* L2_RQSTS.RFOS                */
+		[ C(RESULT_MISS)   ] = 0x0824, /* L2_RQSTS.RFO_MISS            */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f2e, /* LLC Reference                */
+		[ C(RESULT_MISS)   ] = 0x412e, /* LLC Misses                   */
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x010b, /* MEM_INST_RETIRED.LOADS       */
+		[ C(RESULT_MISS)   ] = 0x0108, /* DTLB_LOAD_MISSES.ANY         */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x020b, /* MEM_INST_RETURED.STORES      */
+		[ C(RESULT_MISS)   ] = 0x010c, /* MEM_STORE_RETIRED.DTLB_MISS  */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01c0, /* INST_RETIRED.ANY_P           */
+		[ C(RESULT_MISS)   ] = 0x0185, /* ITLB_MISSES.ANY              */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(BPU ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ALL_BRANCHES */
+		[ C(RESULT_MISS)   ] = 0x03e8, /* BPU_CLEARS.ANY               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
 static __initconst u64 nehalem_hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
@@ -527,11 +667,11 @@ static u64 intel_pmu_raw_event(u64 hw_event)
 #define CORE_EVNTSEL_REG_MASK		0xFF000000ULL
 
 #define CORE_EVNTSEL_MASK		\
-	(CORE_EVNTSEL_EVENT_MASK |	\
-	 CORE_EVNTSEL_UNIT_MASK  |	\
-	 CORE_EVNTSEL_EDGE_MASK  |	\
-	 CORE_EVNTSEL_INV_MASK  |	\
-	 CORE_EVNTSEL_REG_MASK)
+	(INTEL_ARCH_EVTSEL_MASK |	\
+	 INTEL_ARCH_UNIT_MASK   |	\
+	 INTEL_ARCH_EDGE_MASK   |	\
+	 INTEL_ARCH_INV_MASK    |	\
+	 INTEL_ARCH_CNT_MASK)
 
 	return hw_event & CORE_EVNTSEL_MASK;
 }
@@ -1058,15 +1198,8 @@ static int __hw_perf_event_init(struct perf_event *event)
 
 static void p6_pmu_disable_all(void)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 val;
 
-	if (!cpuc->enabled)
-		return;
-
-	cpuc->enabled = 0;
-	barrier();
-
 	/* p6 only has one enable register */
 	rdmsrl(MSR_P6_EVNTSEL0, val);
 	val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
@@ -1077,12 +1210,6 @@ static void intel_pmu_disable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
-	if (!cpuc->enabled)
-		return;
-
-	cpuc->enabled = 0;
-	barrier();
-
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
 
 	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask))
@@ -1094,17 +1221,6 @@ static void amd_pmu_disable_all(void)
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx;
 
-	if (!cpuc->enabled)
-		return;
-
-	cpuc->enabled = 0;
-	/*
-	 * ensure we write the disable before we start disabling the
-	 * events proper, so that amd_pmu_enable_event() does the
-	 * right thing.
-	 */
-	barrier();
-
 	for (idx = 0; idx < x86_pmu.num_events; idx++) {
 		u64 val;
 
@@ -1120,22 +1236,25 @@ static void amd_pmu_disable_all(void)
 
 void hw_perf_disable(void)
 {
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
 	if (!x86_pmu_initialized())
 		return;
-	return x86_pmu.disable_all();
+
+	if (!cpuc->enabled)
+		return;
+
+	cpuc->n_added = 0;
+	cpuc->enabled = 0;
+	barrier();
+
+	x86_pmu.disable_all();
 }
 
 static void p6_pmu_enable_all(void)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	unsigned long val;
 
-	if (cpuc->enabled)
-		return;
-
-	cpuc->enabled = 1;
-	barrier();
-
 	/* p6 only has one enable register */
 	rdmsrl(MSR_P6_EVNTSEL0, val);
 	val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
@@ -1146,12 +1265,6 @@ static void intel_pmu_enable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
-	if (cpuc->enabled)
-		return;
-
-	cpuc->enabled = 1;
-	barrier();
-
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
 
 	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
@@ -1170,12 +1283,6 @@ static void amd_pmu_enable_all(void)
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx;
 
-	if (cpuc->enabled)
-		return;
-
-	cpuc->enabled = 1;
-	barrier();
-
 	for (idx = 0; idx < x86_pmu.num_events; idx++) {
 		struct perf_event *event = cpuc->events[idx];
 		u64 val;
@@ -1189,10 +1296,239 @@ static void amd_pmu_enable_all(void)
 	}
 }
 
+static const struct pmu pmu;
+
+static inline int is_x86_event(struct perf_event *event)
+{
+	return event->pmu == &pmu;
+}
+
+static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
+{
+	struct event_constraint *c, *constraints[X86_PMC_IDX_MAX];
+	unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	int i, j, w, wmax, num = 0;
+	struct hw_perf_event *hwc;
+
+	bitmap_zero(used_mask, X86_PMC_IDX_MAX);
+
+	for (i = 0; i < n; i++) {
+		constraints[i] =
+		  x86_pmu.get_event_constraints(cpuc, cpuc->event_list[i]);
+	}
+
+	/*
+	 * fastpath, try to reuse previous register
+	 */
+	for (i = 0; i < n; i++) {
+		hwc = &cpuc->event_list[i]->hw;
+		c = constraints[i];
+
+		/* never assigned */
+		if (hwc->idx == -1)
+			break;
+
+		/* constraint still honored */
+		if (!test_bit(hwc->idx, c->idxmsk))
+			break;
+
+		/* not already used */
+		if (test_bit(hwc->idx, used_mask))
+			break;
+
+		set_bit(hwc->idx, used_mask);
+		if (assign)
+			assign[i] = hwc->idx;
+	}
+	if (i == n)
+		goto done;
+
+	/*
+	 * begin slow path
+	 */
+
+	bitmap_zero(used_mask, X86_PMC_IDX_MAX);
+
+	/*
+	 * weight = number of possible counters
+	 *
+	 * 1    = most constrained, only works on one counter
+	 * wmax = least constrained, works on any counter
+	 *
+	 * assign events to counters starting with most
+	 * constrained events.
+	 */
+	wmax = x86_pmu.num_events;
+
+	/*
+	 * when fixed event counters are present,
+	 * wmax is incremented by 1 to account
+	 * for one more choice
+	 */
+	if (x86_pmu.num_events_fixed)
+		wmax++;
+
+	for (w = 1, num = n; num && w <= wmax; w++) {
+		/* for each event */
+		for (i = 0; num && i < n; i++) {
+			c = constraints[i];
+			hwc = &cpuc->event_list[i]->hw;
+
+			if (c->weight != w)
+				continue;
+
+			for_each_bit(j, c->idxmsk, X86_PMC_IDX_MAX) {
+				if (!test_bit(j, used_mask))
+					break;
+			}
+
+			if (j == X86_PMC_IDX_MAX)
+				break;
+
+			set_bit(j, used_mask);
+
+			if (assign)
+				assign[i] = j;
+			num--;
+		}
+	}
+done:
+	/*
+	 * scheduling failed or is just a simulation,
+	 * free resources if necessary
+	 */
+	if (!assign || num) {
+		for (i = 0; i < n; i++) {
+			if (x86_pmu.put_event_constraints)
+				x86_pmu.put_event_constraints(cpuc, cpuc->event_list[i]);
+		}
+	}
+	return num ? -ENOSPC : 0;
+}
+
+/*
+ * dogrp: true if must collect siblings events (group)
+ * returns total number of events and error code
+ */
+static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, bool dogrp)
+{
+	struct perf_event *event;
+	int n, max_count;
+
+	max_count = x86_pmu.num_events + x86_pmu.num_events_fixed;
+
+	/* current number of events already accepted */
+	n = cpuc->n_events;
+
+	if (is_x86_event(leader)) {
+		if (n >= max_count)
+			return -ENOSPC;
+		cpuc->event_list[n] = leader;
+		n++;
+	}
+	if (!dogrp)
+		return n;
+
+	list_for_each_entry(event, &leader->sibling_list, group_entry) {
+		if (!is_x86_event(event) ||
+		    event->state <= PERF_EVENT_STATE_OFF)
+			continue;
+
+		if (n >= max_count)
+			return -ENOSPC;
+
+		cpuc->event_list[n] = event;
+		n++;
+	}
+	return n;
+}
+
+
+static inline void x86_assign_hw_event(struct perf_event *event,
+				struct hw_perf_event *hwc, int idx)
+{
+	hwc->idx = idx;
+
+	if (hwc->idx == X86_PMC_IDX_FIXED_BTS) {
+		hwc->config_base = 0;
+		hwc->event_base	= 0;
+	} else if (hwc->idx >= X86_PMC_IDX_FIXED) {
+		hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
+		/*
+		 * We set it so that event_base + idx in wrmsr/rdmsr maps to
+		 * MSR_ARCH_PERFMON_FIXED_CTR0 ... CTR2:
+		 */
+		hwc->event_base =
+			MSR_ARCH_PERFMON_FIXED_CTR0 - X86_PMC_IDX_FIXED;
+	} else {
+		hwc->config_base = x86_pmu.eventsel;
+		hwc->event_base  = x86_pmu.perfctr;
+	}
+}
+
+static void __x86_pmu_disable(struct perf_event *event, struct cpu_hw_events *cpuc);
+
 void hw_perf_enable(void)
 {
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct perf_event *event;
+	struct hw_perf_event *hwc;
+	int i;
+
 	if (!x86_pmu_initialized())
 		return;
+
+	if (cpuc->enabled)
+		return;
+
+	if (cpuc->n_added) {
+		/*
+		 * apply assignment obtained either from
+		 * hw_perf_group_sched_in() or x86_pmu_enable()
+		 *
+		 * step1: save events moving to new counters
+		 * step2: reprogram moved events into new counters
+		 */
+		for (i = 0; i < cpuc->n_events; i++) {
+
+			event = cpuc->event_list[i];
+			hwc = &event->hw;
+
+			if (hwc->idx == -1 || hwc->idx == cpuc->assign[i])
+				continue;
+
+			__x86_pmu_disable(event, cpuc);
+
+			hwc->idx = -1;
+		}
+
+		for (i = 0; i < cpuc->n_events; i++) {
+
+			event = cpuc->event_list[i];
+			hwc = &event->hw;
+
+			if (hwc->idx == -1) {
+				x86_assign_hw_event(event, hwc, cpuc->assign[i]);
+				x86_perf_event_set_period(event, hwc, hwc->idx);
+			}
+			/*
+			 * need to mark as active because x86_pmu_disable()
+			 * clear active_mask and eventsp[] yet it preserves
+			 * idx
+			 */
+			set_bit(hwc->idx, cpuc->active_mask);
+			cpuc->events[hwc->idx] = event;
+
+			x86_pmu.enable(hwc, hwc->idx);
+			perf_event_update_userpage(event);
+		}
+		cpuc->n_added = 0;
+		perf_events_lapic_init();
+	}
+
+	cpuc->enabled = 1;
+	barrier();
+
 	x86_pmu.enable_all();
 }
 
@@ -1398,148 +1734,40 @@ static void amd_pmu_enable_event(struct hw_perf_event *hwc, int idx)
 		x86_pmu_enable_event(hwc, idx);
 }
 
-static int fixed_mode_idx(struct hw_perf_event *hwc)
-{
-	unsigned int hw_event;
-
-	hw_event = hwc->config & ARCH_PERFMON_EVENT_MASK;
-
-	if (unlikely((hw_event ==
-		      x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) &&
-		     (hwc->sample_period == 1)))
-		return X86_PMC_IDX_FIXED_BTS;
-
-	if (!x86_pmu.num_events_fixed)
-		return -1;
-
-	/*
-	 * fixed counters do not take all possible filters
-	 */
-	if (hwc->config & ARCH_PERFMON_EVENT_FILTER_MASK)
-		return -1;
-
-	if (unlikely(hw_event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS)))
-		return X86_PMC_IDX_FIXED_INSTRUCTIONS;
-	if (unlikely(hw_event == x86_pmu.event_map(PERF_COUNT_HW_CPU_CYCLES)))
-		return X86_PMC_IDX_FIXED_CPU_CYCLES;
-	if (unlikely(hw_event == x86_pmu.event_map(PERF_COUNT_HW_BUS_CYCLES)))
-		return X86_PMC_IDX_FIXED_BUS_CYCLES;
-
-	return -1;
-}
-
-/*
- * generic counter allocator: get next free counter
- */
-static int
-gen_get_event_idx(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc)
-{
-	int idx;
-
-	idx = find_first_zero_bit(cpuc->used_mask, x86_pmu.num_events);
-	return idx == x86_pmu.num_events ? -1 : idx;
-}
-
 /*
- * intel-specific counter allocator: check event constraints
- */
-static int
-intel_get_event_idx(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc)
-{
-	const struct event_constraint *event_constraint;
-	int i, code;
-
-	if (!event_constraints)
-		goto skip;
-
-	code = hwc->config & CORE_EVNTSEL_EVENT_MASK;
-
-	for_each_event_constraint(event_constraint, event_constraints) {
-		if (code == event_constraint->code) {
-			for_each_bit(i, event_constraint->idxmsk, X86_PMC_IDX_MAX) {
-				if (!test_and_set_bit(i, cpuc->used_mask))
-					return i;
-			}
-			return -1;
-		}
-	}
-skip:
-	return gen_get_event_idx(cpuc, hwc);
-}
-
-static int
-x86_schedule_event(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc)
-{
-	int idx;
-
-	idx = fixed_mode_idx(hwc);
-	if (idx == X86_PMC_IDX_FIXED_BTS) {
-		/* BTS is already occupied. */
-		if (test_and_set_bit(idx, cpuc->used_mask))
-			return -EAGAIN;
-
-		hwc->config_base	= 0;
-		hwc->event_base		= 0;
-		hwc->idx		= idx;
-	} else if (idx >= 0) {
-		/*
-		 * Try to get the fixed event, if that is already taken
-		 * then try to get a generic event:
-		 */
-		if (test_and_set_bit(idx, cpuc->used_mask))
-			goto try_generic;
-
-		hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
-		/*
-		 * We set it so that event_base + idx in wrmsr/rdmsr maps to
-		 * MSR_ARCH_PERFMON_FIXED_CTR0 ... CTR2:
-		 */
-		hwc->event_base =
-			MSR_ARCH_PERFMON_FIXED_CTR0 - X86_PMC_IDX_FIXED;
-		hwc->idx = idx;
-	} else {
-		idx = hwc->idx;
-		/* Try to get the previous generic event again */
-		if (idx == -1 || test_and_set_bit(idx, cpuc->used_mask)) {
-try_generic:
-			idx = x86_pmu.get_event_idx(cpuc, hwc);
-			if (idx == -1)
-				return -EAGAIN;
-
-			set_bit(idx, cpuc->used_mask);
-			hwc->idx = idx;
-		}
-		hwc->config_base = x86_pmu.eventsel;
-		hwc->event_base  = x86_pmu.perfctr;
-	}
-
-	return idx;
-}
-
-/*
- * Find a PMC slot for the freshly enabled / scheduled in event:
+ * activate a single event
+ *
+ * The event is added to the group of enabled events
+ * but only if it can be scehduled with existing events.
+ *
+ * Called with PMU disabled. If successful and return value 1,
+ * then guaranteed to call perf_enable() and hw_perf_enable()
  */
 static int x86_pmu_enable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct hw_perf_event *hwc = &event->hw;
-	int idx;
-
-	idx = x86_schedule_event(cpuc, hwc);
-	if (idx < 0)
-		return idx;
-
-	perf_events_lapic_init();
+	struct hw_perf_event *hwc;
+	int assign[X86_PMC_IDX_MAX];
+	int n, n0, ret;
 
-	x86_pmu.disable(hwc, idx);
+	hwc = &event->hw;
 
-	cpuc->events[idx] = event;
-	set_bit(idx, cpuc->active_mask);
+	n0 = cpuc->n_events;
+	n = collect_events(cpuc, event, false);
+	if (n < 0)
+		return n;
 
-	x86_perf_event_set_period(event, hwc, idx);
-	x86_pmu.enable(hwc, idx);
+	ret = x86_schedule_events(cpuc, n, assign);
+	if (ret)
+		return ret;
+	/*
+	 * copy new assignment, now we know it is possible
+	 * will be used by hw_perf_enable()
+	 */
+	memcpy(cpuc->assign, assign, n*sizeof(int));
 
-	perf_event_update_userpage(event);
+	cpuc->n_events = n;
+	cpuc->n_added  = n - n0;
 
 	return 0;
 }
@@ -1583,7 +1811,7 @@ void perf_event_print_debug(void)
 		pr_info("CPU#%d: overflow:   %016llx\n", cpu, overflow);
 		pr_info("CPU#%d: fixed:      %016llx\n", cpu, fixed);
 	}
-	pr_info("CPU#%d: used:       %016llx\n", cpu, *(u64 *)cpuc->used_mask);
+	pr_info("CPU#%d: active:       %016llx\n", cpu, *(u64 *)cpuc->active_mask);
 
 	for (idx = 0; idx < x86_pmu.num_events; idx++) {
 		rdmsrl(x86_pmu.eventsel + idx, pmc_ctrl);
@@ -1667,9 +1895,8 @@ static void intel_pmu_drain_bts_buffer(struct cpu_hw_events *cpuc)
 	event->pending_kill = POLL_IN;
 }
 
-static void x86_pmu_disable(struct perf_event *event)
+static void __x86_pmu_disable(struct perf_event *event, struct cpu_hw_events *cpuc)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
 
@@ -1681,12 +1908,6 @@ static void x86_pmu_disable(struct perf_event *event)
 	x86_pmu.disable(hwc, idx);
 
 	/*
-	 * Make sure the cleared pointer becomes visible before we
-	 * (potentially) free the event:
-	 */
-	barrier();
-
-	/*
 	 * Drain the remaining delta count out of a event
 	 * that we are disabling:
 	 */
@@ -1697,8 +1918,28 @@ static void x86_pmu_disable(struct perf_event *event)
 		intel_pmu_drain_bts_buffer(cpuc);
 
 	cpuc->events[idx] = NULL;
-	clear_bit(idx, cpuc->used_mask);
+}
+
+static void x86_pmu_disable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int i;
+
+	__x86_pmu_disable(event, cpuc);
+
+	for (i = 0; i < cpuc->n_events; i++) {
+		if (event == cpuc->event_list[i]) {
 
+			if (x86_pmu.put_event_constraints)
+				x86_pmu.put_event_constraints(cpuc, event);
+
+			while (++i < cpuc->n_events)
+				cpuc->event_list[i-1] = cpuc->event_list[i];
+
+			--cpuc->n_events;
+			break;
+		}
+	}
 	perf_event_update_userpage(event);
 }
 
@@ -1969,6 +2210,162 @@ perf_event_nmi_handler(struct notifier_block *self,
 	return NOTIFY_STOP;
 }
 
+static struct event_constraint unconstrained;
+
+static struct event_constraint bts_constraint =
+	EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0);
+
+static struct event_constraint *
+intel_special_constraints(struct perf_event *event)
+{
+	unsigned int hw_event;
+
+	hw_event = event->hw.config & INTEL_ARCH_EVENT_MASK;
+
+	if (unlikely((hw_event ==
+		      x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) &&
+		     (event->hw.sample_period == 1))) {
+
+		return &bts_constraint;
+	}
+	return NULL;
+}
+
+static struct event_constraint *
+intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	c = intel_special_constraints(event);
+	if (c)
+		return c;
+
+	if (x86_pmu.event_constraints) {
+		for_each_event_constraint(c, x86_pmu.event_constraints) {
+			if ((event->hw.config & c->cmask) == c->code)
+				return c;
+		}
+	}
+
+	return &unconstrained;
+}
+
+static struct event_constraint *
+amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+{
+	return &unconstrained;
+}
+
+static int x86_event_sched_in(struct perf_event *event,
+			  struct perf_cpu_context *cpuctx, int cpu)
+{
+	int ret = 0;
+
+	event->state = PERF_EVENT_STATE_ACTIVE;
+	event->oncpu = cpu;
+	event->tstamp_running += event->ctx->time - event->tstamp_stopped;
+
+	if (!is_x86_event(event))
+		ret = event->pmu->enable(event);
+
+	if (!ret && !is_software_event(event))
+		cpuctx->active_oncpu++;
+
+	if (!ret && event->attr.exclusive)
+		cpuctx->exclusive = 1;
+
+	return ret;
+}
+
+static void x86_event_sched_out(struct perf_event *event,
+			    struct perf_cpu_context *cpuctx, int cpu)
+{
+	event->state = PERF_EVENT_STATE_INACTIVE;
+	event->oncpu = -1;
+
+	if (!is_x86_event(event))
+		event->pmu->disable(event);
+
+	event->tstamp_running -= event->ctx->time - event->tstamp_stopped;
+
+	if (!is_software_event(event))
+		cpuctx->active_oncpu--;
+
+	if (event->attr.exclusive || !cpuctx->active_oncpu)
+		cpuctx->exclusive = 0;
+}
+
+/*
+ * Called to enable a whole group of events.
+ * Returns 1 if the group was enabled, or -EAGAIN if it could not be.
+ * Assumes the caller has disabled interrupts and has
+ * frozen the PMU with hw_perf_save_disable.
+ *
+ * called with PMU disabled. If successful and return value 1,
+ * then guaranteed to call perf_enable() and hw_perf_enable()
+ */
+int hw_perf_group_sched_in(struct perf_event *leader,
+	       struct perf_cpu_context *cpuctx,
+	       struct perf_event_context *ctx, int cpu)
+{
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+	struct perf_event *sub;
+	int assign[X86_PMC_IDX_MAX];
+	int n0, n1, ret;
+
+	/* n0 = total number of events */
+	n0 = collect_events(cpuc, leader, true);
+	if (n0 < 0)
+		return n0;
+
+	ret = x86_schedule_events(cpuc, n0, assign);
+	if (ret)
+		return ret;
+
+	ret = x86_event_sched_in(leader, cpuctx, cpu);
+	if (ret)
+		return ret;
+
+	n1 = 1;
+	list_for_each_entry(sub, &leader->sibling_list, group_entry) {
+		if (sub->state > PERF_EVENT_STATE_OFF) {
+			ret = x86_event_sched_in(sub, cpuctx, cpu);
+			if (ret)
+				goto undo;
+			++n1;
+		}
+	}
+	/*
+	 * copy new assignment, now we know it is possible
+	 * will be used by hw_perf_enable()
+	 */
+	memcpy(cpuc->assign, assign, n0*sizeof(int));
+
+	cpuc->n_events  = n0;
+	cpuc->n_added   = n1;
+	ctx->nr_active += n1;
+
+	/*
+	 * 1 means successful and events are active
+	 * This is not quite true because we defer
+	 * actual activation until hw_perf_enable() but
+	 * this way we* ensure caller won't try to enable
+	 * individual events
+	 */
+	return 1;
+undo:
+	x86_event_sched_out(leader, cpuctx, cpu);
+	n0  = 1;
+	list_for_each_entry(sub, &leader->sibling_list, group_entry) {
+		if (sub->state == PERF_EVENT_STATE_ACTIVE) {
+			x86_event_sched_out(sub, cpuctx, cpu);
+			if (++n0 == n1)
+				break;
+		}
+	}
+	return ret;
+}
+
 static __read_mostly struct notifier_block perf_event_nmi_notifier = {
 	.notifier_call		= perf_event_nmi_handler,
 	.next			= NULL,
@@ -2000,7 +2397,8 @@ static __initconst struct x86_pmu p6_pmu = {
 	 */
 	.event_bits		= 32,
 	.event_mask		= (1ULL << 32) - 1,
-	.get_event_idx		= intel_get_event_idx,
+	.get_event_constraints	= intel_get_event_constraints,
+	.event_constraints	= intel_p6_event_constraints
 };
 
 static __initconst struct x86_pmu intel_pmu = {
@@ -2024,7 +2422,7 @@ static __initconst struct x86_pmu intel_pmu = {
 	.max_period		= (1ULL << 31) - 1,
 	.enable_bts		= intel_pmu_enable_bts,
 	.disable_bts		= intel_pmu_disable_bts,
-	.get_event_idx		= intel_get_event_idx,
+	.get_event_constraints	= intel_get_event_constraints
 };
 
 static __initconst struct x86_pmu amd_pmu = {
@@ -2045,7 +2443,7 @@ static __initconst struct x86_pmu amd_pmu = {
 	.apic			= 1,
 	/* use highest bit to detect overflow */
 	.max_period		= (1ULL << 47) - 1,
-	.get_event_idx		= gen_get_event_idx,
+	.get_event_constraints	= amd_get_event_constraints
 };
 
 static __init int p6_pmu_init(void)
@@ -2058,12 +2456,9 @@ static __init int p6_pmu_init(void)
 	case 7:
 	case 8:
 	case 11: /* Pentium III */
-		event_constraints = intel_p6_event_constraints;
-		break;
 	case 9:
 	case 13:
 		/* Pentium M */
-		event_constraints = intel_p6_event_constraints;
 		break;
 	default:
 		pr_cont("unsupported p6 CPU model %d ",
@@ -2128,23 +2523,40 @@ static __init int intel_pmu_init(void)
 		memcpy(hw_cache_event_ids, core2_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
+		x86_pmu.event_constraints = intel_core_event_constraints;
 		pr_cont("Core2 events, ");
-		event_constraints = intel_core_event_constraints;
 		break;
-	default:
-	case 26:
+
+	case 26: /* 45 nm nehalem, "Bloomfield" */
+	case 30: /* 45 nm nehalem, "Lynnfield" */
 		memcpy(hw_cache_event_ids, nehalem_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
-		event_constraints = intel_nehalem_event_constraints;
+		x86_pmu.event_constraints = intel_nehalem_event_constraints;
 		pr_cont("Nehalem/Corei7 events, ");
 		break;
 	case 28:
 		memcpy(hw_cache_event_ids, atom_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
+		x86_pmu.event_constraints = intel_gen_event_constraints;
 		pr_cont("Atom events, ");
 		break;
+
+	case 37: /* 32 nm nehalem, "Clarkdale" */
+	case 44: /* 32 nm nehalem, "Gulftown" */
+		memcpy(hw_cache_event_ids, westmere_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+
+		x86_pmu.event_constraints = intel_westmere_event_constraints;
+		pr_cont("Westmere events, ");
+		break;
+	default:
+		/*
+		 * default constraints for v2 and up
+		 */
+		x86_pmu.event_constraints = intel_gen_event_constraints;
+		pr_cont("generic architected perfmon, ");
 	}
 	return 0;
 }
@@ -2220,6 +2632,9 @@ void __init init_hw_perf_events(void)
 	perf_events_lapic_init();
 	register_die_notifier(&perf_event_nmi_notifier);
 
+	unconstrained = (struct event_constraint)
+		EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_events) - 1, 0);
+
 	pr_info("... version:                %d\n",     x86_pmu.version);
 	pr_info("... bit width:              %d\n",     x86_pmu.event_bits);
 	pr_info("... generic registers:      %d\n",     x86_pmu.num_events);
@@ -2241,46 +2656,73 @@ static const struct pmu pmu = {
 	.unthrottle	= x86_pmu_unthrottle,
 };
 
-static int
-validate_event(struct cpu_hw_events *cpuc, struct perf_event *event)
-{
-	struct hw_perf_event fake_event = event->hw;
-
-	if (event->pmu && event->pmu != &pmu)
-		return 0;
-
-	return x86_schedule_event(cpuc, &fake_event) >= 0;
-}
-
+/*
+ * validate a single event group
+ *
+ * validation include:
+ *	- check events are compatible which each other
+ *	- events do not compete for the same counter
+ *	- number of events <= number of counters
+ *
+ * validation ensures the group can be loaded onto the
+ * PMU if it was the only group available.
+ */
 static int validate_group(struct perf_event *event)
 {
-	struct perf_event *sibling, *leader = event->group_leader;
-	struct cpu_hw_events fake_pmu;
+	struct perf_event *leader = event->group_leader;
+	struct cpu_hw_events *fake_cpuc;
+	int ret, n;
 
-	memset(&fake_pmu, 0, sizeof(fake_pmu));
+	ret = -ENOMEM;
+	fake_cpuc = kmalloc(sizeof(*fake_cpuc), GFP_KERNEL | __GFP_ZERO);
+	if (!fake_cpuc)
+		goto out;
 
-	if (!validate_event(&fake_pmu, leader))
-		return -ENOSPC;
+	/*
+	 * the event is not yet connected with its
+	 * siblings therefore we must first collect
+	 * existing siblings, then add the new event
+	 * before we can simulate the scheduling
+	 */
+	ret = -ENOSPC;
+	n = collect_events(fake_cpuc, leader, true);
+	if (n < 0)
+		goto out_free;
 
-	list_for_each_entry(sibling, &leader->sibling_list, group_entry) {
-		if (!validate_event(&fake_pmu, sibling))
-			return -ENOSPC;
-	}
+	fake_cpuc->n_events = n;
+	n = collect_events(fake_cpuc, event, false);
+	if (n < 0)
+		goto out_free;
 
-	if (!validate_event(&fake_pmu, event))
-		return -ENOSPC;
+	fake_cpuc->n_events = n;
 
-	return 0;
+	ret = x86_schedule_events(fake_cpuc, n, NULL);
+
+out_free:
+	kfree(fake_cpuc);
+out:
+	return ret;
 }
 
 const struct pmu *hw_perf_event_init(struct perf_event *event)
 {
+	const struct pmu *tmp;
 	int err;
 
 	err = __hw_perf_event_init(event);
 	if (!err) {
+		/*
+		 * we temporarily connect event to its pmu
+		 * such that validate_group() can classify
+		 * it as an x86 event using is_x86_event()
+		 */
+		tmp = event->pmu;
+		event->pmu = &pmu;
+
 		if (event->group_leader != event)
 			err = validate_group(event);
+
+		event->pmu = tmp;
 	}
 	if (err) {
 		if (event->destroy)
@@ -2304,7 +2746,6 @@ void callchain_store(struct perf_callchain_entry *entry, u64 ip)
 
 static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
 static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry);
-static DEFINE_PER_CPU(int, in_ignored_frame);
 
 
 static void
@@ -2320,10 +2761,6 @@ static void backtrace_warning(void *data, char *msg)
 
 static int backtrace_stack(void *data, char *name)
 {
-	per_cpu(in_ignored_frame, smp_processor_id()) =
-			x86_is_stack_id(NMI_STACK, name) ||
-			x86_is_stack_id(DEBUG_STACK, name);
-
 	return 0;
 }
 
@@ -2331,9 +2768,6 @@ static void backtrace_address(void *data, unsigned long addr, int reliable)
 {
 	struct perf_callchain_entry *entry = data;
 
-	if (per_cpu(in_ignored_frame, smp_processor_id()))
-		return;
-
 	if (reliable)
 		callchain_store(entry, addr);
 }
@@ -2440,9 +2874,6 @@ perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
 
 	is_user = user_mode(regs);
 
-	if (!current || current->pid == 0)
-		return;
-
 	if (is_user && current->state != TASK_RUNNING)
 		return;
 
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index 898df9719afb..74f4e85a5727 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -115,17 +115,6 @@ int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
 
 	return !test_bit(counter, perfctr_nmi_owner);
 }
-
-/* checks the an msr for availability */
-int avail_to_resrv_perfctr_nmi(unsigned int msr)
-{
-	unsigned int counter;
-
-	counter = nmi_perfctr_msr_to_bit(msr);
-	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
-
-	return !test_bit(counter, perfctr_nmi_owner);
-}
 EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
 
 int reserve_perfctr_nmi(unsigned int msr)
diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c
index ae775ca47b25..11540a189d93 100644
--- a/arch/x86/kernel/dumpstack_32.c
+++ b/arch/x86/kernel/dumpstack_32.c
@@ -18,11 +18,6 @@
 
 #include "dumpstack.h"
 
-/* Just a stub for now */
-int x86_is_stack_id(int id, char *name)
-{
-	return 0;
-}
 
 void dump_trace(struct task_struct *task, struct pt_regs *regs,
 		unsigned long *stack, unsigned long bp,
diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c
index 0ad9597073f5..676bc051252e 100644
--- a/arch/x86/kernel/dumpstack_64.c
+++ b/arch/x86/kernel/dumpstack_64.c
@@ -33,11 +33,6 @@ static char x86_stack_ids[][8] = {
 #endif
 };
 
-int x86_is_stack_id(int id, char *name)
-{
-	return x86_stack_ids[id - 1] == name;
-}
-
 static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 					 unsigned *usedp, char **idp)
 {
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index a1a7876cadcb..a966b753e496 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -517,11 +517,19 @@ u64 __init e820_remove_range(u64 start, u64 size, unsigned old_type,
 			     int checktype)
 {
 	int i;
+	u64 end;
 	u64 real_removed_size = 0;
 
 	if (size > (ULLONG_MAX - start))
 		size = ULLONG_MAX - start;
 
+	end = start + size;
+	printk(KERN_DEBUG "e820 remove range: %016Lx - %016Lx ",
+		       (unsigned long long) start,
+		       (unsigned long long) end);
+	e820_print_type(old_type);
+	printk(KERN_CONT "\n");
+
 	for (i = 0; i < e820.nr_map; i++) {
 		struct e820entry *ei = &e820.map[i];
 		u64 final_start, final_end;
diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c
index 05d5fec64a94..ae90b4739435 100644
--- a/arch/x86/kernel/hw_breakpoint.c
+++ b/arch/x86/kernel/hw_breakpoint.c
@@ -502,8 +502,6 @@ static int __kprobes hw_breakpoint_handler(struct die_args *args)
 		rcu_read_lock();
 
 		bp = per_cpu(bp_per_reg[i], cpu);
-		if (bp)
-			rc = NOTIFY_DONE;
 		/*
 		 * Reset the 'i'th TRAP bit in dr6 to denote completion of
 		 * exception handling
@@ -522,7 +520,13 @@ static int __kprobes hw_breakpoint_handler(struct die_args *args)
 
 		rcu_read_unlock();
 	}
-	if (dr6 & (~DR_TRAP_BITS))
+	/*
+	 * Further processing in do_debug() is needed for a) user-space
+	 * breakpoints (to generate signals) and b) when the system has
+	 * taken exception due to multiple causes
+	 */
+	if ((current->thread.debugreg6 & DR_TRAP_BITS) ||
+	    (dr6 & (~DR_TRAP_BITS)))
 		rc = NOTIFY_DONE;
 
 	set_debugreg(dr7, 7);
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index f2f8540a7f3d..dffdf91c8370 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -224,6 +224,85 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
 	return ret;
 }
 
+int xstateregs_active(struct task_struct *target,
+		      const struct user_regset *regset)
+{
+	return (cpu_has_xsave && tsk_used_math(target)) ? xstate_size : 0;
+}
+
+int xstateregs_get(struct task_struct *target, const struct user_regset *regset,
+		unsigned int pos, unsigned int count,
+		void *kbuf, void __user *ubuf)
+{
+	int ret;
+
+	if (!cpu_has_xsave)
+		return -ENODEV;
+
+	ret = init_fpu(target);
+	if (ret)
+		return ret;
+
+	/*
+	 * First copy the fxsave bytes 0..463
+	 */
+	ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+				  &target->thread.xstate->xsave, 0,
+				  (sizeof(struct i387_fxsave_struct) -
+				   sizeof(xstate_fx_sw_bytes)));
+	/*
+	 * Copy the 48bytes defined by software
+	 */
+	ret |= user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+				   xstate_fx_sw_bytes,
+				   (sizeof(struct i387_fxsave_struct) -
+				    sizeof(xstate_fx_sw_bytes)),
+				   sizeof(struct i387_fxsave_struct));
+	/*
+	 * Copy the rest of xstate memory layout
+	 */
+	ret |= user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+				   &target->thread.xstate->xsave.xsave_hdr,
+				   sizeof(struct i387_fxsave_struct), -1);
+	return ret;
+}
+
+int xstateregs_set(struct task_struct *target, const struct user_regset *regset,
+		  unsigned int pos, unsigned int count,
+		  const void *kbuf, const void __user *ubuf)
+{
+	int ret;
+	struct xsave_hdr_struct *xsave_hdr =
+				&current->thread.xstate->xsave.xsave_hdr;
+
+
+	if (!cpu_has_xsave)
+		return -ENODEV;
+
+	ret = init_fpu(target);
+	if (ret)
+		return ret;
+
+	set_stopped_child_used_math(target);
+
+	ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+				 &target->thread.xstate->xsave, 0, -1);
+
+	/*
+	 * mxcsr reserved bits must be masked to zero for security reasons.
+	 */
+	target->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask;
+
+	xsave_hdr->xstate_bv &= pcntxt_mask;
+	/*
+	 * These bits must be zero.
+	 */
+	xsave_hdr->reserved1[0] = xsave_hdr->reserved1[1] = 0;
+
+
+	return ret;
+}
+
 #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
 
 /*
diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c
index d5932226614f..fce55d532631 100644
--- a/arch/x86/kernel/irqinit.c
+++ b/arch/x86/kernel/irqinit.c
@@ -84,24 +84,7 @@ static struct irqaction irq2 = {
 };
 
 DEFINE_PER_CPU(vector_irq_t, vector_irq) = {
-	[0 ... IRQ0_VECTOR - 1] = -1,
-	[IRQ0_VECTOR] = 0,
-	[IRQ1_VECTOR] = 1,
-	[IRQ2_VECTOR] = 2,
-	[IRQ3_VECTOR] = 3,
-	[IRQ4_VECTOR] = 4,
-	[IRQ5_VECTOR] = 5,
-	[IRQ6_VECTOR] = 6,
-	[IRQ7_VECTOR] = 7,
-	[IRQ8_VECTOR] = 8,
-	[IRQ9_VECTOR] = 9,
-	[IRQ10_VECTOR] = 10,
-	[IRQ11_VECTOR] = 11,
-	[IRQ12_VECTOR] = 12,
-	[IRQ13_VECTOR] = 13,
-	[IRQ14_VECTOR] = 14,
-	[IRQ15_VECTOR] = 15,
-	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
+	[0 ... NR_VECTORS - 1] = -1,
 };
 
 int vector_used_by_percpu_irq(unsigned int vector)
@@ -116,6 +99,9 @@ int vector_used_by_percpu_irq(unsigned int vector)
 	return 0;
 }
 
+/* Number of legacy interrupts */
+int nr_legacy_irqs __read_mostly = NR_IRQS_LEGACY;
+
 void __init init_ISA_irqs(void)
 {
 	int i;
@@ -142,6 +128,19 @@ void __init init_ISA_irqs(void)
 
 void __init init_IRQ(void)
 {
+	int i;
+
+	/*
+	 * On cpu 0, Assign IRQ0_VECTOR..IRQ15_VECTOR's to IRQ 0..15.
+	 * If these IRQ's are handled by legacy interrupt-controllers like PIC,
+	 * then this configuration will likely be static after the boot. If
+	 * these IRQ's are handled by more mordern controllers like IO-APIC,
+	 * then this vector space can be freed and re-used dynamically as the
+	 * irq's migrate etc.
+	 */
+	for (i = 0; i < nr_legacy_irqs; i++)
+		per_cpu(vector_irq, 0)[IRQ0_VECTOR + i] = i;
+
 	x86_init.irqs.intr_init();
 }
 
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index c9b3522b6b46..02d678065d7d 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -92,6 +92,13 @@ void exit_thread(void)
 	}
 }
 
+void show_regs(struct pt_regs *regs)
+{
+	show_registers(regs);
+	show_trace(NULL, regs, (unsigned long *)kernel_stack_pointer(regs),
+		   regs->bp);
+}
+
 void show_regs_common(void)
 {
 	const char *board, *product;
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 37ad1e046aae..f6c62667e30c 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -174,12 +174,6 @@ void __show_regs(struct pt_regs *regs, int all)
 			d6, d7);
 }
 
-void show_regs(struct pt_regs *regs)
-{
-	show_registers(regs);
-	show_trace(NULL, regs, &regs->sp, regs->bp);
-}
-
 void release_thread(struct task_struct *dead_task)
 {
 	BUG_ON(dead_task->mm);
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 41a26a82470a..fc6ad4b95c6b 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -211,12 +211,6 @@ void __show_regs(struct pt_regs *regs, int all)
 	printk(KERN_DEFAULT "DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7);
 }
 
-void show_regs(struct pt_regs *regs)
-{
-	show_registers(regs);
-	show_trace(NULL, regs, (void *)(regs + 1), regs->bp);
-}
-
 void release_thread(struct task_struct *dead_task)
 {
 	if (dead_task->mm) {
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 118428085ea2..aeac66324895 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -48,6 +48,7 @@ enum x86_regset {
 	REGSET_FP,
 	REGSET_XFP,
 	REGSET_IOPERM64 = REGSET_XFP,
+	REGSET_XSTATE,
 	REGSET_TLS,
 	REGSET_IOPERM32,
 };
@@ -140,30 +141,6 @@ static const int arg_offs_table[] = {
 #endif
 };
 
-/**
- * regs_get_argument_nth() - get Nth argument at function call
- * @regs:	pt_regs which contains registers at function entry.
- * @n:		argument number.
- *
- * regs_get_argument_nth() returns @n th argument of a function call.
- * Since usually the kernel stack will be changed right after function entry,
- * you must use this at function entry. If the @n th entry is NOT in the
- * kernel stack or pt_regs, this returns 0.
- */
-unsigned long regs_get_argument_nth(struct pt_regs *regs, unsigned int n)
-{
-	if (n < ARRAY_SIZE(arg_offs_table))
-		return *(unsigned long *)((char *)regs + arg_offs_table[n]);
-	else {
-		/*
-		 * The typical case: arg n is on the stack.
-		 * (Note: stack[0] = return address, so skip it)
-		 */
-		n -= ARRAY_SIZE(arg_offs_table);
-		return regs_get_kernel_stack_nth(regs, 1 + n);
-	}
-}
-
 /*
  * does not yet catch signals sent when the child dies.
  * in exit.c or in signal.c.
@@ -1232,6 +1209,20 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 					     0, sizeof(struct user_i387_struct),
 					     datap);
 
+	case PTRACE_GETXSTATEREGS:	/* Get the child extended state. */
+		return copy_regset_to_user(child,
+					   task_user_regset_view(current),
+					   REGSET_XSTATE,
+					   0, xstate_size,
+					   datap);
+
+	case PTRACE_SETXSTATEREGS:	/* Set the child extended state. */
+		return copy_regset_from_user(child,
+					     task_user_regset_view(current),
+					     REGSET_XSTATE,
+					     0, xstate_size,
+					     datap);
+
 #ifdef CONFIG_X86_32
 	case PTRACE_GETFPXREGS:	/* Get the child extended FPU state. */
 		return copy_regset_to_user(child, &user_x86_32_view,
@@ -1561,6 +1552,16 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 					     sizeof(struct user32_fxsr_struct),
 					     datap);
 
+	case PTRACE_GETXSTATEREGS:	/* Get the child extended state. */
+		return copy_regset_to_user(child, &user_x86_32_view,
+					   REGSET_XSTATE, 0, xstate_size,
+					   datap);
+
+	case PTRACE_SETXSTATEREGS:	/* Set the child extended state. */
+		return copy_regset_from_user(child, &user_x86_32_view,
+					     REGSET_XSTATE, 0, xstate_size,
+					     datap);
+
 	case PTRACE_GET_THREAD_AREA:
 	case PTRACE_SET_THREAD_AREA:
 #ifdef CONFIG_X86_PTRACE_BTS
@@ -1584,7 +1585,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 
 #ifdef CONFIG_X86_64
 
-static const struct user_regset x86_64_regsets[] = {
+static struct user_regset x86_64_regsets[] __read_mostly = {
 	[REGSET_GENERAL] = {
 		.core_note_type = NT_PRSTATUS,
 		.n = sizeof(struct user_regs_struct) / sizeof(long),
@@ -1597,6 +1598,12 @@ static const struct user_regset x86_64_regsets[] = {
 		.size = sizeof(long), .align = sizeof(long),
 		.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
 	},
+	[REGSET_XSTATE] = {
+		.core_note_type = NT_X86_XSTATE,
+		.size = sizeof(long), .align = sizeof(long),
+		.active = xstateregs_active, .get = xstateregs_get,
+		.set = xstateregs_set
+	},
 	[REGSET_IOPERM64] = {
 		.core_note_type = NT_386_IOPERM,
 		.n = IO_BITMAP_LONGS,
@@ -1622,7 +1629,7 @@ static const struct user_regset_view user_x86_64_view = {
 #endif	/* CONFIG_X86_64 */
 
 #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
-static const struct user_regset x86_32_regsets[] = {
+static struct user_regset x86_32_regsets[] __read_mostly = {
 	[REGSET_GENERAL] = {
 		.core_note_type = NT_PRSTATUS,
 		.n = sizeof(struct user_regs_struct32) / sizeof(u32),
@@ -1641,6 +1648,12 @@ static const struct user_regset x86_32_regsets[] = {
 		.size = sizeof(u32), .align = sizeof(u32),
 		.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
 	},
+	[REGSET_XSTATE] = {
+		.core_note_type = NT_X86_XSTATE,
+		.size = sizeof(u32), .align = sizeof(u32),
+		.active = xstateregs_active, .get = xstateregs_get,
+		.set = xstateregs_set
+	},
 	[REGSET_TLS] = {
 		.core_note_type = NT_386_TLS,
 		.n = GDT_ENTRY_TLS_ENTRIES, .bias = GDT_ENTRY_TLS_MIN,
@@ -1663,6 +1676,18 @@ static const struct user_regset_view user_x86_32_view = {
 };
 #endif
 
+u64 xstate_fx_sw_bytes[6];
+void update_regset_xstate_info(unsigned int size, u64 xstate_mask)
+{
+#ifdef CONFIG_X86_64
+	x86_64_regsets[REGSET_XSTATE].n = size / sizeof(long);
+#endif
+#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
+	x86_32_regsets[REGSET_XSTATE].n = size / sizeof(u32);
+#endif
+	xstate_fx_sw_bytes[0] = xstate_mask;
+}
+
 const struct user_regset_view *task_user_regset_view(struct task_struct *task)
 {
 #ifdef CONFIG_IA32_EMULATION
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 5d9e40c58628..3499b4fabc94 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -667,6 +667,23 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = {
 	{}
 };
 
+static void __init trim_bios_range(void)
+{
+	/*
+	 * A special case is the first 4Kb of memory;
+	 * This is a BIOS owned area, not kernel ram, but generally
+	 * not listed as such in the E820 table.
+	 */
+	e820_update_range(0, PAGE_SIZE, E820_RAM, E820_RESERVED);
+	/*
+	 * special case: Some BIOSen report the PC BIOS
+	 * area (640->1Mb) as ram even though it is not.
+	 * take them out.
+	 */
+	e820_remove_range(BIOS_BEGIN, BIOS_END - BIOS_BEGIN, E820_RAM, 1);
+	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+}
+
 /*
  * Determine if we were loaded by an EFI loader.  If so, then we have also been
  * passed the efi memmap, systab, etc., so we should use these data structures
@@ -830,7 +847,7 @@ void __init setup_arch(char **cmdline_p)
 	insert_resource(&iomem_resource, &data_resource);
 	insert_resource(&iomem_resource, &bss_resource);
 
-
+	trim_bios_range();
 #ifdef CONFIG_X86_32
 	if (ppro_with_ram_bug()) {
 		e820_update_range(0x70000000ULL, 0x40000ULL, E820_RAM,
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 678d0b8c26f3..b2ebcba729d9 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -241,6 +241,11 @@ static void __cpuinit smp_callin(void)
 	map_cpu_to_logical_apicid();
 
 	notify_cpu_starting(cpuid);
+
+	/*
+	 * Need to setup vector mappings before we enable interrupts.
+	 */
+	__setup_vector_irq(smp_processor_id());
 	/*
 	 * Get our bogomips.
 	 *
@@ -315,7 +320,6 @@ notrace static void __cpuinit start_secondary(void *unused)
 	 */
 	ipi_call_lock();
 	lock_vector_lock();
-	__setup_vector_irq(smp_processor_id());
 	set_cpu_online(smp_processor_id(), true);
 	unlock_vector_lock();
 	ipi_call_unlock();
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 6ae2122467dc..7296bd231490 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -540,6 +540,9 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
 
 	get_debugreg(dr6, 6);
 
+	/* Filter out all the reserved bits which are preset to 1 */
+	dr6 &= ~DR6_RESERVED;
+
 	/* Catch kmemcheck conditions first of all! */
 	if ((dr6 & DR_STEP) && kmemcheck_trap(regs))
 		return;
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 597683aa5ba0..23066ecf12fa 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -806,7 +806,7 @@ static void __init check_system_tsc_reliable(void)
 	unsigned long res_low, res_high;
 
 	rdmsr_safe(MSR_GEODE_BUSCONT_CONF0, &res_low, &res_high);
-	/* Geode_LX - the OLPC CPU has a possibly a very reliable TSC */
+	/* Geode_LX - the OLPC CPU has a very reliable TSC */
 	if (res_low & RTSC_SUSP)
 		tsc_clocksource_reliable = 1;
 #endif
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
index 25bbb9bfc312..5e1ff66ecd73 100644
--- a/arch/x86/kernel/vmiclock_32.c
+++ b/arch/x86/kernel/vmiclock_32.c
@@ -79,11 +79,7 @@ unsigned long vmi_tsc_khz(void)
 
 static inline unsigned int vmi_get_timer_vector(void)
 {
-#ifdef CONFIG_X86_IO_APIC
-	return FIRST_DEVICE_VECTOR;
-#else
-	return FIRST_EXTERNAL_VECTOR;
-#endif
+	return IRQ0_VECTOR;
 }
 
 /** vmi clockchip */
diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c
index 619f7f88b8cc..693920b22496 100644
--- a/arch/x86/kernel/x8664_ksyms_64.c
+++ b/arch/x86/kernel/x8664_ksyms_64.c
@@ -26,7 +26,8 @@ EXPORT_SYMBOL(__put_user_2);
 EXPORT_SYMBOL(__put_user_4);
 EXPORT_SYMBOL(__put_user_8);
 
-EXPORT_SYMBOL(copy_user_generic);
+EXPORT_SYMBOL(copy_user_generic_string);
+EXPORT_SYMBOL(copy_user_generic_unrolled);
 EXPORT_SYMBOL(__copy_user_nocache);
 EXPORT_SYMBOL(_copy_from_user);
 EXPORT_SYMBOL(_copy_to_user);
diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c
index c5ee17e8c6d9..782c3a362ec6 100644
--- a/arch/x86/kernel/xsave.c
+++ b/arch/x86/kernel/xsave.c
@@ -337,6 +337,7 @@ void __ref xsave_cntxt_init(void)
 	cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx);
 	xstate_size = ebx;
 
+	update_regset_xstate_info(xstate_size, pcntxt_mask);
 	prepare_fx_sw_frame();
 
 	setup_xstate_init();
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index cffd754f3039..ddef409c2816 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -14,7 +14,7 @@ $(obj)/inat.o: $(obj)/inat-tables.c
 
 clean-files := inat-tables.c
 
-obj-$(CONFIG_SMP) += msr-smp.o
+obj-$(CONFIG_SMP) += msr-smp.o cache-smp.o
 
 lib-y := delay.o
 lib-y += thunk_$(BITS).o
@@ -39,4 +39,5 @@ else
         lib-y += thunk_64.o clear_page_64.o copy_page_64.o
         lib-y += memmove_64.o memset_64.o
         lib-y += copy_user_64.o rwlock_64.o copy_user_nocache_64.o
+	lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem_64.o
 endif
diff --git a/arch/x86/lib/cache-smp.c b/arch/x86/lib/cache-smp.c
new file mode 100644
index 000000000000..a3c668875038
--- /dev/null
+++ b/arch/x86/lib/cache-smp.c
@@ -0,0 +1,19 @@
+#include <linux/smp.h>
+#include <linux/module.h>
+
+static void __wbinvd(void *dummy)
+{
+	wbinvd();
+}
+
+void wbinvd_on_cpu(int cpu)
+{
+	smp_call_function_single(cpu, __wbinvd, NULL, 1);
+}
+EXPORT_SYMBOL(wbinvd_on_cpu);
+
+int wbinvd_on_all_cpus(void)
+{
+	return on_each_cpu(__wbinvd, NULL, 1);
+}
+EXPORT_SYMBOL(wbinvd_on_all_cpus);
diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S
index cf889d4e076a..71100c98e337 100644
--- a/arch/x86/lib/copy_user_64.S
+++ b/arch/x86/lib/copy_user_64.S
@@ -90,12 +90,6 @@ ENTRY(_copy_from_user)
 	CFI_ENDPROC
 ENDPROC(_copy_from_user)
 
-ENTRY(copy_user_generic)
-	CFI_STARTPROC
-	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
-	CFI_ENDPROC
-ENDPROC(copy_user_generic)
-
 	.section .fixup,"ax"
 	/* must zero dest */
 ENTRY(bad_from_user)
diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S
index ad5441ed1b57..f82e884928af 100644
--- a/arch/x86/lib/memcpy_64.S
+++ b/arch/x86/lib/memcpy_64.S
@@ -20,12 +20,11 @@
 /*
  * memcpy_c() - fast string ops (REP MOVSQ) based variant.
  *
- * Calls to this get patched into the kernel image via the
+ * This gets patched over the unrolled variant (below) via the
  * alternative instructions framework:
  */
-	ALIGN
-memcpy_c:
-	CFI_STARTPROC
+	.section .altinstr_replacement, "ax", @progbits
+.Lmemcpy_c:
 	movq %rdi, %rax
 
 	movl %edx, %ecx
@@ -35,8 +34,8 @@ memcpy_c:
 	movl %edx, %ecx
 	rep movsb
 	ret
-	CFI_ENDPROC
-ENDPROC(memcpy_c)
+.Lmemcpy_e:
+	.previous
 
 ENTRY(__memcpy)
 ENTRY(memcpy)
@@ -128,16 +127,10 @@ ENDPROC(__memcpy)
 	 * It is also a lot simpler. Use this when possible:
 	 */
 
-	.section .altinstr_replacement, "ax"
-1:	.byte 0xeb				/* jmp <disp8> */
-	.byte (memcpy_c - memcpy) - (2f - 1b)	/* offset */
-2:
-	.previous
-
 	.section .altinstructions, "a"
 	.align 8
 	.quad memcpy
-	.quad 1b
+	.quad .Lmemcpy_c
 	.byte X86_FEATURE_REP_GOOD
 
 	/*
@@ -145,6 +138,6 @@ ENDPROC(__memcpy)
 	 * so it is silly to overwrite itself with nops - reboot is the
 	 * only outcome...
 	 */
-	.byte 2b - 1b
-	.byte 2b - 1b
+	.byte .Lmemcpy_e - .Lmemcpy_c
+	.byte .Lmemcpy_e - .Lmemcpy_c
 	.previous
diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S
index 2c5948116bd2..e88d3b81644a 100644
--- a/arch/x86/lib/memset_64.S
+++ b/arch/x86/lib/memset_64.S
@@ -12,9 +12,8 @@
  * 
  * rax   original destination
  */	
-	ALIGN
-memset_c:
-	CFI_STARTPROC
+	.section .altinstr_replacement, "ax", @progbits
+.Lmemset_c:
 	movq %rdi,%r9
 	movl %edx,%r8d
 	andl $7,%r8d
@@ -29,8 +28,8 @@ memset_c:
 	rep stosb
 	movq %r9,%rax
 	ret
-	CFI_ENDPROC
-ENDPROC(memset_c)
+.Lmemset_e:
+	.previous
 
 ENTRY(memset)
 ENTRY(__memset)
@@ -118,16 +117,11 @@ ENDPROC(__memset)
 
 #include <asm/cpufeature.h>
 
-	.section .altinstr_replacement,"ax"
-1:	.byte 0xeb				/* jmp <disp8> */
-	.byte (memset_c - memset) - (2f - 1b)	/* offset */
-2:
-	.previous
 	.section .altinstructions,"a"
 	.align 8
 	.quad memset
-	.quad 1b
+	.quad .Lmemset_c
 	.byte X86_FEATURE_REP_GOOD
 	.byte .Lfinal - memset
-	.byte 2b - 1b
+	.byte .Lmemset_e - .Lmemset_c
 	.previous
diff --git a/arch/x86/lib/rwsem_64.S b/arch/x86/lib/rwsem_64.S
new file mode 100644
index 000000000000..15acecf0d7aa
--- /dev/null
+++ b/arch/x86/lib/rwsem_64.S
@@ -0,0 +1,81 @@
+/*
+ * x86-64 rwsem wrappers
+ *
+ * This interfaces the inline asm code to the slow-path
+ * C routines. We need to save the call-clobbered regs
+ * that the asm does not mark as clobbered, and move the
+ * argument from %rax to %rdi.
+ *
+ * NOTE! We don't need to save %rax, because the functions
+ * will always return the semaphore pointer in %rax (which
+ * is also the input argument to these helpers)
+ *
+ * The following can clobber %rdx because the asm clobbers it:
+ *   call_rwsem_down_write_failed
+ *   call_rwsem_wake
+ * but %rdi, %rsi, %rcx, %r8-r11 always need saving.
+ */
+
+#include <linux/linkage.h>
+#include <asm/rwlock.h>
+#include <asm/alternative-asm.h>
+#include <asm/frame.h>
+#include <asm/dwarf2.h>
+
+#define save_common_regs \
+	pushq %rdi; \
+	pushq %rsi; \
+	pushq %rcx; \
+	pushq %r8; \
+	pushq %r9; \
+	pushq %r10; \
+	pushq %r11
+
+#define restore_common_regs \
+	popq %r11; \
+	popq %r10; \
+	popq %r9; \
+	popq %r8; \
+	popq %rcx; \
+	popq %rsi; \
+	popq %rdi
+
+/* Fix up special calling conventions */
+ENTRY(call_rwsem_down_read_failed)
+	save_common_regs
+	pushq %rdx
+	movq %rax,%rdi
+	call rwsem_down_read_failed
+	popq %rdx
+	restore_common_regs
+	ret
+	ENDPROC(call_rwsem_down_read_failed)
+
+ENTRY(call_rwsem_down_write_failed)
+	save_common_regs
+	movq %rax,%rdi
+	call rwsem_down_write_failed
+	restore_common_regs
+	ret
+	ENDPROC(call_rwsem_down_write_failed)
+
+ENTRY(call_rwsem_wake)
+	decw %dx    /* do nothing if still outstanding active readers */
+	jnz 1f
+	save_common_regs
+	movq %rax,%rdi
+	call rwsem_wake
+	restore_common_regs
+1:	ret
+	ENDPROC(call_rwsem_wake)
+
+/* Fix up special calling conventions */
+ENTRY(call_rwsem_downgrade_wake)
+	save_common_regs
+	pushq %rdx
+	movq %rax,%rdi
+	call rwsem_downgrade_wake
+	popq %rdx
+	restore_common_regs
+	ret
+	ENDPROC(call_rwsem_downgrade_wake)
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index 71da1bca13cb..738e6593799d 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -18,7 +18,7 @@ static inline pte_t gup_get_pte(pte_t *ptep)
 #else
 	/*
 	 * With get_user_pages_fast, we walk down the pagetables without taking
-	 * any locks.  For this we would like to load the pointers atoimcally,
+	 * any locks.  For this we would like to load the pointers atomically,
 	 * but that is not possible (without expensive cmpxchg8b) on PAE.  What
 	 * we do have is the guarantee that a pte will only either go from not
 	 * present to present, or present to not present or both -- it will not
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index c246d259822d..5eb1ba74a3a9 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -24,43 +24,6 @@
 
 #include "physaddr.h"
 
-int page_is_ram(unsigned long pagenr)
-{
-	resource_size_t addr, end;
-	int i;
-
-	/*
-	 * A special case is the first 4Kb of memory;
-	 * This is a BIOS owned area, not kernel ram, but generally
-	 * not listed as such in the E820 table.
-	 */
-	if (pagenr == 0)
-		return 0;
-
-	/*
-	 * Second special case: Some BIOSen report the PC BIOS
-	 * area (640->1Mb) as ram even though it is not.
-	 */
-	if (pagenr >= (BIOS_BEGIN >> PAGE_SHIFT) &&
-		    pagenr < (BIOS_END >> PAGE_SHIFT))
-		return 0;
-
-	for (i = 0; i < e820.nr_map; i++) {
-		/*
-		 * Not usable memory:
-		 */
-		if (e820.map[i].type != E820_RAM)
-			continue;
-		addr = (e820.map[i].addr + PAGE_SIZE-1) >> PAGE_SHIFT;
-		end = (e820.map[i].addr + e820.map[i].size) >> PAGE_SHIFT;
-
-
-		if ((pagenr >= addr) && (pagenr < end))
-			return 1;
-	}
-	return 0;
-}
-
 /*
  * Fix up the linear direct mapping of the kernel to avoid cache attribute
  * conflicts.
@@ -422,6 +385,10 @@ void __init early_ioremap_init(void)
 	 * The boot-ioremap range spans multiple pmds, for which
 	 * we are not prepared:
 	 */
+#define __FIXADDR_TOP (-PAGE_SIZE)
+	BUILD_BUG_ON((__fix_to_virt(FIX_BTMAP_BEGIN) >> PMD_SHIFT)
+		     != (__fix_to_virt(FIX_BTMAP_END) >> PMD_SHIFT));
+#undef __FIXADDR_TOP
 	if (pmd != early_ioremap_pmd(fix_to_virt(FIX_BTMAP_END))) {
 		WARN_ON(1);
 		printk(KERN_WARNING "pmd %p != %p\n",
diff --git a/arch/x86/mm/mmap.c b/arch/x86/mm/mmap.c
index c8191defc38a..1dab5194fd9d 100644
--- a/arch/x86/mm/mmap.c
+++ b/arch/x86/mm/mmap.c
@@ -71,7 +71,7 @@ static int mmap_is_legacy(void)
 	if (current->personality & ADDR_COMPAT_LAYOUT)
 		return 1;
 
-	if (current->signal->rlim[RLIMIT_STACK].rlim_cur == RLIM_INFINITY)
+	if (rlimit(RLIMIT_STACK) == RLIM_INFINITY)
 		return 1;
 
 	return sysctl_legacy_va_layout;
@@ -96,7 +96,7 @@ static unsigned long mmap_rnd(void)
 
 static unsigned long mmap_base(void)
 {
-	unsigned long gap = current->signal->rlim[RLIMIT_STACK].rlim_cur;
+	unsigned long gap = rlimit(RLIMIT_STACK);
 
 	if (gap < MIN_GAP)
 		gap = MIN_GAP;
diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c
index 3999a5f25f38..8a713f1e9653 100644
--- a/drivers/char/agp/intel-agp.c
+++ b/drivers/char/agp/intel-agp.c
@@ -8,6 +8,7 @@
 #include <linux/kernel.h>
 #include <linux/pagemap.h>
 #include <linux/agp_backend.h>
+#include <asm/smp.h>
 #include "agp.h"
 
 /*
@@ -815,12 +816,6 @@ static void intel_i830_setup_flush(void)
 		intel_i830_fini_flush();
 }
 
-static void
-do_wbinvd(void *null)
-{
-	wbinvd();
-}
-
 /* The chipset_flush interface needs to get data that has already been
  * flushed out of the CPU all the way out to main memory, because the GPU
  * doesn't snoop those buffers.
@@ -837,12 +832,10 @@ static void intel_i830_chipset_flush(struct agp_bridge_data *bridge)
 
 	memset(pg, 0, 1024);
 
-	if (cpu_has_clflush) {
+	if (cpu_has_clflush)
 		clflush_cache_range(pg, 1024);
-	} else {
-		if (on_each_cpu(do_wbinvd, NULL, 1) != 0)
-			printk(KERN_ERR "Timed out waiting for cache flush.\n");
-	}
+	else if (wbinvd_on_all_cpus() != 0)
+		printk(KERN_ERR "Timed out waiting for cache flush.\n");
 }
 
 /* The intel i830 automatically initializes the agp aperture during POST.
diff --git a/drivers/char/nvram.c b/drivers/char/nvram.c
index fdbcc9fd6d31..5eb83c3ca20d 100644
--- a/drivers/char/nvram.c
+++ b/drivers/char/nvram.c
@@ -336,14 +336,12 @@ static int nvram_ioctl(struct inode *inode, struct file *file,
 
 static int nvram_open(struct inode *inode, struct file *file)
 {
-	lock_kernel();
 	spin_lock(&nvram_state_lock);
 
 	if ((nvram_open_cnt && (file->f_flags & O_EXCL)) ||
 	    (nvram_open_mode & NVRAM_EXCL) ||
 	    ((file->f_mode & FMODE_WRITE) && (nvram_open_mode & NVRAM_WRITE))) {
 		spin_unlock(&nvram_state_lock);
-		unlock_kernel();
 		return -EBUSY;
 	}
 
@@ -354,7 +352,6 @@ static int nvram_open(struct inode *inode, struct file *file)
 	nvram_open_cnt++;
 
 	spin_unlock(&nvram_state_lock);
-	unlock_kernel();
 
 	return 0;
 }
diff --git a/include/linux/bitops.h b/include/linux/bitops.h
index c05a29cb9bb2..ba0fd1eb4af7 100644
--- a/include/linux/bitops.h
+++ b/include/linux/bitops.h
@@ -25,7 +25,7 @@
 static __inline__ int get_bitmask_order(unsigned int count)
 {
 	int order;
-	
+
 	order = fls(count);
 	return order;	/* We could be slightly more clever with -1 here... */
 }
@@ -33,7 +33,7 @@ static __inline__ int get_bitmask_order(unsigned int count)
 static __inline__ int get_count_order(unsigned int count)
 {
 	int order;
-	
+
 	order = fls(count) - 1;
 	if (count & (count - 1))
 		order++;
@@ -45,6 +45,20 @@ static inline unsigned long hweight_long(unsigned long w)
 	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
 }
 
+#define HWEIGHT8(w)			\
+      (	(!!((w) & (1ULL << 0))) +	\
+	(!!((w) & (1ULL << 1))) +	\
+	(!!((w) & (1ULL << 2))) +	\
+	(!!((w) & (1ULL << 3))) +	\
+	(!!((w) & (1ULL << 4))) +	\
+	(!!((w) & (1ULL << 5))) +	\
+	(!!((w) & (1ULL << 6))) +	\
+	(!!((w) & (1ULL << 7)))	)
+
+#define HWEIGHT16(w) (HWEIGHT8(w)  + HWEIGHT8(w >> 8))
+#define HWEIGHT32(w) (HWEIGHT16(w) + HWEIGHT16(w >> 16))
+#define HWEIGHT64(w) (HWEIGHT32(w) + HWEIGHT32(w >> 32))
+
 /**
  * rol32 - rotate a 32-bit value left
  * @word: value to rotate
diff --git a/include/linux/elf.h b/include/linux/elf.h
index 0cc4d55151b7..a8c4af073ce9 100644
--- a/include/linux/elf.h
+++ b/include/linux/elf.h
@@ -361,6 +361,7 @@ typedef struct elf64_shdr {
 #define NT_PPC_VSX	0x102		/* PowerPC VSX registers */
 #define NT_386_TLS	0x200		/* i386 TLS slots (struct user_desc) */
 #define NT_386_IOPERM	0x201		/* x86 io permission bitmap (1=deny) */
+#define NT_X86_XSTATE	0x202		/* x86 extended state using xsave */
 #define NT_S390_HIGH_GPRS	0x300	/* s390 upper register halves */
 
 
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 2233c98d80df..6b7c444ab8f6 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -5,6 +5,7 @@
 #include <linux/trace_seq.h>
 #include <linux/percpu.h>
 #include <linux/hardirq.h>
+#include <linux/perf_event.h>
 
 struct trace_array;
 struct tracer;
@@ -121,9 +122,8 @@ struct ftrace_event_call {
 	int			(*regfunc)(struct ftrace_event_call *);
 	void			(*unregfunc)(struct ftrace_event_call *);
 	int			id;
+	const char		*print_fmt;
 	int			(*raw_init)(struct ftrace_event_call *);
-	int			(*show_format)(struct ftrace_event_call *,
-					       struct trace_seq *);
 	int			(*define_fields)(struct ftrace_event_call *);
 	struct list_head	fields;
 	int			filter_active;
@@ -138,9 +138,6 @@ struct ftrace_event_call {
 
 #define FTRACE_MAX_PROFILE_SIZE	2048
 
-extern char *perf_trace_buf;
-extern char *perf_trace_buf_nmi;
-
 #define MAX_FILTER_PRED		32
 #define MAX_FILTER_STR_VAL	256	/* Should handle KSYM_SYMBOL_LEN */
 
@@ -188,13 +185,27 @@ do {									\
 		__trace_printk(ip, fmt, ##args);			\
 } while (0)
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 struct perf_event;
 extern int ftrace_profile_enable(int event_id);
 extern void ftrace_profile_disable(int event_id);
 extern int ftrace_profile_set_filter(struct perf_event *event, int event_id,
 				     char *filter_str);
 extern void ftrace_profile_free_filter(struct perf_event *event);
+extern void *
+ftrace_perf_buf_prepare(int size, unsigned short type, int *rctxp,
+			 unsigned long *irq_flags);
+
+static inline void
+ftrace_perf_buf_submit(void *raw_data, int size, int rctx, u64 addr,
+		       u64 count, unsigned long irq_flags)
+{
+	struct trace_entry *entry = raw_data;
+
+	perf_tp_event(entry->type, addr, count, raw_data, size);
+	perf_swevent_put_recursion_context(rctx);
+	local_irq_restore(irq_flags);
+}
 #endif
 
 #endif /* _LINUX_FTRACE_EVENT_H */
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 451481c082b5..d13492df57a1 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -90,7 +90,7 @@ struct msi_desc;
  * @startup:		start up the interrupt (defaults to ->enable if NULL)
  * @shutdown:		shut down the interrupt (defaults to ->disable if NULL)
  * @enable:		enable the interrupt (defaults to chip->unmask if NULL)
- * @disable:		disable the interrupt (defaults to chip->mask if NULL)
+ * @disable:		disable the interrupt
  * @ack:		start of a new interrupt
  * @mask:		mask an interrupt source
  * @mask_ack:		ack and mask an interrupt source
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index b632226ddc2f..c5cd65fea0a3 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -124,7 +124,7 @@ extern int _cond_resched(void);
 #endif
 
 #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP
-  void __might_sleep(char *file, int line, int preempt_offset);
+  void __might_sleep(const char *file, int line, int preempt_offset);
 /**
  * might_sleep - annotation for functions that can sleep
  *
@@ -138,7 +138,8 @@ extern int _cond_resched(void);
 # define might_sleep() \
 	do { __might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0)
 #else
-  static inline void __might_sleep(char *file, int line, int preempt_offset) { }
+  static inline void __might_sleep(const char *file, int line,
+				   int preempt_offset) { }
 # define might_sleep() do { might_resched(); } while (0)
 #endif
 
diff --git a/include/linux/list.h b/include/linux/list.h
index 969f6e92d089..5d9c6558e8ab 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -206,6 +206,20 @@ static inline int list_empty_careful(const struct list_head *head)
 }
 
 /**
+ * list_rotate_left - rotate the list to the left
+ * @head: the head of the list
+ */
+static inline void list_rotate_left(struct list_head *head)
+{
+	struct list_head *first;
+
+	if (!list_empty(head)) {
+		first = head->next;
+		list_move_tail(first, head);
+	}
+}
+
+/**
  * list_is_singular - tests whether a list has just one entry.
  * @head: the list to test.
  */
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 60c467bfbabd..8b2fa8593c61 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -265,6 +265,8 @@ static inline int get_page_unless_zero(struct page *page)
 	return atomic_inc_not_zero(&page->_count);
 }
 
+extern int page_is_ram(unsigned long pfn);
+
 /* Support for virtually mapped pages */
 struct page *vmalloc_to_page(const void *addr);
 unsigned long vmalloc_to_pfn(const void *addr);
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 8fa71874113f..556b0f4a668e 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -290,7 +290,7 @@ struct perf_event_mmap_page {
 };
 
 #define PERF_RECORD_MISC_CPUMODE_MASK		(3 << 0)
-#define PERF_RECORD_MISC_CPUMODE_UNKNOWN		(0 << 0)
+#define PERF_RECORD_MISC_CPUMODE_UNKNOWN	(0 << 0)
 #define PERF_RECORD_MISC_KERNEL			(1 << 0)
 #define PERF_RECORD_MISC_USER			(2 << 0)
 #define PERF_RECORD_MISC_HYPERVISOR		(3 << 0)
@@ -356,8 +356,8 @@ enum perf_event_type {
 	 *	u64				stream_id;
 	 * };
 	 */
-	PERF_RECORD_THROTTLE		= 5,
-	PERF_RECORD_UNTHROTTLE		= 6,
+	PERF_RECORD_THROTTLE			= 5,
+	PERF_RECORD_UNTHROTTLE			= 6,
 
 	/*
 	 * struct {
@@ -371,10 +371,10 @@ enum perf_event_type {
 
 	/*
 	 * struct {
-	 * 	struct perf_event_header	header;
-	 * 	u32				pid, tid;
+	 *	struct perf_event_header	header;
+	 *	u32				pid, tid;
 	 *
-	 * 	struct read_format		values;
+	 *	struct read_format		values;
 	 * };
 	 */
 	PERF_RECORD_READ			= 8,
@@ -412,7 +412,7 @@ enum perf_event_type {
 	 *	  char                  data[size];}&& PERF_SAMPLE_RAW
 	 * };
 	 */
-	PERF_RECORD_SAMPLE		= 9,
+	PERF_RECORD_SAMPLE			= 9,
 
 	PERF_RECORD_MAX,			/* non-ABI */
 };
@@ -498,9 +498,8 @@ struct hw_perf_event {
 	atomic64_t			period_left;
 	u64				interrupts;
 
-	u64				freq_count;
-	u64				freq_interrupts;
-	u64				freq_stamp;
+	u64				freq_time_stamp;
+	u64				freq_count_stamp;
 #endif
 };
 
@@ -565,6 +564,10 @@ typedef void (*perf_overflow_handler_t)(struct perf_event *, int,
 					struct perf_sample_data *,
 					struct pt_regs *regs);
 
+enum perf_group_flag {
+	PERF_GROUP_SOFTWARE = 0x1,
+};
+
 /**
  * struct perf_event - performance event kernel representation:
  */
@@ -574,6 +577,7 @@ struct perf_event {
 	struct list_head		event_entry;
 	struct list_head		sibling_list;
 	int				nr_siblings;
+	int				group_flags;
 	struct perf_event		*group_leader;
 	struct perf_event		*output;
 	const struct pmu		*pmu;
@@ -658,7 +662,7 @@ struct perf_event {
 
 	perf_overflow_handler_t		overflow_handler;
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_EVENT_TRACING
 	struct event_filter		*filter;
 #endif
 
@@ -683,7 +687,8 @@ struct perf_event_context {
 	 */
 	struct mutex			mutex;
 
-	struct list_head		group_list;
+	struct list_head		pinned_groups;
+	struct list_head		flexible_groups;
 	struct list_head		event_list;
 	int				nr_events;
 	int				nr_active;
@@ -746,10 +751,9 @@ extern int perf_max_events;
 
 extern const struct pmu *hw_perf_event_init(struct perf_event *event);
 
-extern void perf_event_task_sched_in(struct task_struct *task, int cpu);
-extern void perf_event_task_sched_out(struct task_struct *task,
-					struct task_struct *next, int cpu);
-extern void perf_event_task_tick(struct task_struct *task, int cpu);
+extern void perf_event_task_sched_in(struct task_struct *task);
+extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
+extern void perf_event_task_tick(struct task_struct *task);
 extern int perf_event_init_task(struct task_struct *child);
 extern void perf_event_exit_task(struct task_struct *child);
 extern void perf_event_free_task(struct task_struct *task);
@@ -853,8 +857,7 @@ extern int sysctl_perf_event_mlock;
 extern int sysctl_perf_event_sample_rate;
 
 extern void perf_event_init(void);
-extern void perf_tp_event(int event_id, u64 addr, u64 count,
-				 void *record, int entry_size);
+extern void perf_tp_event(int event_id, u64 addr, u64 count, void *record, int entry_size);
 extern void perf_bp_event(struct perf_event *event, void *data);
 
 #ifndef perf_misc_flags
@@ -875,12 +878,12 @@ extern void perf_event_enable(struct perf_event *event);
 extern void perf_event_disable(struct perf_event *event);
 #else
 static inline void
-perf_event_task_sched_in(struct task_struct *task, int cpu)		{ }
+perf_event_task_sched_in(struct task_struct *task)			{ }
 static inline void
 perf_event_task_sched_out(struct task_struct *task,
-			    struct task_struct *next, int cpu)		{ }
+			    struct task_struct *next)			{ }
 static inline void
-perf_event_task_tick(struct task_struct *task, int cpu)			{ }
+perf_event_task_tick(struct task_struct *task)				{ }
 static inline int perf_event_init_task(struct task_struct *child)	{ return 0; }
 static inline void perf_event_exit_task(struct task_struct *child)	{ }
 static inline void perf_event_free_task(struct task_struct *task)	{ }
@@ -895,13 +898,13 @@ static inline void
 perf_sw_event(u32 event_id, u64 nr, int nmi,
 		     struct pt_regs *regs, u64 addr)			{ }
 static inline void
-perf_bp_event(struct perf_event *event, void *data)		{ }
+perf_bp_event(struct perf_event *event, void *data)			{ }
 
 static inline void perf_event_mmap(struct vm_area_struct *vma)		{ }
 static inline void perf_event_comm(struct task_struct *tsk)		{ }
 static inline void perf_event_fork(struct task_struct *tsk)		{ }
 static inline void perf_event_init(void)				{ }
-static inline int  perf_swevent_get_recursion_context(void)  { return -1; }
+static inline int  perf_swevent_get_recursion_context(void)		{ return -1; }
 static inline void perf_swevent_put_recursion_context(int rctx)		{ }
 static inline void perf_event_enable(struct perf_event *event)		{ }
 static inline void perf_event_disable(struct perf_event *event)		{ }
diff --git a/include/linux/plist.h b/include/linux/plist.h
index 8227f717c70f..6898985e7b38 100644
--- a/include/linux/plist.h
+++ b/include/linux/plist.h
@@ -45,7 +45,7 @@
  * the insertion of new nodes. There are no nodes with duplicate
  * priorites on the list.
  *
- * The nodes on the node_list is ordered by priority and can contain
+ * The nodes on the node_list are ordered by priority and can contain
  * entries which have the same priority. Those entries are ordered
  * FIFO
  *
@@ -265,7 +265,7 @@ static inline int plist_node_empty(const struct plist_node *node)
  *
  * Assumes the plist is _not_ empty.
  */
-static inline struct plist_node* plist_first(const struct plist_head *head)
+static inline struct plist_node *plist_first(const struct plist_head *head)
 {
 	return list_entry(head->node_list.next,
 			  struct plist_node, plist.node_list);
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index 96cc307ed9f4..2b70d4e37383 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -62,6 +62,18 @@ static inline long rcu_batches_completed_bh(void)
 
 extern int rcu_expedited_torture_stats(char *page);
 
+static inline void rcu_force_quiescent_state(void)
+{
+}
+
+static inline void rcu_bh_force_quiescent_state(void)
+{
+}
+
+static inline void rcu_sched_force_quiescent_state(void)
+{
+}
+
 #define synchronize_rcu synchronize_sched
 
 static inline void synchronize_rcu_expedited(void)
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index 8044b1b94333..704a010f686c 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -99,6 +99,9 @@ extern void rcu_check_callbacks(int cpu, int user);
 extern long rcu_batches_completed(void);
 extern long rcu_batches_completed_bh(void);
 extern long rcu_batches_completed_sched(void);
+extern void rcu_force_quiescent_state(void);
+extern void rcu_bh_force_quiescent_state(void);
+extern void rcu_sched_force_quiescent_state(void);
 
 #ifdef CONFIG_NO_HZ
 void rcu_enter_nohz(void);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 78efe7c485ac..b1b8d84c5805 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -740,14 +740,6 @@ struct user_struct {
 	uid_t uid;
 	struct user_namespace *user_ns;
 
-#ifdef CONFIG_USER_SCHED
-	struct task_group *tg;
-#ifdef CONFIG_SYSFS
-	struct kobject kobj;
-	struct delayed_work work;
-#endif
-#endif
-
 #ifdef CONFIG_PERF_EVENTS
 	atomic_long_t locked_vm;
 #endif
@@ -1084,7 +1076,8 @@ struct sched_domain;
 struct sched_class {
 	const struct sched_class *next;
 
-	void (*enqueue_task) (struct rq *rq, struct task_struct *p, int wakeup);
+	void (*enqueue_task) (struct rq *rq, struct task_struct *p, int wakeup,
+			      bool head);
 	void (*dequeue_task) (struct rq *rq, struct task_struct *p, int sleep);
 	void (*yield_task) (struct rq *rq);
 
@@ -1096,14 +1089,6 @@ struct sched_class {
 #ifdef CONFIG_SMP
 	int  (*select_task_rq)(struct task_struct *p, int sd_flag, int flags);
 
-	unsigned long (*load_balance) (struct rq *this_rq, int this_cpu,
-			struct rq *busiest, unsigned long max_load_move,
-			struct sched_domain *sd, enum cpu_idle_type idle,
-			int *all_pinned, int *this_best_prio);
-
-	int (*move_one_task) (struct rq *this_rq, int this_cpu,
-			      struct rq *busiest, struct sched_domain *sd,
-			      enum cpu_idle_type idle);
 	void (*pre_schedule) (struct rq *this_rq, struct task_struct *task);
 	void (*post_schedule) (struct rq *this_rq);
 	void (*task_waking) (struct rq *this_rq, struct task_struct *task);
@@ -2517,13 +2502,9 @@ extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
 
 extern void normalize_rt_tasks(void);
 
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
 
 extern struct task_group init_task_group;
-#ifdef CONFIG_USER_SCHED
-extern struct task_group root_task_group;
-extern void set_tg_uid(struct user_struct *user);
-#endif
 
 extern struct task_group *sched_create_group(struct task_group *parent);
 extern void sched_destroy_group(struct task_group *tg);
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index e4b90e16ee1b..52fe33963e82 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -99,7 +99,7 @@ struct perf_event_attr;
 #define __SC_TEST5(t5, a5, ...)	__SC_TEST(t5); __SC_TEST4(__VA_ARGS__)
 #define __SC_TEST6(t6, a6, ...)	__SC_TEST(t6); __SC_TEST5(__VA_ARGS__)
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 
 #define TRACE_SYS_ENTER_PROFILE_INIT(sname)				       \
 	.profile_enable = prof_sysenter_enable,				       \
@@ -113,7 +113,7 @@ struct perf_event_attr;
 #define TRACE_SYS_ENTER_PROFILE_INIT(sname)
 #define TRACE_SYS_EXIT_PROFILE(sname)
 #define TRACE_SYS_EXIT_PROFILE_INIT(sname)
-#endif
+#endif /* CONFIG_PERF_EVENTS */
 
 #ifdef CONFIG_FTRACE_SYSCALLS
 #define __SC_STR_ADECL1(t, a)		#a
@@ -143,8 +143,7 @@ struct perf_event_attr;
 		.name                   = "sys_enter"#sname,		\
 		.system                 = "syscalls",			\
 		.event                  = &enter_syscall_print_##sname,	\
-		.raw_init		= trace_event_raw_init,		\
-		.show_format		= syscall_enter_format,		\
+		.raw_init		= init_syscall_trace,		\
 		.define_fields		= syscall_enter_define_fields,	\
 		.regfunc		= reg_event_syscall_enter,	\
 		.unregfunc		= unreg_event_syscall_enter,	\
@@ -165,8 +164,7 @@ struct perf_event_attr;
 		.name                   = "sys_exit"#sname,		\
 		.system                 = "syscalls",			\
 		.event                  = &exit_syscall_print_##sname,	\
-		.raw_init		= trace_event_raw_init,		\
-		.show_format		= syscall_exit_format,		\
+		.raw_init		= init_syscall_trace,		\
 		.define_fields		= syscall_exit_define_fields,	\
 		.regfunc		= reg_event_syscall_exit,	\
 		.unregfunc		= unreg_event_syscall_exit,	\
diff --git a/include/linux/timex.h b/include/linux/timex.h
index 94f8faecdcbc..7a082b32d8e1 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -238,9 +238,6 @@ extern int tickadj;			/* amount of adjustment per tick */
  * phase-lock loop variables
  */
 extern int time_status;		/* clock synchronization status bits */
-extern long time_maxerror;	/* maximum error */
-extern long time_esterror;	/* estimated error */
-
 extern long time_adjust;	/* The amount of adjtime left */
 
 extern void ntp_init(void);
diff --git a/include/trace/events/lock.h b/include/trace/events/lock.h
index a870ba125aa8..5c1dcfc16c60 100644
--- a/include/trace/events/lock.h
+++ b/include/trace/events/lock.h
@@ -20,14 +20,17 @@ TRACE_EVENT(lock_acquire,
 	TP_STRUCT__entry(
 		__field(unsigned int, flags)
 		__string(name, lock->name)
+		__field(void *, lockdep_addr)
 	),
 
 	TP_fast_assign(
 		__entry->flags = (trylock ? 1 : 0) | (read ? 2 : 0);
 		__assign_str(name, lock->name);
+		__entry->lockdep_addr = lock;
 	),
 
-	TP_printk("%s%s%s", (__entry->flags & 1) ? "try " : "",
+	TP_printk("%p %s%s%s", __entry->lockdep_addr,
+		  (__entry->flags & 1) ? "try " : "",
 		  (__entry->flags & 2) ? "read " : "",
 		  __get_str(name))
 );
@@ -40,13 +43,16 @@ TRACE_EVENT(lock_release,
 
 	TP_STRUCT__entry(
 		__string(name, lock->name)
+		__field(void *, lockdep_addr)
 	),
 
 	TP_fast_assign(
 		__assign_str(name, lock->name);
+		__entry->lockdep_addr = lock;
 	),
 
-	TP_printk("%s", __get_str(name))
+	TP_printk("%p %s",
+		  __entry->lockdep_addr, __get_str(name))
 );
 
 #ifdef CONFIG_LOCK_STAT
@@ -59,13 +65,16 @@ TRACE_EVENT(lock_contended,
 
 	TP_STRUCT__entry(
 		__string(name, lock->name)
+		__field(void *, lockdep_addr)
 	),
 
 	TP_fast_assign(
 		__assign_str(name, lock->name);
+		__entry->lockdep_addr = lock;
 	),
 
-	TP_printk("%s", __get_str(name))
+	TP_printk("%p %s",
+		  __entry->lockdep_addr, __get_str(name))
 );
 
 TRACE_EVENT(lock_acquired,
@@ -75,16 +84,18 @@ TRACE_EVENT(lock_acquired,
 
 	TP_STRUCT__entry(
 		__string(name, lock->name)
-		__field(unsigned long, wait_usec)
-		__field(unsigned long, wait_nsec_rem)
+		__field(s64, wait_nsec)
+		__field(void *, lockdep_addr)
 	),
+
 	TP_fast_assign(
 		__assign_str(name, lock->name);
-		__entry->wait_nsec_rem = do_div(waittime, NSEC_PER_USEC);
-		__entry->wait_usec = (unsigned long) waittime;
+		__entry->wait_nsec = waittime;
+		__entry->lockdep_addr = lock;
 	),
-	TP_printk("%s (%lu.%03lu us)", __get_str(name), __entry->wait_usec,
-				       __entry->wait_nsec_rem)
+	TP_printk("%p %s (%llu ns)", __entry->lockdep_addr,
+		  __get_str(name),
+		  __entry->wait_nsec)
 );
 
 #endif
diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h
index c6fe03e902ca..fb2c5bd457d7 100644
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
@@ -131,130 +131,6 @@
 #include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
 
 /*
- * Setup the showing format of trace point.
- *
- * int
- * ftrace_format_##call(struct trace_seq *s)
- * {
- *	struct ftrace_raw_##call field;
- *	int ret;
- *
- *	ret = trace_seq_printf(s, #type " " #item ";"
- *			       " offset:%u; size:%u;\n",
- *			       offsetof(struct ftrace_raw_##call, item),
- *			       sizeof(field.type));
- *
- * }
- */
-
-#undef TP_STRUCT__entry
-#define TP_STRUCT__entry(args...) args
-
-#undef __field
-#define __field(type, item)					\
-	ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t"	\
-			       "offset:%u;\tsize:%u;\tsigned:%u;\n",	\
-			       (unsigned int)offsetof(typeof(field), item), \
-			       (unsigned int)sizeof(field.item),	\
-			       (unsigned int)is_signed_type(type));	\
-	if (!ret)							\
-		return 0;
-
-#undef __field_ext
-#define __field_ext(type, item, filter_type)	__field(type, item)
-
-#undef __array
-#define __array(type, item, len)						\
-	ret = trace_seq_printf(s, "\tfield:" #type " " #item "[" #len "];\t"	\
-			       "offset:%u;\tsize:%u;\tsigned:%u;\n",	\
-			       (unsigned int)offsetof(typeof(field), item), \
-			       (unsigned int)sizeof(field.item),	\
-			       (unsigned int)is_signed_type(type));	\
-	if (!ret)							\
-		return 0;
-
-#undef __dynamic_array
-#define __dynamic_array(type, item, len)				       \
-	ret = trace_seq_printf(s, "\tfield:__data_loc " #type "[] " #item ";\t"\
-			       "offset:%u;\tsize:%u;\tsigned:%u;\n",	       \
-			       (unsigned int)offsetof(typeof(field),	       \
-					__data_loc_##item),		       \
-			       (unsigned int)sizeof(field.__data_loc_##item), \
-			       (unsigned int)is_signed_type(type));	\
-	if (!ret)							       \
-		return 0;
-
-#undef __string
-#define __string(item, src) __dynamic_array(char, item, -1)
-
-#undef __entry
-#define __entry REC
-
-#undef __print_symbolic
-#undef __get_dynamic_array
-#undef __get_str
-
-#undef TP_printk
-#define TP_printk(fmt, args...) "\"%s\", %s\n", fmt, __stringify(args)
-
-#undef TP_fast_assign
-#define TP_fast_assign(args...) args
-
-#undef TP_perf_assign
-#define TP_perf_assign(args...)
-
-#undef DECLARE_EVENT_CLASS
-#define DECLARE_EVENT_CLASS(call, proto, args, tstruct, func, print)	\
-static int								\
-ftrace_format_setup_##call(struct ftrace_event_call *unused,		\
-			   struct trace_seq *s)				\
-{									\
-	struct ftrace_raw_##call field __attribute__((unused));		\
-	int ret = 0;							\
-									\
-	tstruct;							\
-									\
-	return ret;							\
-}									\
-									\
-static int								\
-ftrace_format_##call(struct ftrace_event_call *unused,			\
-		     struct trace_seq *s)				\
-{									\
-	int ret = 0;							\
-									\
-	ret = ftrace_format_setup_##call(unused, s);			\
-	if (!ret)							\
-		return ret;						\
-									\
-	ret = trace_seq_printf(s, "\nprint fmt: " print);		\
-									\
-	return ret;							\
-}
-
-#undef DEFINE_EVENT
-#define DEFINE_EVENT(template, name, proto, args)
-
-#undef DEFINE_EVENT_PRINT
-#define DEFINE_EVENT_PRINT(template, name, proto, args, print)		\
-static int								\
-ftrace_format_##name(struct ftrace_event_call *unused,			\
-		      struct trace_seq *s)				\
-{									\
-	int ret = 0;							\
-									\
-	ret = ftrace_format_setup_##template(unused, s);		\
-	if (!ret)							\
-		return ret;						\
-									\
-	trace_seq_printf(s, "\nprint fmt: " print);			\
-									\
-	return ret;							\
-}
-
-#include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
-
-/*
  * Stage 3 of the trace events.
  *
  * Override the macros in <trace/trace_events.h> to include the following:
@@ -499,7 +375,7 @@ static inline int ftrace_get_offsets_##call(				\
 
 #include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 
 /*
  * Generate the functions needed for tracepoint perf_event support.
@@ -542,7 +418,7 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\
 
 #include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
 
-#endif
+#endif /* CONFIG_PERF_EVENTS */
 
 /*
  * Stage 4 of the trace events.
@@ -622,12 +498,11 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\
  *	.raw_init		= trace_event_raw_init,
  *	.regfunc		= ftrace_reg_event_<call>,
  *	.unregfunc		= ftrace_unreg_event_<call>,
- *	.show_format		= ftrace_format_<call>,
  * }
  *
  */
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 
 #define _TRACE_PROFILE_INIT(call)					\
 	.profile_enable = ftrace_profile_enable_##call,			\
@@ -635,7 +510,7 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\
 
 #else
 #define _TRACE_PROFILE_INIT(call)
-#endif
+#endif /* CONFIG_PERF_EVENTS */
 
 #undef __entry
 #define __entry entry
@@ -657,6 +532,12 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\
 #define __assign_str(dst, src)						\
 	strcpy(__get_str(dst), src);
 
+#undef TP_fast_assign
+#define TP_fast_assign(args...) args
+
+#undef TP_perf_assign
+#define TP_perf_assign(args...)
+
 #undef DECLARE_EVENT_CLASS
 #define DECLARE_EVENT_CLASS(call, proto, args, tstruct, assign, print)	\
 									\
@@ -722,8 +603,20 @@ static struct trace_event ftrace_event_type_##call = {			\
 
 #include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
 
+#undef __entry
+#define __entry REC
+
+#undef __print_flags
+#undef __print_symbolic
+#undef __get_dynamic_array
+#undef __get_str
+
+#undef TP_printk
+#define TP_printk(fmt, args...) "\"" fmt "\", "  __stringify(args)
+
 #undef DECLARE_EVENT_CLASS
-#define DECLARE_EVENT_CLASS(call, proto, args, tstruct, assign, print)
+#define DECLARE_EVENT_CLASS(call, proto, args, tstruct, assign, print)	\
+static const char print_fmt_##call[] = print;
 
 #undef DEFINE_EVENT
 #define DEFINE_EVENT(template, call, proto, args)			\
@@ -737,7 +630,7 @@ __attribute__((section("_ftrace_events"))) event_##call = {		\
 	.raw_init		= trace_event_raw_init,			\
 	.regfunc		= ftrace_raw_reg_event_##call,		\
 	.unregfunc		= ftrace_raw_unreg_event_##call,	\
-	.show_format		= ftrace_format_##template,		\
+	.print_fmt		= print_fmt_##template,			\
 	.define_fields		= ftrace_define_fields_##template,	\
 	_TRACE_PROFILE_INIT(call)					\
 }
@@ -745,6 +638,8 @@ __attribute__((section("_ftrace_events"))) event_##call = {		\
 #undef DEFINE_EVENT_PRINT
 #define DEFINE_EVENT_PRINT(template, call, proto, args, print)		\
 									\
+static const char print_fmt_##call[] = print;				\
+									\
 static struct ftrace_event_call __used					\
 __attribute__((__aligned__(4)))						\
 __attribute__((section("_ftrace_events"))) event_##call = {		\
@@ -754,7 +649,7 @@ __attribute__((section("_ftrace_events"))) event_##call = {		\
 	.raw_init		= trace_event_raw_init,			\
 	.regfunc		= ftrace_raw_reg_event_##call,		\
 	.unregfunc		= ftrace_raw_unreg_event_##call,	\
-	.show_format		= ftrace_format_##call,			\
+	.print_fmt		= print_fmt_##call,			\
 	.define_fields		= ftrace_define_fields_##template,	\
 	_TRACE_PROFILE_INIT(call)					\
 }
@@ -835,7 +730,17 @@ __attribute__((section("_ftrace_events"))) event_##call = {		\
  * }
  */
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
+
+#undef __entry
+#define __entry entry
+
+#undef __get_dynamic_array
+#define __get_dynamic_array(field)	\
+		((void *)__entry + (__entry->__data_loc_##field & 0xffff))
+
+#undef __get_str
+#define __get_str(field) (char *)__get_dynamic_array(field)
 
 #undef __perf_addr
 #define __perf_addr(a) __addr = (a)
@@ -850,22 +755,12 @@ ftrace_profile_templ_##call(struct ftrace_event_call *event_call,	\
 			    proto)					\
 {									\
 	struct ftrace_data_offsets_##call __maybe_unused __data_offsets;\
-	extern int perf_swevent_get_recursion_context(void);		\
-	extern void perf_swevent_put_recursion_context(int rctx);	\
-	extern void perf_tp_event(int, u64, u64, void *, int);		\
 	struct ftrace_raw_##call *entry;				\
 	u64 __addr = 0, __count = 1;					\
 	unsigned long irq_flags;					\
-	struct trace_entry *ent;					\
 	int __entry_size;						\
 	int __data_size;						\
-	char *trace_buf;						\
-	char *raw_data;							\
-	int __cpu;							\
 	int rctx;							\
-	int pc;								\
-									\
-	pc = preempt_count();						\
 									\
 	__data_size = ftrace_get_offsets_##call(&__data_offsets, args); \
 	__entry_size = ALIGN(__data_size + sizeof(*entry) + sizeof(u32),\
@@ -875,42 +770,16 @@ ftrace_profile_templ_##call(struct ftrace_event_call *event_call,	\
 	if (WARN_ONCE(__entry_size > FTRACE_MAX_PROFILE_SIZE,		\
 		      "profile buffer not large enough"))		\
 		return;							\
-									\
-	local_irq_save(irq_flags);					\
-									\
-	rctx = perf_swevent_get_recursion_context();			\
-	if (rctx < 0)							\
-		goto end_recursion;					\
-									\
-	__cpu = smp_processor_id();					\
-									\
-	if (in_nmi())							\
-		trace_buf = rcu_dereference(perf_trace_buf_nmi);	\
-	else								\
-		trace_buf = rcu_dereference(perf_trace_buf);		\
-									\
-	if (!trace_buf)							\
-		goto end;						\
-									\
-	raw_data = per_cpu_ptr(trace_buf, __cpu);			\
-									\
-	*(u64 *)(&raw_data[__entry_size - sizeof(u64)]) = 0ULL;		\
-	entry = (struct ftrace_raw_##call *)raw_data;			\
-	ent = &entry->ent;						\
-	tracing_generic_entry_update(ent, irq_flags, pc);		\
-	ent->type = event_call->id;					\
-									\
+	entry = (struct ftrace_raw_##call *)ftrace_perf_buf_prepare(	\
+		__entry_size, event_call->id, &rctx, &irq_flags);	\
+	if (!entry)							\
+		return;							\
 	tstruct								\
 									\
 	{ assign; }							\
 									\
-	perf_tp_event(event_call->id, __addr, __count, entry,		\
-			     __entry_size);				\
-									\
-end:									\
-	perf_swevent_put_recursion_context(rctx);			\
-end_recursion:								\
-	local_irq_restore(irq_flags);					\
+	ftrace_perf_buf_submit(entry, __entry_size, rctx, __addr,	\
+			       __count, irq_flags);			\
 }
 
 #undef DEFINE_EVENT
@@ -927,7 +796,7 @@ static void ftrace_profile_##call(proto)			\
 	DEFINE_EVENT(template, name, PARAMS(proto), PARAMS(args))
 
 #include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
-#endif /* CONFIG_EVENT_PROFILE */
+#endif /* CONFIG_PERF_EVENTS */
 
 #undef _TRACE_PROFILE_INIT
 
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index 961fda3556bb..0387100752f0 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -34,10 +34,6 @@ struct syscall_metadata {
 extern unsigned long arch_syscall_addr(int nr);
 extern int init_syscall_trace(struct ftrace_event_call *call);
 
-extern int syscall_enter_format(struct ftrace_event_call *call,
-				struct trace_seq *s);
-extern int syscall_exit_format(struct ftrace_event_call *call,
-				struct trace_seq *s);
 extern int syscall_enter_define_fields(struct ftrace_event_call *call);
 extern int syscall_exit_define_fields(struct ftrace_event_call *call);
 extern int reg_event_syscall_enter(struct ftrace_event_call *call);
@@ -49,12 +45,12 @@ ftrace_format_syscall(struct ftrace_event_call *call, struct trace_seq *s);
 enum print_line_t print_syscall_enter(struct trace_iterator *iter, int flags);
 enum print_line_t print_syscall_exit(struct trace_iterator *iter, int flags);
 #endif
-#ifdef CONFIG_EVENT_PROFILE
+
+#ifdef CONFIG_PERF_EVENTS
 int prof_sysenter_enable(struct ftrace_event_call *call);
 void prof_sysenter_disable(struct ftrace_event_call *call);
 int prof_sysexit_enable(struct ftrace_event_call *call);
 void prof_sysexit_disable(struct ftrace_event_call *call);
-
 #endif
 
 #endif /* _TRACE_SYSCALL_H */
diff --git a/init/Kconfig b/init/Kconfig
index 79d375d92728..1abdaf68afe7 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -457,57 +457,6 @@ config LOG_BUF_SHIFT
 config HAVE_UNSTABLE_SCHED_CLOCK
 	bool
 
-config GROUP_SCHED
-	bool "Group CPU scheduler"
-	depends on EXPERIMENTAL
-	default n
-	help
-	  This feature lets CPU scheduler recognize task groups and control CPU
-	  bandwidth allocation to such task groups.
-	  In order to create a group from arbitrary set of processes, use
-	  CONFIG_CGROUPS. (See Control Group support.)
-
-config FAIR_GROUP_SCHED
-	bool "Group scheduling for SCHED_OTHER"
-	depends on GROUP_SCHED
-	default GROUP_SCHED
-
-config RT_GROUP_SCHED
-	bool "Group scheduling for SCHED_RR/FIFO"
-	depends on EXPERIMENTAL
-	depends on GROUP_SCHED
-	default n
-	help
-	  This feature lets you explicitly allocate real CPU bandwidth
-	  to users or control groups (depending on the "Basis for grouping tasks"
-	  setting below. If enabled, it will also make it impossible to
-	  schedule realtime tasks for non-root users until you allocate
-	  realtime bandwidth for them.
-	  See Documentation/scheduler/sched-rt-group.txt for more information.
-
-choice
-	depends on GROUP_SCHED
-	prompt "Basis for grouping tasks"
-	default USER_SCHED
-
-config USER_SCHED
-	bool "user id"
-	help
-	  This option will choose userid as the basis for grouping
-	  tasks, thus providing equal CPU bandwidth to each user.
-
-config CGROUP_SCHED
-	bool "Control groups"
- 	depends on CGROUPS
- 	help
-	  This option allows you to create arbitrary task groups
-	  using the "cgroup" pseudo filesystem and control
-	  the cpu bandwidth allocated to each such task group.
-	  Refer to Documentation/cgroups/cgroups.txt for more
-	  information on "cgroup" pseudo filesystem.
-
-endchoice
-
 menuconfig CGROUPS
 	boolean "Control Group support"
 	help
@@ -628,6 +577,36 @@ config CGROUP_MEM_RES_CTLR_SWAP
 	  Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
 	  size is 4096bytes, 512k per 1Gbytes of swap.
 
+menuconfig CGROUP_SCHED
+	bool "Group CPU scheduler"
+	depends on EXPERIMENTAL && CGROUPS
+	default n
+	help
+	  This feature lets CPU scheduler recognize task groups and control CPU
+	  bandwidth allocation to such task groups. It uses cgroups to group
+	  tasks.
+
+if CGROUP_SCHED
+config FAIR_GROUP_SCHED
+	bool "Group scheduling for SCHED_OTHER"
+	depends on CGROUP_SCHED
+	default CGROUP_SCHED
+
+config RT_GROUP_SCHED
+	bool "Group scheduling for SCHED_RR/FIFO"
+	depends on EXPERIMENTAL
+	depends on CGROUP_SCHED
+	default n
+	help
+	  This feature lets you explicitly allocate real CPU bandwidth
+	  to users or control groups (depending on the "Basis for grouping tasks"
+	  setting below. If enabled, it will also make it impossible to
+	  schedule realtime tasks for non-root users until you allocate
+	  realtime bandwidth for them.
+	  See Documentation/scheduler/sched-rt-group.txt for more information.
+
+endif #CGROUP_SCHED
+
 endif # CGROUPS
 
 config MM_OWNER
@@ -988,19 +967,6 @@ config PERF_EVENTS
 
 	  Say Y if unsure.
 
-config EVENT_PROFILE
-	bool "Tracepoint profiling sources"
-	depends on PERF_EVENTS && EVENT_TRACING
-	default y
-	help
-	 Allow the use of tracepoints as software performance events.
-
-	 When this is enabled, you can create perf events based on
-	 tracepoints using PERF_TYPE_TRACEPOINT and the tracepoint ID
-	 found in debugfs://tracing/events/*/*/id. (The -e/--events
-	 option to the perf tool can parse and interpret symbolic
-	 tracepoints, in the subsystem:tracepoint_name format.)
-
 config PERF_COUNTERS
 	bool "Kernel performance counters (old config option)"
 	depends on HAVE_PERF_EVENTS
diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c
index 235716556bf1..d49afb2395e5 100644
--- a/kernel/futex_compat.c
+++ b/kernel/futex_compat.c
@@ -146,7 +146,7 @@ compat_sys_get_robust_list(int pid, compat_uptr_t __user *head_ptr,
 		struct task_struct *p;
 
 		ret = -ESRCH;
-		read_lock(&tasklist_lock);
+		rcu_read_lock();
 		p = find_task_by_vpid(pid);
 		if (!p)
 			goto err_unlock;
@@ -157,7 +157,7 @@ compat_sys_get_robust_list(int pid, compat_uptr_t __user *head_ptr,
 		    !capable(CAP_SYS_PTRACE))
 			goto err_unlock;
 		head = p->compat_robust_list;
-		read_unlock(&tasklist_lock);
+		rcu_read_unlock();
 	}
 
 	if (put_user(sizeof(*head), len_ptr))
@@ -165,7 +165,7 @@ compat_sys_get_robust_list(int pid, compat_uptr_t __user *head_ptr,
 	return put_user(ptr_to_compat(head), head_ptr);
 
 err_unlock:
-	read_unlock(&tasklist_lock);
+	rcu_read_unlock();
 
 	return ret;
 }
diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
index ac08efca54c3..21fe3c426948 100644
--- a/kernel/ksysfs.c
+++ b/kernel/ksysfs.c
@@ -197,16 +197,8 @@ static int __init ksysfs_init(void)
 			goto group_exit;
 	}
 
-	/* create the /sys/kernel/uids/ directory */
-	error = uids_sysfs_init();
-	if (error)
-		goto notes_exit;
-
 	return 0;
 
-notes_exit:
-	if (notes_size > 0)
-		sysfs_remove_bin_file(kernel_kobj, &notes_attr);
 group_exit:
 	sysfs_remove_group(kernel_kobj, &kernel_attr_group);
 kset_exit:
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index d27746bd3a06..40f8b07c5601 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -289,6 +289,15 @@ static void update_event_times(struct perf_event *event)
 	event->total_time_running = run_end - event->tstamp_running;
 }
 
+static struct list_head *
+ctx_group_list(struct perf_event *event, struct perf_event_context *ctx)
+{
+	if (event->attr.pinned)
+		return &ctx->pinned_groups;
+	else
+		return &ctx->flexible_groups;
+}
+
 /*
  * Add a event from the lists for its context.
  * Must be called with ctx->mutex and ctx->lock held.
@@ -303,9 +312,19 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 	 * add it straight to the context's event list, or to the group
 	 * leader's sibling list:
 	 */
-	if (group_leader == event)
-		list_add_tail(&event->group_entry, &ctx->group_list);
-	else {
+	if (group_leader == event) {
+		struct list_head *list;
+
+		if (is_software_event(event))
+			event->group_flags |= PERF_GROUP_SOFTWARE;
+
+		list = ctx_group_list(event, ctx);
+		list_add_tail(&event->group_entry, list);
+	} else {
+		if (group_leader->group_flags & PERF_GROUP_SOFTWARE &&
+		    !is_software_event(event))
+			group_leader->group_flags &= ~PERF_GROUP_SOFTWARE;
+
 		list_add_tail(&event->group_entry, &group_leader->sibling_list);
 		group_leader->nr_siblings++;
 	}
@@ -355,9 +374,14 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 	 * to the context list directly:
 	 */
 	list_for_each_entry_safe(sibling, tmp, &event->sibling_list, group_entry) {
+		struct list_head *list;
 
-		list_move_tail(&sibling->group_entry, &ctx->group_list);
+		list = ctx_group_list(event, ctx);
+		list_move_tail(&sibling->group_entry, list);
 		sibling->group_leader = sibling;
+
+		/* Inherit group flags from the previous leader */
+		sibling->group_flags = event->group_flags;
 	}
 }
 
@@ -686,24 +710,6 @@ group_error:
 }
 
 /*
- * Return 1 for a group consisting entirely of software events,
- * 0 if the group contains any hardware events.
- */
-static int is_software_only_group(struct perf_event *leader)
-{
-	struct perf_event *event;
-
-	if (!is_software_event(leader))
-		return 0;
-
-	list_for_each_entry(event, &leader->sibling_list, group_entry)
-		if (!is_software_event(event))
-			return 0;
-
-	return 1;
-}
-
-/*
  * Work out whether we can put this event group on the CPU now.
  */
 static int group_can_go_on(struct perf_event *event,
@@ -713,7 +719,7 @@ static int group_can_go_on(struct perf_event *event,
 	/*
 	 * Groups consisting entirely of software events can always go on.
 	 */
-	if (is_software_only_group(event))
+	if (event->group_flags & PERF_GROUP_SOFTWARE)
 		return 1;
 	/*
 	 * If an exclusive group is already on, no other hardware
@@ -1043,8 +1049,15 @@ static int perf_event_refresh(struct perf_event *event, int refresh)
 	return 0;
 }
 
-void __perf_event_sched_out(struct perf_event_context *ctx,
-			      struct perf_cpu_context *cpuctx)
+enum event_type_t {
+	EVENT_FLEXIBLE = 0x1,
+	EVENT_PINNED = 0x2,
+	EVENT_ALL = EVENT_FLEXIBLE | EVENT_PINNED,
+};
+
+static void ctx_sched_out(struct perf_event_context *ctx,
+			  struct perf_cpu_context *cpuctx,
+			  enum event_type_t event_type)
 {
 	struct perf_event *event;
 
@@ -1055,10 +1068,18 @@ void __perf_event_sched_out(struct perf_event_context *ctx,
 	update_context_time(ctx);
 
 	perf_disable();
-	if (ctx->nr_active) {
-		list_for_each_entry(event, &ctx->group_list, group_entry)
+	if (!ctx->nr_active)
+		goto out_enable;
+
+	if (event_type & EVENT_PINNED)
+		list_for_each_entry(event, &ctx->pinned_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
-	}
+
+	if (event_type & EVENT_FLEXIBLE)
+		list_for_each_entry(event, &ctx->flexible_groups, group_entry)
+			group_sched_out(event, cpuctx, ctx);
+
+ out_enable:
 	perf_enable();
  out:
 	raw_spin_unlock(&ctx->lock);
@@ -1170,9 +1191,9 @@ static void perf_event_sync_stat(struct perf_event_context *ctx,
  * not restart the event.
  */
 void perf_event_task_sched_out(struct task_struct *task,
-				 struct task_struct *next, int cpu)
+				 struct task_struct *next)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = task->perf_event_ctxp;
 	struct perf_event_context *next_ctx;
 	struct perf_event_context *parent;
@@ -1220,15 +1241,13 @@ void perf_event_task_sched_out(struct task_struct *task,
 	rcu_read_unlock();
 
 	if (do_switch) {
-		__perf_event_sched_out(ctx, cpuctx);
+		ctx_sched_out(ctx, cpuctx, EVENT_ALL);
 		cpuctx->task_ctx = NULL;
 	}
 }
 
-/*
- * Called with IRQs disabled
- */
-static void __perf_event_task_sched_out(struct perf_event_context *ctx)
+static void task_ctx_sched_out(struct perf_event_context *ctx,
+			       enum event_type_t event_type)
 {
 	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 
@@ -1238,41 +1257,36 @@ static void __perf_event_task_sched_out(struct perf_event_context *ctx)
 	if (WARN_ON_ONCE(ctx != cpuctx->task_ctx))
 		return;
 
-	__perf_event_sched_out(ctx, cpuctx);
+	ctx_sched_out(ctx, cpuctx, event_type);
 	cpuctx->task_ctx = NULL;
 }
 
 /*
  * Called with IRQs disabled
  */
-static void perf_event_cpu_sched_out(struct perf_cpu_context *cpuctx)
+static void __perf_event_task_sched_out(struct perf_event_context *ctx)
 {
-	__perf_event_sched_out(&cpuctx->ctx, cpuctx);
+	task_ctx_sched_out(ctx, EVENT_ALL);
+}
+
+/*
+ * Called with IRQs disabled
+ */
+static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx,
+			      enum event_type_t event_type)
+{
+	ctx_sched_out(&cpuctx->ctx, cpuctx, event_type);
 }
 
 static void
-__perf_event_sched_in(struct perf_event_context *ctx,
-			struct perf_cpu_context *cpuctx, int cpu)
+ctx_pinned_sched_in(struct perf_event_context *ctx,
+		    struct perf_cpu_context *cpuctx,
+		    int cpu)
 {
 	struct perf_event *event;
-	int can_add_hw = 1;
-
-	raw_spin_lock(&ctx->lock);
-	ctx->is_active = 1;
-	if (likely(!ctx->nr_events))
-		goto out;
-
-	ctx->timestamp = perf_clock();
 
-	perf_disable();
-
-	/*
-	 * First go through the list and put on any pinned groups
-	 * in order to give them the best chance of going on.
-	 */
-	list_for_each_entry(event, &ctx->group_list, group_entry) {
-		if (event->state <= PERF_EVENT_STATE_OFF ||
-		    !event->attr.pinned)
+	list_for_each_entry(event, &ctx->pinned_groups, group_entry) {
+		if (event->state <= PERF_EVENT_STATE_OFF)
 			continue;
 		if (event->cpu != -1 && event->cpu != cpu)
 			continue;
@@ -1289,16 +1303,20 @@ __perf_event_sched_in(struct perf_event_context *ctx,
 			event->state = PERF_EVENT_STATE_ERROR;
 		}
 	}
+}
 
-	list_for_each_entry(event, &ctx->group_list, group_entry) {
-		/*
-		 * Ignore events in OFF or ERROR state, and
-		 * ignore pinned events since we did them already.
-		 */
-		if (event->state <= PERF_EVENT_STATE_OFF ||
-		    event->attr.pinned)
-			continue;
+static void
+ctx_flexible_sched_in(struct perf_event_context *ctx,
+		      struct perf_cpu_context *cpuctx,
+		      int cpu)
+{
+	struct perf_event *event;
+	int can_add_hw = 1;
 
+	list_for_each_entry(event, &ctx->flexible_groups, group_entry) {
+		/* Ignore events in OFF or ERROR state */
+		if (event->state <= PERF_EVENT_STATE_OFF)
+			continue;
 		/*
 		 * Listen to the 'cpu' scheduling filter constraint
 		 * of events:
@@ -1310,11 +1328,61 @@ __perf_event_sched_in(struct perf_event_context *ctx,
 			if (group_sched_in(event, cpuctx, ctx, cpu))
 				can_add_hw = 0;
 	}
+}
+
+static void
+ctx_sched_in(struct perf_event_context *ctx,
+	     struct perf_cpu_context *cpuctx,
+	     enum event_type_t event_type)
+{
+	int cpu = smp_processor_id();
+
+	raw_spin_lock(&ctx->lock);
+	ctx->is_active = 1;
+	if (likely(!ctx->nr_events))
+		goto out;
+
+	ctx->timestamp = perf_clock();
+
+	perf_disable();
+
+	/*
+	 * First go through the list and put on any pinned groups
+	 * in order to give them the best chance of going on.
+	 */
+	if (event_type & EVENT_PINNED)
+		ctx_pinned_sched_in(ctx, cpuctx, cpu);
+
+	/* Then walk through the lower prio flexible groups */
+	if (event_type & EVENT_FLEXIBLE)
+		ctx_flexible_sched_in(ctx, cpuctx, cpu);
+
 	perf_enable();
  out:
 	raw_spin_unlock(&ctx->lock);
 }
 
+static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx,
+			     enum event_type_t event_type)
+{
+	struct perf_event_context *ctx = &cpuctx->ctx;
+
+	ctx_sched_in(ctx, cpuctx, event_type);
+}
+
+static void task_ctx_sched_in(struct task_struct *task,
+			      enum event_type_t event_type)
+{
+	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct perf_event_context *ctx = task->perf_event_ctxp;
+
+	if (likely(!ctx))
+		return;
+	if (cpuctx->task_ctx == ctx)
+		return;
+	ctx_sched_in(ctx, cpuctx, event_type);
+	cpuctx->task_ctx = ctx;
+}
 /*
  * Called from scheduler to add the events of the current task
  * with interrupts disabled.
@@ -1326,38 +1394,112 @@ __perf_event_sched_in(struct perf_event_context *ctx,
  * accessing the event control register. If a NMI hits, then it will
  * keep the event running.
  */
-void perf_event_task_sched_in(struct task_struct *task, int cpu)
+void perf_event_task_sched_in(struct task_struct *task)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = task->perf_event_ctxp;
 
 	if (likely(!ctx))
 		return;
+
 	if (cpuctx->task_ctx == ctx)
 		return;
-	__perf_event_sched_in(ctx, cpuctx, cpu);
-	cpuctx->task_ctx = ctx;
-}
 
-static void perf_event_cpu_sched_in(struct perf_cpu_context *cpuctx, int cpu)
-{
-	struct perf_event_context *ctx = &cpuctx->ctx;
+	/*
+	 * We want to keep the following priority order:
+	 * cpu pinned (that don't need to move), task pinned,
+	 * cpu flexible, task flexible.
+	 */
+	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 
-	__perf_event_sched_in(ctx, cpuctx, cpu);
+	ctx_sched_in(ctx, cpuctx, EVENT_PINNED);
+	cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
+	ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE);
+
+	cpuctx->task_ctx = ctx;
 }
 
 #define MAX_INTERRUPTS (~0ULL)
 
 static void perf_log_throttle(struct perf_event *event, int enable);
 
-static void perf_adjust_period(struct perf_event *event, u64 events)
+static u64 perf_calculate_period(struct perf_event *event, u64 nsec, u64 count)
+{
+	u64 frequency = event->attr.sample_freq;
+	u64 sec = NSEC_PER_SEC;
+	u64 divisor, dividend;
+
+	int count_fls, nsec_fls, frequency_fls, sec_fls;
+
+	count_fls = fls64(count);
+	nsec_fls = fls64(nsec);
+	frequency_fls = fls64(frequency);
+	sec_fls = 30;
+
+	/*
+	 * We got @count in @nsec, with a target of sample_freq HZ
+	 * the target period becomes:
+	 *
+	 *             @count * 10^9
+	 * period = -------------------
+	 *          @nsec * sample_freq
+	 *
+	 */
+
+	/*
+	 * Reduce accuracy by one bit such that @a and @b converge
+	 * to a similar magnitude.
+	 */
+#define REDUCE_FLS(a, b) 		\
+do {					\
+	if (a##_fls > b##_fls) {	\
+		a >>= 1;		\
+		a##_fls--;		\
+	} else {			\
+		b >>= 1;		\
+		b##_fls--;		\
+	}				\
+} while (0)
+
+	/*
+	 * Reduce accuracy until either term fits in a u64, then proceed with
+	 * the other, so that finally we can do a u64/u64 division.
+	 */
+	while (count_fls + sec_fls > 64 && nsec_fls + frequency_fls > 64) {
+		REDUCE_FLS(nsec, frequency);
+		REDUCE_FLS(sec, count);
+	}
+
+	if (count_fls + sec_fls > 64) {
+		divisor = nsec * frequency;
+
+		while (count_fls + sec_fls > 64) {
+			REDUCE_FLS(count, sec);
+			divisor >>= 1;
+		}
+
+		dividend = count * sec;
+	} else {
+		dividend = count * sec;
+
+		while (nsec_fls + frequency_fls > 64) {
+			REDUCE_FLS(nsec, frequency);
+			dividend >>= 1;
+		}
+
+		divisor = nsec * frequency;
+	}
+
+	return div64_u64(dividend, divisor);
+}
+
+static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	u64 period, sample_period;
 	s64 delta;
 
-	events *= hwc->sample_period;
-	period = div64_u64(events, event->attr.sample_freq);
+	period = perf_calculate_period(event, nsec, count);
 
 	delta = (s64)(period - hwc->sample_period);
 	delta = (delta + 7) / 8; /* low pass filter */
@@ -1368,13 +1510,22 @@ static void perf_adjust_period(struct perf_event *event, u64 events)
 		sample_period = 1;
 
 	hwc->sample_period = sample_period;
+
+	if (atomic64_read(&hwc->period_left) > 8*sample_period) {
+		perf_disable();
+		event->pmu->disable(event);
+		atomic64_set(&hwc->period_left, 0);
+		event->pmu->enable(event);
+		perf_enable();
+	}
 }
 
 static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
 {
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
-	u64 interrupts, freq;
+	u64 interrupts, now;
+	s64 delta;
 
 	raw_spin_lock(&ctx->lock);
 	list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
@@ -1395,44 +1546,18 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
 		if (interrupts == MAX_INTERRUPTS) {
 			perf_log_throttle(event, 1);
 			event->pmu->unthrottle(event);
-			interrupts = 2*sysctl_perf_event_sample_rate/HZ;
 		}
 
 		if (!event->attr.freq || !event->attr.sample_freq)
 			continue;
 
-		/*
-		 * if the specified freq < HZ then we need to skip ticks
-		 */
-		if (event->attr.sample_freq < HZ) {
-			freq = event->attr.sample_freq;
-
-			hwc->freq_count += freq;
-			hwc->freq_interrupts += interrupts;
-
-			if (hwc->freq_count < HZ)
-				continue;
-
-			interrupts = hwc->freq_interrupts;
-			hwc->freq_interrupts = 0;
-			hwc->freq_count -= HZ;
-		} else
-			freq = HZ;
-
-		perf_adjust_period(event, freq * interrupts);
+		event->pmu->read(event);
+		now = atomic64_read(&event->count);
+		delta = now - hwc->freq_count_stamp;
+		hwc->freq_count_stamp = now;
 
-		/*
-		 * In order to avoid being stalled by an (accidental) huge
-		 * sample period, force reset the sample period if we didn't
-		 * get any events in this freq period.
-		 */
-		if (!interrupts) {
-			perf_disable();
-			event->pmu->disable(event);
-			atomic64_set(&hwc->period_left, 0);
-			event->pmu->enable(event);
-			perf_enable();
-		}
+		if (delta > 0)
+			perf_adjust_period(event, TICK_NSEC, delta);
 	}
 	raw_spin_unlock(&ctx->lock);
 }
@@ -1442,26 +1567,22 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
  */
 static void rotate_ctx(struct perf_event_context *ctx)
 {
-	struct perf_event *event;
-
 	if (!ctx->nr_events)
 		return;
 
 	raw_spin_lock(&ctx->lock);
-	/*
-	 * Rotate the first entry last (works just fine for group events too):
-	 */
+
+	/* Rotate the first entry last of non-pinned groups */
 	perf_disable();
-	list_for_each_entry(event, &ctx->group_list, group_entry) {
-		list_move_tail(&event->group_entry, &ctx->group_list);
-		break;
-	}
+
+	list_rotate_left(&ctx->flexible_groups);
+
 	perf_enable();
 
 	raw_spin_unlock(&ctx->lock);
 }
 
-void perf_event_task_tick(struct task_struct *curr, int cpu)
+void perf_event_task_tick(struct task_struct *curr)
 {
 	struct perf_cpu_context *cpuctx;
 	struct perf_event_context *ctx;
@@ -1469,24 +1590,39 @@ void perf_event_task_tick(struct task_struct *curr, int cpu)
 	if (!atomic_read(&nr_events))
 		return;
 
-	cpuctx = &per_cpu(perf_cpu_context, cpu);
+	cpuctx = &__get_cpu_var(perf_cpu_context);
 	ctx = curr->perf_event_ctxp;
 
 	perf_ctx_adjust_freq(&cpuctx->ctx);
 	if (ctx)
 		perf_ctx_adjust_freq(ctx);
 
-	perf_event_cpu_sched_out(cpuctx);
+	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
-		__perf_event_task_sched_out(ctx);
+		task_ctx_sched_out(ctx, EVENT_FLEXIBLE);
 
 	rotate_ctx(&cpuctx->ctx);
 	if (ctx)
 		rotate_ctx(ctx);
 
-	perf_event_cpu_sched_in(cpuctx, cpu);
+	cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
-		perf_event_task_sched_in(curr, cpu);
+		task_ctx_sched_in(curr, EVENT_FLEXIBLE);
+}
+
+static int event_enable_on_exec(struct perf_event *event,
+				struct perf_event_context *ctx)
+{
+	if (!event->attr.enable_on_exec)
+		return 0;
+
+	event->attr.enable_on_exec = 0;
+	if (event->state >= PERF_EVENT_STATE_INACTIVE)
+		return 0;
+
+	__perf_event_mark_enabled(event, ctx);
+
+	return 1;
 }
 
 /*
@@ -1499,6 +1635,7 @@ static void perf_event_enable_on_exec(struct task_struct *task)
 	struct perf_event *event;
 	unsigned long flags;
 	int enabled = 0;
+	int ret;
 
 	local_irq_save(flags);
 	ctx = task->perf_event_ctxp;
@@ -1509,14 +1646,16 @@ static void perf_event_enable_on_exec(struct task_struct *task)
 
 	raw_spin_lock(&ctx->lock);
 
-	list_for_each_entry(event, &ctx->group_list, group_entry) {
-		if (!event->attr.enable_on_exec)
-			continue;
-		event->attr.enable_on_exec = 0;
-		if (event->state >= PERF_EVENT_STATE_INACTIVE)
-			continue;
-		__perf_event_mark_enabled(event, ctx);
-		enabled = 1;
+	list_for_each_entry(event, &ctx->pinned_groups, group_entry) {
+		ret = event_enable_on_exec(event, ctx);
+		if (ret)
+			enabled = 1;
+	}
+
+	list_for_each_entry(event, &ctx->flexible_groups, group_entry) {
+		ret = event_enable_on_exec(event, ctx);
+		if (ret)
+			enabled = 1;
 	}
 
 	/*
@@ -1527,7 +1666,7 @@ static void perf_event_enable_on_exec(struct task_struct *task)
 
 	raw_spin_unlock(&ctx->lock);
 
-	perf_event_task_sched_in(task, smp_processor_id());
+	perf_event_task_sched_in(task);
  out:
 	local_irq_restore(flags);
 }
@@ -1590,7 +1729,8 @@ __perf_event_init_context(struct perf_event_context *ctx,
 {
 	raw_spin_lock_init(&ctx->lock);
 	mutex_init(&ctx->mutex);
-	INIT_LIST_HEAD(&ctx->group_list);
+	INIT_LIST_HEAD(&ctx->pinned_groups);
+	INIT_LIST_HEAD(&ctx->flexible_groups);
 	INIT_LIST_HEAD(&ctx->event_list);
 	atomic_set(&ctx->refcount, 1);
 	ctx->task = task;
@@ -3689,12 +3829,12 @@ static int __perf_event_overflow(struct perf_event *event, int nmi,
 
 	if (event->attr.freq) {
 		u64 now = perf_clock();
-		s64 delta = now - hwc->freq_stamp;
+		s64 delta = now - hwc->freq_time_stamp;
 
-		hwc->freq_stamp = now;
+		hwc->freq_time_stamp = now;
 
-		if (delta > 0 && delta < TICK_NSEC)
-			perf_adjust_period(event, NSEC_PER_SEC / (int)delta);
+		if (delta > 0 && delta < 2*TICK_NSEC)
+			perf_adjust_period(event, delta, hwc->last_period);
 	}
 
 	/*
@@ -4185,7 +4325,7 @@ static const struct pmu perf_ops_task_clock = {
 	.read		= task_clock_perf_event_read,
 };
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_EVENT_TRACING
 
 void perf_tp_event(int event_id, u64 addr, u64 count, void *record,
 			  int entry_size)
@@ -4290,7 +4430,7 @@ static void perf_event_free_filter(struct perf_event *event)
 {
 }
 
-#endif /* CONFIG_EVENT_PROFILE */
+#endif /* CONFIG_EVENT_TRACING */
 
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 static void bp_perf_event_destroy(struct perf_event *event)
@@ -4871,8 +5011,15 @@ inherit_event(struct perf_event *parent_event,
 	else
 		child_event->state = PERF_EVENT_STATE_OFF;
 
-	if (parent_event->attr.freq)
-		child_event->hw.sample_period = parent_event->hw.sample_period;
+	if (parent_event->attr.freq) {
+		u64 sample_period = parent_event->hw.sample_period;
+		struct hw_perf_event *hwc = &child_event->hw;
+
+		hwc->sample_period = sample_period;
+		hwc->last_period   = sample_period;
+
+		atomic64_set(&hwc->period_left, sample_period);
+	}
 
 	child_event->overflow_handler = parent_event->overflow_handler;
 
@@ -5040,7 +5187,11 @@ void perf_event_exit_task(struct task_struct *child)
 	mutex_lock_nested(&child_ctx->mutex, SINGLE_DEPTH_NESTING);
 
 again:
-	list_for_each_entry_safe(child_event, tmp, &child_ctx->group_list,
+	list_for_each_entry_safe(child_event, tmp, &child_ctx->pinned_groups,
+				 group_entry)
+		__perf_event_exit_task(child_event, child_ctx, child);
+
+	list_for_each_entry_safe(child_event, tmp, &child_ctx->flexible_groups,
 				 group_entry)
 		__perf_event_exit_task(child_event, child_ctx, child);
 
@@ -5049,7 +5200,8 @@ again:
 	 * its siblings to the list, but we obtained 'tmp' before that which
 	 * will still point to the list head terminating the iteration.
 	 */
-	if (!list_empty(&child_ctx->group_list))
+	if (!list_empty(&child_ctx->pinned_groups) ||
+	    !list_empty(&child_ctx->flexible_groups))
 		goto again;
 
 	mutex_unlock(&child_ctx->mutex);
@@ -5057,6 +5209,24 @@ again:
 	put_ctx(child_ctx);
 }
 
+static void perf_free_event(struct perf_event *event,
+			    struct perf_event_context *ctx)
+{
+	struct perf_event *parent = event->parent;
+
+	if (WARN_ON_ONCE(!parent))
+		return;
+
+	mutex_lock(&parent->child_mutex);
+	list_del_init(&event->child_list);
+	mutex_unlock(&parent->child_mutex);
+
+	fput(parent->filp);
+
+	list_del_event(event, ctx);
+	free_event(event);
+}
+
 /*
  * free an unexposed, unused context as created by inheritance by
  * init_task below, used by fork() in case of fail.
@@ -5071,36 +5241,70 @@ void perf_event_free_task(struct task_struct *task)
 
 	mutex_lock(&ctx->mutex);
 again:
-	list_for_each_entry_safe(event, tmp, &ctx->group_list, group_entry) {
-		struct perf_event *parent = event->parent;
+	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
+		perf_free_event(event, ctx);
 
-		if (WARN_ON_ONCE(!parent))
-			continue;
+	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups,
+				 group_entry)
+		perf_free_event(event, ctx);
 
-		mutex_lock(&parent->child_mutex);
-		list_del_init(&event->child_list);
-		mutex_unlock(&parent->child_mutex);
+	if (!list_empty(&ctx->pinned_groups) ||
+	    !list_empty(&ctx->flexible_groups))
+		goto again;
 
-		fput(parent->filp);
+	mutex_unlock(&ctx->mutex);
 
-		list_del_event(event, ctx);
-		free_event(event);
+	put_ctx(ctx);
+}
+
+static int
+inherit_task_group(struct perf_event *event, struct task_struct *parent,
+		   struct perf_event_context *parent_ctx,
+		   struct task_struct *child,
+		   int *inherited_all)
+{
+	int ret;
+	struct perf_event_context *child_ctx = child->perf_event_ctxp;
+
+	if (!event->attr.inherit) {
+		*inherited_all = 0;
+		return 0;
 	}
 
-	if (!list_empty(&ctx->group_list))
-		goto again;
+	if (!child_ctx) {
+		/*
+		 * This is executed from the parent task context, so
+		 * inherit events that have been marked for cloning.
+		 * First allocate and initialize a context for the
+		 * child.
+		 */
 
-	mutex_unlock(&ctx->mutex);
+		child_ctx = kzalloc(sizeof(struct perf_event_context),
+				    GFP_KERNEL);
+		if (!child_ctx)
+			return -ENOMEM;
 
-	put_ctx(ctx);
+		__perf_event_init_context(child_ctx, child);
+		child->perf_event_ctxp = child_ctx;
+		get_task_struct(child);
+	}
+
+	ret = inherit_group(event, parent, parent_ctx,
+			    child, child_ctx);
+
+	if (ret)
+		*inherited_all = 0;
+
+	return ret;
 }
 
+
 /*
  * Initialize the perf_event context in task_struct
  */
 int perf_event_init_task(struct task_struct *child)
 {
-	struct perf_event_context *child_ctx = NULL, *parent_ctx;
+	struct perf_event_context *child_ctx, *parent_ctx;
 	struct perf_event_context *cloned_ctx;
 	struct perf_event *event;
 	struct task_struct *parent = current;
@@ -5138,41 +5342,22 @@ int perf_event_init_task(struct task_struct *child)
 	 * We dont have to disable NMIs - we are only looking at
 	 * the list, not manipulating it:
 	 */
-	list_for_each_entry(event, &parent_ctx->group_list, group_entry) {
-
-		if (!event->attr.inherit) {
-			inherited_all = 0;
-			continue;
-		}
-
-		if (!child->perf_event_ctxp) {
-			/*
-			 * This is executed from the parent task context, so
-			 * inherit events that have been marked for cloning.
-			 * First allocate and initialize a context for the
-			 * child.
-			 */
-
-			child_ctx = kzalloc(sizeof(struct perf_event_context),
-					    GFP_KERNEL);
-			if (!child_ctx) {
-				ret = -ENOMEM;
-				break;
-			}
-
-			__perf_event_init_context(child_ctx, child);
-			child->perf_event_ctxp = child_ctx;
-			get_task_struct(child);
-		}
+	list_for_each_entry(event, &parent_ctx->pinned_groups, group_entry) {
+		ret = inherit_task_group(event, parent, parent_ctx, child,
+					 &inherited_all);
+		if (ret)
+			break;
+	}
 
-		ret = inherit_group(event, parent, parent_ctx,
-					     child, child_ctx);
-		if (ret) {
-			inherited_all = 0;
+	list_for_each_entry(event, &parent_ctx->flexible_groups, group_entry) {
+		ret = inherit_task_group(event, parent, parent_ctx, child,
+					 &inherited_all);
+		if (ret)
 			break;
-		}
 	}
 
+	child_ctx = child->perf_event_ctxp;
+
 	if (child_ctx && inherited_all) {
 		/*
 		 * Mark the child context as a clone of the parent
@@ -5221,7 +5406,9 @@ static void __perf_event_exit_cpu(void *info)
 	struct perf_event_context *ctx = &cpuctx->ctx;
 	struct perf_event *event, *tmp;
 
-	list_for_each_entry_safe(event, tmp, &ctx->group_list, group_entry)
+	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
+		__perf_event_remove_from_context(event);
+	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, group_entry)
 		__perf_event_remove_from_context(event);
 }
 static void perf_event_exit_cpu(int cpu)
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index 9bb52177af02..adda92bfafac 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -61,6 +61,9 @@ static int test_no_idle_hz;	/* Test RCU's support for tickless idle CPUs. */
 static int shuffle_interval = 3; /* Interval between shuffles (in sec)*/
 static int stutter = 5;		/* Start/stop testing interval (in sec) */
 static int irqreader = 1;	/* RCU readers from irq (timers). */
+static int fqs_duration = 0;	/* Duration of bursts (us), 0 to disable. */
+static int fqs_holdoff = 0;	/* Hold time within burst (us). */
+static int fqs_stutter = 3;	/* Wait time between bursts (s). */
 static char *torture_type = "rcu"; /* What RCU implementation to torture. */
 
 module_param(nreaders, int, 0444);
@@ -79,6 +82,12 @@ module_param(stutter, int, 0444);
 MODULE_PARM_DESC(stutter, "Number of seconds to run/halt test");
 module_param(irqreader, int, 0444);
 MODULE_PARM_DESC(irqreader, "Allow RCU readers from irq handlers");
+module_param(fqs_duration, int, 0444);
+MODULE_PARM_DESC(fqs_duration, "Duration of fqs bursts (us)");
+module_param(fqs_holdoff, int, 0444);
+MODULE_PARM_DESC(fqs_holdoff, "Holdoff time within fqs bursts (us)");
+module_param(fqs_stutter, int, 0444);
+MODULE_PARM_DESC(fqs_stutter, "Wait time between fqs bursts (s)");
 module_param(torture_type, charp, 0444);
 MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu, rcu_bh, srcu)");
 
@@ -99,6 +108,7 @@ static struct task_struct **reader_tasks;
 static struct task_struct *stats_task;
 static struct task_struct *shuffler_task;
 static struct task_struct *stutter_task;
+static struct task_struct *fqs_task;
 
 #define RCU_TORTURE_PIPE_LEN 10
 
@@ -263,6 +273,7 @@ struct rcu_torture_ops {
 	void (*deferred_free)(struct rcu_torture *p);
 	void (*sync)(void);
 	void (*cb_barrier)(void);
+	void (*fqs)(void);
 	int (*stats)(char *page);
 	int irq_capable;
 	char *name;
@@ -347,6 +358,7 @@ static struct rcu_torture_ops rcu_ops = {
 	.deferred_free	= rcu_torture_deferred_free,
 	.sync		= synchronize_rcu,
 	.cb_barrier	= rcu_barrier,
+	.fqs		= rcu_force_quiescent_state,
 	.stats		= NULL,
 	.irq_capable	= 1,
 	.name		= "rcu"
@@ -388,6 +400,7 @@ static struct rcu_torture_ops rcu_sync_ops = {
 	.deferred_free	= rcu_sync_torture_deferred_free,
 	.sync		= synchronize_rcu,
 	.cb_barrier	= NULL,
+	.fqs		= rcu_force_quiescent_state,
 	.stats		= NULL,
 	.irq_capable	= 1,
 	.name		= "rcu_sync"
@@ -403,6 +416,7 @@ static struct rcu_torture_ops rcu_expedited_ops = {
 	.deferred_free	= rcu_sync_torture_deferred_free,
 	.sync		= synchronize_rcu_expedited,
 	.cb_barrier	= NULL,
+	.fqs		= rcu_force_quiescent_state,
 	.stats		= NULL,
 	.irq_capable	= 1,
 	.name		= "rcu_expedited"
@@ -465,6 +479,7 @@ static struct rcu_torture_ops rcu_bh_ops = {
 	.deferred_free	= rcu_bh_torture_deferred_free,
 	.sync		= rcu_bh_torture_synchronize,
 	.cb_barrier	= rcu_barrier_bh,
+	.fqs		= rcu_bh_force_quiescent_state,
 	.stats		= NULL,
 	.irq_capable	= 1,
 	.name		= "rcu_bh"
@@ -480,6 +495,7 @@ static struct rcu_torture_ops rcu_bh_sync_ops = {
 	.deferred_free	= rcu_sync_torture_deferred_free,
 	.sync		= rcu_bh_torture_synchronize,
 	.cb_barrier	= NULL,
+	.fqs		= rcu_bh_force_quiescent_state,
 	.stats		= NULL,
 	.irq_capable	= 1,
 	.name		= "rcu_bh_sync"
@@ -621,6 +637,7 @@ static struct rcu_torture_ops sched_ops = {
 	.deferred_free	= rcu_sched_torture_deferred_free,
 	.sync		= sched_torture_synchronize,
 	.cb_barrier	= rcu_barrier_sched,
+	.fqs		= rcu_sched_force_quiescent_state,
 	.stats		= NULL,
 	.irq_capable	= 1,
 	.name		= "sched"
@@ -636,6 +653,7 @@ static struct rcu_torture_ops sched_sync_ops = {
 	.deferred_free	= rcu_sync_torture_deferred_free,
 	.sync		= sched_torture_synchronize,
 	.cb_barrier	= NULL,
+	.fqs		= rcu_sched_force_quiescent_state,
 	.stats		= NULL,
 	.name		= "sched_sync"
 };
@@ -650,12 +668,45 @@ static struct rcu_torture_ops sched_expedited_ops = {
 	.deferred_free	= rcu_sync_torture_deferred_free,
 	.sync		= synchronize_sched_expedited,
 	.cb_barrier	= NULL,
+	.fqs		= rcu_sched_force_quiescent_state,
 	.stats		= rcu_expedited_torture_stats,
 	.irq_capable	= 1,
 	.name		= "sched_expedited"
 };
 
 /*
+ * RCU torture force-quiescent-state kthread.  Repeatedly induces
+ * bursts of calls to force_quiescent_state(), increasing the probability
+ * of occurrence of some important types of race conditions.
+ */
+static int
+rcu_torture_fqs(void *arg)
+{
+	unsigned long fqs_resume_time;
+	int fqs_burst_remaining;
+
+	VERBOSE_PRINTK_STRING("rcu_torture_fqs task started");
+	do {
+		fqs_resume_time = jiffies + fqs_stutter * HZ;
+		while (jiffies - fqs_resume_time > LONG_MAX) {
+			schedule_timeout_interruptible(1);
+		}
+		fqs_burst_remaining = fqs_duration;
+		while (fqs_burst_remaining > 0) {
+			cur_ops->fqs();
+			udelay(fqs_holdoff);
+			fqs_burst_remaining -= fqs_holdoff;
+		}
+		rcu_stutter_wait("rcu_torture_fqs");
+	} while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP);
+	VERBOSE_PRINTK_STRING("rcu_torture_fqs task stopping");
+	rcutorture_shutdown_absorb("rcu_torture_fqs");
+	while (!kthread_should_stop())
+		schedule_timeout_uninterruptible(1);
+	return 0;
+}
+
+/*
  * RCU torture writer kthread.  Repeatedly substitutes a new structure
  * for that pointed to by rcu_torture_current, freeing the old structure
  * after a series of grace periods (the "pipeline").
@@ -1030,10 +1081,11 @@ rcu_torture_print_module_parms(char *tag)
 	printk(KERN_ALERT "%s" TORTURE_FLAG
 		"--- %s: nreaders=%d nfakewriters=%d "
 		"stat_interval=%d verbose=%d test_no_idle_hz=%d "
-		"shuffle_interval=%d stutter=%d irqreader=%d\n",
+		"shuffle_interval=%d stutter=%d irqreader=%d "
+		"fqs_duration=%d fqs_holdoff=%d fqs_stutter=%d\n",
 		torture_type, tag, nrealreaders, nfakewriters,
 		stat_interval, verbose, test_no_idle_hz, shuffle_interval,
-		stutter, irqreader);
+		stutter, irqreader, fqs_duration, fqs_holdoff, fqs_stutter);
 }
 
 static struct notifier_block rcutorture_nb = {
@@ -1109,6 +1161,12 @@ rcu_torture_cleanup(void)
 	}
 	stats_task = NULL;
 
+	if (fqs_task) {
+		VERBOSE_PRINTK_STRING("Stopping rcu_torture_fqs task");
+		kthread_stop(fqs_task);
+	}
+	fqs_task = NULL;
+
 	/* Wait for all RCU callbacks to fire.  */
 
 	if (cur_ops->cb_barrier != NULL)
@@ -1154,6 +1212,11 @@ rcu_torture_init(void)
 		mutex_unlock(&fullstop_mutex);
 		return -EINVAL;
 	}
+	if (cur_ops->fqs == NULL && fqs_duration != 0) {
+		printk(KERN_ALERT "rcu-torture: ->fqs NULL and non-zero "
+				  "fqs_duration, fqs disabled.\n");
+		fqs_duration = 0;
+	}
 	if (cur_ops->init)
 		cur_ops->init(); /* no "goto unwind" prior to this point!!! */
 
@@ -1282,6 +1345,19 @@ rcu_torture_init(void)
 			goto unwind;
 		}
 	}
+	if (fqs_duration < 0)
+		fqs_duration = 0;
+	if (fqs_duration) {
+		/* Create the stutter thread */
+		fqs_task = kthread_run(rcu_torture_fqs, NULL,
+				       "rcu_torture_fqs");
+		if (IS_ERR(fqs_task)) {
+			firsterr = PTR_ERR(fqs_task);
+			VERBOSE_PRINTK_ERRSTRING("Failed to create fqs");
+			fqs_task = NULL;
+			goto unwind;
+		}
+	}
 	register_reboot_notifier(&rcutorture_nb);
 	mutex_unlock(&fullstop_mutex);
 	return 0;
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 53ae9598f798..099a255ede4c 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -157,6 +157,24 @@ long rcu_batches_completed_bh(void)
 EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
 
 /*
+ * Force a quiescent state for RCU BH.
+ */
+void rcu_bh_force_quiescent_state(void)
+{
+	force_quiescent_state(&rcu_bh_state, 0);
+}
+EXPORT_SYMBOL_GPL(rcu_bh_force_quiescent_state);
+
+/*
+ * Force a quiescent state for RCU-sched.
+ */
+void rcu_sched_force_quiescent_state(void)
+{
+	force_quiescent_state(&rcu_sched_state, 0);
+}
+EXPORT_SYMBOL_GPL(rcu_sched_force_quiescent_state);
+
+/*
  * Does the CPU have callbacks ready to be invoked?
  */
 static int
@@ -659,7 +677,9 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags)
 	struct rcu_data *rdp = rsp->rda[smp_processor_id()];
 	struct rcu_node *rnp = rcu_get_root(rsp);
 
-	if (!cpu_needs_another_gp(rsp, rdp)) {
+	if (!cpu_needs_another_gp(rsp, rdp) || rsp->fqs_active) {
+		if (cpu_needs_another_gp(rsp, rdp))
+			rsp->fqs_need_gp = 1;
 		if (rnp->completed == rsp->completed) {
 			spin_unlock_irqrestore(&rnp->lock, flags);
 			return;
@@ -1144,11 +1164,9 @@ void rcu_check_callbacks(int cpu, int user)
 /*
  * Scan the leaf rcu_node structures, processing dyntick state for any that
  * have not yet encountered a quiescent state, using the function specified.
- * Returns 1 if the current grace period ends while scanning (possibly
- * because we made it end).
+ * The caller must have suppressed start of new grace periods.
  */
-static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp,
-			       int (*f)(struct rcu_data *))
+static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *))
 {
 	unsigned long bit;
 	int cpu;
@@ -1159,9 +1177,9 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp,
 	rcu_for_each_leaf_node(rsp, rnp) {
 		mask = 0;
 		spin_lock_irqsave(&rnp->lock, flags);
-		if (rnp->completed != lastcomp) {
+		if (!rcu_gp_in_progress(rsp)) {
 			spin_unlock_irqrestore(&rnp->lock, flags);
-			return 1;
+			return;
 		}
 		if (rnp->qsmask == 0) {
 			spin_unlock_irqrestore(&rnp->lock, flags);
@@ -1173,7 +1191,7 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp,
 			if ((rnp->qsmask & bit) != 0 && f(rsp->rda[cpu]))
 				mask |= bit;
 		}
-		if (mask != 0 && rnp->completed == lastcomp) {
+		if (mask != 0) {
 
 			/* rcu_report_qs_rnp() releases rnp->lock. */
 			rcu_report_qs_rnp(mask, rsp, rnp, flags);
@@ -1181,7 +1199,6 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp,
 		}
 		spin_unlock_irqrestore(&rnp->lock, flags);
 	}
-	return 0;
 }
 
 /*
@@ -1191,10 +1208,7 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp,
 static void force_quiescent_state(struct rcu_state *rsp, int relaxed)
 {
 	unsigned long flags;
-	long lastcomp;
 	struct rcu_node *rnp = rcu_get_root(rsp);
-	u8 signaled;
-	u8 forcenow;
 
 	if (!rcu_gp_in_progress(rsp))
 		return;  /* No grace period in progress, nothing to force. */
@@ -1204,19 +1218,17 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed)
 	}
 	if (relaxed &&
 	    (long)(rsp->jiffies_force_qs - jiffies) >= 0)
-		goto unlock_ret; /* no emergency and done recently. */
+		goto unlock_fqs_ret; /* no emergency and done recently. */
 	rsp->n_force_qs++;
-	spin_lock(&rnp->lock);
-	lastcomp = rsp->gpnum - 1;
-	signaled = rsp->signaled;
+	spin_lock(&rnp->lock);  /* irqs already disabled */
 	rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS;
 	if(!rcu_gp_in_progress(rsp)) {
 		rsp->n_force_qs_ngp++;
-		spin_unlock(&rnp->lock);
-		goto unlock_ret;  /* no GP in progress, time updated. */
+		spin_unlock(&rnp->lock);  /* irqs remain disabled */
+		goto unlock_fqs_ret;  /* no GP in progress, time updated. */
 	}
-	spin_unlock(&rnp->lock);
-	switch (signaled) {
+	rsp->fqs_active = 1;
+	switch (rsp->signaled) {
 	case RCU_GP_IDLE:
 	case RCU_GP_INIT:
 
@@ -1224,44 +1236,37 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed)
 
 	case RCU_SAVE_DYNTICK:
 
+		spin_unlock(&rnp->lock);  /* irqs remain disabled */
 		if (RCU_SIGNAL_INIT != RCU_SAVE_DYNTICK)
 			break; /* So gcc recognizes the dead code. */
 
 		/* Record dyntick-idle state. */
-		if (rcu_process_dyntick(rsp, lastcomp,
-					dyntick_save_progress_counter))
-			goto unlock_ret;
-		/* fall into next case. */
-
-	case RCU_SAVE_COMPLETED:
-
-		/* Update state, record completion counter. */
-		forcenow = 0;
-		spin_lock(&rnp->lock);
-		if (lastcomp + 1 == rsp->gpnum &&
-		    lastcomp == rsp->completed &&
-		    rsp->signaled == signaled) {
+		force_qs_rnp(rsp, dyntick_save_progress_counter);
+		spin_lock(&rnp->lock);  /* irqs already disabled */
+		if (rcu_gp_in_progress(rsp))
 			rsp->signaled = RCU_FORCE_QS;
-			rsp->completed_fqs = lastcomp;
-			forcenow = signaled == RCU_SAVE_COMPLETED;
-		}
-		spin_unlock(&rnp->lock);
-		if (!forcenow)
-			break;
-		/* fall into next case. */
+		break;
 
 	case RCU_FORCE_QS:
 
 		/* Check dyntick-idle state, send IPI to laggarts. */
-		if (rcu_process_dyntick(rsp, rsp->completed_fqs,
-					rcu_implicit_dynticks_qs))
-			goto unlock_ret;
+		spin_unlock(&rnp->lock);  /* irqs remain disabled */
+		force_qs_rnp(rsp, rcu_implicit_dynticks_qs);
 
 		/* Leave state in case more forcing is required. */
 
+		spin_lock(&rnp->lock);  /* irqs already disabled */
 		break;
 	}
-unlock_ret:
+	rsp->fqs_active = 0;
+	if (rsp->fqs_need_gp) {
+		spin_unlock(&rsp->fqslock); /* irqs remain disabled */
+		rsp->fqs_need_gp = 0;
+		rcu_start_gp(rsp, flags); /* releases rnp->lock */
+		return;
+	}
+	spin_unlock(&rnp->lock);  /* irqs remain disabled */
+unlock_fqs_ret:
 	spin_unlock_irqrestore(&rsp->fqslock, flags);
 }
 
@@ -1806,11 +1811,17 @@ static void __init rcu_init_levelspread(struct rcu_state *rsp)
  */
 static void __init rcu_init_one(struct rcu_state *rsp)
 {
+	static char *buf[] = { "rcu_node_level_0",
+			       "rcu_node_level_1",
+			       "rcu_node_level_2",
+			       "rcu_node_level_3" };  /* Match MAX_RCU_LVLS */
 	int cpustride = 1;
 	int i;
 	int j;
 	struct rcu_node *rnp;
 
+	BUILD_BUG_ON(MAX_RCU_LVLS > ARRAY_SIZE(buf));  /* Fix buf[] init! */
+
 	/* Initialize the level-tracking arrays. */
 
 	for (i = 1; i < NUM_RCU_LVLS; i++)
@@ -1824,7 +1835,8 @@ static void __init rcu_init_one(struct rcu_state *rsp)
 		rnp = rsp->level[i];
 		for (j = 0; j < rsp->levelcnt[i]; j++, rnp++) {
 			spin_lock_init(&rnp->lock);
-			lockdep_set_class(&rnp->lock, &rcu_node_class[i]);
+			lockdep_set_class_and_name(&rnp->lock,
+						   &rcu_node_class[i], buf[i]);
 			rnp->gpnum = 0;
 			rnp->qsmask = 0;
 			rnp->qsmaskinit = 0;
@@ -1876,7 +1888,7 @@ do { \
 
 void __init rcu_init(void)
 {
-	int i;
+	int cpu;
 
 	rcu_bootup_announce();
 #ifdef CONFIG_RCU_CPU_STALL_DETECTOR
@@ -1896,8 +1908,8 @@ void __init rcu_init(void)
 	 * or the scheduler are operational.
 	 */
 	cpu_notifier(rcu_cpu_notify, 0);
-	for_each_online_cpu(i)
-		rcu_cpu_notify(NULL, CPU_UP_PREPARE, (void *)(long)i);
+	for_each_online_cpu(cpu)
+		rcu_cpu_notify(NULL, CPU_UP_PREPARE, (void *)(long)cpu);
 }
 
 #include "rcutree_plugin.h"
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index d2a0046f63b2..d9d032abd665 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -237,12 +237,11 @@ struct rcu_data {
 #define RCU_GP_IDLE		0	/* No grace period in progress. */
 #define RCU_GP_INIT		1	/* Grace period being initialized. */
 #define RCU_SAVE_DYNTICK	2	/* Need to scan dyntick state. */
-#define RCU_SAVE_COMPLETED	3	/* Need to save rsp->completed. */
-#define RCU_FORCE_QS		4	/* Need to force quiescent state. */
+#define RCU_FORCE_QS		3	/* Need to force quiescent state. */
 #ifdef CONFIG_NO_HZ
 #define RCU_SIGNAL_INIT		RCU_SAVE_DYNTICK
 #else /* #ifdef CONFIG_NO_HZ */
-#define RCU_SIGNAL_INIT		RCU_SAVE_COMPLETED
+#define RCU_SIGNAL_INIT		RCU_FORCE_QS
 #endif /* #else #ifdef CONFIG_NO_HZ */
 
 #define RCU_JIFFIES_TILL_FORCE_QS	 3	/* for rsp->jiffies_force_qs */
@@ -277,6 +276,13 @@ struct rcu_state {
 
 	u8	signaled ____cacheline_internodealigned_in_smp;
 						/* Force QS state. */
+	u8	fqs_active;			/* force_quiescent_state() */
+						/*  is running. */
+	u8	fqs_need_gp;			/* A CPU was prevented from */
+						/*  starting a new grace */
+						/*  period because */
+						/*  force_quiescent_state() */
+						/*  was running. */
 	long	gpnum;				/* Current gp number. */
 	long	completed;			/* # of last completed gp. */
 
@@ -294,8 +300,6 @@ struct rcu_state {
 	long orphan_qlen;			/* Number of orphaned cbs. */
 	spinlock_t fqslock;			/* Only one task forcing */
 						/*  quiescent states. */
-	long	completed_fqs;			/* Value of completed @ snap. */
-						/*  Protected by fqslock. */
 	unsigned long jiffies_force_qs;		/* Time at which to invoke */
 						/*  force_quiescent_state(). */
 	unsigned long n_force_qs;		/* Number of calls to */
@@ -319,8 +323,6 @@ struct rcu_state {
 #define RCU_OFL_TASKS_EXP_GP	0x2		/* Tasks blocking expedited */
 						/*  GP were moved to root. */
 
-#ifdef RCU_TREE_NONCORE
-
 /*
  * RCU implementation internal declarations:
  */
@@ -335,7 +337,7 @@ extern struct rcu_state rcu_preempt_state;
 DECLARE_PER_CPU(struct rcu_data, rcu_preempt_data);
 #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
 
-#else /* #ifdef RCU_TREE_NONCORE */
+#ifndef RCU_TREE_NONCORE
 
 /* Forward declarations for rcutree_plugin.h */
 static void rcu_bootup_announce(void);
@@ -368,4 +370,4 @@ static void __cpuinit rcu_preempt_init_percpu_data(int cpu);
 static void rcu_preempt_send_cbs_to_orphanage(void);
 static void __init __rcu_init_preempt(void);
 
-#endif /* #else #ifdef RCU_TREE_NONCORE */
+#endif /* #ifndef RCU_TREE_NONCORE */
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 37fbccdf41d5..e77cdf321e13 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -62,6 +62,15 @@ long rcu_batches_completed(void)
 EXPORT_SYMBOL_GPL(rcu_batches_completed);
 
 /*
+ * Force a quiescent state for preemptible RCU.
+ */
+void rcu_force_quiescent_state(void)
+{
+	force_quiescent_state(&rcu_preempt_state, 0);
+}
+EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
+
+/*
  * Record a preemptable-RCU quiescent state for the specified CPU.  Note
  * that this just means that the task currently running on the CPU is
  * not in a quiescent state.  There might be any number of tasks blocked
@@ -295,6 +304,9 @@ void __rcu_read_unlock(void)
 	if (--ACCESS_ONCE(t->rcu_read_lock_nesting) == 0 &&
 	    unlikely(ACCESS_ONCE(t->rcu_read_unlock_special)))
 		rcu_read_unlock_special(t);
+#ifdef CONFIG_PROVE_LOCKING
+	WARN_ON_ONCE(ACCESS_ONCE(t->rcu_read_lock_nesting) < 0);
+#endif /* #ifdef CONFIG_PROVE_LOCKING */
 }
 EXPORT_SYMBOL_GPL(__rcu_read_unlock);
 
@@ -713,6 +725,16 @@ long rcu_batches_completed(void)
 EXPORT_SYMBOL_GPL(rcu_batches_completed);
 
 /*
+ * Force a quiescent state for RCU, which, because there is no preemptible
+ * RCU, becomes the same as rcu-sched.
+ */
+void rcu_force_quiescent_state(void)
+{
+	rcu_sched_force_quiescent_state();
+}
+EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
+
+/*
  * Because preemptable RCU does not exist, we never have to check for
  * CPUs being in quiescent states.
  */
diff --git a/kernel/resource.c b/kernel/resource.c
index 24e9e60c1459..4e9d87fd7bc5 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -327,6 +327,19 @@ int walk_system_ram_range(unsigned long start_pfn, unsigned long nr_pages,
 
 #endif
 
+static int __is_ram(unsigned long pfn, unsigned long nr_pages, void *arg)
+{
+	return 1;
+}
+/*
+ * This generic page_is_ram() returns true if specified address is
+ * registered as "System RAM" in iomem_resource list.
+ */
+int __weak page_is_ram(unsigned long pfn)
+{
+	return walk_system_ram_range(pfn, 1, NULL, __is_ram) == 1;
+}
+
 /*
  * Find empty slot in the resource tree given range and alignment.
  */
diff --git a/kernel/sched.c b/kernel/sched.c
index 92ab8b8eff22..673d3867804a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -233,7 +233,7 @@ static void destroy_rt_bandwidth(struct rt_bandwidth *rt_b)
  */
 static DEFINE_MUTEX(sched_domains_mutex);
 
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
 
 #include <linux/cgroup.h>
 
@@ -243,13 +243,7 @@ static LIST_HEAD(task_groups);
 
 /* task group related information */
 struct task_group {
-#ifdef CONFIG_CGROUP_SCHED
 	struct cgroup_subsys_state css;
-#endif
-
-#ifdef CONFIG_USER_SCHED
-	uid_t uid;
-#endif
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	/* schedulable entities of this group on each cpu */
@@ -274,35 +268,7 @@ struct task_group {
 	struct list_head children;
 };
 
-#ifdef CONFIG_USER_SCHED
-
-/* Helper function to pass uid information to create_sched_user() */
-void set_tg_uid(struct user_struct *user)
-{
-	user->tg->uid = user->uid;
-}
-
-/*
- * Root task group.
- *	Every UID task group (including init_task_group aka UID-0) will
- *	be a child to this group.
- */
-struct task_group root_task_group;
-
-#ifdef CONFIG_FAIR_GROUP_SCHED
-/* Default task group's sched entity on each cpu */
-static DEFINE_PER_CPU(struct sched_entity, init_sched_entity);
-/* Default task group's cfs_rq on each cpu */
-static DEFINE_PER_CPU_SHARED_ALIGNED(struct cfs_rq, init_tg_cfs_rq);
-#endif /* CONFIG_FAIR_GROUP_SCHED */
-
-#ifdef CONFIG_RT_GROUP_SCHED
-static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity);
-static DEFINE_PER_CPU_SHARED_ALIGNED(struct rt_rq, init_rt_rq_var);
-#endif /* CONFIG_RT_GROUP_SCHED */
-#else /* !CONFIG_USER_SCHED */
 #define root_task_group init_task_group
-#endif /* CONFIG_USER_SCHED */
 
 /* task_group_lock serializes add/remove of task groups and also changes to
  * a task group's cpu shares.
@@ -318,11 +284,7 @@ static int root_task_group_empty(void)
 }
 #endif
 
-#ifdef CONFIG_USER_SCHED
-# define INIT_TASK_GROUP_LOAD	(2*NICE_0_LOAD)
-#else /* !CONFIG_USER_SCHED */
 # define INIT_TASK_GROUP_LOAD	NICE_0_LOAD
-#endif /* CONFIG_USER_SCHED */
 
 /*
  * A weight of 0 or 1 can cause arithmetics problems.
@@ -348,11 +310,7 @@ static inline struct task_group *task_group(struct task_struct *p)
 {
 	struct task_group *tg;
 
-#ifdef CONFIG_USER_SCHED
-	rcu_read_lock();
-	tg = __task_cred(p)->user->tg;
-	rcu_read_unlock();
-#elif defined(CONFIG_CGROUP_SCHED)
+#ifdef CONFIG_CGROUP_SCHED
 	tg = container_of(task_subsys_state(p, cpu_cgroup_subsys_id),
 				struct task_group, css);
 #else
@@ -383,7 +341,7 @@ static inline struct task_group *task_group(struct task_struct *p)
 	return NULL;
 }
 
-#endif	/* CONFIG_GROUP_SCHED */
+#endif	/* CONFIG_CGROUP_SCHED */
 
 /* CFS-related fields in a runqueue */
 struct cfs_rq {
@@ -1390,32 +1348,6 @@ static const u32 prio_to_wmult[40] = {
  /*  15 */ 119304647, 148102320, 186737708, 238609294, 286331153,
 };
 
-static void activate_task(struct rq *rq, struct task_struct *p, int wakeup);
-
-/*
- * runqueue iterator, to support SMP load-balancing between different
- * scheduling classes, without having to expose their internal data
- * structures to the load-balancing proper:
- */
-struct rq_iterator {
-	void *arg;
-	struct task_struct *(*start)(void *);
-	struct task_struct *(*next)(void *);
-};
-
-#ifdef CONFIG_SMP
-static unsigned long
-balance_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest,
-	      unsigned long max_load_move, struct sched_domain *sd,
-	      enum cpu_idle_type idle, int *all_pinned,
-	      int *this_best_prio, struct rq_iterator *iterator);
-
-static int
-iter_move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		   struct sched_domain *sd, enum cpu_idle_type idle,
-		   struct rq_iterator *iterator);
-#endif
-
 /* Time spent by the tasks of the cpu accounting group executing in ... */
 enum cpuacct_stat_index {
 	CPUACCT_STAT_USER,	/* ... user mode */
@@ -1701,16 +1633,6 @@ static void update_shares(struct sched_domain *sd)
 	}
 }
 
-static void update_shares_locked(struct rq *rq, struct sched_domain *sd)
-{
-	if (root_task_group_empty())
-		return;
-
-	raw_spin_unlock(&rq->lock);
-	update_shares(sd);
-	raw_spin_lock(&rq->lock);
-}
-
 static void update_h_load(long cpu)
 {
 	if (root_task_group_empty())
@@ -1725,10 +1647,6 @@ static inline void update_shares(struct sched_domain *sd)
 {
 }
 
-static inline void update_shares_locked(struct rq *rq, struct sched_domain *sd)
-{
-}
-
 #endif
 
 #ifdef CONFIG_PREEMPT
@@ -1805,6 +1723,51 @@ static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest)
 	raw_spin_unlock(&busiest->lock);
 	lock_set_subclass(&this_rq->lock.dep_map, 0, _RET_IP_);
 }
+
+/*
+ * double_rq_lock - safely lock two runqueues
+ *
+ * Note this does not disable interrupts like task_rq_lock,
+ * you need to do so manually before calling.
+ */
+static void double_rq_lock(struct rq *rq1, struct rq *rq2)
+	__acquires(rq1->lock)
+	__acquires(rq2->lock)
+{
+	BUG_ON(!irqs_disabled());
+	if (rq1 == rq2) {
+		raw_spin_lock(&rq1->lock);
+		__acquire(rq2->lock);	/* Fake it out ;) */
+	} else {
+		if (rq1 < rq2) {
+			raw_spin_lock(&rq1->lock);
+			raw_spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING);
+		} else {
+			raw_spin_lock(&rq2->lock);
+			raw_spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING);
+		}
+	}
+	update_rq_clock(rq1);
+	update_rq_clock(rq2);
+}
+
+/*
+ * double_rq_unlock - safely unlock two runqueues
+ *
+ * Note this does not restore interrupts like task_rq_unlock,
+ * you need to do so manually after calling.
+ */
+static void double_rq_unlock(struct rq *rq1, struct rq *rq2)
+	__releases(rq1->lock)
+	__releases(rq2->lock)
+{
+	raw_spin_unlock(&rq1->lock);
+	if (rq1 != rq2)
+		raw_spin_unlock(&rq2->lock);
+	else
+		__release(rq2->lock);
+}
+
 #endif
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
@@ -1834,18 +1797,14 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 #endif
 }
 
-#include "sched_stats.h"
-#include "sched_idletask.c"
-#include "sched_fair.c"
-#include "sched_rt.c"
-#ifdef CONFIG_SCHED_DEBUG
-# include "sched_debug.c"
-#endif
+static const struct sched_class rt_sched_class;
 
 #define sched_class_highest (&rt_sched_class)
 #define for_each_class(class) \
    for (class = sched_class_highest; class; class = class->next)
 
+#include "sched_stats.h"
+
 static void inc_nr_running(struct rq *rq)
 {
 	rq->nr_running++;
@@ -1883,13 +1842,14 @@ static void update_avg(u64 *avg, u64 sample)
 	*avg += diff >> 3;
 }
 
-static void enqueue_task(struct rq *rq, struct task_struct *p, int wakeup)
+static void
+enqueue_task(struct rq *rq, struct task_struct *p, int wakeup, bool head)
 {
 	if (wakeup)
 		p->se.start_runtime = p->se.sum_exec_runtime;
 
 	sched_info_queued(p);
-	p->sched_class->enqueue_task(rq, p, wakeup);
+	p->sched_class->enqueue_task(rq, p, wakeup, head);
 	p->se.on_rq = 1;
 }
 
@@ -1912,6 +1872,37 @@ static void dequeue_task(struct rq *rq, struct task_struct *p, int sleep)
 }
 
 /*
+ * activate_task - move a task to the runqueue.
+ */
+static void activate_task(struct rq *rq, struct task_struct *p, int wakeup)
+{
+	if (task_contributes_to_load(p))
+		rq->nr_uninterruptible--;
+
+	enqueue_task(rq, p, wakeup, false);
+	inc_nr_running(rq);
+}
+
+/*
+ * deactivate_task - remove a task from the runqueue.
+ */
+static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep)
+{
+	if (task_contributes_to_load(p))
+		rq->nr_uninterruptible++;
+
+	dequeue_task(rq, p, sleep);
+	dec_nr_running(rq);
+}
+
+#include "sched_idletask.c"
+#include "sched_fair.c"
+#include "sched_rt.c"
+#ifdef CONFIG_SCHED_DEBUG
+# include "sched_debug.c"
+#endif
+
+/*
  * __normal_prio - return the priority that is based on the static prio
  */
 static inline int __normal_prio(struct task_struct *p)
@@ -1957,30 +1948,6 @@ static int effective_prio(struct task_struct *p)
 	return p->prio;
 }
 
-/*
- * activate_task - move a task to the runqueue.
- */
-static void activate_task(struct rq *rq, struct task_struct *p, int wakeup)
-{
-	if (task_contributes_to_load(p))
-		rq->nr_uninterruptible--;
-
-	enqueue_task(rq, p, wakeup);
-	inc_nr_running(rq);
-}
-
-/*
- * deactivate_task - remove a task from the runqueue.
- */
-static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep)
-{
-	if (task_contributes_to_load(p))
-		rq->nr_uninterruptible++;
-
-	dequeue_task(rq, p, sleep);
-	dec_nr_running(rq);
-}
-
 /**
  * task_curr - is this task currently executing on a CPU?
  * @p: the task in question.
@@ -2794,7 +2761,13 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev)
 	 */
 	prev_state = prev->state;
 	finish_arch_switch(prev);
-	perf_event_task_sched_in(current, cpu_of(rq));
+#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
+	local_irq_disable();
+#endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */
+	perf_event_task_sched_in(current);
+#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
+	local_irq_enable();
+#endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */
 	finish_lock_switch(rq, prev);
 
 	fire_sched_in_preempt_notifiers(current);
@@ -3099,50 +3072,6 @@ static void update_cpu_load(struct rq *this_rq)
 #ifdef CONFIG_SMP
 
 /*
- * double_rq_lock - safely lock two runqueues
- *
- * Note this does not disable interrupts like task_rq_lock,
- * you need to do so manually before calling.
- */
-static void double_rq_lock(struct rq *rq1, struct rq *rq2)
-	__acquires(rq1->lock)
-	__acquires(rq2->lock)
-{
-	BUG_ON(!irqs_disabled());
-	if (rq1 == rq2) {
-		raw_spin_lock(&rq1->lock);
-		__acquire(rq2->lock);	/* Fake it out ;) */
-	} else {
-		if (rq1 < rq2) {
-			raw_spin_lock(&rq1->lock);
-			raw_spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING);
-		} else {
-			raw_spin_lock(&rq2->lock);
-			raw_spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING);
-		}
-	}
-	update_rq_clock(rq1);
-	update_rq_clock(rq2);
-}
-
-/*
- * double_rq_unlock - safely unlock two runqueues
- *
- * Note this does not restore interrupts like task_rq_unlock,
- * you need to do so manually after calling.
- */
-static void double_rq_unlock(struct rq *rq1, struct rq *rq2)
-	__releases(rq1->lock)
-	__releases(rq2->lock)
-{
-	raw_spin_unlock(&rq1->lock);
-	if (rq1 != rq2)
-		raw_spin_unlock(&rq2->lock);
-	else
-		__release(rq2->lock);
-}
-
-/*
  * sched_exec - execve() is a valuable balancing opportunity, because at
  * this point the task has the smallest effective memory and cache footprint.
  */
@@ -3190,1771 +3119,6 @@ again:
 	task_rq_unlock(rq, &flags);
 }
 
-/*
- * pull_task - move a task from a remote runqueue to the local runqueue.
- * Both runqueues must be locked.
- */
-static void pull_task(struct rq *src_rq, struct task_struct *p,
-		      struct rq *this_rq, int this_cpu)
-{
-	deactivate_task(src_rq, p, 0);
-	set_task_cpu(p, this_cpu);
-	activate_task(this_rq, p, 0);
-	check_preempt_curr(this_rq, p, 0);
-}
-
-/*
- * can_migrate_task - may task p from runqueue rq be migrated to this_cpu?
- */
-static
-int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu,
-		     struct sched_domain *sd, enum cpu_idle_type idle,
-		     int *all_pinned)
-{
-	int tsk_cache_hot = 0;
-	/*
-	 * We do not migrate tasks that are:
-	 * 1) running (obviously), or
-	 * 2) cannot be migrated to this CPU due to cpus_allowed, or
-	 * 3) are cache-hot on their current CPU.
-	 */
-	if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) {
-		schedstat_inc(p, se.nr_failed_migrations_affine);
-		return 0;
-	}
-	*all_pinned = 0;
-
-	if (task_running(rq, p)) {
-		schedstat_inc(p, se.nr_failed_migrations_running);
-		return 0;
-	}
-
-	/*
-	 * Aggressive migration if:
-	 * 1) task is cache cold, or
-	 * 2) too many balance attempts have failed.
-	 */
-
-	tsk_cache_hot = task_hot(p, rq->clock, sd);
-	if (!tsk_cache_hot ||
-		sd->nr_balance_failed > sd->cache_nice_tries) {
-#ifdef CONFIG_SCHEDSTATS
-		if (tsk_cache_hot) {
-			schedstat_inc(sd, lb_hot_gained[idle]);
-			schedstat_inc(p, se.nr_forced_migrations);
-		}
-#endif
-		return 1;
-	}
-
-	if (tsk_cache_hot) {
-		schedstat_inc(p, se.nr_failed_migrations_hot);
-		return 0;
-	}
-	return 1;
-}
-
-static unsigned long
-balance_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest,
-	      unsigned long max_load_move, struct sched_domain *sd,
-	      enum cpu_idle_type idle, int *all_pinned,
-	      int *this_best_prio, struct rq_iterator *iterator)
-{
-	int loops = 0, pulled = 0, pinned = 0;
-	struct task_struct *p;
-	long rem_load_move = max_load_move;
-
-	if (max_load_move == 0)
-		goto out;
-
-	pinned = 1;
-
-	/*
-	 * Start the load-balancing iterator:
-	 */
-	p = iterator->start(iterator->arg);
-next:
-	if (!p || loops++ > sysctl_sched_nr_migrate)
-		goto out;
-
-	if ((p->se.load.weight >> 1) > rem_load_move ||
-	    !can_migrate_task(p, busiest, this_cpu, sd, idle, &pinned)) {
-		p = iterator->next(iterator->arg);
-		goto next;
-	}
-
-	pull_task(busiest, p, this_rq, this_cpu);
-	pulled++;
-	rem_load_move -= p->se.load.weight;
-
-#ifdef CONFIG_PREEMPT
-	/*
-	 * NEWIDLE balancing is a source of latency, so preemptible kernels
-	 * will stop after the first task is pulled to minimize the critical
-	 * section.
-	 */
-	if (idle == CPU_NEWLY_IDLE)
-		goto out;
-#endif
-
-	/*
-	 * We only want to steal up to the prescribed amount of weighted load.
-	 */
-	if (rem_load_move > 0) {
-		if (p->prio < *this_best_prio)
-			*this_best_prio = p->prio;
-		p = iterator->next(iterator->arg);
-		goto next;
-	}
-out:
-	/*
-	 * Right now, this is one of only two places pull_task() is called,
-	 * so we can safely collect pull_task() stats here rather than
-	 * inside pull_task().
-	 */
-	schedstat_add(sd, lb_gained[idle], pulled);
-
-	if (all_pinned)
-		*all_pinned = pinned;
-
-	return max_load_move - rem_load_move;
-}
-
-/*
- * move_tasks tries to move up to max_load_move weighted load from busiest to
- * this_rq, as part of a balancing operation within domain "sd".
- * Returns 1 if successful and 0 otherwise.
- *
- * Called with both runqueues locked.
- */
-static int move_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		      unsigned long max_load_move,
-		      struct sched_domain *sd, enum cpu_idle_type idle,
-		      int *all_pinned)
-{
-	const struct sched_class *class = sched_class_highest;
-	unsigned long total_load_moved = 0;
-	int this_best_prio = this_rq->curr->prio;
-
-	do {
-		total_load_moved +=
-			class->load_balance(this_rq, this_cpu, busiest,
-				max_load_move - total_load_moved,
-				sd, idle, all_pinned, &this_best_prio);
-		class = class->next;
-
-#ifdef CONFIG_PREEMPT
-		/*
-		 * NEWIDLE balancing is a source of latency, so preemptible
-		 * kernels will stop after the first task is pulled to minimize
-		 * the critical section.
-		 */
-		if (idle == CPU_NEWLY_IDLE && this_rq->nr_running)
-			break;
-#endif
-	} while (class && max_load_move > total_load_moved);
-
-	return total_load_moved > 0;
-}
-
-static int
-iter_move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		   struct sched_domain *sd, enum cpu_idle_type idle,
-		   struct rq_iterator *iterator)
-{
-	struct task_struct *p = iterator->start(iterator->arg);
-	int pinned = 0;
-
-	while (p) {
-		if (can_migrate_task(p, busiest, this_cpu, sd, idle, &pinned)) {
-			pull_task(busiest, p, this_rq, this_cpu);
-			/*
-			 * Right now, this is only the second place pull_task()
-			 * is called, so we can safely collect pull_task()
-			 * stats here rather than inside pull_task().
-			 */
-			schedstat_inc(sd, lb_gained[idle]);
-
-			return 1;
-		}
-		p = iterator->next(iterator->arg);
-	}
-
-	return 0;
-}
-
-/*
- * move_one_task tries to move exactly one task from busiest to this_rq, as
- * part of active balancing operations within "domain".
- * Returns 1 if successful and 0 otherwise.
- *
- * Called with both runqueues locked.
- */
-static int move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest,
-			 struct sched_domain *sd, enum cpu_idle_type idle)
-{
-	const struct sched_class *class;
-
-	for_each_class(class) {
-		if (class->move_one_task(this_rq, this_cpu, busiest, sd, idle))
-			return 1;
-	}
-
-	return 0;
-}
-/********** Helpers for find_busiest_group ************************/
-/*
- * sd_lb_stats - Structure to store the statistics of a sched_domain
- * 		during load balancing.
- */
-struct sd_lb_stats {
-	struct sched_group *busiest; /* Busiest group in this sd */
-	struct sched_group *this;  /* Local group in this sd */
-	unsigned long total_load;  /* Total load of all groups in sd */
-	unsigned long total_pwr;   /*	Total power of all groups in sd */
-	unsigned long avg_load;	   /* Average load across all groups in sd */
-
-	/** Statistics of this group */
-	unsigned long this_load;
-	unsigned long this_load_per_task;
-	unsigned long this_nr_running;
-
-	/* Statistics of the busiest group */
-	unsigned long max_load;
-	unsigned long busiest_load_per_task;
-	unsigned long busiest_nr_running;
-
-	int group_imb; /* Is there imbalance in this sd */
-#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
-	int power_savings_balance; /* Is powersave balance needed for this sd */
-	struct sched_group *group_min; /* Least loaded group in sd */
-	struct sched_group *group_leader; /* Group which relieves group_min */
-	unsigned long min_load_per_task; /* load_per_task in group_min */
-	unsigned long leader_nr_running; /* Nr running of group_leader */
-	unsigned long min_nr_running; /* Nr running of group_min */
-#endif
-};
-
-/*
- * sg_lb_stats - stats of a sched_group required for load_balancing
- */
-struct sg_lb_stats {
-	unsigned long avg_load; /*Avg load across the CPUs of the group */
-	unsigned long group_load; /* Total load over the CPUs of the group */
-	unsigned long sum_nr_running; /* Nr tasks running in the group */
-	unsigned long sum_weighted_load; /* Weighted load of group's tasks */
-	unsigned long group_capacity;
-	int group_imb; /* Is there an imbalance in the group ? */
-};
-
-/**
- * group_first_cpu - Returns the first cpu in the cpumask of a sched_group.
- * @group: The group whose first cpu is to be returned.
- */
-static inline unsigned int group_first_cpu(struct sched_group *group)
-{
-	return cpumask_first(sched_group_cpus(group));
-}
-
-/**
- * get_sd_load_idx - Obtain the load index for a given sched domain.
- * @sd: The sched_domain whose load_idx is to be obtained.
- * @idle: The Idle status of the CPU for whose sd load_icx is obtained.
- */
-static inline int get_sd_load_idx(struct sched_domain *sd,
-					enum cpu_idle_type idle)
-{
-	int load_idx;
-
-	switch (idle) {
-	case CPU_NOT_IDLE:
-		load_idx = sd->busy_idx;
-		break;
-
-	case CPU_NEWLY_IDLE:
-		load_idx = sd->newidle_idx;
-		break;
-	default:
-		load_idx = sd->idle_idx;
-		break;
-	}
-
-	return load_idx;
-}
-
-
-#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
-/**
- * init_sd_power_savings_stats - Initialize power savings statistics for
- * the given sched_domain, during load balancing.
- *
- * @sd: Sched domain whose power-savings statistics are to be initialized.
- * @sds: Variable containing the statistics for sd.
- * @idle: Idle status of the CPU at which we're performing load-balancing.
- */
-static inline void init_sd_power_savings_stats(struct sched_domain *sd,
-	struct sd_lb_stats *sds, enum cpu_idle_type idle)
-{
-	/*
-	 * Busy processors will not participate in power savings
-	 * balance.
-	 */
-	if (idle == CPU_NOT_IDLE || !(sd->flags & SD_POWERSAVINGS_BALANCE))
-		sds->power_savings_balance = 0;
-	else {
-		sds->power_savings_balance = 1;
-		sds->min_nr_running = ULONG_MAX;
-		sds->leader_nr_running = 0;
-	}
-}
-
-/**
- * update_sd_power_savings_stats - Update the power saving stats for a
- * sched_domain while performing load balancing.
- *
- * @group: sched_group belonging to the sched_domain under consideration.
- * @sds: Variable containing the statistics of the sched_domain
- * @local_group: Does group contain the CPU for which we're performing
- * 		load balancing ?
- * @sgs: Variable containing the statistics of the group.
- */
-static inline void update_sd_power_savings_stats(struct sched_group *group,
-	struct sd_lb_stats *sds, int local_group, struct sg_lb_stats *sgs)
-{
-
-	if (!sds->power_savings_balance)
-		return;
-
-	/*
-	 * If the local group is idle or completely loaded
-	 * no need to do power savings balance at this domain
-	 */
-	if (local_group && (sds->this_nr_running >= sgs->group_capacity ||
-				!sds->this_nr_running))
-		sds->power_savings_balance = 0;
-
-	/*
-	 * If a group is already running at full capacity or idle,
-	 * don't include that group in power savings calculations
-	 */
-	if (!sds->power_savings_balance ||
-		sgs->sum_nr_running >= sgs->group_capacity ||
-		!sgs->sum_nr_running)
-		return;
-
-	/*
-	 * Calculate the group which has the least non-idle load.
-	 * This is the group from where we need to pick up the load
-	 * for saving power
-	 */
-	if ((sgs->sum_nr_running < sds->min_nr_running) ||
-	    (sgs->sum_nr_running == sds->min_nr_running &&
-	     group_first_cpu(group) > group_first_cpu(sds->group_min))) {
-		sds->group_min = group;
-		sds->min_nr_running = sgs->sum_nr_running;
-		sds->min_load_per_task = sgs->sum_weighted_load /
-						sgs->sum_nr_running;
-	}
-
-	/*
-	 * Calculate the group which is almost near its
-	 * capacity but still has some space to pick up some load
-	 * from other group and save more power
-	 */
-	if (sgs->sum_nr_running + 1 > sgs->group_capacity)
-		return;
-
-	if (sgs->sum_nr_running > sds->leader_nr_running ||
-	    (sgs->sum_nr_running == sds->leader_nr_running &&
-	     group_first_cpu(group) < group_first_cpu(sds->group_leader))) {
-		sds->group_leader = group;
-		sds->leader_nr_running = sgs->sum_nr_running;
-	}
-}
-
-/**
- * check_power_save_busiest_group - see if there is potential for some power-savings balance
- * @sds: Variable containing the statistics of the sched_domain
- *	under consideration.
- * @this_cpu: Cpu at which we're currently performing load-balancing.
- * @imbalance: Variable to store the imbalance.
- *
- * Description:
- * Check if we have potential to perform some power-savings balance.
- * If yes, set the busiest group to be the least loaded group in the
- * sched_domain, so that it's CPUs can be put to idle.
- *
- * Returns 1 if there is potential to perform power-savings balance.
- * Else returns 0.
- */
-static inline int check_power_save_busiest_group(struct sd_lb_stats *sds,
-					int this_cpu, unsigned long *imbalance)
-{
-	if (!sds->power_savings_balance)
-		return 0;
-
-	if (sds->this != sds->group_leader ||
-			sds->group_leader == sds->group_min)
-		return 0;
-
-	*imbalance = sds->min_load_per_task;
-	sds->busiest = sds->group_min;
-
-	return 1;
-
-}
-#else /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */
-static inline void init_sd_power_savings_stats(struct sched_domain *sd,
-	struct sd_lb_stats *sds, enum cpu_idle_type idle)
-{
-	return;
-}
-
-static inline void update_sd_power_savings_stats(struct sched_group *group,
-	struct sd_lb_stats *sds, int local_group, struct sg_lb_stats *sgs)
-{
-	return;
-}
-
-static inline int check_power_save_busiest_group(struct sd_lb_stats *sds,
-					int this_cpu, unsigned long *imbalance)
-{
-	return 0;
-}
-#endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */
-
-
-unsigned long default_scale_freq_power(struct sched_domain *sd, int cpu)
-{
-	return SCHED_LOAD_SCALE;
-}
-
-unsigned long __weak arch_scale_freq_power(struct sched_domain *sd, int cpu)
-{
-	return default_scale_freq_power(sd, cpu);
-}
-
-unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu)
-{
-	unsigned long weight = cpumask_weight(sched_domain_span(sd));
-	unsigned long smt_gain = sd->smt_gain;
-
-	smt_gain /= weight;
-
-	return smt_gain;
-}
-
-unsigned long __weak arch_scale_smt_power(struct sched_domain *sd, int cpu)
-{
-	return default_scale_smt_power(sd, cpu);
-}
-
-unsigned long scale_rt_power(int cpu)
-{
-	struct rq *rq = cpu_rq(cpu);
-	u64 total, available;
-
-	sched_avg_update(rq);
-
-	total = sched_avg_period() + (rq->clock - rq->age_stamp);
-	available = total - rq->rt_avg;
-
-	if (unlikely((s64)total < SCHED_LOAD_SCALE))
-		total = SCHED_LOAD_SCALE;
-
-	total >>= SCHED_LOAD_SHIFT;
-
-	return div_u64(available, total);
-}
-
-static void update_cpu_power(struct sched_domain *sd, int cpu)
-{
-	unsigned long weight = cpumask_weight(sched_domain_span(sd));
-	unsigned long power = SCHED_LOAD_SCALE;
-	struct sched_group *sdg = sd->groups;
-
-	if (sched_feat(ARCH_POWER))
-		power *= arch_scale_freq_power(sd, cpu);
-	else
-		power *= default_scale_freq_power(sd, cpu);
-
-	power >>= SCHED_LOAD_SHIFT;
-
-	if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) {
-		if (sched_feat(ARCH_POWER))
-			power *= arch_scale_smt_power(sd, cpu);
-		else
-			power *= default_scale_smt_power(sd, cpu);
-
-		power >>= SCHED_LOAD_SHIFT;
-	}
-
-	power *= scale_rt_power(cpu);
-	power >>= SCHED_LOAD_SHIFT;
-
-	if (!power)
-		power = 1;
-
-	sdg->cpu_power = power;
-}
-
-static void update_group_power(struct sched_domain *sd, int cpu)
-{
-	struct sched_domain *child = sd->child;
-	struct sched_group *group, *sdg = sd->groups;
-	unsigned long power;
-
-	if (!child) {
-		update_cpu_power(sd, cpu);
-		return;
-	}
-
-	power = 0;
-
-	group = child->groups;
-	do {
-		power += group->cpu_power;
-		group = group->next;
-	} while (group != child->groups);
-
-	sdg->cpu_power = power;
-}
-
-/**
- * update_sg_lb_stats - Update sched_group's statistics for load balancing.
- * @sd: The sched_domain whose statistics are to be updated.
- * @group: sched_group whose statistics are to be updated.
- * @this_cpu: Cpu for which load balance is currently performed.
- * @idle: Idle status of this_cpu
- * @load_idx: Load index of sched_domain of this_cpu for load calc.
- * @sd_idle: Idle status of the sched_domain containing group.
- * @local_group: Does group contain this_cpu.
- * @cpus: Set of cpus considered for load balancing.
- * @balance: Should we balance.
- * @sgs: variable to hold the statistics for this group.
- */
-static inline void update_sg_lb_stats(struct sched_domain *sd,
-			struct sched_group *group, int this_cpu,
-			enum cpu_idle_type idle, int load_idx, int *sd_idle,
-			int local_group, const struct cpumask *cpus,
-			int *balance, struct sg_lb_stats *sgs)
-{
-	unsigned long load, max_cpu_load, min_cpu_load;
-	int i;
-	unsigned int balance_cpu = -1, first_idle_cpu = 0;
-	unsigned long sum_avg_load_per_task;
-	unsigned long avg_load_per_task;
-
-	if (local_group) {
-		balance_cpu = group_first_cpu(group);
-		if (balance_cpu == this_cpu)
-			update_group_power(sd, this_cpu);
-	}
-
-	/* Tally up the load of all CPUs in the group */
-	sum_avg_load_per_task = avg_load_per_task = 0;
-	max_cpu_load = 0;
-	min_cpu_load = ~0UL;
-
-	for_each_cpu_and(i, sched_group_cpus(group), cpus) {
-		struct rq *rq = cpu_rq(i);
-
-		if (*sd_idle && rq->nr_running)
-			*sd_idle = 0;
-
-		/* Bias balancing toward cpus of our domain */
-		if (local_group) {
-			if (idle_cpu(i) && !first_idle_cpu) {
-				first_idle_cpu = 1;
-				balance_cpu = i;
-			}
-
-			load = target_load(i, load_idx);
-		} else {
-			load = source_load(i, load_idx);
-			if (load > max_cpu_load)
-				max_cpu_load = load;
-			if (min_cpu_load > load)
-				min_cpu_load = load;
-		}
-
-		sgs->group_load += load;
-		sgs->sum_nr_running += rq->nr_running;
-		sgs->sum_weighted_load += weighted_cpuload(i);
-
-		sum_avg_load_per_task += cpu_avg_load_per_task(i);
-	}
-
-	/*
-	 * First idle cpu or the first cpu(busiest) in this sched group
-	 * is eligible for doing load balancing at this and above
-	 * domains. In the newly idle case, we will allow all the cpu's
-	 * to do the newly idle load balance.
-	 */
-	if (idle != CPU_NEWLY_IDLE && local_group &&
-	    balance_cpu != this_cpu && balance) {
-		*balance = 0;
-		return;
-	}
-
-	/* Adjust by relative CPU power of the group */
-	sgs->avg_load = (sgs->group_load * SCHED_LOAD_SCALE) / group->cpu_power;
-
-
-	/*
-	 * Consider the group unbalanced when the imbalance is larger
-	 * than the average weight of two tasks.
-	 *
-	 * APZ: with cgroup the avg task weight can vary wildly and
-	 *      might not be a suitable number - should we keep a
-	 *      normalized nr_running number somewhere that negates
-	 *      the hierarchy?
-	 */
-	avg_load_per_task = (sum_avg_load_per_task * SCHED_LOAD_SCALE) /
-		group->cpu_power;
-
-	if ((max_cpu_load - min_cpu_load) > 2*avg_load_per_task)
-		sgs->group_imb = 1;
-
-	sgs->group_capacity =
-		DIV_ROUND_CLOSEST(group->cpu_power, SCHED_LOAD_SCALE);
-}
-
-/**
- * update_sd_lb_stats - Update sched_group's statistics for load balancing.
- * @sd: sched_domain whose statistics are to be updated.
- * @this_cpu: Cpu for which load balance is currently performed.
- * @idle: Idle status of this_cpu
- * @sd_idle: Idle status of the sched_domain containing group.
- * @cpus: Set of cpus considered for load balancing.
- * @balance: Should we balance.
- * @sds: variable to hold the statistics for this sched_domain.
- */
-static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu,
-			enum cpu_idle_type idle, int *sd_idle,
-			const struct cpumask *cpus, int *balance,
-			struct sd_lb_stats *sds)
-{
-	struct sched_domain *child = sd->child;
-	struct sched_group *group = sd->groups;
-	struct sg_lb_stats sgs;
-	int load_idx, prefer_sibling = 0;
-
-	if (child && child->flags & SD_PREFER_SIBLING)
-		prefer_sibling = 1;
-
-	init_sd_power_savings_stats(sd, sds, idle);
-	load_idx = get_sd_load_idx(sd, idle);
-
-	do {
-		int local_group;
-
-		local_group = cpumask_test_cpu(this_cpu,
-					       sched_group_cpus(group));
-		memset(&sgs, 0, sizeof(sgs));
-		update_sg_lb_stats(sd, group, this_cpu, idle, load_idx, sd_idle,
-				local_group, cpus, balance, &sgs);
-
-		if (local_group && balance && !(*balance))
-			return;
-
-		sds->total_load += sgs.group_load;
-		sds->total_pwr += group->cpu_power;
-
-		/*
-		 * In case the child domain prefers tasks go to siblings
-		 * first, lower the group capacity to one so that we'll try
-		 * and move all the excess tasks away.
-		 */
-		if (prefer_sibling)
-			sgs.group_capacity = min(sgs.group_capacity, 1UL);
-
-		if (local_group) {
-			sds->this_load = sgs.avg_load;
-			sds->this = group;
-			sds->this_nr_running = sgs.sum_nr_running;
-			sds->this_load_per_task = sgs.sum_weighted_load;
-		} else if (sgs.avg_load > sds->max_load &&
-			   (sgs.sum_nr_running > sgs.group_capacity ||
-				sgs.group_imb)) {
-			sds->max_load = sgs.avg_load;
-			sds->busiest = group;
-			sds->busiest_nr_running = sgs.sum_nr_running;
-			sds->busiest_load_per_task = sgs.sum_weighted_load;
-			sds->group_imb = sgs.group_imb;
-		}
-
-		update_sd_power_savings_stats(group, sds, local_group, &sgs);
-		group = group->next;
-	} while (group != sd->groups);
-}
-
-/**
- * fix_small_imbalance - Calculate the minor imbalance that exists
- *			amongst the groups of a sched_domain, during
- *			load balancing.
- * @sds: Statistics of the sched_domain whose imbalance is to be calculated.
- * @this_cpu: The cpu at whose sched_domain we're performing load-balance.
- * @imbalance: Variable to store the imbalance.
- */
-static inline void fix_small_imbalance(struct sd_lb_stats *sds,
-				int this_cpu, unsigned long *imbalance)
-{
-	unsigned long tmp, pwr_now = 0, pwr_move = 0;
-	unsigned int imbn = 2;
-
-	if (sds->this_nr_running) {
-		sds->this_load_per_task /= sds->this_nr_running;
-		if (sds->busiest_load_per_task >
-				sds->this_load_per_task)
-			imbn = 1;
-	} else
-		sds->this_load_per_task =
-			cpu_avg_load_per_task(this_cpu);
-
-	if (sds->max_load - sds->this_load + sds->busiest_load_per_task >=
-			sds->busiest_load_per_task * imbn) {
-		*imbalance = sds->busiest_load_per_task;
-		return;
-	}
-
-	/*
-	 * OK, we don't have enough imbalance to justify moving tasks,
-	 * however we may be able to increase total CPU power used by
-	 * moving them.
-	 */
-
-	pwr_now += sds->busiest->cpu_power *
-			min(sds->busiest_load_per_task, sds->max_load);
-	pwr_now += sds->this->cpu_power *
-			min(sds->this_load_per_task, sds->this_load);
-	pwr_now /= SCHED_LOAD_SCALE;
-
-	/* Amount of load we'd subtract */
-	tmp = (sds->busiest_load_per_task * SCHED_LOAD_SCALE) /
-		sds->busiest->cpu_power;
-	if (sds->max_load > tmp)
-		pwr_move += sds->busiest->cpu_power *
-			min(sds->busiest_load_per_task, sds->max_load - tmp);
-
-	/* Amount of load we'd add */
-	if (sds->max_load * sds->busiest->cpu_power <
-		sds->busiest_load_per_task * SCHED_LOAD_SCALE)
-		tmp = (sds->max_load * sds->busiest->cpu_power) /
-			sds->this->cpu_power;
-	else
-		tmp = (sds->busiest_load_per_task * SCHED_LOAD_SCALE) /
-			sds->this->cpu_power;
-	pwr_move += sds->this->cpu_power *
-			min(sds->this_load_per_task, sds->this_load + tmp);
-	pwr_move /= SCHED_LOAD_SCALE;
-
-	/* Move if we gain throughput */
-	if (pwr_move > pwr_now)
-		*imbalance = sds->busiest_load_per_task;
-}
-
-/**
- * calculate_imbalance - Calculate the amount of imbalance present within the
- *			 groups of a given sched_domain during load balance.
- * @sds: statistics of the sched_domain whose imbalance is to be calculated.
- * @this_cpu: Cpu for which currently load balance is being performed.
- * @imbalance: The variable to store the imbalance.
- */
-static inline void calculate_imbalance(struct sd_lb_stats *sds, int this_cpu,
-		unsigned long *imbalance)
-{
-	unsigned long max_pull;
-	/*
-	 * In the presence of smp nice balancing, certain scenarios can have
-	 * max load less than avg load(as we skip the groups at or below
-	 * its cpu_power, while calculating max_load..)
-	 */
-	if (sds->max_load < sds->avg_load) {
-		*imbalance = 0;
-		return fix_small_imbalance(sds, this_cpu, imbalance);
-	}
-
-	/* Don't want to pull so many tasks that a group would go idle */
-	max_pull = min(sds->max_load - sds->avg_load,
-			sds->max_load - sds->busiest_load_per_task);
-
-	/* How much load to actually move to equalise the imbalance */
-	*imbalance = min(max_pull * sds->busiest->cpu_power,
-		(sds->avg_load - sds->this_load) * sds->this->cpu_power)
-			/ SCHED_LOAD_SCALE;
-
-	/*
-	 * if *imbalance is less than the average load per runnable task
-	 * there is no gaurantee that any tasks will be moved so we'll have
-	 * a think about bumping its value to force at least one task to be
-	 * moved
-	 */
-	if (*imbalance < sds->busiest_load_per_task)
-		return fix_small_imbalance(sds, this_cpu, imbalance);
-
-}
-/******* find_busiest_group() helpers end here *********************/
-
-/**
- * find_busiest_group - Returns the busiest group within the sched_domain
- * if there is an imbalance. If there isn't an imbalance, and
- * the user has opted for power-savings, it returns a group whose
- * CPUs can be put to idle by rebalancing those tasks elsewhere, if
- * such a group exists.
- *
- * Also calculates the amount of weighted load which should be moved
- * to restore balance.
- *
- * @sd: The sched_domain whose busiest group is to be returned.
- * @this_cpu: The cpu for which load balancing is currently being performed.
- * @imbalance: Variable which stores amount of weighted load which should
- *		be moved to restore balance/put a group to idle.
- * @idle: The idle status of this_cpu.
- * @sd_idle: The idleness of sd
- * @cpus: The set of CPUs under consideration for load-balancing.
- * @balance: Pointer to a variable indicating if this_cpu
- *	is the appropriate cpu to perform load balancing at this_level.
- *
- * Returns:	- the busiest group if imbalance exists.
- *		- If no imbalance and user has opted for power-savings balance,
- *		   return the least loaded group whose CPUs can be
- *		   put to idle by rebalancing its tasks onto our group.
- */
-static struct sched_group *
-find_busiest_group(struct sched_domain *sd, int this_cpu,
-		   unsigned long *imbalance, enum cpu_idle_type idle,
-		   int *sd_idle, const struct cpumask *cpus, int *balance)
-{
-	struct sd_lb_stats sds;
-
-	memset(&sds, 0, sizeof(sds));
-
-	/*
-	 * Compute the various statistics relavent for load balancing at
-	 * this level.
-	 */
-	update_sd_lb_stats(sd, this_cpu, idle, sd_idle, cpus,
-					balance, &sds);
-
-	/* Cases where imbalance does not exist from POV of this_cpu */
-	/* 1) this_cpu is not the appropriate cpu to perform load balancing
-	 *    at this level.
-	 * 2) There is no busy sibling group to pull from.
-	 * 3) This group is the busiest group.
-	 * 4) This group is more busy than the avg busieness at this
-	 *    sched_domain.
-	 * 5) The imbalance is within the specified limit.
-	 * 6) Any rebalance would lead to ping-pong
-	 */
-	if (balance && !(*balance))
-		goto ret;
-
-	if (!sds.busiest || sds.busiest_nr_running == 0)
-		goto out_balanced;
-
-	if (sds.this_load >= sds.max_load)
-		goto out_balanced;
-
-	sds.avg_load = (SCHED_LOAD_SCALE * sds.total_load) / sds.total_pwr;
-
-	if (sds.this_load >= sds.avg_load)
-		goto out_balanced;
-
-	if (100 * sds.max_load <= sd->imbalance_pct * sds.this_load)
-		goto out_balanced;
-
-	sds.busiest_load_per_task /= sds.busiest_nr_running;
-	if (sds.group_imb)
-		sds.busiest_load_per_task =
-			min(sds.busiest_load_per_task, sds.avg_load);
-
-	/*
-	 * We're trying to get all the cpus to the average_load, so we don't
-	 * want to push ourselves above the average load, nor do we wish to
-	 * reduce the max loaded cpu below the average load, as either of these
-	 * actions would just result in more rebalancing later, and ping-pong
-	 * tasks around. Thus we look for the minimum possible imbalance.
-	 * Negative imbalances (*we* are more loaded than anyone else) will
-	 * be counted as no imbalance for these purposes -- we can't fix that
-	 * by pulling tasks to us. Be careful of negative numbers as they'll
-	 * appear as very large values with unsigned longs.
-	 */
-	if (sds.max_load <= sds.busiest_load_per_task)
-		goto out_balanced;
-
-	/* Looks like there is an imbalance. Compute it */
-	calculate_imbalance(&sds, this_cpu, imbalance);
-	return sds.busiest;
-
-out_balanced:
-	/*
-	 * There is no obvious imbalance. But check if we can do some balancing
-	 * to save power.
-	 */
-	if (check_power_save_busiest_group(&sds, this_cpu, imbalance))
-		return sds.busiest;
-ret:
-	*imbalance = 0;
-	return NULL;
-}
-
-/*
- * find_busiest_queue - find the busiest runqueue among the cpus in group.
- */
-static struct rq *
-find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
-		   unsigned long imbalance, const struct cpumask *cpus)
-{
-	struct rq *busiest = NULL, *rq;
-	unsigned long max_load = 0;
-	int i;
-
-	for_each_cpu(i, sched_group_cpus(group)) {
-		unsigned long power = power_of(i);
-		unsigned long capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE);
-		unsigned long wl;
-
-		if (!cpumask_test_cpu(i, cpus))
-			continue;
-
-		rq = cpu_rq(i);
-		wl = weighted_cpuload(i) * SCHED_LOAD_SCALE;
-		wl /= power;
-
-		if (capacity && rq->nr_running == 1 && wl > imbalance)
-			continue;
-
-		if (wl > max_load) {
-			max_load = wl;
-			busiest = rq;
-		}
-	}
-
-	return busiest;
-}
-
-/*
- * Max backoff if we encounter pinned tasks. Pretty arbitrary value, but
- * so long as it is large enough.
- */
-#define MAX_PINNED_INTERVAL	512
-
-/* Working cpumask for load_balance and load_balance_newidle. */
-static DEFINE_PER_CPU(cpumask_var_t, load_balance_tmpmask);
-
-/*
- * Check this_cpu to ensure it is balanced within domain. Attempt to move
- * tasks if there is an imbalance.
- */
-static int load_balance(int this_cpu, struct rq *this_rq,
-			struct sched_domain *sd, enum cpu_idle_type idle,
-			int *balance)
-{
-	int ld_moved, all_pinned = 0, active_balance = 0, sd_idle = 0;
-	struct sched_group *group;
-	unsigned long imbalance;
-	struct rq *busiest;
-	unsigned long flags;
-	struct cpumask *cpus = __get_cpu_var(load_balance_tmpmask);
-
-	cpumask_copy(cpus, cpu_active_mask);
-
-	/*
-	 * When power savings policy is enabled for the parent domain, idle
-	 * sibling can pick up load irrespective of busy siblings. In this case,
-	 * let the state of idle sibling percolate up as CPU_IDLE, instead of
-	 * portraying it as CPU_NOT_IDLE.
-	 */
-	if (idle != CPU_NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER &&
-	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-		sd_idle = 1;
-
-	schedstat_inc(sd, lb_count[idle]);
-
-redo:
-	update_shares(sd);
-	group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle,
-				   cpus, balance);
-
-	if (*balance == 0)
-		goto out_balanced;
-
-	if (!group) {
-		schedstat_inc(sd, lb_nobusyg[idle]);
-		goto out_balanced;
-	}
-
-	busiest = find_busiest_queue(group, idle, imbalance, cpus);
-	if (!busiest) {
-		schedstat_inc(sd, lb_nobusyq[idle]);
-		goto out_balanced;
-	}
-
-	BUG_ON(busiest == this_rq);
-
-	schedstat_add(sd, lb_imbalance[idle], imbalance);
-
-	ld_moved = 0;
-	if (busiest->nr_running > 1) {
-		/*
-		 * Attempt to move tasks. If find_busiest_group has found
-		 * an imbalance but busiest->nr_running <= 1, the group is
-		 * still unbalanced. ld_moved simply stays zero, so it is
-		 * correctly treated as an imbalance.
-		 */
-		local_irq_save(flags);
-		double_rq_lock(this_rq, busiest);
-		ld_moved = move_tasks(this_rq, this_cpu, busiest,
-				      imbalance, sd, idle, &all_pinned);
-		double_rq_unlock(this_rq, busiest);
-		local_irq_restore(flags);
-
-		/*
-		 * some other cpu did the load balance for us.
-		 */
-		if (ld_moved && this_cpu != smp_processor_id())
-			resched_cpu(this_cpu);
-
-		/* All tasks on this runqueue were pinned by CPU affinity */
-		if (unlikely(all_pinned)) {
-			cpumask_clear_cpu(cpu_of(busiest), cpus);
-			if (!cpumask_empty(cpus))
-				goto redo;
-			goto out_balanced;
-		}
-	}
-
-	if (!ld_moved) {
-		schedstat_inc(sd, lb_failed[idle]);
-		sd->nr_balance_failed++;
-
-		if (unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2)) {
-
-			raw_spin_lock_irqsave(&busiest->lock, flags);
-
-			/* don't kick the migration_thread, if the curr
-			 * task on busiest cpu can't be moved to this_cpu
-			 */
-			if (!cpumask_test_cpu(this_cpu,
-					      &busiest->curr->cpus_allowed)) {
-				raw_spin_unlock_irqrestore(&busiest->lock,
-							    flags);
-				all_pinned = 1;
-				goto out_one_pinned;
-			}
-
-			if (!busiest->active_balance) {
-				busiest->active_balance = 1;
-				busiest->push_cpu = this_cpu;
-				active_balance = 1;
-			}
-			raw_spin_unlock_irqrestore(&busiest->lock, flags);
-			if (active_balance)
-				wake_up_process(busiest->migration_thread);
-
-			/*
-			 * We've kicked active balancing, reset the failure
-			 * counter.
-			 */
-			sd->nr_balance_failed = sd->cache_nice_tries+1;
-		}
-	} else
-		sd->nr_balance_failed = 0;
-
-	if (likely(!active_balance)) {
-		/* We were unbalanced, so reset the balancing interval */
-		sd->balance_interval = sd->min_interval;
-	} else {
-		/*
-		 * If we've begun active balancing, start to back off. This
-		 * case may not be covered by the all_pinned logic if there
-		 * is only 1 task on the busy runqueue (because we don't call
-		 * move_tasks).
-		 */
-		if (sd->balance_interval < sd->max_interval)
-			sd->balance_interval *= 2;
-	}
-
-	if (!ld_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
-	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-		ld_moved = -1;
-
-	goto out;
-
-out_balanced:
-	schedstat_inc(sd, lb_balanced[idle]);
-
-	sd->nr_balance_failed = 0;
-
-out_one_pinned:
-	/* tune up the balancing interval */
-	if ((all_pinned && sd->balance_interval < MAX_PINNED_INTERVAL) ||
-			(sd->balance_interval < sd->max_interval))
-		sd->balance_interval *= 2;
-
-	if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
-	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-		ld_moved = -1;
-	else
-		ld_moved = 0;
-out:
-	if (ld_moved)
-		update_shares(sd);
-	return ld_moved;
-}
-
-/*
- * Check this_cpu to ensure it is balanced within domain. Attempt to move
- * tasks if there is an imbalance.
- *
- * Called from schedule when this_rq is about to become idle (CPU_NEWLY_IDLE).
- * this_rq is locked.
- */
-static int
-load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd)
-{
-	struct sched_group *group;
-	struct rq *busiest = NULL;
-	unsigned long imbalance;
-	int ld_moved = 0;
-	int sd_idle = 0;
-	int all_pinned = 0;
-	struct cpumask *cpus = __get_cpu_var(load_balance_tmpmask);
-
-	cpumask_copy(cpus, cpu_active_mask);
-
-	/*
-	 * When power savings policy is enabled for the parent domain, idle
-	 * sibling can pick up load irrespective of busy siblings. In this case,
-	 * let the state of idle sibling percolate up as IDLE, instead of
-	 * portraying it as CPU_NOT_IDLE.
-	 */
-	if (sd->flags & SD_SHARE_CPUPOWER &&
-	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-		sd_idle = 1;
-
-	schedstat_inc(sd, lb_count[CPU_NEWLY_IDLE]);
-redo:
-	update_shares_locked(this_rq, sd);
-	group = find_busiest_group(sd, this_cpu, &imbalance, CPU_NEWLY_IDLE,
-				   &sd_idle, cpus, NULL);
-	if (!group) {
-		schedstat_inc(sd, lb_nobusyg[CPU_NEWLY_IDLE]);
-		goto out_balanced;
-	}
-
-	busiest = find_busiest_queue(group, CPU_NEWLY_IDLE, imbalance, cpus);
-	if (!busiest) {
-		schedstat_inc(sd, lb_nobusyq[CPU_NEWLY_IDLE]);
-		goto out_balanced;
-	}
-
-	BUG_ON(busiest == this_rq);
-
-	schedstat_add(sd, lb_imbalance[CPU_NEWLY_IDLE], imbalance);
-
-	ld_moved = 0;
-	if (busiest->nr_running > 1) {
-		/* Attempt to move tasks */
-		double_lock_balance(this_rq, busiest);
-		/* this_rq->clock is already updated */
-		update_rq_clock(busiest);
-		ld_moved = move_tasks(this_rq, this_cpu, busiest,
-					imbalance, sd, CPU_NEWLY_IDLE,
-					&all_pinned);
-		double_unlock_balance(this_rq, busiest);
-
-		if (unlikely(all_pinned)) {
-			cpumask_clear_cpu(cpu_of(busiest), cpus);
-			if (!cpumask_empty(cpus))
-				goto redo;
-		}
-	}
-
-	if (!ld_moved) {
-		int active_balance = 0;
-
-		schedstat_inc(sd, lb_failed[CPU_NEWLY_IDLE]);
-		if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
-		    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-			return -1;
-
-		if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
-			return -1;
-
-		if (sd->nr_balance_failed++ < 2)
-			return -1;
-
-		/*
-		 * The only task running in a non-idle cpu can be moved to this
-		 * cpu in an attempt to completely freeup the other CPU
-		 * package. The same method used to move task in load_balance()
-		 * have been extended for load_balance_newidle() to speedup
-		 * consolidation at sched_mc=POWERSAVINGS_BALANCE_WAKEUP (2)
-		 *
-		 * The package power saving logic comes from
-		 * find_busiest_group().  If there are no imbalance, then
-		 * f_b_g() will return NULL.  However when sched_mc={1,2} then
-		 * f_b_g() will select a group from which a running task may be
-		 * pulled to this cpu in order to make the other package idle.
-		 * If there is no opportunity to make a package idle and if
-		 * there are no imbalance, then f_b_g() will return NULL and no
-		 * action will be taken in load_balance_newidle().
-		 *
-		 * Under normal task pull operation due to imbalance, there
-		 * will be more than one task in the source run queue and
-		 * move_tasks() will succeed.  ld_moved will be true and this
-		 * active balance code will not be triggered.
-		 */
-
-		/* Lock busiest in correct order while this_rq is held */
-		double_lock_balance(this_rq, busiest);
-
-		/*
-		 * don't kick the migration_thread, if the curr
-		 * task on busiest cpu can't be moved to this_cpu
-		 */
-		if (!cpumask_test_cpu(this_cpu, &busiest->curr->cpus_allowed)) {
-			double_unlock_balance(this_rq, busiest);
-			all_pinned = 1;
-			return ld_moved;
-		}
-
-		if (!busiest->active_balance) {
-			busiest->active_balance = 1;
-			busiest->push_cpu = this_cpu;
-			active_balance = 1;
-		}
-
-		double_unlock_balance(this_rq, busiest);
-		/*
-		 * Should not call ttwu while holding a rq->lock
-		 */
-		raw_spin_unlock(&this_rq->lock);
-		if (active_balance)
-			wake_up_process(busiest->migration_thread);
-		raw_spin_lock(&this_rq->lock);
-
-	} else
-		sd->nr_balance_failed = 0;
-
-	update_shares_locked(this_rq, sd);
-	return ld_moved;
-
-out_balanced:
-	schedstat_inc(sd, lb_balanced[CPU_NEWLY_IDLE]);
-	if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
-	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-		return -1;
-	sd->nr_balance_failed = 0;
-
-	return 0;
-}
-
-/*
- * idle_balance is called by schedule() if this_cpu is about to become
- * idle. Attempts to pull tasks from other CPUs.
- */
-static void idle_balance(int this_cpu, struct rq *this_rq)
-{
-	struct sched_domain *sd;
-	int pulled_task = 0;
-	unsigned long next_balance = jiffies + HZ;
-
-	this_rq->idle_stamp = this_rq->clock;
-
-	if (this_rq->avg_idle < sysctl_sched_migration_cost)
-		return;
-
-	for_each_domain(this_cpu, sd) {
-		unsigned long interval;
-
-		if (!(sd->flags & SD_LOAD_BALANCE))
-			continue;
-
-		if (sd->flags & SD_BALANCE_NEWIDLE)
-			/* If we've pulled tasks over stop searching: */
-			pulled_task = load_balance_newidle(this_cpu, this_rq,
-							   sd);
-
-		interval = msecs_to_jiffies(sd->balance_interval);
-		if (time_after(next_balance, sd->last_balance + interval))
-			next_balance = sd->last_balance + interval;
-		if (pulled_task) {
-			this_rq->idle_stamp = 0;
-			break;
-		}
-	}
-	if (pulled_task || time_after(jiffies, this_rq->next_balance)) {
-		/*
-		 * We are going idle. next_balance may be set based on
-		 * a busy processor. So reset next_balance.
-		 */
-		this_rq->next_balance = next_balance;
-	}
-}
-
-/*
- * active_load_balance is run by migration threads. It pushes running tasks
- * off the busiest CPU onto idle CPUs. It requires at least 1 task to be
- * running on each physical CPU where possible, and avoids physical /
- * logical imbalances.
- *
- * Called with busiest_rq locked.
- */
-static void active_load_balance(struct rq *busiest_rq, int busiest_cpu)
-{
-	int target_cpu = busiest_rq->push_cpu;
-	struct sched_domain *sd;
-	struct rq *target_rq;
-
-	/* Is there any task to move? */
-	if (busiest_rq->nr_running <= 1)
-		return;
-
-	target_rq = cpu_rq(target_cpu);
-
-	/*
-	 * This condition is "impossible", if it occurs
-	 * we need to fix it. Originally reported by
-	 * Bjorn Helgaas on a 128-cpu setup.
-	 */
-	BUG_ON(busiest_rq == target_rq);
-
-	/* move a task from busiest_rq to target_rq */
-	double_lock_balance(busiest_rq, target_rq);
-	update_rq_clock(busiest_rq);
-	update_rq_clock(target_rq);
-
-	/* Search for an sd spanning us and the target CPU. */
-	for_each_domain(target_cpu, sd) {
-		if ((sd->flags & SD_LOAD_BALANCE) &&
-		    cpumask_test_cpu(busiest_cpu, sched_domain_span(sd)))
-				break;
-	}
-
-	if (likely(sd)) {
-		schedstat_inc(sd, alb_count);
-
-		if (move_one_task(target_rq, target_cpu, busiest_rq,
-				  sd, CPU_IDLE))
-			schedstat_inc(sd, alb_pushed);
-		else
-			schedstat_inc(sd, alb_failed);
-	}
-	double_unlock_balance(busiest_rq, target_rq);
-}
-
-#ifdef CONFIG_NO_HZ
-static struct {
-	atomic_t load_balancer;
-	cpumask_var_t cpu_mask;
-	cpumask_var_t ilb_grp_nohz_mask;
-} nohz ____cacheline_aligned = {
-	.load_balancer = ATOMIC_INIT(-1),
-};
-
-int get_nohz_load_balancer(void)
-{
-	return atomic_read(&nohz.load_balancer);
-}
-
-#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
-/**
- * lowest_flag_domain - Return lowest sched_domain containing flag.
- * @cpu:	The cpu whose lowest level of sched domain is to
- *		be returned.
- * @flag:	The flag to check for the lowest sched_domain
- *		for the given cpu.
- *
- * Returns the lowest sched_domain of a cpu which contains the given flag.
- */
-static inline struct sched_domain *lowest_flag_domain(int cpu, int flag)
-{
-	struct sched_domain *sd;
-
-	for_each_domain(cpu, sd)
-		if (sd && (sd->flags & flag))
-			break;
-
-	return sd;
-}
-
-/**
- * for_each_flag_domain - Iterates over sched_domains containing the flag.
- * @cpu:	The cpu whose domains we're iterating over.
- * @sd:		variable holding the value of the power_savings_sd
- *		for cpu.
- * @flag:	The flag to filter the sched_domains to be iterated.
- *
- * Iterates over all the scheduler domains for a given cpu that has the 'flag'
- * set, starting from the lowest sched_domain to the highest.
- */
-#define for_each_flag_domain(cpu, sd, flag) \
-	for (sd = lowest_flag_domain(cpu, flag); \
-		(sd && (sd->flags & flag)); sd = sd->parent)
-
-/**
- * is_semi_idle_group - Checks if the given sched_group is semi-idle.
- * @ilb_group:	group to be checked for semi-idleness
- *
- * Returns:	1 if the group is semi-idle. 0 otherwise.
- *
- * We define a sched_group to be semi idle if it has atleast one idle-CPU
- * and atleast one non-idle CPU. This helper function checks if the given
- * sched_group is semi-idle or not.
- */
-static inline int is_semi_idle_group(struct sched_group *ilb_group)
-{
-	cpumask_and(nohz.ilb_grp_nohz_mask, nohz.cpu_mask,
-					sched_group_cpus(ilb_group));
-
-	/*
-	 * A sched_group is semi-idle when it has atleast one busy cpu
-	 * and atleast one idle cpu.
-	 */
-	if (cpumask_empty(nohz.ilb_grp_nohz_mask))
-		return 0;
-
-	if (cpumask_equal(nohz.ilb_grp_nohz_mask, sched_group_cpus(ilb_group)))
-		return 0;
-
-	return 1;
-}
-/**
- * find_new_ilb - Finds the optimum idle load balancer for nomination.
- * @cpu:	The cpu which is nominating a new idle_load_balancer.
- *
- * Returns:	Returns the id of the idle load balancer if it exists,
- *		Else, returns >= nr_cpu_ids.
- *
- * This algorithm picks the idle load balancer such that it belongs to a
- * semi-idle powersavings sched_domain. The idea is to try and avoid
- * completely idle packages/cores just for the purpose of idle load balancing
- * when there are other idle cpu's which are better suited for that job.
- */
-static int find_new_ilb(int cpu)
-{
-	struct sched_domain *sd;
-	struct sched_group *ilb_group;
-
-	/*
-	 * Have idle load balancer selection from semi-idle packages only
-	 * when power-aware load balancing is enabled
-	 */
-	if (!(sched_smt_power_savings || sched_mc_power_savings))
-		goto out_done;
-
-	/*
-	 * Optimize for the case when we have no idle CPUs or only one
-	 * idle CPU. Don't walk the sched_domain hierarchy in such cases
-	 */
-	if (cpumask_weight(nohz.cpu_mask) < 2)
-		goto out_done;
-
-	for_each_flag_domain(cpu, sd, SD_POWERSAVINGS_BALANCE) {
-		ilb_group = sd->groups;
-
-		do {
-			if (is_semi_idle_group(ilb_group))
-				return cpumask_first(nohz.ilb_grp_nohz_mask);
-
-			ilb_group = ilb_group->next;
-
-		} while (ilb_group != sd->groups);
-	}
-
-out_done:
-	return cpumask_first(nohz.cpu_mask);
-}
-#else /*  (CONFIG_SCHED_MC || CONFIG_SCHED_SMT) */
-static inline int find_new_ilb(int call_cpu)
-{
-	return cpumask_first(nohz.cpu_mask);
-}
-#endif
-
-/*
- * This routine will try to nominate the ilb (idle load balancing)
- * owner among the cpus whose ticks are stopped. ilb owner will do the idle
- * load balancing on behalf of all those cpus. If all the cpus in the system
- * go into this tickless mode, then there will be no ilb owner (as there is
- * no need for one) and all the cpus will sleep till the next wakeup event
- * arrives...
- *
- * For the ilb owner, tick is not stopped. And this tick will be used
- * for idle load balancing. ilb owner will still be part of
- * nohz.cpu_mask..
- *
- * While stopping the tick, this cpu will become the ilb owner if there
- * is no other owner. And will be the owner till that cpu becomes busy
- * or if all cpus in the system stop their ticks at which point
- * there is no need for ilb owner.
- *
- * When the ilb owner becomes busy, it nominates another owner, during the
- * next busy scheduler_tick()
- */
-int select_nohz_load_balancer(int stop_tick)
-{
-	int cpu = smp_processor_id();
-
-	if (stop_tick) {
-		cpu_rq(cpu)->in_nohz_recently = 1;
-
-		if (!cpu_active(cpu)) {
-			if (atomic_read(&nohz.load_balancer) != cpu)
-				return 0;
-
-			/*
-			 * If we are going offline and still the leader,
-			 * give up!
-			 */
-			if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
-				BUG();
-
-			return 0;
-		}
-
-		cpumask_set_cpu(cpu, nohz.cpu_mask);
-
-		/* time for ilb owner also to sleep */
-		if (cpumask_weight(nohz.cpu_mask) == num_active_cpus()) {
-			if (atomic_read(&nohz.load_balancer) == cpu)
-				atomic_set(&nohz.load_balancer, -1);
-			return 0;
-		}
-
-		if (atomic_read(&nohz.load_balancer) == -1) {
-			/* make me the ilb owner */
-			if (atomic_cmpxchg(&nohz.load_balancer, -1, cpu) == -1)
-				return 1;
-		} else if (atomic_read(&nohz.load_balancer) == cpu) {
-			int new_ilb;
-
-			if (!(sched_smt_power_savings ||
-						sched_mc_power_savings))
-				return 1;
-			/*
-			 * Check to see if there is a more power-efficient
-			 * ilb.
-			 */
-			new_ilb = find_new_ilb(cpu);
-			if (new_ilb < nr_cpu_ids && new_ilb != cpu) {
-				atomic_set(&nohz.load_balancer, -1);
-				resched_cpu(new_ilb);
-				return 0;
-			}
-			return 1;
-		}
-	} else {
-		if (!cpumask_test_cpu(cpu, nohz.cpu_mask))
-			return 0;
-
-		cpumask_clear_cpu(cpu, nohz.cpu_mask);
-
-		if (atomic_read(&nohz.load_balancer) == cpu)
-			if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
-				BUG();
-	}
-	return 0;
-}
-#endif
-
-static DEFINE_SPINLOCK(balancing);
-
-/*
- * It checks each scheduling domain to see if it is due to be balanced,
- * and initiates a balancing operation if so.
- *
- * Balancing parameters are set up in arch_init_sched_domains.
- */
-static void rebalance_domains(int cpu, enum cpu_idle_type idle)
-{
-	int balance = 1;
-	struct rq *rq = cpu_rq(cpu);
-	unsigned long interval;
-	struct sched_domain *sd;
-	/* Earliest time when we have to do rebalance again */
-	unsigned long next_balance = jiffies + 60*HZ;
-	int update_next_balance = 0;
-	int need_serialize;
-
-	for_each_domain(cpu, sd) {
-		if (!(sd->flags & SD_LOAD_BALANCE))
-			continue;
-
-		interval = sd->balance_interval;
-		if (idle != CPU_IDLE)
-			interval *= sd->busy_factor;
-
-		/* scale ms to jiffies */
-		interval = msecs_to_jiffies(interval);
-		if (unlikely(!interval))
-			interval = 1;
-		if (interval > HZ*NR_CPUS/10)
-			interval = HZ*NR_CPUS/10;
-
-		need_serialize = sd->flags & SD_SERIALIZE;
-
-		if (need_serialize) {
-			if (!spin_trylock(&balancing))
-				goto out;
-		}
-
-		if (time_after_eq(jiffies, sd->last_balance + interval)) {
-			if (load_balance(cpu, rq, sd, idle, &balance)) {
-				/*
-				 * We've pulled tasks over so either we're no
-				 * longer idle, or one of our SMT siblings is
-				 * not idle.
-				 */
-				idle = CPU_NOT_IDLE;
-			}
-			sd->last_balance = jiffies;
-		}
-		if (need_serialize)
-			spin_unlock(&balancing);
-out:
-		if (time_after(next_balance, sd->last_balance + interval)) {
-			next_balance = sd->last_balance + interval;
-			update_next_balance = 1;
-		}
-
-		/*
-		 * Stop the load balance at this level. There is another
-		 * CPU in our sched group which is doing load balancing more
-		 * actively.
-		 */
-		if (!balance)
-			break;
-	}
-
-	/*
-	 * next_balance will be updated only when there is a need.
-	 * When the cpu is attached to null domain for ex, it will not be
-	 * updated.
-	 */
-	if (likely(update_next_balance))
-		rq->next_balance = next_balance;
-}
-
-/*
- * run_rebalance_domains is triggered when needed from the scheduler tick.
- * In CONFIG_NO_HZ case, the idle load balance owner will do the
- * rebalancing for all the cpus for whom scheduler ticks are stopped.
- */
-static void run_rebalance_domains(struct softirq_action *h)
-{
-	int this_cpu = smp_processor_id();
-	struct rq *this_rq = cpu_rq(this_cpu);
-	enum cpu_idle_type idle = this_rq->idle_at_tick ?
-						CPU_IDLE : CPU_NOT_IDLE;
-
-	rebalance_domains(this_cpu, idle);
-
-#ifdef CONFIG_NO_HZ
-	/*
-	 * If this cpu is the owner for idle load balancing, then do the
-	 * balancing on behalf of the other idle cpus whose ticks are
-	 * stopped.
-	 */
-	if (this_rq->idle_at_tick &&
-	    atomic_read(&nohz.load_balancer) == this_cpu) {
-		struct rq *rq;
-		int balance_cpu;
-
-		for_each_cpu(balance_cpu, nohz.cpu_mask) {
-			if (balance_cpu == this_cpu)
-				continue;
-
-			/*
-			 * If this cpu gets work to do, stop the load balancing
-			 * work being done for other cpus. Next load
-			 * balancing owner will pick it up.
-			 */
-			if (need_resched())
-				break;
-
-			rebalance_domains(balance_cpu, CPU_IDLE);
-
-			rq = cpu_rq(balance_cpu);
-			if (time_after(this_rq->next_balance, rq->next_balance))
-				this_rq->next_balance = rq->next_balance;
-		}
-	}
-#endif
-}
-
-static inline int on_null_domain(int cpu)
-{
-	return !rcu_dereference(cpu_rq(cpu)->sd);
-}
-
-/*
- * Trigger the SCHED_SOFTIRQ if it is time to do periodic load balancing.
- *
- * In case of CONFIG_NO_HZ, this is the place where we nominate a new
- * idle load balancing owner or decide to stop the periodic load balancing,
- * if the whole system is idle.
- */
-static inline void trigger_load_balance(struct rq *rq, int cpu)
-{
-#ifdef CONFIG_NO_HZ
-	/*
-	 * If we were in the nohz mode recently and busy at the current
-	 * scheduler tick, then check if we need to nominate new idle
-	 * load balancer.
-	 */
-	if (rq->in_nohz_recently && !rq->idle_at_tick) {
-		rq->in_nohz_recently = 0;
-
-		if (atomic_read(&nohz.load_balancer) == cpu) {
-			cpumask_clear_cpu(cpu, nohz.cpu_mask);
-			atomic_set(&nohz.load_balancer, -1);
-		}
-
-		if (atomic_read(&nohz.load_balancer) == -1) {
-			int ilb = find_new_ilb(cpu);
-
-			if (ilb < nr_cpu_ids)
-				resched_cpu(ilb);
-		}
-	}
-
-	/*
-	 * If this cpu is idle and doing idle load balancing for all the
-	 * cpus with ticks stopped, is it time for that to stop?
-	 */
-	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) == cpu &&
-	    cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
-		resched_cpu(cpu);
-		return;
-	}
-
-	/*
-	 * If this cpu is idle and the idle load balancing is done by
-	 * someone else, then no need raise the SCHED_SOFTIRQ
-	 */
-	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) != cpu &&
-	    cpumask_test_cpu(cpu, nohz.cpu_mask))
-		return;
-#endif
-	/* Don't need to rebalance while attached to NULL domain */
-	if (time_after_eq(jiffies, rq->next_balance) &&
-	    likely(!on_null_domain(cpu)))
-		raise_softirq(SCHED_SOFTIRQ);
-}
-
-#else	/* CONFIG_SMP */
-
-/*
- * on UP we do not need to balance between CPUs:
- */
-static inline void idle_balance(int cpu, struct rq *rq)
-{
-}
-
 #endif
 
 DEFINE_PER_CPU(struct kernel_stat, kstat);
@@ -5309,7 +3473,7 @@ void scheduler_tick(void)
 	curr->sched_class->task_tick(rq, curr, 0);
 	raw_spin_unlock(&rq->lock);
 
-	perf_event_task_tick(curr, cpu);
+	perf_event_task_tick(curr);
 
 #ifdef CONFIG_SMP
 	rq->idle_at_tick = idle_cpu(cpu);
@@ -5523,7 +3687,7 @@ need_resched_nonpreemptible:
 
 	if (likely(prev != next)) {
 		sched_info_switch(prev, next);
-		perf_event_task_sched_out(prev, next, cpu);
+		perf_event_task_sched_out(prev, next);
 
 		rq->nr_switches++;
 		rq->curr = next;
@@ -6079,7 +4243,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 	if (running)
 		p->sched_class->set_curr_task(rq);
 	if (on_rq) {
-		enqueue_task(rq, p, 0);
+		enqueue_task(rq, p, 0, oldprio < prio);
 
 		check_class_changed(rq, p, prev_class, oldprio, running);
 	}
@@ -6123,7 +4287,7 @@ void set_user_nice(struct task_struct *p, long nice)
 	delta = p->prio - old_prio;
 
 	if (on_rq) {
-		enqueue_task(rq, p, 0);
+		enqueue_task(rq, p, 0, false);
 		/*
 		 * If the task increased its priority or is running and
 		 * lowered its priority, then reschedule its CPU:
@@ -9483,9 +7647,6 @@ void __init sched_init(void)
 #ifdef CONFIG_RT_GROUP_SCHED
 	alloc_size += 2 * nr_cpu_ids * sizeof(void **);
 #endif
-#ifdef CONFIG_USER_SCHED
-	alloc_size *= 2;
-#endif
 #ifdef CONFIG_CPUMASK_OFFSTACK
 	alloc_size += num_possible_cpus() * cpumask_size();
 #endif
@@ -9499,13 +7660,6 @@ void __init sched_init(void)
 		init_task_group.cfs_rq = (struct cfs_rq **)ptr;
 		ptr += nr_cpu_ids * sizeof(void **);
 
-#ifdef CONFIG_USER_SCHED
-		root_task_group.se = (struct sched_entity **)ptr;
-		ptr += nr_cpu_ids * sizeof(void **);
-
-		root_task_group.cfs_rq = (struct cfs_rq **)ptr;
-		ptr += nr_cpu_ids * sizeof(void **);
-#endif /* CONFIG_USER_SCHED */
 #endif /* CONFIG_FAIR_GROUP_SCHED */
 #ifdef CONFIG_RT_GROUP_SCHED
 		init_task_group.rt_se = (struct sched_rt_entity **)ptr;
@@ -9514,13 +7668,6 @@ void __init sched_init(void)
 		init_task_group.rt_rq = (struct rt_rq **)ptr;
 		ptr += nr_cpu_ids * sizeof(void **);
 
-#ifdef CONFIG_USER_SCHED
-		root_task_group.rt_se = (struct sched_rt_entity **)ptr;
-		ptr += nr_cpu_ids * sizeof(void **);
-
-		root_task_group.rt_rq = (struct rt_rq **)ptr;
-		ptr += nr_cpu_ids * sizeof(void **);
-#endif /* CONFIG_USER_SCHED */
 #endif /* CONFIG_RT_GROUP_SCHED */
 #ifdef CONFIG_CPUMASK_OFFSTACK
 		for_each_possible_cpu(i) {
@@ -9540,22 +7687,13 @@ void __init sched_init(void)
 #ifdef CONFIG_RT_GROUP_SCHED
 	init_rt_bandwidth(&init_task_group.rt_bandwidth,
 			global_rt_period(), global_rt_runtime());
-#ifdef CONFIG_USER_SCHED
-	init_rt_bandwidth(&root_task_group.rt_bandwidth,
-			global_rt_period(), RUNTIME_INF);
-#endif /* CONFIG_USER_SCHED */
 #endif /* CONFIG_RT_GROUP_SCHED */
 
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
 	list_add(&init_task_group.list, &task_groups);
 	INIT_LIST_HEAD(&init_task_group.children);
 
-#ifdef CONFIG_USER_SCHED
-	INIT_LIST_HEAD(&root_task_group.children);
-	init_task_group.parent = &root_task_group;
-	list_add(&init_task_group.siblings, &root_task_group.children);
-#endif /* CONFIG_USER_SCHED */
-#endif /* CONFIG_GROUP_SCHED */
+#endif /* CONFIG_CGROUP_SCHED */
 
 #if defined CONFIG_FAIR_GROUP_SCHED && defined CONFIG_SMP
 	update_shares_data = __alloc_percpu(nr_cpu_ids * sizeof(unsigned long),
@@ -9595,25 +7733,6 @@ void __init sched_init(void)
 		 * directly in rq->cfs (i.e init_task_group->se[] = NULL).
 		 */
 		init_tg_cfs_entry(&init_task_group, &rq->cfs, NULL, i, 1, NULL);
-#elif defined CONFIG_USER_SCHED
-		root_task_group.shares = NICE_0_LOAD;
-		init_tg_cfs_entry(&root_task_group, &rq->cfs, NULL, i, 0, NULL);
-		/*
-		 * In case of task-groups formed thr' the user id of tasks,
-		 * init_task_group represents tasks belonging to root user.
-		 * Hence it forms a sibling of all subsequent groups formed.
-		 * In this case, init_task_group gets only a fraction of overall
-		 * system cpu resource, based on the weight assigned to root
-		 * user's cpu share (INIT_TASK_GROUP_LOAD). This is accomplished
-		 * by letting tasks of init_task_group sit in a separate cfs_rq
-		 * (init_tg_cfs_rq) and having one entity represent this group of
-		 * tasks in rq->cfs (i.e init_task_group->se[] != NULL).
-		 */
-		init_tg_cfs_entry(&init_task_group,
-				&per_cpu(init_tg_cfs_rq, i),
-				&per_cpu(init_sched_entity, i), i, 1,
-				root_task_group.se[i]);
-
 #endif
 #endif /* CONFIG_FAIR_GROUP_SCHED */
 
@@ -9622,12 +7741,6 @@ void __init sched_init(void)
 		INIT_LIST_HEAD(&rq->leaf_rt_rq_list);
 #ifdef CONFIG_CGROUP_SCHED
 		init_tg_rt_entry(&init_task_group, &rq->rt, NULL, i, 1, NULL);
-#elif defined CONFIG_USER_SCHED
-		init_tg_rt_entry(&root_task_group, &rq->rt, NULL, i, 0, NULL);
-		init_tg_rt_entry(&init_task_group,
-				&per_cpu(init_rt_rq_var, i),
-				&per_cpu(init_sched_rt_entity, i), i, 1,
-				root_task_group.rt_se[i]);
 #endif
 #endif
 
@@ -9712,7 +7825,7 @@ static inline int preempt_count_equals(int preempt_offset)
 	return (nested == PREEMPT_INATOMIC_BASE + preempt_offset);
 }
 
-void __might_sleep(char *file, int line, int preempt_offset)
+void __might_sleep(const char *file, int line, int preempt_offset)
 {
 #ifdef in_atomic
 	static unsigned long prev_jiffy;	/* ratelimiting */
@@ -10026,7 +8139,7 @@ static inline void unregister_rt_sched_group(struct task_group *tg, int cpu)
 }
 #endif /* CONFIG_RT_GROUP_SCHED */
 
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
 static void free_sched_group(struct task_group *tg)
 {
 	free_fair_sched_group(tg);
@@ -10131,11 +8244,11 @@ void sched_move_task(struct task_struct *tsk)
 	if (unlikely(running))
 		tsk->sched_class->set_curr_task(rq);
 	if (on_rq)
-		enqueue_task(rq, tsk, 0);
+		enqueue_task(rq, tsk, 0, false);
 
 	task_rq_unlock(rq, &flags);
 }
-#endif /* CONFIG_GROUP_SCHED */
+#endif /* CONFIG_CGROUP_SCHED */
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 static void __set_se_shares(struct sched_entity *se, unsigned long shares)
@@ -10277,13 +8390,6 @@ static int tg_schedulable(struct task_group *tg, void *data)
 		runtime = d->rt_runtime;
 	}
 
-#ifdef CONFIG_USER_SCHED
-	if (tg == &root_task_group) {
-		period = global_rt_period();
-		runtime = global_rt_runtime();
-	}
-#endif
-
 	/*
 	 * Cannot have more runtime than the period.
 	 */
diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c
index 3db4b1a0e921..370012af3044 100644
--- a/kernel/sched_cpupri.c
+++ b/kernel/sched_cpupri.c
@@ -47,9 +47,7 @@ static int convert_prio(int prio)
 }
 
 #define for_each_cpupri_active(array, idx)                    \
-  for (idx = find_first_bit(array, CPUPRI_NR_PRIORITIES);     \
-       idx < CPUPRI_NR_PRIORITIES;                            \
-       idx = find_next_bit(array, CPUPRI_NR_PRIORITIES, idx+1))
+	for_each_bit(idx, array, CPUPRI_NR_PRIORITIES)
 
 /**
  * cpupri_find - find the best (lowest-pri) CPU in the system
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 8fe7ee81c552..b45abbe55067 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1053,7 +1053,8 @@ static inline void hrtick_update(struct rq *rq)
  * increased. Here we update the fair scheduling stats and
  * then put the task into the rbtree:
  */
-static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup)
+static void
+enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup, bool head)
 {
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se = &p->se;
@@ -1815,57 +1816,164 @@ static void put_prev_task_fair(struct rq *rq, struct task_struct *prev)
  */
 
 /*
- * Load-balancing iterator. Note: while the runqueue stays locked
- * during the whole iteration, the current task might be
- * dequeued so the iterator has to be dequeue-safe. Here we
- * achieve that by always pre-iterating before returning
- * the current task:
+ * pull_task - move a task from a remote runqueue to the local runqueue.
+ * Both runqueues must be locked.
  */
-static struct task_struct *
-__load_balance_iterator(struct cfs_rq *cfs_rq, struct list_head *next)
+static void pull_task(struct rq *src_rq, struct task_struct *p,
+		      struct rq *this_rq, int this_cpu)
 {
-	struct task_struct *p = NULL;
-	struct sched_entity *se;
+	deactivate_task(src_rq, p, 0);
+	set_task_cpu(p, this_cpu);
+	activate_task(this_rq, p, 0);
+	check_preempt_curr(this_rq, p, 0);
+}
 
-	if (next == &cfs_rq->tasks)
-		return NULL;
+/*
+ * can_migrate_task - may task p from runqueue rq be migrated to this_cpu?
+ */
+static
+int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu,
+		     struct sched_domain *sd, enum cpu_idle_type idle,
+		     int *all_pinned)
+{
+	int tsk_cache_hot = 0;
+	/*
+	 * We do not migrate tasks that are:
+	 * 1) running (obviously), or
+	 * 2) cannot be migrated to this CPU due to cpus_allowed, or
+	 * 3) are cache-hot on their current CPU.
+	 */
+	if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) {
+		schedstat_inc(p, se.nr_failed_migrations_affine);
+		return 0;
+	}
+	*all_pinned = 0;
 
-	se = list_entry(next, struct sched_entity, group_node);
-	p = task_of(se);
-	cfs_rq->balance_iterator = next->next;
+	if (task_running(rq, p)) {
+		schedstat_inc(p, se.nr_failed_migrations_running);
+		return 0;
+	}
 
-	return p;
-}
+	/*
+	 * Aggressive migration if:
+	 * 1) task is cache cold, or
+	 * 2) too many balance attempts have failed.
+	 */
 
-static struct task_struct *load_balance_start_fair(void *arg)
-{
-	struct cfs_rq *cfs_rq = arg;
+	tsk_cache_hot = task_hot(p, rq->clock, sd);
+	if (!tsk_cache_hot ||
+		sd->nr_balance_failed > sd->cache_nice_tries) {
+#ifdef CONFIG_SCHEDSTATS
+		if (tsk_cache_hot) {
+			schedstat_inc(sd, lb_hot_gained[idle]);
+			schedstat_inc(p, se.nr_forced_migrations);
+		}
+#endif
+		return 1;
+	}
 
-	return __load_balance_iterator(cfs_rq, cfs_rq->tasks.next);
+	if (tsk_cache_hot) {
+		schedstat_inc(p, se.nr_failed_migrations_hot);
+		return 0;
+	}
+	return 1;
 }
 
-static struct task_struct *load_balance_next_fair(void *arg)
+/*
+ * move_one_task tries to move exactly one task from busiest to this_rq, as
+ * part of active balancing operations within "domain".
+ * Returns 1 if successful and 0 otherwise.
+ *
+ * Called with both runqueues locked.
+ */
+static int
+move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest,
+	      struct sched_domain *sd, enum cpu_idle_type idle)
 {
-	struct cfs_rq *cfs_rq = arg;
+	struct task_struct *p, *n;
+	struct cfs_rq *cfs_rq;
+	int pinned = 0;
+
+	for_each_leaf_cfs_rq(busiest, cfs_rq) {
+		list_for_each_entry_safe(p, n, &cfs_rq->tasks, se.group_node) {
+
+			if (!can_migrate_task(p, busiest, this_cpu,
+						sd, idle, &pinned))
+				continue;
+
+			pull_task(busiest, p, this_rq, this_cpu);
+			/*
+			 * Right now, this is only the second place pull_task()
+			 * is called, so we can safely collect pull_task()
+			 * stats here rather than inside pull_task().
+			 */
+			schedstat_inc(sd, lb_gained[idle]);
+			return 1;
+		}
+	}
 
-	return __load_balance_iterator(cfs_rq, cfs_rq->balance_iterator);
+	return 0;
 }
 
 static unsigned long
-__load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		unsigned long max_load_move, struct sched_domain *sd,
-		enum cpu_idle_type idle, int *all_pinned, int *this_best_prio,
-		struct cfs_rq *cfs_rq)
+balance_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest,
+	      unsigned long max_load_move, struct sched_domain *sd,
+	      enum cpu_idle_type idle, int *all_pinned,
+	      int *this_best_prio, struct cfs_rq *busiest_cfs_rq)
 {
-	struct rq_iterator cfs_rq_iterator;
+	int loops = 0, pulled = 0, pinned = 0;
+	long rem_load_move = max_load_move;
+	struct task_struct *p, *n;
 
-	cfs_rq_iterator.start = load_balance_start_fair;
-	cfs_rq_iterator.next = load_balance_next_fair;
-	cfs_rq_iterator.arg = cfs_rq;
+	if (max_load_move == 0)
+		goto out;
 
-	return balance_tasks(this_rq, this_cpu, busiest,
-			max_load_move, sd, idle, all_pinned,
-			this_best_prio, &cfs_rq_iterator);
+	pinned = 1;
+
+	list_for_each_entry_safe(p, n, &busiest_cfs_rq->tasks, se.group_node) {
+		if (loops++ > sysctl_sched_nr_migrate)
+			break;
+
+		if ((p->se.load.weight >> 1) > rem_load_move ||
+		    !can_migrate_task(p, busiest, this_cpu, sd, idle, &pinned))
+			continue;
+
+		pull_task(busiest, p, this_rq, this_cpu);
+		pulled++;
+		rem_load_move -= p->se.load.weight;
+
+#ifdef CONFIG_PREEMPT
+		/*
+		 * NEWIDLE balancing is a source of latency, so preemptible
+		 * kernels will stop after the first task is pulled to minimize
+		 * the critical section.
+		 */
+		if (idle == CPU_NEWLY_IDLE)
+			break;
+#endif
+
+		/*
+		 * We only want to steal up to the prescribed amount of
+		 * weighted load.
+		 */
+		if (rem_load_move <= 0)
+			break;
+
+		if (p->prio < *this_best_prio)
+			*this_best_prio = p->prio;
+	}
+out:
+	/*
+	 * Right now, this is one of only two places pull_task() is called,
+	 * so we can safely collect pull_task() stats here rather than
+	 * inside pull_task().
+	 */
+	schedstat_add(sd, lb_gained[idle], pulled);
+
+	if (all_pinned)
+		*all_pinned = pinned;
+
+	return max_load_move - rem_load_move;
 }
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
@@ -1897,9 +2005,9 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		rem_load = (u64)rem_load_move * busiest_weight;
 		rem_load = div_u64(rem_load, busiest_h_load + 1);
 
-		moved_load = __load_balance_fair(this_rq, this_cpu, busiest,
+		moved_load = balance_tasks(this_rq, this_cpu, busiest,
 				rem_load, sd, idle, all_pinned, this_best_prio,
-				tg->cfs_rq[busiest_cpu]);
+				busiest_cfs_rq);
 
 		if (!moved_load)
 			continue;
@@ -1922,35 +2030,1488 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		  struct sched_domain *sd, enum cpu_idle_type idle,
 		  int *all_pinned, int *this_best_prio)
 {
-	return __load_balance_fair(this_rq, this_cpu, busiest,
+	return balance_tasks(this_rq, this_cpu, busiest,
 			max_load_move, sd, idle, all_pinned,
 			this_best_prio, &busiest->cfs);
 }
 #endif
 
-static int
-move_one_task_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		   struct sched_domain *sd, enum cpu_idle_type idle)
+/*
+ * move_tasks tries to move up to max_load_move weighted load from busiest to
+ * this_rq, as part of a balancing operation within domain "sd".
+ * Returns 1 if successful and 0 otherwise.
+ *
+ * Called with both runqueues locked.
+ */
+static int move_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest,
+		      unsigned long max_load_move,
+		      struct sched_domain *sd, enum cpu_idle_type idle,
+		      int *all_pinned)
 {
-	struct cfs_rq *busy_cfs_rq;
-	struct rq_iterator cfs_rq_iterator;
+	unsigned long total_load_moved = 0, load_moved;
+	int this_best_prio = this_rq->curr->prio;
+
+	do {
+		load_moved = load_balance_fair(this_rq, this_cpu, busiest,
+				max_load_move - total_load_moved,
+				sd, idle, all_pinned, &this_best_prio);
 
-	cfs_rq_iterator.start = load_balance_start_fair;
-	cfs_rq_iterator.next = load_balance_next_fair;
+		total_load_moved += load_moved;
 
-	for_each_leaf_cfs_rq(busiest, busy_cfs_rq) {
+#ifdef CONFIG_PREEMPT
 		/*
-		 * pass busy_cfs_rq argument into
-		 * load_balance_[start|next]_fair iterators
+		 * NEWIDLE balancing is a source of latency, so preemptible
+		 * kernels will stop after the first task is pulled to minimize
+		 * the critical section.
 		 */
-		cfs_rq_iterator.arg = busy_cfs_rq;
-		if (iter_move_one_task(this_rq, this_cpu, busiest, sd, idle,
-				       &cfs_rq_iterator))
-		    return 1;
+		if (idle == CPU_NEWLY_IDLE && this_rq->nr_running)
+			break;
+
+		if (raw_spin_is_contended(&this_rq->lock) ||
+				raw_spin_is_contended(&busiest->lock))
+			break;
+#endif
+	} while (load_moved && max_load_move > total_load_moved);
+
+	return total_load_moved > 0;
+}
+
+/********** Helpers for find_busiest_group ************************/
+/*
+ * sd_lb_stats - Structure to store the statistics of a sched_domain
+ * 		during load balancing.
+ */
+struct sd_lb_stats {
+	struct sched_group *busiest; /* Busiest group in this sd */
+	struct sched_group *this;  /* Local group in this sd */
+	unsigned long total_load;  /* Total load of all groups in sd */
+	unsigned long total_pwr;   /*	Total power of all groups in sd */
+	unsigned long avg_load;	   /* Average load across all groups in sd */
+
+	/** Statistics of this group */
+	unsigned long this_load;
+	unsigned long this_load_per_task;
+	unsigned long this_nr_running;
+
+	/* Statistics of the busiest group */
+	unsigned long max_load;
+	unsigned long busiest_load_per_task;
+	unsigned long busiest_nr_running;
+
+	int group_imb; /* Is there imbalance in this sd */
+#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
+	int power_savings_balance; /* Is powersave balance needed for this sd */
+	struct sched_group *group_min; /* Least loaded group in sd */
+	struct sched_group *group_leader; /* Group which relieves group_min */
+	unsigned long min_load_per_task; /* load_per_task in group_min */
+	unsigned long leader_nr_running; /* Nr running of group_leader */
+	unsigned long min_nr_running; /* Nr running of group_min */
+#endif
+};
+
+/*
+ * sg_lb_stats - stats of a sched_group required for load_balancing
+ */
+struct sg_lb_stats {
+	unsigned long avg_load; /*Avg load across the CPUs of the group */
+	unsigned long group_load; /* Total load over the CPUs of the group */
+	unsigned long sum_nr_running; /* Nr tasks running in the group */
+	unsigned long sum_weighted_load; /* Weighted load of group's tasks */
+	unsigned long group_capacity;
+	int group_imb; /* Is there an imbalance in the group ? */
+};
+
+/**
+ * group_first_cpu - Returns the first cpu in the cpumask of a sched_group.
+ * @group: The group whose first cpu is to be returned.
+ */
+static inline unsigned int group_first_cpu(struct sched_group *group)
+{
+	return cpumask_first(sched_group_cpus(group));
+}
+
+/**
+ * get_sd_load_idx - Obtain the load index for a given sched domain.
+ * @sd: The sched_domain whose load_idx is to be obtained.
+ * @idle: The Idle status of the CPU for whose sd load_icx is obtained.
+ */
+static inline int get_sd_load_idx(struct sched_domain *sd,
+					enum cpu_idle_type idle)
+{
+	int load_idx;
+
+	switch (idle) {
+	case CPU_NOT_IDLE:
+		load_idx = sd->busy_idx;
+		break;
+
+	case CPU_NEWLY_IDLE:
+		load_idx = sd->newidle_idx;
+		break;
+	default:
+		load_idx = sd->idle_idx;
+		break;
+	}
+
+	return load_idx;
+}
+
+
+#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
+/**
+ * init_sd_power_savings_stats - Initialize power savings statistics for
+ * the given sched_domain, during load balancing.
+ *
+ * @sd: Sched domain whose power-savings statistics are to be initialized.
+ * @sds: Variable containing the statistics for sd.
+ * @idle: Idle status of the CPU at which we're performing load-balancing.
+ */
+static inline void init_sd_power_savings_stats(struct sched_domain *sd,
+	struct sd_lb_stats *sds, enum cpu_idle_type idle)
+{
+	/*
+	 * Busy processors will not participate in power savings
+	 * balance.
+	 */
+	if (idle == CPU_NOT_IDLE || !(sd->flags & SD_POWERSAVINGS_BALANCE))
+		sds->power_savings_balance = 0;
+	else {
+		sds->power_savings_balance = 1;
+		sds->min_nr_running = ULONG_MAX;
+		sds->leader_nr_running = 0;
+	}
+}
+
+/**
+ * update_sd_power_savings_stats - Update the power saving stats for a
+ * sched_domain while performing load balancing.
+ *
+ * @group: sched_group belonging to the sched_domain under consideration.
+ * @sds: Variable containing the statistics of the sched_domain
+ * @local_group: Does group contain the CPU for which we're performing
+ * 		load balancing ?
+ * @sgs: Variable containing the statistics of the group.
+ */
+static inline void update_sd_power_savings_stats(struct sched_group *group,
+	struct sd_lb_stats *sds, int local_group, struct sg_lb_stats *sgs)
+{
+
+	if (!sds->power_savings_balance)
+		return;
+
+	/*
+	 * If the local group is idle or completely loaded
+	 * no need to do power savings balance at this domain
+	 */
+	if (local_group && (sds->this_nr_running >= sgs->group_capacity ||
+				!sds->this_nr_running))
+		sds->power_savings_balance = 0;
+
+	/*
+	 * If a group is already running at full capacity or idle,
+	 * don't include that group in power savings calculations
+	 */
+	if (!sds->power_savings_balance ||
+		sgs->sum_nr_running >= sgs->group_capacity ||
+		!sgs->sum_nr_running)
+		return;
+
+	/*
+	 * Calculate the group which has the least non-idle load.
+	 * This is the group from where we need to pick up the load
+	 * for saving power
+	 */
+	if ((sgs->sum_nr_running < sds->min_nr_running) ||
+	    (sgs->sum_nr_running == sds->min_nr_running &&
+	     group_first_cpu(group) > group_first_cpu(sds->group_min))) {
+		sds->group_min = group;
+		sds->min_nr_running = sgs->sum_nr_running;
+		sds->min_load_per_task = sgs->sum_weighted_load /
+						sgs->sum_nr_running;
 	}
 
+	/*
+	 * Calculate the group which is almost near its
+	 * capacity but still has some space to pick up some load
+	 * from other group and save more power
+	 */
+	if (sgs->sum_nr_running + 1 > sgs->group_capacity)
+		return;
+
+	if (sgs->sum_nr_running > sds->leader_nr_running ||
+	    (sgs->sum_nr_running == sds->leader_nr_running &&
+	     group_first_cpu(group) < group_first_cpu(sds->group_leader))) {
+		sds->group_leader = group;
+		sds->leader_nr_running = sgs->sum_nr_running;
+	}
+}
+
+/**
+ * check_power_save_busiest_group - see if there is potential for some power-savings balance
+ * @sds: Variable containing the statistics of the sched_domain
+ *	under consideration.
+ * @this_cpu: Cpu at which we're currently performing load-balancing.
+ * @imbalance: Variable to store the imbalance.
+ *
+ * Description:
+ * Check if we have potential to perform some power-savings balance.
+ * If yes, set the busiest group to be the least loaded group in the
+ * sched_domain, so that it's CPUs can be put to idle.
+ *
+ * Returns 1 if there is potential to perform power-savings balance.
+ * Else returns 0.
+ */
+static inline int check_power_save_busiest_group(struct sd_lb_stats *sds,
+					int this_cpu, unsigned long *imbalance)
+{
+	if (!sds->power_savings_balance)
+		return 0;
+
+	if (sds->this != sds->group_leader ||
+			sds->group_leader == sds->group_min)
+		return 0;
+
+	*imbalance = sds->min_load_per_task;
+	sds->busiest = sds->group_min;
+
+	return 1;
+
+}
+#else /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */
+static inline void init_sd_power_savings_stats(struct sched_domain *sd,
+	struct sd_lb_stats *sds, enum cpu_idle_type idle)
+{
+	return;
+}
+
+static inline void update_sd_power_savings_stats(struct sched_group *group,
+	struct sd_lb_stats *sds, int local_group, struct sg_lb_stats *sgs)
+{
+	return;
+}
+
+static inline int check_power_save_busiest_group(struct sd_lb_stats *sds,
+					int this_cpu, unsigned long *imbalance)
+{
 	return 0;
 }
+#endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */
+
+
+unsigned long default_scale_freq_power(struct sched_domain *sd, int cpu)
+{
+	return SCHED_LOAD_SCALE;
+}
+
+unsigned long __weak arch_scale_freq_power(struct sched_domain *sd, int cpu)
+{
+	return default_scale_freq_power(sd, cpu);
+}
+
+unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu)
+{
+	unsigned long weight = cpumask_weight(sched_domain_span(sd));
+	unsigned long smt_gain = sd->smt_gain;
+
+	smt_gain /= weight;
+
+	return smt_gain;
+}
+
+unsigned long __weak arch_scale_smt_power(struct sched_domain *sd, int cpu)
+{
+	return default_scale_smt_power(sd, cpu);
+}
+
+unsigned long scale_rt_power(int cpu)
+{
+	struct rq *rq = cpu_rq(cpu);
+	u64 total, available;
+
+	sched_avg_update(rq);
+
+	total = sched_avg_period() + (rq->clock - rq->age_stamp);
+	available = total - rq->rt_avg;
+
+	if (unlikely((s64)total < SCHED_LOAD_SCALE))
+		total = SCHED_LOAD_SCALE;
+
+	total >>= SCHED_LOAD_SHIFT;
+
+	return div_u64(available, total);
+}
+
+static void update_cpu_power(struct sched_domain *sd, int cpu)
+{
+	unsigned long weight = cpumask_weight(sched_domain_span(sd));
+	unsigned long power = SCHED_LOAD_SCALE;
+	struct sched_group *sdg = sd->groups;
+
+	if (sched_feat(ARCH_POWER))
+		power *= arch_scale_freq_power(sd, cpu);
+	else
+		power *= default_scale_freq_power(sd, cpu);
+
+	power >>= SCHED_LOAD_SHIFT;
+
+	if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) {
+		if (sched_feat(ARCH_POWER))
+			power *= arch_scale_smt_power(sd, cpu);
+		else
+			power *= default_scale_smt_power(sd, cpu);
+
+		power >>= SCHED_LOAD_SHIFT;
+	}
+
+	power *= scale_rt_power(cpu);
+	power >>= SCHED_LOAD_SHIFT;
+
+	if (!power)
+		power = 1;
+
+	sdg->cpu_power = power;
+}
+
+static void update_group_power(struct sched_domain *sd, int cpu)
+{
+	struct sched_domain *child = sd->child;
+	struct sched_group *group, *sdg = sd->groups;
+	unsigned long power;
+
+	if (!child) {
+		update_cpu_power(sd, cpu);
+		return;
+	}
+
+	power = 0;
+
+	group = child->groups;
+	do {
+		power += group->cpu_power;
+		group = group->next;
+	} while (group != child->groups);
+
+	sdg->cpu_power = power;
+}
+
+/**
+ * update_sg_lb_stats - Update sched_group's statistics for load balancing.
+ * @sd: The sched_domain whose statistics are to be updated.
+ * @group: sched_group whose statistics are to be updated.
+ * @this_cpu: Cpu for which load balance is currently performed.
+ * @idle: Idle status of this_cpu
+ * @load_idx: Load index of sched_domain of this_cpu for load calc.
+ * @sd_idle: Idle status of the sched_domain containing group.
+ * @local_group: Does group contain this_cpu.
+ * @cpus: Set of cpus considered for load balancing.
+ * @balance: Should we balance.
+ * @sgs: variable to hold the statistics for this group.
+ */
+static inline void update_sg_lb_stats(struct sched_domain *sd,
+			struct sched_group *group, int this_cpu,
+			enum cpu_idle_type idle, int load_idx, int *sd_idle,
+			int local_group, const struct cpumask *cpus,
+			int *balance, struct sg_lb_stats *sgs)
+{
+	unsigned long load, max_cpu_load, min_cpu_load;
+	int i;
+	unsigned int balance_cpu = -1, first_idle_cpu = 0;
+	unsigned long sum_avg_load_per_task;
+	unsigned long avg_load_per_task;
+
+	if (local_group)
+		balance_cpu = group_first_cpu(group);
+
+	/* Tally up the load of all CPUs in the group */
+	sum_avg_load_per_task = avg_load_per_task = 0;
+	max_cpu_load = 0;
+	min_cpu_load = ~0UL;
+
+	for_each_cpu_and(i, sched_group_cpus(group), cpus) {
+		struct rq *rq = cpu_rq(i);
+
+		if (*sd_idle && rq->nr_running)
+			*sd_idle = 0;
+
+		/* Bias balancing toward cpus of our domain */
+		if (local_group) {
+			if (idle_cpu(i) && !first_idle_cpu) {
+				first_idle_cpu = 1;
+				balance_cpu = i;
+			}
+
+			load = target_load(i, load_idx);
+		} else {
+			load = source_load(i, load_idx);
+			if (load > max_cpu_load)
+				max_cpu_load = load;
+			if (min_cpu_load > load)
+				min_cpu_load = load;
+		}
+
+		sgs->group_load += load;
+		sgs->sum_nr_running += rq->nr_running;
+		sgs->sum_weighted_load += weighted_cpuload(i);
+
+		sum_avg_load_per_task += cpu_avg_load_per_task(i);
+	}
+
+	/*
+	 * First idle cpu or the first cpu(busiest) in this sched group
+	 * is eligible for doing load balancing at this and above
+	 * domains. In the newly idle case, we will allow all the cpu's
+	 * to do the newly idle load balance.
+	 */
+	if (idle != CPU_NEWLY_IDLE && local_group &&
+	    balance_cpu != this_cpu) {
+		*balance = 0;
+		return;
+	}
+
+	update_group_power(sd, this_cpu);
+
+	/* Adjust by relative CPU power of the group */
+	sgs->avg_load = (sgs->group_load * SCHED_LOAD_SCALE) / group->cpu_power;
+
+
+	/*
+	 * Consider the group unbalanced when the imbalance is larger
+	 * than the average weight of two tasks.
+	 *
+	 * APZ: with cgroup the avg task weight can vary wildly and
+	 *      might not be a suitable number - should we keep a
+	 *      normalized nr_running number somewhere that negates
+	 *      the hierarchy?
+	 */
+	avg_load_per_task = (sum_avg_load_per_task * SCHED_LOAD_SCALE) /
+		group->cpu_power;
+
+	if ((max_cpu_load - min_cpu_load) > 2*avg_load_per_task)
+		sgs->group_imb = 1;
+
+	sgs->group_capacity =
+		DIV_ROUND_CLOSEST(group->cpu_power, SCHED_LOAD_SCALE);
+}
+
+/**
+ * update_sd_lb_stats - Update sched_group's statistics for load balancing.
+ * @sd: sched_domain whose statistics are to be updated.
+ * @this_cpu: Cpu for which load balance is currently performed.
+ * @idle: Idle status of this_cpu
+ * @sd_idle: Idle status of the sched_domain containing group.
+ * @cpus: Set of cpus considered for load balancing.
+ * @balance: Should we balance.
+ * @sds: variable to hold the statistics for this sched_domain.
+ */
+static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu,
+			enum cpu_idle_type idle, int *sd_idle,
+			const struct cpumask *cpus, int *balance,
+			struct sd_lb_stats *sds)
+{
+	struct sched_domain *child = sd->child;
+	struct sched_group *group = sd->groups;
+	struct sg_lb_stats sgs;
+	int load_idx, prefer_sibling = 0;
+
+	if (child && child->flags & SD_PREFER_SIBLING)
+		prefer_sibling = 1;
+
+	init_sd_power_savings_stats(sd, sds, idle);
+	load_idx = get_sd_load_idx(sd, idle);
+
+	do {
+		int local_group;
+
+		local_group = cpumask_test_cpu(this_cpu,
+					       sched_group_cpus(group));
+		memset(&sgs, 0, sizeof(sgs));
+		update_sg_lb_stats(sd, group, this_cpu, idle, load_idx, sd_idle,
+				local_group, cpus, balance, &sgs);
+
+		if (local_group && !(*balance))
+			return;
+
+		sds->total_load += sgs.group_load;
+		sds->total_pwr += group->cpu_power;
+
+		/*
+		 * In case the child domain prefers tasks go to siblings
+		 * first, lower the group capacity to one so that we'll try
+		 * and move all the excess tasks away.
+		 */
+		if (prefer_sibling)
+			sgs.group_capacity = min(sgs.group_capacity, 1UL);
+
+		if (local_group) {
+			sds->this_load = sgs.avg_load;
+			sds->this = group;
+			sds->this_nr_running = sgs.sum_nr_running;
+			sds->this_load_per_task = sgs.sum_weighted_load;
+		} else if (sgs.avg_load > sds->max_load &&
+			   (sgs.sum_nr_running > sgs.group_capacity ||
+				sgs.group_imb)) {
+			sds->max_load = sgs.avg_load;
+			sds->busiest = group;
+			sds->busiest_nr_running = sgs.sum_nr_running;
+			sds->busiest_load_per_task = sgs.sum_weighted_load;
+			sds->group_imb = sgs.group_imb;
+		}
+
+		update_sd_power_savings_stats(group, sds, local_group, &sgs);
+		group = group->next;
+	} while (group != sd->groups);
+}
+
+/**
+ * fix_small_imbalance - Calculate the minor imbalance that exists
+ *			amongst the groups of a sched_domain, during
+ *			load balancing.
+ * @sds: Statistics of the sched_domain whose imbalance is to be calculated.
+ * @this_cpu: The cpu at whose sched_domain we're performing load-balance.
+ * @imbalance: Variable to store the imbalance.
+ */
+static inline void fix_small_imbalance(struct sd_lb_stats *sds,
+				int this_cpu, unsigned long *imbalance)
+{
+	unsigned long tmp, pwr_now = 0, pwr_move = 0;
+	unsigned int imbn = 2;
+
+	if (sds->this_nr_running) {
+		sds->this_load_per_task /= sds->this_nr_running;
+		if (sds->busiest_load_per_task >
+				sds->this_load_per_task)
+			imbn = 1;
+	} else
+		sds->this_load_per_task =
+			cpu_avg_load_per_task(this_cpu);
+
+	if (sds->max_load - sds->this_load + sds->busiest_load_per_task >=
+			sds->busiest_load_per_task * imbn) {
+		*imbalance = sds->busiest_load_per_task;
+		return;
+	}
+
+	/*
+	 * OK, we don't have enough imbalance to justify moving tasks,
+	 * however we may be able to increase total CPU power used by
+	 * moving them.
+	 */
+
+	pwr_now += sds->busiest->cpu_power *
+			min(sds->busiest_load_per_task, sds->max_load);
+	pwr_now += sds->this->cpu_power *
+			min(sds->this_load_per_task, sds->this_load);
+	pwr_now /= SCHED_LOAD_SCALE;
+
+	/* Amount of load we'd subtract */
+	tmp = (sds->busiest_load_per_task * SCHED_LOAD_SCALE) /
+		sds->busiest->cpu_power;
+	if (sds->max_load > tmp)
+		pwr_move += sds->busiest->cpu_power *
+			min(sds->busiest_load_per_task, sds->max_load - tmp);
+
+	/* Amount of load we'd add */
+	if (sds->max_load * sds->busiest->cpu_power <
+		sds->busiest_load_per_task * SCHED_LOAD_SCALE)
+		tmp = (sds->max_load * sds->busiest->cpu_power) /
+			sds->this->cpu_power;
+	else
+		tmp = (sds->busiest_load_per_task * SCHED_LOAD_SCALE) /
+			sds->this->cpu_power;
+	pwr_move += sds->this->cpu_power *
+			min(sds->this_load_per_task, sds->this_load + tmp);
+	pwr_move /= SCHED_LOAD_SCALE;
+
+	/* Move if we gain throughput */
+	if (pwr_move > pwr_now)
+		*imbalance = sds->busiest_load_per_task;
+}
+
+/**
+ * calculate_imbalance - Calculate the amount of imbalance present within the
+ *			 groups of a given sched_domain during load balance.
+ * @sds: statistics of the sched_domain whose imbalance is to be calculated.
+ * @this_cpu: Cpu for which currently load balance is being performed.
+ * @imbalance: The variable to store the imbalance.
+ */
+static inline void calculate_imbalance(struct sd_lb_stats *sds, int this_cpu,
+		unsigned long *imbalance)
+{
+	unsigned long max_pull;
+	/*
+	 * In the presence of smp nice balancing, certain scenarios can have
+	 * max load less than avg load(as we skip the groups at or below
+	 * its cpu_power, while calculating max_load..)
+	 */
+	if (sds->max_load < sds->avg_load) {
+		*imbalance = 0;
+		return fix_small_imbalance(sds, this_cpu, imbalance);
+	}
+
+	/* Don't want to pull so many tasks that a group would go idle */
+	max_pull = min(sds->max_load - sds->avg_load,
+			sds->max_load - sds->busiest_load_per_task);
+
+	/* How much load to actually move to equalise the imbalance */
+	*imbalance = min(max_pull * sds->busiest->cpu_power,
+		(sds->avg_load - sds->this_load) * sds->this->cpu_power)
+			/ SCHED_LOAD_SCALE;
+
+	/*
+	 * if *imbalance is less than the average load per runnable task
+	 * there is no gaurantee that any tasks will be moved so we'll have
+	 * a think about bumping its value to force at least one task to be
+	 * moved
+	 */
+	if (*imbalance < sds->busiest_load_per_task)
+		return fix_small_imbalance(sds, this_cpu, imbalance);
+
+}
+/******* find_busiest_group() helpers end here *********************/
+
+/**
+ * find_busiest_group - Returns the busiest group within the sched_domain
+ * if there is an imbalance. If there isn't an imbalance, and
+ * the user has opted for power-savings, it returns a group whose
+ * CPUs can be put to idle by rebalancing those tasks elsewhere, if
+ * such a group exists.
+ *
+ * Also calculates the amount of weighted load which should be moved
+ * to restore balance.
+ *
+ * @sd: The sched_domain whose busiest group is to be returned.
+ * @this_cpu: The cpu for which load balancing is currently being performed.
+ * @imbalance: Variable which stores amount of weighted load which should
+ *		be moved to restore balance/put a group to idle.
+ * @idle: The idle status of this_cpu.
+ * @sd_idle: The idleness of sd
+ * @cpus: The set of CPUs under consideration for load-balancing.
+ * @balance: Pointer to a variable indicating if this_cpu
+ *	is the appropriate cpu to perform load balancing at this_level.
+ *
+ * Returns:	- the busiest group if imbalance exists.
+ *		- If no imbalance and user has opted for power-savings balance,
+ *		   return the least loaded group whose CPUs can be
+ *		   put to idle by rebalancing its tasks onto our group.
+ */
+static struct sched_group *
+find_busiest_group(struct sched_domain *sd, int this_cpu,
+		   unsigned long *imbalance, enum cpu_idle_type idle,
+		   int *sd_idle, const struct cpumask *cpus, int *balance)
+{
+	struct sd_lb_stats sds;
+
+	memset(&sds, 0, sizeof(sds));
+
+	/*
+	 * Compute the various statistics relavent for load balancing at
+	 * this level.
+	 */
+	update_sd_lb_stats(sd, this_cpu, idle, sd_idle, cpus,
+					balance, &sds);
+
+	/* Cases where imbalance does not exist from POV of this_cpu */
+	/* 1) this_cpu is not the appropriate cpu to perform load balancing
+	 *    at this level.
+	 * 2) There is no busy sibling group to pull from.
+	 * 3) This group is the busiest group.
+	 * 4) This group is more busy than the avg busieness at this
+	 *    sched_domain.
+	 * 5) The imbalance is within the specified limit.
+	 * 6) Any rebalance would lead to ping-pong
+	 */
+	if (!(*balance))
+		goto ret;
+
+	if (!sds.busiest || sds.busiest_nr_running == 0)
+		goto out_balanced;
+
+	if (sds.this_load >= sds.max_load)
+		goto out_balanced;
+
+	sds.avg_load = (SCHED_LOAD_SCALE * sds.total_load) / sds.total_pwr;
+
+	if (sds.this_load >= sds.avg_load)
+		goto out_balanced;
+
+	if (100 * sds.max_load <= sd->imbalance_pct * sds.this_load)
+		goto out_balanced;
+
+	sds.busiest_load_per_task /= sds.busiest_nr_running;
+	if (sds.group_imb)
+		sds.busiest_load_per_task =
+			min(sds.busiest_load_per_task, sds.avg_load);
+
+	/*
+	 * We're trying to get all the cpus to the average_load, so we don't
+	 * want to push ourselves above the average load, nor do we wish to
+	 * reduce the max loaded cpu below the average load, as either of these
+	 * actions would just result in more rebalancing later, and ping-pong
+	 * tasks around. Thus we look for the minimum possible imbalance.
+	 * Negative imbalances (*we* are more loaded than anyone else) will
+	 * be counted as no imbalance for these purposes -- we can't fix that
+	 * by pulling tasks to us. Be careful of negative numbers as they'll
+	 * appear as very large values with unsigned longs.
+	 */
+	if (sds.max_load <= sds.busiest_load_per_task)
+		goto out_balanced;
+
+	/* Looks like there is an imbalance. Compute it */
+	calculate_imbalance(&sds, this_cpu, imbalance);
+	return sds.busiest;
+
+out_balanced:
+	/*
+	 * There is no obvious imbalance. But check if we can do some balancing
+	 * to save power.
+	 */
+	if (check_power_save_busiest_group(&sds, this_cpu, imbalance))
+		return sds.busiest;
+ret:
+	*imbalance = 0;
+	return NULL;
+}
+
+/*
+ * find_busiest_queue - find the busiest runqueue among the cpus in group.
+ */
+static struct rq *
+find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
+		   unsigned long imbalance, const struct cpumask *cpus)
+{
+	struct rq *busiest = NULL, *rq;
+	unsigned long max_load = 0;
+	int i;
+
+	for_each_cpu(i, sched_group_cpus(group)) {
+		unsigned long power = power_of(i);
+		unsigned long capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE);
+		unsigned long wl;
+
+		if (!cpumask_test_cpu(i, cpus))
+			continue;
+
+		rq = cpu_rq(i);
+		wl = weighted_cpuload(i) * SCHED_LOAD_SCALE;
+		wl /= power;
+
+		if (capacity && rq->nr_running == 1 && wl > imbalance)
+			continue;
+
+		if (wl > max_load) {
+			max_load = wl;
+			busiest = rq;
+		}
+	}
+
+	return busiest;
+}
+
+/*
+ * Max backoff if we encounter pinned tasks. Pretty arbitrary value, but
+ * so long as it is large enough.
+ */
+#define MAX_PINNED_INTERVAL	512
+
+/* Working cpumask for load_balance and load_balance_newidle. */
+static DEFINE_PER_CPU(cpumask_var_t, load_balance_tmpmask);
+
+static int need_active_balance(struct sched_domain *sd, int sd_idle, int idle)
+{
+	if (idle == CPU_NEWLY_IDLE) {
+		/*
+		 * The only task running in a non-idle cpu can be moved to this
+		 * cpu in an attempt to completely freeup the other CPU
+		 * package.
+		 *
+		 * The package power saving logic comes from
+		 * find_busiest_group(). If there are no imbalance, then
+		 * f_b_g() will return NULL. However when sched_mc={1,2} then
+		 * f_b_g() will select a group from which a running task may be
+		 * pulled to this cpu in order to make the other package idle.
+		 * If there is no opportunity to make a package idle and if
+		 * there are no imbalance, then f_b_g() will return NULL and no
+		 * action will be taken in load_balance_newidle().
+		 *
+		 * Under normal task pull operation due to imbalance, there
+		 * will be more than one task in the source run queue and
+		 * move_tasks() will succeed.  ld_moved will be true and this
+		 * active balance code will not be triggered.
+		 */
+		if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
+		    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
+			return 0;
+
+		if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
+			return 0;
+	}
+
+	return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2);
+}
+
+/*
+ * Check this_cpu to ensure it is balanced within domain. Attempt to move
+ * tasks if there is an imbalance.
+ */
+static int load_balance(int this_cpu, struct rq *this_rq,
+			struct sched_domain *sd, enum cpu_idle_type idle,
+			int *balance)
+{
+	int ld_moved, all_pinned = 0, active_balance = 0, sd_idle = 0;
+	struct sched_group *group;
+	unsigned long imbalance;
+	struct rq *busiest;
+	unsigned long flags;
+	struct cpumask *cpus = __get_cpu_var(load_balance_tmpmask);
+
+	cpumask_copy(cpus, cpu_active_mask);
+
+	/*
+	 * When power savings policy is enabled for the parent domain, idle
+	 * sibling can pick up load irrespective of busy siblings. In this case,
+	 * let the state of idle sibling percolate up as CPU_IDLE, instead of
+	 * portraying it as CPU_NOT_IDLE.
+	 */
+	if (idle != CPU_NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER &&
+	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
+		sd_idle = 1;
+
+	schedstat_inc(sd, lb_count[idle]);
+
+redo:
+	update_shares(sd);
+	group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle,
+				   cpus, balance);
+
+	if (*balance == 0)
+		goto out_balanced;
+
+	if (!group) {
+		schedstat_inc(sd, lb_nobusyg[idle]);
+		goto out_balanced;
+	}
+
+	busiest = find_busiest_queue(group, idle, imbalance, cpus);
+	if (!busiest) {
+		schedstat_inc(sd, lb_nobusyq[idle]);
+		goto out_balanced;
+	}
+
+	BUG_ON(busiest == this_rq);
+
+	schedstat_add(sd, lb_imbalance[idle], imbalance);
+
+	ld_moved = 0;
+	if (busiest->nr_running > 1) {
+		/*
+		 * Attempt to move tasks. If find_busiest_group has found
+		 * an imbalance but busiest->nr_running <= 1, the group is
+		 * still unbalanced. ld_moved simply stays zero, so it is
+		 * correctly treated as an imbalance.
+		 */
+		local_irq_save(flags);
+		double_rq_lock(this_rq, busiest);
+		ld_moved = move_tasks(this_rq, this_cpu, busiest,
+				      imbalance, sd, idle, &all_pinned);
+		double_rq_unlock(this_rq, busiest);
+		local_irq_restore(flags);
+
+		/*
+		 * some other cpu did the load balance for us.
+		 */
+		if (ld_moved && this_cpu != smp_processor_id())
+			resched_cpu(this_cpu);
+
+		/* All tasks on this runqueue were pinned by CPU affinity */
+		if (unlikely(all_pinned)) {
+			cpumask_clear_cpu(cpu_of(busiest), cpus);
+			if (!cpumask_empty(cpus))
+				goto redo;
+			goto out_balanced;
+		}
+	}
+
+	if (!ld_moved) {
+		schedstat_inc(sd, lb_failed[idle]);
+		sd->nr_balance_failed++;
+
+		if (need_active_balance(sd, sd_idle, idle)) {
+			raw_spin_lock_irqsave(&busiest->lock, flags);
+
+			/* don't kick the migration_thread, if the curr
+			 * task on busiest cpu can't be moved to this_cpu
+			 */
+			if (!cpumask_test_cpu(this_cpu,
+					      &busiest->curr->cpus_allowed)) {
+				raw_spin_unlock_irqrestore(&busiest->lock,
+							    flags);
+				all_pinned = 1;
+				goto out_one_pinned;
+			}
+
+			if (!busiest->active_balance) {
+				busiest->active_balance = 1;
+				busiest->push_cpu = this_cpu;
+				active_balance = 1;
+			}
+			raw_spin_unlock_irqrestore(&busiest->lock, flags);
+			if (active_balance)
+				wake_up_process(busiest->migration_thread);
+
+			/*
+			 * We've kicked active balancing, reset the failure
+			 * counter.
+			 */
+			sd->nr_balance_failed = sd->cache_nice_tries+1;
+		}
+	} else
+		sd->nr_balance_failed = 0;
+
+	if (likely(!active_balance)) {
+		/* We were unbalanced, so reset the balancing interval */
+		sd->balance_interval = sd->min_interval;
+	} else {
+		/*
+		 * If we've begun active balancing, start to back off. This
+		 * case may not be covered by the all_pinned logic if there
+		 * is only 1 task on the busy runqueue (because we don't call
+		 * move_tasks).
+		 */
+		if (sd->balance_interval < sd->max_interval)
+			sd->balance_interval *= 2;
+	}
+
+	if (!ld_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
+	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
+		ld_moved = -1;
+
+	goto out;
+
+out_balanced:
+	schedstat_inc(sd, lb_balanced[idle]);
+
+	sd->nr_balance_failed = 0;
+
+out_one_pinned:
+	/* tune up the balancing interval */
+	if ((all_pinned && sd->balance_interval < MAX_PINNED_INTERVAL) ||
+			(sd->balance_interval < sd->max_interval))
+		sd->balance_interval *= 2;
+
+	if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
+	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
+		ld_moved = -1;
+	else
+		ld_moved = 0;
+out:
+	if (ld_moved)
+		update_shares(sd);
+	return ld_moved;
+}
+
+/*
+ * idle_balance is called by schedule() if this_cpu is about to become
+ * idle. Attempts to pull tasks from other CPUs.
+ */
+static void idle_balance(int this_cpu, struct rq *this_rq)
+{
+	struct sched_domain *sd;
+	int pulled_task = 0;
+	unsigned long next_balance = jiffies + HZ;
+
+	this_rq->idle_stamp = this_rq->clock;
+
+	if (this_rq->avg_idle < sysctl_sched_migration_cost)
+		return;
+
+	/*
+	 * Drop the rq->lock, but keep IRQ/preempt disabled.
+	 */
+	raw_spin_unlock(&this_rq->lock);
+
+	for_each_domain(this_cpu, sd) {
+		unsigned long interval;
+		int balance = 1;
+
+		if (!(sd->flags & SD_LOAD_BALANCE))
+			continue;
+
+		if (sd->flags & SD_BALANCE_NEWIDLE) {
+			/* If we've pulled tasks over stop searching: */
+			pulled_task = load_balance(this_cpu, this_rq,
+						   sd, CPU_NEWLY_IDLE, &balance);
+		}
+
+		interval = msecs_to_jiffies(sd->balance_interval);
+		if (time_after(next_balance, sd->last_balance + interval))
+			next_balance = sd->last_balance + interval;
+		if (pulled_task) {
+			this_rq->idle_stamp = 0;
+			break;
+		}
+	}
+
+	raw_spin_lock(&this_rq->lock);
+
+	if (pulled_task || time_after(jiffies, this_rq->next_balance)) {
+		/*
+		 * We are going idle. next_balance may be set based on
+		 * a busy processor. So reset next_balance.
+		 */
+		this_rq->next_balance = next_balance;
+	}
+}
+
+/*
+ * active_load_balance is run by migration threads. It pushes running tasks
+ * off the busiest CPU onto idle CPUs. It requires at least 1 task to be
+ * running on each physical CPU where possible, and avoids physical /
+ * logical imbalances.
+ *
+ * Called with busiest_rq locked.
+ */
+static void active_load_balance(struct rq *busiest_rq, int busiest_cpu)
+{
+	int target_cpu = busiest_rq->push_cpu;
+	struct sched_domain *sd;
+	struct rq *target_rq;
+
+	/* Is there any task to move? */
+	if (busiest_rq->nr_running <= 1)
+		return;
+
+	target_rq = cpu_rq(target_cpu);
+
+	/*
+	 * This condition is "impossible", if it occurs
+	 * we need to fix it. Originally reported by
+	 * Bjorn Helgaas on a 128-cpu setup.
+	 */
+	BUG_ON(busiest_rq == target_rq);
+
+	/* move a task from busiest_rq to target_rq */
+	double_lock_balance(busiest_rq, target_rq);
+	update_rq_clock(busiest_rq);
+	update_rq_clock(target_rq);
+
+	/* Search for an sd spanning us and the target CPU. */
+	for_each_domain(target_cpu, sd) {
+		if ((sd->flags & SD_LOAD_BALANCE) &&
+		    cpumask_test_cpu(busiest_cpu, sched_domain_span(sd)))
+				break;
+	}
+
+	if (likely(sd)) {
+		schedstat_inc(sd, alb_count);
+
+		if (move_one_task(target_rq, target_cpu, busiest_rq,
+				  sd, CPU_IDLE))
+			schedstat_inc(sd, alb_pushed);
+		else
+			schedstat_inc(sd, alb_failed);
+	}
+	double_unlock_balance(busiest_rq, target_rq);
+}
+
+#ifdef CONFIG_NO_HZ
+static struct {
+	atomic_t load_balancer;
+	cpumask_var_t cpu_mask;
+	cpumask_var_t ilb_grp_nohz_mask;
+} nohz ____cacheline_aligned = {
+	.load_balancer = ATOMIC_INIT(-1),
+};
+
+int get_nohz_load_balancer(void)
+{
+	return atomic_read(&nohz.load_balancer);
+}
+
+#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
+/**
+ * lowest_flag_domain - Return lowest sched_domain containing flag.
+ * @cpu:	The cpu whose lowest level of sched domain is to
+ *		be returned.
+ * @flag:	The flag to check for the lowest sched_domain
+ *		for the given cpu.
+ *
+ * Returns the lowest sched_domain of a cpu which contains the given flag.
+ */
+static inline struct sched_domain *lowest_flag_domain(int cpu, int flag)
+{
+	struct sched_domain *sd;
+
+	for_each_domain(cpu, sd)
+		if (sd && (sd->flags & flag))
+			break;
+
+	return sd;
+}
+
+/**
+ * for_each_flag_domain - Iterates over sched_domains containing the flag.
+ * @cpu:	The cpu whose domains we're iterating over.
+ * @sd:		variable holding the value of the power_savings_sd
+ *		for cpu.
+ * @flag:	The flag to filter the sched_domains to be iterated.
+ *
+ * Iterates over all the scheduler domains for a given cpu that has the 'flag'
+ * set, starting from the lowest sched_domain to the highest.
+ */
+#define for_each_flag_domain(cpu, sd, flag) \
+	for (sd = lowest_flag_domain(cpu, flag); \
+		(sd && (sd->flags & flag)); sd = sd->parent)
+
+/**
+ * is_semi_idle_group - Checks if the given sched_group is semi-idle.
+ * @ilb_group:	group to be checked for semi-idleness
+ *
+ * Returns:	1 if the group is semi-idle. 0 otherwise.
+ *
+ * We define a sched_group to be semi idle if it has atleast one idle-CPU
+ * and atleast one non-idle CPU. This helper function checks if the given
+ * sched_group is semi-idle or not.
+ */
+static inline int is_semi_idle_group(struct sched_group *ilb_group)
+{
+	cpumask_and(nohz.ilb_grp_nohz_mask, nohz.cpu_mask,
+					sched_group_cpus(ilb_group));
+
+	/*
+	 * A sched_group is semi-idle when it has atleast one busy cpu
+	 * and atleast one idle cpu.
+	 */
+	if (cpumask_empty(nohz.ilb_grp_nohz_mask))
+		return 0;
+
+	if (cpumask_equal(nohz.ilb_grp_nohz_mask, sched_group_cpus(ilb_group)))
+		return 0;
+
+	return 1;
+}
+/**
+ * find_new_ilb - Finds the optimum idle load balancer for nomination.
+ * @cpu:	The cpu which is nominating a new idle_load_balancer.
+ *
+ * Returns:	Returns the id of the idle load balancer if it exists,
+ *		Else, returns >= nr_cpu_ids.
+ *
+ * This algorithm picks the idle load balancer such that it belongs to a
+ * semi-idle powersavings sched_domain. The idea is to try and avoid
+ * completely idle packages/cores just for the purpose of idle load balancing
+ * when there are other idle cpu's which are better suited for that job.
+ */
+static int find_new_ilb(int cpu)
+{
+	struct sched_domain *sd;
+	struct sched_group *ilb_group;
+
+	/*
+	 * Have idle load balancer selection from semi-idle packages only
+	 * when power-aware load balancing is enabled
+	 */
+	if (!(sched_smt_power_savings || sched_mc_power_savings))
+		goto out_done;
+
+	/*
+	 * Optimize for the case when we have no idle CPUs or only one
+	 * idle CPU. Don't walk the sched_domain hierarchy in such cases
+	 */
+	if (cpumask_weight(nohz.cpu_mask) < 2)
+		goto out_done;
+
+	for_each_flag_domain(cpu, sd, SD_POWERSAVINGS_BALANCE) {
+		ilb_group = sd->groups;
+
+		do {
+			if (is_semi_idle_group(ilb_group))
+				return cpumask_first(nohz.ilb_grp_nohz_mask);
+
+			ilb_group = ilb_group->next;
+
+		} while (ilb_group != sd->groups);
+	}
+
+out_done:
+	return cpumask_first(nohz.cpu_mask);
+}
+#else /*  (CONFIG_SCHED_MC || CONFIG_SCHED_SMT) */
+static inline int find_new_ilb(int call_cpu)
+{
+	return cpumask_first(nohz.cpu_mask);
+}
+#endif
+
+/*
+ * This routine will try to nominate the ilb (idle load balancing)
+ * owner among the cpus whose ticks are stopped. ilb owner will do the idle
+ * load balancing on behalf of all those cpus. If all the cpus in the system
+ * go into this tickless mode, then there will be no ilb owner (as there is
+ * no need for one) and all the cpus will sleep till the next wakeup event
+ * arrives...
+ *
+ * For the ilb owner, tick is not stopped. And this tick will be used
+ * for idle load balancing. ilb owner will still be part of
+ * nohz.cpu_mask..
+ *
+ * While stopping the tick, this cpu will become the ilb owner if there
+ * is no other owner. And will be the owner till that cpu becomes busy
+ * or if all cpus in the system stop their ticks at which point
+ * there is no need for ilb owner.
+ *
+ * When the ilb owner becomes busy, it nominates another owner, during the
+ * next busy scheduler_tick()
+ */
+int select_nohz_load_balancer(int stop_tick)
+{
+	int cpu = smp_processor_id();
+
+	if (stop_tick) {
+		cpu_rq(cpu)->in_nohz_recently = 1;
+
+		if (!cpu_active(cpu)) {
+			if (atomic_read(&nohz.load_balancer) != cpu)
+				return 0;
+
+			/*
+			 * If we are going offline and still the leader,
+			 * give up!
+			 */
+			if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
+				BUG();
+
+			return 0;
+		}
+
+		cpumask_set_cpu(cpu, nohz.cpu_mask);
+
+		/* time for ilb owner also to sleep */
+		if (cpumask_weight(nohz.cpu_mask) == num_active_cpus()) {
+			if (atomic_read(&nohz.load_balancer) == cpu)
+				atomic_set(&nohz.load_balancer, -1);
+			return 0;
+		}
+
+		if (atomic_read(&nohz.load_balancer) == -1) {
+			/* make me the ilb owner */
+			if (atomic_cmpxchg(&nohz.load_balancer, -1, cpu) == -1)
+				return 1;
+		} else if (atomic_read(&nohz.load_balancer) == cpu) {
+			int new_ilb;
+
+			if (!(sched_smt_power_savings ||
+						sched_mc_power_savings))
+				return 1;
+			/*
+			 * Check to see if there is a more power-efficient
+			 * ilb.
+			 */
+			new_ilb = find_new_ilb(cpu);
+			if (new_ilb < nr_cpu_ids && new_ilb != cpu) {
+				atomic_set(&nohz.load_balancer, -1);
+				resched_cpu(new_ilb);
+				return 0;
+			}
+			return 1;
+		}
+	} else {
+		if (!cpumask_test_cpu(cpu, nohz.cpu_mask))
+			return 0;
+
+		cpumask_clear_cpu(cpu, nohz.cpu_mask);
+
+		if (atomic_read(&nohz.load_balancer) == cpu)
+			if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
+				BUG();
+	}
+	return 0;
+}
+#endif
+
+static DEFINE_SPINLOCK(balancing);
+
+/*
+ * It checks each scheduling domain to see if it is due to be balanced,
+ * and initiates a balancing operation if so.
+ *
+ * Balancing parameters are set up in arch_init_sched_domains.
+ */
+static void rebalance_domains(int cpu, enum cpu_idle_type idle)
+{
+	int balance = 1;
+	struct rq *rq = cpu_rq(cpu);
+	unsigned long interval;
+	struct sched_domain *sd;
+	/* Earliest time when we have to do rebalance again */
+	unsigned long next_balance = jiffies + 60*HZ;
+	int update_next_balance = 0;
+	int need_serialize;
+
+	for_each_domain(cpu, sd) {
+		if (!(sd->flags & SD_LOAD_BALANCE))
+			continue;
+
+		interval = sd->balance_interval;
+		if (idle != CPU_IDLE)
+			interval *= sd->busy_factor;
+
+		/* scale ms to jiffies */
+		interval = msecs_to_jiffies(interval);
+		if (unlikely(!interval))
+			interval = 1;
+		if (interval > HZ*NR_CPUS/10)
+			interval = HZ*NR_CPUS/10;
+
+		need_serialize = sd->flags & SD_SERIALIZE;
+
+		if (need_serialize) {
+			if (!spin_trylock(&balancing))
+				goto out;
+		}
+
+		if (time_after_eq(jiffies, sd->last_balance + interval)) {
+			if (load_balance(cpu, rq, sd, idle, &balance)) {
+				/*
+				 * We've pulled tasks over so either we're no
+				 * longer idle, or one of our SMT siblings is
+				 * not idle.
+				 */
+				idle = CPU_NOT_IDLE;
+			}
+			sd->last_balance = jiffies;
+		}
+		if (need_serialize)
+			spin_unlock(&balancing);
+out:
+		if (time_after(next_balance, sd->last_balance + interval)) {
+			next_balance = sd->last_balance + interval;
+			update_next_balance = 1;
+		}
+
+		/*
+		 * Stop the load balance at this level. There is another
+		 * CPU in our sched group which is doing load balancing more
+		 * actively.
+		 */
+		if (!balance)
+			break;
+	}
+
+	/*
+	 * next_balance will be updated only when there is a need.
+	 * When the cpu is attached to null domain for ex, it will not be
+	 * updated.
+	 */
+	if (likely(update_next_balance))
+		rq->next_balance = next_balance;
+}
+
+/*
+ * run_rebalance_domains is triggered when needed from the scheduler tick.
+ * In CONFIG_NO_HZ case, the idle load balance owner will do the
+ * rebalancing for all the cpus for whom scheduler ticks are stopped.
+ */
+static void run_rebalance_domains(struct softirq_action *h)
+{
+	int this_cpu = smp_processor_id();
+	struct rq *this_rq = cpu_rq(this_cpu);
+	enum cpu_idle_type idle = this_rq->idle_at_tick ?
+						CPU_IDLE : CPU_NOT_IDLE;
+
+	rebalance_domains(this_cpu, idle);
+
+#ifdef CONFIG_NO_HZ
+	/*
+	 * If this cpu is the owner for idle load balancing, then do the
+	 * balancing on behalf of the other idle cpus whose ticks are
+	 * stopped.
+	 */
+	if (this_rq->idle_at_tick &&
+	    atomic_read(&nohz.load_balancer) == this_cpu) {
+		struct rq *rq;
+		int balance_cpu;
+
+		for_each_cpu(balance_cpu, nohz.cpu_mask) {
+			if (balance_cpu == this_cpu)
+				continue;
+
+			/*
+			 * If this cpu gets work to do, stop the load balancing
+			 * work being done for other cpus. Next load
+			 * balancing owner will pick it up.
+			 */
+			if (need_resched())
+				break;
+
+			rebalance_domains(balance_cpu, CPU_IDLE);
+
+			rq = cpu_rq(balance_cpu);
+			if (time_after(this_rq->next_balance, rq->next_balance))
+				this_rq->next_balance = rq->next_balance;
+		}
+	}
+#endif
+}
+
+static inline int on_null_domain(int cpu)
+{
+	return !rcu_dereference(cpu_rq(cpu)->sd);
+}
+
+/*
+ * Trigger the SCHED_SOFTIRQ if it is time to do periodic load balancing.
+ *
+ * In case of CONFIG_NO_HZ, this is the place where we nominate a new
+ * idle load balancing owner or decide to stop the periodic load balancing,
+ * if the whole system is idle.
+ */
+static inline void trigger_load_balance(struct rq *rq, int cpu)
+{
+#ifdef CONFIG_NO_HZ
+	/*
+	 * If we were in the nohz mode recently and busy at the current
+	 * scheduler tick, then check if we need to nominate new idle
+	 * load balancer.
+	 */
+	if (rq->in_nohz_recently && !rq->idle_at_tick) {
+		rq->in_nohz_recently = 0;
+
+		if (atomic_read(&nohz.load_balancer) == cpu) {
+			cpumask_clear_cpu(cpu, nohz.cpu_mask);
+			atomic_set(&nohz.load_balancer, -1);
+		}
+
+		if (atomic_read(&nohz.load_balancer) == -1) {
+			int ilb = find_new_ilb(cpu);
+
+			if (ilb < nr_cpu_ids)
+				resched_cpu(ilb);
+		}
+	}
+
+	/*
+	 * If this cpu is idle and doing idle load balancing for all the
+	 * cpus with ticks stopped, is it time for that to stop?
+	 */
+	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) == cpu &&
+	    cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
+		resched_cpu(cpu);
+		return;
+	}
+
+	/*
+	 * If this cpu is idle and the idle load balancing is done by
+	 * someone else, then no need raise the SCHED_SOFTIRQ
+	 */
+	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) != cpu &&
+	    cpumask_test_cpu(cpu, nohz.cpu_mask))
+		return;
+#endif
+	/* Don't need to rebalance while attached to NULL domain */
+	if (time_after_eq(jiffies, rq->next_balance) &&
+	    likely(!on_null_domain(cpu)))
+		raise_softirq(SCHED_SOFTIRQ);
+}
 
 static void rq_online_fair(struct rq *rq)
 {
@@ -1962,6 +3523,15 @@ static void rq_offline_fair(struct rq *rq)
 	update_sysctl();
 }
 
+#else	/* CONFIG_SMP */
+
+/*
+ * on UP we do not need to balance between CPUs:
+ */
+static inline void idle_balance(int cpu, struct rq *rq)
+{
+}
+
 #endif /* CONFIG_SMP */
 
 /*
@@ -2076,7 +3646,7 @@ static void moved_group_fair(struct task_struct *p, int on_rq)
 }
 #endif
 
-unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task)
+static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task)
 {
 	struct sched_entity *se = &task->se;
 	unsigned int rr_interval = 0;
@@ -2108,8 +3678,6 @@ static const struct sched_class fair_sched_class = {
 #ifdef CONFIG_SMP
 	.select_task_rq		= select_task_rq_fair,
 
-	.load_balance		= load_balance_fair,
-	.move_one_task		= move_one_task_fair,
 	.rq_online		= rq_online_fair,
 	.rq_offline		= rq_offline_fair,
 
diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c
index 5f93b570d383..a8a6d8a50947 100644
--- a/kernel/sched_idletask.c
+++ b/kernel/sched_idletask.c
@@ -44,24 +44,6 @@ static void put_prev_task_idle(struct rq *rq, struct task_struct *prev)
 {
 }
 
-#ifdef CONFIG_SMP
-static unsigned long
-load_balance_idle(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		  unsigned long max_load_move,
-		  struct sched_domain *sd, enum cpu_idle_type idle,
-		  int *all_pinned, int *this_best_prio)
-{
-	return 0;
-}
-
-static int
-move_one_task_idle(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		   struct sched_domain *sd, enum cpu_idle_type idle)
-{
-	return 0;
-}
-#endif
-
 static void task_tick_idle(struct rq *rq, struct task_struct *curr, int queued)
 {
 }
@@ -97,7 +79,7 @@ static void prio_changed_idle(struct rq *rq, struct task_struct *p,
 		check_preempt_curr(rq, p, 0);
 }
 
-unsigned int get_rr_interval_idle(struct rq *rq, struct task_struct *task)
+static unsigned int get_rr_interval_idle(struct rq *rq, struct task_struct *task)
 {
 	return 0;
 }
@@ -119,9 +101,6 @@ static const struct sched_class idle_sched_class = {
 
 #ifdef CONFIG_SMP
 	.select_task_rq		= select_task_rq_idle,
-
-	.load_balance		= load_balance_idle,
-	.move_one_task		= move_one_task_idle,
 #endif
 
 	.set_curr_task          = set_curr_task_idle,
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index f48328ac216f..ca49ceb01201 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -194,7 +194,7 @@ static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
 	return rt_se->my_q;
 }
 
-static void enqueue_rt_entity(struct sched_rt_entity *rt_se);
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se, bool head);
 static void dequeue_rt_entity(struct sched_rt_entity *rt_se);
 
 static void sched_rt_rq_enqueue(struct rt_rq *rt_rq)
@@ -204,7 +204,7 @@ static void sched_rt_rq_enqueue(struct rt_rq *rt_rq)
 
 	if (rt_rq->rt_nr_running) {
 		if (rt_se && !on_rt_rq(rt_se))
-			enqueue_rt_entity(rt_se);
+			enqueue_rt_entity(rt_se, false);
 		if (rt_rq->highest_prio.curr < curr->prio)
 			resched_task(curr);
 	}
@@ -803,7 +803,7 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 	dec_rt_group(rt_se, rt_rq);
 }
 
-static void __enqueue_rt_entity(struct sched_rt_entity *rt_se)
+static void __enqueue_rt_entity(struct sched_rt_entity *rt_se, bool head)
 {
 	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
 	struct rt_prio_array *array = &rt_rq->active;
@@ -819,7 +819,10 @@ static void __enqueue_rt_entity(struct sched_rt_entity *rt_se)
 	if (group_rq && (rt_rq_throttled(group_rq) || !group_rq->rt_nr_running))
 		return;
 
-	list_add_tail(&rt_se->run_list, queue);
+	if (head)
+		list_add(&rt_se->run_list, queue);
+	else
+		list_add_tail(&rt_se->run_list, queue);
 	__set_bit(rt_se_prio(rt_se), array->bitmap);
 
 	inc_rt_tasks(rt_se, rt_rq);
@@ -856,11 +859,11 @@ static void dequeue_rt_stack(struct sched_rt_entity *rt_se)
 	}
 }
 
-static void enqueue_rt_entity(struct sched_rt_entity *rt_se)
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se, bool head)
 {
 	dequeue_rt_stack(rt_se);
 	for_each_sched_rt_entity(rt_se)
-		__enqueue_rt_entity(rt_se);
+		__enqueue_rt_entity(rt_se, head);
 }
 
 static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
@@ -871,21 +874,22 @@ static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
 		struct rt_rq *rt_rq = group_rt_rq(rt_se);
 
 		if (rt_rq && rt_rq->rt_nr_running)
-			__enqueue_rt_entity(rt_se);
+			__enqueue_rt_entity(rt_se, false);
 	}
 }
 
 /*
  * Adding/removing a task to/from a priority array:
  */
-static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
+static void
+enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup, bool head)
 {
 	struct sched_rt_entity *rt_se = &p->rt;
 
 	if (wakeup)
 		rt_se->timeout = 0;
 
-	enqueue_rt_entity(rt_se);
+	enqueue_rt_entity(rt_se, head);
 
 	if (!task_current(rq, p) && p->rt.nr_cpus_allowed > 1)
 		enqueue_pushable_task(rq, p);
@@ -1481,24 +1485,6 @@ static void task_woken_rt(struct rq *rq, struct task_struct *p)
 		push_rt_tasks(rq);
 }
 
-static unsigned long
-load_balance_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		unsigned long max_load_move,
-		struct sched_domain *sd, enum cpu_idle_type idle,
-		int *all_pinned, int *this_best_prio)
-{
-	/* don't touch RT tasks */
-	return 0;
-}
-
-static int
-move_one_task_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		 struct sched_domain *sd, enum cpu_idle_type idle)
-{
-	/* don't touch RT tasks */
-	return 0;
-}
-
 static void set_cpus_allowed_rt(struct task_struct *p,
 				const struct cpumask *new_mask)
 {
@@ -1721,7 +1707,7 @@ static void set_curr_task_rt(struct rq *rq)
 	dequeue_pushable_task(rq, p);
 }
 
-unsigned int get_rr_interval_rt(struct rq *rq, struct task_struct *task)
+static unsigned int get_rr_interval_rt(struct rq *rq, struct task_struct *task)
 {
 	/*
 	 * Time slice is 0 for SCHED_FIFO tasks
@@ -1746,8 +1732,6 @@ static const struct sched_class rt_sched_class = {
 #ifdef CONFIG_SMP
 	.select_task_rq		= select_task_rq_rt,
 
-	.load_balance		= load_balance_rt,
-	.move_one_task		= move_one_task_rt,
 	.set_cpus_allowed       = set_cpus_allowed_rt,
 	.rq_online              = rq_online_rt,
 	.rq_offline             = rq_offline_rt,
diff --git a/kernel/smp.c b/kernel/smp.c
index f10408422444..9867b6bfefce 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -12,8 +12,6 @@
 #include <linux/smp.h>
 #include <linux/cpu.h>
 
-static DEFINE_PER_CPU(struct call_single_queue, call_single_queue);
-
 static struct {
 	struct list_head	queue;
 	raw_spinlock_t		lock;
@@ -33,12 +31,14 @@ struct call_function_data {
 	cpumask_var_t		cpumask;
 };
 
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_function_data, cfd_data);
+
 struct call_single_queue {
 	struct list_head	list;
 	raw_spinlock_t		lock;
 };
 
-static DEFINE_PER_CPU(struct call_function_data, cfd_data);
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_queue, call_single_queue);
 
 static int
 hotplug_cfd(struct notifier_block *nfb, unsigned long action, void *hcpu)
@@ -256,7 +256,7 @@ void generic_smp_call_function_single_interrupt(void)
 	}
 }
 
-static DEFINE_PER_CPU(struct call_single_data, csd_data);
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_data, csd_data);
 
 /*
  * smp_call_function_single - Run a function on a specific CPU
diff --git a/kernel/srcu.c b/kernel/srcu.c
index 818d7d9aa03c..31b275b9c112 100644
--- a/kernel/srcu.c
+++ b/kernel/srcu.c
@@ -144,7 +144,7 @@ EXPORT_SYMBOL_GPL(srcu_read_unlock);
 /*
  * Helper function for synchronize_srcu() and synchronize_srcu_expedited().
  */
-void __synchronize_srcu(struct srcu_struct *sp, void (*sync_func)(void))
+static void __synchronize_srcu(struct srcu_struct *sp, void (*sync_func)(void))
 {
 	int idx;
 
diff --git a/kernel/sys.c b/kernel/sys.c
index c801a55c169f..0e91346be20a 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -569,11 +569,6 @@ static int set_user(struct cred *new)
 	if (!new_user)
 		return -EAGAIN;
 
-	if (!task_can_switch_user(new_user, current)) {
-		free_uid(new_user);
-		return -EINVAL;
-	}
-
 	if (atomic_read(&new_user->processes) >=
 				current->signal->rlim[RLIMIT_NPROC].rlim_cur &&
 			new_user != INIT_USER) {
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 4800f933910e..7c0f180d6e9d 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -58,10 +58,10 @@ static s64			time_offset;
 static long			time_constant = 2;
 
 /* maximum error (usecs):						*/
-long				time_maxerror = NTP_PHASE_LIMIT;
+static long			time_maxerror = NTP_PHASE_LIMIT;
 
 /* estimated error (usecs):						*/
-long				time_esterror = NTP_PHASE_LIMIT;
+static long			time_esterror = NTP_PHASE_LIMIT;
 
 /* frequency offset (scaled nsecs/secs):				*/
 static s64			time_freq;
@@ -142,11 +142,11 @@ static void ntp_update_offset(long offset)
 	 * Select how the frequency is to be controlled
 	 * and in which mode (PLL or FLL).
 	 */
-	secs = xtime.tv_sec - time_reftime;
+	secs = get_seconds() - time_reftime;
 	if (unlikely(time_status & STA_FREQHOLD))
 		secs = 0;
 
-	time_reftime = xtime.tv_sec;
+	time_reftime = get_seconds();
 
 	offset64    = offset;
 	freq_adj    = (offset64 * secs) <<
@@ -368,7 +368,7 @@ static inline void process_adj_status(struct timex *txc, struct timespec *ts)
 	 * reference time to current time.
 	 */
 	if (!(time_status & STA_PLL) && (txc->status & STA_PLL))
-		time_reftime = xtime.tv_sec;
+		time_reftime = get_seconds();
 
 	/* only set allowed bits */
 	time_status &= STA_RONLY;
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index cd9ecd89ec77..d00c6fe23f54 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -51,7 +51,9 @@ endif
 obj-$(CONFIG_EVENT_TRACING) += trace_events.o
 obj-$(CONFIG_EVENT_TRACING) += trace_export.o
 obj-$(CONFIG_FTRACE_SYSCALLS) += trace_syscalls.o
-obj-$(CONFIG_EVENT_PROFILE) += trace_event_profile.o
+ifeq ($(CONFIG_PERF_EVENTS),y)
+obj-$(CONFIG_EVENT_TRACING) += trace_event_profile.o
+endif
 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o
 obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o
 obj-$(CONFIG_KSYM_TRACER) += trace_ksym.o
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index eac6875cb990..032c57ca6502 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -32,6 +32,7 @@
 #include <linux/splice.h>
 #include <linux/kdebug.h>
 #include <linux/string.h>
+#include <linux/rwsem.h>
 #include <linux/ctype.h>
 #include <linux/init.h>
 #include <linux/poll.h>
@@ -102,9 +103,6 @@ static inline void ftrace_enable_cpu(void)
 
 static cpumask_var_t __read_mostly	tracing_buffer_mask;
 
-/* Define which cpu buffers are currently read in trace_pipe */
-static cpumask_var_t			tracing_reader_cpumask;
-
 #define for_each_tracing_cpu(cpu)	\
 	for_each_cpu(cpu, tracing_buffer_mask)
 
@@ -243,12 +241,91 @@ static struct tracer		*current_trace __read_mostly;
 
 /*
  * trace_types_lock is used to protect the trace_types list.
- * This lock is also used to keep user access serialized.
- * Accesses from userspace will grab this lock while userspace
- * activities happen inside the kernel.
  */
 static DEFINE_MUTEX(trace_types_lock);
 
+/*
+ * serialize the access of the ring buffer
+ *
+ * ring buffer serializes readers, but it is low level protection.
+ * The validity of the events (which returns by ring_buffer_peek() ..etc)
+ * are not protected by ring buffer.
+ *
+ * The content of events may become garbage if we allow other process consumes
+ * these events concurrently:
+ *   A) the page of the consumed events may become a normal page
+ *      (not reader page) in ring buffer, and this page will be rewrited
+ *      by events producer.
+ *   B) The page of the consumed events may become a page for splice_read,
+ *      and this page will be returned to system.
+ *
+ * These primitives allow multi process access to different cpu ring buffer
+ * concurrently.
+ *
+ * These primitives don't distinguish read-only and read-consume access.
+ * Multi read-only access are also serialized.
+ */
+
+#ifdef CONFIG_SMP
+static DECLARE_RWSEM(all_cpu_access_lock);
+static DEFINE_PER_CPU(struct mutex, cpu_access_lock);
+
+static inline void trace_access_lock(int cpu)
+{
+	if (cpu == TRACE_PIPE_ALL_CPU) {
+		/* gain it for accessing the whole ring buffer. */
+		down_write(&all_cpu_access_lock);
+	} else {
+		/* gain it for accessing a cpu ring buffer. */
+
+		/* Firstly block other trace_access_lock(TRACE_PIPE_ALL_CPU). */
+		down_read(&all_cpu_access_lock);
+
+		/* Secondly block other access to this @cpu ring buffer. */
+		mutex_lock(&per_cpu(cpu_access_lock, cpu));
+	}
+}
+
+static inline void trace_access_unlock(int cpu)
+{
+	if (cpu == TRACE_PIPE_ALL_CPU) {
+		up_write(&all_cpu_access_lock);
+	} else {
+		mutex_unlock(&per_cpu(cpu_access_lock, cpu));
+		up_read(&all_cpu_access_lock);
+	}
+}
+
+static inline void trace_access_lock_init(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu)
+		mutex_init(&per_cpu(cpu_access_lock, cpu));
+}
+
+#else
+
+static DEFINE_MUTEX(access_lock);
+
+static inline void trace_access_lock(int cpu)
+{
+	(void)cpu;
+	mutex_lock(&access_lock);
+}
+
+static inline void trace_access_unlock(int cpu)
+{
+	(void)cpu;
+	mutex_unlock(&access_lock);
+}
+
+static inline void trace_access_lock_init(void)
+{
+}
+
+#endif
+
 /* trace_wait is a waitqueue for tasks blocked on trace_poll */
 static DECLARE_WAIT_QUEUE_HEAD(trace_wait);
 
@@ -1320,8 +1397,10 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args)
 	entry->fmt			= fmt;
 
 	memcpy(entry->buf, trace_buf, sizeof(u32) * len);
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!filter_check_discard(call, entry, buffer, event)) {
 		ring_buffer_unlock_commit(buffer, event);
+		ftrace_trace_stack(buffer, flags, 6, pc);
+	}
 
 out_unlock:
 	arch_spin_unlock(&trace_buf_lock);
@@ -1394,8 +1473,10 @@ int trace_array_vprintk(struct trace_array *tr,
 
 	memcpy(&entry->buf, trace_buf, len);
 	entry->buf[len] = '\0';
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!filter_check_discard(call, entry, buffer, event)) {
 		ring_buffer_unlock_commit(buffer, event);
+		ftrace_trace_stack(buffer, irq_flags, 6, pc);
+	}
 
  out_unlock:
 	arch_spin_unlock(&trace_buf_lock);
@@ -1585,12 +1666,6 @@ static void tracing_iter_reset(struct trace_iterator *iter, int cpu)
 }
 
 /*
- * No necessary locking here. The worst thing which can
- * happen is loosing events consumed at the same time
- * by a trace_pipe reader.
- * Other than that, we don't risk to crash the ring buffer
- * because it serializes the readers.
- *
  * The current tracer is copied to avoid a global locking
  * all around.
  */
@@ -1645,12 +1720,16 @@ static void *s_start(struct seq_file *m, loff_t *pos)
 	}
 
 	trace_event_read_lock();
+	trace_access_lock(cpu_file);
 	return p;
 }
 
 static void s_stop(struct seq_file *m, void *p)
 {
+	struct trace_iterator *iter = m->private;
+
 	atomic_dec(&trace_record_cmdline_disabled);
+	trace_access_unlock(iter->cpu_file);
 	trace_event_read_unlock();
 }
 
@@ -2841,22 +2920,6 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp)
 
 	mutex_lock(&trace_types_lock);
 
-	/* We only allow one reader per cpu */
-	if (cpu_file == TRACE_PIPE_ALL_CPU) {
-		if (!cpumask_empty(tracing_reader_cpumask)) {
-			ret = -EBUSY;
-			goto out;
-		}
-		cpumask_setall(tracing_reader_cpumask);
-	} else {
-		if (!cpumask_test_cpu(cpu_file, tracing_reader_cpumask))
-			cpumask_set_cpu(cpu_file, tracing_reader_cpumask);
-		else {
-			ret = -EBUSY;
-			goto out;
-		}
-	}
-
 	/* create a buffer to store the information to pass to userspace */
 	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
 	if (!iter) {
@@ -2912,12 +2975,6 @@ static int tracing_release_pipe(struct inode *inode, struct file *file)
 
 	mutex_lock(&trace_types_lock);
 
-	if (iter->cpu_file == TRACE_PIPE_ALL_CPU)
-		cpumask_clear(tracing_reader_cpumask);
-	else
-		cpumask_clear_cpu(iter->cpu_file, tracing_reader_cpumask);
-
-
 	if (iter->trace->pipe_close)
 		iter->trace->pipe_close(iter);
 
@@ -3079,6 +3136,7 @@ waitagain:
 	iter->pos = -1;
 
 	trace_event_read_lock();
+	trace_access_lock(iter->cpu_file);
 	while (find_next_entry_inc(iter) != NULL) {
 		enum print_line_t ret;
 		int len = iter->seq.len;
@@ -3095,6 +3153,7 @@ waitagain:
 		if (iter->seq.len >= cnt)
 			break;
 	}
+	trace_access_unlock(iter->cpu_file);
 	trace_event_read_unlock();
 
 	/* Now copy what we have to the user */
@@ -3220,6 +3279,7 @@ static ssize_t tracing_splice_read_pipe(struct file *filp,
 	}
 
 	trace_event_read_lock();
+	trace_access_lock(iter->cpu_file);
 
 	/* Fill as many pages as possible. */
 	for (i = 0, rem = len; i < PIPE_BUFFERS && rem; i++) {
@@ -3243,6 +3303,7 @@ static ssize_t tracing_splice_read_pipe(struct file *filp,
 		trace_seq_init(&iter->seq);
 	}
 
+	trace_access_unlock(iter->cpu_file);
 	trace_event_read_unlock();
 	mutex_unlock(&iter->mutex);
 
@@ -3544,10 +3605,12 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 
 	info->read = 0;
 
+	trace_access_lock(info->cpu);
 	ret = ring_buffer_read_page(info->tr->buffer,
 				    &info->spare,
 				    count,
 				    info->cpu, 0);
+	trace_access_unlock(info->cpu);
 	if (ret < 0)
 		return 0;
 
@@ -3675,6 +3738,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 		len &= PAGE_MASK;
 	}
 
+	trace_access_lock(info->cpu);
 	entries = ring_buffer_entries_cpu(info->tr->buffer, info->cpu);
 
 	for (i = 0; i < PIPE_BUFFERS && len && entries; i++, len -= PAGE_SIZE) {
@@ -3722,6 +3786,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 		entries = ring_buffer_entries_cpu(info->tr->buffer, info->cpu);
 	}
 
+	trace_access_unlock(info->cpu);
 	spd.nr_pages = i;
 
 	/* did we read anything? */
@@ -4158,6 +4223,8 @@ static __init int tracer_init_debugfs(void)
 	struct dentry *d_tracer;
 	int cpu;
 
+	trace_access_lock_init();
+
 	d_tracer = tracing_init_dentry();
 
 	trace_create_file("tracing_enabled", 0644, d_tracer,
@@ -4392,9 +4459,6 @@ __init static int tracer_alloc_buffers(void)
 	if (!alloc_cpumask_var(&tracing_cpumask, GFP_KERNEL))
 		goto out_free_buffer_mask;
 
-	if (!zalloc_cpumask_var(&tracing_reader_cpumask, GFP_KERNEL))
-		goto out_free_tracing_cpumask;
-
 	/* To save memory, keep the ring buffer size to its minimum */
 	if (ring_buffer_expanded)
 		ring_buf_size = trace_buf_size;
@@ -4452,8 +4516,6 @@ __init static int tracer_alloc_buffers(void)
 	return 0;
 
 out_free_cpumask:
-	free_cpumask_var(tracing_reader_cpumask);
-out_free_tracing_cpumask:
 	free_cpumask_var(tracing_cpumask);
 out_free_buffer_mask:
 	free_cpumask_var(tracing_buffer_mask);
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index e4b32c8aa85f..eae53f958bd5 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -504,7 +504,7 @@ static inline int ftrace_graph_addr(unsigned long addr)
 {
 	int i;
 
-	if (!ftrace_graph_count || test_tsk_trace_graph(current))
+	if (!ftrace_graph_count)
 		return 1;
 
 	for (i = 0; i < ftrace_graph_count; i++) {
diff --git a/kernel/trace/trace_event_profile.c b/kernel/trace/trace_event_profile.c
index 9e25573242cf..f0d693005075 100644
--- a/kernel/trace/trace_event_profile.c
+++ b/kernel/trace/trace_event_profile.c
@@ -6,14 +6,12 @@
  */
 
 #include <linux/module.h>
+#include <linux/kprobes.h>
 #include "trace.h"
 
 
-char *perf_trace_buf;
-EXPORT_SYMBOL_GPL(perf_trace_buf);
-
-char *perf_trace_buf_nmi;
-EXPORT_SYMBOL_GPL(perf_trace_buf_nmi);
+static char *perf_trace_buf;
+static char *perf_trace_buf_nmi;
 
 typedef typeof(char [FTRACE_MAX_PROFILE_SIZE]) perf_trace_t ;
 
@@ -120,3 +118,47 @@ void ftrace_profile_disable(int event_id)
 	}
 	mutex_unlock(&event_mutex);
 }
+
+__kprobes void *ftrace_perf_buf_prepare(int size, unsigned short type,
+					int *rctxp, unsigned long *irq_flags)
+{
+	struct trace_entry *entry;
+	char *trace_buf, *raw_data;
+	int pc, cpu;
+
+	pc = preempt_count();
+
+	/* Protect the per cpu buffer, begin the rcu read side */
+	local_irq_save(*irq_flags);
+
+	*rctxp = perf_swevent_get_recursion_context();
+	if (*rctxp < 0)
+		goto err_recursion;
+
+	cpu = smp_processor_id();
+
+	if (in_nmi())
+		trace_buf = rcu_dereference(perf_trace_buf_nmi);
+	else
+		trace_buf = rcu_dereference(perf_trace_buf);
+
+	if (!trace_buf)
+		goto err;
+
+	raw_data = per_cpu_ptr(trace_buf, cpu);
+
+	/* zero the dead bytes from align to not leak stack to user */
+	*(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL;
+
+	entry = (struct trace_entry *)raw_data;
+	tracing_generic_entry_update(entry, *irq_flags, pc);
+	entry->type = type;
+
+	return raw_data;
+err:
+	perf_swevent_put_recursion_context(*rctxp);
+err_recursion:
+	local_irq_restore(*irq_flags);
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(ftrace_perf_buf_prepare);
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 189b09baf4fb..c2a3077b7353 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -520,41 +520,16 @@ out:
 	return ret;
 }
 
-extern char *__bad_type_size(void);
-
-#undef FIELD
-#define FIELD(type, name)						\
-	sizeof(type) != sizeof(field.name) ? __bad_type_size() :	\
-	#type, "common_" #name, offsetof(typeof(field), name),		\
-		sizeof(field.name), is_signed_type(type)
-
-static int trace_write_header(struct trace_seq *s)
-{
-	struct trace_entry field;
-
-	/* struct trace_entry */
-	return trace_seq_printf(s,
-			"\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n"
-			"\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n"
-			"\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n"
-			"\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n"
-			"\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n"
-			"\n",
-			FIELD(unsigned short, type),
-			FIELD(unsigned char, flags),
-			FIELD(unsigned char, preempt_count),
-			FIELD(int, pid),
-			FIELD(int, lock_depth));
-}
-
 static ssize_t
 event_format_read(struct file *filp, char __user *ubuf, size_t cnt,
 		  loff_t *ppos)
 {
 	struct ftrace_event_call *call = filp->private_data;
+	struct ftrace_event_field *field;
 	struct trace_seq *s;
+	int common_field_count = 5;
 	char *buf;
-	int r;
+	int r = 0;
 
 	if (*ppos)
 		return 0;
@@ -565,14 +540,48 @@ event_format_read(struct file *filp, char __user *ubuf, size_t cnt,
 
 	trace_seq_init(s);
 
-	/* If any of the first writes fail, so will the show_format. */
-
 	trace_seq_printf(s, "name: %s\n", call->name);
 	trace_seq_printf(s, "ID: %d\n", call->id);
 	trace_seq_printf(s, "format:\n");
-	trace_write_header(s);
 
-	r = call->show_format(call, s);
+	list_for_each_entry_reverse(field, &call->fields, link) {
+		/*
+		 * Smartly shows the array type(except dynamic array).
+		 * Normal:
+		 *	field:TYPE VAR
+		 * If TYPE := TYPE[LEN], it is shown:
+		 *	field:TYPE VAR[LEN]
+		 */
+		const char *array_descriptor = strchr(field->type, '[');
+
+		if (!strncmp(field->type, "__data_loc", 10))
+			array_descriptor = NULL;
+
+		if (!array_descriptor) {
+			r = trace_seq_printf(s, "\tfield:%s %s;\toffset:%u;"
+					"\tsize:%u;\tsigned:%d;\n",
+					field->type, field->name, field->offset,
+					field->size, !!field->is_signed);
+		} else {
+			r = trace_seq_printf(s, "\tfield:%.*s %s%s;\toffset:%u;"
+					"\tsize:%u;\tsigned:%d;\n",
+					(int)(array_descriptor - field->type),
+					field->type, field->name,
+					array_descriptor, field->offset,
+					field->size, !!field->is_signed);
+		}
+
+		if (--common_field_count == 0)
+			r = trace_seq_printf(s, "\n");
+
+		if (!r)
+			break;
+	}
+
+	if (r)
+		r = trace_seq_printf(s, "\nprint fmt: %s\n",
+				call->print_fmt);
+
 	if (!r) {
 		/*
 		 * ug!  The format output is bigger than a PAGE!!
@@ -948,10 +957,6 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events,
 				  filter);
 	}
 
-	/* A trace may not want to export its format */
-	if (!call->show_format)
-		return 0;
-
 	trace_create_file("format", 0444, call->dir, call,
 			  format);
 
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index e42af9aad69f..4615f62a04f1 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -1371,7 +1371,7 @@ out_unlock:
 	return err;
 }
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 
 void ftrace_profile_free_filter(struct perf_event *event)
 {
@@ -1439,5 +1439,5 @@ out_unlock:
 	return err;
 }
 
-#endif /* CONFIG_EVENT_PROFILE */
+#endif /* CONFIG_PERF_EVENTS */
 
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index d4fa5dc1ee4e..e091f64ba6ce 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -62,78 +62,6 @@ static void __always_unused ____ftrace_check_##name(void)	\
 
 #include "trace_entries.h"
 
-
-#undef __field
-#define __field(type, item)						\
-	ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t"	\
-			       "offset:%zu;\tsize:%zu;\tsigned:%u;\n",	\
-			       offsetof(typeof(field), item),		\
-			       sizeof(field.item), is_signed_type(type)); \
-	if (!ret)							\
-		return 0;
-
-#undef __field_desc
-#define __field_desc(type, container, item)				\
-	ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t"	\
-			       "offset:%zu;\tsize:%zu;\tsigned:%u;\n",	\
-			       offsetof(typeof(field), container.item),	\
-			       sizeof(field.container.item),		\
-			       is_signed_type(type));			\
-	if (!ret)							\
-		return 0;
-
-#undef __array
-#define __array(type, item, len)					\
-	ret = trace_seq_printf(s, "\tfield:" #type " " #item "[" #len "];\t" \
-			       "offset:%zu;\tsize:%zu;\tsigned:%u;\n",	\
-			       offsetof(typeof(field), item),		\
-			       sizeof(field.item), is_signed_type(type)); \
-	if (!ret)							\
-		return 0;
-
-#undef __array_desc
-#define __array_desc(type, container, item, len)			\
-	ret = trace_seq_printf(s, "\tfield:" #type " " #item "[" #len "];\t" \
-			       "offset:%zu;\tsize:%zu;\tsigned:%u;\n",	\
-			       offsetof(typeof(field), container.item),	\
-			       sizeof(field.container.item),		\
-			       is_signed_type(type));			\
-	if (!ret)							\
-		return 0;
-
-#undef __dynamic_array
-#define __dynamic_array(type, item)					\
-	ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t"	\
-			       "offset:%zu;\tsize:0;\tsigned:%u;\n",	\
-			       offsetof(typeof(field), item),		\
-			       is_signed_type(type));			\
-	if (!ret)							\
-		return 0;
-
-#undef F_printk
-#define F_printk(fmt, args...) "%s, %s\n", #fmt, __stringify(args)
-
-#undef __entry
-#define __entry REC
-
-#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print)		\
-static int								\
-ftrace_format_##name(struct ftrace_event_call *unused,			\
-		     struct trace_seq *s)				\
-{									\
-	struct struct_name field __attribute__((unused));		\
-	int ret = 0;							\
-									\
-	tstruct;							\
-									\
-	trace_seq_printf(s, "\nprint fmt: " print);			\
-									\
-	return ret;							\
-}
-
-#include "trace_entries.h"
-
 #undef __field
 #define __field(type, item)						\
 	ret = trace_define_field(event_call, #type, #item,		\
@@ -175,7 +103,12 @@ ftrace_format_##name(struct ftrace_event_call *unused,			\
 		return ret;
 
 #undef __dynamic_array
-#define __dynamic_array(type, item)
+#define __dynamic_array(type, item)					\
+	ret = trace_define_field(event_call, #type, #item,		\
+				 offsetof(typeof(field), item),		\
+				 0, is_signed_type(type), FILTER_OTHER);\
+	if (ret)							\
+		return ret;
 
 #undef FTRACE_ENTRY
 #define FTRACE_ENTRY(name, struct_name, id, tstruct, print)		\
@@ -198,6 +131,9 @@ static int ftrace_raw_init_event(struct ftrace_event_call *call)
 	return 0;
 }
 
+#undef __entry
+#define __entry REC
+
 #undef __field
 #define __field(type, item)
 
@@ -213,6 +149,9 @@ static int ftrace_raw_init_event(struct ftrace_event_call *call)
 #undef __dynamic_array
 #define __dynamic_array(type, item)
 
+#undef F_printk
+#define F_printk(fmt, args...) #fmt ", "  __stringify(args)
+
 #undef FTRACE_ENTRY
 #define FTRACE_ENTRY(call, struct_name, type, tstruct, print)		\
 									\
@@ -223,7 +162,7 @@ __attribute__((section("_ftrace_events"))) event_##call = {		\
 	.id			= type,					\
 	.system			= __stringify(TRACE_SYSTEM),		\
 	.raw_init		= ftrace_raw_init_event,		\
-	.show_format		= ftrace_format_##call,			\
+	.print_fmt		= print,				\
 	.define_fields		= ftrace_define_fields_##call,		\
 };									\
 
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index b1342c5d37cf..616b135c9eb9 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -212,13 +212,11 @@ int trace_graph_entry(struct ftrace_graph_ent *trace)
 	int cpu;
 	int pc;
 
-	if (unlikely(!tr))
-		return 0;
-
 	if (!ftrace_trace_task(current))
 		return 0;
 
-	if (!ftrace_graph_addr(trace->func))
+	/* trace it when it is-nested-in or is a function enabled. */
+	if (!(trace->depth || ftrace_graph_addr(trace->func)))
 		return 0;
 
 	local_irq_save(flags);
@@ -231,9 +229,6 @@ int trace_graph_entry(struct ftrace_graph_ent *trace)
 	} else {
 		ret = 0;
 	}
-	/* Only do the atomic if it is not already set */
-	if (!test_tsk_trace_graph(current))
-		set_tsk_trace_graph(current);
 
 	atomic_dec(&data->disabled);
 	local_irq_restore(flags);
@@ -281,17 +276,24 @@ void trace_graph_return(struct ftrace_graph_ret *trace)
 		pc = preempt_count();
 		__trace_graph_return(tr, trace, flags, pc);
 	}
-	if (!trace->depth)
-		clear_tsk_trace_graph(current);
 	atomic_dec(&data->disabled);
 	local_irq_restore(flags);
 }
 
+void set_graph_array(struct trace_array *tr)
+{
+	graph_array = tr;
+
+	/* Make graph_array visible before we start tracing */
+
+	smp_mb();
+}
+
 static int graph_trace_init(struct trace_array *tr)
 {
 	int ret;
 
-	graph_array = tr;
+	set_graph_array(tr);
 	ret = register_ftrace_graph(&trace_graph_return,
 				    &trace_graph_entry);
 	if (ret)
@@ -301,11 +303,6 @@ static int graph_trace_init(struct trace_array *tr)
 	return 0;
 }
 
-void set_graph_array(struct trace_array *tr)
-{
-	graph_array = tr;
-}
-
 static void graph_trace_reset(struct trace_array *tr)
 {
 	tracing_stop_cmdline_record();
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 6ea90c0e2c96..9d085d3cbdef 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -91,11 +91,6 @@ static __kprobes unsigned long fetch_memory(struct pt_regs *regs, void *addr)
 	return retval;
 }
 
-static __kprobes unsigned long fetch_argument(struct pt_regs *regs, void *num)
-{
-	return regs_get_argument_nth(regs, (unsigned int)((unsigned long)num));
-}
-
 static __kprobes unsigned long fetch_retvalue(struct pt_regs *regs,
 					      void *dummy)
 {
@@ -231,9 +226,7 @@ static int probe_arg_string(char *buf, size_t n, struct fetch_func *ff)
 {
 	int ret = -EINVAL;
 
-	if (ff->func == fetch_argument)
-		ret = snprintf(buf, n, "$arg%lu", (unsigned long)ff->data);
-	else if (ff->func == fetch_register) {
+	if (ff->func == fetch_register) {
 		const char *name;
 		name = regs_query_register_name((unsigned int)((long)ff->data));
 		ret = snprintf(buf, n, "%%%s", name);
@@ -489,14 +482,6 @@ static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return)
 			}
 		} else
 			ret = -EINVAL;
-	} else if (strncmp(arg, "arg", 3) == 0 && isdigit(arg[3])) {
-		ret = strict_strtoul(arg + 3, 10, &param);
-		if (ret || param > PARAM_MAX_ARGS)
-			ret = -EINVAL;
-		else {
-			ff->func = fetch_argument;
-			ff->data = (void *)param;
-		}
 	} else
 		ret = -EINVAL;
 	return ret;
@@ -611,7 +596,6 @@ static int create_trace_probe(int argc, char **argv)
 	 *  - Add kprobe: p[:[GRP/]EVENT] KSYM[+OFFS]|KADDR [FETCHARGS]
 	 *  - Add kretprobe: r[:[GRP/]EVENT] KSYM[+0] [FETCHARGS]
 	 * Fetch args:
-	 *  $argN	: fetch Nth of function argument. (N:0-)
 	 *  $retval	: fetch return value
 	 *  $stack	: fetch stack address
 	 *  $stackN	: fetch Nth of stack (N:0-)
@@ -958,7 +942,7 @@ static const struct file_operations kprobe_profile_ops = {
 };
 
 /* Kprobe handler */
-static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
+static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
 	struct kprobe_trace_entry *entry;
@@ -978,7 +962,7 @@ static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 	event = trace_current_buffer_lock_reserve(&buffer, call->id, size,
 						  irq_flags, pc);
 	if (!event)
-		return 0;
+		return;
 
 	entry = ring_buffer_event_data(event);
 	entry->nargs = tp->nr_args;
@@ -988,11 +972,10 @@ static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 
 	if (!filter_current_check_discard(buffer, call, entry, event))
 		trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc);
-	return 0;
 }
 
 /* Kretprobe handler */
-static __kprobes int kretprobe_trace_func(struct kretprobe_instance *ri,
+static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 					  struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
@@ -1011,7 +994,7 @@ static __kprobes int kretprobe_trace_func(struct kretprobe_instance *ri,
 	event = trace_current_buffer_lock_reserve(&buffer, call->id, size,
 						  irq_flags, pc);
 	if (!event)
-		return 0;
+		return;
 
 	entry = ring_buffer_event_data(event);
 	entry->nargs = tp->nr_args;
@@ -1022,8 +1005,6 @@ static __kprobes int kretprobe_trace_func(struct kretprobe_instance *ri,
 
 	if (!filter_current_check_discard(buffer, call, entry, event))
 		trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc);
-
-	return 0;
 }
 
 /* Event entry printers */
@@ -1174,213 +1155,123 @@ static int kretprobe_event_define_fields(struct ftrace_event_call *event_call)
 	return 0;
 }
 
-static int __probe_event_show_format(struct trace_seq *s,
-				     struct trace_probe *tp, const char *fmt,
-				     const char *arg)
+static int __set_print_fmt(struct trace_probe *tp, char *buf, int len)
 {
 	int i;
+	int pos = 0;
 
-	/* Show format */
-	if (!trace_seq_printf(s, "\nprint fmt: \"%s", fmt))
-		return 0;
+	const char *fmt, *arg;
 
-	for (i = 0; i < tp->nr_args; i++)
-		if (!trace_seq_printf(s, " %s=%%lx", tp->args[i].name))
-			return 0;
+	if (!probe_is_return(tp)) {
+		fmt = "(%lx)";
+		arg = "REC->" FIELD_STRING_IP;
+	} else {
+		fmt = "(%lx <- %lx)";
+		arg = "REC->" FIELD_STRING_FUNC ", REC->" FIELD_STRING_RETIP;
+	}
 
-	if (!trace_seq_printf(s, "\", %s", arg))
-		return 0;
+	/* When len=0, we just calculate the needed length */
+#define LEN_OR_ZERO (len ? len - pos : 0)
 
-	for (i = 0; i < tp->nr_args; i++)
-		if (!trace_seq_printf(s, ", REC->%s", tp->args[i].name))
-			return 0;
-
-	return trace_seq_puts(s, "\n");
-}
+	pos += snprintf(buf + pos, LEN_OR_ZERO, "\"%s", fmt);
 
-#undef SHOW_FIELD
-#define SHOW_FIELD(type, item, name)					\
-	do {								\
-		ret = trace_seq_printf(s, "\tfield:" #type " %s;\t"	\
-				"offset:%u;\tsize:%u;\tsigned:%d;\n", name,\
-				(unsigned int)offsetof(typeof(field), item),\
-				(unsigned int)sizeof(type),		\
-				is_signed_type(type));			\
-		if (!ret)						\
-			return 0;					\
-	} while (0)
+	for (i = 0; i < tp->nr_args; i++) {
+		pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%%lx",
+				tp->args[i].name);
+	}
 
-static int kprobe_event_show_format(struct ftrace_event_call *call,
-				    struct trace_seq *s)
-{
-	struct kprobe_trace_entry field __attribute__((unused));
-	int ret, i;
-	struct trace_probe *tp = (struct trace_probe *)call->data;
+	pos += snprintf(buf + pos, LEN_OR_ZERO, "\", %s", arg);
 
-	SHOW_FIELD(unsigned long, ip, FIELD_STRING_IP);
-	SHOW_FIELD(int, nargs, FIELD_STRING_NARGS);
+	for (i = 0; i < tp->nr_args; i++) {
+		pos += snprintf(buf + pos, LEN_OR_ZERO, ", REC->%s",
+				tp->args[i].name);
+	}
 
-	/* Show fields */
-	for (i = 0; i < tp->nr_args; i++)
-		SHOW_FIELD(unsigned long, args[i], tp->args[i].name);
-	trace_seq_puts(s, "\n");
+#undef LEN_OR_ZERO
 
-	return __probe_event_show_format(s, tp, "(%lx)",
-					 "REC->" FIELD_STRING_IP);
+	/* return the length of print_fmt */
+	return pos;
 }
 
-static int kretprobe_event_show_format(struct ftrace_event_call *call,
-				       struct trace_seq *s)
+static int set_print_fmt(struct trace_probe *tp)
 {
-	struct kretprobe_trace_entry field __attribute__((unused));
-	int ret, i;
-	struct trace_probe *tp = (struct trace_probe *)call->data;
+	int len;
+	char *print_fmt;
 
-	SHOW_FIELD(unsigned long, func, FIELD_STRING_FUNC);
-	SHOW_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP);
-	SHOW_FIELD(int, nargs, FIELD_STRING_NARGS);
+	/* First: called with 0 length to calculate the needed length */
+	len = __set_print_fmt(tp, NULL, 0);
+	print_fmt = kmalloc(len + 1, GFP_KERNEL);
+	if (!print_fmt)
+		return -ENOMEM;
 
-	/* Show fields */
-	for (i = 0; i < tp->nr_args; i++)
-		SHOW_FIELD(unsigned long, args[i], tp->args[i].name);
-	trace_seq_puts(s, "\n");
+	/* Second: actually write the @print_fmt */
+	__set_print_fmt(tp, print_fmt, len + 1);
+	tp->call.print_fmt = print_fmt;
 
-	return __probe_event_show_format(s, tp, "(%lx <- %lx)",
-					 "REC->" FIELD_STRING_FUNC
-					 ", REC->" FIELD_STRING_RETIP);
+	return 0;
 }
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 
 /* Kprobe profile handler */
-static __kprobes int kprobe_profile_func(struct kprobe *kp,
+static __kprobes void kprobe_profile_func(struct kprobe *kp,
 					 struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
 	struct ftrace_event_call *call = &tp->call;
 	struct kprobe_trace_entry *entry;
-	struct trace_entry *ent;
-	int size, __size, i, pc, __cpu;
+	int size, __size, i;
 	unsigned long irq_flags;
-	char *trace_buf;
-	char *raw_data;
 	int rctx;
 
-	pc = preempt_count();
 	__size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args);
 	size = ALIGN(__size + sizeof(u32), sizeof(u64));
 	size -= sizeof(u32);
 	if (WARN_ONCE(size > FTRACE_MAX_PROFILE_SIZE,
 		     "profile buffer not large enough"))
-		return 0;
-
-	/*
-	 * Protect the non nmi buffer
-	 * This also protects the rcu read side
-	 */
-	local_irq_save(irq_flags);
-
-	rctx = perf_swevent_get_recursion_context();
-	if (rctx < 0)
-		goto end_recursion;
-
-	__cpu = smp_processor_id();
-
-	if (in_nmi())
-		trace_buf = rcu_dereference(perf_trace_buf_nmi);
-	else
-		trace_buf = rcu_dereference(perf_trace_buf);
+		return;
 
-	if (!trace_buf)
-		goto end;
-
-	raw_data = per_cpu_ptr(trace_buf, __cpu);
-
-	/* Zero dead bytes from alignment to avoid buffer leak to userspace */
-	*(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL;
-	entry = (struct kprobe_trace_entry *)raw_data;
-	ent = &entry->ent;
+	entry = ftrace_perf_buf_prepare(size, call->id, &rctx, &irq_flags);
+	if (!entry)
+		return;
 
-	tracing_generic_entry_update(ent, irq_flags, pc);
-	ent->type = call->id;
 	entry->nargs = tp->nr_args;
 	entry->ip = (unsigned long)kp->addr;
 	for (i = 0; i < tp->nr_args; i++)
 		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
-	perf_tp_event(call->id, entry->ip, 1, entry, size);
-
-end:
-	perf_swevent_put_recursion_context(rctx);
-end_recursion:
-	local_irq_restore(irq_flags);
 
-	return 0;
+	ftrace_perf_buf_submit(entry, size, rctx, entry->ip, 1, irq_flags);
 }
 
 /* Kretprobe profile handler */
-static __kprobes int kretprobe_profile_func(struct kretprobe_instance *ri,
+static __kprobes void kretprobe_profile_func(struct kretprobe_instance *ri,
 					    struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
 	struct ftrace_event_call *call = &tp->call;
 	struct kretprobe_trace_entry *entry;
-	struct trace_entry *ent;
-	int size, __size, i, pc, __cpu;
+	int size, __size, i;
 	unsigned long irq_flags;
-	char *trace_buf;
-	char *raw_data;
 	int rctx;
 
-	pc = preempt_count();
 	__size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args);
 	size = ALIGN(__size + sizeof(u32), sizeof(u64));
 	size -= sizeof(u32);
 	if (WARN_ONCE(size > FTRACE_MAX_PROFILE_SIZE,
 		     "profile buffer not large enough"))
-		return 0;
-
-	/*
-	 * Protect the non nmi buffer
-	 * This also protects the rcu read side
-	 */
-	local_irq_save(irq_flags);
-
-	rctx = perf_swevent_get_recursion_context();
-	if (rctx < 0)
-		goto end_recursion;
-
-	__cpu = smp_processor_id();
+		return;
 
-	if (in_nmi())
-		trace_buf = rcu_dereference(perf_trace_buf_nmi);
-	else
-		trace_buf = rcu_dereference(perf_trace_buf);
-
-	if (!trace_buf)
-		goto end;
-
-	raw_data = per_cpu_ptr(trace_buf, __cpu);
-
-	/* Zero dead bytes from alignment to avoid buffer leak to userspace */
-	*(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL;
-	entry = (struct kretprobe_trace_entry *)raw_data;
-	ent = &entry->ent;
+	entry = ftrace_perf_buf_prepare(size, call->id, &rctx, &irq_flags);
+	if (!entry)
+		return;
 
-	tracing_generic_entry_update(ent, irq_flags, pc);
-	ent->type = call->id;
 	entry->nargs = tp->nr_args;
 	entry->func = (unsigned long)tp->rp.kp.addr;
 	entry->ret_ip = (unsigned long)ri->ret_addr;
 	for (i = 0; i < tp->nr_args; i++)
 		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
-	perf_tp_event(call->id, entry->ret_ip, 1, entry, size);
-
-end:
-	perf_swevent_put_recursion_context(rctx);
-end_recursion:
-	local_irq_restore(irq_flags);
 
-	return 0;
+	ftrace_perf_buf_submit(entry, size, rctx, entry->ret_ip, 1, irq_flags);
 }
 
 static int probe_profile_enable(struct ftrace_event_call *call)
@@ -1408,7 +1299,7 @@ static void probe_profile_disable(struct ftrace_event_call *call)
 			disable_kprobe(&tp->rp.kp);
 	}
 }
-#endif	/* CONFIG_EVENT_PROFILE */
+#endif	/* CONFIG_PERF_EVENTS */
 
 
 static __kprobes
@@ -1418,10 +1309,10 @@ int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs)
 
 	if (tp->flags & TP_FLAG_TRACE)
 		kprobe_trace_func(kp, regs);
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 	if (tp->flags & TP_FLAG_PROFILE)
 		kprobe_profile_func(kp, regs);
-#endif	/* CONFIG_EVENT_PROFILE */
+#endif
 	return 0;	/* We don't tweek kernel, so just return 0 */
 }
 
@@ -1432,10 +1323,10 @@ int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs)
 
 	if (tp->flags & TP_FLAG_TRACE)
 		kretprobe_trace_func(ri, regs);
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 	if (tp->flags & TP_FLAG_PROFILE)
 		kretprobe_profile_func(ri, regs);
-#endif	/* CONFIG_EVENT_PROFILE */
+#endif
 	return 0;	/* We don't tweek kernel, so just return 0 */
 }
 
@@ -1448,23 +1339,25 @@ static int register_probe_event(struct trace_probe *tp)
 	if (probe_is_return(tp)) {
 		tp->event.trace = print_kretprobe_event;
 		call->raw_init = probe_event_raw_init;
-		call->show_format = kretprobe_event_show_format;
 		call->define_fields = kretprobe_event_define_fields;
 	} else {
 		tp->event.trace = print_kprobe_event;
 		call->raw_init = probe_event_raw_init;
-		call->show_format = kprobe_event_show_format;
 		call->define_fields = kprobe_event_define_fields;
 	}
+	if (set_print_fmt(tp) < 0)
+		return -ENOMEM;
 	call->event = &tp->event;
 	call->id = register_ftrace_event(&tp->event);
-	if (!call->id)
+	if (!call->id) {
+		kfree(call->print_fmt);
 		return -ENODEV;
+	}
 	call->enabled = 0;
 	call->regfunc = probe_event_enable;
 	call->unregfunc = probe_event_disable;
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 	call->profile_enable = probe_profile_enable;
 	call->profile_disable = probe_profile_disable;
 #endif
@@ -1472,6 +1365,7 @@ static int register_probe_event(struct trace_probe *tp)
 	ret = trace_add_event_call(call);
 	if (ret) {
 		pr_info("Failed to register kprobe event: %s\n", call->name);
+		kfree(call->print_fmt);
 		unregister_ftrace_event(&tp->event);
 	}
 	return ret;
@@ -1481,6 +1375,7 @@ static void unregister_probe_event(struct trace_probe *tp)
 {
 	/* tp->event is unregistered in trace_remove_event_call() */
 	trace_remove_event_call(&tp->call);
+	kfree(tp->call.print_fmt);
 }
 
 /* Make a debugfs interface for controling probe points */
@@ -1523,28 +1418,67 @@ static int kprobe_trace_selftest_target(int a1, int a2, int a3,
 
 static __init int kprobe_trace_self_tests_init(void)
 {
-	int ret;
+	int ret, warn = 0;
 	int (*target)(int, int, int, int, int, int);
+	struct trace_probe *tp;
 
 	target = kprobe_trace_selftest_target;
 
 	pr_info("Testing kprobe tracing: ");
 
 	ret = command_trace_probe("p:testprobe kprobe_trace_selftest_target "
-				  "$arg1 $arg2 $arg3 $arg4 $stack $stack0");
-	if (WARN_ON_ONCE(ret))
-		pr_warning("error enabling function entry\n");
+				  "$stack $stack0 +0($stack)");
+	if (WARN_ON_ONCE(ret)) {
+		pr_warning("error on probing function entry.\n");
+		warn++;
+	} else {
+		/* Enable trace point */
+		tp = find_probe_event("testprobe", KPROBE_EVENT_SYSTEM);
+		if (WARN_ON_ONCE(tp == NULL)) {
+			pr_warning("error on getting new probe.\n");
+			warn++;
+		} else
+			probe_event_enable(&tp->call);
+	}
 
 	ret = command_trace_probe("r:testprobe2 kprobe_trace_selftest_target "
 				  "$retval");
-	if (WARN_ON_ONCE(ret))
-		pr_warning("error enabling function return\n");
+	if (WARN_ON_ONCE(ret)) {
+		pr_warning("error on probing function return.\n");
+		warn++;
+	} else {
+		/* Enable trace point */
+		tp = find_probe_event("testprobe2", KPROBE_EVENT_SYSTEM);
+		if (WARN_ON_ONCE(tp == NULL)) {
+			pr_warning("error on getting new probe.\n");
+			warn++;
+		} else
+			probe_event_enable(&tp->call);
+	}
+
+	if (warn)
+		goto end;
 
 	ret = target(1, 2, 3, 4, 5, 6);
 
-	cleanup_all_probes();
+	ret = command_trace_probe("-:testprobe");
+	if (WARN_ON_ONCE(ret)) {
+		pr_warning("error on deleting a probe.\n");
+		warn++;
+	}
 
-	pr_cont("OK\n");
+	ret = command_trace_probe("-:testprobe2");
+	if (WARN_ON_ONCE(ret)) {
+		pr_warning("error on deleting a probe.\n");
+		warn++;
+	}
+
+end:
+	cleanup_all_probes();
+	if (warn)
+		pr_cont("NG: Some tests are failed. Please check them.\n");
+	else
+		pr_cont("OK\n");
 	return 0;
 }
 
diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c
index 678a5120ee30..f4bc9b27de5f 100644
--- a/kernel/trace/trace_stack.c
+++ b/kernel/trace/trace_stack.c
@@ -157,6 +157,7 @@ stack_max_size_write(struct file *filp, const char __user *ubuf,
 	unsigned long val, flags;
 	char buf[64];
 	int ret;
+	int cpu;
 
 	if (count >= sizeof(buf))
 		return -EINVAL;
@@ -171,9 +172,20 @@ stack_max_size_write(struct file *filp, const char __user *ubuf,
 		return ret;
 
 	local_irq_save(flags);
+
+	/*
+	 * In case we trace inside arch_spin_lock() or after (NMI),
+	 * we will cause circular lock, so we also need to increase
+	 * the percpu trace_active here.
+	 */
+	cpu = smp_processor_id();
+	per_cpu(trace_active, cpu)++;
+
 	arch_spin_lock(&max_stack_lock);
 	*ptr = val;
 	arch_spin_unlock(&max_stack_lock);
+
+	per_cpu(trace_active, cpu)--;
 	local_irq_restore(flags);
 
 	return count;
@@ -206,7 +218,13 @@ t_next(struct seq_file *m, void *v, loff_t *pos)
 
 static void *t_start(struct seq_file *m, loff_t *pos)
 {
+	int cpu;
+
 	local_irq_disable();
+
+	cpu = smp_processor_id();
+	per_cpu(trace_active, cpu)++;
+
 	arch_spin_lock(&max_stack_lock);
 
 	if (*pos == 0)
@@ -217,7 +235,13 @@ static void *t_start(struct seq_file *m, loff_t *pos)
 
 static void t_stop(struct seq_file *m, void *p)
 {
+	int cpu;
+
 	arch_spin_unlock(&max_stack_lock);
+
+	cpu = smp_processor_id();
+	per_cpu(trace_active, cpu)--;
+
 	local_irq_enable();
 }
 
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 75289f372dd2..6cce6a808949 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -143,70 +143,65 @@ extern char *__bad_type_size(void);
 		#type, #name, offsetof(typeof(trace), name),		\
 		sizeof(trace.name), is_signed_type(type)
 
-int syscall_enter_format(struct ftrace_event_call *call, struct trace_seq *s)
+static
+int  __set_enter_print_fmt(struct syscall_metadata *entry, char *buf, int len)
 {
 	int i;
-	int ret;
-	struct syscall_metadata *entry = call->data;
-	struct syscall_trace_enter trace;
-	int offset = offsetof(struct syscall_trace_enter, args);
+	int pos = 0;
 
-	ret = trace_seq_printf(s, "\tfield:%s %s;\toffset:%zu;\tsize:%zu;"
-			       "\tsigned:%u;\n",
-			       SYSCALL_FIELD(int, nr));
-	if (!ret)
-		return 0;
+	/* When len=0, we just calculate the needed length */
+#define LEN_OR_ZERO (len ? len - pos : 0)
 
+	pos += snprintf(buf + pos, LEN_OR_ZERO, "\"");
 	for (i = 0; i < entry->nb_args; i++) {
-		ret = trace_seq_printf(s, "\tfield:%s %s;", entry->types[i],
-				        entry->args[i]);
-		if (!ret)
-			return 0;
-		ret = trace_seq_printf(s, "\toffset:%d;\tsize:%zu;"
-				       "\tsigned:%u;\n", offset,
-				       sizeof(unsigned long),
-				       is_signed_type(unsigned long));
-		if (!ret)
-			return 0;
-		offset += sizeof(unsigned long);
+		pos += snprintf(buf + pos, LEN_OR_ZERO, "%s: 0x%%0%zulx%s",
+				entry->args[i], sizeof(unsigned long),
+				i == entry->nb_args - 1 ? "" : ", ");
 	}
+	pos += snprintf(buf + pos, LEN_OR_ZERO, "\"");
 
-	trace_seq_puts(s, "\nprint fmt: \"");
 	for (i = 0; i < entry->nb_args; i++) {
-		ret = trace_seq_printf(s, "%s: 0x%%0%zulx%s", entry->args[i],
-				        sizeof(unsigned long),
-					i == entry->nb_args - 1 ? "" : ", ");
-		if (!ret)
-			return 0;
+		pos += snprintf(buf + pos, LEN_OR_ZERO,
+				", ((unsigned long)(REC->%s))", entry->args[i]);
 	}
-	trace_seq_putc(s, '"');
 
-	for (i = 0; i < entry->nb_args; i++) {
-		ret = trace_seq_printf(s, ", ((unsigned long)(REC->%s))",
-				       entry->args[i]);
-		if (!ret)
-			return 0;
-	}
+#undef LEN_OR_ZERO
 
-	return trace_seq_putc(s, '\n');
+	/* return the length of print_fmt */
+	return pos;
 }
 
-int syscall_exit_format(struct ftrace_event_call *call, struct trace_seq *s)
+static int set_syscall_print_fmt(struct ftrace_event_call *call)
 {
-	int ret;
-	struct syscall_trace_exit trace;
+	char *print_fmt;
+	int len;
+	struct syscall_metadata *entry = call->data;
 
-	ret = trace_seq_printf(s,
-			       "\tfield:%s %s;\toffset:%zu;\tsize:%zu;"
-			       "\tsigned:%u;\n"
-			       "\tfield:%s %s;\toffset:%zu;\tsize:%zu;"
-			       "\tsigned:%u;\n",
-			       SYSCALL_FIELD(int, nr),
-			       SYSCALL_FIELD(long, ret));
-	if (!ret)
+	if (entry->enter_event != call) {
+		call->print_fmt = "\"0x%lx\", REC->ret";
 		return 0;
+	}
+
+	/* First: called with 0 length to calculate the needed length */
+	len = __set_enter_print_fmt(entry, NULL, 0);
+
+	print_fmt = kmalloc(len + 1, GFP_KERNEL);
+	if (!print_fmt)
+		return -ENOMEM;
+
+	/* Second: actually write the @print_fmt */
+	__set_enter_print_fmt(entry, print_fmt, len + 1);
+	call->print_fmt = print_fmt;
 
-	return trace_seq_printf(s, "\nprint fmt: \"0x%%lx\", REC->ret\n");
+	return 0;
+}
+
+static void free_syscall_print_fmt(struct ftrace_event_call *call)
+{
+	struct syscall_metadata *entry = call->data;
+
+	if (entry->enter_event == call)
+		kfree(call->print_fmt);
 }
 
 int syscall_enter_define_fields(struct ftrace_event_call *call)
@@ -386,12 +381,17 @@ int init_syscall_trace(struct ftrace_event_call *call)
 {
 	int id;
 
-	id = register_ftrace_event(call->event);
-	if (!id)
-		return -ENODEV;
-	call->id = id;
-	INIT_LIST_HEAD(&call->fields);
-	return 0;
+	if (set_syscall_print_fmt(call) < 0)
+		return -ENOMEM;
+
+	id = trace_event_raw_init(call);
+
+	if (id < 0) {
+		free_syscall_print_fmt(call);
+		return id;
+	}
+
+	return id;
 }
 
 int __init init_ftrace_syscalls(void)
@@ -421,7 +421,7 @@ int __init init_ftrace_syscalls(void)
 }
 core_initcall(init_ftrace_syscalls);
 
-#ifdef CONFIG_EVENT_PROFILE
+#ifdef CONFIG_PERF_EVENTS
 
 static DECLARE_BITMAP(enabled_prof_enter_syscalls, NR_syscalls);
 static DECLARE_BITMAP(enabled_prof_exit_syscalls, NR_syscalls);
@@ -433,12 +433,9 @@ static void prof_syscall_enter(struct pt_regs *regs, long id)
 	struct syscall_metadata *sys_data;
 	struct syscall_trace_enter *rec;
 	unsigned long flags;
-	char *trace_buf;
-	char *raw_data;
 	int syscall_nr;
 	int rctx;
 	int size;
-	int cpu;
 
 	syscall_nr = syscall_get_nr(current, regs);
 	if (!test_bit(syscall_nr, enabled_prof_enter_syscalls))
@@ -457,37 +454,15 @@ static void prof_syscall_enter(struct pt_regs *regs, long id)
 		      "profile buffer not large enough"))
 		return;
 
-	/* Protect the per cpu buffer, begin the rcu read side */
-	local_irq_save(flags);
-
-	rctx = perf_swevent_get_recursion_context();
-	if (rctx < 0)
-		goto end_recursion;
-
-	cpu = smp_processor_id();
-
-	trace_buf = rcu_dereference(perf_trace_buf);
-
-	if (!trace_buf)
-		goto end;
-
-	raw_data = per_cpu_ptr(trace_buf, cpu);
-
-	/* zero the dead bytes from align to not leak stack to user */
-	*(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL;
+	rec = (struct syscall_trace_enter *)ftrace_perf_buf_prepare(size,
+				sys_data->enter_event->id, &rctx, &flags);
+	if (!rec)
+		return;
 
-	rec = (struct syscall_trace_enter *) raw_data;
-	tracing_generic_entry_update(&rec->ent, 0, 0);
-	rec->ent.type = sys_data->enter_event->id;
 	rec->nr = syscall_nr;
 	syscall_get_arguments(current, regs, 0, sys_data->nb_args,
 			       (unsigned long *)&rec->args);
-	perf_tp_event(sys_data->enter_event->id, 0, 1, rec, size);
-
-end:
-	perf_swevent_put_recursion_context(rctx);
-end_recursion:
-	local_irq_restore(flags);
+	ftrace_perf_buf_submit(rec, size, rctx, 0, 1, flags);
 }
 
 int prof_sysenter_enable(struct ftrace_event_call *call)
@@ -531,11 +506,8 @@ static void prof_syscall_exit(struct pt_regs *regs, long ret)
 	struct syscall_trace_exit *rec;
 	unsigned long flags;
 	int syscall_nr;
-	char *trace_buf;
-	char *raw_data;
 	int rctx;
 	int size;
-	int cpu;
 
 	syscall_nr = syscall_get_nr(current, regs);
 	if (!test_bit(syscall_nr, enabled_prof_exit_syscalls))
@@ -557,38 +529,15 @@ static void prof_syscall_exit(struct pt_regs *regs, long ret)
 		"exit event has grown above profile buffer size"))
 		return;
 
-	/* Protect the per cpu buffer, begin the rcu read side */
-	local_irq_save(flags);
-
-	rctx = perf_swevent_get_recursion_context();
-	if (rctx < 0)
-		goto end_recursion;
-
-	cpu = smp_processor_id();
-
-	trace_buf = rcu_dereference(perf_trace_buf);
-
-	if (!trace_buf)
-		goto end;
-
-	raw_data = per_cpu_ptr(trace_buf, cpu);
-
-	/* zero the dead bytes from align to not leak stack to user */
-	*(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL;
-
-	rec = (struct syscall_trace_exit *)raw_data;
+	rec = (struct syscall_trace_exit *)ftrace_perf_buf_prepare(size,
+				sys_data->exit_event->id, &rctx, &flags);
+	if (!rec)
+		return;
 
-	tracing_generic_entry_update(&rec->ent, 0, 0);
-	rec->ent.type = sys_data->exit_event->id;
 	rec->nr = syscall_nr;
 	rec->ret = syscall_get_return_value(current, regs);
 
-	perf_tp_event(sys_data->exit_event->id, 0, 1, rec, size);
-
-end:
-	perf_swevent_put_recursion_context(rctx);
-end_recursion:
-	local_irq_restore(flags);
+	ftrace_perf_buf_submit(rec, size, rctx, 0, 1, flags);
 }
 
 int prof_sysexit_enable(struct ftrace_event_call *call)
@@ -626,6 +575,5 @@ void prof_sysexit_disable(struct ftrace_event_call *call)
 	mutex_unlock(&syscall_trace_lock);
 }
 
-#endif
-
+#endif /* CONFIG_PERF_EVENTS */
 
diff --git a/kernel/user.c b/kernel/user.c
index 46d0165ca70c..766467b3bcb7 100644
--- a/kernel/user.c
+++ b/kernel/user.c
@@ -56,9 +56,6 @@ struct user_struct root_user = {
 	.sigpending	= ATOMIC_INIT(0),
 	.locked_shm     = 0,
 	.user_ns	= &init_user_ns,
-#ifdef CONFIG_USER_SCHED
-	.tg		= &init_task_group,
-#endif
 };
 
 /*
@@ -75,268 +72,6 @@ static void uid_hash_remove(struct user_struct *up)
 	put_user_ns(up->user_ns);
 }
 
-#ifdef CONFIG_USER_SCHED
-
-static void sched_destroy_user(struct user_struct *up)
-{
-	sched_destroy_group(up->tg);
-}
-
-static int sched_create_user(struct user_struct *up)
-{
-	int rc = 0;
-
-	up->tg = sched_create_group(&root_task_group);
-	if (IS_ERR(up->tg))
-		rc = -ENOMEM;
-
-	set_tg_uid(up);
-
-	return rc;
-}
-
-#else	/* CONFIG_USER_SCHED */
-
-static void sched_destroy_user(struct user_struct *up) { }
-static int sched_create_user(struct user_struct *up) { return 0; }
-
-#endif	/* CONFIG_USER_SCHED */
-
-#if defined(CONFIG_USER_SCHED) && defined(CONFIG_SYSFS)
-
-static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)
-{
-	struct user_struct *user;
-	struct hlist_node *h;
-
-	hlist_for_each_entry(user, h, hashent, uidhash_node) {
-		if (user->uid == uid) {
-			/* possibly resurrect an "almost deleted" object */
-			if (atomic_inc_return(&user->__count) == 1)
-				cancel_delayed_work(&user->work);
-			return user;
-		}
-	}
-
-	return NULL;
-}
-
-static struct kset *uids_kset; /* represents the /sys/kernel/uids/ directory */
-static DEFINE_MUTEX(uids_mutex);
-
-static inline void uids_mutex_lock(void)
-{
-	mutex_lock(&uids_mutex);
-}
-
-static inline void uids_mutex_unlock(void)
-{
-	mutex_unlock(&uids_mutex);
-}
-
-/* uid directory attributes */
-#ifdef CONFIG_FAIR_GROUP_SCHED
-static ssize_t cpu_shares_show(struct kobject *kobj,
-			       struct kobj_attribute *attr,
-			       char *buf)
-{
-	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-
-	return sprintf(buf, "%lu\n", sched_group_shares(up->tg));
-}
-
-static ssize_t cpu_shares_store(struct kobject *kobj,
-				struct kobj_attribute *attr,
-				const char *buf, size_t size)
-{
-	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-	unsigned long shares;
-	int rc;
-
-	sscanf(buf, "%lu", &shares);
-
-	rc = sched_group_set_shares(up->tg, shares);
-
-	return (rc ? rc : size);
-}
-
-static struct kobj_attribute cpu_share_attr =
-	__ATTR(cpu_share, 0644, cpu_shares_show, cpu_shares_store);
-#endif
-
-#ifdef CONFIG_RT_GROUP_SCHED
-static ssize_t cpu_rt_runtime_show(struct kobject *kobj,
-				   struct kobj_attribute *attr,
-				   char *buf)
-{
-	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-
-	return sprintf(buf, "%ld\n", sched_group_rt_runtime(up->tg));
-}
-
-static ssize_t cpu_rt_runtime_store(struct kobject *kobj,
-				    struct kobj_attribute *attr,
-				    const char *buf, size_t size)
-{
-	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-	unsigned long rt_runtime;
-	int rc;
-
-	sscanf(buf, "%ld", &rt_runtime);
-
-	rc = sched_group_set_rt_runtime(up->tg, rt_runtime);
-
-	return (rc ? rc : size);
-}
-
-static struct kobj_attribute cpu_rt_runtime_attr =
-	__ATTR(cpu_rt_runtime, 0644, cpu_rt_runtime_show, cpu_rt_runtime_store);
-
-static ssize_t cpu_rt_period_show(struct kobject *kobj,
-				   struct kobj_attribute *attr,
-				   char *buf)
-{
-	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-
-	return sprintf(buf, "%lu\n", sched_group_rt_period(up->tg));
-}
-
-static ssize_t cpu_rt_period_store(struct kobject *kobj,
-				    struct kobj_attribute *attr,
-				    const char *buf, size_t size)
-{
-	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-	unsigned long rt_period;
-	int rc;
-
-	sscanf(buf, "%lu", &rt_period);
-
-	rc = sched_group_set_rt_period(up->tg, rt_period);
-
-	return (rc ? rc : size);
-}
-
-static struct kobj_attribute cpu_rt_period_attr =
-	__ATTR(cpu_rt_period, 0644, cpu_rt_period_show, cpu_rt_period_store);
-#endif
-
-/* default attributes per uid directory */
-static struct attribute *uids_attributes[] = {
-#ifdef CONFIG_FAIR_GROUP_SCHED
-	&cpu_share_attr.attr,
-#endif
-#ifdef CONFIG_RT_GROUP_SCHED
-	&cpu_rt_runtime_attr.attr,
-	&cpu_rt_period_attr.attr,
-#endif
-	NULL
-};
-
-/* the lifetime of user_struct is not managed by the core (now) */
-static void uids_release(struct kobject *kobj)
-{
-	return;
-}
-
-static struct kobj_type uids_ktype = {
-	.sysfs_ops = &kobj_sysfs_ops,
-	.default_attrs = uids_attributes,
-	.release = uids_release,
-};
-
-/*
- * Create /sys/kernel/uids/<uid>/cpu_share file for this user
- * We do not create this file for users in a user namespace (until
- * sysfs tagging is implemented).
- *
- * See Documentation/scheduler/sched-design-CFS.txt for ramifications.
- */
-static int uids_user_create(struct user_struct *up)
-{
-	struct kobject *kobj = &up->kobj;
-	int error;
-
-	memset(kobj, 0, sizeof(struct kobject));
-	if (up->user_ns != &init_user_ns)
-		return 0;
-	kobj->kset = uids_kset;
-	error = kobject_init_and_add(kobj, &uids_ktype, NULL, "%d", up->uid);
-	if (error) {
-		kobject_put(kobj);
-		goto done;
-	}
-
-	kobject_uevent(kobj, KOBJ_ADD);
-done:
-	return error;
-}
-
-/* create these entries in sysfs:
- * 	"/sys/kernel/uids" directory
- * 	"/sys/kernel/uids/0" directory (for root user)
- * 	"/sys/kernel/uids/0/cpu_share" file (for root user)
- */
-int __init uids_sysfs_init(void)
-{
-	uids_kset = kset_create_and_add("uids", NULL, kernel_kobj);
-	if (!uids_kset)
-		return -ENOMEM;
-
-	return uids_user_create(&root_user);
-}
-
-/* delayed work function to remove sysfs directory for a user and free up
- * corresponding structures.
- */
-static void cleanup_user_struct(struct work_struct *w)
-{
-	struct user_struct *up = container_of(w, struct user_struct, work.work);
-	unsigned long flags;
-	int remove_user = 0;
-
-	/* Make uid_hash_remove() + sysfs_remove_file() + kobject_del()
-	 * atomic.
-	 */
-	uids_mutex_lock();
-
-	spin_lock_irqsave(&uidhash_lock, flags);
-	if (atomic_read(&up->__count) == 0) {
-		uid_hash_remove(up);
-		remove_user = 1;
-	}
-	spin_unlock_irqrestore(&uidhash_lock, flags);
-
-	if (!remove_user)
-		goto done;
-
-	if (up->user_ns == &init_user_ns) {
-		kobject_uevent(&up->kobj, KOBJ_REMOVE);
-		kobject_del(&up->kobj);
-		kobject_put(&up->kobj);
-	}
-
-	sched_destroy_user(up);
-	key_put(up->uid_keyring);
-	key_put(up->session_keyring);
-	kmem_cache_free(uid_cachep, up);
-
-done:
-	uids_mutex_unlock();
-}
-
-/* IRQs are disabled and uidhash_lock is held upon function entry.
- * IRQ state (as stored in flags) is restored and uidhash_lock released
- * upon function exit.
- */
-static void free_user(struct user_struct *up, unsigned long flags)
-{
-	INIT_DELAYED_WORK(&up->work, cleanup_user_struct);
-	schedule_delayed_work(&up->work, msecs_to_jiffies(1000));
-	spin_unlock_irqrestore(&uidhash_lock, flags);
-}
-
-#else	/* CONFIG_USER_SCHED && CONFIG_SYSFS */
-
 static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)
 {
 	struct user_struct *user;
@@ -352,11 +87,6 @@ static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)
 	return NULL;
 }
 
-int uids_sysfs_init(void) { return 0; }
-static inline int uids_user_create(struct user_struct *up) { return 0; }
-static inline void uids_mutex_lock(void) { }
-static inline void uids_mutex_unlock(void) { }
-
 /* IRQs are disabled and uidhash_lock is held upon function entry.
  * IRQ state (as stored in flags) is restored and uidhash_lock released
  * upon function exit.
@@ -365,32 +95,11 @@ static void free_user(struct user_struct *up, unsigned long flags)
 {
 	uid_hash_remove(up);
 	spin_unlock_irqrestore(&uidhash_lock, flags);
-	sched_destroy_user(up);
 	key_put(up->uid_keyring);
 	key_put(up->session_keyring);
 	kmem_cache_free(uid_cachep, up);
 }
 
-#endif
-
-#if defined(CONFIG_RT_GROUP_SCHED) && defined(CONFIG_USER_SCHED)
-/*
- * We need to check if a setuid can take place. This function should be called
- * before successfully completing the setuid.
- */
-int task_can_switch_user(struct user_struct *up, struct task_struct *tsk)
-{
-
-	return sched_rt_can_attach(up->tg, tsk);
-
-}
-#else
-int task_can_switch_user(struct user_struct *up, struct task_struct *tsk)
-{
-	return 1;
-}
-#endif
-
 /*
  * Locate the user_struct for the passed UID.  If found, take a ref on it.  The
  * caller must undo that ref with free_uid().
@@ -431,8 +140,6 @@ struct user_struct *alloc_uid(struct user_namespace *ns, uid_t uid)
 	/* Make uid_hash_find() + uids_user_create() + uid_hash_insert()
 	 * atomic.
 	 */
-	uids_mutex_lock();
-
 	spin_lock_irq(&uidhash_lock);
 	up = uid_hash_find(uid, hashent);
 	spin_unlock_irq(&uidhash_lock);
@@ -445,14 +152,8 @@ struct user_struct *alloc_uid(struct user_namespace *ns, uid_t uid)
 		new->uid = uid;
 		atomic_set(&new->__count, 1);
 
-		if (sched_create_user(new) < 0)
-			goto out_free_user;
-
 		new->user_ns = get_user_ns(ns);
 
-		if (uids_user_create(new))
-			goto out_destoy_sched;
-
 		/*
 		 * Before adding this, check whether we raced
 		 * on adding the same user already..
@@ -475,17 +176,11 @@ struct user_struct *alloc_uid(struct user_namespace *ns, uid_t uid)
 		spin_unlock_irq(&uidhash_lock);
 	}
 
-	uids_mutex_unlock();
-
 	return up;
 
-out_destoy_sched:
-	sched_destroy_user(new);
 	put_user_ns(new->user_ns);
-out_free_user:
 	kmem_cache_free(uid_cachep, new);
 out_unlock:
-	uids_mutex_unlock();
 	return NULL;
 }
 
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index d62e3cdab357..6cf8638639b9 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -765,9 +765,9 @@ config RCU_CPU_STALL_DETECTOR
 	  CPUs are delaying the current grace period, but only when
 	  the grace period extends for excessive time periods.
 
-	  Say Y if you want RCU to perform such checks.
+	  Say N if you want to disable such checks.
 
-	  Say N if you are unsure.
+	  Say Y if you are unsure.
 
 config KPROBES_SANITY_TEST
 	bool "Kprobes sanity tests"
diff --git a/lib/hweight.c b/lib/hweight.c
index 389424ecb129..63ee4eb1228d 100644
--- a/lib/hweight.c
+++ b/lib/hweight.c
@@ -11,11 +11,18 @@
 
 unsigned int hweight32(unsigned int w)
 {
+#ifdef ARCH_HAS_FAST_MULTIPLIER
+	w -= (w >> 1) & 0x55555555;
+	w =  (w & 0x33333333) + ((w >> 2) & 0x33333333);
+	w =  (w + (w >> 4)) & 0x0f0f0f0f;
+	return (w * 0x01010101) >> 24;
+#else
 	unsigned int res = w - ((w >> 1) & 0x55555555);
 	res = (res & 0x33333333) + ((res >> 2) & 0x33333333);
 	res = (res + (res >> 4)) & 0x0F0F0F0F;
 	res = res + (res >> 8);
 	return (res + (res >> 16)) & 0x000000FF;
+#endif
 }
 EXPORT_SYMBOL(hweight32);
 
diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
index ea6f6e3adaea..f3c9c0a90b98 100755
--- a/scripts/recordmcount.pl
+++ b/scripts/recordmcount.pl
@@ -136,13 +136,14 @@ my %text_sections = (
      ".text.unlikely" => 1,
 );
 
-$objdump = "objdump" if ((length $objdump) == 0);
-$objcopy = "objcopy" if ((length $objcopy) == 0);
-$cc = "gcc" if ((length $cc) == 0);
-$ld = "ld" if ((length $ld) == 0);
-$nm = "nm" if ((length $nm) == 0);
-$rm = "rm" if ((length $rm) == 0);
-$mv = "mv" if ((length $mv) == 0);
+# Note: we are nice to C-programmers here, thus we skip the '||='-idiom.
+$objdump = 'objdump' if (!$objdump);
+$objcopy = 'objcopy' if (!$objcopy);
+$cc = 'gcc' if (!$cc);
+$ld = 'ld' if (!$ld);
+$nm = 'nm' if (!$nm);
+$rm = 'rm' if (!$rm);
+$mv = 'mv' if (!$mv);
 
 #print STDERR "running: $P '$arch' '$objdump' '$objcopy' '$cc' '$ld' " .
 #    "'$nm' '$rm' '$mv' '$inputfile'\n";
@@ -432,14 +433,14 @@ sub update_funcs
 
     # Loop through all the mcount caller offsets and print a reference
     # to the caller based from the ref_func.
-    for (my $i=0; $i <= $#offsets; $i++) {
-	if (!$opened) {
-	    open(FILE, ">$mcount_s") || die "can't create $mcount_s\n";
-	    $opened = 1;
-	    print FILE "\t.section $mcount_section,\"a\",$section_type\n";
-	    print FILE "\t.align $alignment\n" if (defined($alignment));
-	}
-	printf FILE "\t%s %s + %d\n", $type, $ref_func, $offsets[$i] - $offset;
+    if (!$opened) {
+	open(FILE, ">$mcount_s") || die "can't create $mcount_s\n";
+	$opened = 1;
+	print FILE "\t.section $mcount_section,\"a\",$section_type\n";
+	print FILE "\t.align $alignment\n" if (defined($alignment));
+    }
+    foreach my $cur_offset (@offsets) {
+	printf FILE "\t%s %s + %d\n", $type, $ref_func, $cur_offset - $offset;
     }
 }
 
@@ -476,11 +477,7 @@ while (<IN>) {
 	$read_headers = 0;
 
 	# Only record text sections that we know are safe
-	if (defined($text_sections{$1})) {
-	    $read_function = 1;
-	} else {
-	    $read_function = 0;
-	}
+	$read_function = defined($text_sections{$1});
 	# print out any recorded offsets
 	update_funcs();
 
@@ -514,7 +511,7 @@ while (<IN>) {
     }
     # is this a call site to mcount? If so, record it to print later
     if ($text_found && /$mcount_regex/) {
-	$offsets[$#offsets + 1] = hex $1;
+	push(@offsets, hex $1);
     }
 }
 
diff --git a/tools/perf/.gitignore b/tools/perf/.gitignore
index 124760bb37b5..e1d60d780784 100644
--- a/tools/perf/.gitignore
+++ b/tools/perf/.gitignore
@@ -14,6 +14,7 @@ perf*.html
 common-cmds.h
 perf.data
 perf.data.old
+perf-archive
 tags
 TAGS
 cscope*
diff --git a/tools/perf/Documentation/perf-archive.txt b/tools/perf/Documentation/perf-archive.txt
new file mode 100644
index 000000000000..fae174dc7d01
--- /dev/null
+++ b/tools/perf/Documentation/perf-archive.txt
@@ -0,0 +1,22 @@
+perf-archive(1)
+===============
+
+NAME
+----
+perf-archive - Create archive with object files with build-ids found in perf.data file
+
+SYNOPSIS
+--------
+[verse]
+'perf archive' [file]
+
+DESCRIPTION
+-----------
+This command runs runs perf-buildid-list --with-hits, and collects the files
+with the buildids found so that analisys of perf.data contents can be possible
+on another machine.
+
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-buildid-list[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-buildid-cache.txt b/tools/perf/Documentation/perf-buildid-cache.txt
new file mode 100644
index 000000000000..88bc3b519746
--- /dev/null
+++ b/tools/perf/Documentation/perf-buildid-cache.txt
@@ -0,0 +1,33 @@
+perf-buildid-cache(1)
+=====================
+
+NAME
+----
+perf-buildid-cache - Manage build-id cache.
+
+SYNOPSIS
+--------
+[verse]
+'perf buildid-list <options>'
+
+DESCRIPTION
+-----------
+This command manages the build-id cache. It can add and remove files to the
+cache. In the future it should as well purge older entries, set upper limits
+for the space used by the cache, etc.
+
+OPTIONS
+-------
+-a::
+--add=::
+        Add specified file to the cache.
+-r::
+--remove=::
+        Remove specified file to the cache.
+-v::
+--verbose::
+	Be more verbose.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt
index 250e391b4bc8..2de34075f6a4 100644
--- a/tools/perf/Documentation/perf-probe.txt
+++ b/tools/perf/Documentation/perf-probe.txt
@@ -15,6 +15,8 @@ or
 'perf probe' [options] --del='[GROUP:]EVENT' [...]
 or
 'perf probe' --list
+or
+'perf probe' --line='FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]'
 
 DESCRIPTION
 -----------
@@ -45,6 +47,11 @@ OPTIONS
 --list::
 	List up current probe events.
 
+-L::
+--line=::
+	Show source code lines which can be probed. This needs an argument
+	which specifies a range of the source code.
+
 PROBE SYNTAX
 ------------
 Probe points are defined by following syntax.
@@ -56,6 +63,19 @@ Probe points are defined by following syntax.
 It is also possible to specify a probe point by the source line number by using 'SRC:ALN' syntax, where 'SRC' is the source file path and 'ALN' is the line number.
 'ARG' specifies the arguments of this probe point. You can use the name of local variable, or kprobe-tracer argument format (e.g. $retval, %ax, etc).
 
+LINE SYNTAX
+-----------
+Line range is descripted by following syntax.
+
+ "FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]"
+
+FUNC specifies the function name of showing lines. 'RLN' is the start line
+number from function entry line, and 'RLN2' is the end line number. As same as
+probe syntax, 'SRC' means the source file path, 'ALN' is start line number,
+and 'ALN2' is end line number in the file. It is also possible to specify how
+many lines to show by using 'NUM'.
+So, "source.c:100-120" shows lines between 100th to l20th in source.c file. And "func:10+20" shows 20 lines from 10th line of func function.
+
 SEE ALSO
 --------
 linkperf:perf-trace[1], linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf.txt b/tools/perf/Documentation/perf.txt
index 69c832557199..0eeb247dc7d2 100644
--- a/tools/perf/Documentation/perf.txt
+++ b/tools/perf/Documentation/perf.txt
@@ -12,7 +12,7 @@ SYNOPSIS
 
 DESCRIPTION
 -----------
-Performance counters for Linux are are a new kernel-based subsystem
+Performance counters for Linux are a new kernel-based subsystem
 that provide a framework for all things performance analysis. It
 covers hardware level (CPU/PMU, Performance Monitoring Unit) features
 and software features (software counters, tracepoints) as well.
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 2e7fa3a06806..42969303e20b 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -286,11 +286,7 @@ SCRIPT_PERL =
 SCRIPT_SH =
 TEST_PROGRAMS =
 
-#
-# No scripts right now:
-#
-
-# SCRIPT_SH += perf-am.sh
+SCRIPT_SH += perf-archive.sh
 
 #
 # No Perl scripts right now:
@@ -315,9 +311,6 @@ PROGRAMS += perf
 # List built-in command $C whose implementation cmd_$C() is not in
 # builtin-$C.o but is linked in as part of some other command.
 #
-# None right now:
-#
-# BUILT_INS += perf-init $X
 
 # what 'all' will build and 'install' will install, in perfexecdir
 ALL_PROGRAMS = $(PROGRAMS) $(SCRIPTS)
@@ -340,6 +333,7 @@ LIB_FILE=libperf.a
 LIB_H += ../../include/linux/perf_event.h
 LIB_H += ../../include/linux/rbtree.h
 LIB_H += ../../include/linux/list.h
+LIB_H += ../../include/linux/hash.h
 LIB_H += ../../include/linux/stringify.h
 LIB_H += util/include/linux/bitmap.h
 LIB_H += util/include/linux/bitops.h
@@ -369,6 +363,7 @@ LIB_H += util/event.h
 LIB_H += util/exec_cmd.h
 LIB_H += util/types.h
 LIB_H += util/levenshtein.h
+LIB_H += util/map.h
 LIB_H += util/parse-options.h
 LIB_H += util/parse-events.h
 LIB_H += util/quote.h
@@ -435,8 +430,8 @@ LIB_OBJS += util/trace-event-perl.o
 LIB_OBJS += util/svghelper.o
 LIB_OBJS += util/sort.o
 LIB_OBJS += util/hist.o
-LIB_OBJS += util/data_map.o
 LIB_OBJS += util/probe-event.o
+LIB_OBJS += util/util.o
 
 BUILTIN_OBJS += builtin-annotate.o
 
@@ -451,6 +446,7 @@ BUILTIN_OBJS += builtin-diff.o
 BUILTIN_OBJS += builtin-help.o
 BUILTIN_OBJS += builtin-sched.o
 BUILTIN_OBJS += builtin-buildid-list.o
+BUILTIN_OBJS += builtin-buildid-cache.o
 BUILTIN_OBJS += builtin-list.o
 BUILTIN_OBJS += builtin-record.o
 BUILTIN_OBJS += builtin-report.o
@@ -460,6 +456,7 @@ BUILTIN_OBJS += builtin-top.o
 BUILTIN_OBJS += builtin-trace.o
 BUILTIN_OBJS += builtin-probe.o
 BUILTIN_OBJS += builtin-kmem.o
+BUILTIN_OBJS += builtin-lock.o
 
 PERFLIBS = $(LIB_FILE)
 
@@ -1009,6 +1006,7 @@ install: all
 	$(INSTALL) perf$X '$(DESTDIR_SQ)$(bindir_SQ)'
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
+	$(INSTALL) perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
 	$(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) scripts/perl/*.pl -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl'
 	$(INSTALL) scripts/perl/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 593ff25006de..73c202ee0882 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -132,8 +132,8 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 {
 	struct addr_location al;
 
-	dump_printf("(IP, %d): %d: %p\n", event->header.misc,
-		    event->ip.pid, (void *)(long)event->ip.ip);
+	dump_printf("(IP, %d): %d: %#Lx\n", event->header.misc,
+		    event->ip.pid, event->ip.ip);
 
 	if (event__preprocess_sample(event, session, &al, symbol_filter) < 0) {
 		fprintf(stderr, "problem processing %d event, skipping it.\n",
@@ -451,10 +451,10 @@ static void perf_session__find_annotations(struct perf_session *self)
 }
 
 static struct perf_event_ops event_ops = {
-	.process_sample_event	= process_sample_event,
-	.process_mmap_event	= event__process_mmap,
-	.process_comm_event	= event__process_comm,
-	.process_fork_event	= event__process_task,
+	.sample	= process_sample_event,
+	.mmap	= event__process_mmap,
+	.comm	= event__process_comm,
+	.fork	= event__process_task,
 };
 
 static int __cmd_annotate(void)
diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c
new file mode 100644
index 000000000000..30a05f552c96
--- /dev/null
+++ b/tools/perf/builtin-buildid-cache.c
@@ -0,0 +1,133 @@
+/*
+ * builtin-buildid-cache.c
+ *
+ * Builtin buildid-cache command: Manages build-id cache
+ *
+ * Copyright (C) 2010, Red Hat Inc.
+ * Copyright (C) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "builtin.h"
+#include "perf.h"
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/strlist.h"
+#include "util/symbol.h"
+
+static char const *add_name_list_str, *remove_name_list_str;
+
+static const char * const buildid_cache_usage[] = {
+	"perf buildid-cache [<options>]",
+	NULL
+};
+
+static const struct option buildid_cache_options[] = {
+	OPT_STRING('a', "add", &add_name_list_str,
+		   "file list", "file(s) to add"),
+	OPT_STRING('r', "remove", &remove_name_list_str, "file list",
+		    "file(s) to remove"),
+	OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose"),
+	OPT_END()
+};
+
+static int build_id_cache__add_file(const char *filename, const char *debugdir)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+	u8 build_id[BUILD_ID_SIZE];
+	int err;
+
+	if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) {
+		pr_debug("Couldn't read a build-id in %s\n", filename);
+		return -1;
+	}
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+	err = build_id_cache__add_s(sbuild_id, debugdir, filename, false);
+	if (verbose)
+		pr_info("Adding %s %s: %s\n", sbuild_id, filename,
+			err ? "FAIL" : "Ok");
+	return err;
+}
+
+static int build_id_cache__remove_file(const char *filename __used,
+				       const char *debugdir __used)
+{
+	u8 build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	int err;
+
+	if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) {
+		pr_debug("Couldn't read a build-id in %s\n", filename);
+		return -1;
+	}
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+	err = build_id_cache__remove_s(sbuild_id, debugdir);
+	if (verbose)
+		pr_info("Removing %s %s: %s\n", sbuild_id, filename,
+			err ? "FAIL" : "Ok");
+
+	return err;
+}
+
+static int __cmd_buildid_cache(void)
+{
+	struct strlist *list;
+	struct str_node *pos;
+	char debugdir[PATH_MAX];
+
+	snprintf(debugdir, sizeof(debugdir), "%s/%s", getenv("HOME"),
+		 DEBUG_CACHE_DIR);
+
+	if (add_name_list_str) {
+		list = strlist__new(true, add_name_list_str);
+		if (list) {
+			strlist__for_each(pos, list)
+				if (build_id_cache__add_file(pos->s, debugdir)) {
+					if (errno == EEXIST) {
+						pr_debug("%s already in the cache\n",
+							 pos->s);
+						continue;
+					}
+					pr_warning("Couldn't add %s: %s\n",
+						   pos->s, strerror(errno));
+				}
+
+			strlist__delete(list);
+		}
+	}
+
+	if (remove_name_list_str) {
+		list = strlist__new(true, remove_name_list_str);
+		if (list) {
+			strlist__for_each(pos, list)
+				if (build_id_cache__remove_file(pos->s, debugdir)) {
+					if (errno == ENOENT) {
+						pr_debug("%s wasn't in the cache\n",
+							 pos->s);
+						continue;
+					}
+					pr_warning("Couldn't remove %s: %s\n",
+						   pos->s, strerror(errno));
+				}
+
+			strlist__delete(list);
+		}
+	}
+
+	return 0;
+}
+
+int cmd_buildid_cache(int argc, const char **argv, const char *prefix __used)
+{
+	argc = parse_options(argc, argv, buildid_cache_options,
+			     buildid_cache_usage, 0);
+
+	if (symbol__init() < 0)
+		return -1;
+
+	setup_pager();
+	return __cmd_buildid_cache();
+}
diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c
index 1e99ac806913..431f204bde64 100644
--- a/tools/perf/builtin-buildid-list.c
+++ b/tools/perf/builtin-buildid-list.c
@@ -16,6 +16,7 @@
 
 static char const *input_name = "perf.data";
 static int force;
+static bool with_hits;
 
 static const char * const buildid_list_usage[] = {
 	"perf buildid-list [<options>]",
@@ -23,6 +24,7 @@ static const char * const buildid_list_usage[] = {
 };
 
 static const struct option options[] = {
+	OPT_BOOLEAN('H', "with-hits", &with_hits, "Show only DSOs with hits"),
 	OPT_STRING('i', "input", &input_name, "file",
 		    "input file name"),
 	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
@@ -31,26 +33,34 @@ static const struct option options[] = {
 	OPT_END()
 };
 
-static int perf_file_section__process_buildids(struct perf_file_section *self,
-					       int feat, int fd)
+static int build_id_list__process_event(event_t *event,
+					struct perf_session *session)
 {
-	if (feat != HEADER_BUILD_ID)
-		return 0;
+	struct addr_location al;
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread = perf_session__findnew(session, event->ip.pid);
 
-	if (lseek(fd, self->offset, SEEK_SET) < 0) {
-		pr_warning("Failed to lseek to %Ld offset for buildids!\n",
-			   self->offset);
+	if (thread == NULL) {
+		pr_err("problem processing %d event, skipping it.\n",
+			event->header.type);
 		return -1;
 	}
 
-	if (perf_header__read_build_ids(fd, self->offset, self->size)) {
-		pr_warning("Failed to read buildids!\n");
-		return -1;
-	}
+	thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+			      event->ip.ip, &al);
+
+	if (al.map != NULL)
+		al.map->dso->hit = 1;
 
 	return 0;
 }
 
+static struct perf_event_ops build_id_list__event_ops = {
+	.sample	= build_id_list__process_event,
+	.mmap	= event__process_mmap,
+	.fork	= event__process_task,
+};
+
 static int __cmd_buildid_list(void)
 {
 	int err = -1;
@@ -60,10 +70,10 @@ static int __cmd_buildid_list(void)
 	if (session == NULL)
 		return -1;
 
-	err = perf_header__process_sections(&session->header, session->fd,
-				         perf_file_section__process_buildids);
-	if (err >= 0)
-		dsos__fprintf_buildid(stdout);
+	if (with_hits)
+		perf_session__process_events(session, &build_id_list__event_ops);
+
+	dsos__fprintf_buildid(stdout, with_hits);
 
 	perf_session__delete(session);
 	return err;
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
index bd71b8ceafb7..18b3f505f9db 100644
--- a/tools/perf/builtin-diff.c
+++ b/tools/perf/builtin-diff.c
@@ -42,8 +42,8 @@ static int diff__process_sample_event(event_t *event, struct perf_session *sessi
 	struct addr_location al;
 	struct sample_data data = { .period = 1, };
 
-	dump_printf("(IP, %d): %d: %p\n", event->header.misc,
-		    event->ip.pid, (void *)(long)event->ip.ip);
+	dump_printf("(IP, %d): %d: %#Lx\n", event->header.misc,
+		    event->ip.pid, event->ip.ip);
 
 	if (event__preprocess_sample(event, session, &al, NULL) < 0) {
 		pr_warning("problem processing %d event, skipping it.\n",
@@ -51,12 +51,12 @@ static int diff__process_sample_event(event_t *event, struct perf_session *sessi
 		return -1;
 	}
 
-	if (al.filtered)
+	if (al.filtered || al.sym == NULL)
 		return 0;
 
 	event__parse_sample(event, session->sample_type, &data);
 
-	if (al.sym && perf_session__add_hist_entry(session, &al, data.period)) {
+	if (perf_session__add_hist_entry(session, &al, data.period)) {
 		pr_warning("problem incrementing symbol count, skipping event\n");
 		return -1;
 	}
@@ -66,12 +66,12 @@ static int diff__process_sample_event(event_t *event, struct perf_session *sessi
 }
 
 static struct perf_event_ops event_ops = {
-	.process_sample_event = diff__process_sample_event,
-	.process_mmap_event   = event__process_mmap,
-	.process_comm_event   = event__process_comm,
-	.process_exit_event   = event__process_task,
-	.process_fork_event   = event__process_task,
-	.process_lost_event   = event__process_lost,
+	.sample	= diff__process_sample_event,
+	.mmap	= event__process_mmap,
+	.comm	= event__process_comm,
+	.exit	= event__process_task,
+	.fork	= event__process_task,
+	.lost	= event__process_lost,
 };
 
 static void perf_session__insert_hist_entry_by_name(struct rb_root *root,
@@ -82,29 +82,19 @@ static void perf_session__insert_hist_entry_by_name(struct rb_root *root,
 	struct hist_entry *iter;
 
 	while (*p != NULL) {
-		int cmp;
 		parent = *p;
 		iter = rb_entry(parent, struct hist_entry, rb_node);
-
-		cmp = strcmp(he->map->dso->name, iter->map->dso->name);
-		if (cmp > 0)
+		if (hist_entry__cmp(he, iter) < 0)
 			p = &(*p)->rb_left;
-		else if (cmp < 0)
+		else
 			p = &(*p)->rb_right;
-		else {
-			cmp = strcmp(he->sym->name, iter->sym->name);
-			if (cmp > 0)
-				p = &(*p)->rb_left;
-			else
-				p = &(*p)->rb_right;
-		}
 	}
 
 	rb_link_node(&he->rb_node, parent, p);
 	rb_insert_color(&he->rb_node, root);
 }
 
-static void perf_session__resort_by_name(struct perf_session *self)
+static void perf_session__resort_hist_entries(struct perf_session *self)
 {
 	unsigned long position = 1;
 	struct rb_root tmp = RB_ROOT;
@@ -122,29 +112,28 @@ static void perf_session__resort_by_name(struct perf_session *self)
 	self->hists = tmp;
 }
 
+static void perf_session__set_hist_entries_positions(struct perf_session *self)
+{
+	perf_session__output_resort(self, self->events_stats.total);
+	perf_session__resort_hist_entries(self);
+}
+
 static struct hist_entry *
-perf_session__find_hist_entry_by_name(struct perf_session *self,
-				      struct hist_entry *he)
+perf_session__find_hist_entry(struct perf_session *self,
+			      struct hist_entry *he)
 {
 	struct rb_node *n = self->hists.rb_node;
 
 	while (n) {
 		struct hist_entry *iter = rb_entry(n, struct hist_entry, rb_node);
-		int cmp = strcmp(he->map->dso->name, iter->map->dso->name);
+		int64_t cmp = hist_entry__cmp(he, iter);
 
-		if (cmp > 0)
+		if (cmp < 0)
 			n = n->rb_left;
-		else if (cmp < 0)
+		else if (cmp > 0)
 			n = n->rb_right;
-		else {
-			cmp = strcmp(he->sym->name, iter->sym->name);
-			if (cmp > 0)
-				n = n->rb_left;
-			else if (cmp < 0)
-				n = n->rb_right;
-			else
-				return iter;
-		}
+		else 
+			return iter;
 	}
 
 	return NULL;
@@ -155,11 +144,9 @@ static void perf_session__match_hists(struct perf_session *old_session,
 {
 	struct rb_node *nd;
 
-	perf_session__resort_by_name(old_session);
-
 	for (nd = rb_first(&new_session->hists); nd; nd = rb_next(nd)) {
 		struct hist_entry *pos = rb_entry(nd, struct hist_entry, rb_node);
-		pos->pair = perf_session__find_hist_entry_by_name(old_session, pos);
+		pos->pair = perf_session__find_hist_entry(old_session, pos);
 	}
 }
 
@@ -177,9 +164,12 @@ static int __cmd_diff(void)
 		ret = perf_session__process_events(session[i], &event_ops);
 		if (ret)
 			goto out_delete;
-		perf_session__output_resort(session[i], session[i]->events_stats.total);
 	}
 
+	perf_session__output_resort(session[1], session[1]->events_stats.total);
+	if (show_displacement)
+		perf_session__set_hist_entries_positions(session[0]);
+
 	perf_session__match_hists(session[0], session[1]);
 	perf_session__fprintf_hists(session[1], session[0],
 				    show_displacement, stdout);
@@ -204,7 +194,7 @@ static const struct option options[] = {
 	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
 	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
 		    "load module symbols - WARNING: use only with -k and LIVE kernel"),
-	OPT_BOOLEAN('P', "full-paths", &event_ops.full_paths,
+	OPT_BOOLEAN('P', "full-paths", &symbol_conf.full_paths,
 		    "Don't shorten the pathnames taking into account the cwd"),
 	OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
 		   "only consider symbols in these dsos"),
diff --git a/tools/perf/builtin-help.c b/tools/perf/builtin-help.c
index 9f810b17c25c..215b584007b1 100644
--- a/tools/perf/builtin-help.c
+++ b/tools/perf/builtin-help.c
@@ -286,8 +286,7 @@ void list_common_cmds_help(void)
 
 	puts(" The most commonly used perf commands are:");
 	for (i = 0; i < ARRAY_SIZE(common_cmds); i++) {
-		printf("   %s   ", common_cmds[i].name);
-		mput_char(' ', longest - strlen(common_cmds[i].name));
+		printf("   %-*s   ", longest, common_cmds[i].name);
 		puts(common_cmds[i].help);
 	}
 }
@@ -314,8 +313,6 @@ static const char *cmd_to_page(const char *perf_cmd)
 		return "perf";
 	else if (!prefixcmp(perf_cmd, "perf"))
 		return perf_cmd;
-	else if (is_perf_command(perf_cmd))
-		return prepend("perf-", perf_cmd);
 	else
 		return prepend("perf-", perf_cmd);
 }
diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c
index 93c67bf53d2c..5d5dc6b09617 100644
--- a/tools/perf/builtin-kmem.c
+++ b/tools/perf/builtin-kmem.c
@@ -92,23 +92,18 @@ static void setup_cpunode_map(void)
 	if (!dir1)
 		return;
 
-	while (true) {
-		dent1 = readdir(dir1);
-		if (!dent1)
-			break;
-
-		if (sscanf(dent1->d_name, "node%u", &mem) < 1)
+	while ((dent1 = readdir(dir1)) != NULL) {
+		if (dent1->d_type != DT_DIR ||
+		    sscanf(dent1->d_name, "node%u", &mem) < 1)
 			continue;
 
 		snprintf(buf, PATH_MAX, "%s/%s", PATH_SYS_NODE, dent1->d_name);
 		dir2 = opendir(buf);
 		if (!dir2)
 			continue;
-		while (true) {
-			dent2 = readdir(dir2);
-			if (!dent2)
-				break;
-			if (sscanf(dent2->d_name, "cpu%u", &cpu) < 1)
+		while ((dent2 = readdir(dir2)) != NULL) {
+			if (dent2->d_type != DT_LNK ||
+			    sscanf(dent2->d_name, "cpu%u", &cpu) < 1)
 				continue;
 			cpunode_map[cpu] = mem;
 		}
@@ -321,11 +316,8 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 
 	event__parse_sample(event, session->sample_type, &data);
 
-	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
-		event->header.misc,
-		data.pid, data.tid,
-		(void *)(long)data.ip,
-		(long long)data.period);
+	dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc,
+		    data.pid, data.tid, data.ip, data.period);
 
 	thread = perf_session__findnew(session, event->ip.pid);
 	if (thread == NULL) {
@@ -342,22 +334,9 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 	return 0;
 }
 
-static int sample_type_check(struct perf_session *session)
-{
-	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
-		fprintf(stderr,
-			"No trace sample to read. Did you call perf record "
-			"without -R?");
-		return -1;
-	}
-
-	return 0;
-}
-
 static struct perf_event_ops event_ops = {
-	.process_sample_event	= process_sample_event,
-	.process_comm_event	= event__process_comm,
-	.sample_type_check	= sample_type_check,
+	.sample	= process_sample_event,
+	.comm	= event__process_comm,
 };
 
 static double fragmentation(unsigned long n_req, unsigned long n_alloc)
@@ -504,11 +483,14 @@ static void sort_result(void)
 
 static int __cmd_kmem(void)
 {
-	int err;
+	int err = -EINVAL;
 	struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0);
 	if (session == NULL)
 		return -ENOMEM;
 
+	if (!perf_session__has_traces(session, "kmem record"))
+		goto out_delete;
+
 	setup_pager();
 	err = perf_session__process_events(session, &event_ops);
 	if (err != 0)
diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c
new file mode 100644
index 000000000000..fb9ab2ad3f92
--- /dev/null
+++ b/tools/perf/builtin-lock.c
@@ -0,0 +1,678 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+#include "util/session.h"
+
+#include <sys/types.h>
+#include <sys/prctl.h>
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+#include <limits.h>
+
+#include <linux/list.h>
+#include <linux/hash.h>
+
+/* based on kernel/lockdep.c */
+#define LOCKHASH_BITS		12
+#define LOCKHASH_SIZE		(1UL << LOCKHASH_BITS)
+
+static struct list_head lockhash_table[LOCKHASH_SIZE];
+
+#define __lockhashfn(key)	hash_long((unsigned long)key, LOCKHASH_BITS)
+#define lockhashentry(key)	(lockhash_table + __lockhashfn((key)))
+
+#define LOCK_STATE_UNLOCKED	0	       /* initial state */
+#define LOCK_STATE_LOCKED	1
+
+struct lock_stat {
+	struct list_head	hash_entry;
+	struct rb_node		rb;		/* used for sorting */
+
+	/*
+	 * FIXME: raw_field_value() returns unsigned long long,
+	 * so address of lockdep_map should be dealed as 64bit.
+	 * Is there more better solution?
+	 */
+	void			*addr;		/* address of lockdep_map, used as ID */
+	char			*name;		/* for strcpy(), we cannot use const */
+
+	int			state;
+	u64			prev_event_time; /* timestamp of previous event */
+
+	unsigned int		nr_acquired;
+	unsigned int		nr_acquire;
+	unsigned int		nr_contended;
+	unsigned int		nr_release;
+
+	/* these times are in nano sec. */
+	u64			wait_time_total;
+	u64			wait_time_min;
+	u64			wait_time_max;
+};
+
+/* build simple key function one is bigger than two */
+#define SINGLE_KEY(member)						\
+	static int lock_stat_key_ ## member(struct lock_stat *one,	\
+					 struct lock_stat *two)		\
+	{								\
+		return one->member > two->member;			\
+	}
+
+SINGLE_KEY(nr_acquired)
+SINGLE_KEY(nr_contended)
+SINGLE_KEY(wait_time_total)
+SINGLE_KEY(wait_time_min)
+SINGLE_KEY(wait_time_max)
+
+struct lock_key {
+	/*
+	 * name: the value for specify by user
+	 * this should be simpler than raw name of member
+	 * e.g. nr_acquired -> acquired, wait_time_total -> wait_total
+	 */
+	const char		*name;
+	int			(*key)(struct lock_stat*, struct lock_stat*);
+};
+
+static const char		*sort_key = "acquired";
+
+static int			(*compare)(struct lock_stat *, struct lock_stat *);
+
+static struct rb_root		result;	/* place to store sorted data */
+
+#define DEF_KEY_LOCK(name, fn_suffix)	\
+	{ #name, lock_stat_key_ ## fn_suffix }
+struct lock_key keys[] = {
+	DEF_KEY_LOCK(acquired, nr_acquired),
+	DEF_KEY_LOCK(contended, nr_contended),
+	DEF_KEY_LOCK(wait_total, wait_time_total),
+	DEF_KEY_LOCK(wait_min, wait_time_min),
+	DEF_KEY_LOCK(wait_max, wait_time_max),
+
+	/* extra comparisons much complicated should be here */
+
+	{ NULL, NULL }
+};
+
+static void select_key(void)
+{
+	int i;
+
+	for (i = 0; keys[i].name; i++) {
+		if (!strcmp(keys[i].name, sort_key)) {
+			compare = keys[i].key;
+			return;
+		}
+	}
+
+	die("Unknown compare key:%s\n", sort_key);
+}
+
+static void insert_to_result(struct lock_stat *st,
+			     int (*bigger)(struct lock_stat *, struct lock_stat *))
+{
+	struct rb_node **rb = &result.rb_node;
+	struct rb_node *parent = NULL;
+	struct lock_stat *p;
+
+	while (*rb) {
+		p = container_of(*rb, struct lock_stat, rb);
+		parent = *rb;
+
+		if (bigger(st, p))
+			rb = &(*rb)->rb_left;
+		else
+			rb = &(*rb)->rb_right;
+	}
+
+	rb_link_node(&st->rb, parent, rb);
+	rb_insert_color(&st->rb, &result);
+}
+
+/* returns left most element of result, and erase it */
+static struct lock_stat *pop_from_result(void)
+{
+	struct rb_node *node = result.rb_node;
+
+	if (!node)
+		return NULL;
+
+	while (node->rb_left)
+		node = node->rb_left;
+
+	rb_erase(node, &result);
+	return container_of(node, struct lock_stat, rb);
+}
+
+static struct lock_stat *lock_stat_findnew(void *addr, const char *name)
+{
+	struct list_head *entry = lockhashentry(addr);
+	struct lock_stat *ret, *new;
+
+	list_for_each_entry(ret, entry, hash_entry) {
+		if (ret->addr == addr)
+			return ret;
+	}
+
+	new = zalloc(sizeof(struct lock_stat));
+	if (!new)
+		goto alloc_failed;
+
+	new->addr = addr;
+	new->name = zalloc(sizeof(char) * strlen(name) + 1);
+	if (!new->name)
+		goto alloc_failed;
+	strcpy(new->name, name);
+
+	/* LOCK_STATE_UNLOCKED == 0 isn't guaranteed forever */
+	new->state = LOCK_STATE_UNLOCKED;
+	new->wait_time_min = ULLONG_MAX;
+
+	list_add(&new->hash_entry, entry);
+	return new;
+
+alloc_failed:
+	die("memory allocation failed\n");
+}
+
+static char			const *input_name = "perf.data";
+
+static int			profile_cpu = -1;
+
+struct raw_event_sample {
+	u32			size;
+	char			data[0];
+};
+
+struct trace_acquire_event {
+	void			*addr;
+	const char		*name;
+};
+
+struct trace_acquired_event {
+	void			*addr;
+	const char		*name;
+};
+
+struct trace_contended_event {
+	void			*addr;
+	const char		*name;
+};
+
+struct trace_release_event {
+	void			*addr;
+	const char		*name;
+};
+
+struct trace_lock_handler {
+	void (*acquire_event)(struct trace_acquire_event *,
+			      struct event *,
+			      int cpu,
+			      u64 timestamp,
+			      struct thread *thread);
+
+	void (*acquired_event)(struct trace_acquired_event *,
+			       struct event *,
+			       int cpu,
+			       u64 timestamp,
+			       struct thread *thread);
+
+	void (*contended_event)(struct trace_contended_event *,
+				struct event *,
+				int cpu,
+				u64 timestamp,
+				struct thread *thread);
+
+	void (*release_event)(struct trace_release_event *,
+			      struct event *,
+			      int cpu,
+			      u64 timestamp,
+			      struct thread *thread);
+};
+
+static void
+report_lock_acquire_event(struct trace_acquire_event *acquire_event,
+			struct event *__event __used,
+			int cpu __used,
+			u64 timestamp,
+			struct thread *thread __used)
+{
+	struct lock_stat *st;
+
+	st = lock_stat_findnew(acquire_event->addr, acquire_event->name);
+
+	switch (st->state) {
+	case LOCK_STATE_UNLOCKED:
+		break;
+	case LOCK_STATE_LOCKED:
+		break;
+	default:
+		BUG_ON(1);
+		break;
+	}
+
+	st->prev_event_time = timestamp;
+}
+
+static void
+report_lock_acquired_event(struct trace_acquired_event *acquired_event,
+			 struct event *__event __used,
+			 int cpu __used,
+			 u64 timestamp,
+			 struct thread *thread __used)
+{
+	struct lock_stat *st;
+
+	st = lock_stat_findnew(acquired_event->addr, acquired_event->name);
+
+	switch (st->state) {
+	case LOCK_STATE_UNLOCKED:
+		st->state = LOCK_STATE_LOCKED;
+		st->nr_acquired++;
+		break;
+	case LOCK_STATE_LOCKED:
+		break;
+	default:
+		BUG_ON(1);
+		break;
+	}
+
+	st->prev_event_time = timestamp;
+}
+
+static void
+report_lock_contended_event(struct trace_contended_event *contended_event,
+			  struct event *__event __used,
+			  int cpu __used,
+			  u64 timestamp,
+			  struct thread *thread __used)
+{
+	struct lock_stat *st;
+
+	st = lock_stat_findnew(contended_event->addr, contended_event->name);
+
+	switch (st->state) {
+	case LOCK_STATE_UNLOCKED:
+		break;
+	case LOCK_STATE_LOCKED:
+		st->nr_contended++;
+		break;
+	default:
+		BUG_ON(1);
+		break;
+	}
+
+	st->prev_event_time = timestamp;
+}
+
+static void
+report_lock_release_event(struct trace_release_event *release_event,
+			struct event *__event __used,
+			int cpu __used,
+			u64 timestamp,
+			struct thread *thread __used)
+{
+	struct lock_stat *st;
+	u64 hold_time;
+
+	st = lock_stat_findnew(release_event->addr, release_event->name);
+
+	switch (st->state) {
+	case LOCK_STATE_UNLOCKED:
+		break;
+	case LOCK_STATE_LOCKED:
+		st->state = LOCK_STATE_UNLOCKED;
+		hold_time = timestamp - st->prev_event_time;
+
+		if (timestamp < st->prev_event_time) {
+			/* terribly, this can happen... */
+			goto end;
+		}
+
+		if (st->wait_time_min > hold_time)
+			st->wait_time_min = hold_time;
+		if (st->wait_time_max < hold_time)
+			st->wait_time_max = hold_time;
+		st->wait_time_total += hold_time;
+
+		st->nr_release++;
+		break;
+	default:
+		BUG_ON(1);
+		break;
+	}
+
+end:
+	st->prev_event_time = timestamp;
+}
+
+/* lock oriented handlers */
+/* TODO: handlers for CPU oriented, thread oriented */
+static struct trace_lock_handler report_lock_ops  = {
+	.acquire_event		= report_lock_acquire_event,
+	.acquired_event		= report_lock_acquired_event,
+	.contended_event	= report_lock_contended_event,
+	.release_event		= report_lock_release_event,
+};
+
+static struct trace_lock_handler *trace_handler;
+
+static void
+process_lock_acquire_event(void *data,
+			   struct event *event __used,
+			   int cpu __used,
+			   u64 timestamp __used,
+			   struct thread *thread __used)
+{
+	struct trace_acquire_event acquire_event;
+	u64 tmp;		/* this is required for casting... */
+
+	tmp = raw_field_value(event, "lockdep_addr", data);
+	memcpy(&acquire_event.addr, &tmp, sizeof(void *));
+	acquire_event.name = (char *)raw_field_ptr(event, "name", data);
+
+	if (trace_handler->acquire_event)
+		trace_handler->acquire_event(&acquire_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_lock_acquired_event(void *data,
+			    struct event *event __used,
+			    int cpu __used,
+			    u64 timestamp __used,
+			    struct thread *thread __used)
+{
+	struct trace_acquired_event acquired_event;
+	u64 tmp;		/* this is required for casting... */
+
+	tmp = raw_field_value(event, "lockdep_addr", data);
+	memcpy(&acquired_event.addr, &tmp, sizeof(void *));
+	acquired_event.name = (char *)raw_field_ptr(event, "name", data);
+
+	if (trace_handler->acquire_event)
+		trace_handler->acquired_event(&acquired_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_lock_contended_event(void *data,
+			     struct event *event __used,
+			     int cpu __used,
+			     u64 timestamp __used,
+			     struct thread *thread __used)
+{
+	struct trace_contended_event contended_event;
+	u64 tmp;		/* this is required for casting... */
+
+	tmp = raw_field_value(event, "lockdep_addr", data);
+	memcpy(&contended_event.addr, &tmp, sizeof(void *));
+	contended_event.name = (char *)raw_field_ptr(event, "name", data);
+
+	if (trace_handler->acquire_event)
+		trace_handler->contended_event(&contended_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_lock_release_event(void *data,
+			   struct event *event __used,
+			   int cpu __used,
+			   u64 timestamp __used,
+			   struct thread *thread __used)
+{
+	struct trace_release_event release_event;
+	u64 tmp;		/* this is required for casting... */
+
+	tmp = raw_field_value(event, "lockdep_addr", data);
+	memcpy(&release_event.addr, &tmp, sizeof(void *));
+	release_event.name = (char *)raw_field_ptr(event, "name", data);
+
+	if (trace_handler->acquire_event)
+		trace_handler->release_event(&release_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_raw_event(void *data, int cpu,
+		  u64 timestamp, struct thread *thread)
+{
+	struct event *event;
+	int type;
+
+	type = trace_parse_common_type(data);
+	event = trace_find_event(type);
+
+	if (!strcmp(event->name, "lock_acquire"))
+		process_lock_acquire_event(data, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "lock_acquired"))
+		process_lock_acquired_event(data, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "lock_contended"))
+		process_lock_contended_event(data, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "lock_release"))
+		process_lock_release_event(data, event, cpu, timestamp, thread);
+}
+
+static int process_sample_event(event_t *event, struct perf_session *session)
+{
+	struct thread *thread;
+	struct sample_data data;
+
+	bzero(&data, sizeof(struct sample_data));
+	event__parse_sample(event, session->sample_type, &data);
+	thread = perf_session__findnew(session, data.pid);
+
+	if (thread == NULL) {
+		pr_debug("problem processing %d event, skipping it.\n",
+			 event->header.type);
+		return -1;
+	}
+
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
+
+	if (profile_cpu != -1 && profile_cpu != (int) data.cpu)
+		return 0;
+
+	process_raw_event(data.raw_data, data.cpu, data.time, thread);
+
+	return 0;
+}
+
+/* TODO: various way to print, coloring, nano or milli sec */
+static void print_result(void)
+{
+	struct lock_stat *st;
+	char cut_name[20];
+
+	printf("%18s ", "ID");
+	printf("%20s ", "Name");
+	printf("%10s ", "acquired");
+	printf("%10s ", "contended");
+
+	printf("%15s ", "total wait (ns)");
+	printf("%15s ", "max wait (ns)");
+	printf("%15s ", "min wait (ns)");
+
+	printf("\n\n");
+
+	while ((st = pop_from_result())) {
+		bzero(cut_name, 20);
+
+		printf("%p ", st->addr);
+
+		if (strlen(st->name) < 16) {
+			/* output raw name */
+			printf("%20s ", st->name);
+		} else {
+			strncpy(cut_name, st->name, 16);
+			cut_name[16] = '.';
+			cut_name[17] = '.';
+			cut_name[18] = '.';
+			cut_name[19] = '\0';
+			/* cut off name for saving output style */
+			printf("%20s ", cut_name);
+		}
+
+		printf("%10u ", st->nr_acquired);
+		printf("%10u ", st->nr_contended);
+
+		printf("%15llu ", st->wait_time_total);
+		printf("%15llu ", st->wait_time_max);
+		printf("%15llu ", st->wait_time_min == ULLONG_MAX ?
+		       0 : st->wait_time_min);
+		printf("\n");
+	}
+}
+
+static void dump_map(void)
+{
+	unsigned int i;
+	struct lock_stat *st;
+
+	for (i = 0; i < LOCKHASH_SIZE; i++) {
+		list_for_each_entry(st, &lockhash_table[i], hash_entry) {
+			printf("%p: %s\n", st->addr, st->name);
+		}
+	}
+}
+
+static struct perf_event_ops eops = {
+	.sample			= process_sample_event,
+	.comm			= event__process_comm,
+};
+
+static struct perf_session *session;
+
+static int read_events(void)
+{
+	session = perf_session__new(input_name, O_RDONLY, 0);
+	if (!session)
+		die("Initializing perf session failed\n");
+
+	return perf_session__process_events(session, &eops);
+}
+
+static void sort_result(void)
+{
+	unsigned int i;
+	struct lock_stat *st;
+
+	for (i = 0; i < LOCKHASH_SIZE; i++) {
+		list_for_each_entry(st, &lockhash_table[i], hash_entry) {
+			insert_to_result(st, compare);
+		}
+	}
+}
+
+static void __cmd_report(void)
+{
+	setup_pager();
+	select_key();
+	read_events();
+	sort_result();
+	print_result();
+}
+
+static const char * const report_usage[] = {
+	"perf lock report [<options>]",
+	NULL
+};
+
+static const struct option report_options[] = {
+	OPT_STRING('k', "key", &sort_key, "acquired",
+		    "key for sorting"),
+	/* TODO: type */
+	OPT_END()
+};
+
+static const char * const lock_usage[] = {
+	"perf lock [<options>] {record|trace|report}",
+	NULL
+};
+
+static const struct option lock_options[] = {
+	OPT_STRING('i', "input", &input_name, "file", "input file name"),
+	OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"),
+	OPT_END()
+};
+
+static const char *record_args[] = {
+	"record",
+	"-a",
+	"-R",
+	"-M",
+	"-f",
+	"-m", "1024",
+	"-c", "1",
+	"-e", "lock:lock_acquire:r",
+	"-e", "lock:lock_acquired:r",
+	"-e", "lock:lock_contended:r",
+	"-e", "lock:lock_release:r",
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+
+	rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[i] = strdup(record_args[i]);
+
+	for (j = 1; j < (unsigned int)argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_record(i, rec_argv, NULL);
+}
+
+int cmd_lock(int argc, const char **argv, const char *prefix __used)
+{
+	unsigned int i;
+
+	symbol__init();
+	for (i = 0; i < LOCKHASH_SIZE; i++)
+		INIT_LIST_HEAD(lockhash_table + i);
+
+	argc = parse_options(argc, argv, lock_options, lock_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc)
+		usage_with_options(lock_usage, lock_options);
+
+	if (!strncmp(argv[0], "rec", 3)) {
+		return __cmd_record(argc, argv);
+	} else if (!strncmp(argv[0], "report", 6)) {
+		trace_handler = &report_lock_ops;
+		if (argc) {
+			argc = parse_options(argc, argv,
+					     report_options, report_usage, 0);
+			if (argc)
+				usage_with_options(report_usage, report_options);
+		}
+		__cmd_report();
+	} else if (!strcmp(argv[0], "trace")) {
+		/* Aliased to 'perf trace' */
+		return cmd_trace(argc, argv, prefix);
+	} else if (!strcmp(argv[0], "map")) {
+		/* recycling report_lock_ops */
+		trace_handler = &report_lock_ops;
+		setup_pager();
+		read_events();
+		dump_map();
+	} else {
+		usage_with_options(lock_usage, lock_options);
+	}
+
+	return 0;
+}
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
index c1e6774fd3ed..34f2acb1cc88 100644
--- a/tools/perf/builtin-probe.c
+++ b/tools/perf/builtin-probe.c
@@ -55,11 +55,13 @@ static struct {
 	bool need_dwarf;
 	bool list_events;
 	bool force_add;
+	bool show_lines;
 	int nr_probe;
 	struct probe_point probes[MAX_PROBES];
 	struct strlist *dellist;
 	struct perf_session *psession;
 	struct map *kmap;
+	struct line_range line_range;
 } session;
 
 
@@ -137,6 +139,16 @@ static int open_vmlinux(void)
 	pr_debug("Try to open %s\n", session.kmap->dso->long_name);
 	return open(session.kmap->dso->long_name, O_RDONLY);
 }
+
+static int opt_show_lines(const struct option *opt __used,
+			  const char *str, int unset __used)
+{
+	if (str)
+		parse_line_range_desc(str, &session.line_range);
+	INIT_LIST_HEAD(&session.line_range.line_list);
+	session.show_lines = true;
+	return 0;
+}
 #endif
 
 static const char * const probe_usage[] = {
@@ -144,6 +156,7 @@ static const char * const probe_usage[] = {
 	"perf probe [<options>] --add 'PROBEDEF' [--add 'PROBEDEF' ...]",
 	"perf probe [<options>] --del '[GROUP:]EVENT' ...",
 	"perf probe --list",
+	"perf probe --line 'LINEDESC'",
 	NULL
 };
 
@@ -182,9 +195,32 @@ static const struct option options[] = {
 		opt_add_probe_event),
 	OPT_BOOLEAN('f', "force", &session.force_add, "forcibly add events"
 		    " with existing name"),
+#ifndef NO_LIBDWARF
+	OPT_CALLBACK('L', "line", NULL,
+		     "FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]",
+		     "Show source code lines.", opt_show_lines),
+#endif
 	OPT_END()
 };
 
+/* Initialize symbol maps for vmlinux */
+static void init_vmlinux(void)
+{
+	symbol_conf.sort_by_name = true;
+	if (symbol_conf.vmlinux_name == NULL)
+		symbol_conf.try_vmlinux_path = true;
+	else
+		pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name);
+	if (symbol__init() < 0)
+		die("Failed to init symbol map.");
+	session.psession = perf_session__new(NULL, O_WRONLY, false);
+	if (session.psession == NULL)
+		die("Failed to init perf_session.");
+	session.kmap = session.psession->vmlinux_maps[MAP__FUNCTION];
+	if (!session.kmap)
+		die("Could not find kernel map.\n");
+}
+
 int cmd_probe(int argc, const char **argv, const char *prefix __used)
 {
 	int i, ret;
@@ -203,7 +239,8 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used)
 		parse_probe_event_argv(argc, argv);
 	}
 
-	if ((!session.nr_probe && !session.dellist && !session.list_events))
+	if ((!session.nr_probe && !session.dellist && !session.list_events &&
+	     !session.show_lines))
 		usage_with_options(probe_usage, options);
 
 	if (debugfs_valid_mountpoint(debugfs_path) < 0)
@@ -215,10 +252,34 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used)
 				   " --add/--del.\n");
 			usage_with_options(probe_usage, options);
 		}
+		if (session.show_lines) {
+			pr_warning("  Error: Don't use --list with --line.\n");
+			usage_with_options(probe_usage, options);
+		}
 		show_perf_probe_events();
 		return 0;
 	}
 
+#ifndef NO_LIBDWARF
+	if (session.show_lines) {
+		if (session.nr_probe != 0 || session.dellist) {
+			pr_warning("  Error: Don't use --line with"
+				   " --add/--del.\n");
+			usage_with_options(probe_usage, options);
+		}
+		init_vmlinux();
+		fd = open_vmlinux();
+		if (fd < 0)
+			die("Could not open debuginfo file.");
+		ret = find_line_range(fd, &session.line_range);
+		if (ret <= 0)
+			die("Source line is not found.\n");
+		close(fd);
+		show_line_range(&session.line_range);
+		return 0;
+	}
+#endif
+
 	if (session.dellist) {
 		del_trace_kprobe_events(session.dellist);
 		strlist__delete(session.dellist);
@@ -226,20 +287,8 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used)
 			return 0;
 	}
 
-	/* Initialize symbol maps for vmlinux */
-	symbol_conf.sort_by_name = true;
-	if (symbol_conf.vmlinux_name == NULL)
-		symbol_conf.try_vmlinux_path = true;
-	if (symbol__init() < 0)
-		die("Failed to init symbol map.");
-	session.psession = perf_session__new(NULL, O_WRONLY, false);
-	if (session.psession == NULL)
-		die("Failed to init perf_session.");
-	session.kmap = map_groups__find_by_name(&session.psession->kmaps,
-						MAP__FUNCTION,
-						"[kernel.kallsyms]");
-	if (!session.kmap)
-		die("Could not find kernel map.\n");
+	/* Add probes */
+	init_vmlinux();
 
 	if (session.need_dwarf)
 #ifdef NO_LIBDWARF
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 265425322734..949167efa1ed 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -5,6 +5,9 @@
  * (or a CPU, or a PID) into the perf.data output file - for
  * later analysis via perf report.
  */
+#define _LARGEFILE64_SOURCE
+#define _FILE_OFFSET_BITS 64
+
 #include "builtin.h"
 
 #include "perf.h"
@@ -117,8 +120,12 @@ static void write_event(event_t *buf, size_t size)
 	* Add it to the list of DSOs, so that when we finish this
 	 * record session we can pick the available build-ids.
 	 */
-	if (buf->header.type == PERF_RECORD_MMAP)
-		dsos__findnew(buf->mmap.filename);
+	if (buf->header.type == PERF_RECORD_MMAP) {
+		struct list_head *head = &dsos__user;
+		if (buf->mmap.header.misc == 1)
+			head = &dsos__kernel;
+		__dsos__findnew(head, buf->mmap.filename);
+	}
 
 	write_output(buf, size);
 }
@@ -447,7 +454,7 @@ static int __cmd_record(int argc, const char **argv)
 		append_file = 0;
 	}
 
-	flags = O_CREAT|O_RDWR;
+	flags = O_CREAT|O_RDWR|O_LARGEFILE;
 	if (append_file)
 		file_new = 0;
 	else
@@ -551,6 +558,19 @@ static int __cmd_record(int argc, const char **argv)
 			return err;
 	}
 
+	err = event__synthesize_kernel_mmap(process_synthesized_event,
+					    session, "_text");
+	if (err < 0) {
+		pr_err("Couldn't record kernel reference relocation symbol.\n");
+		return err;
+	}
+
+	err = event__synthesize_modules(process_synthesized_event, session);
+	if (err < 0) {
+		pr_err("Couldn't record kernel reference relocation symbol.\n");
+		return err;
+	}
+
 	if (!system_wide && profile_cpu == -1)
 		event__synthesize_thread(pid, process_synthesized_event,
 					 session);
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 860f1eeeea7d..cfc655d40bb7 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -34,6 +34,8 @@
 static char		const *input_name = "perf.data";
 
 static int		force;
+static bool		hide_unresolved;
+static bool		dont_use_callchains;
 
 static int		show_threads;
 static struct perf_read_values	show_threads_values;
@@ -91,11 +93,8 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 
 	event__parse_sample(event, session->sample_type, &data);
 
-	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
-		event->header.misc,
-		data.pid, data.tid,
-		(void *)(long)data.ip,
-		(long long)data.period);
+	dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc,
+		    data.pid, data.tid, data.ip, data.period);
 
 	if (session->sample_type & PERF_SAMPLE_CALLCHAIN) {
 		unsigned int i;
@@ -121,7 +120,7 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 		return -1;
 	}
 
-	if (al.filtered)
+	if (al.filtered || (hide_unresolved && al.sym == NULL))
 		return 0;
 
 	if (perf_session__add_hist_entry(session, &al, data.callchain, data.period)) {
@@ -156,14 +155,14 @@ static int process_read_event(event_t *event, struct perf_session *session __use
 	return 0;
 }
 
-static int sample_type_check(struct perf_session *session)
+static int perf_session__setup_sample_type(struct perf_session *self)
 {
-	if (!(session->sample_type & PERF_SAMPLE_CALLCHAIN)) {
+	if (!(self->sample_type & PERF_SAMPLE_CALLCHAIN)) {
 		if (sort__has_parent) {
 			fprintf(stderr, "selected --sort parent, but no"
 					" callchain data. Did you call"
 					" perf record without -g?\n");
-			return -1;
+			return -EINVAL;
 		}
 		if (symbol_conf.use_callchain) {
 			fprintf(stderr, "selected -g but no callchain data."
@@ -171,12 +170,13 @@ static int sample_type_check(struct perf_session *session)
 					" -g?\n");
 			return -1;
 		}
-	} else if (callchain_param.mode != CHAIN_NONE && !symbol_conf.use_callchain) {
+	} else if (!dont_use_callchains && callchain_param.mode != CHAIN_NONE &&
+		   !symbol_conf.use_callchain) {
 			symbol_conf.use_callchain = true;
 			if (register_callchain_param(&callchain_param) < 0) {
 				fprintf(stderr, "Can't register callchain"
 						" params\n");
-				return -1;
+				return -EINVAL;
 			}
 	}
 
@@ -184,20 +184,18 @@ static int sample_type_check(struct perf_session *session)
 }
 
 static struct perf_event_ops event_ops = {
-	.process_sample_event	= process_sample_event,
-	.process_mmap_event	= event__process_mmap,
-	.process_comm_event	= event__process_comm,
-	.process_exit_event	= event__process_task,
-	.process_fork_event	= event__process_task,
-	.process_lost_event	= event__process_lost,
-	.process_read_event	= process_read_event,
-	.sample_type_check	= sample_type_check,
+	.sample	= process_sample_event,
+	.mmap	= event__process_mmap,
+	.comm	= event__process_comm,
+	.exit	= event__process_task,
+	.fork	= event__process_task,
+	.lost	= event__process_lost,
+	.read	= process_read_event,
 };
 
-
 static int __cmd_report(void)
 {
-	int ret;
+	int ret = -EINVAL;
 	struct perf_session *session;
 
 	session = perf_session__new(input_name, O_RDONLY, force);
@@ -207,6 +205,10 @@ static int __cmd_report(void)
 	if (show_threads)
 		perf_read_values_init(&show_threads_values);
 
+	ret = perf_session__setup_sample_type(session);
+	if (ret)
+		goto out_delete;
+
 	ret = perf_session__process_events(session, &event_ops);
 	if (ret)
 		goto out_delete;
@@ -243,11 +245,19 @@ out_delete:
 
 static int
 parse_callchain_opt(const struct option *opt __used, const char *arg,
-		    int unset __used)
+		    int unset)
 {
 	char *tok;
 	char *endptr;
 
+	/*
+	 * --no-call-graph
+	 */
+	if (unset) {
+		dont_use_callchains = true;
+		return 0;
+	}
+
 	symbol_conf.use_callchain = true;
 
 	if (!arg)
@@ -319,7 +329,7 @@ static const struct option options[] = {
 		   "pretty printing style key: normal raw"),
 	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
 		   "sort by key(s): pid, comm, dso, symbol, parent"),
-	OPT_BOOLEAN('P', "full-paths", &event_ops.full_paths,
+	OPT_BOOLEAN('P', "full-paths", &symbol_conf.full_paths,
 		    "Don't shorten the pathnames taking into account the cwd"),
 	OPT_STRING('p', "parent", &parent_pattern, "regex",
 		   "regex filter to identify parent, see: '--sort parent'"),
@@ -340,6 +350,8 @@ static const struct option options[] = {
 	OPT_STRING('t', "field-separator", &symbol_conf.field_sep, "separator",
 		   "separator for columns, no spaces will be added between "
 		   "columns '.' is reserved."),
+	OPT_BOOLEAN('U', "hide-unresolved", &hide_unresolved,
+		    "Only display entries resolved to a symbol"),
 	OPT_END()
 };
 
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index 80209df6cfe8..4f5a03e43444 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -1621,11 +1621,8 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 
 	event__parse_sample(event, session->sample_type, &data);
 
-	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
-		event->header.misc,
-		data.pid, data.tid,
-		(void *)(long)data.ip,
-		(long long)data.period);
+	dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc,
+		    data.pid, data.tid, data.ip, data.period);
 
 	thread = perf_session__findnew(session, data.pid);
 	if (thread == NULL) {
@@ -1653,33 +1650,22 @@ static int process_lost_event(event_t *event __used,
 	return 0;
 }
 
-static int sample_type_check(struct perf_session *session __used)
-{
-	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
-		fprintf(stderr,
-			"No trace sample to read. Did you call perf record "
-			"without -R?");
-		return -1;
-	}
-
-	return 0;
-}
-
 static struct perf_event_ops event_ops = {
-	.process_sample_event	= process_sample_event,
-	.process_comm_event	= event__process_comm,
-	.process_lost_event	= process_lost_event,
-	.sample_type_check	= sample_type_check,
+	.sample	= process_sample_event,
+	.comm	= event__process_comm,
+	.lost	= process_lost_event,
 };
 
 static int read_events(void)
 {
-	int err;
+	int err = -EINVAL;
 	struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0);
 	if (session == NULL)
 		return -ENOMEM;
 
-	err = perf_session__process_events(session, &event_ops);
+	if (perf_session__has_traces(session, "record -R"))
+		err = perf_session__process_events(session, &event_ops);
+
 	perf_session__delete(session);
 	return err;
 }
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index c70d72003557..e8c85d5aec41 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -44,6 +44,7 @@
 #include "util/parse-events.h"
 #include "util/event.h"
 #include "util/debug.h"
+#include "util/header.h"
 
 #include <sys/prctl.h>
 #include <math.h>
@@ -79,6 +80,8 @@ static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
 
 static int			event_scaled[MAX_COUNTERS];
 
+static volatile int done = 0;
+
 struct stats
 {
 	double n, mean, M2;
@@ -247,61 +250,64 @@ static int run_perf_stat(int argc __used, const char **argv)
 	unsigned long long t0, t1;
 	int status = 0;
 	int counter;
-	int pid;
+	int pid = target_pid;
 	int child_ready_pipe[2], go_pipe[2];
+	const bool forks = (target_pid == -1 && argc > 0);
 	char buf;
 
 	if (!system_wide)
 		nr_cpus = 1;
 
-	if (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0) {
+	if (forks && (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0)) {
 		perror("failed to create pipes");
 		exit(1);
 	}
 
-	if ((pid = fork()) < 0)
-		perror("failed to fork");
+	if (forks) {
+		if ((pid = fork()) < 0)
+			perror("failed to fork");
+
+		if (!pid) {
+			close(child_ready_pipe[0]);
+			close(go_pipe[1]);
+			fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
+
+			/*
+			 * Do a dummy execvp to get the PLT entry resolved,
+			 * so we avoid the resolver overhead on the real
+			 * execvp call.
+			 */
+			execvp("", (char **)argv);
+
+			/*
+			 * Tell the parent we're ready to go
+			 */
+			close(child_ready_pipe[1]);
+
+			/*
+			 * Wait until the parent tells us to go.
+			 */
+			if (read(go_pipe[0], &buf, 1) == -1)
+				perror("unable to read pipe");
+
+			execvp(argv[0], (char **)argv);
+
+			perror(argv[0]);
+			exit(-1);
+		}
 
-	if (!pid) {
-		close(child_ready_pipe[0]);
-		close(go_pipe[1]);
-		fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
+		child_pid = pid;
 
 		/*
-		 * Do a dummy execvp to get the PLT entry resolved,
-		 * so we avoid the resolver overhead on the real
-		 * execvp call.
-		 */
-		execvp("", (char **)argv);
-
-		/*
-		 * Tell the parent we're ready to go
+		 * Wait for the child to be ready to exec.
 		 */
 		close(child_ready_pipe[1]);
-
-		/*
-		 * Wait until the parent tells us to go.
-		 */
-		if (read(go_pipe[0], &buf, 1) == -1)
+		close(go_pipe[0]);
+		if (read(child_ready_pipe[0], &buf, 1) == -1)
 			perror("unable to read pipe");
-
-		execvp(argv[0], (char **)argv);
-
-		perror(argv[0]);
-		exit(-1);
+		close(child_ready_pipe[0]);
 	}
 
-	child_pid = pid;
-
-	/*
-	 * Wait for the child to be ready to exec.
-	 */
-	close(child_ready_pipe[1]);
-	close(go_pipe[0]);
-	if (read(child_ready_pipe[0], &buf, 1) == -1)
-		perror("unable to read pipe");
-	close(child_ready_pipe[0]);
-
 	for (counter = 0; counter < nr_counters; counter++)
 		create_perf_stat_counter(counter, pid);
 
@@ -310,8 +316,12 @@ static int run_perf_stat(int argc __used, const char **argv)
 	 */
 	t0 = rdclock();
 
-	close(go_pipe[1]);
-	wait(&status);
+	if (forks) {
+		close(go_pipe[1]);
+		wait(&status);
+	} else {
+		while(!done);
+	}
 
 	t1 = rdclock();
 
@@ -417,10 +427,13 @@ static void print_stat(int argc, const char **argv)
 	fflush(stdout);
 
 	fprintf(stderr, "\n");
-	fprintf(stderr, " Performance counter stats for \'%s", argv[0]);
-
-	for (i = 1; i < argc; i++)
-		fprintf(stderr, " %s", argv[i]);
+	fprintf(stderr, " Performance counter stats for ");
+	if(target_pid == -1) {
+		fprintf(stderr, "\'%s", argv[0]);
+		for (i = 1; i < argc; i++)
+			fprintf(stderr, " %s", argv[i]);
+	}else
+		fprintf(stderr, "task pid \'%d", target_pid);
 
 	fprintf(stderr, "\'");
 	if (run_count > 1)
@@ -445,6 +458,9 @@ static volatile int signr = -1;
 
 static void skip_signal(int signo)
 {
+	if(target_pid != -1)
+		done = 1;
+
 	signr = signo;
 }
 
@@ -461,7 +477,7 @@ static void sig_atexit(void)
 }
 
 static const char * const stat_usage[] = {
-	"perf stat [<options>] <command>",
+	"perf stat [<options>] [<command>]",
 	NULL
 };
 
@@ -492,7 +508,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
 
 	argc = parse_options(argc, argv, options, stat_usage,
 		PARSE_OPT_STOP_AT_NON_OPTION);
-	if (!argc)
+	if (!argc && target_pid == -1)
 		usage_with_options(stat_usage, options);
 	if (run_count <= 0)
 		usage_with_options(stat_usage, options);
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index 3f8bbcfb1e9b..0d4d8ff7914b 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -1029,33 +1029,24 @@ static void process_samples(struct perf_session *session)
 	}
 }
 
-static int sample_type_check(struct perf_session *session)
-{
-	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
-		fprintf(stderr, "No trace samples found in the file.\n"
-				"Have you used 'perf timechart record' to record it?\n");
-		return -1;
-	}
-
-	return 0;
-}
-
 static struct perf_event_ops event_ops = {
-	.process_comm_event	= process_comm_event,
-	.process_fork_event	= process_fork_event,
-	.process_exit_event	= process_exit_event,
-	.process_sample_event	= queue_sample_event,
-	.sample_type_check	= sample_type_check,
+	.comm	= process_comm_event,
+	.fork	= process_fork_event,
+	.exit	= process_exit_event,
+	.sample	= queue_sample_event,
 };
 
 static int __cmd_timechart(void)
 {
 	struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0);
-	int ret;
+	int ret = -EINVAL;
 
 	if (session == NULL)
 		return -ENOMEM;
 
+	if (!perf_session__has_traces(session, "timechart record"))
+		goto out_delete;
+
 	ret = perf_session__process_events(session, &event_ops);
 	if (ret)
 		goto out_delete;
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index ddc584b64871..1fc018e048e1 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -667,7 +667,7 @@ static void prompt_symbol(struct sym_entry **target, const char *msg)
 	}
 
 	if (!found) {
-		fprintf(stderr, "Sorry, %s is not active.\n", sym_filter);
+		fprintf(stderr, "Sorry, %s is not active.\n", buf);
 		sleep(1);
 		return;
 	} else
@@ -934,8 +934,11 @@ static void event__process_sample(const event_t *self,
 	struct addr_location al;
 	u8 origin = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
 
+	++samples;
+
 	switch (origin) {
 	case PERF_RECORD_MISC_USER:
+		++userspace_samples;
 		if (hide_user_symbols)
 			return;
 		break;
@@ -948,9 +951,31 @@ static void event__process_sample(const event_t *self,
 	}
 
 	if (event__preprocess_sample(self, session, &al, symbol_filter) < 0 ||
-	    al.sym == NULL || al.filtered)
+	    al.filtered)
 		return;
 
+	if (al.sym == NULL) {
+		/*
+		 * As we do lazy loading of symtabs we only will know if the
+		 * specified vmlinux file is invalid when we actually have a
+		 * hit in kernel space and then try to load it. So if we get
+		 * here and there are _no_ symbols in the DSO backing the
+		 * kernel map, bail out.
+		 *
+		 * We may never get here, for instance, if we use -K/
+		 * --hide-kernel-symbols, even if the user specifies an
+		 * invalid --vmlinux ;-)
+		 */
+		if (al.map == session->vmlinux_maps[MAP__FUNCTION] &&
+		    RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION])) {
+			pr_err("The %s file can't be used\n",
+			       symbol_conf.vmlinux_name);
+			exit(1);
+		}
+
+		return;
+	}
+
 	syme = symbol__priv(al.sym);
 	if (!syme->skip) {
 		syme->count[counter]++;
@@ -960,9 +985,6 @@ static void event__process_sample(const event_t *self,
 		if (list_empty(&syme->node) || !syme->node.next)
 			__list_insert_active_sym(syme);
 		pthread_mutex_unlock(&active_symbols_lock);
-		if (origin == PERF_RECORD_MISC_USER)
-			++userspace_samples;
-		++samples;
 	}
 }
 
@@ -975,6 +997,10 @@ static int event__process(event_t *event, struct perf_session *session)
 	case PERF_RECORD_MMAP:
 		event__process_mmap(event, session);
 		break;
+	case PERF_RECORD_FORK:
+	case PERF_RECORD_EXIT:
+		event__process_task(event, session);
+		break;
 	default:
 		break;
 	}
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 574a215e800b..0b65779e3c10 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -75,11 +75,8 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 
 	event__parse_sample(event, session->sample_type, &data);
 
-	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
-		event->header.misc,
-		data.pid, data.tid,
-		(void *)(long)data.ip,
-		(long long)data.period);
+	dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc,
+		    data.pid, data.tid, data.ip, data.period);
 
 	thread = perf_session__findnew(session, event->ip.pid);
 	if (thread == NULL) {
@@ -103,22 +100,9 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 	return 0;
 }
 
-static int sample_type_check(struct perf_session *session)
-{
-	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
-		fprintf(stderr,
-			"No trace sample to read. Did you call perf record "
-			"without -R?");
-		return -1;
-	}
-
-	return 0;
-}
-
 static struct perf_event_ops event_ops = {
-	.process_sample_event	= process_sample_event,
-	.process_comm_event	= event__process_comm,
-	.sample_type_check	= sample_type_check,
+	.sample	= process_sample_event,
+	.comm	= event__process_comm,
 };
 
 static int __cmd_trace(struct perf_session *session)
@@ -531,6 +515,8 @@ static const struct option options[] = {
 		     parse_scriptname),
 	OPT_STRING('g', "gen-script", &generate_script_lang, "lang",
 		   "generate perf-trace.xx script in specified language"),
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
 
 	OPT_END()
 };
@@ -592,6 +578,9 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used)
 	if (session == NULL)
 		return -ENOMEM;
 
+	if (!perf_session__has_traces(session, "record -R"))
+		return -EINVAL;
+
 	if (generate_script_lang) {
 		struct stat perf_stat;
 
diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h
index 18035b1f16c7..10fe49e7048a 100644
--- a/tools/perf/builtin.h
+++ b/tools/perf/builtin.h
@@ -16,6 +16,7 @@ extern int check_pager_config(const char *cmd);
 
 extern int cmd_annotate(int argc, const char **argv, const char *prefix);
 extern int cmd_bench(int argc, const char **argv, const char *prefix);
+extern int cmd_buildid_cache(int argc, const char **argv, const char *prefix);
 extern int cmd_buildid_list(int argc, const char **argv, const char *prefix);
 extern int cmd_diff(int argc, const char **argv, const char *prefix);
 extern int cmd_help(int argc, const char **argv, const char *prefix);
@@ -30,5 +31,6 @@ extern int cmd_trace(int argc, const char **argv, const char *prefix);
 extern int cmd_version(int argc, const char **argv, const char *prefix);
 extern int cmd_probe(int argc, const char **argv, const char *prefix);
 extern int cmd_kmem(int argc, const char **argv, const char *prefix);
+extern int cmd_lock(int argc, const char **argv, const char *prefix);
 
 #endif
diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt
index 71dc7c3fe7b2..9afcff2e3ae5 100644
--- a/tools/perf/command-list.txt
+++ b/tools/perf/command-list.txt
@@ -3,7 +3,9 @@
 # command name			category [deprecated] [common]
 #
 perf-annotate			mainporcelain common
+perf-archive			mainporcelain common
 perf-bench			mainporcelain common
+perf-buildid-cache		mainporcelain common
 perf-buildid-list		mainporcelain common
 perf-diff			mainporcelain common
 perf-list			mainporcelain common
diff --git a/tools/perf/design.txt b/tools/perf/design.txt
index 8d0de5130db3..bd0bb1b1279b 100644
--- a/tools/perf/design.txt
+++ b/tools/perf/design.txt
@@ -101,10 +101,10 @@ enum hw_event_ids {
 	 */
 	PERF_COUNT_HW_CPU_CYCLES		= 0,
 	PERF_COUNT_HW_INSTRUCTIONS		= 1,
-	PERF_COUNT_HW_CACHE_REFERENCES	= 2,
+	PERF_COUNT_HW_CACHE_REFERENCES		= 2,
 	PERF_COUNT_HW_CACHE_MISSES		= 3,
 	PERF_COUNT_HW_BRANCH_INSTRUCTIONS	= 4,
-	PERF_COUNT_HW_BRANCH_MISSES	= 5,
+	PERF_COUNT_HW_BRANCH_MISSES		= 5,
 	PERF_COUNT_HW_BUS_CYCLES		= 6,
 };
 
@@ -131,8 +131,8 @@ software events, selected by 'event_id':
  */
 enum sw_event_ids {
 	PERF_COUNT_SW_CPU_CLOCK		= 0,
-	PERF_COUNT_SW_TASK_CLOCK		= 1,
-	PERF_COUNT_SW_PAGE_FAULTS		= 2,
+	PERF_COUNT_SW_TASK_CLOCK	= 1,
+	PERF_COUNT_SW_PAGE_FAULTS	= 2,
 	PERF_COUNT_SW_CONTEXT_SWITCHES	= 3,
 	PERF_COUNT_SW_CPU_MIGRATIONS	= 4,
 	PERF_COUNT_SW_PAGE_FAULTS_MIN	= 5,
diff --git a/tools/perf/perf-archive.sh b/tools/perf/perf-archive.sh
new file mode 100644
index 000000000000..45fbe2f07b15
--- /dev/null
+++ b/tools/perf/perf-archive.sh
@@ -0,0 +1,32 @@
+#!/bin/bash
+# perf archive
+# Arnaldo Carvalho de Melo <acme@redhat.com>
+
+PERF_DATA=perf.data
+if [ $# -ne 0 ] ; then
+	PERF_DATA=$1
+fi
+
+DEBUGDIR=~/.debug/
+BUILDIDS=$(mktemp /tmp/perf-archive-buildids.XXXXXX)
+
+perf buildid-list -i $PERF_DATA --with-hits > $BUILDIDS
+if [ ! -s $BUILDIDS ] ; then
+	echo "perf archive: no build-ids found"
+	rm -f $BUILDIDS
+	exit 1
+fi
+
+MANIFEST=$(mktemp /tmp/perf-archive-manifest.XXXXXX)
+
+cut -d ' ' -f 1 $BUILDIDS | \
+while read build_id ; do
+	linkname=$DEBUGDIR.build-id/${build_id:0:2}/${build_id:2}
+	filename=$(readlink -f $linkname)
+	echo ${linkname#$DEBUGDIR} >> $MANIFEST
+	echo ${filename#$DEBUGDIR} >> $MANIFEST
+done
+
+tar cfj $PERF_DATA.tar.bz2 -C $DEBUGDIR -T $MANIFEST
+rm -f $MANIFEST $BUILDIDS
+exit 0
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index 89eae67a358e..cd32c200cdb3 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -48,7 +48,8 @@ int check_pager_config(const char *cmd)
 	return c.val;
 }
 
-static void commit_pager_choice(void) {
+static void commit_pager_choice(void)
+{
 	switch (use_pager) {
 	case 0:
 		setenv("PERF_PAGER", "cat", 1);
@@ -70,7 +71,7 @@ static void set_debugfs_path(void)
 		 "tracing/events");
 }
 
-static int handle_options(const char*** argv, int* argc, int* envchanged)
+static int handle_options(const char ***argv, int *argc, int *envchanged)
 {
 	int handled = 0;
 
@@ -109,7 +110,7 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
 				*envchanged = 1;
 		} else if (!strcmp(cmd, "--perf-dir")) {
 			if (*argc < 2) {
-				fprintf(stderr, "No directory given for --perf-dir.\n" );
+				fprintf(stderr, "No directory given for --perf-dir.\n");
 				usage(perf_usage_string);
 			}
 			setenv(PERF_DIR_ENVIRONMENT, (*argv)[1], 1);
@@ -124,7 +125,7 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
 				*envchanged = 1;
 		} else if (!strcmp(cmd, "--work-tree")) {
 			if (*argc < 2) {
-				fprintf(stderr, "No directory given for --work-tree.\n" );
+				fprintf(stderr, "No directory given for --work-tree.\n");
 				usage(perf_usage_string);
 			}
 			setenv(PERF_WORK_TREE_ENVIRONMENT, (*argv)[1], 1);
@@ -168,7 +169,7 @@ static int handle_alias(int *argcp, const char ***argv)
 {
 	int envchanged = 0, ret = 0, saved_errno = errno;
 	int count, option_count;
-	const char** new_argv;
+	const char **new_argv;
 	const char *alias_command;
 	char *alias_string;
 
@@ -210,11 +211,11 @@ static int handle_alias(int *argcp, const char ***argv)
 		if (!strcmp(alias_command, new_argv[0]))
 			die("recursive alias: %s", alias_command);
 
-		new_argv = realloc(new_argv, sizeof(char*) *
+		new_argv = realloc(new_argv, sizeof(char *) *
 				    (count + *argcp + 1));
 		/* insert after command name */
-		memcpy(new_argv + count, *argv + 1, sizeof(char*) * *argcp);
-		new_argv[count+*argcp] = NULL;
+		memcpy(new_argv + count, *argv + 1, sizeof(char *) * *argcp);
+		new_argv[count + *argcp] = NULL;
 
 		*argv = new_argv;
 		*argcp += count - 1;
@@ -285,6 +286,7 @@ static void handle_internal_command(int argc, const char **argv)
 {
 	const char *cmd = argv[0];
 	static struct cmd_struct commands[] = {
+		{ "buildid-cache", cmd_buildid_cache, 0 },
 		{ "buildid-list", cmd_buildid_list, 0 },
 		{ "diff",	cmd_diff,	0 },
 		{ "help",	cmd_help,	0 },
@@ -301,6 +303,7 @@ static void handle_internal_command(int argc, const char **argv)
 		{ "sched",	cmd_sched,	0 },
 		{ "probe",	cmd_probe,	0 },
 		{ "kmem",	cmd_kmem,	0 },
+		{ "lock",	cmd_lock,	0 },
 	};
 	unsigned int i;
 	static const char ext[] = STRIP_EXTENSION;
@@ -388,7 +391,7 @@ static int run_argv(int *argcp, const char ***argv)
 /* mini /proc/mounts parser: searching for "^blah /mount/point debugfs" */
 static void get_debugfs_mntpt(void)
 {
-	const char *path = debugfs_find_mountpoint();
+	const char *path = debugfs_mount(NULL);
 
 	if (path)
 		strncpy(debugfs_mntpt, path, sizeof(debugfs_mntpt));
@@ -449,8 +452,8 @@ int main(int argc, const char **argv)
 	setup_path();
 
 	while (1) {
-		static int done_help = 0;
-		static int was_alias = 0;
+		static int done_help;
+		static int was_alias;
 
 		was_alias = run_argv(&argc, &argv);
 		if (errno != ENOENT)
diff --git a/tools/perf/util/data_map.c b/tools/perf/util/data_map.c
deleted file mode 100644
index b557b836de3d..000000000000
--- a/tools/perf/util/data_map.c
+++ /dev/null
@@ -1,252 +0,0 @@
-#include "symbol.h"
-#include "util.h"
-#include "debug.h"
-#include "thread.h"
-#include "session.h"
-
-static int process_event_stub(event_t *event __used,
-			      struct perf_session *session __used)
-{
-	dump_printf(": unhandled!\n");
-	return 0;
-}
-
-static void perf_event_ops__fill_defaults(struct perf_event_ops *handler)
-{
-	if (!handler->process_sample_event)
-		handler->process_sample_event = process_event_stub;
-	if (!handler->process_mmap_event)
-		handler->process_mmap_event = process_event_stub;
-	if (!handler->process_comm_event)
-		handler->process_comm_event = process_event_stub;
-	if (!handler->process_fork_event)
-		handler->process_fork_event = process_event_stub;
-	if (!handler->process_exit_event)
-		handler->process_exit_event = process_event_stub;
-	if (!handler->process_lost_event)
-		handler->process_lost_event = process_event_stub;
-	if (!handler->process_read_event)
-		handler->process_read_event = process_event_stub;
-	if (!handler->process_throttle_event)
-		handler->process_throttle_event = process_event_stub;
-	if (!handler->process_unthrottle_event)
-		handler->process_unthrottle_event = process_event_stub;
-}
-
-static const char *event__name[] = {
-	[0]			 = "TOTAL",
-	[PERF_RECORD_MMAP]	 = "MMAP",
-	[PERF_RECORD_LOST]	 = "LOST",
-	[PERF_RECORD_COMM]	 = "COMM",
-	[PERF_RECORD_EXIT]	 = "EXIT",
-	[PERF_RECORD_THROTTLE]	 = "THROTTLE",
-	[PERF_RECORD_UNTHROTTLE] = "UNTHROTTLE",
-	[PERF_RECORD_FORK]	 = "FORK",
-	[PERF_RECORD_READ]	 = "READ",
-	[PERF_RECORD_SAMPLE]	 = "SAMPLE",
-};
-
-unsigned long event__total[PERF_RECORD_MAX];
-
-void event__print_totals(void)
-{
-	int i;
-	for (i = 0; i < PERF_RECORD_MAX; ++i)
-		pr_info("%10s events: %10ld\n",
-			event__name[i], event__total[i]);
-}
-
-static int process_event(event_t *event, struct perf_session *session,
-			 struct perf_event_ops *ops,
-			 unsigned long offset, unsigned long head)
-{
-	trace_event(event);
-
-	if (event->header.type < PERF_RECORD_MAX) {
-		dump_printf("%p [%p]: PERF_RECORD_%s",
-			    (void *)(offset + head),
-			    (void *)(long)(event->header.size),
-			    event__name[event->header.type]);
-		++event__total[0];
-		++event__total[event->header.type];
-	}
-
-	switch (event->header.type) {
-	case PERF_RECORD_SAMPLE:
-		return ops->process_sample_event(event, session);
-	case PERF_RECORD_MMAP:
-		return ops->process_mmap_event(event, session);
-	case PERF_RECORD_COMM:
-		return ops->process_comm_event(event, session);
-	case PERF_RECORD_FORK:
-		return ops->process_fork_event(event, session);
-	case PERF_RECORD_EXIT:
-		return ops->process_exit_event(event, session);
-	case PERF_RECORD_LOST:
-		return ops->process_lost_event(event, session);
-	case PERF_RECORD_READ:
-		return ops->process_read_event(event, session);
-	case PERF_RECORD_THROTTLE:
-		return ops->process_throttle_event(event, session);
-	case PERF_RECORD_UNTHROTTLE:
-		return ops->process_unthrottle_event(event, session);
-	default:
-		ops->total_unknown++;
-		return -1;
-	}
-}
-
-int perf_header__read_build_ids(int input, u64 offset, u64 size)
-{
-	struct build_id_event bev;
-	char filename[PATH_MAX];
-	u64 limit = offset + size;
-	int err = -1;
-
-	while (offset < limit) {
-		struct dso *dso;
-		ssize_t len;
-
-		if (read(input, &bev, sizeof(bev)) != sizeof(bev))
-			goto out;
-
-		len = bev.header.size - sizeof(bev);
-		if (read(input, filename, len) != len)
-			goto out;
-
-		dso = dsos__findnew(filename);
-		if (dso != NULL)
-			dso__set_build_id(dso, &bev.build_id);
-
-		offset += bev.header.size;
-	}
-	err = 0;
-out:
-	return err;
-}
-
-static struct thread *perf_session__register_idle_thread(struct perf_session *self)
-{
-	struct thread *thread = perf_session__findnew(self, 0);
-
-	if (!thread || thread__set_comm(thread, "swapper")) {
-		pr_err("problem inserting idle task.\n");
-		thread = NULL;
-	}
-
-	return thread;
-}
-
-int perf_session__process_events(struct perf_session *self,
-				 struct perf_event_ops *ops)
-{
-	int err;
-	unsigned long head, shift;
-	unsigned long offset = 0;
-	size_t	page_size;
-	event_t *event;
-	uint32_t size;
-	char *buf;
-
-	if (perf_session__register_idle_thread(self) == NULL)
-		return -ENOMEM;
-
-	perf_event_ops__fill_defaults(ops);
-
-	page_size = getpagesize();
-
-	head = self->header.data_offset;
-	self->sample_type = perf_header__sample_type(&self->header);
-
-	err = -EINVAL;
-	if (ops->sample_type_check && ops->sample_type_check(self) < 0)
-		goto out_err;
-
-	if (!ops->full_paths) {
-		char bf[PATH_MAX];
-
-		if (getcwd(bf, sizeof(bf)) == NULL) {
-			err = -errno;
-out_getcwd_err:
-			pr_err("failed to get the current directory\n");
-			goto out_err;
-		}
-		self->cwd = strdup(bf);
-		if (self->cwd == NULL) {
-			err = -ENOMEM;
-			goto out_getcwd_err;
-		}
-		self->cwdlen = strlen(self->cwd);
-	}
-
-	shift = page_size * (head / page_size);
-	offset += shift;
-	head -= shift;
-
-remap:
-	buf = mmap(NULL, page_size * self->mmap_window, PROT_READ,
-		   MAP_SHARED, self->fd, offset);
-	if (buf == MAP_FAILED) {
-		pr_err("failed to mmap file\n");
-		err = -errno;
-		goto out_err;
-	}
-
-more:
-	event = (event_t *)(buf + head);
-
-	size = event->header.size;
-	if (!size)
-		size = 8;
-
-	if (head + event->header.size >= page_size * self->mmap_window) {
-		int munmap_ret;
-
-		shift = page_size * (head / page_size);
-
-		munmap_ret = munmap(buf, page_size * self->mmap_window);
-		assert(munmap_ret == 0);
-
-		offset += shift;
-		head -= shift;
-		goto remap;
-	}
-
-	size = event->header.size;
-
-	dump_printf("\n%p [%p]: event: %d\n",
-			(void *)(offset + head),
-			(void *)(long)event->header.size,
-			event->header.type);
-
-	if (!size || process_event(event, self, ops, offset, head) < 0) {
-
-		dump_printf("%p [%p]: skipping unknown header type: %d\n",
-			(void *)(offset + head),
-			(void *)(long)(event->header.size),
-			event->header.type);
-
-		/*
-		 * assume we lost track of the stream, check alignment, and
-		 * increment a single u64 in the hope to catch on again 'soon'.
-		 */
-
-		if (unlikely(head & 7))
-			head &= ~7ULL;
-
-		size = 8;
-	}
-
-	head += size;
-
-	if (offset + head >= self->header.data_offset + self->header.data_size)
-		goto done;
-
-	if (offset + head < self->size)
-		goto more;
-
-done:
-	err = 0;
-out_err:
-	return err;
-}
diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c
index 28d520d5a1fb..0905600c3851 100644
--- a/tools/perf/util/debug.c
+++ b/tools/perf/util/debug.c
@@ -9,6 +9,7 @@
 #include "color.h"
 #include "event.h"
 #include "debug.h"
+#include "util.h"
 
 int verbose = 0;
 int dump_trace = 0;
diff --git a/tools/perf/util/debugfs.c b/tools/perf/util/debugfs.c
index 06b73ee02c49..a88fefc0cc0a 100644
--- a/tools/perf/util/debugfs.c
+++ b/tools/perf/util/debugfs.c
@@ -106,16 +106,14 @@ int debugfs_valid_entry(const char *path)
 	return 0;
 }
 
-/* mount the debugfs somewhere */
+/* mount the debugfs somewhere if it's not mounted */
 
-int debugfs_mount(const char *mountpoint)
+char *debugfs_mount(const char *mountpoint)
 {
-	char mountcmd[128];
-
 	/* see if it's already mounted */
 	if (debugfs_find_mountpoint()) {
 		debugfs_premounted = 1;
-		return 0;
+		return debugfs_mountpoint;
 	}
 
 	/* if not mounted and no argument */
@@ -127,13 +125,14 @@ int debugfs_mount(const char *mountpoint)
 			mountpoint = "/sys/kernel/debug";
 	}
 
+	if (mount(NULL, mountpoint, "debugfs", 0, NULL) < 0)
+		return NULL;
+
 	/* save the mountpoint */
 	strncpy(debugfs_mountpoint, mountpoint, sizeof(debugfs_mountpoint));
+	debugfs_found = 1;
 
-	/* mount it */
-	snprintf(mountcmd, sizeof(mountcmd),
-		 "/bin/mount -t debugfs debugfs %s", mountpoint);
-	return system(mountcmd);
+	return debugfs_mountpoint;
 }
 
 /* umount the debugfs */
diff --git a/tools/perf/util/debugfs.h b/tools/perf/util/debugfs.h
index 3cd14f9ae784..83a02879745f 100644
--- a/tools/perf/util/debugfs.h
+++ b/tools/perf/util/debugfs.h
@@ -15,7 +15,7 @@
 extern const char *debugfs_find_mountpoint(void);
 extern int debugfs_valid_mountpoint(const char *debugfs);
 extern int debugfs_valid_entry(const char *path);
-extern int debugfs_mount(const char *mountpoint);
+extern char *debugfs_mount(const char *mountpoint);
 extern int debugfs_umount(void);
 extern int debugfs_write(const char *entry, const char *value);
 extern int debugfs_read(const char *entry, char *buffer, size_t size);
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index bb0fd6da2d56..bbaee61c1683 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -8,8 +8,7 @@
 #include "thread.h"
 
 static pid_t event__synthesize_comm(pid_t pid, int full,
-				    int (*process)(event_t *event,
-						   struct perf_session *session),
+				    event__handler_t process,
 				    struct perf_session *session)
 {
 	event_t ev;
@@ -91,8 +90,7 @@ out_failure:
 }
 
 static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
-					 int (*process)(event_t *event,
-							struct perf_session *session),
+					 event__handler_t process,
 					 struct perf_session *session)
 {
 	char filename[PATH_MAX];
@@ -112,7 +110,10 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
 	while (1) {
 		char bf[BUFSIZ], *pbf = bf;
 		event_t ev = {
-			.header = { .type = PERF_RECORD_MMAP },
+			.header = {
+				.type = PERF_RECORD_MMAP,
+				.misc = 0, /* Just like the kernel, see kernel/perf_event.c __perf_event_mmap */
+			 },
 		};
 		int n;
 		size_t size;
@@ -156,9 +157,38 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
 	return 0;
 }
 
-int event__synthesize_thread(pid_t pid,
-			     int (*process)(event_t *event,
-					    struct perf_session *session),
+int event__synthesize_modules(event__handler_t process,
+			      struct perf_session *session)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&session->kmaps.maps[MAP__FUNCTION]);
+	     nd; nd = rb_next(nd)) {
+		event_t ev;
+		size_t size;
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+
+		if (pos->dso->kernel)
+			continue;
+
+		size = ALIGN(pos->dso->long_name_len + 1, sizeof(u64));
+		memset(&ev, 0, sizeof(ev));
+		ev.mmap.header.misc = 1; /* kernel uses 0 for user space maps, see kernel/perf_event.c __perf_event_mmap */
+		ev.mmap.header.type = PERF_RECORD_MMAP;
+		ev.mmap.header.size = (sizeof(ev.mmap) -
+				        (sizeof(ev.mmap.filename) - size));
+		ev.mmap.start = pos->start;
+		ev.mmap.len   = pos->end - pos->start;
+
+		memcpy(ev.mmap.filename, pos->dso->long_name,
+		       pos->dso->long_name_len + 1);
+		process(&ev, session);
+	}
+
+	return 0;
+}
+
+int event__synthesize_thread(pid_t pid, event__handler_t process,
 			     struct perf_session *session)
 {
 	pid_t tgid = event__synthesize_comm(pid, 1, process, session);
@@ -167,8 +197,7 @@ int event__synthesize_thread(pid_t pid,
 	return event__synthesize_mmap_events(pid, tgid, process, session);
 }
 
-void event__synthesize_threads(int (*process)(event_t *event,
-					      struct perf_session *session),
+void event__synthesize_threads(event__handler_t process,
 			       struct perf_session *session)
 {
 	DIR *proc;
@@ -189,6 +218,59 @@ void event__synthesize_threads(int (*process)(event_t *event,
 	closedir(proc);
 }
 
+struct process_symbol_args {
+	const char *name;
+	u64	   start;
+};
+
+static int find_symbol_cb(void *arg, const char *name, char type, u64 start)
+{
+	struct process_symbol_args *args = arg;
+
+	/*
+	 * Must be a function or at least an alias, as in PARISC64, where "_text" is
+	 * an 'A' to the same address as "_stext".
+	 */
+	if (!(symbol_type__is_a(type, MAP__FUNCTION) ||
+	      type == 'A') || strcmp(name, args->name))
+		return 0;
+
+	args->start = start;
+	return 1;
+}
+
+int event__synthesize_kernel_mmap(event__handler_t process,
+				  struct perf_session *session,
+				  const char *symbol_name)
+{
+	size_t size;
+	event_t ev = {
+		.header = {
+			.type = PERF_RECORD_MMAP,
+			.misc = 1, /* kernel uses 0 for user space maps, see kernel/perf_event.c __perf_event_mmap */
+		},
+	};
+	/*
+	 * We should get this from /sys/kernel/sections/.text, but till that is
+	 * available use this, and after it is use this as a fallback for older
+	 * kernels.
+	 */
+	struct process_symbol_args args = { .name = symbol_name, };
+
+	if (kallsyms__parse("/proc/kallsyms", &args, find_symbol_cb) <= 0)
+		return -ENOENT;
+
+	size = snprintf(ev.mmap.filename, sizeof(ev.mmap.filename),
+			"[kernel.kallsyms.%s]", symbol_name) + 1;
+	size = ALIGN(size, sizeof(u64));
+	ev.mmap.header.size = (sizeof(ev.mmap) - (sizeof(ev.mmap.filename) - size));
+	ev.mmap.pgoff = args.start;
+	ev.mmap.start = session->vmlinux_maps[MAP__FUNCTION]->start;
+	ev.mmap.len   = session->vmlinux_maps[MAP__FUNCTION]->end - ev.mmap.start ;
+
+	return process(&ev, session);
+}
+
 static void thread__comm_adjust(struct thread *self)
 {
 	char *comm = self->comm;
@@ -240,22 +322,84 @@ int event__process_lost(event_t *self, struct perf_session *session)
 
 int event__process_mmap(event_t *self, struct perf_session *session)
 {
-	struct thread *thread = perf_session__findnew(session, self->mmap.pid);
-	struct map *map = map__new(&self->mmap, MAP__FUNCTION,
-				   session->cwd, session->cwdlen);
+	struct thread *thread;
+	struct map *map;
+
+	dump_printf(" %d/%d: [%#Lx(%#Lx) @ %#Lx]: %s\n",
+		    self->mmap.pid, self->mmap.tid, self->mmap.start,
+		    self->mmap.len, self->mmap.pgoff, self->mmap.filename);
+
+	if (self->mmap.pid == 0) {
+		static const char kmmap_prefix[] = "[kernel.kallsyms.";
+
+		if (self->mmap.filename[0] == '/') {
+			char short_module_name[1024];
+			char *name = strrchr(self->mmap.filename, '/'), *dot;
+
+			if (name == NULL)
+				goto out_problem;
+
+			++name; /* skip / */
+			dot = strrchr(name, '.');
+			if (dot == NULL)
+				goto out_problem;
+
+			snprintf(short_module_name, sizeof(short_module_name),
+				 "[%.*s]", (int)(dot - name), name);
+			strxfrchar(short_module_name, '-', '_');
+
+			map = perf_session__new_module_map(session,
+							   self->mmap.start,
+							   self->mmap.filename);
+			if (map == NULL)
+				goto out_problem;
+
+			name = strdup(short_module_name);
+			if (name == NULL)
+				goto out_problem;
+
+			map->dso->short_name = name;
+			map->end = map->start + self->mmap.len;
+		} else if (memcmp(self->mmap.filename, kmmap_prefix,
+				sizeof(kmmap_prefix) - 1) == 0) {
+			const char *symbol_name = (self->mmap.filename +
+						   sizeof(kmmap_prefix) - 1);
+			/*
+			 * Should be there already, from the build-id table in
+			 * the header.
+			 */
+			struct dso *kernel = __dsos__findnew(&dsos__kernel,
+							     "[kernel.kallsyms]");
+			if (kernel == NULL)
+				goto out_problem;
+
+			kernel->kernel = 1;
+			if (__map_groups__create_kernel_maps(&session->kmaps,
+							     session->vmlinux_maps,
+							     kernel) < 0)
+				goto out_problem;
+
+			session->vmlinux_maps[MAP__FUNCTION]->start = self->mmap.start;
+			session->vmlinux_maps[MAP__FUNCTION]->end   = self->mmap.start + self->mmap.len;
+
+			perf_session__set_kallsyms_ref_reloc_sym(session, symbol_name,
+								 self->mmap.pgoff);
+		}
+		return 0;
+	}
 
-	dump_printf(" %d/%d: [%p(%p) @ %p]: %s\n",
-		    self->mmap.pid, self->mmap.tid,
-		    (void *)(long)self->mmap.start,
-		    (void *)(long)self->mmap.len,
-		    (void *)(long)self->mmap.pgoff,
-		    self->mmap.filename);
+	thread = perf_session__findnew(session, self->mmap.pid);
+	map = map__new(&self->mmap, MAP__FUNCTION,
+		       session->cwd, session->cwdlen);
 
 	if (thread == NULL || map == NULL)
-		dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
-	else
-		thread__insert_map(thread, map);
+		goto out_problem;
 
+	thread__insert_map(thread, map);
+	return 0;
+
+out_problem:
+	dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
 	return 0;
 }
 
@@ -284,11 +428,10 @@ int event__process_task(event_t *self, struct perf_session *session)
 	return 0;
 }
 
-void thread__find_addr_location(struct thread *self,
-				struct perf_session *session, u8 cpumode,
-				enum map_type type, u64 addr,
-				struct addr_location *al,
-				symbol_filter_t filter)
+void thread__find_addr_map(struct thread *self,
+			   struct perf_session *session, u8 cpumode,
+			   enum map_type type, u64 addr,
+			   struct addr_location *al)
 {
 	struct map_groups *mg = &self->mg;
 
@@ -303,7 +446,6 @@ void thread__find_addr_location(struct thread *self,
 	else {
 		al->level = 'H';
 		al->map = NULL;
-		al->sym = NULL;
 		return;
 	}
 try_again:
@@ -322,11 +464,21 @@ try_again:
 			mg = &session->kmaps;
 			goto try_again;
 		}
-		al->sym = NULL;
-	} else {
+	} else
 		al->addr = al->map->map_ip(al->map, al->addr);
+}
+
+void thread__find_addr_location(struct thread *self,
+				struct perf_session *session, u8 cpumode,
+				enum map_type type, u64 addr,
+				struct addr_location *al,
+				symbol_filter_t filter)
+{
+	thread__find_addr_map(self, session, cpumode, type, addr, al);
+	if (al->map != NULL)
 		al->sym = map__find_symbol(al->map, session, al->addr, filter);
-	}
+	else
+		al->sym = NULL;
 }
 
 static void dso__calc_col_width(struct dso *self)
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index 690a96d0467c..50a7132887f5 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -1,10 +1,10 @@
 #ifndef __PERF_RECORD_H
 #define __PERF_RECORD_H
 
+#include <limits.h>
+
 #include "../perf.h"
-#include "util.h"
-#include <linux/list.h>
-#include <linux/rbtree.h>
+#include "map.h"
 
 /*
  * PERF_SAMPLE_IP | PERF_SAMPLE_TID | *
@@ -101,74 +101,19 @@ struct events_stats {
 
 void event__print_totals(void);
 
-enum map_type {
-	MAP__FUNCTION = 0,
-	MAP__VARIABLE,
-};
-
-#define MAP__NR_TYPES (MAP__VARIABLE + 1)
-
-struct map {
-	union {
-		struct rb_node	rb_node;
-		struct list_head node;
-	};
-	u64			start;
-	u64			end;
-	enum map_type		type;
-	u64			pgoff;
-	u64			(*map_ip)(struct map *, u64);
-	u64			(*unmap_ip)(struct map *, u64);
-	struct dso		*dso;
-};
-
-static inline u64 map__map_ip(struct map *map, u64 ip)
-{
-	return ip - map->start + map->pgoff;
-}
-
-static inline u64 map__unmap_ip(struct map *map, u64 ip)
-{
-	return ip + map->start - map->pgoff;
-}
-
-static inline u64 identity__map_ip(struct map *map __used, u64 ip)
-{
-	return ip;
-}
-
-struct symbol;
-
-typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
-
-void map__init(struct map *self, enum map_type type,
-	       u64 start, u64 end, u64 pgoff, struct dso *dso);
-struct map *map__new(struct mmap_event *event, enum map_type,
-		     char *cwd, int cwdlen);
-void map__delete(struct map *self);
-struct map *map__clone(struct map *self);
-int map__overlap(struct map *l, struct map *r);
-size_t map__fprintf(struct map *self, FILE *fp);
-
 struct perf_session;
 
-int map__load(struct map *self, struct perf_session *session,
-	      symbol_filter_t filter);
-struct symbol *map__find_symbol(struct map *self, struct perf_session *session,
-				u64 addr, symbol_filter_t filter);
-struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
-					struct perf_session *session,
-					symbol_filter_t filter);
-void map__fixup_start(struct map *self);
-void map__fixup_end(struct map *self);
-
-int event__synthesize_thread(pid_t pid,
-			     int (*process)(event_t *event,
-					    struct perf_session *session),
+typedef int (*event__handler_t)(event_t *event, struct perf_session *session);
+
+int event__synthesize_thread(pid_t pid, event__handler_t process,
 			     struct perf_session *session);
-void event__synthesize_threads(int (*process)(event_t *event,
-					      struct perf_session *session),
+void event__synthesize_threads(event__handler_t process,
 			       struct perf_session *session);
+int event__synthesize_kernel_mmap(event__handler_t process,
+				  struct perf_session *session,
+				  const char *symbol_name);
+int event__synthesize_modules(event__handler_t process,
+			      struct perf_session *session);
 
 int event__process_comm(event_t *self, struct perf_session *session);
 int event__process_lost(event_t *self, struct perf_session *session);
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 8a0bca55106f..ed3efd728b41 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -1,8 +1,13 @@
+#define _LARGEFILE64_SOURCE
+#define _FILE_OFFSET_BITS 64
+
 #include <sys/types.h>
+#include <byteswap.h>
 #include <unistd.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <linux/list.h>
+#include <linux/kernel.h>
 
 #include "util.h"
 #include "header.h"
@@ -105,24 +110,28 @@ struct perf_trace_event_type {
 static int event_count;
 static struct perf_trace_event_type *events;
 
-void perf_header__push_event(u64 id, const char *name)
+int perf_header__push_event(u64 id, const char *name)
 {
 	if (strlen(name) > MAX_EVENT_NAME)
 		pr_warning("Event %s will be truncated\n", name);
 
 	if (!events) {
 		events = malloc(sizeof(struct perf_trace_event_type));
-		if (!events)
-			die("nomem");
+		if (events == NULL)
+			return -ENOMEM;
 	} else {
-		events = realloc(events, (event_count + 1) * sizeof(struct perf_trace_event_type));
-		if (!events)
-			die("nomem");
+		struct perf_trace_event_type *nevents;
+
+		nevents = realloc(events, (event_count + 1) * sizeof(*events));
+		if (nevents == NULL)
+			return -ENOMEM;
+		events = nevents;
 	}
 	memset(&events[event_count], 0, sizeof(struct perf_trace_event_type));
 	events[event_count].event_id = id;
 	strncpy(events[event_count].name, name, MAX_EVENT_NAME - 1);
 	event_count++;
+	return 0;
 }
 
 char *perf_header__find_event(u64 id)
@@ -169,31 +178,45 @@ static int do_write(int fd, const void *buf, size_t size)
 	return 0;
 }
 
-static int __dsos__write_buildid_table(struct list_head *head, int fd)
+#define NAME_ALIGN 64
+
+static int write_padded(int fd, const void *bf, size_t count,
+			size_t count_aligned)
 {
-#define NAME_ALIGN	64
-	struct dso *pos;
 	static const char zero_buf[NAME_ALIGN];
+	int err = do_write(fd, bf, count);
+
+	if (!err)
+		err = do_write(fd, zero_buf, count_aligned - count);
+
+	return err;
+}
 
-	list_for_each_entry(pos, head, node) {
+#define dsos__for_each_with_build_id(pos, head)	\
+	list_for_each_entry(pos, head, node)	\
+		if (!pos->has_build_id)		\
+			continue;		\
+		else
+
+static int __dsos__write_buildid_table(struct list_head *head, u16 misc, int fd)
+{
+	struct dso *pos;
+
+	dsos__for_each_with_build_id(pos, head) {
 		int err;
 		struct build_id_event b;
-		size_t len;
+		size_t len = pos->long_name_len + 1;
 
-		if (!pos->has_build_id)
-			continue;
-		len = pos->long_name_len + 1;
 		len = ALIGN(len, NAME_ALIGN);
 		memset(&b, 0, sizeof(b));
 		memcpy(&b.build_id, pos->build_id, sizeof(pos->build_id));
+		b.header.misc = misc;
 		b.header.size = sizeof(b) + len;
 		err = do_write(fd, &b, sizeof(b));
 		if (err < 0)
 			return err;
-		err = do_write(fd, pos->long_name, pos->long_name_len + 1);
-		if (err < 0)
-			return err;
-		err = do_write(fd, zero_buf, len - pos->long_name_len - 1);
+		err = write_padded(fd, pos->long_name,
+				   pos->long_name_len + 1, len);
 		if (err < 0)
 			return err;
 	}
@@ -203,12 +226,143 @@ static int __dsos__write_buildid_table(struct list_head *head, int fd)
 
 static int dsos__write_buildid_table(int fd)
 {
-	int err = __dsos__write_buildid_table(&dsos__kernel, fd);
+	int err = __dsos__write_buildid_table(&dsos__kernel,
+					      PERF_RECORD_MISC_KERNEL, fd);
 	if (err == 0)
-		err = __dsos__write_buildid_table(&dsos__user, fd);
+		err = __dsos__write_buildid_table(&dsos__user,
+						  PERF_RECORD_MISC_USER, fd);
 	return err;
 }
 
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+			  const char *name, bool is_kallsyms)
+{
+	const size_t size = PATH_MAX;
+	char *filename = malloc(size),
+	     *linkname = malloc(size), *targetname;
+	int len, err = -1;
+
+	if (filename == NULL || linkname == NULL)
+		goto out_free;
+
+	len = snprintf(filename, size, "%s%s%s",
+		       debugdir, is_kallsyms ? "/" : "", name);
+	if (mkdir_p(filename, 0755))
+		goto out_free;
+
+	snprintf(filename + len, sizeof(filename) - len, "/%s", sbuild_id);
+
+	if (access(filename, F_OK)) {
+		if (is_kallsyms) {
+			 if (copyfile("/proc/kallsyms", filename))
+				goto out_free;
+		} else if (link(name, filename) && copyfile(name, filename))
+			goto out_free;
+	}
+
+	len = snprintf(linkname, size, "%s/.build-id/%.2s",
+		       debugdir, sbuild_id);
+
+	if (access(linkname, X_OK) && mkdir_p(linkname, 0755))
+		goto out_free;
+
+	snprintf(linkname + len, size - len, "/%s", sbuild_id + 2);
+	targetname = filename + strlen(debugdir) - 5;
+	memcpy(targetname, "../..", 5);
+
+	if (symlink(targetname, linkname) == 0)
+		err = 0;
+out_free:
+	free(filename);
+	free(linkname);
+	return err;
+}
+
+static int build_id_cache__add_b(const u8 *build_id, size_t build_id_size,
+				 const char *name, const char *debugdir,
+				 bool is_kallsyms)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	build_id__sprintf(build_id, build_id_size, sbuild_id);
+
+	return build_id_cache__add_s(sbuild_id, debugdir, name, is_kallsyms);
+}
+
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir)
+{
+	const size_t size = PATH_MAX;
+	char *filename = malloc(size),
+	     *linkname = malloc(size);
+	int err = -1;
+
+	if (filename == NULL || linkname == NULL)
+		goto out_free;
+
+	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+		 debugdir, sbuild_id, sbuild_id + 2);
+
+	if (access(linkname, F_OK))
+		goto out_free;
+
+	if (readlink(linkname, filename, size) < 0)
+		goto out_free;
+
+	if (unlink(linkname))
+		goto out_free;
+
+	/*
+	 * Since the link is relative, we must make it absolute:
+	 */
+	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+		 debugdir, sbuild_id, filename);
+
+	if (unlink(linkname))
+		goto out_free;
+
+	err = 0;
+out_free:
+	free(filename);
+	free(linkname);
+	return err;
+}
+
+static int dso__cache_build_id(struct dso *self, const char *debugdir)
+{
+	bool is_kallsyms = self->kernel && self->long_name[0] != '/';
+
+	return build_id_cache__add_b(self->build_id, sizeof(self->build_id),
+				     self->long_name, debugdir, is_kallsyms);
+}
+
+static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir)
+{
+	struct dso *pos;
+	int err = 0;
+
+	dsos__for_each_with_build_id(pos, head)
+		if (dso__cache_build_id(pos, debugdir))
+			err = -1;
+
+	return err;
+}
+
+static int dsos__cache_build_ids(void)
+{
+	int err_kernel, err_user;
+	char debugdir[PATH_MAX];
+
+	snprintf(debugdir, sizeof(debugdir), "%s/%s", getenv("HOME"),
+		 DEBUG_CACHE_DIR);
+
+	if (mkdir(debugdir, 0755) != 0 && errno != EEXIST)
+		return -1;
+
+	err_kernel = __dsos__cache_build_ids(&dsos__kernel, debugdir);
+	err_user   = __dsos__cache_build_ids(&dsos__user, debugdir);
+	return err_kernel || err_user ? -1 : 0;
+}
+
 static int perf_header__adds_write(struct perf_header *self, int fd)
 {
 	int nr_sections;
@@ -231,7 +385,7 @@ static int perf_header__adds_write(struct perf_header *self, int fd)
 	sec_size = sizeof(*feat_sec) * nr_sections;
 
 	sec_start = self->data_offset + self->data_size;
-	lseek(fd, sec_start + sec_size, SEEK_SET);
+	lseek64(fd, sec_start + sec_size, SEEK_SET);
 
 	if (perf_header__has_feat(self, HEADER_TRACE_INFO)) {
 		struct perf_file_section *trace_sec;
@@ -239,9 +393,9 @@ static int perf_header__adds_write(struct perf_header *self, int fd)
 		trace_sec = &feat_sec[idx++];
 
 		/* Write trace info */
-		trace_sec->offset = lseek(fd, 0, SEEK_CUR);
+		trace_sec->offset = lseek64(fd, 0, SEEK_CUR);
 		read_tracing_data(fd, attrs, nr_counters);
-		trace_sec->size = lseek(fd, 0, SEEK_CUR) - trace_sec->offset;
+		trace_sec->size = lseek64(fd, 0, SEEK_CUR) - trace_sec->offset;
 	}
 
 
@@ -251,16 +405,18 @@ static int perf_header__adds_write(struct perf_header *self, int fd)
 		buildid_sec = &feat_sec[idx++];
 
 		/* Write build-ids */
-		buildid_sec->offset = lseek(fd, 0, SEEK_CUR);
+		buildid_sec->offset = lseek64(fd, 0, SEEK_CUR);
 		err = dsos__write_buildid_table(fd);
 		if (err < 0) {
 			pr_debug("failed to write buildid table\n");
 			goto out_free;
 		}
-		buildid_sec->size = lseek(fd, 0, SEEK_CUR) - buildid_sec->offset;
+		buildid_sec->size = lseek64(fd, 0, SEEK_CUR) -
+					    buildid_sec->offset;
+		dsos__cache_build_ids();
 	}
 
-	lseek(fd, sec_start, SEEK_SET);
+	lseek64(fd, sec_start, SEEK_SET);
 	err = do_write(fd, feat_sec, sec_size);
 	if (err < 0)
 		pr_debug("failed to write feature section\n");
@@ -354,36 +510,49 @@ int perf_header__write(struct perf_header *self, int fd, bool at_exit)
 		pr_debug("failed to write perf header\n");
 		return err;
 	}
-	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+	lseek64(fd, self->data_offset + self->data_size, SEEK_SET);
 
 	self->frozen = 1;
 	return 0;
 }
 
-static void do_read(int fd, void *buf, size_t size)
+static int do_read(int fd, void *buf, size_t size)
 {
 	while (size) {
 		int ret = read(fd, buf, size);
 
-		if (ret < 0)
-			die("failed to read");
-		if (ret == 0)
-			die("failed to read: missing data");
+		if (ret <= 0)
+			return -1;
 
 		size -= ret;
 		buf += ret;
 	}
+
+	return 0;
+}
+
+static int perf_header__getbuffer64(struct perf_header *self,
+				    int fd, void *buf, size_t size)
+{
+	if (do_read(fd, buf, size))
+		return -1;
+
+	if (self->needs_swap)
+		mem_bswap_64(buf, size);
+
+	return 0;
 }
 
 int perf_header__process_sections(struct perf_header *self, int fd,
 				  int (*process)(struct perf_file_section *self,
+						 struct perf_header *ph,
 						 int feat, int fd))
 {
 	struct perf_file_section *feat_sec;
 	int nr_sections;
 	int sec_size;
 	int idx = 0;
-	int err = 0, feat = 1;
+	int err = -1, feat = 1;
 
 	nr_sections = bitmap_weight(self->adds_features, HEADER_FEAT_BITS);
 	if (!nr_sections)
@@ -395,35 +564,47 @@ int perf_header__process_sections(struct perf_header *self, int fd,
 
 	sec_size = sizeof(*feat_sec) * nr_sections;
 
-	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+	lseek64(fd, self->data_offset + self->data_size, SEEK_SET);
 
-	do_read(fd, feat_sec, sec_size);
+	if (perf_header__getbuffer64(self, fd, feat_sec, sec_size))
+		goto out_free;
 
+	err = 0;
 	while (idx < nr_sections && feat < HEADER_LAST_FEATURE) {
 		if (perf_header__has_feat(self, feat)) {
 			struct perf_file_section *sec = &feat_sec[idx++];
 
-			err = process(sec, feat, fd);
+			err = process(sec, self, feat, fd);
 			if (err < 0)
 				break;
 		}
 		++feat;
 	}
-
+out_free:
 	free(feat_sec);
 	return err;
-};
+}
 
 int perf_file_header__read(struct perf_file_header *self,
 			   struct perf_header *ph, int fd)
 {
 	lseek(fd, 0, SEEK_SET);
-	do_read(fd, self, sizeof(*self));
 
-	if (self->magic     != PERF_MAGIC ||
-	    self->attr_size != sizeof(struct perf_file_attr))
+	if (do_read(fd, self, sizeof(*self)) ||
+	    memcmp(&self->magic, __perf_magic, sizeof(self->magic)))
 		return -1;
 
+	if (self->attr_size != sizeof(struct perf_file_attr)) {
+		u64 attr_size = bswap_64(self->attr_size);
+
+		if (attr_size != sizeof(struct perf_file_attr))
+			return -1;
+
+		mem_bswap_64(self, offsetof(struct perf_file_header,
+					    adds_features));
+		ph->needs_swap = true;
+	}
+
 	if (self->size != sizeof(*self)) {
 		/* Support the previous format */
 		if (self->size == offsetof(typeof(*self), adds_features))
@@ -433,19 +614,31 @@ int perf_file_header__read(struct perf_file_header *self,
 	}
 
 	memcpy(&ph->adds_features, &self->adds_features,
-	       sizeof(self->adds_features));
+	       sizeof(ph->adds_features));
+	/*
+	 * FIXME: hack that assumes that if we need swap the perf.data file
+	 * may be coming from an arch with a different word-size, ergo different
+	 * DEFINE_BITMAP format, investigate more later, but for now its mostly
+	 * safe to assume that we have a build-id section. Trace files probably
+	 * have several other issues in this realm anyway...
+	 */
+	if (ph->needs_swap) {
+		memset(&ph->adds_features, 0, sizeof(ph->adds_features));
+		perf_header__set_feat(ph, HEADER_BUILD_ID);
+	}
 
 	ph->event_offset = self->event_types.offset;
-	ph->event_size	 = self->event_types.size;
-	ph->data_offset	 = self->data.offset;
+	ph->event_size   = self->event_types.size;
+	ph->data_offset  = self->data.offset;
 	ph->data_size	 = self->data.size;
 	return 0;
 }
 
 static int perf_file_section__process(struct perf_file_section *self,
+				      struct perf_header *ph,
 				      int feat, int fd)
 {
-	if (lseek(fd, self->offset, SEEK_SET) < 0) {
+	if (lseek64(fd, self->offset, SEEK_SET) < 0) {
 		pr_debug("Failed to lseek to %Ld offset for feature %d, "
 			 "continuing...\n", self->offset, feat);
 		return 0;
@@ -457,7 +650,7 @@ static int perf_file_section__process(struct perf_file_section *self,
 		break;
 
 	case HEADER_BUILD_ID:
-		if (perf_header__read_build_ids(fd, self->offset, self->size))
+		if (perf_header__read_build_ids(ph, fd, self->offset, self->size))
 			pr_debug("Failed to read buildids, continuing...\n");
 		break;
 	default:
@@ -469,7 +662,7 @@ static int perf_file_section__process(struct perf_file_section *self,
 
 int perf_header__read(struct perf_header *self, int fd)
 {
-	struct perf_file_header f_header;
+	struct perf_file_header	f_header;
 	struct perf_file_attr	f_attr;
 	u64			f_id;
 	int nr_attrs, nr_ids, i, j;
@@ -486,7 +679,9 @@ int perf_header__read(struct perf_header *self, int fd)
 		struct perf_header_attr *attr;
 		off_t tmp;
 
-		do_read(fd, &f_attr, sizeof(f_attr));
+		if (perf_header__getbuffer64(self, fd, &f_attr, sizeof(f_attr)))
+			goto out_errno;
+
 		tmp = lseek(fd, 0, SEEK_CUR);
 
 		attr = perf_header_attr__new(&f_attr.attr);
@@ -497,7 +692,8 @@ int perf_header__read(struct perf_header *self, int fd)
 		lseek(fd, f_attr.ids.offset, SEEK_SET);
 
 		for (j = 0; j < nr_ids; j++) {
-			do_read(fd, &f_id, sizeof(f_id));
+			if (perf_header__getbuffer64(self, fd, &f_id, sizeof(f_id)))
+				goto out_errno;
 
 			if (perf_header_attr__add_id(attr, f_id) < 0) {
 				perf_header_attr__delete(attr);
@@ -517,7 +713,9 @@ int perf_header__read(struct perf_header *self, int fd)
 		events = malloc(f_header.event_types.size);
 		if (events == NULL)
 			return -ENOMEM;
-		do_read(fd, events, f_header.event_types.size);
+		if (perf_header__getbuffer64(self, fd, events,
+					     f_header.event_types.size))
+			goto out_errno;
 		event_count =  f_header.event_types.size / sizeof(struct perf_trace_event_type);
 	}
 
@@ -527,6 +725,8 @@ int perf_header__read(struct perf_header *self, int fd)
 
 	self->frozen = 1;
 	return 0;
+out_errno:
+	return -errno;
 }
 
 u64 perf_header__sample_type(struct perf_header *header)
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
index d118d05d3abe..82a6af72d4cc 100644
--- a/tools/perf/util/header.h
+++ b/tools/perf/util/header.h
@@ -5,6 +5,7 @@
 #include <sys/types.h>
 #include <stdbool.h>
 #include "types.h"
+#include "event.h"
 
 #include <linux/bitmap.h>
 
@@ -52,6 +53,7 @@ struct perf_header {
 	u64			data_size;
 	u64			event_offset;
 	u64			event_size;
+	bool			needs_swap;
 	DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
 };
 
@@ -64,7 +66,7 @@ int perf_header__write(struct perf_header *self, int fd, bool at_exit);
 int perf_header__add_attr(struct perf_header *self,
 			  struct perf_header_attr *attr);
 
-void perf_header__push_event(u64 id, const char *name);
+int perf_header__push_event(u64 id, const char *name);
 char *perf_header__find_event(u64 id);
 
 struct perf_header_attr *perf_header_attr__new(struct perf_event_attr *attr);
@@ -80,6 +82,11 @@ bool perf_header__has_feat(const struct perf_header *self, int feat);
 
 int perf_header__process_sections(struct perf_header *self, int fd,
 				  int (*process)(struct perf_file_section *self,
+						 struct perf_header *ph,
 						 int feat, int fd));
 
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+			  const char *name, bool is_kallsyms);
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir);
+
 #endif /* __PERF_HEADER_H */
diff --git a/tools/perf/util/include/linux/hash.h b/tools/perf/util/include/linux/hash.h
new file mode 100644
index 000000000000..201f57397997
--- /dev/null
+++ b/tools/perf/util/include/linux/hash.h
@@ -0,0 +1,5 @@
+#include "../../../../include/linux/hash.h"
+
+#ifndef PERF_HASH_H
+#define PERF_HASH_H
+#endif
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
new file mode 100644
index 000000000000..72f0b6ab5ea5
--- /dev/null
+++ b/tools/perf/util/map.h
@@ -0,0 +1,73 @@
+#ifndef __PERF_MAP_H
+#define __PERF_MAP_H
+
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <linux/types.h>
+
+enum map_type {
+	MAP__FUNCTION = 0,
+	MAP__VARIABLE,
+};
+
+#define MAP__NR_TYPES (MAP__VARIABLE + 1)
+
+struct dso;
+
+struct map {
+	union {
+		struct rb_node	rb_node;
+		struct list_head node;
+	};
+	u64			start;
+	u64			end;
+	enum map_type		type;
+	u64			pgoff;
+	u64			(*map_ip)(struct map *, u64);
+	u64			(*unmap_ip)(struct map *, u64);
+	struct dso		*dso;
+};
+
+static inline u64 map__map_ip(struct map *map, u64 ip)
+{
+	return ip - map->start + map->pgoff;
+}
+
+static inline u64 map__unmap_ip(struct map *map, u64 ip)
+{
+	return ip + map->start - map->pgoff;
+}
+
+static inline u64 identity__map_ip(struct map *map __used, u64 ip)
+{
+	return ip;
+}
+
+struct symbol;
+struct mmap_event;
+
+typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
+
+void map__init(struct map *self, enum map_type type,
+	       u64 start, u64 end, u64 pgoff, struct dso *dso);
+struct map *map__new(struct mmap_event *event, enum map_type,
+		     char *cwd, int cwdlen);
+void map__delete(struct map *self);
+struct map *map__clone(struct map *self);
+int map__overlap(struct map *l, struct map *r);
+size_t map__fprintf(struct map *self, FILE *fp);
+
+struct perf_session;
+
+int map__load(struct map *self, struct perf_session *session,
+	      symbol_filter_t filter);
+struct symbol *map__find_symbol(struct map *self, struct perf_session *session,
+				u64 addr, symbol_filter_t filter);
+struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
+					struct perf_session *session,
+					symbol_filter_t filter);
+void map__fixup_start(struct map *self);
+void map__fixup_end(struct map *self);
+
+#endif /* __PERF_MAP_H */
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index e5bc0fb016b2..05d0c5c2030c 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -450,7 +450,8 @@ parse_single_tracepoint_event(char *sys_name,
 /* sys + ':' + event + ':' + flags*/
 #define MAX_EVOPT_LEN	(MAX_EVENT_LENGTH * 2 + 2 + 128)
 static enum event_result
-parse_subsystem_tracepoint_event(char *sys_name, char *flags)
+parse_multiple_tracepoint_event(char *sys_name, const char *evt_exp,
+				char *flags)
 {
 	char evt_path[MAXPATHLEN];
 	struct dirent *evt_ent;
@@ -474,6 +475,9 @@ parse_subsystem_tracepoint_event(char *sys_name, char *flags)
 		    || !strcmp(evt_ent->d_name, "filter"))
 			continue;
 
+		if (!strglobmatch(evt_ent->d_name, evt_exp))
+			continue;
+
 		len = snprintf(event_opt, MAX_EVOPT_LEN, "%s:%s%s%s", sys_name,
 			       evt_ent->d_name, flags ? ":" : "",
 			       flags ?: "");
@@ -522,9 +526,10 @@ static enum event_result parse_tracepoint_event(const char **strp,
 	if (evt_length >= MAX_EVENT_LENGTH)
 		return EVT_FAILED;
 
-	if (!strcmp(evt_name, "*")) {
+	if (strpbrk(evt_name, "*?")) {
 		*strp = evt_name + evt_length;
-		return parse_subsystem_tracepoint_event(sys_name, flags);
+		return parse_multiple_tracepoint_event(sys_name, evt_name,
+						       flags);
 	} else
 		return parse_single_tracepoint_event(sys_name, evt_name,
 						     evt_length, flags,
@@ -753,11 +758,11 @@ modifier:
 	return ret;
 }
 
-static void store_event_type(const char *orgname)
+static int store_event_type(const char *orgname)
 {
 	char filename[PATH_MAX], *c;
 	FILE *file;
-	int id;
+	int id, n;
 
 	sprintf(filename, "%s/", debugfs_path);
 	strncat(filename, orgname, strlen(orgname));
@@ -769,11 +774,14 @@ static void store_event_type(const char *orgname)
 
 	file = fopen(filename, "r");
 	if (!file)
-		return;
-	if (fscanf(file, "%i", &id) < 1)
-		die("cannot store event ID");
+		return 0;
+	n = fscanf(file, "%i", &id);
 	fclose(file);
-	perf_header__push_event(id, orgname);
+	if (n < 1) {
+		pr_err("cannot store event ID\n");
+		return -EINVAL;
+	}
+	return perf_header__push_event(id, orgname);
 }
 
 int parse_events(const struct option *opt __used, const char *str, int unset __used)
@@ -782,7 +790,8 @@ int parse_events(const struct option *opt __used, const char *str, int unset __u
 	enum event_result ret;
 
 	if (strchr(str, ':'))
-		store_event_type(str);
+		if (store_event_type(str) < 0)
+			return -1;
 
 	for (;;) {
 		if (nr_counters == MAX_COUNTERS)
@@ -835,11 +844,12 @@ int parse_filter(const struct option *opt __used, const char *str,
 }
 
 static const char * const event_type_descriptors[] = {
-	"",
 	"Hardware event",
 	"Software event",
 	"Tracepoint event",
 	"Hardware cache event",
+	"Raw hardware event descriptor",
+	"Hardware breakpoint",
 };
 
 /*
@@ -872,7 +882,7 @@ static void print_tracepoint_events(void)
 			snprintf(evt_path, MAXPATHLEN, "%s:%s",
 				 sys_dirent.d_name, evt_dirent.d_name);
 			printf("  %-42s [%s]\n", evt_path,
-				event_type_descriptors[PERF_TYPE_TRACEPOINT+1]);
+				event_type_descriptors[PERF_TYPE_TRACEPOINT]);
 		}
 		closedir(evt_dir);
 	}
@@ -892,9 +902,7 @@ void print_events(void)
 	printf("List of pre-defined events (to be used in -e):\n");
 
 	for (i = 0; i < ARRAY_SIZE(event_symbols); i++, syms++) {
-		type = syms->type + 1;
-		if (type >= ARRAY_SIZE(event_type_descriptors))
-			type = 0;
+		type = syms->type;
 
 		if (type != prev_type)
 			printf("\n");
@@ -919,17 +927,19 @@ void print_events(void)
 			for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) {
 				printf("  %-42s [%s]\n",
 					event_cache_name(type, op, i),
-					event_type_descriptors[4]);
+					event_type_descriptors[PERF_TYPE_HW_CACHE]);
 			}
 		}
 	}
 
 	printf("\n");
-	printf("  %-42s [raw hardware event descriptor]\n",
-		"rNNN");
+	printf("  %-42s [%s]\n",
+		"rNNN", event_type_descriptors[PERF_TYPE_RAW]);
 	printf("\n");
 
-	printf("  %-42s [hardware breakpoint]\n", "mem:<addr>[:access]");
+	printf("  %-42s [%s]\n",
+			"mem:<addr>[:access]",
+			event_type_descriptors[PERF_TYPE_BREAKPOINT]);
 	printf("\n");
 
 	print_tracepoint_events();
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 29465d440043..71b0dd590a37 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -37,6 +37,8 @@
 #include "string.h"
 #include "strlist.h"
 #include "debug.h"
+#include "cache.h"
+#include "color.h"
 #include "parse-events.h"  /* For debugfs_path */
 #include "probe-event.h"
 
@@ -62,6 +64,42 @@ static int e_snprintf(char *str, size_t size, const char *format, ...)
 	return ret;
 }
 
+void parse_line_range_desc(const char *arg, struct line_range *lr)
+{
+	const char *ptr;
+	char *tmp;
+	/*
+	 * <Syntax>
+	 * SRC:SLN[+NUM|-ELN]
+	 * FUNC[:SLN[+NUM|-ELN]]
+	 */
+	ptr = strchr(arg, ':');
+	if (ptr) {
+		lr->start = (unsigned int)strtoul(ptr + 1, &tmp, 0);
+		if (*tmp == '+')
+			lr->end = lr->start + (unsigned int)strtoul(tmp + 1,
+								    &tmp, 0);
+		else if (*tmp == '-')
+			lr->end = (unsigned int)strtoul(tmp + 1, &tmp, 0);
+		else
+			lr->end = 0;
+		pr_debug("Line range is %u to %u\n", lr->start, lr->end);
+		if (lr->end && lr->start > lr->end)
+			semantic_error("Start line must be smaller"
+				       " than end line.");
+		if (*tmp != '\0')
+			semantic_error("Tailing with invalid character '%d'.",
+				       *tmp);
+		tmp = strndup(arg, (ptr - arg));
+	} else
+		tmp = strdup(arg);
+
+	if (strchr(tmp, '.'))
+		lr->file = tmp;
+	else
+		lr->function = tmp;
+}
+
 /* Check the name is good for event/group */
 static bool check_event_name(const char *name)
 {
@@ -368,7 +406,7 @@ static int open_kprobe_events(int flags, int mode)
 	if (ret < 0) {
 		if (errno == ENOENT)
 			die("kprobe_events file does not exist -"
-			    " please rebuild with CONFIG_KPROBE_TRACER.");
+			    " please rebuild with CONFIG_KPROBE_EVENT.");
 		else
 			die("Could not open kprobe_events file: %s",
 			    strerror(errno));
@@ -455,6 +493,8 @@ void show_perf_probe_events(void)
 	struct strlist *rawlist;
 	struct str_node *ent;
 
+	setup_pager();
+
 	fd = open_kprobe_events(O_RDONLY, 0);
 	rawlist = get_trace_kprobe_event_rawlist(fd);
 	close(fd);
@@ -675,3 +715,66 @@ void del_trace_kprobe_events(struct strlist *dellist)
 	close(fd);
 }
 
+#define LINEBUF_SIZE 256
+
+static void show_one_line(FILE *fp, unsigned int l, bool skip, bool show_num)
+{
+	char buf[LINEBUF_SIZE];
+	const char *color = PERF_COLOR_BLUE;
+
+	if (fgets(buf, LINEBUF_SIZE, fp) == NULL)
+		goto error;
+	if (!skip) {
+		if (show_num)
+			fprintf(stdout, "%7u  %s", l, buf);
+		else
+			color_fprintf(stdout, color, "         %s", buf);
+	}
+
+	while (strlen(buf) == LINEBUF_SIZE - 1 &&
+	       buf[LINEBUF_SIZE - 2] != '\n') {
+		if (fgets(buf, LINEBUF_SIZE, fp) == NULL)
+			goto error;
+		if (!skip) {
+			if (show_num)
+				fprintf(stdout, "%s", buf);
+			else
+				color_fprintf(stdout, color, "%s", buf);
+		}
+	}
+	return;
+error:
+	if (feof(fp))
+		die("Source file is shorter than expected.");
+	else
+		die("File read error: %s", strerror(errno));
+}
+
+void show_line_range(struct line_range *lr)
+{
+	unsigned int l = 1;
+	struct line_node *ln;
+	FILE *fp;
+
+	setup_pager();
+
+	if (lr->function)
+		fprintf(stdout, "<%s:%d>\n", lr->function,
+			lr->start - lr->offset);
+	else
+		fprintf(stdout, "<%s:%d>\n", lr->file, lr->start);
+
+	fp = fopen(lr->path, "r");
+	if (fp == NULL)
+		die("Failed to open %s: %s", lr->path, strerror(errno));
+	/* Skip to starting line number */
+	while (l < lr->start)
+		show_one_line(fp, l++, true, false);
+
+	list_for_each_entry(ln, &lr->line_list, list) {
+		while (ln->line > l)
+			show_one_line(fp, (l++) - lr->offset, false, false);
+		show_one_line(fp, (l++) - lr->offset, false, true);
+	}
+	fclose(fp);
+}
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
index 7f1d499118c0..711287d4baea 100644
--- a/tools/perf/util/probe-event.h
+++ b/tools/perf/util/probe-event.h
@@ -5,6 +5,7 @@
 #include "probe-finder.h"
 #include "strlist.h"
 
+extern void parse_line_range_desc(const char *arg, struct line_range *lr);
 extern void parse_perf_probe_event(const char *str, struct probe_point *pp,
 				   bool *need_dwarf);
 extern int synthesize_perf_probe_point(struct probe_point *pp);
@@ -15,6 +16,7 @@ extern void add_trace_kprobe_events(struct probe_point *probes, int nr_probes,
 				    bool force_add);
 extern void del_trace_kprobe_events(struct strlist *dellist);
 extern void show_perf_probe_events(void);
+extern void show_line_range(struct line_range *lr);
 
 /* Maximum index number of event-name postfix */
 #define MAX_EVENT_INDEX	1024
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 4b852c0d16a5..1b2124d12f68 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -140,6 +140,31 @@ static Dwarf_Unsigned cu_find_fileno(Dwarf_Die cu_die, const char *fname)
 	return found;
 }
 
+static int cu_get_filename(Dwarf_Die cu_die, Dwarf_Unsigned fno, char **buf)
+{
+	Dwarf_Signed cnt, i;
+	char **srcs;
+	int ret = 0;
+
+	if (!buf || !fno)
+		return -EINVAL;
+
+	ret = dwarf_srcfiles(cu_die, &srcs, &cnt, &__dw_error);
+	if (ret == DW_DLV_OK) {
+		if ((Dwarf_Unsigned)cnt > fno - 1) {
+			*buf = strdup(srcs[fno - 1]);
+			ret = 0;
+			pr_debug("found filename: %s\n", *buf);
+		} else
+			ret = -ENOENT;
+		for (i = 0; i < cnt; i++)
+			dwarf_dealloc(__dw_debug, srcs[i], DW_DLA_STRING);
+		dwarf_dealloc(__dw_debug, srcs, DW_DLA_LIST);
+	} else
+		ret = -EINVAL;
+	return ret;
+}
+
 /* Compare diename and tname */
 static int die_compare_name(Dwarf_Die dw_die, const char *tname)
 {
@@ -402,11 +427,11 @@ static void show_location(Dwarf_Loc *loc, struct probe_finder *pf)
 	} else if (op == DW_OP_regx) {
 		regn = loc->lr_number;
 	} else
-		die("Dwarf_OP %d is not supported.\n", op);
+		die("Dwarf_OP %d is not supported.", op);
 
 	regs = get_arch_regstr(regn);
 	if (!regs)
-		die("%lld exceeds max register number.\n", regn);
+		die("%lld exceeds max register number.", regn);
 
 	if (deref)
 		ret = snprintf(pf->buf, pf->len,
@@ -438,7 +463,7 @@ static void show_variable(Dwarf_Die vr_die, struct probe_finder *pf)
 	return ;
 error:
 	die("Failed to find the location of %s at this address.\n"
-	    " Perhaps, it has been optimized out.\n", pf->var);
+	    " Perhaps, it has been optimized out.", pf->var);
 }
 
 static int variable_callback(struct die_link *dlink, void *data)
@@ -476,7 +501,7 @@ static void find_variable(Dwarf_Die sp_die, struct probe_finder *pf)
 	/* Search child die for local variables and parameters. */
 	ret = search_die_from_children(sp_die, variable_callback, pf);
 	if (!ret)
-		die("Failed to find '%s' in this function.\n", pf->var);
+		die("Failed to find '%s' in this function.", pf->var);
 }
 
 /* Get a frame base on the address */
@@ -567,7 +592,7 @@ static int probeaddr_callback(struct die_link *dlink, void *data)
 }
 
 /* Find probe point from its line number */
-static void find_by_line(struct probe_finder *pf)
+static void find_probe_point_by_line(struct probe_finder *pf)
 {
 	Dwarf_Signed cnt, i, clm;
 	Dwarf_Line *lines;
@@ -602,7 +627,7 @@ static void find_by_line(struct probe_finder *pf)
 		ret = search_die_from_children(pf->cu_die,
 					       probeaddr_callback, pf);
 		if (ret == 0)
-			die("Probe point is not found in subprograms.\n");
+			die("Probe point is not found in subprograms.");
 		/* Continuing, because target line might be inlined. */
 	}
 	dwarf_srclines_dealloc(__dw_debug, lines, cnt);
@@ -626,7 +651,7 @@ static int probefunc_callback(struct die_link *dlink, void *data)
 				pf->fno = die_get_decl_file(dlink->die);
 				pf->lno = die_get_decl_line(dlink->die)
 					 + pp->line;
-				find_by_line(pf);
+				find_probe_point_by_line(pf);
 				return 1;
 			}
 			if (die_inlined_subprogram(dlink->die)) {
@@ -661,7 +686,7 @@ static int probefunc_callback(struct die_link *dlink, void *data)
 				    !die_inlined_subprogram(lk->die))
 					goto found;
 			}
-			die("Failed to find real subprogram.\n");
+			die("Failed to find real subprogram.");
 found:
 			/* Get offset from subprogram */
 			ret = die_within_subprogram(lk->die, pf->addr, &offs);
@@ -673,7 +698,7 @@ found:
 	return 0;
 }
 
-static void find_by_func(struct probe_finder *pf)
+static void find_probe_point_by_func(struct probe_finder *pf)
 {
 	search_die_from_children(pf->cu_die, probefunc_callback, pf);
 }
@@ -714,10 +739,10 @@ int find_probepoint(int fd, struct probe_point *pp)
 			if (ret == DW_DLV_NO_ENTRY)
 				pf.cu_base = 0;
 			if (pp->function)
-				find_by_func(&pf);
+				find_probe_point_by_func(&pf);
 			else {
 				pf.lno = pp->line;
-				find_by_line(&pf);
+				find_probe_point_by_line(&pf);
 			}
 		}
 		dwarf_dealloc(__dw_debug, pf.cu_die, DW_DLA_DIE);
@@ -728,3 +753,159 @@ int find_probepoint(int fd, struct probe_point *pp)
 	return pp->found;
 }
 
+
+static void line_range_add_line(struct line_range *lr, unsigned int line)
+{
+	struct line_node *ln;
+	struct list_head *p;
+
+	/* Reverse search, because new line will be the last one */
+	list_for_each_entry_reverse(ln, &lr->line_list, list) {
+		if (ln->line < line) {
+			p = &ln->list;
+			goto found;
+		} else if (ln->line == line)	/* Already exist */
+			return ;
+	}
+	/* List is empty, or the smallest entry */
+	p = &lr->line_list;
+found:
+	pr_debug("Debug: add a line %u\n", line);
+	ln = zalloc(sizeof(struct line_node));
+	DIE_IF(ln == NULL);
+	ln->line = line;
+	INIT_LIST_HEAD(&ln->list);
+	list_add(&ln->list, p);
+}
+
+/* Find line range from its line number */
+static void find_line_range_by_line(struct line_finder *lf)
+{
+	Dwarf_Signed cnt, i;
+	Dwarf_Line *lines;
+	Dwarf_Unsigned lineno = 0;
+	Dwarf_Unsigned fno;
+	Dwarf_Addr addr;
+	int ret;
+
+	ret = dwarf_srclines(lf->cu_die, &lines, &cnt, &__dw_error);
+	DIE_IF(ret != DW_DLV_OK);
+
+	for (i = 0; i < cnt; i++) {
+		ret = dwarf_line_srcfileno(lines[i], &fno, &__dw_error);
+		DIE_IF(ret != DW_DLV_OK);
+		if (fno != lf->fno)
+			continue;
+
+		ret = dwarf_lineno(lines[i], &lineno, &__dw_error);
+		DIE_IF(ret != DW_DLV_OK);
+		if (lf->lno_s > lineno || lf->lno_e < lineno)
+			continue;
+
+		/* Filter line in the function address range */
+		if (lf->addr_s && lf->addr_e) {
+			ret = dwarf_lineaddr(lines[i], &addr, &__dw_error);
+			DIE_IF(ret != DW_DLV_OK);
+			if (lf->addr_s > addr || lf->addr_e <= addr)
+				continue;
+		}
+		line_range_add_line(lf->lr, (unsigned int)lineno);
+	}
+	dwarf_srclines_dealloc(__dw_debug, lines, cnt);
+	if (!list_empty(&lf->lr->line_list))
+		lf->found = 1;
+}
+
+/* Search function from function name */
+static int linefunc_callback(struct die_link *dlink, void *data)
+{
+	struct line_finder *lf = (struct line_finder *)data;
+	struct line_range *lr = lf->lr;
+	Dwarf_Half tag;
+	int ret;
+
+	ret = dwarf_tag(dlink->die, &tag, &__dw_error);
+	DIE_IF(ret == DW_DLV_ERROR);
+	if (tag == DW_TAG_subprogram &&
+	    die_compare_name(dlink->die, lr->function) == 0) {
+		/* Get the address range of this function */
+		ret = dwarf_highpc(dlink->die, &lf->addr_e, &__dw_error);
+		if (ret == DW_DLV_OK)
+			ret = dwarf_lowpc(dlink->die, &lf->addr_s, &__dw_error);
+		DIE_IF(ret == DW_DLV_ERROR);
+		if (ret == DW_DLV_NO_ENTRY) {
+			lf->addr_s = 0;
+			lf->addr_e = 0;
+		}
+
+		lf->fno = die_get_decl_file(dlink->die);
+		lr->offset = die_get_decl_line(dlink->die);;
+		lf->lno_s = lr->offset + lr->start;
+		if (!lr->end)
+			lf->lno_e = (Dwarf_Unsigned)-1;
+		else
+			lf->lno_e = lr->offset + lr->end;
+		lr->start = lf->lno_s;
+		lr->end = lf->lno_e;
+		find_line_range_by_line(lf);
+		/* If we find a target function, this should be end. */
+		lf->found = 1;
+		return 1;
+	}
+	return 0;
+}
+
+static void find_line_range_by_func(struct line_finder *lf)
+{
+	search_die_from_children(lf->cu_die, linefunc_callback, lf);
+}
+
+int find_line_range(int fd, struct line_range *lr)
+{
+	Dwarf_Half addr_size = 0;
+	Dwarf_Unsigned next_cuh = 0;
+	int ret;
+	struct line_finder lf = {.lr = lr};
+
+	ret = dwarf_init(fd, DW_DLC_READ, 0, 0, &__dw_debug, &__dw_error);
+	if (ret != DW_DLV_OK)
+		return -ENOENT;
+
+	while (!lf.found) {
+		/* Search CU (Compilation Unit) */
+		ret = dwarf_next_cu_header(__dw_debug, NULL, NULL, NULL,
+			&addr_size, &next_cuh, &__dw_error);
+		DIE_IF(ret == DW_DLV_ERROR);
+		if (ret == DW_DLV_NO_ENTRY)
+			break;
+
+		/* Get the DIE(Debugging Information Entry) of this CU */
+		ret = dwarf_siblingof(__dw_debug, 0, &lf.cu_die, &__dw_error);
+		DIE_IF(ret != DW_DLV_OK);
+
+		/* Check if target file is included. */
+		if (lr->file)
+			lf.fno = cu_find_fileno(lf.cu_die, lr->file);
+
+		if (!lr->file || lf.fno) {
+			if (lr->function)
+				find_line_range_by_func(&lf);
+			else {
+				lf.lno_s = lr->start;
+				if (!lr->end)
+					lf.lno_e = (Dwarf_Unsigned)-1;
+				else
+					lf.lno_e = lr->end;
+				find_line_range_by_line(&lf);
+			}
+			/* Get the real file path */
+			if (lf.found)
+				cu_get_filename(lf.cu_die, lf.fno, &lr->path);
+		}
+		dwarf_dealloc(__dw_debug, lf.cu_die, DW_DLA_DIE);
+	}
+	ret = dwarf_finish(__dw_debug, &__dw_error);
+	DIE_IF(ret != DW_DLV_OK);
+	return lf.found;
+}
+
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
index a4086aaddb73..972b386116f1 100644
--- a/tools/perf/util/probe-finder.h
+++ b/tools/perf/util/probe-finder.h
@@ -1,6 +1,8 @@
 #ifndef _PROBE_FINDER_H
 #define _PROBE_FINDER_H
 
+#include "util.h"
+
 #define MAX_PATH_LEN		 256
 #define MAX_PROBE_BUFFER	1024
 #define MAX_PROBES		 128
@@ -32,8 +34,26 @@ struct probe_point {
 	char			*probes[MAX_PROBES];	/* Output buffers (will be allocated)*/
 };
 
+/* Line number container */
+struct line_node {
+	struct list_head	list;
+	unsigned int		line;
+};
+
+/* Line range */
+struct line_range {
+	char			*file;			/* File name */
+	char			*function;		/* Function name */
+	unsigned int		start;			/* Start line number */
+	unsigned int		end;			/* End line number */
+	unsigned int		offset;			/* Start line offset */
+	char			*path;			/* Real path name */
+	struct list_head	line_list;		/* Visible lines */
+};
+
 #ifndef NO_LIBDWARF
 extern int find_probepoint(int fd, struct probe_point *pp);
+extern int find_line_range(int fd, struct line_range *lr);
 
 /* Workaround for undefined _MIPS_SZLONG bug in libdwarf.h: */
 #ifndef _MIPS_SZLONG
@@ -60,6 +80,19 @@ struct probe_finder {
 	char			*buf;			/* Current output buffer */
 	int			len;			/* Length of output buffer */
 };
+
+struct line_finder {
+	struct line_range	*lr;			/* Target line range */
+
+	Dwarf_Unsigned		fno;			/* File number */
+	Dwarf_Unsigned		lno_s;			/* Start line number */
+	Dwarf_Unsigned		lno_e;			/* End line number */
+	Dwarf_Addr		addr_s;			/* Start address */
+	Dwarf_Addr		addr_e;			/* End address */
+	Dwarf_Die		cu_die;			/* Current CU */
+	int			found;
+};
+
 #endif /* NO_LIBDWARF */
 
 #endif /*_PROBE_FINDER_H */
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index ce3a6c8abe76..cf91d099f0aa 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1,5 +1,9 @@
+#define _LARGEFILE64_SOURCE
+#define _FILE_OFFSET_BITS 64
+
 #include <linux/kernel.h>
 
+#include <byteswap.h>
 #include <unistd.h>
 #include <sys/types.h>
 
@@ -11,7 +15,7 @@ static int perf_session__open(struct perf_session *self, bool force)
 {
 	struct stat input_stat;
 
-	self->fd = open(self->filename, O_RDONLY);
+	self->fd = open(self->filename, O_RDONLY|O_LARGEFILE);
 	if (self->fd < 0) {
 		pr_err("failed to open file: %s", self->filename);
 		if (!strcmp(self->filename, "perf.data"))
@@ -66,13 +70,22 @@ struct perf_session *perf_session__new(const char *filename, int mode, bool forc
 	self->mmap_window = 32;
 	self->cwd = NULL;
 	self->cwdlen = 0;
+	self->unknown_events = 0;
 	map_groups__init(&self->kmaps);
 
-	if (perf_session__create_kernel_maps(self) < 0)
-		goto out_delete;
+	if (mode == O_RDONLY) {
+		if (perf_session__open(self, force) < 0)
+			goto out_delete;
+	} else if (mode == O_WRONLY) {
+		/*
+		 * In O_RDONLY mode this will be performed when reading the
+		 * kernel MMAP event, in event__process_mmap().
+		 */
+		if (perf_session__create_kernel_maps(self) < 0)
+			goto out_delete;
+	}
 
-	if (mode == O_RDONLY && perf_session__open(self, force) < 0)
-		goto out_delete;
+	self->sample_type = perf_header__sample_type(&self->header);
 out:
 	return self;
 out_free:
@@ -148,3 +161,389 @@ struct symbol **perf_session__resolve_callchain(struct perf_session *self,
 
 	return syms;
 }
+
+static int process_event_stub(event_t *event __used,
+			      struct perf_session *session __used)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static void perf_event_ops__fill_defaults(struct perf_event_ops *handler)
+{
+	if (handler->sample == NULL)
+		handler->sample = process_event_stub;
+	if (handler->mmap == NULL)
+		handler->mmap = process_event_stub;
+	if (handler->comm == NULL)
+		handler->comm = process_event_stub;
+	if (handler->fork == NULL)
+		handler->fork = process_event_stub;
+	if (handler->exit == NULL)
+		handler->exit = process_event_stub;
+	if (handler->lost == NULL)
+		handler->lost = process_event_stub;
+	if (handler->read == NULL)
+		handler->read = process_event_stub;
+	if (handler->throttle == NULL)
+		handler->throttle = process_event_stub;
+	if (handler->unthrottle == NULL)
+		handler->unthrottle = process_event_stub;
+}
+
+static const char *event__name[] = {
+	[0]			 = "TOTAL",
+	[PERF_RECORD_MMAP]	 = "MMAP",
+	[PERF_RECORD_LOST]	 = "LOST",
+	[PERF_RECORD_COMM]	 = "COMM",
+	[PERF_RECORD_EXIT]	 = "EXIT",
+	[PERF_RECORD_THROTTLE]	 = "THROTTLE",
+	[PERF_RECORD_UNTHROTTLE] = "UNTHROTTLE",
+	[PERF_RECORD_FORK]	 = "FORK",
+	[PERF_RECORD_READ]	 = "READ",
+	[PERF_RECORD_SAMPLE]	 = "SAMPLE",
+};
+
+unsigned long event__total[PERF_RECORD_MAX];
+
+void event__print_totals(void)
+{
+	int i;
+	for (i = 0; i < PERF_RECORD_MAX; ++i)
+		pr_info("%10s events: %10ld\n",
+			event__name[i], event__total[i]);
+}
+
+void mem_bswap_64(void *src, int byte_size)
+{
+	u64 *m = src;
+
+	while (byte_size > 0) {
+		*m = bswap_64(*m);
+		byte_size -= sizeof(u64);
+		++m;
+	}
+}
+
+static void event__all64_swap(event_t *self)
+{
+	struct perf_event_header *hdr = &self->header;
+	mem_bswap_64(hdr + 1, self->header.size - sizeof(*hdr));
+}
+
+static void event__comm_swap(event_t *self)
+{
+	self->comm.pid = bswap_32(self->comm.pid);
+	self->comm.tid = bswap_32(self->comm.tid);
+}
+
+static void event__mmap_swap(event_t *self)
+{
+	self->mmap.pid	 = bswap_32(self->mmap.pid);
+	self->mmap.tid	 = bswap_32(self->mmap.tid);
+	self->mmap.start = bswap_64(self->mmap.start);
+	self->mmap.len	 = bswap_64(self->mmap.len);
+	self->mmap.pgoff = bswap_64(self->mmap.pgoff);
+}
+
+static void event__task_swap(event_t *self)
+{
+	self->fork.pid	= bswap_32(self->fork.pid);
+	self->fork.tid	= bswap_32(self->fork.tid);
+	self->fork.ppid	= bswap_32(self->fork.ppid);
+	self->fork.ptid	= bswap_32(self->fork.ptid);
+	self->fork.time	= bswap_64(self->fork.time);
+}
+
+static void event__read_swap(event_t *self)
+{
+	self->read.pid		= bswap_32(self->read.pid);
+	self->read.tid		= bswap_32(self->read.tid);
+	self->read.value	= bswap_64(self->read.value);
+	self->read.time_enabled	= bswap_64(self->read.time_enabled);
+	self->read.time_running	= bswap_64(self->read.time_running);
+	self->read.id		= bswap_64(self->read.id);
+}
+
+typedef void (*event__swap_op)(event_t *self);
+
+static event__swap_op event__swap_ops[] = {
+	[PERF_RECORD_MMAP]   = event__mmap_swap,
+	[PERF_RECORD_COMM]   = event__comm_swap,
+	[PERF_RECORD_FORK]   = event__task_swap,
+	[PERF_RECORD_EXIT]   = event__task_swap,
+	[PERF_RECORD_LOST]   = event__all64_swap,
+	[PERF_RECORD_READ]   = event__read_swap,
+	[PERF_RECORD_SAMPLE] = event__all64_swap,
+	[PERF_RECORD_MAX]    = NULL,
+};
+
+static int perf_session__process_event(struct perf_session *self,
+				       event_t *event,
+				       struct perf_event_ops *ops,
+				       u64 offset, u64 head)
+{
+	trace_event(event);
+
+	if (event->header.type < PERF_RECORD_MAX) {
+		dump_printf("%#Lx [%#x]: PERF_RECORD_%s",
+			    offset + head, event->header.size,
+			    event__name[event->header.type]);
+		++event__total[0];
+		++event__total[event->header.type];
+	}
+
+	if (self->header.needs_swap && event__swap_ops[event->header.type])
+		event__swap_ops[event->header.type](event);
+
+	switch (event->header.type) {
+	case PERF_RECORD_SAMPLE:
+		return ops->sample(event, self);
+	case PERF_RECORD_MMAP:
+		return ops->mmap(event, self);
+	case PERF_RECORD_COMM:
+		return ops->comm(event, self);
+	case PERF_RECORD_FORK:
+		return ops->fork(event, self);
+	case PERF_RECORD_EXIT:
+		return ops->exit(event, self);
+	case PERF_RECORD_LOST:
+		return ops->lost(event, self);
+	case PERF_RECORD_READ:
+		return ops->read(event, self);
+	case PERF_RECORD_THROTTLE:
+		return ops->throttle(event, self);
+	case PERF_RECORD_UNTHROTTLE:
+		return ops->unthrottle(event, self);
+	default:
+		self->unknown_events++;
+		return -1;
+	}
+}
+
+void perf_event_header__bswap(struct perf_event_header *self)
+{
+	self->type = bswap_32(self->type);
+	self->misc = bswap_16(self->misc);
+	self->size = bswap_16(self->size);
+}
+
+int perf_header__read_build_ids(struct perf_header *self,
+				int input, u64 offset, u64 size)
+{
+	struct build_id_event bev;
+	char filename[PATH_MAX];
+	u64 limit = offset + size;
+	int err = -1;
+
+	while (offset < limit) {
+		struct dso *dso;
+		ssize_t len;
+		struct list_head *head = &dsos__user;
+
+		if (read(input, &bev, sizeof(bev)) != sizeof(bev))
+			goto out;
+
+		if (self->needs_swap)
+			perf_event_header__bswap(&bev.header);
+
+		len = bev.header.size - sizeof(bev);
+		if (read(input, filename, len) != len)
+			goto out;
+
+		if (bev.header.misc & PERF_RECORD_MISC_KERNEL)
+			head = &dsos__kernel;
+
+		dso = __dsos__findnew(head, filename);
+		if (dso != NULL) {
+			dso__set_build_id(dso, &bev.build_id);
+			if (head == &dsos__kernel && filename[0] == '[')
+				dso->kernel = 1;
+		}
+
+		offset += bev.header.size;
+	}
+	err = 0;
+out:
+	return err;
+}
+
+static struct thread *perf_session__register_idle_thread(struct perf_session *self)
+{
+	struct thread *thread = perf_session__findnew(self, 0);
+
+	if (thread == NULL || thread__set_comm(thread, "swapper")) {
+		pr_err("problem inserting idle task.\n");
+		thread = NULL;
+	}
+
+	return thread;
+}
+
+int perf_session__process_events(struct perf_session *self,
+				 struct perf_event_ops *ops)
+{
+	int err, mmap_prot, mmap_flags;
+	u64 head, shift;
+	u64 offset = 0;
+	size_t	page_size;
+	event_t *event;
+	uint32_t size;
+	char *buf;
+
+	if (perf_session__register_idle_thread(self) == NULL)
+		return -ENOMEM;
+
+	perf_event_ops__fill_defaults(ops);
+
+	page_size = sysconf(_SC_PAGESIZE);
+
+	head = self->header.data_offset;
+
+	if (!symbol_conf.full_paths) {
+		char bf[PATH_MAX];
+
+		if (getcwd(bf, sizeof(bf)) == NULL) {
+			err = -errno;
+out_getcwd_err:
+			pr_err("failed to get the current directory\n");
+			goto out_err;
+		}
+		self->cwd = strdup(bf);
+		if (self->cwd == NULL) {
+			err = -ENOMEM;
+			goto out_getcwd_err;
+		}
+		self->cwdlen = strlen(self->cwd);
+	}
+
+	shift = page_size * (head / page_size);
+	offset += shift;
+	head -= shift;
+
+	mmap_prot  = PROT_READ;
+	mmap_flags = MAP_SHARED;
+
+	if (self->header.needs_swap) {
+		mmap_prot  |= PROT_WRITE;
+		mmap_flags = MAP_PRIVATE;
+	}
+remap:
+	buf = mmap(NULL, page_size * self->mmap_window, mmap_prot,
+		   mmap_flags, self->fd, offset);
+	if (buf == MAP_FAILED) {
+		pr_err("failed to mmap file\n");
+		err = -errno;
+		goto out_err;
+	}
+
+more:
+	event = (event_t *)(buf + head);
+
+	if (self->header.needs_swap)
+		perf_event_header__bswap(&event->header);
+	size = event->header.size;
+	if (size == 0)
+		size = 8;
+
+	if (head + event->header.size >= page_size * self->mmap_window) {
+		int munmap_ret;
+
+		shift = page_size * (head / page_size);
+
+		munmap_ret = munmap(buf, page_size * self->mmap_window);
+		assert(munmap_ret == 0);
+
+		offset += shift;
+		head -= shift;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+	dump_printf("\n%#Lx [%#x]: event: %d\n",
+		    offset + head, event->header.size, event->header.type);
+
+	if (size == 0 ||
+	    perf_session__process_event(self, event, ops, offset, head) < 0) {
+		dump_printf("%#Lx [%#x]: skipping unknown header type: %d\n",
+			    offset + head, event->header.size,
+			    event->header.type);
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+
+	if (offset + head >= self->header.data_offset + self->header.data_size)
+		goto done;
+
+	if (offset + head < self->size)
+		goto more;
+done:
+	err = 0;
+out_err:
+	return err;
+}
+
+bool perf_session__has_traces(struct perf_session *self, const char *msg)
+{
+	if (!(self->sample_type & PERF_SAMPLE_RAW)) {
+		pr_err("No trace sample to read. Did you call 'perf %s'?\n", msg);
+		return false;
+	}
+
+	return true;
+}
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct perf_session *self,
+					     const char *symbol_name,
+					     u64 addr)
+{
+	char *bracket;
+
+	self->ref_reloc_sym.name = strdup(symbol_name);
+	if (self->ref_reloc_sym.name == NULL)
+		return -ENOMEM;
+
+	bracket = strchr(self->ref_reloc_sym.name, ']');
+	if (bracket)
+		*bracket = '\0';
+
+	self->ref_reloc_sym.addr = addr;
+	return 0;
+}
+
+static u64 map__reloc_map_ip(struct map *map, u64 ip)
+{
+	return ip + (s64)map->pgoff;
+}
+
+static u64 map__reloc_unmap_ip(struct map *map, u64 ip)
+{
+	return ip - (s64)map->pgoff;
+}
+
+void perf_session__reloc_vmlinux_maps(struct perf_session *self,
+				      u64 unrelocated_addr)
+{
+	enum map_type type;
+	s64 reloc = unrelocated_addr - self->ref_reloc_sym.addr;
+
+	if (!reloc)
+		return;
+
+	for (type = 0; type < MAP__NR_TYPES; ++type) {
+		struct map *map = self->vmlinux_maps[type];
+
+		map->map_ip = map__reloc_map_ip;
+		map->unmap_ip = map__reloc_unmap_ip;
+		map->pgoff = reloc;
+	}
+}
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
index 32eaa1bada06..36d1a80c0b6c 100644
--- a/tools/perf/util/session.h
+++ b/tools/perf/util/session.h
@@ -18,10 +18,16 @@ struct perf_session {
 	struct map_groups	kmaps;
 	struct rb_root		threads;
 	struct thread		*last_match;
+	struct map		*vmlinux_maps[MAP__NR_TYPES];
 	struct events_stats	events_stats;
 	unsigned long		event_total[PERF_RECORD_MAX];
+	unsigned long		unknown_events;
 	struct rb_root		hists;
 	u64			sample_type;
+	struct {
+		const char	*name;
+		u64		addr;
+	}			ref_reloc_sym;
 	int			fd;
 	int			cwdlen;
 	char			*cwd;
@@ -31,23 +37,22 @@ struct perf_session {
 typedef int (*event_op)(event_t *self, struct perf_session *session);
 
 struct perf_event_ops {
-	event_op	process_sample_event;
-	event_op	process_mmap_event;
-	event_op	process_comm_event;
-	event_op	process_fork_event;
-	event_op	process_exit_event;
-	event_op	process_lost_event;
-	event_op	process_read_event;
-	event_op	process_throttle_event;
-	event_op	process_unthrottle_event;
-	int		(*sample_type_check)(struct perf_session *session);
-	unsigned long	total_unknown;
-	bool		full_paths;
+	event_op sample,
+		 mmap,
+		 comm,
+		 fork,
+		 exit,
+		 lost,
+		 read,
+		 throttle,
+		 unthrottle;
 };
 
 struct perf_session *perf_session__new(const char *filename, int mode, bool force);
 void perf_session__delete(struct perf_session *self);
 
+void perf_event_header__bswap(struct perf_event_header *self);
+
 int perf_session__process_events(struct perf_session *self,
 				 struct perf_event_ops *event_ops);
 
@@ -56,6 +61,17 @@ struct symbol **perf_session__resolve_callchain(struct perf_session *self,
 						struct ip_callchain *chain,
 						struct symbol **parent);
 
-int perf_header__read_build_ids(int input, u64 offset, u64 file_size);
+bool perf_session__has_traces(struct perf_session *self, const char *msg);
+
+int perf_header__read_build_ids(struct perf_header *self, int input,
+				u64 offset, u64 file_size);
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct perf_session *self,
+					     const char *symbol_name,
+					     u64 addr);
+void perf_session__reloc_vmlinux_maps(struct perf_session *self,
+				      u64 unrelocated_addr);
+
+void mem_bswap_64(void *src, int byte_size);
 
 #endif /* __PERF_SESSION_H */
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
index 5352d7dccc61..c397d4f6f748 100644
--- a/tools/perf/util/string.c
+++ b/tools/perf/util/string.c
@@ -227,16 +227,73 @@ fail:
 	return NULL;
 }
 
-/* Glob expression pattern matching */
+/* Character class matching */
+static bool __match_charclass(const char *pat, char c, const char **npat)
+{
+	bool complement = false, ret = true;
+
+	if (*pat == '!') {
+		complement = true;
+		pat++;
+	}
+	if (*pat++ == c)	/* First character is special */
+		goto end;
+
+	while (*pat && *pat != ']') {	/* Matching */
+		if (*pat == '-' && *(pat + 1) != ']') {	/* Range */
+			if (*(pat - 1) <= c && c <= *(pat + 1))
+				goto end;
+			if (*(pat - 1) > *(pat + 1))
+				goto error;
+			pat += 2;
+		} else if (*pat++ == c)
+			goto end;
+	}
+	if (!*pat)
+		goto error;
+	ret = false;
+
+end:
+	while (*pat && *pat != ']')	/* Searching closing */
+		pat++;
+	if (!*pat)
+		goto error;
+	*npat = pat + 1;
+	return complement ? !ret : ret;
+
+error:
+	return false;
+}
+
+/**
+ * strglobmatch - glob expression pattern matching
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This returns true if the @str matches @pat. @pat can includes wildcards
+ * ('*','?') and character classes ([CHARS], complementation and ranges are
+ * also supported). Also, this supports escape character ('\') to use special
+ * characters as normal character.
+ *
+ * Note: if @pat syntax is broken, this always returns false.
+ */
 bool strglobmatch(const char *str, const char *pat)
 {
 	while (*str && *pat && *pat != '*') {
-		if (*pat == '?') {
+		if (*pat == '?') {	/* Matches any single character */
 			str++;
 			pat++;
-		} else
-			if (*str++ != *pat++)
+			continue;
+		} else if (*pat == '[')	/* Character classes/Ranges */
+			if (__match_charclass(pat + 1, *str, &pat)) {
+				str++;
+				continue;
+			} else
 				return false;
+		else if (*pat == '\\') /* Escaped char match as normal char */
+			pat++;
+		if (*str++ != *pat++)
+			return false;
 	}
 	/* Check wild card */
 	if (*pat == '*') {
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index ab92763edb03..f9049d12ead6 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -22,6 +22,7 @@
 enum dso_origin {
 	DSO__ORIG_KERNEL = 0,
 	DSO__ORIG_JAVA_JIT,
+	DSO__ORIG_BUILD_ID_CACHE,
 	DSO__ORIG_FEDORA,
 	DSO__ORIG_UBUNTU,
 	DSO__ORIG_BUILDID,
@@ -63,7 +64,7 @@ static void dso__set_sorted_by_name(struct dso *self, enum map_type type)
 	self->sorted_by_name |= (1 << type);
 }
 
-static bool symbol_type__is_a(char symbol_type, enum map_type map_type)
+bool symbol_type__is_a(char symbol_type, enum map_type map_type)
 {
 	switch (map_type) {
 	case MAP__FUNCTION:
@@ -160,7 +161,7 @@ static size_t symbol__fprintf(struct symbol *self, FILE *fp)
 		       self->start, self->end, self->name);
 }
 
-static void dso__set_long_name(struct dso *self, char *name)
+void dso__set_long_name(struct dso *self, char *name)
 {
 	if (name == NULL)
 		return;
@@ -175,7 +176,7 @@ static void dso__set_basename(struct dso *self)
 
 struct dso *dso__new(const char *name)
 {
-	struct dso *self = malloc(sizeof(*self) + strlen(name) + 1);
+	struct dso *self = zalloc(sizeof(*self) + strlen(name) + 1);
 
 	if (self != NULL) {
 		int i;
@@ -344,10 +345,10 @@ void dso__sort_by_name(struct dso *self, enum map_type type)
 				     &self->symbols[type]);
 }
 
-int build_id__sprintf(u8 *self, int len, char *bf)
+int build_id__sprintf(const u8 *self, int len, char *bf)
 {
 	char *bid = bf;
-	u8 *raw = self;
+	const u8 *raw = self;
 	int i;
 
 	for (i = 0; i < len; ++i) {
@@ -382,24 +383,20 @@ size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp)
 	return ret;
 }
 
-/*
- * Loads the function entries in /proc/kallsyms into kernel_map->dso,
- * so that we can in the next step set the symbol ->end address and then
- * call kernel_maps__split_kallsyms.
- */
-static int dso__load_all_kallsyms(struct dso *self, struct map *map)
+int kallsyms__parse(const char *filename, void *arg,
+		    int (*process_symbol)(void *arg, const char *name,
+						     char type, u64 start))
 {
 	char *line = NULL;
 	size_t n;
-	struct rb_root *root = &self->symbols[map->type];
-	FILE *file = fopen("/proc/kallsyms", "r");
+	int err = 0;
+	FILE *file = fopen(filename, "r");
 
 	if (file == NULL)
 		goto out_failure;
 
 	while (!feof(file)) {
 		u64 start;
-		struct symbol *sym;
 		int line_len, len;
 		char symbol_type;
 		char *symbol_name;
@@ -420,35 +417,63 @@ static int dso__load_all_kallsyms(struct dso *self, struct map *map)
 			continue;
 
 		symbol_type = toupper(line[len]);
-		if (!symbol_type__is_a(symbol_type, map->type))
-			continue;
-
 		symbol_name = line + len + 2;
-		/*
-		 * Will fix up the end later, when we have all symbols sorted.
-		 */
-		sym = symbol__new(start, 0, symbol_name);
 
-		if (sym == NULL)
-			goto out_delete_line;
-		/*
-		 * We will pass the symbols to the filter later, in
-		 * map__split_kallsyms, when we have split the maps per module
-		 */
-		symbols__insert(root, sym);
+		err = process_symbol(arg, symbol_name, symbol_type, start);
+		if (err)
+			break;
 	}
 
 	free(line);
 	fclose(file);
+	return err;
 
-	return 0;
-
-out_delete_line:
-	free(line);
 out_failure:
 	return -1;
 }
 
+struct process_kallsyms_args {
+	struct map *map;
+	struct dso *dso;
+};
+
+static int map__process_kallsym_symbol(void *arg, const char *name,
+				       char type, u64 start)
+{
+	struct symbol *sym;
+	struct process_kallsyms_args *a = arg;
+	struct rb_root *root = &a->dso->symbols[a->map->type];
+
+	if (!symbol_type__is_a(type, a->map->type))
+		return 0;
+
+	/*
+	 * Will fix up the end later, when we have all symbols sorted.
+	 */
+	sym = symbol__new(start, 0, name);
+
+	if (sym == NULL)
+		return -ENOMEM;
+	/*
+	 * We will pass the symbols to the filter later, in
+	 * map__split_kallsyms, when we have split the maps per module
+	 */
+	symbols__insert(root, sym);
+	return 0;
+}
+
+/*
+ * Loads the function entries in /proc/kallsyms into kernel_map->dso,
+ * so that we can in the next step set the symbol ->end address and then
+ * call kernel_maps__split_kallsyms.
+ */
+static int dso__load_all_kallsyms(struct dso *self, const char *filename,
+				  struct map *map)
+{
+	struct process_kallsyms_args args = { .map = map, .dso = self, };
+	return kallsyms__parse(filename, &args, map__process_kallsym_symbol);
+}
+
 /*
  * Split the symbols into maps, making sure there are no overlaps, i.e. the
  * kernel range is broken in several maps, named [kernel].N, as we don't have
@@ -477,13 +502,17 @@ static int dso__split_kallsyms(struct dso *self, struct map *map,
 
 			*module++ = '\0';
 
-			if (strcmp(self->name, module)) {
+			if (strcmp(curr_map->dso->short_name, module)) {
 				curr_map = map_groups__find_by_name(&session->kmaps, map->type, module);
 				if (curr_map == NULL) {
 					pr_debug("/proc/{kallsyms,modules} "
-					         "inconsistency!\n");
+					         "inconsistency while looking "
+						 "for \"%s\" module!\n", module);
 					return -1;
 				}
+
+				if (curr_map->dso->loaded)
+					goto discard_symbol;
 			}
 			/*
 			 * So that we look just like we get from .ko files,
@@ -529,10 +558,10 @@ discard_symbol:		rb_erase(&pos->rb_node, root);
 }
 
 
-static int dso__load_kallsyms(struct dso *self, struct map *map,
+static int dso__load_kallsyms(struct dso *self, const char *filename, struct map *map,
 			      struct perf_session *session, symbol_filter_t filter)
 {
-	if (dso__load_all_kallsyms(self, map) < 0)
+	if (dso__load_all_kallsyms(self, filename, map) < 0)
 		return -1;
 
 	symbols__fixup_end(&self->symbols[map->type]);
@@ -933,11 +962,15 @@ static int dso__load_sym(struct dso *self, struct map *map,
 
 	elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) {
 		struct symbol *f;
-		const char *elf_name;
+		const char *elf_name = elf_sym__name(&sym, symstrs);
 		char *demangled = NULL;
 		int is_label = elf_sym__is_label(&sym);
 		const char *section_name;
 
+		if (kernel && session->ref_reloc_sym.name != NULL &&
+		    strcmp(elf_name, session->ref_reloc_sym.name) == 0)
+			perf_session__reloc_vmlinux_maps(session, sym.st_value);
+
 		if (!is_label && !elf_sym__is_a(&sym, map->type))
 			continue;
 
@@ -950,7 +983,6 @@ static int dso__load_sym(struct dso *self, struct map *map,
 		if (is_label && !elf_sec__is_a(&shdr, secstrs, map->type))
 			continue;
 
-		elf_name = elf_sym__name(&sym, symstrs);
 		section_name = elf_sec__name(&shdr, secstrs);
 
 		if (kernel || kmodule) {
@@ -1191,6 +1223,7 @@ char dso__symtab_origin(const struct dso *self)
 	static const char origin[] = {
 		[DSO__ORIG_KERNEL] =   'k',
 		[DSO__ORIG_JAVA_JIT] = 'j',
+		[DSO__ORIG_BUILD_ID_CACHE] = 'B',
 		[DSO__ORIG_FEDORA] =   'f',
 		[DSO__ORIG_UBUNTU] =   'u',
 		[DSO__ORIG_BUILDID] =  'b',
@@ -1209,6 +1242,7 @@ int dso__load(struct dso *self, struct map *map, struct perf_session *session,
 	int size = PATH_MAX;
 	char *name;
 	u8 build_id[BUILD_ID_SIZE];
+	char build_id_hex[BUILD_ID_SIZE * 2 + 1];
 	int ret = -1;
 	int fd;
 
@@ -1230,8 +1264,16 @@ int dso__load(struct dso *self, struct map *map, struct perf_session *session,
 		return ret;
 	}
 
-	self->origin = DSO__ORIG_FEDORA - 1;
+	self->origin = DSO__ORIG_BUILD_ID_CACHE;
 
+	if (self->has_build_id) {
+		build_id__sprintf(self->build_id, sizeof(self->build_id),
+				  build_id_hex);
+		snprintf(name, size, "%s/%s/.build-id/%.2s/%s",
+			 getenv("HOME"), DEBUG_CACHE_DIR,
+			 build_id_hex, build_id_hex + 2);
+		goto open_file;
+	}
 more:
 	do {
 		self->origin++;
@@ -1247,8 +1289,6 @@ more:
 		case DSO__ORIG_BUILDID:
 			if (filename__read_build_id(self->long_name, build_id,
 						    sizeof(build_id))) {
-				char build_id_hex[BUILD_ID_SIZE * 2 + 1];
-
 				build_id__sprintf(build_id, sizeof(build_id),
 						  build_id_hex);
 				snprintf(name, size,
@@ -1276,7 +1316,7 @@ compare_build_id:
 			if (!dso__build_id_equal(self, build_id))
 				goto more;
 		}
-
+open_file:
 		fd = open(name, O_RDONLY);
 	} while (fd < 0);
 
@@ -1309,13 +1349,33 @@ struct map *map_groups__find_by_name(struct map_groups *self,
 	for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) {
 		struct map *map = rb_entry(nd, struct map, rb_node);
 
-		if (map->dso && strcmp(map->dso->name, name) == 0)
+		if (map->dso && strcmp(map->dso->short_name, name) == 0)
 			return map;
 	}
 
 	return NULL;
 }
 
+static int dso__kernel_module_get_build_id(struct dso *self)
+{
+	char filename[PATH_MAX];
+	/*
+	 * kernel module short names are of the form "[module]" and
+	 * we need just "module" here.
+	 */
+	const char *name = self->short_name + 1;
+
+	snprintf(filename, sizeof(filename),
+		 "/sys/module/%.*s/notes/.note.gnu.build-id",
+		 (int)strlen(name - 1), name);
+
+	if (sysfs__read_build_id(filename, self->build_id,
+				 sizeof(self->build_id)) == 0)
+		self->has_build_id = true;
+
+	return 0;
+}
+
 static int perf_session__set_modules_path_dir(struct perf_session *self, char *dirname)
 {
 	struct dirent *dent;
@@ -1361,6 +1421,7 @@ static int perf_session__set_modules_path_dir(struct perf_session *self, char *d
 			if (long_name == NULL)
 				goto failure;
 			dso__set_long_name(map->dso, long_name);
+			dso__kernel_module_get_build_id(map->dso);
 		}
 	}
 
@@ -1403,6 +1464,24 @@ static struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
 	return self;
 }
 
+struct map *perf_session__new_module_map(struct perf_session *self, u64 start,
+					 const char *filename)
+{
+	struct map *map;
+	struct dso *dso = __dsos__findnew(&dsos__kernel, filename);
+
+	if (dso == NULL)
+		return NULL;
+
+	map = map__new2(start, dso, MAP__FUNCTION);
+	if (map == NULL)
+		return NULL;
+
+	dso->origin = DSO__ORIG_KMODULE;
+	map_groups__insert(&self->kmaps, map);
+	return map;
+}
+
 static int perf_session__create_module_maps(struct perf_session *self)
 {
 	char *line = NULL;
@@ -1416,7 +1495,6 @@ static int perf_session__create_module_maps(struct perf_session *self)
 	while (!feof(file)) {
 		char name[PATH_MAX];
 		u64 start;
-		struct dso *dso;
 		char *sep;
 		int line_len;
 
@@ -1442,26 +1520,10 @@ static int perf_session__create_module_maps(struct perf_session *self)
 		*sep = '\0';
 
 		snprintf(name, sizeof(name), "[%s]", line);
-		dso = dso__new(name);
-
-		if (dso == NULL)
+		map = perf_session__new_module_map(self, start, name);
+		if (map == NULL)
 			goto out_delete_line;
-
-		map = map__new2(start, dso, MAP__FUNCTION);
-		if (map == NULL) {
-			dso__delete(dso);
-			goto out_delete_line;
-		}
-
-		snprintf(name, sizeof(name),
-			 "/sys/module/%s/notes/.note.gnu.build-id", line);
-		if (sysfs__read_build_id(name, dso->build_id,
-					 sizeof(dso->build_id)) == 0)
-			dso->has_build_id = true;
-
-		dso->origin = DSO__ORIG_KMODULE;
-		map_groups__insert(&self->kmaps, map);
-		dsos__add(&dsos__kernel, dso);
+		dso__kernel_module_get_build_id(map->dso);
 	}
 
 	free(line);
@@ -1510,51 +1572,117 @@ static int dso__load_vmlinux(struct dso *self, struct map *map,
 		return -1;
 
 	dso__set_loaded(self, map->type);
-	err = dso__load_sym(self, map, session, self->long_name, fd, filter, 1, 0);
+	err = dso__load_sym(self, map, session, vmlinux, fd, filter, 1, 0);
 	close(fd);
 
 	return err;
 }
 
+int dso__load_vmlinux_path(struct dso *self, struct map *map,
+			   struct perf_session *session, symbol_filter_t filter)
+{
+	int i, err = 0;
+
+	pr_debug("Looking at the vmlinux_path (%d entries long)\n",
+		 vmlinux_path__nr_entries);
+
+	for (i = 0; i < vmlinux_path__nr_entries; ++i) {
+		err = dso__load_vmlinux(self, map, session, vmlinux_path[i],
+					filter);
+		if (err > 0) {
+			pr_debug("Using %s for symbols\n", vmlinux_path[i]);
+			dso__set_long_name(self, strdup(vmlinux_path[i]));
+			break;
+		}
+	}
+
+	return err;
+}
+
 static int dso__load_kernel_sym(struct dso *self, struct map *map,
 				struct perf_session *session, symbol_filter_t filter)
 {
 	int err;
-	bool is_kallsyms;
+	const char *kallsyms_filename = NULL;
+	char *kallsyms_allocated_filename = NULL;
+	/*
+	 * Step 1: if the user specified a vmlinux filename, use it and only
+	 * it, reporting errors to the user if it cannot be used.
+	 *
+	 * For instance, try to analyse an ARM perf.data file _without_ a
+	 * build-id, or if the user specifies the wrong path to the right
+	 * vmlinux file, obviously we can't fallback to another vmlinux (a
+	 * x86_86 one, on the machine where analysis is being performed, say),
+	 * or worse, /proc/kallsyms.
+	 *
+	 * If the specified file _has_ a build-id and there is a build-id
+	 * section in the perf.data file, we will still do the expected
+	 * validation in dso__load_vmlinux and will bail out if they don't
+	 * match.
+	 */
+	if (symbol_conf.vmlinux_name != NULL) {
+		err = dso__load_vmlinux(self, map, session,
+					symbol_conf.vmlinux_name, filter);
+		goto out_try_fixup;
+	}
 
 	if (vmlinux_path != NULL) {
-		int i;
-		pr_debug("Looking at the vmlinux_path (%d entries long)\n",
-			 vmlinux_path__nr_entries);
-		for (i = 0; i < vmlinux_path__nr_entries; ++i) {
-			err = dso__load_vmlinux(self, map, session,
-						vmlinux_path[i], filter);
-			if (err > 0) {
-				pr_debug("Using %s for symbols\n",
-					 vmlinux_path[i]);
-				dso__set_long_name(self,
-						   strdup(vmlinux_path[i]));
-				goto out_fixup;
+		err = dso__load_vmlinux_path(self, map, session, filter);
+		if (err > 0)
+			goto out_fixup;
+	}
+
+	/*
+	 * Say the kernel DSO was created when processing the build-id header table,
+	 * we have a build-id, so check if it is the same as the running kernel,
+	 * using it if it is.
+	 */
+	if (self->has_build_id) {
+		u8 kallsyms_build_id[BUILD_ID_SIZE];
+		char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+		if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id,
+					 sizeof(kallsyms_build_id)) == 0) {
+			if (dso__build_id_equal(self, kallsyms_build_id)) {
+				kallsyms_filename = "/proc/kallsyms";
+				goto do_kallsyms;
 			}
 		}
-	}
+		/*
+		 * Now look if we have it on the build-id cache in
+		 * $HOME/.debug/[kernel.kallsyms].
+		 */
+		build_id__sprintf(self->build_id, sizeof(self->build_id),
+				  sbuild_id);
+
+		if (asprintf(&kallsyms_allocated_filename,
+			     "%s/.debug/[kernel.kallsyms]/%s",
+			     getenv("HOME"), sbuild_id) == -1)
+			return -1;
 
-	is_kallsyms = self->long_name[0] == '[';
-	if (is_kallsyms)
-		goto do_kallsyms;
+		kallsyms_filename = kallsyms_allocated_filename;
 
-	err = dso__load_vmlinux(self, map, session, self->long_name, filter);
-	if (err <= 0) {
-		pr_info("The file %s cannot be used, "
-			"trying to use /proc/kallsyms...", self->long_name);
-do_kallsyms:
-		err = dso__load_kallsyms(self, map, session, filter);
-		if (err > 0 && !is_kallsyms)
-                        dso__set_long_name(self, strdup("[kernel.kallsyms]"));
+		if (access(kallsyms_filename, F_OK)) {
+			free(kallsyms_allocated_filename);
+			return -1;
+		}
+	} else {
+		/*
+		 * Last resort, if we don't have a build-id and couldn't find
+		 * any vmlinux file, try the running kernel kallsyms table.
+		 */
+		kallsyms_filename = "/proc/kallsyms";
 	}
 
+do_kallsyms:
+	err = dso__load_kallsyms(self, kallsyms_filename, map, session, filter);
+	free(kallsyms_allocated_filename);
+
+out_try_fixup:
 	if (err > 0) {
 out_fixup:
+		if (kallsyms_filename != NULL)
+			dso__set_long_name(self, strdup("[kernel.kallsyms]"));
 		map__fixup_start(map);
 		map__fixup_end(map);
 	}
@@ -1576,19 +1704,19 @@ static struct dso *dsos__find(struct list_head *head, const char *name)
 	struct dso *pos;
 
 	list_for_each_entry(pos, head, node)
-		if (strcmp(pos->name, name) == 0)
+		if (strcmp(pos->long_name, name) == 0)
 			return pos;
 	return NULL;
 }
 
-struct dso *dsos__findnew(const char *name)
+struct dso *__dsos__findnew(struct list_head *head, const char *name)
 {
-	struct dso *dso = dsos__find(&dsos__user, name);
+	struct dso *dso = dsos__find(head, name);
 
 	if (!dso) {
 		dso = dso__new(name);
 		if (dso != NULL) {
-			dsos__add(&dsos__user, dso);
+			dsos__add(head, dso);
 			dso__set_basename(dso);
 		}
 	}
@@ -1613,42 +1741,59 @@ void dsos__fprintf(FILE *fp)
 	__dsos__fprintf(&dsos__user, fp);
 }
 
-static size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp)
+static size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp,
+				      bool with_hits)
 {
 	struct dso *pos;
 	size_t ret = 0;
 
 	list_for_each_entry(pos, head, node) {
+		if (with_hits && !pos->hit)
+			continue;
 		ret += dso__fprintf_buildid(pos, fp);
 		ret += fprintf(fp, " %s\n", pos->long_name);
 	}
 	return ret;
 }
 
-size_t dsos__fprintf_buildid(FILE *fp)
+size_t dsos__fprintf_buildid(FILE *fp, bool with_hits)
 {
-	return (__dsos__fprintf_buildid(&dsos__kernel, fp) +
-		__dsos__fprintf_buildid(&dsos__user, fp));
+	return (__dsos__fprintf_buildid(&dsos__kernel, fp, with_hits) +
+		__dsos__fprintf_buildid(&dsos__user, fp, with_hits));
 }
 
-static struct dso *dsos__create_kernel( const char *vmlinux)
+struct dso *dso__new_kernel(const char *name)
 {
-	struct dso *kernel = dso__new(vmlinux ?: "[kernel.kallsyms]");
+	struct dso *self = dso__new(name ?: "[kernel.kallsyms]");
+
+	if (self != NULL) {
+		self->short_name = "[kernel]";
+		self->kernel	 = 1;
+	}
+
+	return self;
+}
+
+void dso__read_running_kernel_build_id(struct dso *self)
+{
+	if (sysfs__read_build_id("/sys/kernel/notes", self->build_id,
+				 sizeof(self->build_id)) == 0)
+		self->has_build_id = true;
+}
+
+static struct dso *dsos__create_kernel(const char *vmlinux)
+{
+	struct dso *kernel = dso__new_kernel(vmlinux);
 
 	if (kernel == NULL)
 		return NULL;
 
-	kernel->short_name = "[kernel]";
-	kernel->kernel	   = 1;
-
 	vdso = dso__new("[vdso]");
 	if (vdso == NULL)
 		goto out_delete_kernel_dso;
 	dso__set_loaded(vdso, MAP__FUNCTION);
 
-	if (sysfs__read_build_id("/sys/kernel/notes", kernel->build_id,
-				 sizeof(kernel->build_id)) == 0)
-		kernel->has_build_id = true;
+	dso__read_running_kernel_build_id(kernel);
 
 	dsos__add(&dsos__kernel, kernel);
 	dsos__add(&dsos__user, vdso);
@@ -1660,32 +1805,37 @@ out_delete_kernel_dso:
 	return NULL;
 }
 
-static int map_groups__create_kernel_maps(struct map_groups *self, const char *vmlinux)
+int __map_groups__create_kernel_maps(struct map_groups *self,
+				     struct map *vmlinux_maps[MAP__NR_TYPES],
+				     struct dso *kernel)
 {
-	struct map *functions, *variables;
-	struct dso *kernel = dsos__create_kernel(vmlinux);
-
-	if (kernel == NULL)
-		return -1;
+	enum map_type type;
 
-	functions = map__new2(0, kernel, MAP__FUNCTION);
-	if (functions == NULL)
-		return -1;
+	for (type = 0; type < MAP__NR_TYPES; ++type) {
+		vmlinux_maps[type] = map__new2(0, kernel, type);
+		if (vmlinux_maps[type] == NULL)
+			return -1;
 
-	variables = map__new2(0, kernel, MAP__VARIABLE);
-	if (variables == NULL) {
-		map__delete(functions);
-		return -1;
+		vmlinux_maps[type]->map_ip =
+			vmlinux_maps[type]->unmap_ip = identity__map_ip;
+		map_groups__insert(self, vmlinux_maps[type]);
 	}
 
-	functions->map_ip = functions->unmap_ip =
-		variables->map_ip = variables->unmap_ip = identity__map_ip;
-	map_groups__insert(self, functions);
-	map_groups__insert(self, variables);
-
 	return 0;
 }
 
+static int map_groups__create_kernel_maps(struct map_groups *self,
+					  struct map *vmlinux_maps[MAP__NR_TYPES],
+					  const char *vmlinux)
+{
+	struct dso *kernel = dsos__create_kernel(vmlinux);
+
+	if (kernel == NULL)
+		return -1;
+
+	return __map_groups__create_kernel_maps(self, vmlinux_maps, kernel);
+}
+
 static void vmlinux_path__exit(void)
 {
 	while (--vmlinux_path__nr_entries >= 0) {
@@ -1793,7 +1943,7 @@ out_free_comm_list:
 
 int perf_session__create_kernel_maps(struct perf_session *self)
 {
-	if (map_groups__create_kernel_maps(&self->kmaps,
+	if (map_groups__create_kernel_maps(&self->kmaps, self->vmlinux_maps,
 					   symbol_conf.vmlinux_name) < 0)
 		return -1;
 
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 8aded2356f79..124302778c09 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -8,6 +8,8 @@
 #include <linux/rbtree.h>
 #include "event.h"
 
+#define DEBUG_CACHE_DIR ".debug"
+
 #ifdef HAVE_CPLUS_DEMANGLE
 extern char *cplus_demangle(const char *, int);
 
@@ -58,7 +60,8 @@ struct symbol_conf {
 			sort_by_name,
 			show_nr_samples,
 			use_callchain,
-			exclude_other;
+			exclude_other,
+			full_paths;
 	const char	*vmlinux_name,
 			*field_sep;
 	char            *dso_list_str,
@@ -94,6 +97,7 @@ struct dso {
 	u8		 slen_calculated:1;
 	u8		 has_build_id:1;
 	u8		 kernel:1;
+	u8		 hit:1;
 	unsigned char	 origin;
 	u8		 sorted_by_name;
 	u8		 loaded;
@@ -105,6 +109,7 @@ struct dso {
 };
 
 struct dso *dso__new(const char *name);
+struct dso *dso__new_kernel(const char *name);
 void dso__delete(struct dso *self);
 
 bool dso__loaded(const struct dso *self, enum map_type type);
@@ -112,18 +117,30 @@ bool dso__sorted_by_name(const struct dso *self, enum map_type type);
 
 void dso__sort_by_name(struct dso *self, enum map_type type);
 
+extern struct list_head dsos__user, dsos__kernel;
+
+struct dso *__dsos__findnew(struct list_head *head, const char *name);
+
+static inline struct dso *dsos__findnew(const char *name)
+{
+	return __dsos__findnew(&dsos__user, name);
+}
+
 struct perf_session;
 
-struct dso *dsos__findnew(const char *name);
 int dso__load(struct dso *self, struct map *map, struct perf_session *session,
 	      symbol_filter_t filter);
+int dso__load_vmlinux_path(struct dso *self, struct map *map,
+			   struct perf_session *session, symbol_filter_t filter);
 void dsos__fprintf(FILE *fp);
-size_t dsos__fprintf_buildid(FILE *fp);
+size_t dsos__fprintf_buildid(FILE *fp, bool with_hits);
 
 size_t dso__fprintf_buildid(struct dso *self, FILE *fp);
 size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp);
 char dso__symtab_origin(const struct dso *self);
+void dso__set_long_name(struct dso *self, char *name);
 void dso__set_build_id(struct dso *self, void *build_id);
+void dso__read_running_kernel_build_id(struct dso *self);
 struct symbol *dso__find_symbol(struct dso *self, enum map_type type, u64 addr);
 struct symbol *dso__find_symbol_by_name(struct dso *self, enum map_type type,
 					const char *name);
@@ -131,11 +148,17 @@ struct symbol *dso__find_symbol_by_name(struct dso *self, enum map_type type,
 int filename__read_build_id(const char *filename, void *bf, size_t size);
 int sysfs__read_build_id(const char *filename, void *bf, size_t size);
 bool dsos__read_build_ids(void);
-int build_id__sprintf(u8 *self, int len, char *bf);
+int build_id__sprintf(const u8 *self, int len, char *bf);
+int kallsyms__parse(const char *filename, void *arg,
+		    int (*process_symbol)(void *arg, const char *name,
+					  char type, u64 start));
 
 int symbol__init(void);
+bool symbol_type__is_a(char symbol_type, enum map_type map_type);
+
 int perf_session__create_kernel_maps(struct perf_session *self);
 
-extern struct list_head dsos__user, dsos__kernel;
+struct map *perf_session__new_module_map(struct perf_session *self, u64 start,
+					 const char *filename);
 extern struct dso *vdso;
 #endif /* __PERF_SYMBOL */
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index c206f72c8881..e35653c1817c 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -48,6 +48,11 @@ static inline struct map *thread__find_map(struct thread *self,
 	return self ? map_groups__find(&self->mg, type, addr) : NULL;
 }
 
+void thread__find_addr_map(struct thread *self,
+			   struct perf_session *session, u8 cpumode,
+			   enum map_type type, u64 addr,
+			   struct addr_location *al);
+
 void thread__find_addr_location(struct thread *self,
 				struct perf_session *session, u8 cpumode,
 				enum map_type type, u64 addr,
@@ -67,4 +72,8 @@ map_groups__find_function(struct map_groups *self, struct perf_session *session,
 
 struct map *map_groups__find_by_name(struct map_groups *self,
 				     enum map_type type, const char *name);
+
+int __map_groups__create_kernel_maps(struct map_groups *self,
+				     struct map *vmlinux_maps[MAP__NR_TYPES],
+				     struct dso *kernel);
 #endif	/* __PERF_THREAD_H */
diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
index cace35595530..5ea8973ad331 100644
--- a/tools/perf/util/trace-event-info.c
+++ b/tools/perf/util/trace-event-info.c
@@ -20,6 +20,7 @@
  */
 #define _GNU_SOURCE
 #include <dirent.h>
+#include <mntent.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -37,6 +38,7 @@
 
 #include "../perf.h"
 #include "trace-event.h"
+#include "debugfs.h"
 
 #define VERSION "0.5"
 
@@ -101,32 +103,12 @@ void *malloc_or_die(unsigned int size)
 
 static const char *find_debugfs(void)
 {
-	static char debugfs[MAX_PATH+1];
-	static int debugfs_found;
-	char type[100];
-	FILE *fp;
-
-	if (debugfs_found)
-		return debugfs;
-
-	if ((fp = fopen("/proc/mounts","r")) == NULL)
-		die("Can't open /proc/mounts for read");
-
-	while (fscanf(fp, "%*s %"
-		      STR(MAX_PATH)
-		      "s %99s %*s %*d %*d\n",
-		      debugfs, type) == 2) {
-		if (strcmp(type, "debugfs") == 0)
-			break;
-	}
-	fclose(fp);
-
-	if (strcmp(type, "debugfs") != 0)
-		die("debugfs not mounted, please mount");
+	const char *path = debugfs_mount(NULL);
 
-	debugfs_found = 1;
+	if (!path)
+		die("Your kernel not support debugfs filesystem");
 
-	return debugfs;
+	return path;
 }
 
 /*
@@ -271,6 +253,8 @@ static void read_header_files(void)
 	write_or_die("header_page", 12);
 	write_or_die(&size, 8);
 	check_size = copy_file_fd(fd);
+	close(fd);
+
 	if (size != check_size)
 		die("wrong size for '%s' size=%lld read=%lld",
 		    path, size, check_size);
@@ -289,6 +273,7 @@ static void read_header_files(void)
 	if (size != check_size)
 		die("wrong size for '%s'", path);
 	put_tracing_file(path);
+	close(fd);
 }
 
 static bool name_in_tp_list(char *sys, struct tracepoint_path *tps)
@@ -317,7 +302,8 @@ static void copy_event_system(const char *sys, struct tracepoint_path *tps)
 		die("can't read directory '%s'", sys);
 
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    !name_in_tp_list(dent->d_name, tps))
 			continue;
@@ -334,7 +320,8 @@ static void copy_event_system(const char *sys, struct tracepoint_path *tps)
 
 	rewinddir(dir);
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    !name_in_tp_list(dent->d_name, tps))
 			continue;
@@ -353,6 +340,7 @@ static void copy_event_system(const char *sys, struct tracepoint_path *tps)
 
 		free(format);
 	}
+	closedir(dir);
 }
 
 static void read_ftrace_files(struct tracepoint_path *tps)
@@ -394,26 +382,21 @@ static void read_event_files(struct tracepoint_path *tps)
 		die("can't read directory '%s'", path);
 
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    strcmp(dent->d_name, "ftrace") == 0 ||
 		    !system_in_tp_list(dent->d_name, tps))
 			continue;
-		sys = malloc_or_die(strlen(path) + strlen(dent->d_name) + 2);
-		sprintf(sys, "%s/%s", path, dent->d_name);
-		ret = stat(sys, &st);
-		free(sys);
-		if (ret < 0)
-			continue;
-		if (S_ISDIR(st.st_mode))
-			count++;
+		count++;
 	}
 
 	write_or_die(&count, 4);
 
 	rewinddir(dir);
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    strcmp(dent->d_name, "ftrace") == 0 ||
 		    !system_in_tp_list(dent->d_name, tps))
@@ -422,14 +405,13 @@ static void read_event_files(struct tracepoint_path *tps)
 		sprintf(sys, "%s/%s", path, dent->d_name);
 		ret = stat(sys, &st);
 		if (ret >= 0) {
-			if (S_ISDIR(st.st_mode)) {
-				write_or_die(dent->d_name, strlen(dent->d_name) + 1);
-				copy_event_system(sys, tps);
-			}
+			write_or_die(dent->d_name, strlen(dent->d_name) + 1);
+			copy_event_system(sys, tps);
 		}
 		free(sys);
 	}
 
+	closedir(dir);
 	put_tracing_file(path);
 }
 
@@ -533,7 +515,7 @@ int read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events)
 	write_or_die(buf, 1);
 
 	/* save page_size */
-	page_size = getpagesize();
+	page_size = sysconf(_SC_PAGESIZE);
 	write_or_die(&page_size, 4);
 
 	read_header_files();
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index c5c32be040bf..c4b3cb8a02b1 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -1925,6 +1925,15 @@ void *raw_field_ptr(struct event *event, const char *name, void *data)
 	if (!field)
 		return NULL;
 
+	if (field->flags & FIELD_IS_STRING) {
+		int offset;
+
+		offset = *(int *)(data + field->offset);
+		offset &= 0xffff;
+
+		return data + offset;
+	}
+
 	return data + field->offset;
 }
 
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
index 1744422cafcb..ca3c26d466f3 100644
--- a/tools/perf/util/trace-event-read.c
+++ b/tools/perf/util/trace-event-read.c
@@ -83,7 +83,7 @@ static char *read_string(void)
 	char *str = NULL;
 	int size = 0;
 	int i;
-	int r;
+	s64 r;
 
 	for (;;) {
 		r = read(input_fd, buf, BUFSIZ);
@@ -117,7 +117,7 @@ static char *read_string(void)
 	i++;
 
 	/* move the file descriptor to the end of the string */
-	r = lseek(input_fd, -(r - i), SEEK_CUR);
+	r = lseek64(input_fd, -(r - i), SEEK_CUR);
 	if (r < 0)
 		die("lseek");
 
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
new file mode 100644
index 000000000000..f9b890fde681
--- /dev/null
+++ b/tools/perf/util/util.c
@@ -0,0 +1,94 @@
+#include "util.h"
+#include <sys/mman.h>
+
+int mkdir_p(char *path, mode_t mode)
+{
+	struct stat st;
+	int err;
+	char *d = path;
+
+	if (*d != '/')
+		return -1;
+
+	if (stat(path, &st) == 0)
+		return 0;
+
+	while (*++d == '/');
+
+	while ((d = strchr(d, '/'))) {
+		*d = '\0';
+		err = stat(path, &st) && mkdir(path, mode);
+		*d++ = '/';
+		if (err)
+			return -1;
+		while (*d == '/')
+			++d;
+	}
+	return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0;
+}
+
+static int slow_copyfile(const char *from, const char *to)
+{
+	int err = 0;
+	char *line = NULL;
+	size_t n;
+	FILE *from_fp = fopen(from, "r"), *to_fp;
+
+	if (from_fp == NULL)
+		goto out;
+
+	to_fp = fopen(to, "w");
+	if (to_fp == NULL)
+		goto out_fclose_from;
+
+	while (getline(&line, &n, from_fp) > 0)
+		if (fputs(line, to_fp) == EOF)
+			goto out_fclose_to;
+	err = 0;
+out_fclose_to:
+	fclose(to_fp);
+	free(line);
+out_fclose_from:
+	fclose(from_fp);
+out:
+	return err;
+}
+
+int copyfile(const char *from, const char *to)
+{
+	int fromfd, tofd;
+	struct stat st;
+	void *addr;
+	int err = -1;
+
+	if (stat(from, &st))
+		goto out;
+
+	if (st.st_size == 0) /* /proc? do it slowly... */
+		return slow_copyfile(from, to);
+
+	fromfd = open(from, O_RDONLY);
+	if (fromfd < 0)
+		goto out;
+
+	tofd = creat(to, 0755);
+	if (tofd < 0)
+		goto out_close_from;
+
+	addr = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fromfd, 0);
+	if (addr == MAP_FAILED)
+		goto out_close_to;
+
+	if (write(tofd, addr, st.st_size) == st.st_size)
+		err = 0;
+
+	munmap(addr, st.st_size);
+out_close_to:
+	close(tofd);
+	if (err)
+		unlink(to);
+out_close_from:
+	close(fromfd);
+out:
+	return err;
+}
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index c673d8825883..0f5b2a6f1080 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -403,4 +403,7 @@ void git_qsort(void *base, size_t nmemb, size_t size,
 #endif
 #endif
 
+int mkdir_p(char *path, mode_t mode);
+int copyfile(const char *from, const char *to);
+
 #endif
diff --git a/tools/perf/util/values.c b/tools/perf/util/values.c
index 1c15e39f99e3..cfa55d686e3b 100644
--- a/tools/perf/util/values.c
+++ b/tools/perf/util/values.c
@@ -169,6 +169,7 @@ static void perf_read_values__display_pretty(FILE *fp,
 				counterwidth[j], values->value[i][j]);
 		fprintf(fp, "\n");
 	}
+	free(counterwidth);
 }
 
 static void perf_read_values__display_raw(FILE *fp,