Since software events are always schedulable, mixing them up with
hardware events (who are not) can lead to funny scheduling oddities.

Giving them their own context solves this.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Provide the infrastructure for multiple task contexts.

A more flexible approach would have resulted in more pointer chases
in the scheduling hot-paths. This approach has the limitation of a
static number of task contexts.

Since I expect most external PMUs to be system wide, or at least node
wide (as per the intel uncore unit) they won't actually need a task
context.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Unify the two perf_event_context allocation sites.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Move all inherit code near each other.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Allocate per-cpu contexts per pmu.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Give each cpu-context its own timer so that it is a self contained
entity, this eases the way for per-pmu-per-cpu contexts as well as
provides the basic infrastructure to allow different rotation
times per pmu.

Things to look at:
 - folding the tick and these TICK_NSEC timers
 - separate task context rotation
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Separate the swevent hash-table from the cpu_context bits in
preparation for per pmu cpu contexts.

This keeps the swevent hash a global entity.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Separate find_get_context() from the event allocation and
initialization so that we may make find_get_context() depend
on the event pmu in a later patch.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Neither the overcommit nor the reservation sysfs parameter were
actually working, remove them as they'll only get in the way.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Replace pmu::{enable,disable,start,stop,unthrottle} with
pmu::{add,del,start,stop}, all of which take a flags argument.

The new interface extends the capability to stop a counter while
keeping it scheduled on the PMU. We replace the throttled state with
the generic stopped state.

This also allows us to efficiently stop/start counters over certain
code paths (like IRQ handlers).

It also allows scheduling a counter without it starting, allowing for
a generic frozen state (useful for rotating stopped counters).

The stopped state is implemented in two different ways, depending on
how the architecture implemented the throttled state:

 1) We disable the counter:
    a) the pmu has per-counter enable bits, we flip that
    b) we program a NOP event, preserving the counter state

 2) We store the counter state and ignore all read/overflow events
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Use hw_perf_event::period_left instead of hw_perf_event::remaining
and win back 8 bytes.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Provide default implementations for the pmu txn methods, this
allows us to remove some conditional code.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Changes perf_disable() into perf_pmu_disable().
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Since the current perf_disable() usage is only an optimization,
remove it for now. This eases the removal of the __weak
hw_perf_enable() interface.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Fixup random annoying style bits.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Simple registration interface for struct pmu, this provides the
infrastructure for removing all the weak functions.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

sed -ie 's/const struct pmu\>/struct pmu/g' `git grep -l "const struct pmu\>"`
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Since we have UP_PREPARE, we should also have UP_CANCELED.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Commit 1c024eca (perf, trace: Optimize tracepoints by using
per-tracepoint-per-cpu hlist to track events) caused a module
refcount leak.
Reported-And-Tested-by: NAvi Kivity <avi@redhat.com>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <4C7E1F12.8030304@cn.fujitsu.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Check the argument name whether it is invalid (not C-like symbol name). This
makes event format simple.
Reported-by: NSrikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
LKML-Reference: <20100827113912.22882.62313.stgit@ltc236.sdl.hitachi.co.jp>
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Set "argN" name for each argument automatically if it has no specified name.
Since dynamic trace event(kprobe_events) accepts special characters for its
argument, its format can show those special characters (e.g. '$', '%', '+').
However, perf can't parse those format because of the character (especially
'%') mess up the format.  This sets "argX" name for those arguments if user
omitted the argument names.

E.g.
 # echo 'p do_fork %ax IP=%ip $stack' > tracing/kprobe_events
 # cat tracing/kprobe_events
 p:kprobes/p_do_fork_0 do_fork arg1=%ax IP=%ip arg3=$stack
Reported-by: NSrikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
LKML-Reference: <20100827113906.22882.59312.stgit@ltc236.sdl.hitachi.co.jp>
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Fix a memory leak which happens when a field name conflicts with others. In
error case, free_trace_probe() will free all arguments until nr_args, so this
increments nr_args the begining of the loop instead of the end.

Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
LKML-Reference: <20100827113846.22882.12670.stgit@ltc236.sdl.hitachi.co.jp>
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Reading the file set_ftrace_filter does three things.

1) shows whether or not filters are set for the function tracer
2) shows what functions are set for the function tracer
3) shows what triggers are set on any functions

3 is independent from 1 and 2.

The way this file currently works is that it is a state machine,
and as you read it, it may change state. But this assumption breaks
when you use lseek() on the file. The state machine gets out of sync
and the t_show() may use the wrong pointer and cause a kernel oops.

Luckily, this will only kill the app that does the lseek, but the app
dies while holding a mutex. This prevents anyone else from using the
set_ftrace_filter file (or any other function tracing file for that matter).

A real fix for this is to rewrite the code, but that is too much for
a -rc release or stable. This patch simply disables llseek on the
set_ftrace_filter() file for now, and we can do the proper fix for the
next major release.
Reported-by: NRobert Swiecki <swiecki@google.com>
Cc: Chris Wright <chrisw@sous-sol.org>
Cc: Tavis Ormandy <taviso@google.com>
Cc: Eugene Teo <eugene@redhat.com>
Cc: vendor-sec@lst.de
Cc: <stable@kernel.org>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

While discussing the strictness of the 80 character limit on the
Kernel Summit Discussion mailing list, I showed examples that I
broke that limit slightly with some algorithms. In discussing with
John Linville, what looked better, I realized that two of the
80 char breaking culprits were an identical expression.

As a clean up, this patch moves the identical expression into its
own helper function and that is used instead. As a side effect,
the offending code is now under the 80 character limit. :-)

This clean up code also changes the expression from

	(A - B) - C  to  A - (B + C)

This makes the code look a little nicer too.

Cc: John W. Linville <linville@tuxdriver.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

During my rewrite, the semantics of touch_nmi_watchdog and
touch_softlockup_watchdog changed enough to break some drivers
(mostly over preemptable regions).

These are cases where long delays on one CPU (due to
print_delay for example) can cause long delays on other
CPUs - so we must 'touch' the nmi_watchdog flag of those
other CPUs as well.

This change brings those touch_*_watchdog() functions back in line
with to how they used to work.
Signed-off-by: NDon Zickus <dzickus@redhat.com>
Acked-by: NCyrill Gorcunov <gorcunov@openvz.org>
Cc: peterz@infradead.org
Cc: fweisbec@gmail.com
LKML-Reference: <1283310009-22168-2-git-send-email-dzickus@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The panic notifer in lockup_detector just set did_panic to 1.
But did_panic is not used anywhere so we can just remove it.
Signed-off-by: NAkinobu Mita <akinobu.mita@gmail.com>
Cc: peterz@infradead.org
Cc: gorcunov@gmail.com
Cc: fweisbec@gmail.com
LKML-Reference: <1283310009-22168-4-git-send-email-dzickus@redhat.com>
Signed-off-by: NDon Zickus <dzickus@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

By the commit e6bde73b
("cpu-hotplug: return better errno on cpu hotplug failure"),
the cpu notifier can return encapsulate errno value, resulting
in more meaningful error codes for CPU hotplug failures.

This converts the cpu notifier to return encapsulate errno value
for the lockup_detector as well.
Signed-off-by: NAkinobu Mita <akinobu.mita@gmail.com>
Cc: peterz@infradead.org
Cc: gorcunov@gmail.com
Cc: fweisbec@gmail.com
LKML-Reference: <1283310009-22168-3-git-send-email-dzickus@redhat.com>
Signed-off-by: NDon Zickus <dzickus@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

While we are reading trace_stat/functionX and someone just
disabled function_profile at that time, we can trigger this:

	divide error: 0000 [#1] PREEMPT SMP
	...
	EIP is at function_stat_show+0x90/0x230
	...

This fix just takes the ftrace_profile_lock and checks if
rec->counter is 0. If it's 0, we know the profile buffer
has been reset.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Cc: stable@kernel.org
LKML-Reference: <4C723644.4040708@cn.fujitsu.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

Per-thread events with a cpu filter, i.e., cpu != -1, were not
reporting correct timings when the thread never ran on the
monitored cpu. The time enabled was reported as a negative
value.

This patch fixes the problem by updating tstamp_stopped,
tstamp_running in event_sched_out() for events with filters and
which are marked as INACTIVE.

The function group_sched_out() is modified to systematically
call into event_sched_out() to avoid duplicating the timing
adjustment code twice.

With the patch, I now get:

$ task_cpu -i -e unhalted_core_cycles,unhalted_core_cycles
noploop 2 noploop for 2 seconds
CPU0 0		   unhalted_core_cycles (ena=1,991,136,594, run=0)
CPU0 0		   unhalted_core_cycles (ena=1,991,136,594, run=0)

CPU1 0		   unhalted_core_cycles (ena=1,991,136,594, run=0)
CPU1 0		   unhalted_core_cycles (ena=1,991,136,594, run=0)

CPU2 0		   unhalted_core_cycles (ena=1,991,136,594, run=0)
CPU2 0		   unhalted_core_cycles (ena=1,991,136,594, run=0)

CPU3 4,747,990,931 unhalted_core_cycles (ena=1,991,136,594, run=1,991,136,594)
CPU3 4,747,990,931 unhalted_core_cycles (ena=1,991,136,594, run=1,991,136,594)
Signed-off-by: NStephane Eranian <eranian@gmail.com>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus@samba.org
Cc: davem@davemloft.net
Cc: fweisbec@gmail.com
Cc: perfmon2-devel@lists.sf.net
Cc: eranian@google.com
LKML-Reference: <4c76802d.aae9d80a.115d.70fe@mx.google.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

save_stack_trace() stores the instruction pointer, not the
function descriptor. On ppc64 the trace stack code currently
dereferences the instruction pointer and shows 8 bytes of
instructions in our backtraces:

 # cat /sys/kernel/debug/tracing/stack_trace
        Depth    Size   Location    (26 entries)
        -----    ----   --------
  0)     5424     112   0x6000000048000004
  1)     5312     160   0x60000000ebad01b0
  2)     5152     160   0x2c23000041c20030
  3)     4992     240   0x600000007c781b79
  4)     4752     160   0xe84100284800000c
  5)     4592     192   0x600000002fa30000
  6)     4400     256   0x7f1800347b7407e0
  7)     4144     208   0xe89f0108f87f0070
  8)     3936     272   0xe84100282fa30000

Since we aren't dealing with function descriptors, use %pS
instead of %pF to fix it:

 # cat /sys/kernel/debug/tracing/stack_trace
        Depth    Size   Location    (26 entries)
        -----    ----   --------
  0)     5424     112   ftrace_call+0x4/0x8
  1)     5312     160   .current_io_context+0x28/0x74
  2)     5152     160   .get_io_context+0x48/0xa0
  3)     4992     240   .cfq_set_request+0x94/0x4c4
  4)     4752     160   .elv_set_request+0x60/0x84
  5)     4592     192   .get_request+0x2d4/0x468
  6)     4400     256   .get_request_wait+0x7c/0x258
  7)     4144     208   .__make_request+0x49c/0x610
  8)     3936     272   .generic_make_request+0x390/0x434
Signed-off-by: NAnton Blanchard <anton@samba.org>
Cc: rostedt@goodmis.org
Cc: fweisbec@gmail.com
LKML-Reference: <20100825013238.GE28360@kryten>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

There is a scalability issue for current implementation of optimistic
mutex spin in the kernel.  It is found on a 8 node 64 core Nehalem-EX
system (HT mode).

The intention of the optimistic mutex spin is to busy wait and spin on a
mutex if the owner of the mutex is running, in the hope that the mutex
will be released soon and be acquired, without the thread trying to
acquire mutex going to sleep. However, when we have a large number of
threads, contending for the mutex, we could have the mutex grabbed by
other thread, and then another ……, and we will keep spinning, wasting cpu
cycles and adding to the contention.  One possible fix is to quit
spinning and put the current thread on wait-list if mutex lock switch to
a new owner while we spin, indicating heavy contention (see the patch
included).

I did some testing on a 8 socket Nehalem-EX system with a total of 64
cores. Using Ingo's test-mutex program that creates/delete files with 256
threads (http://lkml.org/lkml/2006/1/8/50) , I see the following speed up
after putting in the mutex spin fix:

 ./mutex-test V 256 10
                 Ops/sec
 2.6.34          62864
 With fix        197200

Repeating the test with Aim7 fserver workload, again there is a speed up
with the fix:

                 Jobs/min
 2.6.34          91657
 With fix        149325

To look at the impact on the distribution of mutex acquisition time, I
collected the mutex acquisition time on Aim7 fserver workload with some
instrumentation.  The average acquisition time is reduced by 48% and
number of contentions reduced by 32%.

                 #contentions    Time to acquire mutex (cycles)
 2.6.34          72973           44765791
 With fix        49210           23067129

The histogram of mutex acquisition time is listed below.  The acquisition
time is in 2^bin cycles.  We see that without the fix, the acquisition
time is mostly around 2^26 cycles.  With the fix, we the distribution get
spread out a lot more towards the lower cycles, starting from 2^13.
However, there is an increase of the tail distribution with the fix at
2^28 and 2^29 cycles.  It seems a small price to pay for the reduced
average acquisition time and also getting the cpu to do useful work.

 Mutex acquisition time distribution (acq time = 2^bin cycles):
         2.6.34                  With Fix
 bin     #occurrence     %       #occurrence     %
 11      2               0.00%   120             0.24%
 12      10              0.01%   790             1.61%
 13      14              0.02%   2058            4.18%
 14      86              0.12%   3378            6.86%
 15      393             0.54%   4831            9.82%
 16      710             0.97%   4893            9.94%
 17      815             1.12%   4667            9.48%
 18      790             1.08%   5147            10.46%
 19      580             0.80%   6250            12.70%
 20      429             0.59%   6870            13.96%
 21      311             0.43%   1809            3.68%
 22      255             0.35%   2305            4.68%
 23      317             0.44%   916             1.86%
 24      610             0.84%   233             0.47%
 25      3128            4.29%   95              0.19%
 26      63902           87.69%  122             0.25%
 27      619             0.85%   286             0.58%
 28      0               0.00%   3536            7.19%
 29      0               0.00%   903             1.83%
 30      0               0.00%   0               0.00%

I've done similar experiments with 2.6.35 kernel on smaller boxes as
well.  One is on a dual-socket Westmere box (12 cores total, with HT).
Another experiment is on an old dual-socket Core 2 box (4 cores total, no
HT)

On the 12-core Westmere box, I see a 250% increase for Ingo's mutex-test
program with my mutex patch but no significant difference in aim7's
fserver workload.

On the 4-core Core 2 box, I see the difference with the patch for both
mutex-test and aim7 fserver are negligible.

So far, it seems like the patch has not caused regression on smaller
systems.
Signed-off-by: NTim Chen <tim.c.chen@linux.intel.com>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: <stable@kernel.org> # .35.x
LKML-Reference: <1282168827.9542.72.camel@schen9-DESK>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Stephane reported that when the machine locks up, the regular ticks,
which are responsible to resetting the throttle count, stop too.

Hence the NMI watchdog can end up being throttled before it reports on
the locked up state, and we end up being sad..

Cure this by having the watchdog overflow reset its own throttle count.
Reported-by: NStephane Eranian <eranian@google.com>
Tested-by: NStephane Eranian <eranian@google.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1282215916.1926.4696.camel@laptop>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

With the introduction of the new unified work queue thread pools,
we lost one feature: It's no longer possible to know which worker
is causing the CPU to wake out of idle. The result is that PowerTOP
now reports a lot of "kworker/a:b" instead of more readable results.

This patch adds a pair of tracepoints to the new workqueue code,
similar in style to the timer/hrtimer tracepoints.

With this pair of tracepoints, the next PowerTOP can correctly
report which work item caused the wakeup (and how long it took):

Interrupt (43)            i915      time   3.51ms    wakeups 141
Work      ieee80211_iface_work      time   0.81ms    wakeups  29
Work              do_dbs_timer      time   0.55ms    wakeups  24
Process                   Xorg      time  21.36ms    wakeups   4
Timer    sched_rt_period_timer      time   0.01ms    wakeups   1
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It's a really simple list, and several of the users want to go backwards
in it to find the previous vma.  So rather than have to look up the
previous entry with 'find_vma_prev()' or something similar, just make it
doubly linked instead.
Tested-by: NIan Campbell <ijc@hellion.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

kfifo_skip() is currently broken, due to the missing of the internal
helper function.  Add it.
Signed-off-by: NAndrea Righi <arighi@develer.com>
Cc: Greg KH <greg@kroah.com>
Acked-by: NStefani Seibold <stefani@seibold.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

ftrace_event_call->perf_events, perf_trace_buf,
fgraph_data->cpu_data and some local variables are percpu pointers
missing __percpu markups. Add them.
Signed-off-by: NNamhyung Kim <namhyung@gmail.com>
Acked-by: NTejun Heo <tj@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <1281498479-28551-1-git-send-email-namhyung@gmail.com>
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>

Instead of hardcoding the number of contexts for the recursions
barriers, define a cpp constant to make the code more
self-explanatory.
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>

Now that software events don't have interrupt disabled anymore in
the event path, callchains can nest on any context. So seperating
nmi and others contexts in two buffers has become racy.

Fix this by providing one buffer per nesting level. Given the size
of the callchain entries (2040 bytes * 4), we now need to allocate
them dynamically.

v2: Fixed put_callchain_entry call after recursion.
    Fix the type of the recursion, it must be an array.

v3: Use a manual pr cpu allocation (temporary solution until NMIs
    can safely access vmalloc'ed memory).
    Do a better separation between callchain reference tracking and
    allocation. Make the "put" path lockless for non-release cases.

v4: Protect the callchain buffers with rcu.

v5: Do the cpu buffers allocations node affine.
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Tested-by: NWill Deacon <will.deacon@arm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: David Miller <davem@davemloft.net>
Cc: Borislav Petkov <bp@amd64.org>

Store the kernel and user contexts from the generic layer instead
of archs, this gathers some repetitive code.
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Acked-by: NPaul Mackerras <paulus@samba.org>
Tested-by: NWill Deacon <will.deacon@arm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: David Miller <davem@davemloft.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Borislav Petkov <bp@amd64.org>

- Most archs use one callchain buffer per cpu, except x86 that needs
  to deal with NMIs. Provide a default perf_callchain_buffer()
  implementation that x86 overrides.

- Centralize all the kernel/user regs handling and invoke new arch
  handlers from there: perf_callchain_user() / perf_callchain_kernel()
  That avoid all the user_mode(), current->mm checks and so...

- Invert some parameters in perf_callchain_*() helpers: entry to the
  left, regs to the right, following the traditional (dst, src).
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Acked-by: NPaul Mackerras <paulus@samba.org>
Tested-by: NWill Deacon <will.deacon@arm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: David Miller <davem@davemloft.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Borislav Petkov <bp@amd64.org>