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>

callchain_store() is the same on every archs, inline it in
perf_event.h and rename it to perf_callchain_store() to avoid
any collision.

This removes 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>

Drop the TASK_RUNNING test on user tasks for callchains as
this check doesn't seem to make any sense.

Also remove the tests for !current that is not supposed to
happen and current->pid as this should be handled at the
generic level, with exclude_idle attribute.
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: David Miller <davem@davemloft.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Borislav Petkov <bp@amd64.org>

To match what is shown when '?' or 'H' is pressed, i.e. the keybind help
window.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

So that the common tasks of providing a helpline at __run entry and
destroying the window and releasing resourses at exit can be abstracted
away, reducing a bit more the coupling with libnewt.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

The annotate TUI now starts centered on the line with most samples, i.e.
the hottest line in the annotated function. Pressing TAB will center on
the second hottest function and so on. Shift+TAB goes in the other
direction.

This way one can more easily sift thru the function hotspots.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Not just on the annotate one.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Right now it will just sort and position at the hottest line, i.e.
the one where more samples were taken.

It will be at the center of the screen and later TAB/shift-TAB will
cycle thru the hottest lines.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

ARM ELF files use symbols with special names $a, $t, $d to identify regions of
ARM code, Thumb code and data within code sections.  This can cause confusing
output from the perf tools, especially for partially stripped binaries, or
binaries containing user-added zero-sized symbols (which may occur in
hand-written assembler which hasn't been fully annotated with .size
directives).

This patch filters out these symbols at load time.

LKML-Reference: <1281352878-8735-2-git-send-email-dave.martin@linaro.org>
Signed-off-by: NDave Martin <dave.martin@linaro.org>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

As new TUI features get added the newt.c file is growing a lot and its
name is growing misleading as an effort is being made to reduce the
coupling with libnewt.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

So that building other browser based on structures linked via a linked
list can be as easy as it is already for the ones linked via an rb_tree.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Fix several memory leaks of pkgs and tevs in add_perf_probe_events().
Reported-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: linux-kernel@vger.kernel.org
LKML-Reference: <4C577ADC.1000309@hitachi.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Copy type field if it is for raw parameters.
Without this fix, perf probe drops the type if user passes it
for raw parameters (e.g. %ax:u32 will be converted to %ax).

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
LKML-Reference: <4C577AD8.50808@hitachi.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Merge branch 'perf/core' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 into perf/core

When cmd_record exits the whole perf binary will exit right after,
so no need to traverse lots of complex data structures freeing them.

Sticked a comment for leak detectives and for a experiment with obstacks
to be performed so that we can speed up freeing that memory.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Outdent the code following the if.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@r disable braces4@
position p1,p2;
statement S1,S2;
@@

(
if (...) { ... }
|
if (...) S1@p1 S2@p2
)

@script:python@
p1 << r.p1;
p2 << r.p2;
@@

if (p1[0].column == p2[0].column):
  cocci.print_main("branch",p1)
  cocci.print_secs("after",p2)
// </smpl>

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ingo Molnar <mingo@elte.hu>
LKML-Reference: <Pine.LNX.4.64.1008052227330.31692@ask.diku.dk>
Signed-off-by: NJulia Lawall <julia@diku.dk>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Removed duplicated #includes util/trace-event.h and
util/exec_cmd.h.
Grouped and sorted all the #includes.

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <1281016299-23958-14-git-send-email-andrea.gelmini@gelma.net>
Signed-off-by: NAndrea Gelmini <andrea.gelmini@gelma.net>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <1281016299-23958-15-git-send-email-andrea.gelmini@gelma.net>
Signed-off-by: NAndrea Gelmini <andrea.gelmini@gelma.net>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Only in verbose mode so as not to bloat struct symbol too much.

The key used is '/', just like in vi, less, etc.

More work is needed to allocate space on the symbol in a more clear way.

This experiment shows how to do it for the hist_browser, in the main
window.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

By using BITS_PER_LONG/4 as the width specifier.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Otherwise entries will get chopped up on the window.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Press -> and then "Browse map details" to see the DSO long name as the title
and the list of symbols in the DSO used by the map where the current symbol is.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

So that tools that wan't to act only on a subset of (weak, global,
local) symbols can do so, such as the upcoming uprobes support in 'perf
probe'.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Newer Intel processors identifying themselves as model 30 are not recognized by
oprofile.

<cpuinfo snippet>
model           : 30
model name      : Intel(R) Xeon(R) CPU           X3470  @ 2.93GHz
</cpuinfo snippet>

Running oprofile on these machines gives the following:
+ opcontrol --init
+ opcontrol --list-events
oprofile: available events for CPU type "Intel Architectural Perfmon"

See Intel 64 and IA-32 Architectures Software Developer's Manual
Volume 3B (Document 253669) Chapter 18 for architectural perfmon events
This is a limited set of fallback events because oprofile doesn't know your CPU
CPU_CLK_UNHALTED: (counter: all)
        Clock cycles when not halted (min count: 6000)
INST_RETIRED: (counter: all)
        number of instructions retired (min count: 6000)
LLC_MISSES: (counter: all)
        Last level cache demand requests from this core that missed the LLC
(min count: 6000)
        Unit masks (default 0x41)
        ----------
        0x41: No unit mask
LLC_REFS: (counter: all)
        Last level cache demand requests from this core (min count: 6000)
        Unit masks (default 0x4f)
        ----------
        0x4f: No unit mask
BR_MISS_PRED_RETIRED: (counter: all)
        number of mispredicted branches retired (precise) (min count: 500)
+ opcontrol --shutdown

Tested using oprofile 0.9.6.
Signed-off-by: NJosh Hunt <johunt@akamai.com>
Reviewed-by: NAndi Kleen <ak@linux.intel.com>
Signed-off-by: NRobert Richter <robert.richter@amd.com>

Merge branch 'perf/core' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 into perf/core

Conflicts:
	kernel/Makefile

Merge reason: Add the now complete topic, fix the conflict.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Comment in unregister_trace_probe() says probe_lock will be held when it
gets called. However there is a case where it might called without the
probe_lock being held. Also since we are traversing the probe_list and
deleting an element from the probe_list, probe_lock should be held.

This was first pointed in uprobes traceevent review by Frederic
Weisbecker here.  (http://lkml.org/lkml/2010/5/12/106)

Cc: Ingo Molnar <mingo@elte.hu>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Acked-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Acked-by: NSteven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20100630084548.GA10325@linux.vnet.ibm.com>
Signed-off-by: NSrikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

The event__process function is useful in processing /proc/<pid>/maps.  All of
the functions that are called from event__process are defined in util/event.c.
Though its defined in builtin-top.c, it could be reused for perf probe for
uprobes. Hence moving it to util/event.c and exporting the function.

LKML-Reference: <20100802123851.GD22812@linux.vnet.ibm.com>
Signed-off-by: NSrikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Fix buggy-looking code which unnecessarily adjusts the file offset
fields read from /proc/*/maps.

This may have gone unnoticed since the offset is usually 0 (and the
logic in util/symbol.c may work incorrectly for other offset values).

Commiter note:

This fixes a bug introduced in 4af8b35d, there is no need to shift pgoff
twice, the show_map_vma routine in fs/proc/task_mmu.c already converts
it from the number of pages to the size in bytes, and that is what
appears in /proc/PID/map.

Cc: Nicolas Pitre <nicolas.pitre@linaro.org>
Cc: Will Deacon <Will.Deacon@arm.com>
LKML-Reference: <1280836116-6654-2-git-send-email-dave.martin@linaro.org>
Signed-off-by: NDave Martin <dave.martin@linaro.org>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>