Linus reported that the RCU lockdep annotation bits triggered for this
rcu_dereference() because we're not holding rcu_read_lock().

Going over the code I cannot convince myself its correct:

 - holding a ref on the parent_ctx, doesn't avoid it being uncloned
   concurrently (as the comment says), so we can race with a free.

 - holding parent_ctx->mutex doesn't avoid the above free from taking
   place either, it would at best avoid parent_ctx from being freed.

I.e. the warning is correct. To fix the bug, serialize against the
unclone_ctx() call by extending the reach of the parent_ctx->lock.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

It was broken by f006d25a that passed just the event name, not the complete
sys:event that it expected to open the /sys/.../sys/sys:event/id file to get
the id.

Fix it by moving it to after parse_events in cmd_record, as at that point
we can just traverse the evsel_list and use evsel->attr.config +
event_name(evsel) instead of re-opening the /id file.
Reported-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Han Pingtian <phan@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <20110117202801.GG2085@ghostprotocols.net>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

It wasn't accounting the ':' when consuming bytes in the the event
selector string, so parse_events() would fail in this test:

                if (!(*str == 0 || *str == ',' || isspace(*str)))
                        return -1;

as *str would be pointing to '*', the last character in the '-e' arg in:

$ perf record -q -a -D -e sched:sched_* | perf script -i - -s perf-script.py

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

When profiling a benchmark that is almost 100% userspace, I noticed some wildly
inaccurate profiles that showed almost all time spent in the kernel.

Closer examination shows we were programming a tiny number of cycles into the
PMU after each overflow (about ~200 away from the next overflow). This gets us
stuck in a loop which we eventually break out of by throttling the PMU (there
are regular throttle/unthrottle events in the log).

It looks like we aren't setting event->hw.last_period to something same and the
frequency to period calculations in perf are going haywire.

With the following patch we find the correct period after a few interrupts and
stay there. I also see no more throttle events.
Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
LKML-Reference: <20110117161742.5feb3761@kryten>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Merge branch 'tip/perf/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace into perf/urgent

There is no need for syscall_exit_fields as the syscall
exit event class can already host the fields in its structure,
like most other trace events do by default. Use that
default behavior instead.

Following this scheme, we don't need anymore to override the
get_fields() callback of the syscall exit event class either.

Hence both syscall_exit_fields and syscall_get_exit_fields() can
be removed.

Also changed some indentation to keep the following under 80
characters:

".fields		= LIST_HEAD_INIT(event_class_syscall_exit.fields),"
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
LKML-Reference: <4D301C0E.8090408@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

The commit:

 9f987b3141f086de27832514aad9f50a53f754
 tracing: Include module.h in define_trace.h

only solved half the problem. If the trace/events/module.h header is
included at the time of define_trace.h (or in ftrace.h within it),
the module.h TRACE_SYSTEM will override the current TRACE_SYSTEM
macro.

Since define_trace.h is included when CREATE_TRACE_POINTS is set,
and the first thing it does is to #undef CREATE_TRACE_POINTS,
by placing the module.h TRACE_SYSTEM inside a
 #ifdef CREATE_TRACE_POINTS
we can prevent it from overriding the TRACE_SYSTEM that is
being processed, and still process the module.h tracepoints
when the module code defines CREATE_TRACE_POINTS and includes
the trace/events/module.h header.

As with commit 9f987b3141, this is only an issue if module.h
is not included before the trace/events/<event>.h file is
included, which (luckily) has not happened yet.
Reported-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

Sometimes there is a need to use perf in "live-log" mode. The problem
is, for seldom events, actual info output is largely delayed because
perf-record reads sample data in whole pages.

So for such scenarious, add flag for perf-record to go in "nodelay"
mode. To track e.g. what's going on in icmp_rcv while ping is running
Use it with something like this:

(1) $ perf probe -L icmp_rcv | grep -U8 '^ *43\>'
                                    goto error;
                    }
         38         if (!pskb_pull(skb, sizeof(*icmph)))
                            goto error;
                    icmph = icmp_hdr(skb);

         43         ICMPMSGIN_INC_STATS_BH(net, icmph->type);
                    /*
                     *      18 is the highest 'known' ICMP type. Anything else is a mystery
                     *
                     *      RFC 1122: 3.2.2  Unknown ICMP messages types MUST be silently
                     *                discarded.
                     */
         50         if (icmph->type > NR_ICMP_TYPES)
                            goto error;

    $ perf probe icmp_rcv:43 'type=icmph->type'

(2) $ cat trace-icmp.py
    [...]
    def trace_begin():
            print "in trace_begin"

    def trace_end():
            print "in trace_end"

    def probe__icmp_rcv(event_name, context, common_cpu,
            common_secs, common_nsecs, common_pid, common_comm,
            __probe_ip, type):
                    print_header(event_name, common_cpu, common_secs, common_nsecs,
                            common_pid, common_comm)

                    print "__probe_ip=%u, type=%u\n" % \
                    (__probe_ip, type),
    [...]

(3) $ perf record -a -D -e probe:icmp_rcv -o - | \
      perf script -i - -s trace-icmp.py

Thanks to Peter Zijlstra for pointing how to do it.

Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>, Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <20110112140613.GA11698@tugrik.mns.mnsspb.ru>
Signed-off-by: Kirill Smelkov <kirr@mns.spb.ru>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Looks to me like the :r modifier is not supported anymore, so remove it from
the list of events.

Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Robert Richter <robert.richter@amd.com>
LKML-Reference: <AANLkTim=jawJyBj0iFd0r4-LCKzvjFW+NddzJMD5GUB9@mail.gmail.com>
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

This reverts commit aa7bc7ef.

It removed the fallback from hardware profiling to software profiling.
.e.g., in a VM with no PMU.
Reported-by: David Ahern <daahern@cisco.com>
Cc: David Ahern <daahern@cisco.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Before we had sym_counter, it was initialized to zero and we used that
as an index in the global attrs variable, now we have a list of evsel
entries, and sym_counter became sym_evsel, that remained initialized to
zero (NULL): b00m.

Fix it by initializing it to the first entry in the evsel list.

Bug-introduced: 69aad6f1Reported-by: Kirill Smelkov <kirr@mns.spb.ru>
Tested-by: Kirill Smelkov <kirr@mns.spb.ru>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Kirill Smelkov <kirr@mns.spb.ru>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

We need to defer calling perf_evsel_list__delete() till after atexit
registered routines, because we need to traverse the events being
recorded at that time at least on 'perf record'.

This fixes the problem reported by Thomas Renninger where cmd_record
called by cmd_timechart would not write the tracing data to the perf.data
file header because the evsel_list at atexit (control+C on 'perf timechart
record') time would be empty, being already deleted by run_builtin(),
and thus 'perf timechart' when trying to process such perf.data file would
die with:

"no trace data in the file"

Problem introduced in 70d544d0.
Reported-by: Thomas Renninger <trenn@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Renninger <trenn@suse.de>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

In this if statement:

        if (head + event->header.size >= mmap_size) {
                if (mmaps[map_idx]) {
                        munmap(mmaps[map_idx], mmap_size);
                        mmaps[map_idx] = NULL;
                }

                page_offset = page_size * (head / page_size);
                file_offset += page_offset;
                head -= page_offset;
                goto remap;
        }

With, for instance, these values:

head=2992
event->header.size=48
mmap_size=3040

We end up endlessly looping back to remap. Off by one.

Problem introduced in 55b4462.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Reported-by: Ingo Molnar <mingo@elte.hu>
Reported-by: David Ahern <daahern@cisco.com>
Bisected-by: David Ahern <daahern@cisco.com>
Tested-by: David Ahern <daahern@cisco.com>
Cc: David Ahern <daahern@cisco.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

And a test for it:

[acme@felicio linux]$ perf test
 1: vmlinux symtab matches kallsyms: Ok
 2: detect open syscall event: Ok
 3: detect open syscall event on all cpus: Ok
[acme@felicio linux]$

Translating C the test does:

1. generates different number of open syscalls on each CPU
   by using sched_setaffinity
2. Verifies that the expected number of events is generated
   on each CPU

It works as expected.

LKML-Reference: <new-submission>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

on ppc64:
/usr/include/bits/local_lim.h:#define PTHREAD_STACK_MIN	131072

therefore following set of commands:

gives:
perf.2.6.37test: builtin-sched.c:493: create_tasks: Assertion `!(err)' failed.

So make sure we do not set stack size lower than PTHREAD_STACK_MIN.

Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20110110160417.GB2685@psychotron.brq.redhat.com>
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Improve sys_perf_event_open ENOENT return handling in top and record, just
like 5a3446bc does for stat.

Cc: David Ahern <daahern@cisco.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

For unsupported events (e.g., H/W events when running in a VM)
perf stat currently fails with the error message:

      Error: open_counter returned with 2 (No such file or directory).
    /bin/dmesg may provide additional information.

      Fatal: Not all events could be opened.

dmesg is of no help and it is not clear as to why it fails to
open the counter. This patch changes the error message to

      Error: cache-misses event is not supported.
      Fatal: Not all events could be opened.

Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: a.p.zijlstra@chello.nl
LPU-Reference: <1294597272-17335-1-git-send-email-daahern@cisco.com>
Signed-off-by: David Ahern <daahern@cisco.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Bug introduced in ce47dc56.
Reported-by: Mike Galbraith <efault@gmx.de>
Cc: Chris Samuel <chris@csamuel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

Merge branch 'tip/perf/core' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace into perf/urgent

Don found that P4 PMU reads CCCR register instead of counter
itself (in attempt to catch unflagged event) this makes P4
NMI handler to consume all NMIs it observes. So the other
NMI users such as kgdb simply have no chance to get NMI
on their hands.

Side note: at moment there is no way to run nmi-watchdog
together with perf tool. This is because both 'perf top' and
nmi-watchdog use same event. So while nmi-watchdog reserves
one event/counter for own needs there is no room for perf tool
left (there is a way to disable nmi-watchdog on boot of course).

Ming has tested this patch with the following results

 | 1. watchdog disabled
 |
 | kgdb tests on boot OK
 | perf works OK
 |
 | 2. watchdog enabled, without patch perf-x86-p4-nmi-4
 |
 | kgdb tests on boot hang
 |
 | 3. watchdog enabled, without patch perf-x86-p4-nmi-4 and do not run kgdb
 | tests on boot
 |
 | "perf top" partialy works
 |   cpu-cycles            no
 |   instructions          yes
 |   cache-references      no
 |   cache-misses          no
 |   branch-instructions   no
 |   branch-misses         yes
 |   bus-cycles            no
 |
 | 4. watchdog enabled, with patch perf-x86-p4-nmi-4 applied
 |
 | kgdb tests on boot OK
 | perf does not work, NMI "Dazed and confused" messages show up
 |

Which means we still have problems with p4 box due to 'unknown'
nmi happens but at least it should fix kgdb test cases.
Reported-by: Jason Wessel <jason.wessel@windriver.com>
Reported-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Acked-by: Don Zickus <dzickus@redhat.com>
Acked-by: Lin Ming <ming.m.lin@intel.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <4D275E7E.3040903@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

On older gcc (3.3) dynamic debug fails to compile:

include/net/inet_connection_sock.h: In function `inet_csk_reset_xmit_timer':
include/net/inet_connection_sock.h:236: error: duplicate label declaration `do_printk'
include/net/inet_connection_sock.h:219: error: this is a previous declaration
include/net/inet_connection_sock.h:236: error: duplicate label declaration `out'
include/net/inet_connection_sock.h:219: error: this is a previous declaration
include/net/inet_connection_sock.h:236: error: duplicate label `do_printk'
include/net/inet_connection_sock.h:236: error: duplicate label `out'

Fix, by reverting the usage of JUMP_LABEL() in dynamic debug for now.

Cc: <stable@kernel.org>
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Tested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Jason Baron <jbaron@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

DEFINE_TRACE should also exist when CONFIG_EVENT_TRACING=n. Otherwise, setting
only TRACEPOINTS=y is broken.
Acked-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
LKML-Reference: <20101028153117.GA4051@Krystal>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

While running my ftrace stress test, this showed up:

BUG: sleeping function called from invalid context at mm/mmap.c:233
...
note: cat[3293] exited with preempt_count 1

The bug was introduced by commit 91e86e56
("tracing: Fix recursive user stack trace")

Cc: <stable@kernel.org>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <4D0089AC.1020802@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

Add __rcu annotation to :
	(struct tracepoint)->funcs
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
LKML-Reference: <4D22D4F1.50505@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

The check for NULL skb in the kfree_skb trace event is a duplicate from the
check already done in its only caller, kfree_skb(). Remove this duplicate check.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
LKML-Reference: <20110106175319.GA30610@Krystal>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: David S. Miller <davem@davemloft.net>
CC: Steven Rostedt <rostedt@goodmis.org>
CC: Frederic Weisbecker <fweisbec@gmail.com>
CC: Ingo Molnar <mingo@elte.hu>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Zhaolei <zhaolei@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

Add missing comma within the TRACE_EVENT() example in tracepoint.h.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
LKML-Reference: <20110106184532.GA2526@Krystal>
CC: Frederic Weisbecker <fweisbec@gmail.com>
CC: Ingo Molnar <mingo@elte.hu>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

While doing some developing, Peter Zijlstra and I have found
that if a CREATE_TRACE_POINTS include is done before module.h
is included, it can break the build.

We have been lucky so far that this has not broke the build
since module.h is included in almost everything.
Reported-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

From the x86_64 low level interrupt handlers, the frame pointer is
saved right after the partial pt_regs frame.

rbp is not supposed to be part of the irq partial saved registers,
but it only requires to extend the pt_regs frame by 8 bytes to
do so, plus a tiny stack offset fixup on irq exit.

This changes a bit the semantics or get_irq_entry() that is supposed
to provide only the value of caller saved registers and the cpu
saved frame. However it's a win for unwinders that can walk through
stack frames on top of get_irq_regs() snapshots.

A noticeable impact is that it makes perf events cpu-clock and
task-clock events based callchains working on x86_64.

Let's then save rbp into the irq pt_regs.

As a result with:

	perf record -e cpu-clock perf bench sched messaging
	perf report --stdio

Before:
    20.94%             perf  [kernel.kallsyms]        [k] lock_acquire
                       |
                       --- lock_acquire
                          |
                          |--44.01%-- __write_nocancel
                          |
                          |--43.18%-- __read
                          |
                          |--6.08%-- fork
                          |          create_worker
                          |
                          |--0.88%-- _dl_fixup
                          |
                          |--0.65%-- do_lookup_x
                          |
                          |--0.53%-- __GI___libc_read
                           --4.67%-- [...]

After:
    19.23%         perf  [kernel.kallsyms]    [k] __lock_acquire
                   |
                   --- __lock_acquire
                      |
                      |--97.74%-- lock_acquire
                      |          |
                      |          |--21.82%-- _raw_spin_lock
                      |          |          |
                      |          |          |--37.26%-- unix_stream_recvmsg
                      |          |          |          sock_aio_read
                      |          |          |          do_sync_read
                      |          |          |          vfs_read
                      |          |          |          sys_read
                      |          |          |          system_call
                      |          |          |          __read
                      |          |          |
                      |          |          |--24.09%-- unix_stream_sendmsg
                      |          |          |          sock_aio_write
                      |          |          |          do_sync_write
                      |          |          |          vfs_write
                      |          |          |          sys_write
                      |          |          |          system_call
                      |          |          |          __write_nocancel

v2: Fix cfi annotations.
Reported-by: Soeren Sandmann Pedersen <sandmann@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Jan Beulich <JBeulich@novell.com>

In dump_stack function, bp isn't used anymore, which is introduced by
commit 9c0729dc. This patch removes bp
completely.
Signed-off-by: Rakib Mullick <rakib.mullick@gmail.com>
Cc: Soeren Sandmann <sandmann@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <AANLkTik9U_Z0WSZ7YjrykER_pBUfPDdgUUmtYx=R74nL@mail.gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>

Just re-arrange the code a bit to make it easier to follow what is
going on.  Basically un-negating the if-statement and swapping the code
inside the if-statement with code outside.

No functional changes.
Originally-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294348732-15030-7-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

In original NMI handler, NMI reason io port (0x61) is only processed
on BSP.  This makes it impossible to hot-remove BSP.  To solve the
issue, a raw spinlock is used to allow the port to be processed on any
CPU.
Originally-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294348732-15030-6-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

With priorities in place and no one really understanding the difference between
DIE_NMI and DIE_NMI_IPI, just remove DIE_NMI_IPI and convert everyone to DIE_NMI.

This also simplifies default_do_nmi() a little bit.  Instead of calling the
die_notifier in both the if and else part, just pull it out and call it before
the if-statement.  This has the side benefit of avoiding a call to the ioport
to see if there is an external NMI sitting around until after the (more frequent)
internal NMIs are dealt with.
Patch-Inspired-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294348732-15030-5-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

In order to consolidate the NMI die_chain events, we need to setup the priorities
for the die notifiers.

I started by defining a bunch of common priorities that can be used by the
notifier blocks.  Then I modified the notifier blocks to use the newly created
priorities.

Now that the priorities are straightened out, it should be easier to remove the
event DIE_NMI_IPI.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294348732-15030-4-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

They are a handful of places in the code that register a die_notifier
as a catch all in case no claims the NMI.  Unfortunately, they trigger
on events like DIE_NMI and DIE_NMI_IPI, which depending on when they
registered may collide with other handlers that have the ability to
determine if the NMI is theirs or not.

The function unknown_nmi_error() makes one last effort to walk the
die_chain when no one else has claimed the NMI before spitting out
messages that the NMI is unknown.

This is a better spot for these devices to execute any code without
colliding with the other handlers.

The two drivers modified are only compiled on x86 arches I believe, so
they shouldn't be affected by other arches that may not have
DIE_NMIUNKNOWN defined.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Russ Anderson <rja@sgi.com>
Cc: Corey Minyard <minyard@acm.org>
Cc: openipmi-developer@lists.sourceforge.net
Cc: dann frazier <dannf@hp.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294348732-15030-3-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Replace the NMI related magic numbers with symbol constants.

Memory parity error is only valid for IBM PC-AT, newer machine use
bit 7 (0x80) of 0x61 port for PCI SERR. While memory error is usually
reported via MCE. So corresponding function name and kernel log string
is changed.

But on some machines, PCI SERR line is still used to report memory
errors. This is used by EDAC, so corresponding EDAC call is reserved.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294348732-15030-2-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Adds perf_event_time() to try and centralize access to event
timing and in particular ctx->time. Prepares for cgroup support.
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <4d22059c.122ae30a.5e0e.ffff8b8b@mx.google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Replace all occurrences of:
	event->cpu != -1 && event->cpu == smp_processor_id()
by a call to:
	event_filter_match(event)

This makes the code more consistent and will make the cgroup
patch smaller.
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <4d220593.2308e30a.48c5.ffff8ae9@mx.google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

In particular this patch move perf_event_exit_task() before
cgroup_exit() to allow for cgroup support. The cgroup_exit()
function detaches the cgroups attached to a task.

Other movements include hoisting some definitions and inlines
at the top of perf_event.c
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <4d22058b.cdace30a.4657.ffff95b1@mx.google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Since commit 69aad6f1(perf tools: Introduce event selectors), only
perf_event_attr::type and ::config are passed to event selector, which
makes perf tool not work correctly.

For example, PEBS does not work because perf_event_attr::precise_ip is
not passed to the syscall.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1294369869.20563.19.camel@minggr.sh.intel.com>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

It seems that some gcc versions build by default with frame pointers
and some others omit them.

Just build the tools with frame pointers as the callchains can be an
important part of the perf workflow.

Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <1294325513-14276-3-git-send-email-fweisbec@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>