When event group is enabled for forked task (i.e. no target task/cpu
was specified) all events were disabled and marked ->enable_on_exec.
However they wouldn't be counted at all since only group leader will
be enabled on exec actually.
In contrast to perf stat, perf record doesn't have a real problem
as it enables all the event before proceeding. But it needs to be
fixed anyway IMHO.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1331887340-32448-2-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim <namhyung.kim@lge.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When event group is enabled for forked task (i.e. no target task was
specified) all events were disabled and marked ->enable_on_exec.
However they are not counted at all since only group leader will be
enabled on exec actually. So the result looked like below:
$ ./perf stat --group -- sleep 1
Performance counter stats for 'sleep 1':
0.554926 task-clock # 0.001 CPUs utilized
<not counted> context-switches
<not counted> CPU-migrations
<not counted> page-faults
<not counted> cycles
<not supported> stalled-cycles-frontend
<not supported> stalled-cycles-backend
<not counted> instructions
<not counted> branches
<not counted> branch-misses
1.001228093 seconds time elapsed
Fix it by disabling group leader only.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1331887340-32448-1-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim <namhyung.kim@lge.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Added new event rule to the event definition grammar:
event_def: event_pmu |
...
event_pmu: PE_NAME '/' event_config '/'
Using this rule, event could be now specified like:
cpu/config=1,config1=2,config2=3/u
where pmu name 'cpu' is looked up via following path:
${sysfs_mount}/bus/event_source/devices/${pmu}
and config options are bound to the pmu's format definiton:
${sysfs_mount}/bus/event_source/devices/${pmu}/format
The hardcoded config options still stays and have precedence
over any format field defined with same name.
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/n/tip-50d8nr94f8k4wkezutrxvthe@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adding pmu object which provides interface to pmu's sysfs
event format definition located at:
${sysfs_mount}/bus/event_source/devices/${pmu}/format
Following interface is exported:
struct perf_pmu* perf_pmu__find(char *name);
- this function returns pmu object, which is then
passed as a handle to other interface functions
int perf_pmu__config(struct perf_pmu *pmu, struct perf_event_attr *attr,
struct list_head *head_terms);
- this function configures perf_event_attr struct based
on pmu's format definitions and config terms data,
containined in head_terms list.
Parser generator is used to retrive the pmu's format definition.
The generated parser is part of the patch. Added makefile rule
'pmu-parser' to generate the parser code out of the bison/flex
sources.
Added builtin test 'Test perf pmu format parsing', which could
be run like:
perf test pmu
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/n/tip-errz96u1668gj9wlop1zhpht@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adding a new rule to the event grammar to be able to specify
values of additional attributes of symbolic event.
The new syntax for event symbolic definition is:
event_legacy_symbol: PE_NAME_SYM '/' event_config '/' |
PE_NAME_SYM sep_slash_dc
event_config: event_config ',' event_term | event_term
event_term: PE_NAME '=' PE_NAME |
PE_NAME '=' PE_VALUE
PE_NAME
sep_slash_dc: '/' | ':' |
At the moment the config options are hardcoded to be used for legacy
symbol events to define several perf_event_attr fields. It is:
'config' to define perf_event_attr::config
'config1' to define perf_event_attr::config1
'config2' to define perf_event_attr::config2
'period' to define perf_event_attr::sample_period
Legacy events could be now specified as:
cycles/period=100000/
If term is specified without the value assignment, then 1 is
assigned by default.
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/n/tip-mgkavww9790jbt2jdkooyv4q@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add rules to generate pre-processed files (just like are available for
the normal kernel build), and adjust the rule to create assembly files
from C ones to produce its output in the output directory rather than
in the source tree.
E.g.
$ make -C tools/perf/ O=/tmp/perf /tmp/perf/builtin-top.i
make: Entering directory `/home/git/linux/tools/perf'
CC /tmp/perf/builtin-top.i
make: Leaving directory `/home/git/linux/tools/perf'
$ wc -l /tmp/perf/builtin-top.i
31379 /tmp/perf/builtin-top.i
$
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/4F588A0802000078000770FF@nat28.tlf.novell.com
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This patch adds support for taken branch sampling, i.e, the
PERF_SAMPLE_BRANCH_STACK feature to perf report. In other
words, to display histograms based on taken branches rather
than executed instructions addresses.
The new option is called -b and it takes no argument. To
generate meaningful output, the perf.data must have been
obtained using perf record -b xxx ... where xxx is a branch
filter option.
The output shows symbols, modules, sorted by 'who branches
where' the most often. The percentages reported in the first
column refer to the total number of branches captured and
not the usual number of samples.
Here is a quick example.
Here branchy is simple test program which looks as follows:
void f2(void)
{}
void f3(void)
{}
void f1(unsigned long n)
{
if (n & 1UL)
f2();
else
f3();
}
int main(void)
{
unsigned long i;
for (i=0; i < N; i++)
f1(i);
return 0;
}
Here is the output captured on Nehalem, if we are
only interested in user level function calls.
$ perf record -b any_call,u -e cycles:u branchy
$ perf report -b --sort=symbol
52.34% [.] main [.] f1
24.04% [.] f1 [.] f3
23.60% [.] f1 [.] f2
0.01% [k] _IO_new_file_xsputn [k] _IO_file_overflow
0.01% [k] _IO_vfprintf_internal [k] _IO_new_file_xsputn
0.01% [k] _IO_vfprintf_internal [k] strchrnul
0.01% [k] __printf [k] _IO_vfprintf_internal
0.01% [k] main [k] __printf
About half (52%) of the call branches captured are from main()
-> f1(). The second half (24%+23%) is split in two equal shares
between f1() -> f2(), f1() ->f3(). The output is as expected
given the code.
It should be noted, that using -b in perf record does not
eliminate information in the perf.data file. Consequently, a
typical profile can also be obtained by perf report by simply
not using its -b option.
It is possible to sort on branch related columns:
- dso_from, symbol_from
- dso_to, symbol_to
- mispredict
Signed-off-by: Roberto Agostino Vitillo <ravitillo@lbl.gov>
Signed-off-by: Stephane Eranian <eranian@google.com>
Cc: peterz@infradead.org
Cc: acme@redhat.com
Cc: robert.richter@amd.com
Cc: ming.m.lin@intel.com
Cc: andi@firstfloor.org
Cc: asharma@fb.com
Cc: vweaver1@eecs.utk.edu
Cc: khandual@linux.vnet.ibm.com
Cc: dsahern@gmail.com
Link: http://lkml.kernel.org/r/1328826068-11713-14-git-send-email-eranian@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Pull perf fixes from Ingo Molnar:
"It contains three cherry-picked fixes from perf/core, which turned out
to be more urgent than we originally thought."
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf tools: Handle kernels that don't support attr.exclude_{guest,host}
perf tools: Change perf_guest default back to false
perf record: No build id option fails
If perf_evsel__open() failed, the errno was set and returned properly.
However since the perf_evlist__open() called close() on fd's for all of
evsel x cpu x thread after the failure, the errno was overridden by
other code (EBADF). So the caller of the function ended up seeing
different error message and getting confused.
Fit it by restoring original return value. Because one of caller of the
function is in the python extension, and it uses system errno
internally, it'd be better restoring the original value rather than
using the return value of the function directly, IMHO (i.e. I'm not a
python expert :)
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1329966816-23175-1-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim <namhyung.kim@lge.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pulling latest branches from Ingo:
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
memblock: Fix size aligning of memblock_alloc_base_nid()
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf probe: Ensure offset provided is not greater than function length without DWARF info too
perf tools: Ensure comm string is properly terminated
perf probe: Ensure offset provided is not greater than function length
perf evlist: Return first evsel for non-sample event on old kernel
perf/hwbp: Fix a possible memory leak
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
CPU hotplug, cpusets, suspend: Don't touch cpusets during suspend/resume
If threads in a multi-threaded process have names shorter than the main
thread the comm for the named threads is not properly terminated.
E.g., for the process 'namedthreads' where each thread is named noploop%d
where %d is the thread number:
Before:
perf script -f comm,tid,ip,sym,dso
noploop:4ads 21616 400a49 noploop (/tmp/namedthreads)
The 'ads' in the thread comm bleeds over from the process name.
After:
perf script -f comm,tid,ip,sym,dso
noploop:4 21616 400a49 noploop (/tmp/namedthreads)
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1330111898-68071-1-git-send-email-dsahern@gmail.com
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
While demoing ktest at ELC in 2012, it was embarrassing that the
make_min_config test failed to work because the snowball board I was
testing it against had a config that would not build. But the
make_min_config only tested the testing part and ignored build failures.
The end result was a config file that would not boot.
This time, for real.
* tag 'ktest-fix-make-min-failed-build-for-real' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest:
ktest: Fix make_min_config test when build fails
The make_min_config does not take into account when the build fails,
resulting in a invalid MIN_CONFIG .config file. When the build fails,
it is ignored and the boot test is executed, using the previous built
kernel. The configs that should be tested are not tested and they may
be added or removed depending on the result of the last kernel that
succeeded to be built.
If the build fails, mark the current config as a failure and the
configs that were disabled may still be needed.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The following commit:
b52956c perf tools: Allow multiple threads or processes in record, stat, top
introduced a bug in the thread_map code which caused perf record -a to
not setup system-wide monitoring properly.
$ taskset -c 1 noploop 1000 &
$ perf record -a -C 1 sleep 10
$ perf report -D | tail -20
cycles stats:
TOTAL events: 4413
MMAP events: 4025
COMM events: 340
SAMPLE events: 48
Here I was expecting about 10,000 samples and not 48.
In system-wide mode, the PID passed to perf_event_open() must be -1 and
it was 0. That caused the kernel to setup a per-process event on PID:0.
Consequently, the number of samples captured does not correspond to the
requested measurement.
The following one-liner fixes the problem for me with or without -C.
I would also suggest to change the malloc() to something that matches
the struct definition. thread_map->map[] is declared as int map[] and
not pid_t map[]. If map[] can only contain pids, then change the struct
definition.
Acked-by: David Ahern <dsahern@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120221145424.GA6757@quad
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The __print_symbolic() function takes a sequence of key-value pairs for
pretty-printing a constant. The new kvm:kvm_exit print fmt uses the
expression:
__print_symbolic(..., { 0x040 + 1, "DB excp" }, ...)
Currently only atoms are supported and this print fmt fails to parse.
This patch adds support for expressions instead of just atoms so that
0x040 + 1 is parsed successfully. Also add arg_num_eval() support for
the '+' operator.
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1315148939-14313-1-git-send-email-stefanha@linux.vnet.ibm.com
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
