Re: [PATCH v3 3/3] perf script/python: Pass event/thread/dso nameand symbol info to event handler in python

[Feng Tang]

Hi Arnaldo,

On Wed, 27 Jun 2012 15:58:17 -0300
Arnaldo Carvalho de Melo <acme@xxxxxxxxxx> wrote:

> Em Mon, Jun 18, 2012 at 02:10:39PM +0800, Feng Tang escreveu:
> > Also as suggested by Arnaldo, pack all these parameters to a dictionary,
> > which is more expandable for adding new parameters while keep the
> > compatibility for old scripts.
> > 
> > Signed-off-by: Feng Tang <feng.tang@xxxxxxxxx>
> > ---
> >  .../util/scripting-engines/trace-event-python.c    |   32
> > +++++++++++++++---- 1 files changed, 25 insertions(+), 7 deletions(-)
> > 
> > diff --git a/tools/perf/util/scripting-engines/trace-event-python.c
> > b/tools/perf/util/scripting-engines/trace-event-python.c index
> > 97f4fad..e8ac480 100644 ---
> > a/tools/perf/util/scripting-engines/trace-event-python.c +++
> > b/tools/perf/util/scripting-engines/trace-event-python.c @@ -336,15 +336,23
> > @@ static void python_process_general_event(union perf_event *pevent
> > __unused, struct machine *machine __unused, struct addr_location *al)
> >  {
> > -	PyObject *handler, *retval, *t;
> > +	PyObject *handler, *retval, *t, *dict;
> >  	static char handler_name[64];
> >  	unsigned n = 0;
> > -	void *data = sample->raw_data;
> > +	struct thread *thread = al->thread;
> >  
> > +	/*
> > +	 * Use the MAX_FIELDS to make the function expandable, though
> > +	 * currently there is only one itme for the tuple.
> > +	 */
> >  	t = PyTuple_New(MAX_FIELDS);
> >  	if (!t)
> >  		Py_FatalError("couldn't create Python tuple");
> >  
> > +	dict = PyDict_New();
> > +	if (!dict)
> > +		Py_FatalError("couldn't create Python dictionary");
> > +
> >  	snprintf(handler_name, sizeof(handler_name), "%s",
> > "process_event"); 
> >  	handler = PyDict_GetItemString(main_dict, handler_name);
> > @@ -353,14 +361,23 @@ static void python_process_general_event(union
> > perf_event *pevent __unused, goto exit;
> >  	}
> >  
> > -	/* Pass 4 parameters: event_attr, perf_sample, raw data, thread
> > name */
> > -	PyTuple_SetItem(t, n++, PyString_FromStringAndSize(
> > +	PyDict_SetItemString(dict, "ev_name",
> > PyString_FromString(event_name(evsel)));
> > +	PyDict_SetItemString(dict, "attr", PyString_FromStringAndSize(
> >  			(const char *)&evsel->attr, sizeof(evsel->attr)));
> > -	PyTuple_SetItem(t, n++, PyString_FromStringAndSize(
> > +	PyDict_SetItemString(dict, "sample", PyString_FromStringAndSize(
> >  			(const char *)sample, sizeof(*sample)));
> > -	PyTuple_SetItem(t, n++, PyString_FromStringAndSize(
> > -			data, sample->raw_size));
> > +	PyDict_SetItemString(dict, "raw_buf", PyString_FromStringAndSize(
> > +			(const char *)sample->raw_data, sample->raw_size));
> > +	PyDict_SetItemString(dict, "comm",
> > +			PyString_FromString(thread->comm));
> > +	PyDict_SetItemString(dict, "dso",
> > +			PyString_FromString(al->map->dso->name));
> > +	if (al->sym) {
> > +		PyDict_SetItemString(dict, "symbol",
> > +			PyString_FromString(al->sym->name));
> > +	}
> >  
> > +	PyTuple_SetItem(t, n++, dict);
> 
> Now old scrips will break, as the tuple they expect:
> 
>  (attr, sample, raw_data)
> 
> will not be there.

Actually, there are no such scripts out. Current perf python script only
supports the trace point type, while perl scriptt has the support for
general events. It is my first patch "perf script: Add general python 
handler to process non-tracepoint events" which bring this (attr, sample,
raw_data) in to be on the same page as perl. So this won't be a problem
when the 3 patches are merged together. 

> 
> The first two patches in this series don't apply anymore, please find
> them attached fixed, please check if they work as expected.
Thanks a lot for fixing them, I'll generate perf tool patch against the
"perf/core" branch of your git tree on kernel.org from now on.

Btw, I have 2 more python scripts based on these patches, will add them
to this serie for your review.

Thanks,
Feng

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/