Re: [PATCH 4/6] perf script: dump software events too

[Frederic Weisbecker]

2011/3/1 David Ahern <daahern@xxxxxxxxx>:
> I don't follow what you mean.
>
> This is the same place that tracepoints are dumped. In the
> PERF_TYPE_TRACEPOINT case process_event == print_event which dumps the
> lines to stdout.

No it's not. scripting_ops can be either default_scripting_ops,
perl_scripting_ops
or python_scripting_ops.

Hence process_sample_event() doesn't necessarily ends up pretty print events.
It can also invoke a script that may decide to do what it wants with events.

pretty printing must be an endpoint decision.

> I came in added a switch that invokes a different handler for software
> events. Software events will be processed differently than tracepoint.
> The print_event function and its pretty_print function are not designed
> to dump software events (and eventually hardware events).

Why shouldn't it be designed to dump software events? It's called print_event().
Its current version is rather something I would call "limited". But it
was not designed
to be limited.

Ideally, we should have print_tracepoint_event() in
trace-event-parse.c, print_software_event()
where you want, and have print_event() in builtin-script.c that wraps on those.
--
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/