Re: [PATCH/RFC 02/16] perf top: Fix and cleanup perf_top__record_precise_ip()

From: Namhyung Kim
Date: Thu Dec 10 2015 - 21:28:14 EST


Hi Arnaldo,

On Thu, Dec 10, 2015 at 04:04:17PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Thu, Dec 10, 2015 at 04:53:21PM +0900, Namhyung Kim escreveu:
> > At first, it has duplicate ui__has_annotation() and 'sort__has_sym'
> > and 'use_browser' check. In fact, the ui__has_annotation() should be
> > removed as it needs to annotate on --stdio as well. And the
> > top->sym_filter_entry is used only for stdio mode, so make it clear on
> > the condition too.
>
> So this is doing more than one thing and changes decisions about
> non-trivial operations, can you please break it down into smaller
> patches?

Sure.

>
> > Also the check will be simplifed if it sets sym before the check. And
> > omit check for 'he' since its caller (hist_iter__top_callback) only
> > gets called when iter->he is not NULL.
> >
> > Secondly, it converts the 'ip' argument using map__unmap_ip() before
> > the call and then reverts it using map__map_ip(). This seems
> > meaningless and strange since only one of them checks the 'map'.
>
> For isntance: the following change has value and should be on a separate
> patch.

Will do!

Thanks,
Namhyung


>
> > Finally, access the he->hists->lock only if there's an error.
> >
> > Signed-off-by: Namhyung Kim <namhyung@xxxxxxxxxx>
> > ---
--
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/