Re: [PATCH 1/1] perf lock info: Display both map and thread by default

[Nick Forrington]

Hello,

On 08/05/2024 22:18, Namhyung Kim wrote:
> Hello,
>
> On Wed, May 8, 2024 at 6:46 AM Nick Forrington <nick.forrington@xxxxxxx> wrote:
> > Change "perf lock info" argument handling to:
> >
> > Display both map and thread info (rather than an error) when neither are
> > specified.
> >
> > Display both map and thread info (rather than just thread info) when
> > both are requested.
> >
> > Signed-off-by: Nick Forrington <nick.forrington@xxxxxxx>
> > ---
> >   tools/perf/Documentation/perf-lock.txt |  4 ++--
> >   tools/perf/builtin-lock.c              | 22 +++++++++++++++++-----
> >   2 files changed, 19 insertions(+), 7 deletions(-)
> >
> > diff --git a/tools/perf/Documentation/perf-lock.txt b/tools/perf/Documentation/perf-lock.txt
> > index f5938d616d75..57a940399de0 100644
> > --- a/tools/perf/Documentation/perf-lock.txt
> > +++ b/tools/perf/Documentation/perf-lock.txt
> > @@ -111,11 +111,11 @@ INFO OPTIONS
> >
> >   -t::
> >   --threads::
> > -       dump thread list in perf.data
> > +       dump only the thread list in perf.data
> >
> >   -m::
> >   --map::
> > -       dump map of lock instances (address:name table)
> > +       dump only the map of lock instances (address:name table)
> >
> >
> >   CONTENTION OPTIONS
> > diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c
> > index 230461280e45..d0b8645487ad 100644
> > --- a/tools/perf/builtin-lock.c
> > +++ b/tools/perf/builtin-lock.c
> > @@ -1483,11 +1483,16 @@ static int dump_info(void)
> >
> >          if (info_threads)
> >                  dump_threads();
> > -       else if (info_map)
> > +
> > +       if (info_map) {
> > +               if (info_threads)
> > +                       fputc('\n', lock_output);
> it seems you need pr_info("\n").  Where does lock_output come from?


It's a static FILE* that defaults to stderr.


It's used by dump_threads and dump_map (via fprintf), so I think it 
makes sense to use it for the output spacing between them?

> >                  dump_map();
> > -       else {
> > +       }
> > +
> > +       if (!info_threads && !info_map) {
> >                  rc = -1;
> > -               pr_err("Unknown type of information\n");
> > +               pr_err("No lock info specified\n");
> >          }
> I think we can remove this block now.


The original error looked like a function input sanity check (rather 
than for handling user input), so I kept it (even if it should no longer 
be possible for a user to trigger it).


Although I can remove it (or move it to the start of the function) if 
you prefer.


Thanks,
Nick

> Thanks,
> Namhyung
>
>
> >          return rc;
> > @@ -2578,9 +2583,9 @@ int cmd_lock(int argc, const char **argv)
> >
> >          const struct option info_options[] = {
> >          OPT_BOOLEAN('t', "threads", &info_threads,
> > -                   "dump thread list in perf.data"),
> > +                   "dump the thread list in perf.data"),
> >          OPT_BOOLEAN('m', "map", &info_map,
> > -                   "map of lock instances (address:name table)"),
> > +                   "dump the map of lock instances (address:name table)"),
> >          OPT_PARENT(lock_options)
> >          };
> >
> > @@ -2694,6 +2699,13 @@ int cmd_lock(int argc, const char **argv)
> >                          if (argc)
> >                                  usage_with_options(info_usage, info_options);
> >                  }
> > +
> > +               /* If neither threads nor map requested, display both */
> > +               if (!info_threads && !info_map) {
> > +                       info_threads = true;
> > +                       info_map = true;
> > +               }
> > +
> >                  /* recycling report_lock_ops */
> >                  trace_handler = &report_lock_ops;
> >                  rc = __cmd_report(true);
> > --
> > 2.44.0