RE: [PATCH V3 10/12] perf top: add overwrite fall back

From: Liang, Kan
Date: Thu Jan 11 2018 - 16:30:29 EST


> On Thu, Dec 21, 2017 at 10:08:52AM -0800, kan.liang@xxxxxxxxx wrote:
> > From: Kan Liang <kan.liang@xxxxxxxxx>
> >
> > Switch to non-overwrite mode if kernel doesnot support overwrite
> > ringbuffer.
> >
> > It's only effect when overwrite mode is supported.
> > No change to current behavior.
> >
> > Signed-off-by: Kan Liang <kan.liang@xxxxxxxxx>
> > ---
> > tools/perf/builtin-top.c | 36 ++++++++++++++++++++++++++++++++++++
> > 1 file changed, 36 insertions(+)
> >
> > diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
> > index 4b85e7b..8d19ef7 100644
> > --- a/tools/perf/builtin-top.c
> > +++ b/tools/perf/builtin-top.c
> > @@ -931,6 +931,27 @@ static int perf_top_overwrite_check(struct
> perf_top *top)
> > return 0;
> > }
> >
> > +static int perf_top_overwrite_fallback(struct perf_top *top,
> > + struct perf_evsel *evsel)
> > +{
> > + struct record_opts *opts = &top->record_opts;
> > + struct perf_evlist *evlist = top->evlist;
> > + struct perf_evsel *counter;
> > +
> > + if (!opts->overwrite)
> > + return 0;
> > +
> > + /* only fall back when first event fails */
> > + if (evsel != perf_evlist__first(evlist))
> > + return 0;
> > +
> > + evlist__for_each_entry(evlist, counter)
> > + counter->attr.write_backward = false;
> > + opts->overwrite = false;
> > + ui__warning("fall back to non-overwrite mode\n");
> > + return 1;
> > +}
>
> we already do that for evsel in perf_evsel__open.. could we make
> this fallback on one place only?
>
> checking the code, why don't we follow the
> perf_missing_features.write_backward
> like we do for other features in the perf_evsel__open and set
> write_backward accordingly, like in attached patch.
>

No, the per-event fallback is explicitly disabled in perf_evsel__open.
You may refer to commit 32a951b4fd6b ("perf evlist: Drop redundant
evsel->overwrite indicator)".

Now, only perf top supports the whole fallback. So it is specially handled.


Thanks,
Kan

> jirka
>
>
> ---
> diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
> index dbd80268de54..89253f4ff9ee 100644
> --- a/tools/perf/util/evsel.c
> +++ b/tools/perf/util/evsel.c
> @@ -1683,9 +1683,6 @@ int perf_evsel__open(struct perf_evsel *evsel,
> struct cpu_map *cpus,
> int pid = -1, err;
> enum { NO_CHANGE, SET_TO_MAX, INCREASED_MAX } set_rlimit =
> NO_CHANGE;
>
> - if (perf_missing_features.write_backward && evsel-
> >attr.write_backward)
> - return -EINVAL;
> -
> if (cpus == NULL) {
> static struct cpu_map *empty_cpu_map;
>
> @@ -1725,6 +1722,8 @@ int perf_evsel__open(struct perf_evsel *evsel,
> struct cpu_map *cpus,
> }
>
> fallback_missing_features:
> + if (perf_missing_features.write_backward)
> + evsel->attr.write_backward = false;
> if (perf_missing_features.clockid_wrong)
> evsel->attr.clockid = CLOCK_MONOTONIC; /* should always
> work */
> if (perf_missing_features.clockid) {