Re: [PATCH v3] perf pmu: Validate raw event with sysfs exported format bits

From: Jiri Olsa
Date: Mon Mar 08 2021 - 08:15:55 EST


On Mon, Mar 08, 2021 at 08:57:49PM +0800, Jin, Yao wrote:
> Hi Jiri,
>
> On 3/8/2021 6:35 PM, Jiri Olsa wrote:
> > On Mon, Mar 08, 2021 at 11:15:06AM +0800, Jin Yao wrote:
> >
> > SNIP
> >
> > > diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
> > > index 44ef28302fc7..03ab1e6d0418 100644
> > > --- a/tools/perf/util/pmu.c
> > > +++ b/tools/perf/util/pmu.c
> > > @@ -1812,3 +1812,39 @@ int perf_pmu__caps_parse(struct perf_pmu *pmu)
> > > return nr_caps;
> > > }
> > > +
> > > +void perf_pmu__warn_invalid_config(struct perf_pmu *pmu, __u64 config,
> > > + char *name)
> > > +{
> > > + struct perf_pmu_format *format;
> > > + __u64 masks = 0, bits;
> > > + char buf[100];
> > > + unsigned int i;
> > > +
> > > + list_for_each_entry(format, &pmu->format, list) {
> > > + /*
> > > + * Skip extra configs such as config1/config2.
> > > + */
> > > + if (format->value > 0)
> > > + continue;
> >
> > sorry I did not notice before, but could you please use more direct
> > approach like:
> >
> > if (format->value == PERF_PMU_FORMAT_VALUE_CONFIG) {
> > break;
> > }
> >
> > this will be more obvious, also no need for the comment.. I spent some
> > time looking what's the value for ;-)
> >
> > thanks,
> > jirka
> >
>
> Oh, yes, using PERF_PMU_FORMAT_VALUE_CONFIG is much more obvious. Sorry about that!
>
> While it can't break the loop, because we need to iterate over the whole
> list to get the total valid bits. So like:
>
> if (format->value != PERF_PMU_FORMAT_VALUE_CONFIG)
> continue;
>
> Is it right?

sure, what I meant was to process only PERF_PMU_FORMAT_VALUE_CONFIG
and then call break, because there's no need to iterate further

jirka