[PATCH v7 0/9] Add metrics for neoverse-n2-v2

[Ian Rogers]

On Fri, Jan 13, 2023 at 8:32 AM Jing Zhang <renyu.zj at linux.alibaba.com> wrote:
>
>
>
> 在 2023/1/13 下午5:59, John Garry 写道:
> > On 13/01/2023 09:22, Jing Zhang wrote:
> >> Changes since v6:
> >> - Split patch #1 into 3 smaller patches as suggested by Ian.
> >> - Change perf_pmu__get_slots into perf_pmu__cpu_slots_per_cycle,
> >>    per John's suggestion;
> >> - Return NAN instead of 0 in perf_pmu__cpu_slots_per_cycle weak
> >>    function, per John's suggestion;
> >> - Factor out pmu_core__find_same function, per John's suggestion.
> >> - Link:https://urldefense.com/v3/__https://lore.kernel.org/all/1673017529-1429208-1-git-send-email-renyu.zj@linux.alibaba.com/__;!!ACWV5N9M2RV99hQ!LhBq67uDCOsz1k7ZF4aQPHF0Bp8FsMr-ZNgCnBSUKF4qJTFODfnkId7lw_NXqB4qZUCpu-jbY8z8LTckoqFGz2Q8bA$
> >
> > This looks fine. But for this code:
> >
> > On 13/01/2023 09:22, Jing Zhang wrote:
> >> +double perf_pmu__cpu_slots_per_cycle(void)
> >> +{
> >> +    char path[PATH_MAX];
> >> +    unsigned long long slots = 0;
> >
> > I would prefer if this returned NAN (and not 0) for when we can't find a pmu or the value from ./caps/slots is zero, but I am not going to get too hung up on that.
> >
>
> Ok, I like this way too.
>
> > For series:
> >
> > Reviewed-by: John Garry <john.g.garry at oracle.com>
>
> Thank you very much indeed!

Aside a naming nit in 1/9 for series:

Acked-by: Ian Rogers <irogers at google.com>

Thanks,
Ian