Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220926140604.4173723-1-daniel.lezcano@linaro.org>
 <20220926140604.4173723-6-daniel.lezcano@linaro.org> <CAJZ5v0ibt6nj6+E3onu4Ri74gX84VBob-hOWQPZiusGKb_Apjw@mail.gmail.com>
 <657008be-34e3-e2de-a9bd-41d2dc804e51@linaro.org>
In-Reply-To: <657008be-34e3-e2de-a9bd-41d2dc804e51@linaro.org>
From:   "Rafael J. Wysocki" <rafael@kernel.org>
Date:   Tue, 27 Sep 2022 13:08:00 +0200
Message-ID: <CAJZ5v0hiofXuEHqX1sMXdyTqaDu85viL69neJo=QvDa47aHaFg@mail.gmail.com>
Subject: Re: [PATCH v5 05/30] thermal/core/governors: Use thermal_zone_get_trip()
 instead of ops functions
To:     Daniel Lezcano <daniel.lezcano@linaro.org>
Cc:     "Rafael J. Wysocki" <rafael@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux PM <linux-pm@vger.kernel.org>,
        "Zhang, Rui" <rui.zhang@intel.com>,
        Lukasz Luba <lukasz.luba@arm.com>,
        Amit Kucheria <amitk@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Sep 27, 2022 at 12:15 AM Daniel Lezcano
<daniel.lezcano@linaro.org> wrote:
>
> On 26/09/2022 21:34, Rafael J. Wysocki wrote:
> > On Mon, Sep 26, 2022 at 4:06 PM Daniel Lezcano
> > <daniel.lezcano@linaro.org> wrote:
> >>
> >> The governors are using the ops->get_trip_* functions, Replace these
> >> calls with thermal_zone_get_trip().
> >>
> >> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> >> Reviewed-by: Zhang Rui <rui.zhang@intel.com>
> >> Reviewed-by: Lukasz Luba <lukasz.luba@arm.com> # IPA
> >> ---
> >>   drivers/thermal/gov_bang_bang.c       | 29 ++++++++-------
> >>   drivers/thermal/gov_fair_share.c      | 18 ++++------
> >>   drivers/thermal/gov_power_allocator.c | 51 ++++++++++++---------------
> >>   drivers/thermal/gov_step_wise.c       | 22 ++++++------
> >>   4 files changed, 53 insertions(+), 67 deletions(-)
> >>
> >> diff --git a/drivers/thermal/gov_bang_bang.c b/drivers/thermal/gov_bang_bang.c
> >> index a08bbe33be96..f5b85e5ea707 100644
> >> --- a/drivers/thermal/gov_bang_bang.c
> >> +++ b/drivers/thermal/gov_bang_bang.c
> >> @@ -13,26 +13,25 @@
> >>
> >>   #include "thermal_core.h"
> >>
> >> -static void thermal_zone_trip_update(struct thermal_zone_device *tz, int trip)
> >> +static void thermal_zone_trip_update(struct thermal_zone_device *tz, int trip_id)
> >>   {
> >> -       int trip_temp, trip_hyst;
> >> +       struct thermal_trip trip;
> >>          struct thermal_instance *instance;
> >> +       int ret;
> >>
> >> -       tz->ops->get_trip_temp(tz, trip, &trip_temp);
> >> -
> >> -       if (!tz->ops->get_trip_hyst) {
> >> -               pr_warn_once("Undefined get_trip_hyst for thermal zone %s - "
> >> -                               "running with default hysteresis zero\n", tz->type);
> >> -               trip_hyst = 0;
> >> -       } else
> >> -               tz->ops->get_trip_hyst(tz, trip, &trip_hyst);
> >> +       ret = __thermal_zone_get_trip(tz, trip_id, &trip);
> >> +       if (ret)
> >> +               pr_warn_once("Failed to retrieve trip point %d\n", trip_id);
> >
> > Does it even make sense to continue beyond this point if ret is nonzero?
>
> No, I think we can bail out from here
>
> > All of the contents of trip can be garbage then AFAICS.
> >
> >> +
> >> +       if (!trip.hysteresis)
> >> +               pr_warn_once("Zero hysteresis value for thermal zone %s\n", tz->type);
> >
> > Why do you want to warn about this?  Haven't we declared already that
> > zero hysteresis is valid and normal?
>
> Apparently the bang-bang governor is expecting a hysteresis value as the
> check is expecting:
>
>  >> -       if (!tz->ops->get_trip_hyst) {
>  >> -               pr_warn_once("Undefined get_trip_hyst for thermal
> zone %s - "
>  >> -                               "running with default hysteresis
> zero\n", tz->type);
>  >> -               trip_hyst = 0;
>
> It is just to keep a warning as before.

The new message will be different, though.

I think it should be per-tz and the "info" level should be sufficient,
and because thermal_zone_device is based on struct device,
dev_info_once(&tz->dev, ...) should work.