MIME-Version: 1.0
References: <20230314120252.48263-1-mmkurbanov@sberdevices.ru>
 <20230314120252.48263-3-mmkurbanov@sberdevices.ru> <3b920b9e-07dc-7bda-4fe1-d15d07e708cc@sberdevices.ru>
 <CAHp75VcWuOEWZn2E8dG=Pb_KuEv06jYt_+nZSL-ceAQRPmgeGw@mail.gmail.com>
In-Reply-To: <CAHp75VcWuOEWZn2E8dG=Pb_KuEv06jYt_+nZSL-ceAQRPmgeGw@mail.gmail.com>
From:   Andy Shevchenko <andy.shevchenko@gmail.com>
Date:   Tue, 14 Mar 2023 18:49:58 +0200
Message-ID: <CAHp75Vd5goCgubOyaR2FFPi05eCfguh7XqP7MnHd6qP_o7wEmw@mail.gmail.com>
Subject: Re: [PATCH v3 2/2] leds: add aw20xx driver
To:     Martin Kurbanov <mmkurbanov@sberdevices.ru>
Cc:     Pavel Machek <pavel@ucw.cz>, Lee Jones <lee@kernel.org>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        linux-leds@vger.kernel.org, linux-kernel@vger.kernel.org,
        kernel@sberdevices.ru, devicetree@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Mar 14, 2023 at 6:22=E2=80=AFPM Andy Shevchenko
<andy.shevchenko@gmail.com> wrote:
> On Tue, Mar 14, 2023 at 2:12=E2=80=AFPM Martin Kurbanov
> <mmkurbanov@sberdevices.ru> wrote:
> > On 2023-03-01 00:51, Andy Shevchenko wrote:

...

> > >> +static int aw200xx_set_imax(const struct aw200xx *const chip,
> > >> +                           u32 led_imax_microamp)
> > >> +{
> > >> +       struct imax_global {
> > >> +               u32 regval;
> > >> +               u32 microamp;
> > >> +       } imaxs[] =3D {
> > >> +               { 8,  3300 },
> > >> +               { 9,  6700 },
> > >> +               { 0,  10000 },
> > >> +               { 11, 13300 },
> > >> +               { 1,  20000 },
> > >> +               { 13, 26700 },
> > >> +               { 2,  30000 },
> > >> +               { 3,  40000 },
> > >> +               { 15, 53300 },
> > >> +               { 4,  60000 },
> > >> +               { 5,  80000 },
> > >> +               { 6,  120000 },
> > >> +               { 7,  160000 },
> > >
> > > This looks a bit random. Is there any pattern on how value is
> > > connected to the register value?
> >
> > There is no ability to create any pattern here, because this table data
> > doesn=E2=80=99t have any regularity.
>
> There is a clear pattern.
>
> You have two tables, i.e. with multiplier 10000 and second one with
> multiplier 3333 (table in the datasheet seems bad from a math
> perspective). And it's even shown correctly in the datasheet.
>
> With this mix you missed 10.
>
> The coefficient table is 1,2,3,4,6,8,12,16 for both tables.
>
> Hence you need one table and two multipliers.
>
> Please, rewrite accordingly.

JFYI: You may see how I killed a table in one driver due to missing
understanding that there is a pattern.
9df461eca18f ("spi: pxa2xx: replace ugly table by approximation")

--=20
With Best Regards,
Andy Shevchenko