Date:   Thu, 26 Jan 2023 13:48:32 +0100
From:   Wolfram Sang <wsa+renesas@sang-engineering.com>
To:     Geert Uytterhoeven <geert@linux-m68k.org>
Cc:     linux-renesas-soc@vger.kernel.org,
        Prabhakar <prabhakar.csengg@gmail.com>,
        Sergey Shtylyov <s.shtylyov@omp.ru>,
        Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] memory: renesas-rpc-if: Fix PHYCNT.STRTIM setting
Message-ID: <Y9J2oPHHNQoU+7m7@ninjato>
Mail-Followup-To: Wolfram Sang <wsa+renesas@sang-engineering.com>,
        Geert Uytterhoeven <geert@linux-m68k.org>,
        linux-renesas-soc@vger.kernel.org,
        Prabhakar <prabhakar.csengg@gmail.com>,
        Sergey Shtylyov <s.shtylyov@omp.ru>,
        Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        linux-kernel@vger.kernel.org
References: <20230118104656.67706-1-wsa+renesas@sang-engineering.com>
 <CAMuHMdXY0c4qKCi057CtJdhUJ+443rtyCjwYyK3qGgwWTZ2a0A@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha512;
        protocol="application/pgp-signature"; boundary="SBOQttDUaUptgdAG"
Content-Disposition: inline
In-Reply-To: <CAMuHMdXY0c4qKCi057CtJdhUJ+443rtyCjwYyK3qGgwWTZ2a0A@mail.gmail.com>
Precedence: bulk


--SBOQttDUaUptgdAG
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hi Geert,

> > +static const struct soc_device_attribute rpcif_info_match[]  =3D {
> > +       { .soc_id =3D "r8a7795", .revision =3D "ES1.*", .data =3D &rpci=
f_info_r8a7795_es1 },
> > +       { .soc_id =3D "r8a7796", .revision =3D "ES1.*", .data =3D &rpci=
f_info_r8a7796_es1 },
>=20
> As we do have a separate compatible value for R-Car M3-W+ aka R-Car M3-W =
ES3.0
> ("renesas,r8a77961-rpc-if"), and there is no R-Car M3-W ES2.x (see the PRR
> screwup handling in renesas_soc_init()), you can just match against
> "renesas,r8a7796-rpc-if instead.

Right, I missed that! This is awesome news because with us dropping H3
ES1 support, this means we can drop all the soc_device_match() handling
now and use compatibles only. There, everybody happy :D

> > +       regmap_update_bits(rpc->regmap, RPCIF_PHYCNT,
> > +                          /* create mask with all affected bits set */
> > +                          RPCIF_PHYCNT_STRTIM(BIT(fls(rpc->info->strti=
m)) - 1),
>=20
> fls(0) =3D 0, and BIT(-1) is undefined, so this won't work for R-Car
> H3 ES1.x.  So I'm afraid you cannot handle this without storing the
> actual mask ;-)

You misread the parens, it is: BIT(0) - 1 =3D 0

I actually wrote a program to print out the calculations to make sure I
got it right. But yeah, ES1 is obsolete now.

I'll send V3 later today.

Thanks everyone,

   Wolfram


--SBOQttDUaUptgdAG
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQIzBAABCgAdFiEEOZGx6rniZ1Gk92RdFA3kzBSgKbYFAmPSdpwACgkQFA3kzBSg
KbY7IRAAiwQXZG3DVHFKWmGKEXMGRWtzC5nzyYs7xb/taY5bhK/VGaYKgOVSI39w
GvhkAyv38k1jbDgFcKgBh8OE/ygvHfm3XBk0/X7d3U1sCycQA/Qr715HNN0InEqb
OaAUy2j5l5KQgEXZZA3QpAvM30eDasi2FTEimMzhsF91zjoGZkw3SagBsU03gd9d
DIb8hFuinpgOqVrwbzaWDqtj+PbBGvfvz+nQuxtpNEtlF/0ETcuBq2yHoWVm2qZ9
0aSDGXRl06k+l7SyreqBTfZXnsXUfsiA5P7rYyZ2l4DPo5KVKnsadSCvLzFvVfm3
84cvz3cPRtGrAL1AbALy1LQ2A3HIYExRRhvDN4/jl8sBAUNqV6FlPs2vF/OzVk/4
6H5WC/TOq+Zc6Cr0FM2uC541hhKyPeOuJWhsxdIFHy9Ov/kJieaBz9P26UwQKwxI
4+0609PeBjY63OuvGjRVN8FjDUupmpYmga8HzKRgKdFsP+EVEhCtPGwA6tI4AEfx
Lrq4OVnD4zCfZBYdJTFvLxMtkLGeME+T4jRGMeWUwltncKm1dC7hvZ6WkRK7FArX
L+QAZKI5aDW+cPgzjRB9vhK5rb7kxRj+BJC3aGL0ZCtXt9wlzn7z/TNsVB6At3lH
BjDQ/3O3hKu4v0CbyclklAS+2y65+3qG3uQpr9Nlq87Dd3OPkoo=
=povj
-----END PGP SIGNATURE-----

--SBOQttDUaUptgdAG--