Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:4 as permitted sender) client-ip=2620:137:e000::3:4;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.9.0-alpha0-761-gece9e40c48-fm-20230913.001-gece9e40c
MIME-Version: 1.0
Message-Id: <d0587e7d-95a4-4ce3-b7fe-1aaa94352592@app.fastmail.com>
In-Reply-To: <87msxau3up.wl-ysato@users.sourceforge.jp>
References: <20230802184849.1019466-1-arnd@kernel.org>
 <20230802184849.1019466-4-arnd@kernel.org>
 <CAMuHMdVjmD357K-yxxW-jn-6vKsXTg+u1Psw9DftyxH=dQoMEg@mail.gmail.com>
 <5dad2d86-78ea-4a39-8ee1-98e3eb134d36@app.fastmail.com>
 <CAMuHMdVYcvPL+JpPw9sA48=615cdfwa8d0LP-bVp0NWqbQ+JOw@mail.gmail.com>
 <190041c8-2d99-4bc3-adc3-6fbe902c1265@app.fastmail.com>
 <CAMuHMdXyLHitBWOMp74cqtJbSs6q_4sPOEee+x72tE-E2G-KWg@mail.gmail.com>
 <d2f5cdc1-4bff-4f1d-a7b2-38eee6a6a86d@app.fastmail.com>
 <f61e1f218ee4d5a87121c0e5ee0d8694364ea2dd.camel@physik.fu-berlin.de>
 <CAMuHMdXFSvyTGvYrc2af_Bba9hHNQ-taufOMXRPrKJGNiCP8mw@mail.gmail.com>
 <87msxau3up.wl-ysato@users.sourceforge.jp>
Date:   Mon, 25 Sep 2023 09:36:03 +0200
From:   "Arnd Bergmann" <arnd@arndb.de>
To:     "Yoshinori Sato" <ysato@users.sourceforge.jp>,
        "Geert Uytterhoeven" <geert@linux-m68k.org>
Cc:     "John Paul Adrian Glaubitz" <glaubitz@physik.fu-berlin.de>,
        "Arnd Bergmann" <arnd@kernel.org>, linux-sh@vger.kernel.org,
        "Rich Felker" <dalias@libc.org>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 4/4] sh: machvec: remove custom ioport_{un,}map()
Content-Type: text/plain;charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Mon, Sep 25, 2023, at 09:08, Yoshinori Sato wrote:
> On Thu, 21 Sep 2023 17:52:29 +0900, Geert Uytterhoeven wrote:
>> On Thu, Sep 21, 2023 at 9:45=E2=80=AFAM John Paul Adrian Glaubitz <gl=
aubitz@physik.fu-berlin.de> wrote:
>>=20
>> I think including Arnd's cleanups (that is, his v2) in v6.7 is fine.
>> Sato-san's series needs more work, and is easy to fix for Arnd's clea=
nup
>> (just provide sh_io_port_base unconditionally).
>
> For devicetree support, we have been using GENERIC_IOMAP and GENERIC_P=
CI_IOMAP.
> This change has no effect, so it's okay to be merged first.

Ok, thanks for confirming.

I would still suggest that you try to avoid GENERIC_IOMAP altogether
since sh has no custom inb/outb instructions and can just implement
ioport_map() to return an __iomem token that can be passed into=20
readb/writeb, the same as the non-GENERIC_IOMAP version in sh does.

Ideally you can however remove the iomap.c file and the
asm-generic/iomap.h include and instead get the inline definitions
from the defaults in include/asm-generic/io.h.

    Arnd