Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
MIME-Version: 1.0
References: <20220519082552.117736-1-wangkefeng.wang@huawei.com> <20220519082552.117736-2-wangkefeng.wang@huawei.com>
In-Reply-To: <20220519082552.117736-2-wangkefeng.wang@huawei.com>
From:   Arnd Bergmann <arnd@arndb.de>
Date:   Sat, 21 May 2022 22:30:01 +0200
Message-ID: <CAK8P3a3AaYovsu+aOmi62y-2i+0tmHHqpGN_PdToGt_UnGw=Ag@mail.gmail.com>
Subject: Re: [PATCH v3 1/6] ARM: mm: kill unused runtime hook arch_iounmap()
To:     Kefeng Wang <wangkefeng.wang@huawei.com>
Cc:     Catalin Marinas <catalin.marinas@arm.com>,
        Will Deacon <will@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux-MM <linux-mm@kvack.org>,
        Christoph Hellwig <hch@infradead.org>,
        Arnd Bergmann <arnd@arndb.de>,
        Anshuman Khandual <anshuman.khandual@arm.com>,
        Russell King <linux@armlinux.org.uk>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, May 19, 2022 at 10:25 AM Kefeng Wang <wangkefeng.wang@huawei.com> wrote:
>
> Since the following commits,
>
> v5.4
>   commit 59d3ae9a5bf6 ("ARM: remove Intel iop33x and iop13xx support")
> v5.11
>   commit 3e3f354bc383 ("ARM: remove ebsa110 platform")
>
> The runtime hook arch_iounmap() on ARM is useless, kill arch_iounmap()
> and __iounmap(), and the naming of arch_iounmap will be used in
> GENERIC_IOREMAP with the later patch.
>
> Cc: Russell King <linux@armlinux.org.uk>
> Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>

I had a very similar patch prototyped recently,

Reviewed-by: Arnd Bergmann <arnd@arndb.de>

It would be nice to do the same for arch_ioremap_caller(), which now
has two implementations left for mvebu and imx3, previously we had
more for iop13xx, ebsa110, ixp4xx and msm.

For both armada37x/380 and imx3, the only purpose is to override
the mtype argument, and it feels like there should be a better way
to do this, though I'm not sure what that is. Having an overridable
mtype value per 256MB section of physical address space would
be sufficient for both, but I don't know if that's any better than what
we have.

        Arnd