Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20191018154052.1276506-1-arnd@arndb.de> <20191018154201.1276638-44-arnd@arndb.de>
 <4d6920e9fab519ddf69aae9da13a1cd02d13bddd.camel@v3.sk>
In-Reply-To: <4d6920e9fab519ddf69aae9da13a1cd02d13bddd.camel@v3.sk>
From:   Arnd Bergmann <arnd@arndb.de>
Date:   Sat, 19 Oct 2019 21:34:36 +0200
Message-ID: <CAK8P3a1buT7K7YWAdztqnZ83Jq6f4q-7x_G3Ed09cKQxtKLJSQ@mail.gmail.com>
Subject: Re: [PATCH 44/46] ARM: mmp: remove tavorevb board support
To:     Lubomir Rintel <lkundrak@v3.sk>
Cc:     Daniel Mack <daniel@zonque.org>,
        Haojian Zhuang <haojian.zhuang@gmail.com>,
        Robert Jarzmik <robert.jarzmik@free.fr>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Linus Walleij <linus.walleij@linaro.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Sat, Oct 19, 2019 at 4:20 PM Lubomir Rintel <lkundrak@v3.sk> wrote:
> On Fri, 2019-10-18 at 17:41 +0200, Arnd Bergmann wrote:
> >
> > There is a third board named TavorEVB in the Kconfig description,
> > but this refers to the "TTC_DKB" machine. The two are clearly
> > related, so I change the Kconfig description to just list both
> > names.
> >
> > Cc: Lubomir Rintel <lkundrak@v3.sk>
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>
> Reviewed-by: Lubomir Rintel <lkundrak@v3.sk>
>
> In fact, I'd love to see more non-DT boards go from mach-mmp. There are
> good indications nobody is looking after MMP2-based "Jasper", "Flint"
> and "Brownstone" and they probably weren't seen outside Marvell either.
> The latter has a DTS file.
>
> Would anybody miss them?

Probably not, but I had a hard time identifying any boards in mmp
and some other platforms that are actually worth keeping.

Back in the days, a common way the platforms were maintained
was to only have the official development board in mainline Linux,
while many products were left with out of tree board files.
This means it's impossible to see which SoCs actually got used
in the field and which ones did not. It also means the other
machines stopped getting forward-ported and nobody could test
the mainline changes.

I wouldn't mind just removing all of the machines that were clearly
reference hardware rather than actual products unless we know of
someone still using them.

In case of MMP, that doesn't leave a lot though, the gplugD is the
only one that clearly meant as an end-user product. I'd also leave
all the DT based platforms as a rule, mainly because the DT has
made it possible to support additional boards with a custom dt blob
and no kernel changes.

If you have any more insight into what particular boards actually
were used for, and which ones can get removed, that would
be very welcome.

        Arnd