MIME-Version: 1.0
References: <20230221133307.20287-1-andriy.shevchenko@linux.intel.com>
 <20230221133307.20287-3-andriy.shevchenko@linux.intel.com>
 <Y/TJs+Arban0ats8@smile.fi.intel.com> <be203dfd290e67c8ce74d11c5c9478a4@protonic.nl>
 <Y/UD3HWNy8uKYShC@smile.fi.intel.com> <0235f0fed989a8b027db720663699f5d@protonic.nl>
 <Y/ZKdN4nuHcL4DgE@smile.fi.intel.com> <Y/ZOyGo8X7r258EC@smile.fi.intel.com>
 <06f29d66-f16a-039c-ecd0-155bdcce00c1@linaro.org> <Y/ZoyaV10TCWhloT@smile.fi.intel.com>
In-Reply-To: <Y/ZoyaV10TCWhloT@smile.fi.intel.com>
From:   Geert Uytterhoeven <geert@linux-m68k.org>
Date:   Thu, 23 Feb 2023 10:55:15 +0100
Message-ID: <CAMuHMdXbYapqG1eirbRoxS9akue+URKOV338efMp1c0CbwUEKg@mail.gmail.com>
Subject: Re: [PATCH v1 2/3] auxdisplay: ht16k33: Make use of device_get_match_data()
To:     Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc:     Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        Robin van der Gracht <robin@protonic.nl>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>,
        Raul E Rangel <rrangel@chromium.org>,
        Wolfram Sang <wsa@kernel.org>, linux-kernel@vger.kernel.org,
        linux-i2c@vger.kernel.org, linux-usb@vger.kernel.org,
        Miguel Ojeda <ojeda@kernel.org>,
        Heikki Krogerus <heikki.krogerus@linux.intel.com>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        devicetree@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8BIT
Precedence: bulk

Hi Andy,

On Wed, Feb 22, 2023 at 8:21 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> On Wed, Feb 22, 2023 at 07:46:25PM +0100, Krzysztof Kozlowski wrote:
> > On 22/02/2023 18:20, Andy Shevchenko wrote:
> > >>> Which effectively breaks i.e. user-space instantiation for other display
> > >>> types which now do work due to i2c_of_match_device().
> > >>> (so my suggestion above is not sufficient).
> > >>>
> > >>> Are you proposing extending and searching the I2C ID table to work around
> > >>> that?
> > >>
> > >> See (1) above. This is the downside I have noticed after sending this series.
> > >> So, the I²C ID table match has to be restored, but the above mentioned issues
> > >> with existing table are not gone, hence they need to be addressed in the next
> > >> version.
> > >
> > > I see now what you mean. So, we have even more issues in this driver:
> > > - I²C table is not in sync with all devices supported
> >
> > Does anything actually rely on i2c_device_id table? ACPI would match
> > either via ACPI or OF tables. All modern ARM systems (e.g. imx6) are
> > DT-based. Maybe just drop the I2C ID table?
>
> For I²C it's still possible to enumerate the device via sysfs, which is ABI.

Yes, and AFAIK, that worked fine. E.g.

    echo adafruit,3130 0x70 > /sys/class/i2c/i2c-adapter/.../new_device

Cfr. https://lore.kernel.org/all/20211019144520.3613926-3-geert@linux-m68k.org/

Note that that example actually includes the manufacturer.
I didn't check whether the I2C core takes that part into account when
matching, or just strips it.

> > > - the OF ID table seems has something really badly formed for adafruit
> > >   (just a number after a comma)
> >
> > Maybe it is a model number? It was documented:
> > Documentation/devicetree/bindings/auxdisplay/holtek,ht16k33.yaml
>
> Yes, it's not a problem for ACPI/DT platforms, the problem is for the above
> way of enumeration, so if we have more than 1 manufacturer that uses plain
> numbers for the model, I²C framework may not distinguish which driver to use.
>
> I.o.w. the part after comma in the compatible strings of the I²C devices must
> be unique globally to make that enumeration disambiguous.

Which is not unique to this driver?
I bet you can find other compatible values that become non-unique
after stripping the manufacturer.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds