linux-kernel - Re: [PATCH 2/2] gpio: mcp23s08: add pinconf support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CACRpkdZ6Dh9wf8FXnHO5t8WnLfhLyTzNED_oRUzx=+3a0N8fQQ@mail.gmail.com>
Date:   Wed, 1 Feb 2017 16:12:24 +0100
From:   Linus Walleij <linus.walleij@...aro.org>
To:     Sebastian Reichel <sre@...nel.org>
Cc:     Alexandre Courbot <gnurou@...il.com>,
        "linux-gpio@...r.kernel.org" <linux-gpio@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] gpio: mcp23s08: add pinconf support

On Mon, Jan 30, 2017 at 5:40 PM, Sebastian Reichel <sre@...nel.org> wrote:

>> Can we move the patch to drivers/pinctrl/* like all other mixed drivers
>> doing combined pinctrl and GPIO?
>
> Sure. Two questions:
>
>  * Should the config name change to PINCTRL_MCP23s08 (or PINCTRL_MCP23XXX)?
>    This will mean people will have to fix their .config

I never understood what the right way is to deal with this.

What happens actually when you run "make oldconfig"?

Is there a way to supply a replacement Kconfig symbol in a good
way?

Else I guess we can keep a dangling GPIO_MCP23S08 bool
that just selects the new PINCTRL_MCP23XXX for now?
With a comment that it got moved?

(Please verify that it works though...)

>  * Should there be a "SPI or I2C GPIO expanders" sub-menu for it as in the
>    gpio Kconfig?

Maybe, that would be a separate refactoring thing for later.
I think it is this one and SX150X that use I2C right now.

> I did not convert the custom caching in the regmap patch, since it
> should be done in its own cleanup patch. Introducing basic regmap
> was much more straight forward than removing the open-coded caching.

OK I take it that it will be fixed later then, that's fine.

> I guess the custom platform data based pullup config could also
> be removed. I just checked and there are not many users of the
> platform data:
>
> linux/arch $ git grep -l mcp23 | grep -v "/dts/"
> blackfin/mach-bf527/boards/tll6527m.c
> blackfin/mach-bf609/boards/ezkit.c
>
> None of them seems to use the pullups variable.

Just kill it off, good riddance!

As separate patch before/after moving it I guess.

Yours,
Linus Walleij