Received-SPF: pass (google.com: domain of linux-kernel+bounces-52760-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Date: Mon, 5 Feb 2024 15:56:29 +0200
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
To: Arturas Moskvinas <arturas.moskvinas@gmail.com>
Cc: linus.walleij@linaro.org, biju.das.jz@bp.renesas.com,
	akaessens@gmail.com, thomas.preston@codethink.co.uk,
	preid@electromag.com.au, u.kleine-koenig@pengutronix.de,
	linux-gpio@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] pinctrl: mcp23s08: Check only GPIOs which have
 interrupts enabled
Message-ID: <ZcDpDaclAXEa5ETN@smile.fi.intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Organization: Intel Finland Oy - BIC 0357606-4 - Westendinkatu 7, 02160 Espoo


On Fri, Feb 02, 2024 at 12:40:23PM +0200, Arturas Moskvinas wrote:
> 
> On 2/1/24 16:30, Andy Shevchenko wrote:
> > On Thu, Feb 01, 2024 at 04:14:07PM +0200, Arturas Moskvinas wrote:
> > > GPINTEN register contains information about GPIOs with enabled
> > > interrupts no need to check other GPIOs for changes.
> > > 
> > > Signed-off-by: Arturas Moskvinas<arturas.moskvinas@gmail.com>
> > > ---
> > You forgot to add a changelog here, but no need to resend, just you can respond
> > to the email since it's not a big issue in this case.
> Ack.
> > > +	if (mcp_read(mcp, MCP_GPINTEN, &gpinten))
> > > +		goto unlock;
> > Do all hw variants have this register available?
> > Esp. I2C part, wouldn't it be problematic (exception with NACK on the bus)?
> According to specification sheets MCP(s0)17 [1] page 19, MCP(s0)18 [2] page
> 19, MCP(s0)08 [3] page 11 - all supported expanders have that register also
> that register needs to be used [4] to mask/unmask interrupts on given GPIO,
> without it - expander won't even fire an interrupt. I tested on MCP23018 I2C
> expander though but module itself is not treating that expander differently
> for interrupt handling purposes.

Thank you for the clarification!

> Do you want that information to be added as part of commit message or
> information in the mailing thread will be enough?

Up to maintainers. I'm fine with this email. If Linus uses b4 to generate
a Link tag to this discussion it will be enough (in my opinion).

> [1] https://ww1.microchip.com/downloads/aemDocuments/documents/APID/ProductDocuments/DataSheets/MCP23017-Data-Sheet-DS20001952.pdf
> [2] https://ww1.microchip.com/downloads/aemDocuments/documents/APID/ProductDocuments/DataSheets/MCP23018-Data-Sheet-DS20002103.pdf
> [3] https://ww1.microchip.com/downloads/aemDocuments/documents/APID/ProductDocuments/DataSheets/MCP23008-MCP23S08-Data-Sheet-DS20001919.pdf
> [4] https://elixir.bootlin.com/linux/v6.7/source/drivers/pinctrl/pinctrl-mcp23s08.c#L473

-- 
With Best Regards,
Andy Shevchenko