Devicetree
 help / color / mirror / Atom feed
From: Charles Perry <charles.perry@microchip.com>
To: Andrew Lunn <andrew@lunn.ch>
Cc: Charles Perry <charles.perry@microchip.com>,
	<netdev@vger.kernel.org>, Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>,
	"Paolo Abeni" <pabeni@redhat.com>, Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Heiner Kallweit <hkallweit1@gmail.com>,
	Russell King <linux@armlinux.org.uk>,
	<devicetree@vger.kernel.org>, <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH net-next 2/2] net: mdio: add a driver for PIC64-HPSC/HX MDIO controller
Date: Thu, 19 Mar 2026 12:26:45 -0700	[thread overview]
Message-ID: <abxN9aGnCYA6s13u@bby-cbu-swbuild03.eng.microchip.com> (raw)
In-Reply-To: <6dc9a358-4140-457f-bdfb-b4fb22aaf88e@lunn.ch>

On Thu, Mar 19, 2026 at 05:55:44PM +0100, Andrew Lunn wrote:
> > +static int pic64hpsc_mdio_probe(struct platform_device *pdev)
> > +{
> > +	struct device_node *np = pdev->dev.of_node;
> > +	struct device *dev = &pdev->dev;
> > +	struct pic64hpsc_mdio_dev *priv;
> > +	struct mii_bus *bus;
> > +	unsigned long rate;
> > +	struct clk *clk;
> > +	u32 bus_freq;
> > +	u32 div;
> > +	int ret;
> > +
> > +	bus = devm_mdiobus_alloc_size(dev, sizeof(*priv));
> > +	if (!bus)
> > +		return -ENOMEM;
> > +
> > +	priv = bus->priv;
> > +
> > +	priv->regs = devm_platform_ioremap_resource(pdev, 0);
> > +	if (IS_ERR(priv->regs))
> > +		return PTR_ERR(priv->regs);
> > +
> > +	bus->name = KBUILD_MODNAME;
> > +	bus->read = pic64hpsc_mdio_read;
> > +	bus->write = pic64hpsc_mdio_write;
> > +	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
> > +	bus->parent = dev;
> > +
> > +	clk = devm_clk_get_optional_enabled(dev, NULL);
> > +	if (IS_ERR(clk))
> > +		return PTR_ERR(clk);
> 
> What is the use case for not listing the clock? Optional clocks are
> generally because it was forgotten about in the initial driver, and
> added later. In order to not break backwards compatibility, the clock
> needs to be optional.
> 
> But this is a new driver. Why not make it required?
> 

My idea is that if someone wants to use whatever is the hardware default
or what was set by the bootloader, they have an option to do so. For that
reason, I made the clock and the clock-frequency optional. This is
something I can do without if you think it will homogenize better with new
drivers.

Now I just realized that I can achieve this by just making the
clock-frequency optional and not the clock.

Looking at some other MDIO drivers, I can see that there's different
policies on the "clock-frequency" not specified case:

 - mdio-airoha.c: use 2.5MHz if not specified
 - mdio-ipq4019.c: detect if the prescaler is the out of reset value,
                   choose 2.5MHz if that's the case.
 - mdio-mscc-miim.c and mdio-bcm-unimac.c: keep the current settings

I can implement any of the three policy above and don't have strong
opinions about this.

> > +
> > +	of_property_read_u32(np, "clock-frequency", &bus_freq);
> > +
> > +	if (bus_freq) {
> > +		if (!clk) {
> > +			dev_err(dev,
> > +				"cannot use clock-frequency without a clock\n");
> > +			return -EINVAL;
> > +		}
> 
> And this then gets simpler.
> 

Ok, I'll make the clock mandatory.

> > +
> > +		rate = clk_get_rate(clk);
> > +
> > +		div = DIV_ROUND_UP(rate, 2 * bus_freq) - 1;
> > +		if (div == 0 || div & ~MDIO_CFG_PRESCALE_MASK) {
> > +			dev_err(dev, "Incorrect MDIO clock frequency\n");
> 
> I think "Out of range" is more correct.

Ack.

Thank you,
Charles

> 
> 	Andrew

  reply	other threads:[~2026-03-19 19:27 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-17 18:46 [PATCH net-next 0/2] Add support for PIC64-HPSC/HX MDIO controller Charles Perry
2026-03-17 18:46 ` [PATCH net-next 1/2] dt-bindings: net: document Microchip " Charles Perry
2026-03-18 17:48   ` Conor Dooley
2026-03-18 21:23     ` Charles Perry
2026-03-19  1:35       ` Conor Dooley
2026-03-23 13:38         ` Charles Perry
2026-03-23 19:35           ` Conor Dooley
2026-03-19 16:47   ` Andrew Lunn
2026-03-19 16:59   ` Andrew Lunn
2026-03-19 19:36     ` Charles Perry
2026-03-17 18:46 ` [PATCH net-next 2/2] net: mdio: add a driver for " Charles Perry
2026-03-18  9:52   ` Maxime Chevallier
2026-03-18 21:25     ` Charles Perry
2026-03-19 16:55   ` Andrew Lunn
2026-03-19 19:26     ` Charles Perry [this message]
2026-03-19 19:53       ` Andrew Lunn
2026-03-19 21:38         ` Charles Perry
2026-03-19 16:56   ` Andrew Lunn
2026-03-19 19:31     ` Charles Perry
2026-03-19 17:03   ` Andrew Lunn
2026-03-19 19:33     ` Charles Perry

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=abxN9aGnCYA6s13u@bby-cbu-swbuild03.eng.microchip.com \
    --to=charles.perry@microchip.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=andrew@lunn.ch \
    --cc=conor+dt@kernel.org \
    --cc=davem@davemloft.net \
    --cc=devicetree@vger.kernel.org \
    --cc=edumazet@google.com \
    --cc=hkallweit1@gmail.com \
    --cc=krzk+dt@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@armlinux.org.uk \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=robh@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox