[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHp75VdthEZL6GvT5Q=f7rbcDfA5XX=7-VLfVz1kZmBFem_eCA@mail.gmail.com>
Date: Tue, 16 May 2023 10:12:22 +0300
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Jiawen Wu <jiawenwu@...stnetic.com>
Cc: netdev@...r.kernel.org, jarkko.nikula@...ux.intel.com,
andriy.shevchenko@...ux.intel.com, mika.westerberg@...ux.intel.com,
jsd@...ihalf.com, Jose.Abreu@...opsys.com, andrew@...n.ch,
hkallweit1@...il.com, linux@...linux.org.uk, linux-i2c@...r.kernel.org,
linux-gpio@...r.kernel.org, mengyuanlou@...-swift.com
Subject: Re: [PATCH net-next v8 6/9] net: txgbe: Support GPIO to SFP socket
On Tue, May 16, 2023 at 5:39 AM Jiawen Wu <jiawenwu@...stnetic.com> wrote:
...
> > > + struct gpio_irq_chip *girq;
> > > + struct wx *wx = txgbe->wx;
> > > + struct gpio_chip *gc;
> > > + struct device *dev;
> > > + int ret;
> >
> > > + dev = &wx->pdev->dev;
> >
> > This can be united with the defintion above.
> >
> > struct device *dev = &wx->pdev->dev;
> >
>
> This is a question that I often run into, when I want to keep this order,
> i.e. lines longest to shortest, but the line of the pointer which get later
> is longer. For this example:
>
> struct wx *wx = txgbe->wx;
> struct device *dev = &wx->pdev->dev;
So, we locate assignments according to the flow. I do not see an issue here.
> should I split the line, or put the long line abruptly there?
The latter is fine.
...
> > > + gc = devm_kzalloc(dev, sizeof(*gc), GFP_KERNEL);
> > > + if (!gc)
> > > + return -ENOMEM;
> > > +
> > > + gc->label = devm_kasprintf(dev, GFP_KERNEL, "txgbe_gpio-%x",
> > > + (wx->pdev->bus->number << 8) | wx->pdev->devfn);
> > > + gc->base = -1;
> > > + gc->ngpio = 6;
> > > + gc->owner = THIS_MODULE;
> > > + gc->parent = dev;
> > > + gc->fwnode = software_node_fwnode(txgbe->nodes.group[SWNODE_GPIO]);
> >
> > Looking at the I²C case, I'm wondering if gpio-regmap can be used for this piece.
>
> I can access this GPIO region directly, do I really need to use regmap?
It's not a matter of access, it's a matter of using an existing
wrapper that will give you already a lot of code done there, i.o.w.
you don't need to reinvent a wheel.
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists