Re: RE: [PATCH] Add Silicom Platform Driver

[Christophe JAILLET]

Le 28/07/2023 à 14:59, Huibin Shi a écrit :
> Christophe,
>
> Thanks for the comments. See my comments below.
>
> Updated patch will be sent out later after review comments from other reviewers are addressed.
>
> Henry
> -----Original Message-----
> From: Christophe JAILLET <christophe.jaillet@xxxxxxxxxx>
> Sent: Tuesday, July 25, 2023 5:03 PM
> To: Henry Shi <henryshi2018@xxxxxxxxx>; hbshi69@xxxxxxxxxxx; tglx@xxxxxxxxxxxxx; mingo@xxxxxxxxxx; bp@xxxxxxxxx; dave.hansen@xxxxxxxxxxxxxxx; x86@xxxxxxxxxx; hpa@xxxxxxxxx; hdegoede@xxxxxxxxxx; markgross@xxxxxxxxxx; jdelvare@xxxxxxxx; linux@xxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; platform-driver-x86@xxxxxxxxxxxxxxx; linux-hwmon@xxxxxxxxxxxxxxx
> Cc: hb_shi2003@xxxxxxxxx; Huibin Shi <henrys@xxxxxxxxxxxxxxx>; Wen Wang <wenw@xxxxxxxxxxxxxxx>
> Subject: Re: [PATCH] Add Silicom Platform Driver
>
> Caution: This is an external email. Please take care when clicking links or opening attachments.
>
>
> Le 18/07/2023 à 18:01, Henry Shi a écrit :
> > The Silicom platform (silicom-platform) Linux driver for Swisscom
> > Business Box (Swisscom BB) as well as Cordoba family products is a
> > software solution designed to facilitate the efficient management and
> > control of devices through the integration of various Linux
> > frameworks. This platform driver provides seamless support for device
> > management via the Linux LED framework, GPIO framework, Hardware
> > Monitoring (HWMON), and device attributes. The Silicom platform
> > driver's compatibility with these Linux frameworks allows applications
> > to access and control Cordoba family devices using existing software
> > that is compatible with these frameworks. This compatibility
> > simplifies the development process, reduces dependencies on
> > proprietary solutions, and promotes interoperability with other
> > Linux-based systems and software.
> >
> > Signed-off-by: Henry Shi <henryshi2018@xxxxxxxxx>
> > ---
>
> [...]
>
> > +static int __init silicom_mc_leds_register(struct device *dev,
> > +                                        const struct led_classdev_mc
> > +*mc_leds) {
> > +     struct led_classdev_mc *led;
> > +     int i, err;
> > +
> > +     for (i = 0; mc_leds[i].led_cdev.name; i++) {
> > +             /* allocate and copy data from the init constansts */
> > +             led = devm_kzalloc(dev, sizeof(struct led_classdev_mc),
> > + GFP_KERNEL);
>
> sizeof(*led) is shorter.
> Mostly a matter of taste.
>
> Maybe even devm_kmemdup()?
>
> Henry: thanks. Devm_kmemdup() API requires additional argument that is not necessary of this driver. I prefer devm_kzalloc for now.

CJ: The only additionnal parameter I can think of are the one of 
memcpy() ...

> > +             if (IS_ERR_OR_NULL(led)) {
>
> if (!led)
> is enough.
>
> Henry: OK, changed
>
> > +                     dev_err(dev, "Failed to alloc
> > + led_classdev_mc[%d]: %ld\n", i, PTR_ERR(led));
>
> This kind of message is useless and should be removed (checkpatch should warn about it)
>
> Henry: OK, removed.
>
> > +                     return -ENOMEM;
> > +             }
> > +             memcpy(led, &mc_leds[i], sizeof(*led));

... here.

devm_kzalloc() + this memcpy() could be done with only devm_kmemdup().

This is mostly a matter of taste.

> > +
> > +             led->subled_info = devm_kzalloc(dev, led->num_colors * sizeof(struct mc_subled),
> > +                                             GFP_KERNEL);
>
> Maybe even devm_kmemdup()?

Same...

> > +             if (IS_ERR_OR_NULL(led->subled_info)) {
>
> if (!led->subled_info)
> is enough.
>
> Henry: OK, changed.
>
> > +                     dev_err(dev, "Failed to alloc subled_info[%d]: %ld\n",
> > +                             i, PTR_ERR(led->subled_info));
>
> This kind of message is useless and should be removed (checkpatch should warn about it)
>
> Henry: OK, removed.
>
> > +                     return -ENOMEM;
> > +             }
> > +             memcpy(led->subled_info, mc_leds[i].subled_info,
> > +                     led->num_colors * sizeof(struct mc_subled));

... here.

CJ