From: Greg KH <gregkh@linuxfoundation.org>
To: Harshit Shaw <shawharshit116@gmail.com>
Cc: deller@gmx.de, chintanlike@gmail.com, tzimmermann@suse.de,
andriy.shevchenko@intel.com, linux-staging@lists.linux.dev,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] staging: fbtft: convert sysfs attributes to use attribute_group
Date: Tue, 12 May 2026 11:44:20 +0200 [thread overview]
Message-ID: <2026051235-lent-genre-813a@gregkh> (raw)
In-Reply-To: <20260512092817.1941-1-shawharshit116@gmail.com>
On Tue, May 12, 2026 at 09:28:17AM +0000, Harshit Shaw wrote:
> Replace direct device_create_file() and device_remove_file() calls
> with the correct attribute_group API using sysfs_create_group() and
> sysfs_remove_group(). This is the proper way to register sysfs
> attributes in kernel drivers.
>
> Signed-off-by: Harshit Shaw <shawharshit116@gmail.com>
> ---
> drivers/staging/fbtft/fbtft-sysfs.c | 83 ++++++++++++++++++-----------
> 1 file changed, 52 insertions(+), 31 deletions(-)
>
> diff --git a/drivers/staging/fbtft/fbtft-sysfs.c b/drivers/staging/fbtft/fbtft-sysfs.c
> index d05599d80011..8d3c4e49b68e 100644
> --- a/drivers/staging/fbtft/fbtft-sysfs.c
> +++ b/drivers/staging/fbtft/fbtft-sysfs.c
> @@ -142,9 +142,30 @@ static ssize_t show_gamma_curve(struct device *device,
> return sprintf_gamma(par, par->gamma.curves, buf);
> }
>
> -static struct device_attribute gamma_device_attrs[] = {
> - __ATTR(gamma, 0660, show_gamma_curve, store_gamma_curve),
> -};
> +static ssize_t store_debug(struct device *device,
> + struct device_attribute *attr,
> + const char *buf, size_t count)
> +{
> + struct fb_info *fb_info = dev_get_drvdata(device);
> + struct fbtft_par *par = fb_info->par;
> + int ret;
> +
> + ret = kstrtoul(buf, 10, &par->debug);
> + if (ret)
> + return ret;
> + fbtft_expand_debug_value(&par->debug);
> +
> + return count;
> +}
> +
> +static ssize_t show_debug(struct device *device,
> + struct device_attribute *attr, char *buf)
> +{
> + struct fb_info *fb_info = dev_get_drvdata(device);
> + struct fbtft_par *par = fb_info->par;
> +
> + return sysfs_emit(buf, "%lu\n", par->debug);
> +}
>
> void fbtft_expand_debug_value(unsigned long *debug)
> {
> @@ -173,45 +194,45 @@ void fbtft_expand_debug_value(unsigned long *debug)
> }
> }
>
> -static ssize_t store_debug(struct device *device,
> - struct device_attribute *attr,
> - const char *buf, size_t count)
> -{
> - struct fb_info *fb_info = dev_get_drvdata(device);
> - struct fbtft_par *par = fb_info->par;
> - int ret;
> -
> - ret = kstrtoul(buf, 10, &par->debug);
> - if (ret)
> - return ret;
> - fbtft_expand_debug_value(&par->debug);
> +static DEVICE_ATTR(gamma, 0644, show_gamma_curve, store_gamma_curve);
> +static DEVICE_ATTR(debug, 0644, show_debug, store_debug);
>
> - return count;
> -}
> +static struct attribute *fbtft_attrs[] = {
> + &dev_attr_debug.attr,
> + NULL,
> +};
>
> -static ssize_t show_debug(struct device *device,
> - struct device_attribute *attr, char *buf)
> -{
> - struct fb_info *fb_info = dev_get_drvdata(device);
> - struct fbtft_par *par = fb_info->par;
> +static struct attribute *fbtft_gamma_attrs[] = {
> + &dev_attr_gamma.attr,
> + NULL,
> +};
>
> - return sysfs_emit(buf, "%lu\n", par->debug);
> -}
> +static const struct attribute_group fbtft_attr_group = {
> + .attrs = fbtft_attrs,
> +};
>
> -static struct device_attribute debug_device_attr =
> - __ATTR(debug, 0660, show_debug, store_debug);
> +static const struct attribute_group fbtft_gamma_attr_group = {
> + .attrs = fbtft_gamma_attrs,
> +};
Why not use the ATTRIBUTE_GROUPS() macro instead? That's what it is
there for :)
> void fbtft_sysfs_init(struct fbtft_par *par)
> {
> struct device *dev;
> + int ret;
>
> dev = dev_of_fbinfo(par->info);
> if (!dev)
> return;
>
> - device_create_file(dev, &debug_device_attr);
> - if (par->gamma.curves && par->fbtftops.set_gamma)
> - device_create_file(dev, &gamma_device_attrs[0]);
> + ret = sysfs_create_group(&dev->kobj, &fbtft_attr_group);
Close, but if you ever find a driver calling a "raw" sysfs_* function,
something is really wrong.
This group needs to be added to the driver itself, as a default group
pointer, and then the driver core will add/remove the group
automatically for you when a device is bound/unbound from the driver.
No need to ever do this on its own.
thanks,
greg k-h
next prev parent reply other threads:[~2026-05-12 9:44 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-12 9:28 [PATCH] staging: fbtft: convert sysfs attributes to use attribute_group Harshit Shaw
2026-05-12 9:44 ` Greg KH [this message]
2026-05-12 10:58 ` Andy Shevchenko
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=2026051235-lent-genre-813a@gregkh \
--to=gregkh@linuxfoundation.org \
--cc=andriy.shevchenko@intel.com \
--cc=chintanlike@gmail.com \
--cc=deller@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-staging@lists.linux.dev \
--cc=shawharshit116@gmail.com \
--cc=tzimmermann@suse.de \
/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