Linux SOC development
 help / color / mirror / Atom feed
From: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
To: "Marek Behún" <kabel@kernel.org>
Cc: Gregory CLEMENT <gregory.clement@bootlin.com>,
	 Andrew Lunn <andrew@lunn.ch>, Arnd Bergmann <arnd@arndb.de>,
	 soc@kernel.org, arm@kernel.org,
	Andy Shevchenko <andy@kernel.org>,
	 Hans de Goede <hdegoede@redhat.com>
Subject: Re: [PATCH 10/19] firmware: turris-mox-rwtm: Simplify driver kobject code
Date: Thu, 13 Jun 2024 11:31:43 +0300 (EEST)	[thread overview]
Message-ID: <7a6603a0-c207-f500-209a-ab2b9a80f26b@linux.intel.com> (raw)
In-Reply-To: <20240612135443.30239-11-kabel@kernel.org>

[-- Attachment #1: Type: text/plain, Size: 3386 bytes --]

On Wed, 12 Jun 2024, Marek Behún wrote:

> Drop the mox_kobject wrapper that needs to be allocated, instead put the
> kobject directly into the driver private structure. This allows us to
> drop one kzalloc() call.
> 
> Signed-off-by: Marek Behún <kabel@kernel.org>
> ---
>  drivers/firmware/turris-mox-rwtm.c | 36 ++++++++----------------------
>  1 file changed, 9 insertions(+), 27 deletions(-)
> 
> diff --git a/drivers/firmware/turris-mox-rwtm.c b/drivers/firmware/turris-mox-rwtm.c
> index 6d1e0b1dd2b4..84ec72575c4d 100644
> --- a/drivers/firmware/turris-mox-rwtm.c
> +++ b/drivers/firmware/turris-mox-rwtm.c
> @@ -21,7 +21,6 @@
>  #include <linux/module.h>
>  #include <linux/mutex.h>
>  #include <linux/platform_device.h>
> -#include <linux/slab.h>
>  #include <linux/sysfs.h>
>  #include <linux/types.h>
>  
> @@ -58,13 +57,11 @@ enum mbox_cmd {
>  	MBOX_CMD_OTP_WRITE	= 8,
>  };
>  
> -struct mox_kobject;
> -
>  struct mox_rwtm {
>  	struct device *dev;
>  	struct mbox_client mbox_client;
>  	struct mbox_chan *mbox;
> -	struct mox_kobject *kobj;
> +	struct kobject kobj;
>  	struct hwrng hwrng;
>  
>  	struct armada_37xx_rwtm_rx_msg reply;
> @@ -97,19 +94,9 @@ struct mox_rwtm {
>  #endif
>  };
>  
> -struct mox_kobject {
> -	struct kobject kobj;
> -	struct mox_rwtm *rwtm;
> -};
> -
> -static inline struct kobject *rwtm_to_kobj(struct mox_rwtm *rwtm)
> -{
> -	return &rwtm->kobj->kobj;
> -}
> -
>  static inline struct mox_rwtm *to_rwtm(struct kobject *kobj)
>  {
> -	return container_of(kobj, struct mox_kobject, kobj)->rwtm;
> +	return container_of(kobj, struct mox_rwtm, kobj);
>  }
>  
>  #define MOX_ATTR_RO(name, format, cat)				\
> @@ -142,9 +129,8 @@ static struct attribute *mox_rwtm_attrs[] = {
>  };
>  ATTRIBUTE_GROUPS(mox_rwtm);
>  
> -static void mox_kobj_release(struct kobject *kobj)
> +static void mox_kobj_release(struct kobject *)
>  {
> -	kfree(to_rwtm(kobj)->kobj);
>  }

Is empty .release necessary at all? I found some kobj_type structs without 
.release so I'd expect it to be unnecessary.

-- 
 i.

>  static const struct kobj_type mox_kobj_ktype = {
> @@ -155,18 +141,14 @@ static const struct kobj_type mox_kobj_ktype = {
>  
>  static int mox_kobj_create(struct mox_rwtm *rwtm)
>  {
> -	rwtm->kobj = kzalloc(sizeof(*rwtm->kobj), GFP_KERNEL);
> -	if (!rwtm->kobj)
> -		return -ENOMEM;
> +	struct kobject *kobj = &rwtm->kobj;
>  
> -	kobject_init(rwtm_to_kobj(rwtm), &mox_kobj_ktype);
> -	if (kobject_add(rwtm_to_kobj(rwtm), firmware_kobj, "turris-mox-rwtm")) {
> -		kobject_put(rwtm_to_kobj(rwtm));
> +	kobject_init(kobj, &mox_kobj_ktype);
> +	if (kobject_add(kobj, firmware_kobj, "turris-mox-rwtm")) {
> +		kobject_put(kobj);
>  		return -ENXIO;
>  	}
>  
> -	rwtm->kobj->rwtm = rwtm;
> -
>  	return 0;
>  }
>  
> @@ -540,7 +522,7 @@ static int turris_mox_rwtm_probe(struct platform_device *pdev)
>  free_channel:
>  	mbox_free_channel(rwtm->mbox);
>  put_kobj:
> -	kobject_put(rwtm_to_kobj(rwtm));
> +	kobject_put(&rwtm->kobj);
>  	return ret;
>  }
>  
> @@ -548,7 +530,7 @@ static void turris_mox_rwtm_remove(struct platform_device *pdev)
>  {
>  	struct mox_rwtm *rwtm = platform_get_drvdata(pdev);
>  
> -	kobject_put(rwtm_to_kobj(rwtm));
> +	kobject_put(&rwtm->kobj);
>  	mbox_free_channel(rwtm->mbox);
>  }
>  
> 

  reply	other threads:[~2024-06-13  8:31 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-12 13:54 [PATCH 00/19] Updates for turris-mox-rwtm driver Marek Behún
2024-06-12 13:54 ` [PATCH 01/19] firmware: turris-mox-rwtm: Do not complete if there are no waiters Marek Behún
2024-06-12 13:54 ` [PATCH 02/19] firmware: turris-mox-rwtm: Fix checking return value of wait_for_completion_timeout() Marek Behún
2024-06-13  8:06   ` Ilpo Järvinen
2024-06-12 13:54 ` [PATCH 03/19] firmware: turris-mox-rwtm: Use PAGE_SIZE instead of hardcoded 4096 Marek Behún
2024-06-13  8:01   ` Ilpo Järvinen
2024-06-13  9:54     ` Marek Behún
2024-06-12 13:54 ` [PATCH 04/19] firmware: turris-mox-rwtm: Use ETH_ALEN instead of hardcoded 6 Marek Behún
2024-06-13  8:02   ` Ilpo Järvinen
2024-06-12 13:54 ` [PATCH 05/19] firmware: turris-mox-rwtm: Use the boolean type where appropriate Marek Behún
2024-06-12 13:54 ` [PATCH 06/19] firmware: turris-mox-rwtm: Hide signature related constants behind macros Marek Behún
2024-06-12 13:54 ` [PATCH 07/19] firmware: turris-mox-rwtm: Fix driver includes Marek Behún
2024-06-12 13:54 ` [PATCH 08/19] firmware: turris-mox-rwtm: Use kobj_type's default_groups member for automatic sysfs files creation Marek Behún
2024-06-13  8:14   ` Ilpo Järvinen
2024-06-12 13:54 ` [PATCH 09/19] firmware: turris-mox-rwtm: Simplify debugfs code Marek Behún
2024-06-12 13:54 ` [PATCH 10/19] firmware: turris-mox-rwtm: Simplify driver kobject code Marek Behún
2024-06-13  8:31   ` Ilpo Järvinen [this message]
2024-06-13  9:55     ` Marek Behún
2024-06-13 10:19       ` Ilpo Järvinen
2024-06-13 13:31         ` Marek Behún
2024-06-13 13:37           ` Ilpo Järvinen
2024-06-13 15:39             ` Marek Behún
2024-06-13 15:44               ` Ilpo Järvinen
2024-06-12 13:54 ` [PATCH 11/19] firmware: turris-mox-rwtm: Return true error code if kobject_add() fails Marek Behún
2024-06-12 13:54 ` [PATCH 12/19] firmware: turris-mox-rwtm: Convert rest to devm_* and get rid of driver .remove() Marek Behún
2024-06-12 13:54 ` [PATCH 13/19] firmware: turris-mox-rwtm: Use dev_err_probe() where possible Marek Behún
2024-06-12 13:54 ` [PATCH 14/19] firmware: turris-mox-rwtm: Rearrange probe calls Marek Behún
2024-06-12 13:54 ` [PATCH 15/19] firmware: turris-mox-rwtm: Drop redundant device pointer Marek Behún
2024-06-12 13:54 ` [PATCH 16/19] firmware: turris-mox-rwtm: Use devm_mutex_init() instead of mutex_init() Marek Behún
2024-06-12 13:54 ` [PATCH 17/19] firmware: turris-mox-rwtm: Use container_of() instead of hwrng .priv member Marek Behún
2024-06-12 13:54 ` [PATCH 18/19] firmware: turris-mox-rwtm: Deduplicate command execution code Marek Behún
2024-06-12 13:54 ` [PATCH 19/19] firmware: turris-mox-rwtm: Use EOPNOTSUPP instead of ENOSYS Marek Behún

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=7a6603a0-c207-f500-209a-ab2b9a80f26b@linux.intel.com \
    --to=ilpo.jarvinen@linux.intel.com \
    --cc=andrew@lunn.ch \
    --cc=andy@kernel.org \
    --cc=arm@kernel.org \
    --cc=arnd@arndb.de \
    --cc=gregory.clement@bootlin.com \
    --cc=hdegoede@redhat.com \
    --cc=kabel@kernel.org \
    --cc=soc@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