Linux Btrfs filesystem development
 help / color / mirror / Atom feed
From: Qu Wenruo <quwenruo.btrfs@gmx.com>
To: Gladyshev Ilya <foxido@foxido.dev>
Cc: Chris Mason <clm@fb.com>, David Sterba <dsterba@suse.com>,
	linux-btrfs@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH 8/8] btrfs: simplify cleanup in btrfs_add_qgroup_relation
Date: Thu, 13 Nov 2025 07:16:41 +1030	[thread overview]
Message-ID: <ec7e5f41-6585-4817-a901-e7a1e7e7b760@gmx.com> (raw)
In-Reply-To: <d9e7808a976e6325bfdc41100bd9b38892663a8b.1762972845.git.foxido@foxido.dev>



在 2025/11/13 05:19, Gladyshev Ilya 写道:
> Remove from cleanup path mutex_unlock via guard() and kfree via
> __free(kfree) macro. With those two cleanups gone, we can remove `out`
> label and replace all gotos with direct returns.
> 
> Signed-off-by: Gladyshev Ilya <foxido@foxido.dev>
> ---
>   fs/btrfs/qgroup.c | 42 ++++++++++++++++--------------------------
>   1 file changed, 16 insertions(+), 26 deletions(-)
> 
> diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
> index 9b2f2c8ca505..238c17c7d969 100644
> --- a/fs/btrfs/qgroup.c
> +++ b/fs/btrfs/qgroup.c
> @@ -1528,65 +1528,55 @@ static int quick_update_accounting(struct btrfs_fs_info *fs_info,
>    * callers and transferred here (either used or freed on error).
>    */
>   int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans, u64 src, u64 dst,
> -			      struct btrfs_qgroup_list *prealloc)
> +			      struct btrfs_qgroup_list *_prealloc)

We don't use '_' prefix. Even '__' usage is discouraged.

>   {
>   	struct btrfs_fs_info *fs_info = trans->fs_info;
>   	struct btrfs_qgroup *parent;
>   	struct btrfs_qgroup *member;
>   	struct btrfs_qgroup_list *list;
> +	struct btrfs_qgroup_list *prealloc __free(kfree) = _prealloc;

I do not think preallocation is a good use case for scope based cleanup, 
especially when there are two similarly named variables.

This reduces the readability.

Thanks,
Qu

>   	int ret = 0;
>   
>   	ASSERT(prealloc);
>   
>   	/* Check the level of src and dst first */
>   	if (btrfs_qgroup_level(src) >= btrfs_qgroup_level(dst)) {
> -		kfree(prealloc);
>   		return -EINVAL;
>   	}
>   
> -	mutex_lock(&fs_info->qgroup_ioctl_lock);
> -	if (!fs_info->quota_root) {
> -		ret = -ENOTCONN;
> -		goto out;
> -	}
> +	guard(mutex)(&fs_info->qgroup_ioctl_lock);
> +
> +	if (!fs_info->quota_root)
> +		return -ENOTCONN;
> +
>   	member = find_qgroup_rb(fs_info, src);
>   	parent = find_qgroup_rb(fs_info, dst);
> -	if (!member || !parent) {
> -		ret = -EINVAL;
> -		goto out;
> -	}
> +	if (!member || !parent)
> +		return -EINVAL;
>   
>   	/* check if such qgroup relation exist firstly */
>   	list_for_each_entry(list, &member->groups, next_group) {
> -		if (list->group == parent) {
> -			ret = -EEXIST;
> -			goto out;
> -		}
> +		if (list->group == parent)
> +			return -EEXIST;
>   	}
>   
>   	ret = add_qgroup_relation_item(trans, src, dst);
>   	if (ret)
> -		goto out;
> +		return ret;
>   
>   	ret = add_qgroup_relation_item(trans, dst, src);
>   	if (ret) {
>   		del_qgroup_relation_item(trans, src, dst);
> -		goto out;
> +		return ret;
>   	}
>   
> -	spin_lock(&fs_info->qgroup_lock);
> +	guard(spinlock)(&fs_info->qgroup_lock);
>   	ret = __add_relation_rb(prealloc, member, parent);
>   	prealloc = NULL;
>   	if (ret < 0) {
> -		spin_unlock(&fs_info->qgroup_lock);
> -		goto out;
> +		return ret;
>   	}
> -	ret = quick_update_accounting(fs_info, src, dst, 1);
> -	spin_unlock(&fs_info->qgroup_lock);
> -out:
> -	kfree(prealloc);
> -	mutex_unlock(&fs_info->qgroup_ioctl_lock);
> -	return ret;
> +	return quick_update_accounting(fs_info, src, dst, 1);
>   }
>   
>   static int __del_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,


  reply	other threads:[~2025-11-12 20:46 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-12 18:49 [RFC PATCH 0/8] use cleanup.h in btrfs Gladyshev Ilya
2025-11-12 18:49 ` [RFC PATCH 1/8] btrfs: remove redundant label in __del_qgroup_relation Gladyshev Ilya
2025-11-12 18:49 ` [RFC PATCH 2/8] btrfs: move kfree out of btrfs_create_qgroup's cleanup path Gladyshev Ilya
2025-11-12 20:30   ` Qu Wenruo
2025-11-12 18:49 ` [RFC PATCH 3/8] btrfs: simplify control flow in scrub_simple_mirror Gladyshev Ilya
2025-11-12 18:49 ` [RFC PATCH 4/8] btrfs: simplify function protections with guards Gladyshev Ilya
2025-11-13  8:43   ` David Sterba
2025-11-13 10:06     ` Gladyshev Ilya
2025-11-13 11:25       ` David Sterba
2025-11-13 12:30         ` Daniel Vacek
2025-11-12 18:49 ` [RFC PATCH 5/8] btrfs: use cleanup.h guard()s to simplify unlocks on return Gladyshev Ilya
2025-11-12 18:49 ` [RFC PATCH 6/8] btrfs: simplify cleanup via scoped_guard() Gladyshev Ilya
2025-11-13  8:48   ` David Sterba
2025-11-12 18:49 ` [RFC PATCH 7/8] btrfs: simplify return path via cleanup.h Gladyshev Ilya
2025-11-12 20:50   ` Qu Wenruo
2025-11-13  8:54     ` David Sterba
2025-11-13 12:48       ` Daniel Vacek
2025-11-12 18:49 ` [RFC PATCH 8/8] btrfs: simplify cleanup in btrfs_add_qgroup_relation Gladyshev Ilya
2025-11-12 20:46   ` Qu Wenruo [this message]
2025-11-12 20:55 ` [RFC PATCH 0/8] use cleanup.h in btrfs Qu Wenruo
2025-11-13  8:01   ` Gladyshev Ilya

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=ec7e5f41-6585-4817-a901-e7a1e7e7b760@gmx.com \
    --to=quwenruo.btrfs@gmx.com \
    --cc=clm@fb.com \
    --cc=dsterba@suse.com \
    --cc=foxido@foxido.dev \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-kernel@vger.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