linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Zhangfei Gao <zhangfei.gao@linaro.org>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Christian Brauner <brauner@kernel.org>,
	Wenkai Lin <linwenkai6@hisilicon.com>,
	Chenghai Huang <huangchenghai2@huawei.com>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	Zhangfei Gao <zhangfei.gao@linaro.org>
Subject: [PATCH] chardev: fix consistent error handling in cdev_device_add
Date: Wed, 19 Nov 2025 10:15:40 +0000	[thread overview]
Message-ID: <20251119101540.106441-1-zhangfei.gao@linaro.org> (raw)

Currently cdev_device_add has inconsistent error handling:

- If device_add fails, it calls cdev_del(cdev)
- If cdev_add fails, it only returns error without cleanup

This creates a problem because cdev_set_parent(cdev, &dev->kobj)
establishes a parent-child relationship.
When callers use cdev_del(cdev) to clean up after cdev_add failure,
it also decrements the dev's refcount due to the parent relationship,
causing refcount mismatch.

To unify error handling:
- Set cdev->kobj.parent = NULL first to break the parent relationship
- Then call cdev_del(cdev) for cleanup

This ensures that in both error paths,
the dev's refcount remains consistent and callers don't need
special handling for different failure scenarios.

Signed-off-by: Zhangfei Gao <zhangfei.gao@linaro.org>
---
 fs/char_dev.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/char_dev.c b/fs/char_dev.c
index c2ddb998f3c9..fef6ee1aba66 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -549,8 +549,11 @@ int cdev_device_add(struct cdev *cdev, struct device *dev)
 		cdev_set_parent(cdev, &dev->kobj);
 
 		rc = cdev_add(cdev, dev->devt, 1);
-		if (rc)
+		if (rc) {
+			cdev->kobj.parent = NULL;
+			cdev_del(cdev);
 			return rc;
+		}
 	}
 
 	rc = device_add(dev);
-- 
2.25.1


             reply	other threads:[~2025-11-19 10:15 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-19 10:15 Zhangfei Gao [this message]
2025-11-25 15:32 ` [PATCH] chardev: fix consistent error handling in cdev_device_add Zhangfei Gao
2025-11-26 10:27 ` Christian Brauner
2025-11-26 11:01   ` Zhangfei Gao
2025-11-28  2:08     ` Zhangfei Gao

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=20251119101540.106441-1-zhangfei.gao@linaro.org \
    --to=zhangfei.gao@linaro.org \
    --cc=brauner@kernel.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=huangchenghai2@huawei.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linwenkai6@hisilicon.com \
    --cc=viro@zeniv.linux.org.uk \
    /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;
as well as URLs for NNTP newsgroup(s).