[RESEND PATCH] chardev: set variable ret to -EBUSY before checking minor range overlap

[RESEND PATCH] chardev: set variable ret to -EBUSY before checking minor range overlap

[Chengguang Xu]

When allocating dynamic major, the minor range overlap check
in __register_chrdev_region() will not fail, so actually
there is no real case to passing non negative error code to
caller. However, set variable ret to -EBUSY before chekcking
minor range overlap will avoid false-positive warning from
code analyzing tool(like Smatch) and also make the code more
easy to understand.

Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Chengguang Xu <cgxu519@gmx.com>
---
 fs/char_dev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/char_dev.c b/fs/char_dev.c
index d18cad28c1c3..00dfe17871ac 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -98,7 +98,7 @@ __register_chrdev_region(unsigned int major, unsigned int baseminor,
 			   int minorct, const char *name)
 {
 	struct char_device_struct *cd, *curr, *prev = NULL;
-	int ret = -EBUSY;
+	int ret;
 	int i;

 	if (major >= CHRDEV_MAJOR_MAX) {
@@ -129,6 +129,7 @@ __register_chrdev_region(unsigned int major, unsigned int baseminor,
 		major = ret;
 	}

+	ret = -EBUSY;
 	i = major_to_index(major);
 	for (curr = chrdevs[i]; curr; prev = curr, curr = curr->next) {
 		if (curr->major < major)
--
2.20.1

Re: [RESEND PATCH] chardev: set variable ret to -EBUSY before checking minor range overlap

[Dan Carpenter]

Please don't say RESEND, say [PATCH v2].  RESEND is for when we ignored
your patch.  (Maybe we made a mistake or maybe the mailing list tagged
it as spam and deleted it or something).  Use [PATCH v2] instead.

On Fri, Apr 26, 2019 at 03:38:37PM +0800, Chengguang Xu wrote:
> When allocating dynamic major, the minor range overlap check
> in __register_chrdev_region() will not fail, so actually
> there is no real case to passing non negative error code to
> caller. However, set variable ret to -EBUSY before chekcking
> minor range overlap will avoid false-positive warning from
> code analyzing tool(like Smatch) and also make the code more
> easy to understand.
> 
> Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
> Signed-off-by: Chengguang Xu <cgxu519@gmx.com>
> ---

Then here under the --- cut off line put:

v2: rebase against the latest linux-next

That way we will remember why the patch was sent twice and what changed.

>  fs/char_dev.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 

regards,
dan carpenter