[PATCH v2] btrfs: use kbasename in btrfsic_mount

[PATCH v2] btrfs: use kbasename in btrfsic_mount

[Rasmus Villemoes]

This is more readable.

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
---
v2: make p const char* (thanks 0day).

Original comment:

I think the following strlcpy may be somewhat fragile since obviously
ds->name and p overlap. It certainly relies on strlcpy doing a forward
copy, and since different architectures can have their own strlcpy,
that's hard to verify (and also won't necessarily continue to hold).

Maybe

  if (p != ds->name)
    memmove(ds->name, p, strlen(p)+1);

instead.

 fs/btrfs/check-integrity.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index 0340c57bf377..d6fe77c5b4ac 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -95,6 +95,7 @@
 #include <linux/genhd.h>
 #include <linux/blkdev.h>
 #include <linux/vmalloc.h>
+#include <linux/string.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "hash.h"
@@ -3120,7 +3121,7 @@ int btrfsic_mount(struct btrfs_root *root,
 
 	list_for_each_entry(device, dev_head, dev_list) {
 		struct btrfsic_dev_state *ds;
-		char *p;
+		const char *p;
 
 		if (!device->bdev || !device->name)
 			continue;
@@ -3136,11 +3137,7 @@ int btrfsic_mount(struct btrfs_root *root,
 		ds->state = state;
 		bdevname(ds->bdev, ds->name);
 		ds->name[BDEVNAME_SIZE - 1] = '\0';
-		for (p = ds->name; *p != '\0'; p++);
-		while (p > ds->name && *p != '/')
-			p--;
-		if (*p == '/')
-			p++;
+		p = kbasename(ds->name);
 		strlcpy(ds->name, p, sizeof(ds->name));
 		btrfsic_dev_state_hashtable_add(ds,
 						&btrfsic_dev_state_hashtable);
-- 
2.6.1

Re: [PATCH v2] btrfs: use kbasename in btrfsic_mount

[Andy Shevchenko]

On Fri, Nov 27, 2015 at 10:42 AM, Rasmus Villemoes
<linux@rasmusvillemoes.dk> wrote:
> This is more readable.
>
> Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>

You missed David's tag.

Anyway, also mine is here

Reviewed-by Andy Shevchenko <andy.shevchenko@gmail.com>

> ---
> v2: make p const char* (thanks 0day).
>
> Original comment:
>
> I think the following strlcpy may be somewhat fragile since obviously
> ds->name and p overlap. It certainly relies on strlcpy doing a forward
> copy, and since different architectures can have their own strlcpy,
> that's hard to verify (and also won't necessarily continue to hold).
>
> Maybe
>
>   if (p != ds->name)
>     memmove(ds->name, p, strlen(p)+1);
>
> instead.
>
>  fs/btrfs/check-integrity.c | 9 +++------
>  1 file changed, 3 insertions(+), 6 deletions(-)
>
> diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
> index 0340c57bf377..d6fe77c5b4ac 100644
> --- a/fs/btrfs/check-integrity.c
> +++ b/fs/btrfs/check-integrity.c
> @@ -95,6 +95,7 @@
>  #include <linux/genhd.h>
>  #include <linux/blkdev.h>
>  #include <linux/vmalloc.h>
> +#include <linux/string.h>
>  #include "ctree.h"
>  #include "disk-io.h"
>  #include "hash.h"
> @@ -3120,7 +3121,7 @@ int btrfsic_mount(struct btrfs_root *root,
>
>         list_for_each_entry(device, dev_head, dev_list) {
>                 struct btrfsic_dev_state *ds;
> -               char *p;
> +               const char *p;
>
>                 if (!device->bdev || !device->name)
>                         continue;
> @@ -3136,11 +3137,7 @@ int btrfsic_mount(struct btrfs_root *root,
>                 ds->state = state;
>                 bdevname(ds->bdev, ds->name);
>                 ds->name[BDEVNAME_SIZE - 1] = '\0';
> -               for (p = ds->name; *p != '\0'; p++);
> -               while (p > ds->name && *p != '/')
> -                       p--;
> -               if (*p == '/')
> -                       p++;
> +               p = kbasename(ds->name);
>                 strlcpy(ds->name, p, sizeof(ds->name));
>                 btrfsic_dev_state_hashtable_add(ds,
>                                                 &btrfsic_dev_state_hashtable);
> --
> 2.6.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/



-- 
With Best Regards,
Andy Shevchenko