[PATCH] reiserfs: check kstrdup failure

[PATCH] reiserfs: check kstrdup failure

[Mikulas Patocka]

Check out-of-memory failure of the kstrdup option. Note that the argument
"arg" may be NULL (in that case kstrup returns NULL), so out of memory
condition happened if arg was non-NULL and kstrdup returned NULL.

The patch also changes the call to replace_mount_options - if we didn't
pass any filesystem-specific options, we don't call replace_mount_options
(thus we don't erase existing reported options).

Note that to properly report options after remount, the reiserfs 
filesystem should implement the show_options method. Without the 
show_options method, options changed with remount replace existing 
options.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>

---
 fs/reiserfs/super.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

Index: linux-4.6/fs/reiserfs/super.c
===================================================================
--- linux-4.6.orig/fs/reiserfs/super.c	2016-05-19 20:59:46.000000000 +0200
+++ linux-4.6/fs/reiserfs/super.c	2016-05-19 21:00:00.000000000 +0200
@@ -1393,7 +1393,7 @@ static int reiserfs_remount(struct super
 	unsigned long safe_mask = 0;
 	unsigned int commit_max_age = (unsigned int)-1;
 	struct reiserfs_journal *journal = SB_JOURNAL(s);
-	char *new_opts = kstrdup(arg, GFP_KERNEL);
+	char *new_opts;
 	int err;
 	char *qf_names[REISERFS_MAXQUOTAS];
 	unsigned int qfmt = 0;
@@ -1401,6 +1401,10 @@ static int reiserfs_remount(struct super
 	int i;
 #endif
 
+	new_opts = kstrdup(arg, GFP_KERNEL);
+	if (arg && !new_opts)
+		return -ENOMEM;
+
 	sync_filesystem(s);
 	reiserfs_write_lock(s);
 
@@ -1546,7 +1550,8 @@ static int reiserfs_remount(struct super
 	}
 
 out_ok_unlocked:
-	replace_mount_options(s, new_opts);
+	if (new_opts)
+		replace_mount_options(s, new_opts);
 	return 0;
 
 out_err_unlock:

Re: [PATCH] reiserfs: check kstrdup failure

[Jan Kara]

On Tue 24-05-16 22:51:27, Mikulas Patocka wrote:
> Check out-of-memory failure of the kstrdup option. Note that the argument
> "arg" may be NULL (in that case kstrup returns NULL), so out of memory
> condition happened if arg was non-NULL and kstrdup returned NULL.
> 
> The patch also changes the call to replace_mount_options - if we didn't
> pass any filesystem-specific options, we don't call replace_mount_options
> (thus we don't erase existing reported options).
> 
> Note that to properly report options after remount, the reiserfs 
> filesystem should implement the show_options method. Without the 
> show_options method, options changed with remount replace existing 
> options.
> 
> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>

OK, makes sense. I've added the patch to my tree and will push it to Linus.
Thanks for the patch.

								Honza
> 
> ---
>  fs/reiserfs/super.c |    9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 
> Index: linux-4.6/fs/reiserfs/super.c
> ===================================================================
> --- linux-4.6.orig/fs/reiserfs/super.c	2016-05-19 20:59:46.000000000 +0200
> +++ linux-4.6/fs/reiserfs/super.c	2016-05-19 21:00:00.000000000 +0200
> @@ -1393,7 +1393,7 @@ static int reiserfs_remount(struct super
>  	unsigned long safe_mask = 0;
>  	unsigned int commit_max_age = (unsigned int)-1;
>  	struct reiserfs_journal *journal = SB_JOURNAL(s);
> -	char *new_opts = kstrdup(arg, GFP_KERNEL);
> +	char *new_opts;
>  	int err;
>  	char *qf_names[REISERFS_MAXQUOTAS];
>  	unsigned int qfmt = 0;
> @@ -1401,6 +1401,10 @@ static int reiserfs_remount(struct super
>  	int i;
>  #endif
>  
> +	new_opts = kstrdup(arg, GFP_KERNEL);
> +	if (arg && !new_opts)
> +		return -ENOMEM;
> +
>  	sync_filesystem(s);
>  	reiserfs_write_lock(s);
>  
> @@ -1546,7 +1550,8 @@ static int reiserfs_remount(struct super
>  	}
>  
>  out_ok_unlocked:
> -	replace_mount_options(s, new_opts);
> +	if (new_opts)
> +		replace_mount_options(s, new_opts);
>  	return 0;
>  
>  out_err_unlock:
> --
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR