linux-kernel - Re: [PATCH 1/4] fs/ntfs3: In function ntfs_set_acl_ex do not change inode->i_mode if called from function ntfs_init

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20211026184035.jkopoeaqukeofmye@kari-VirtualBox>
Date:   Tue, 26 Oct 2021 21:40:35 +0300
From:   Kari Argillander <kari.argillander@...il.com>
To:     Konstantin Komarov <almaz.alexandrovich@...agon-software.com>
Cc:     ntfs3@...ts.linux.dev, linux-kernel@...r.kernel.org,
        linux-fsdevel@...r.kernel.org, Joe Perches <joe@...ches.com>
Subject: Re: [PATCH 1/4] fs/ntfs3: In function ntfs_set_acl_ex do not change
 inode->i_mode if called from function ntfs_init_acl

Maybe little too long subject line.

On Mon, Oct 25, 2021 at 07:58:26PM +0300, Konstantin Komarov wrote:
> ntfs_init_acl sets mode. ntfs_init_acl calls ntfs_set_acl_ex.
> ntfs_set_acl_ex must not change this mode.
> Fixes xfstest generic/444
> Fixes: 83e8f5032e2d ("fs/ntfs3: Add attrib operations")

Where does this commit id come from? Seems wrong to me.

> 
> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@...agon-software.com>
> ---
>  fs/ntfs3/xattr.c | 13 +++++++------
>  1 file changed, 7 insertions(+), 6 deletions(-)
> 
> diff --git a/fs/ntfs3/xattr.c b/fs/ntfs3/xattr.c
> index 2143099cffdf..97b5f8417d85 100644
> --- a/fs/ntfs3/xattr.c
> +++ b/fs/ntfs3/xattr.c
> @@ -538,7 +538,7 @@ struct posix_acl *ntfs_get_acl(struct inode *inode, int type)
>  
>  static noinline int ntfs_set_acl_ex(struct user_namespace *mnt_userns,
>  				    struct inode *inode, struct posix_acl *acl,
> -				    int type)
> +				    int type, int init_acl)

Like Joe say. Bool here and use true/false

>  {
>  	const char *name;
>  	size_t size, name_len;
> @@ -551,8 +551,9 @@ static noinline int ntfs_set_acl_ex(struct user_namespace *mnt_userns,
>  
>  	switch (type) {
>  	case ACL_TYPE_ACCESS:
> -		if (acl) {
> -			umode_t mode = inode->i_mode;
> +		/* Do not change i_mode if we are in init_acl */
> +		if (acl && !init_acl) {
> +			umode_t mode;
>  
>  			err = posix_acl_update_mode(mnt_userns, inode, &mode,
>  						    &acl);
> @@ -613,7 +614,7 @@ static noinline int ntfs_set_acl_ex(struct user_namespace *mnt_userns,
>  int ntfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode,
>  		 struct posix_acl *acl, int type)
>  {
> -	return ntfs_set_acl_ex(mnt_userns, inode, acl, type);
> +	return ntfs_set_acl_ex(mnt_userns, inode, acl, type, 0);
>  }
>  
>  /*
> @@ -633,7 +634,7 @@ int ntfs_init_acl(struct user_namespace *mnt_userns, struct inode *inode,
>  
>  	if (default_acl) {
>  		err = ntfs_set_acl_ex(mnt_userns, inode, default_acl,
> -				      ACL_TYPE_DEFAULT);
> +				      ACL_TYPE_DEFAULT, 1);
>  		posix_acl_release(default_acl);
>  	} else {
>  		inode->i_default_acl = NULL;
> @@ -644,7 +645,7 @@ int ntfs_init_acl(struct user_namespace *mnt_userns, struct inode *inode,
>  	else {
>  		if (!err)
>  			err = ntfs_set_acl_ex(mnt_userns, inode, acl,
> -					      ACL_TYPE_ACCESS);
> +					      ACL_TYPE_ACCESS, 1);
>  		posix_acl_release(acl);
>  	}
>  
> -- 
> 2.33.0
>