Re: [PATCH 05/13] vfs: allow rmdir to wait for delegation break on parent

On Mon 13-10-25 10:48:03, Jeff Layton wrote:
> In order to add directory delegation support, we need to break
> delegations on the parent whenever there is going to be a change in the
> directory.
> 
> Add a "delegated_inode" return pointer to vfs_rmdir() and populate that
> pointer with the parent inode if it's non-NULL. Most existing in-kernel
> callers pass in a NULL pointer.
> 
> Signed-off-by: Jeff Layton <[email protected]>

Looks good. Feel free to add:

Reviewed-by: Jan Kara <[email protected]>

								Honza

> ---
>  drivers/base/devtmpfs.c  |  2 +-
>  fs/ecryptfs/inode.c      |  2 +-
>  fs/namei.c               | 22 +++++++++++++++++-----
>  fs/nfsd/nfs4recover.c    |  4 ++--
>  fs/nfsd/vfs.c            |  2 +-
>  fs/overlayfs/overlayfs.h |  2 +-
>  fs/smb/server/vfs.c      |  4 ++--
>  include/linux/fs.h       |  3 ++-
>  8 files changed, 27 insertions(+), 14 deletions(-)
> 
> diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c
> index 0e79621cb0f79870003b867ca384199171ded4e0..104025104ef75381984fd94dfbd50feeaa8cdd22 100644
> --- a/drivers/base/devtmpfs.c
> +++ b/drivers/base/devtmpfs.c
> @@ -261,7 +261,7 @@ static int dev_rmdir(const char *name)
>  		return PTR_ERR(dentry);
>  	if (d_inode(dentry)->i_private == &thread)
>  		err = vfs_rmdir(&nop_mnt_idmap, d_inode(parent.dentry),
> -				dentry);
> +				dentry, NULL);
>  	else
>  		err = -EPERM;
>  
> diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
> index 35830b3144f8f71374a78b3e7463b864f4fc216e..88631291b32535f623a3fbe4ea9b6ed48a306ca0 100644
> --- a/fs/ecryptfs/inode.c
> +++ b/fs/ecryptfs/inode.c
> @@ -540,7 +540,7 @@ static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry)
>  		if (d_unhashed(lower_dentry))
>  			rc = -EINVAL;
>  		else
> -			rc = vfs_rmdir(&nop_mnt_idmap, lower_dir, lower_dentry);
> +			rc = vfs_rmdir(&nop_mnt_idmap, lower_dir, lower_dentry, NULL);
>  	}
>  	if (!rc) {
>  		clear_nlink(d_inode(dentry));
> diff --git a/fs/namei.c b/fs/namei.c
> index 86cf6eca1f485361c6732974e4103cf5ea721539..4b5a99653c558397e592715d9d4663cd4a63ef86 100644
> --- a/fs/namei.c
> +++ b/fs/namei.c
> @@ -4522,9 +4522,10 @@ SYSCALL_DEFINE2(mkdir, const char __user *, pathname, umode_t, mode)
>  
>  /**
>   * vfs_rmdir - remove directory
> - * @idmap:	idmap of the mount the inode was found from
> - * @dir:	inode of the parent directory
> - * @dentry:	dentry of the child directory
> + * @idmap:		idmap of the mount the inode was found from
> + * @dir:		inode of the parent directory
> + * @dentry:		dentry of the child directory
> + * @delegated_inode:	returns parent inode, if it's delegated.
>   *
>   * Remove a directory.
>   *
> @@ -4535,7 +4536,7 @@ SYSCALL_DEFINE2(mkdir, const char __user *, pathname, umode_t, mode)
>   * raw inode simply pass @nop_mnt_idmap.
>   */
>  int vfs_rmdir(struct mnt_idmap *idmap, struct inode *dir,
> -		     struct dentry *dentry)
> +	      struct dentry *dentry, struct inode **delegated_inode)
>  {
>  	int error = may_delete(idmap, dir, dentry, 1);
>  
> @@ -4557,6 +4558,10 @@ int vfs_rmdir(struct mnt_idmap *idmap, struct inode *dir,
>  	if (error)
>  		goto out;
>  
> +	error = try_break_deleg(dir, delegated_inode);
> +	if (error)
> +		goto out;
> +
>  	error = dir->i_op->rmdir(dir, dentry);
>  	if (error)
>  		goto out;
> @@ -4583,6 +4588,7 @@ int do_rmdir(int dfd, struct filename *name)
>  	struct qstr last;
>  	int type;
>  	unsigned int lookup_flags = 0;
> +	struct inode *delegated_inode = NULL;
>  retry:
>  	error = filename_parentat(dfd, name, lookup_flags, &path, &last, &type);
>  	if (error)
> @@ -4612,7 +4618,8 @@ int do_rmdir(int dfd, struct filename *name)
>  	error = security_path_rmdir(&path, dentry);
>  	if (error)
>  		goto exit4;
> -	error = vfs_rmdir(mnt_idmap(path.mnt), path.dentry->d_inode, dentry);
> +	error = vfs_rmdir(mnt_idmap(path.mnt), path.dentry->d_inode,
> +			  dentry, &delegated_inode);
>  exit4:
>  	dput(dentry);
>  exit3:
> @@ -4620,6 +4627,11 @@ int do_rmdir(int dfd, struct filename *name)
>  	mnt_drop_write(path.mnt);
>  exit2:
>  	path_put(&path);
> +	if (delegated_inode) {
> +		error = break_deleg_wait(&delegated_inode);
> +		if (!error)
> +			goto retry;
> +	}
>  	if (retry_estale(error, lookup_flags)) {
>  		lookup_flags |= LOOKUP_REVAL;
>  		goto retry;
> diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
> index 3dfbb85c9a1166b56e56eb9f1d6bfd140584730b..ad3acbb956d90cac88f74e5f598719af6f1f8328 100644
> --- a/fs/nfsd/nfs4recover.c
> +++ b/fs/nfsd/nfs4recover.c
> @@ -275,7 +275,7 @@ nfsd4_unlink_clid_dir(char *name, struct nfsd_net *nn)
>  	status = -ENOENT;
>  	if (d_really_is_negative(dentry))
>  		goto out;
> -	status = vfs_rmdir(&nop_mnt_idmap, d_inode(dir), dentry);
> +	status = vfs_rmdir(&nop_mnt_idmap, d_inode(dir), dentry, NULL);
>  out:
>  	dput(dentry);
>  out_unlock:
> @@ -367,7 +367,7 @@ purge_old(struct dentry *parent, char *cname, struct nfsd_net *nn)
>  	inode_lock_nested(d_inode(parent), I_MUTEX_PARENT);
>  	child = lookup_one(&nop_mnt_idmap, &QSTR(cname), parent);
>  	if (!IS_ERR(child)) {
> -		status = vfs_rmdir(&nop_mnt_idmap, d_inode(parent), child);
> +		status = vfs_rmdir(&nop_mnt_idmap, d_inode(parent), child, NULL);
>  		if (status)
>  			printk("failed to remove client recovery directory %pd\n",
>  			       child);
> diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
> index 447f5ab8e0b92288c9f220060ab15f32f2a84de9..7d8cd2595f197be9741ee6320d43ed6651896647 100644
> --- a/fs/nfsd/vfs.c
> +++ b/fs/nfsd/vfs.c
> @@ -2194,7 +2194,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
>  				break;
>  		}
>  	} else {
> -		host_err = vfs_rmdir(&nop_mnt_idmap, dirp, rdentry);
> +		host_err = vfs_rmdir(&nop_mnt_idmap, dirp, rdentry, NULL);
>  	}
>  	fh_fill_post_attrs(fhp);
>  
> diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
> index 0f65f9a5d54d4786b39e4f4f30f416d5b9016e70..d215d7349489686b66bb66e939b27046f7d836f6 100644
> --- a/fs/overlayfs/overlayfs.h
> +++ b/fs/overlayfs/overlayfs.h
> @@ -206,7 +206,7 @@ static inline int ovl_do_notify_change(struct ovl_fs *ofs,
>  static inline int ovl_do_rmdir(struct ovl_fs *ofs,
>  			       struct inode *dir, struct dentry *dentry)
>  {
> -	int err = vfs_rmdir(ovl_upper_mnt_idmap(ofs), dir, dentry);
> +	int err = vfs_rmdir(ovl_upper_mnt_idmap(ofs), dir, dentry, NULL);
>  
>  	pr_debug("rmdir(%pd2) = %i\n", dentry, err);
>  	return err;
> diff --git a/fs/smb/server/vfs.c b/fs/smb/server/vfs.c
> index 3d2190f26623b23ea79c63410905a3c3ad684048..c5f0f3170d586cb2dc4d416b80948c642797fb82 100644
> --- a/fs/smb/server/vfs.c
> +++ b/fs/smb/server/vfs.c
> @@ -609,7 +609,7 @@ int ksmbd_vfs_remove_file(struct ksmbd_work *work, const struct path *path)
>  
>  	idmap = mnt_idmap(path->mnt);
>  	if (S_ISDIR(d_inode(path->dentry)->i_mode)) {
> -		err = vfs_rmdir(idmap, d_inode(parent), path->dentry);
> +		err = vfs_rmdir(idmap, d_inode(parent), path->dentry, NULL);
>  		if (err && err != -ENOTEMPTY)
>  			ksmbd_debug(VFS, "rmdir failed, err %d\n", err);
>  	} else {
> @@ -1090,7 +1090,7 @@ int ksmbd_vfs_unlink(struct file *filp)
>  	dget(dentry);
>  
>  	if (S_ISDIR(d_inode(dentry)->i_mode))
> -		err = vfs_rmdir(idmap, d_inode(dir), dentry);
> +		err = vfs_rmdir(idmap, d_inode(dir), dentry, NULL);
>  	else
>  		err = vfs_unlink(idmap, d_inode(dir), dentry, NULL);
>  
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 1040df3792794cd353b86558b41618294e25b8a6..d8bdaf7c87502ff17775602f5391d375738b4ed8 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -2120,7 +2120,8 @@ int vfs_symlink(struct mnt_idmap *, struct inode *,
>  		struct dentry *, const char *);
>  int vfs_link(struct dentry *, struct mnt_idmap *, struct inode *,
>  	     struct dentry *, struct inode **);
> -int vfs_rmdir(struct mnt_idmap *, struct inode *, struct dentry *);
> +int vfs_rmdir(struct mnt_idmap *, struct inode *, struct dentry *,
> +	      struct inode **);
>  int vfs_unlink(struct mnt_idmap *, struct inode *, struct dentry *,
>  	       struct inode **);
>  
> 
> -- 
> 2.51.0
> 
-- 
Jan Kara <[email protected]>
SUSE Labs, CR