[PATCH RFC 04/24] vfs: allow mkdir to wait for delegation break on parent
From: Jeff Layton <jlayton@kernel.org>
Date: 2024-03-15 16:53:19
Also in:
linux-cifs, linux-fsdevel, linux-nfs, linux-unionfs, lkml, netfs
Subsystem:
cachefiles: fs-cache backend for caching on mounted filesystems, driver core, kobjects, debugfs and sysfs, ecrypt file system, filesystems (vfs and infrastructure), kernel nfsd, sunrpc, and lockd servers, kernel smb3 server (ksmbd), overlay filesystem, the rest · Maintainers:
David Howells, Greg Kroah-Hartman, "Rafael J. Wysocki", Danilo Krummrich, Tyler Hicks, Alexander Viro, Christian Brauner, Chuck Lever, Jeff Layton, Namjae Jeon, Steve French, Miklos Szeredi, Amir Goldstein, Linus Torvalds
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 new delegated_inode parameter to vfs_mkdir. Most callers will set that to NULL, but do_mkdirat can use that to wait for the delegation break to complete and then retry. Signed-off-by: Jeff Layton <jlayton@kernel.org> --- drivers/base/devtmpfs.c | 2 +- fs/cachefiles/namei.c | 2 +- fs/ecryptfs/inode.c | 2 +- fs/init.c | 2 +- fs/namei.c | 17 +++++++++++++---- fs/nfsd/nfs4recover.c | 2 +- fs/nfsd/vfs.c | 2 +- fs/overlayfs/overlayfs.h | 2 +- fs/smb/server/vfs.c | 2 +- include/linux/fs.h | 2 +- 10 files changed, 22 insertions(+), 13 deletions(-)
diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c
index b848764ef018..8d1dbcad69f7 100644
--- a/drivers/base/devtmpfs.c
+++ b/drivers/base/devtmpfs.c@@ -166,7 +166,7 @@ static int dev_mkdir(const char *name, umode_t mode) if (IS_ERR(dentry)) return PTR_ERR(dentry); - err = vfs_mkdir(&nop_mnt_idmap, d_inode(path.dentry), dentry, mode); + err = vfs_mkdir(&nop_mnt_idmap, d_inode(path.dentry), dentry, mode, NULL); if (!err) /* mark as kernel-created inode */ d_inode(dentry)->i_private = &thread;
diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c
index 7ade836beb58..4e3e31320275 100644
--- a/fs/cachefiles/namei.c
+++ b/fs/cachefiles/namei.c@@ -130,7 +130,7 @@ struct dentry *cachefiles_get_directory(struct cachefiles_cache *cache, goto mkdir_error; ret = cachefiles_inject_write_error(); if (ret == 0) - ret = vfs_mkdir(&nop_mnt_idmap, d_inode(dir), subdir, 0700); + ret = vfs_mkdir(&nop_mnt_idmap, d_inode(dir), subdir, 0700, NULL); if (ret < 0) { trace_cachefiles_vfs_error(NULL, d_inode(dir), ret, cachefiles_trace_mkdir_error);
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 5ed1e4cf6c0b..d26b4484fa60 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c@@ -513,7 +513,7 @@ static int ecryptfs_mkdir(struct mnt_idmap *idmap, struct inode *dir, rc = lock_parent(dentry, &lower_dentry, &lower_dir); if (!rc) rc = vfs_mkdir(&nop_mnt_idmap, lower_dir, - lower_dentry, mode); + lower_dentry, mode, NULL); if (rc || d_really_is_negative(lower_dentry)) goto out; rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb);
diff --git a/fs/init.c b/fs/init.c
index e9387b6c4f30..325c9e4d9b20 100644
--- a/fs/init.c
+++ b/fs/init.c@@ -232,7 +232,7 @@ int __init init_mkdir(const char *pathname, umode_t mode) error = security_path_mkdir(&path, dentry, mode); if (!error) error = vfs_mkdir(mnt_idmap(path.mnt), path.dentry->d_inode, - dentry, mode); + dentry, mode, NULL); done_path_create(&path, dentry); return error; }
diff --git a/fs/namei.c b/fs/namei.c
index 56d3ebed7bac..6a22517f9938 100644
--- a/fs/namei.c
+++ b/fs/namei.c@@ -4103,7 +4103,7 @@ SYSCALL_DEFINE3(mknod, const char __user *, filename, umode_t, mode, unsigned, d * raw inode simply pass @nop_mnt_idmap. */ int vfs_mkdir(struct mnt_idmap *idmap, struct inode *dir, - struct dentry *dentry, umode_t mode) + struct dentry *dentry, umode_t mode, struct inode **delegated_inode) { int error; unsigned max_links = dir->i_sb->s_max_links;
@@ -4123,6 +4123,10 @@ int vfs_mkdir(struct mnt_idmap *idmap, struct inode *dir, if (max_links && dir->i_nlink >= max_links) return -EMLINK; + error = try_break_deleg(dir, delegated_inode); + if (error) + return error; + error = dir->i_op->mkdir(idmap, dir, dentry, mode); if (!error) fsnotify_mkdir(dir, dentry);
@@ -4136,6 +4140,7 @@ int do_mkdirat(int dfd, struct filename *name, umode_t mode) struct path path; int error; unsigned int lookup_flags = LOOKUP_DIRECTORY; + struct inode *delegated_inode = NULL; retry: dentry = filename_create(dfd, name, &path, lookup_flags);
@@ -4145,11 +4150,15 @@ int do_mkdirat(int dfd, struct filename *name, umode_t mode) error = security_path_mkdir(&path, dentry, mode_strip_umask(path.dentry->d_inode, mode)); - if (!error) { + if (!error) error = vfs_mkdir(mnt_idmap(path.mnt), path.dentry->d_inode, - dentry, mode); - } + dentry, mode, &delegated_inode); done_path_create(&path, dentry); + 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 2c060e0b1604..5bfced783a70 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c@@ -234,7 +234,7 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) * as well be forgiving and just succeed silently. */ goto out_put; - status = vfs_mkdir(&nop_mnt_idmap, d_inode(dir), dentry, S_IRWXU); + status = vfs_mkdir(&nop_mnt_idmap, d_inode(dir), dentry, S_IRWXU, NULL); out_put: dput(dentry); out_unlock:
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 6a4c506038e0..e42e58825590 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c@@ -1496,7 +1496,7 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp, nfsd_check_ignore_resizing(iap); break; case S_IFDIR: - host_err = vfs_mkdir(&nop_mnt_idmap, dirp, dchild, iap->ia_mode); + host_err = vfs_mkdir(&nop_mnt_idmap, dirp, dchild, iap->ia_mode, NULL); if (!host_err && unlikely(d_unhashed(dchild))) { struct dentry *d; d = lookup_one_len(dchild->d_name.name,
diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
index ee949f3e7c77..baa371947f86 100644
--- a/fs/overlayfs/overlayfs.h
+++ b/fs/overlayfs/overlayfs.h@@ -242,7 +242,7 @@ static inline int ovl_do_mkdir(struct ovl_fs *ofs, struct inode *dir, struct dentry *dentry, umode_t mode) { - int err = vfs_mkdir(ovl_upper_mnt_idmap(ofs), dir, dentry, mode); + int err = vfs_mkdir(ovl_upper_mnt_idmap(ofs), dir, dentry, mode, NULL); pr_debug("mkdir(%pd2, 0%o) = %i\n", dentry, mode, err); return err; }
diff --git a/fs/smb/server/vfs.c b/fs/smb/server/vfs.c
index c487e834331a..3760e0dda349 100644
--- a/fs/smb/server/vfs.c
+++ b/fs/smb/server/vfs.c@@ -227,7 +227,7 @@ int ksmbd_vfs_mkdir(struct ksmbd_work *work, const char *name, umode_t mode) idmap = mnt_idmap(path.mnt); mode |= S_IFDIR; - err = vfs_mkdir(idmap, d_inode(path.dentry), dentry, mode); + err = vfs_mkdir(idmap, d_inode(path.dentry), dentry, mode, NULL); if (!err && d_unhashed(dentry)) { struct dentry *d;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 8dfcddafdb3d..18eb7d628290 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h@@ -1839,7 +1839,7 @@ bool inode_owner_or_capable(struct mnt_idmap *idmap, int vfs_create(struct mnt_idmap *, struct inode *, struct dentry *, umode_t, bool); int vfs_mkdir(struct mnt_idmap *, struct inode *, - struct dentry *, umode_t); + struct dentry *, umode_t, struct inode **); int vfs_mknod(struct mnt_idmap *, struct inode *, struct dentry *, umode_t, dev_t); int vfs_symlink(struct mnt_idmap *, struct inode *,
--
2.44.0