#include <sys/zfs_vfsops.h>
#include <sys/zfs_vnops.h>
+#include <sys/zfs_znode.h>
#include <sys/vfs.h>
#include <sys/zpl.h>
static struct dentry *
+#ifdef HAVE_LOOKUP_NAMEIDATA
zpl_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
+#else
+zpl_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags)
+#endif
{
cred_t *cr = CRED();
struct inode *ip;
int error;
+ if (dlen(dentry) > ZFS_MAXNAMELEN)
+ return ERR_PTR(-ENAMETOOLONG);
+
crhold(cr);
error = -zfs_lookup(dir, dname(dentry), &ip, 0, cr, NULL, NULL);
ASSERT3S(error, <=, 0);
crfree(cr);
+ spin_lock(&dentry->d_lock);
+ dentry->d_time = jiffies;
+#ifndef HAVE_S_D_OP
+ d_set_d_op(dentry, &zpl_dentry_operations);
+#endif /* HAVE_S_D_OP */
+ spin_unlock(&dentry->d_lock);
+
if (error) {
if (error == -ENOENT)
return d_splice_alias(NULL, dentry);
return d_splice_alias(ip, dentry);
}
+void
+zpl_vap_init(vattr_t *vap, struct inode *dir, zpl_umode_t mode, cred_t *cr)
+{
+ vap->va_mask = ATTR_MODE;
+ vap->va_mode = mode;
+ vap->va_uid = crgetfsuid(cr);
+
+ if (dir && dir->i_mode & S_ISGID) {
+ vap->va_gid = dir->i_gid;
+ if (S_ISDIR(mode))
+ vap->va_mode |= S_ISGID;
+ } else {
+ vap->va_gid = crgetfsgid(cr);
+ }
+}
+
static int
-zpl_create(struct inode *dir, struct dentry *dentry, int mode,
+#ifdef HAVE_CREATE_NAMEIDATA
+zpl_create(struct inode *dir, struct dentry *dentry, zpl_umode_t mode,
struct nameidata *nd)
+#else
+zpl_create(struct inode *dir, struct dentry *dentry, zpl_umode_t mode,
+ bool flag)
+#endif
{
cred_t *cr = CRED();
struct inode *ip;
crhold(cr);
vap = kmem_zalloc(sizeof(vattr_t), KM_SLEEP);
- vap->va_mode = mode;
- vap->va_mask = ATTR_MODE;
- vap->va_uid = crgetfsuid(cr);
- vap->va_gid = crgetfsgid(cr);
+ zpl_vap_init(vap, dir, mode, cr);
- error = -zfs_create(dir, (char *)dentry->d_name.name,
- vap, 0, mode, &ip, cr, 0, NULL);
- if (error)
- goto out;
+ error = -zfs_create(dir, dname(dentry), vap, 0, mode, &ip, cr, 0, NULL);
+ if (error == 0) {
+ error = zpl_xattr_security_init(ip, dir, &dentry->d_name);
+ VERIFY3S(error, ==, 0);
+ d_instantiate(dentry, ip);
+ }
- d_instantiate(dentry, ip);
-out:
kmem_free(vap, sizeof(vattr_t));
crfree(cr);
ASSERT3S(error, <=, 0);
}
static int
-zpl_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
+zpl_mknod(struct inode *dir, struct dentry *dentry, zpl_umode_t mode,
+ dev_t rdev)
{
cred_t *cr = CRED();
struct inode *ip;
vattr_t *vap;
int error;
+ /*
+ * We currently expect Linux to supply rdev=0 for all sockets
+ * and fifos, but we want to know if this behavior ever changes.
+ */
+ if (S_ISSOCK(mode) || S_ISFIFO(mode))
+ ASSERT(rdev == 0);
+
crhold(cr);
vap = kmem_zalloc(sizeof(vattr_t), KM_SLEEP);
- vap->va_mode = mode;
- vap->va_mask = ATTR_MODE;
+ zpl_vap_init(vap, dir, mode, cr);
vap->va_rdev = rdev;
- vap->va_uid = crgetfsuid(cr);
- vap->va_gid = crgetfsgid(cr);
- error = -zfs_create(dir, (char *)dentry->d_name.name,
- vap, 0, mode, &ip, cr, 0, NULL);
- if (error)
- goto out;
+ error = -zfs_create(dir, dname(dentry), vap, 0, mode, &ip, cr, 0, NULL);
+ if (error == 0)
+ d_instantiate(dentry, ip);
- d_instantiate(dentry, ip);
-out:
kmem_free(vap, sizeof(vattr_t));
crfree(cr);
ASSERT3S(error, <=, 0);
}
static int
-zpl_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+zpl_mkdir(struct inode *dir, struct dentry *dentry, zpl_umode_t mode)
{
cred_t *cr = CRED();
vattr_t *vap;
crhold(cr);
vap = kmem_zalloc(sizeof(vattr_t), KM_SLEEP);
- vap->va_mode = S_IFDIR | mode;
- vap->va_mask = ATTR_MODE;
- vap->va_uid = crgetfsuid(cr);
- vap->va_gid = crgetfsgid(cr);
+ zpl_vap_init(vap, dir, mode | S_IFDIR, cr);
error = -zfs_mkdir(dir, dname(dentry), vap, &ip, cr, 0, NULL);
- if (error)
- goto out;
+ if (error == 0)
+ d_instantiate(dentry, ip);
- d_instantiate(dentry, ip);
-out:
kmem_free(vap, sizeof(vattr_t));
crfree(cr);
ASSERT3S(error, <=, 0);
static int
zpl_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
{
- cred_t *cr = CRED();
- vattr_t *vap;
- struct inode *ip;
+ boolean_t issnap = ITOZSB(dentry->d_inode)->z_issnap;
int error;
- ip = dentry->d_inode;
- crhold(cr);
- vap = kmem_zalloc(sizeof(vattr_t), KM_SLEEP);
-
- error = -zfs_getattr(ip, vap, 0, cr);
- if (error)
- goto out;
-
- stat->ino = ip->i_ino;
- stat->dev = ip->i_sb->s_dev;
- stat->mode = vap->va_mode;
- stat->nlink = vap->va_nlink;
- stat->uid = vap->va_uid;
- stat->gid = vap->va_gid;
- stat->rdev = vap->va_rdev;
- stat->size = vap->va_size;
- stat->atime = vap->va_atime;
- stat->mtime = vap->va_mtime;
- stat->ctime = vap->va_ctime;
- stat->blksize = vap->va_blksize;
- stat->blocks = vap->va_nblocks;
-out:
- kmem_free(vap, sizeof(vattr_t));
- crfree(cr);
+ /*
+ * Ensure MNT_SHRINKABLE is set on snapshots to ensure they are
+ * unmounted automatically with the parent file system. This
+ * is done on the first getattr because it's not easy to get the
+ * vfsmount structure at mount time. This call path is explicitly
+ * marked unlikely to avoid any performance impact. FWIW, ext4
+ * resorts to a similar trick for sysadmin convenience.
+ */
+ if (unlikely(issnap && !(mnt->mnt_flags & MNT_SHRINKABLE)))
+ mnt->mnt_flags |= MNT_SHRINKABLE;
+
+ error = -zfs_getattr_fast(dentry->d_inode, stat);
ASSERT3S(error, <=, 0);
return (error);
crhold(cr);
vap = kmem_zalloc(sizeof(vattr_t), KM_SLEEP);
- vap->va_mode = S_IFLNK | S_IRWXUGO;
- vap->va_mask = ATTR_MODE;
- vap->va_uid = crgetfsuid(cr);
- vap->va_gid = crgetfsgid(cr);
+ zpl_vap_init(vap, dir, S_IFLNK | S_IRWXUGO, cr);
error = -zfs_symlink(dir, dname(dentry), vap, (char *)name, &ip, cr, 0);
- if (error)
- goto out;
+ if (error == 0)
+ d_instantiate(dentry, ip);
- d_instantiate(dentry, ip);
-out:
kmem_free(vap, sizeof(vattr_t));
crfree(cr);
ASSERT3S(error, <=, 0);
static void
zpl_put_link(struct dentry *dentry, struct nameidata *nd, void *ptr)
{
- char *link;
+ const char *link = nd_get_link(nd);
- link = nd_get_link(nd);
if (!IS_ERR(link))
kmem_free(link, MAXPATHLEN);
}
return (error);
}
+#ifdef HAVE_INODE_TRUNCATE_RANGE
+static void
+zpl_truncate_range(struct inode* ip, loff_t start, loff_t end)
+{
+ cred_t *cr = CRED();
+ flock64_t bf;
+
+ ASSERT3S(start, <=, end);
+
+ /*
+ * zfs_freesp() will interpret (len == 0) as meaning "truncate until
+ * the end of the file". We don't want that.
+ */
+ if (start == end)
+ return;
+
+ crhold(cr);
+
+ bf.l_type = F_WRLCK;
+ bf.l_whence = 0;
+ bf.l_start = start;
+ bf.l_len = end - start;
+ bf.l_pid = 0;
+ zfs_space(ip, F_FREESP, &bf, FWRITE, start, cr);
+
+ crfree(cr);
+}
+#endif /* HAVE_INODE_TRUNCATE_RANGE */
+
+#ifdef HAVE_INODE_FALLOCATE
+static long
+zpl_fallocate(struct inode *ip, int mode, loff_t offset, loff_t len)
+{
+ return zpl_fallocate_common(ip, mode, offset, len);
+}
+#endif /* HAVE_INODE_FALLOCATE */
+
+static int
+#ifdef HAVE_D_REVALIDATE_NAMEIDATA
+zpl_revalidate(struct dentry *dentry, struct nameidata *nd)
+{
+ unsigned int flags = (nd ? nd->flags : 0);
+#else
+zpl_revalidate(struct dentry *dentry, unsigned int flags)
+{
+#endif /* HAVE_D_REVALIDATE_NAMEIDATA */
+ zfs_sb_t *zsb = dentry->d_sb->s_fs_info;
+ int error;
+
+ if (flags & LOOKUP_RCU)
+ return (-ECHILD);
+
+ /*
+ * After a rollback negative dentries created before the rollback
+ * time must be invalidated. Otherwise they can obscure files which
+ * are only present in the rolled back dataset.
+ */
+ if (dentry->d_inode == NULL) {
+ spin_lock(&dentry->d_lock);
+ error = time_before(dentry->d_time, zsb->z_rollback_time);
+ spin_unlock(&dentry->d_lock);
+
+ if (error)
+ return (0);
+ }
+
+ /*
+ * The dentry may reference a stale inode if a mounted file system
+ * was rolled back to a point in time where the object didn't exist.
+ */
+ if (dentry->d_inode && ITOZ(dentry->d_inode)->z_is_stale)
+ return (0);
+
+ return (1);
+}
+
const struct inode_operations zpl_inode_operations = {
.create = zpl_create,
.link = zpl_link,
.getxattr = generic_getxattr,
.removexattr = generic_removexattr,
.listxattr = zpl_xattr_list,
+#ifdef HAVE_INODE_TRUNCATE_RANGE
+ .truncate_range = zpl_truncate_range,
+#endif /* HAVE_INODE_TRUNCATE_RANGE */
+#ifdef HAVE_INODE_FALLOCATE
+ .fallocate = zpl_fallocate,
+#endif /* HAVE_INODE_FALLOCATE */
};
const struct inode_operations zpl_dir_inode_operations = {
.readlink = generic_readlink,
.follow_link = zpl_follow_link,
.put_link = zpl_put_link,
+ .setattr = zpl_setattr,
+ .getattr = zpl_getattr,
+ .setxattr = generic_setxattr,
+ .getxattr = generic_getxattr,
+ .removexattr = generic_removexattr,
+ .listxattr = zpl_xattr_list,
};
const struct inode_operations zpl_special_inode_operations = {
.removexattr = generic_removexattr,
.listxattr = zpl_xattr_list,
};
+
+dentry_operations_t zpl_dentry_operations = {
+ .d_revalidate = zpl_revalidate,
+};