summaryrefslogtreecommitdiffstats
path: root/fs/overlayfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/overlayfs')
-rw-r--r--fs/overlayfs/copy_up.c2
-rw-r--r--fs/overlayfs/dir.c35
-rw-r--r--fs/overlayfs/inode.c46
-rw-r--r--fs/overlayfs/namei.c53
-rw-r--r--fs/overlayfs/overlayfs.h18
-rw-r--r--fs/overlayfs/readdir.c392
-rw-r--r--fs/overlayfs/super.c24
-rw-r--r--fs/overlayfs/util.c31
8 files changed, 493 insertions, 108 deletions
diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c
index acb6f97..aad97b3 100644
--- a/fs/overlayfs/copy_up.c
+++ b/fs/overlayfs/copy_up.c
@@ -241,7 +241,7 @@ struct ovl_fh *ovl_encode_fh(struct dentry *lower, bool is_upper)
int buflen = MAX_HANDLE_SZ;
uuid_t *uuid = &lower->d_sb->s_uuid;
- buf = kmalloc(buflen, GFP_TEMPORARY);
+ buf = kmalloc(buflen, GFP_KERNEL);
if (!buf)
return ERR_PTR(-ENOMEM);
diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c
index 641d9ee..3309b19 100644
--- a/fs/overlayfs/dir.c
+++ b/fs/overlayfs/dir.c
@@ -155,7 +155,7 @@ static int ovl_set_opaque(struct dentry *dentry, struct dentry *upperdentry)
static void ovl_instantiate(struct dentry *dentry, struct inode *inode,
struct dentry *newdentry, bool hardlink)
{
- ovl_dentry_version_inc(dentry->d_parent);
+ ovl_dentry_version_inc(dentry->d_parent, false);
ovl_dentry_set_upper_alias(dentry);
if (!hardlink) {
ovl_inode_update(inode, newdentry);
@@ -481,17 +481,30 @@ out_cleanup:
}
static int ovl_create_or_link(struct dentry *dentry, struct inode *inode,
- struct cattr *attr, struct dentry *hardlink)
+ struct cattr *attr, struct dentry *hardlink,
+ bool origin)
{
int err;
const struct cred *old_cred;
struct cred *override_cred;
+ struct dentry *parent = dentry->d_parent;
- err = ovl_copy_up(dentry->d_parent);
+ err = ovl_copy_up(parent);
if (err)
return err;
old_cred = ovl_override_creds(dentry->d_sb);
+
+ /*
+ * When linking a file with copy up origin into a new parent, mark the
+ * new parent dir "impure".
+ */
+ if (origin) {
+ err = ovl_set_impure(parent, ovl_dentry_upper(parent));
+ if (err)
+ goto out_revert_creds;
+ }
+
err = -ENOMEM;
override_cred = prepare_creds();
if (override_cred) {
@@ -550,7 +563,7 @@ static int ovl_create_object(struct dentry *dentry, int mode, dev_t rdev,
inode_init_owner(inode, dentry->d_parent->d_inode, mode);
attr.mode = inode->i_mode;
- err = ovl_create_or_link(dentry, inode, &attr, NULL);
+ err = ovl_create_or_link(dentry, inode, &attr, NULL, false);
if (err)
iput(inode);
@@ -609,7 +622,8 @@ static int ovl_link(struct dentry *old, struct inode *newdir,
inode = d_inode(old);
ihold(inode);
- err = ovl_create_or_link(new, inode, NULL, ovl_dentry_upper(old));
+ err = ovl_create_or_link(new, inode, NULL, ovl_dentry_upper(old),
+ ovl_type_origin(old));
if (err)
iput(inode);
@@ -678,7 +692,7 @@ static int ovl_remove_and_whiteout(struct dentry *dentry, bool is_dir)
if (flags)
ovl_cleanup(wdir, upper);
- ovl_dentry_version_inc(dentry->d_parent);
+ ovl_dentry_version_inc(dentry->d_parent, true);
out_d_drop:
d_drop(dentry);
dput(whiteout);
@@ -728,7 +742,7 @@ static int ovl_remove_upper(struct dentry *dentry, bool is_dir)
err = vfs_rmdir(dir, upper);
else
err = vfs_unlink(dir, upper, NULL);
- ovl_dentry_version_inc(dentry->d_parent);
+ ovl_dentry_version_inc(dentry->d_parent, ovl_type_origin(dentry));
/*
* Keeping this dentry hashed would mean having to release
@@ -819,7 +833,7 @@ static char *ovl_get_redirect(struct dentry *dentry, bool samedir)
goto out;
}
- buf = ret = kmalloc(buflen, GFP_TEMPORARY);
+ buf = ret = kmalloc(buflen, GFP_KERNEL);
if (!buf)
goto out;
@@ -1075,8 +1089,9 @@ static int ovl_rename(struct inode *olddir, struct dentry *old,
drop_nlink(d_inode(new));
}
- ovl_dentry_version_inc(old->d_parent);
- ovl_dentry_version_inc(new->d_parent);
+ ovl_dentry_version_inc(old->d_parent,
+ !overwrite && ovl_type_origin(new));
+ ovl_dentry_version_inc(new->d_parent, ovl_type_origin(old));
out_dput:
dput(newdentry);
diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c
index 69f4fc2..a619add 100644
--- a/fs/overlayfs/inode.c
+++ b/fs/overlayfs/inode.c
@@ -202,37 +202,38 @@ bool ovl_is_private_xattr(const char *name)
sizeof(OVL_XATTR_PREFIX) - 1) == 0;
}
-int ovl_xattr_set(struct dentry *dentry, const char *name, const void *value,
- size_t size, int flags)
+int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
{
int err;
- struct path realpath;
- enum ovl_path_type type = ovl_path_real(dentry, &realpath);
+ struct dentry *upperdentry = ovl_i_dentry_upper(inode);
+ struct dentry *realdentry = upperdentry ?: ovl_dentry_lower(dentry);
const struct cred *old_cred;
err = ovl_want_write(dentry);
if (err)
goto out;
- if (!value && !OVL_TYPE_UPPER(type)) {
- err = vfs_getxattr(realpath.dentry, name, NULL, 0);
+ if (!value && !upperdentry) {
+ err = vfs_getxattr(realdentry, name, NULL, 0);
if (err < 0)
goto out_drop_write;
}
- err = ovl_copy_up(dentry);
- if (err)
- goto out_drop_write;
+ if (!upperdentry) {
+ err = ovl_copy_up(dentry);
+ if (err)
+ goto out_drop_write;
- if (!OVL_TYPE_UPPER(type))
- ovl_path_upper(dentry, &realpath);
+ realdentry = ovl_dentry_upper(dentry);
+ }
old_cred = ovl_override_creds(dentry->d_sb);
if (value)
- err = vfs_setxattr(realpath.dentry, name, value, size, flags);
+ err = vfs_setxattr(realdentry, name, value, size, flags);
else {
WARN_ON(flags != XATTR_REPLACE);
- err = vfs_removexattr(realpath.dentry, name);
+ err = vfs_removexattr(realdentry, name);
}
revert_creds(old_cred);
@@ -242,12 +243,13 @@ out:
return err;
}
-int ovl_xattr_get(struct dentry *dentry, const char *name,
+int ovl_xattr_get(struct dentry *dentry, struct inode *inode, const char *name,
void *value, size_t size)
{
- struct dentry *realdentry = ovl_dentry_real(dentry);
ssize_t res;
const struct cred *old_cred;
+ struct dentry *realdentry =
+ ovl_i_dentry_upper(inode) ?: ovl_dentry_lower(dentry);
old_cred = ovl_override_creds(dentry->d_sb);
res = vfs_getxattr(realdentry, name, value, size);
@@ -496,6 +498,9 @@ static int ovl_set_nlink_common(struct dentry *dentry,
len = snprintf(buf, sizeof(buf), format,
(int) (inode->i_nlink - realinode->i_nlink));
+ if (WARN_ON(len >= sizeof(buf)))
+ return -EIO;
+
return ovl_do_setxattr(ovl_dentry_upper(dentry),
OVL_XATTR_NLINK, buf, len, 0);
}
@@ -574,10 +579,13 @@ static int ovl_inode_set(struct inode *inode, void *data)
static bool ovl_verify_inode(struct inode *inode, struct dentry *lowerdentry,
struct dentry *upperdentry)
{
- struct inode *lowerinode = lowerdentry ? d_inode(lowerdentry) : NULL;
-
- /* Lower (origin) inode must match, even if NULL */
- if (ovl_inode_lower(inode) != lowerinode)
+ /*
+ * Allow non-NULL lower inode in ovl_inode even if lowerdentry is NULL.
+ * This happens when finding a copied up overlay inode for a renamed
+ * or hardlinked overlay dentry and lower dentry cannot be followed
+ * by origin because lower fs does not support file handles.
+ */
+ if (lowerdentry && ovl_inode_lower(inode) != d_inode(lowerdentry))
return false;
/*
diff --git a/fs/overlayfs/namei.c b/fs/overlayfs/namei.c
index 9bc0e58..c3addd1 100644
--- a/fs/overlayfs/namei.c
+++ b/fs/overlayfs/namei.c
@@ -38,7 +38,7 @@ static int ovl_check_redirect(struct dentry *dentry, struct ovl_lookup_data *d,
return 0;
goto fail;
}
- buf = kzalloc(prelen + res + strlen(post) + 1, GFP_TEMPORARY);
+ buf = kzalloc(prelen + res + strlen(post) + 1, GFP_KERNEL);
if (!buf)
return -ENOMEM;
@@ -103,7 +103,7 @@ static struct ovl_fh *ovl_get_origin_fh(struct dentry *dentry)
if (res == 0)
return NULL;
- fh = kzalloc(res, GFP_TEMPORARY);
+ fh = kzalloc(res, GFP_KERNEL);
if (!fh)
return ERR_PTR(-ENOMEM);
@@ -309,7 +309,7 @@ static int ovl_check_origin(struct dentry *upperdentry,
BUG_ON(*ctrp);
if (!*stackp)
- *stackp = kmalloc(sizeof(struct path), GFP_TEMPORARY);
+ *stackp = kmalloc(sizeof(struct path), GFP_KERNEL);
if (!*stackp) {
dput(origin);
return -ENOMEM;
@@ -397,8 +397,19 @@ int ovl_verify_index(struct dentry *index, struct path *lowerstack,
if (!d_inode(index))
return 0;
- err = -EISDIR;
- if (d_is_dir(index))
+ /*
+ * Directory index entries are going to be used for looking up
+ * redirected upper dirs by lower dir fh when decoding an overlay
+ * file handle of a merge dir. Whiteout index entries are going to be
+ * used as an indication that an exported overlay file handle should
+ * be treated as stale (i.e. after unlink of the overlay inode).
+ * We don't know the verification rules for directory and whiteout
+ * index entries, because they have not been implemented yet, so return
+ * EROFS if those entries are found to avoid corrupting an index that
+ * was created by a newer kernel.
+ */
+ err = -EROFS;
+ if (d_is_dir(index) || ovl_is_whiteout(index))
goto fail;
err = -EINVAL;
@@ -407,7 +418,7 @@ int ovl_verify_index(struct dentry *index, struct path *lowerstack,
err = -ENOMEM;
len = index->d_name.len / 2;
- fh = kzalloc(len, GFP_TEMPORARY);
+ fh = kzalloc(len, GFP_KERNEL);
if (!fh)
goto fail;
@@ -436,8 +447,8 @@ out:
return err;
fail:
- pr_warn_ratelimited("overlayfs: failed to verify index (%pd2, err=%i)\n",
- index, err);
+ pr_warn_ratelimited("overlayfs: failed to verify index (%pd2, ftype=%x, err=%i)\n",
+ index, d_inode(index)->i_mode & S_IFMT, err);
goto out;
}
@@ -467,7 +478,7 @@ int ovl_get_index_name(struct dentry *origin, struct qstr *name)
return PTR_ERR(fh);
err = -ENOMEM;
- n = kzalloc(fh->len * 2, GFP_TEMPORARY);
+ n = kzalloc(fh->len * 2, GFP_KERNEL);
if (n) {
s = bin2hex(n, fh, fh->len);
*name = (struct qstr) QSTR_INIT(n, s - n);
@@ -502,6 +513,7 @@ static struct dentry *ovl_lookup_index(struct dentry *dentry,
goto out;
}
+ inode = d_inode(index);
if (d_is_negative(index)) {
if (upper && d_inode(origin)->i_nlink > 1) {
pr_warn_ratelimited("overlayfs: hard link with origin but no index (ino=%lu).\n",
@@ -511,11 +523,22 @@ static struct dentry *ovl_lookup_index(struct dentry *dentry,
dput(index);
index = NULL;
- } else if (upper && d_inode(index) != d_inode(upper)) {
- inode = d_inode(index);
- pr_warn_ratelimited("overlayfs: wrong index found (index ino: %lu, upper ino: %lu).\n",
- d_inode(index)->i_ino,
- d_inode(upper)->i_ino);
+ } else if (upper && d_inode(upper) != inode) {
+ pr_warn_ratelimited("overlayfs: wrong index found (index=%pd2, ino=%lu, upper ino=%lu).\n",
+ index, inode->i_ino, d_inode(upper)->i_ino);
+ goto fail;
+ } else if (ovl_dentry_weird(index) || ovl_is_whiteout(index) ||
+ ((inode->i_mode ^ d_inode(origin)->i_mode) & S_IFMT)) {
+ /*
+ * Index should always be of the same file type as origin
+ * except for the case of a whiteout index. A whiteout
+ * index should only exist if all lower aliases have been
+ * unlinked, which means that finding a lower origin on lookup
+ * whose index is a whiteout should be treated as an error.
+ */
+ pr_warn_ratelimited("overlayfs: bad index found (index=%pd2, ftype=%x, origin ftype=%x).\n",
+ index, d_inode(index)->i_mode & S_IFMT,
+ d_inode(origin)->i_mode & S_IFMT);
goto fail;
}
@@ -623,7 +646,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
if (!d.stop && poe->numlower) {
err = -ENOMEM;
stack = kcalloc(ofs->numlower, sizeof(struct path),
- GFP_TEMPORARY);
+ GFP_KERNEL);
if (!stack)
goto out_put_upper;
}
diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
index 60d2660..d4e8c1a 100644
--- a/fs/overlayfs/overlayfs.h
+++ b/fs/overlayfs/overlayfs.h
@@ -47,7 +47,8 @@ enum ovl_flag {
/* Is the real inode encoded in fid an upper inode? */
#define OVL_FH_FLAG_PATH_UPPER (1 << 2)
-#define OVL_FH_FLAG_ALL (OVL_FH_FLAG_BIG_ENDIAN | OVL_FH_FLAG_ANY_ENDIAN)
+#define OVL_FH_FLAG_ALL (OVL_FH_FLAG_BIG_ENDIAN | OVL_FH_FLAG_ANY_ENDIAN | \
+ OVL_FH_FLAG_PATH_UPPER)
#if defined(__LITTLE_ENDIAN)
#define OVL_FH_FLAG_CPU_ENDIAN 0
@@ -199,11 +200,12 @@ enum ovl_path_type ovl_path_real(struct dentry *dentry, struct path *path);
struct dentry *ovl_dentry_upper(struct dentry *dentry);
struct dentry *ovl_dentry_lower(struct dentry *dentry);
struct dentry *ovl_dentry_real(struct dentry *dentry);
+struct dentry *ovl_i_dentry_upper(struct inode *inode);
struct inode *ovl_inode_upper(struct inode *inode);
struct inode *ovl_inode_lower(struct inode *inode);
struct inode *ovl_inode_real(struct inode *inode);
-struct ovl_dir_cache *ovl_dir_cache(struct dentry *dentry);
-void ovl_set_dir_cache(struct dentry *dentry, struct ovl_dir_cache *cache);
+struct ovl_dir_cache *ovl_dir_cache(struct inode *inode);
+void ovl_set_dir_cache(struct inode *inode, struct ovl_dir_cache *cache);
bool ovl_dentry_is_opaque(struct dentry *dentry);
bool ovl_dentry_is_whiteout(struct dentry *dentry);
void ovl_dentry_set_opaque(struct dentry *dentry);
@@ -215,7 +217,7 @@ void ovl_dentry_set_redirect(struct dentry *dentry, const char *redirect);
void ovl_inode_init(struct inode *inode, struct dentry *upperdentry,
struct dentry *lowerdentry);
void ovl_inode_update(struct inode *inode, struct dentry *upperdentry);
-void ovl_dentry_version_inc(struct dentry *dentry);
+void ovl_dentry_version_inc(struct dentry *dentry, bool impurity);
u64 ovl_dentry_version_get(struct dentry *dentry);
bool ovl_is_whiteout(struct dentry *dentry);
struct file *ovl_path_open(struct path *path, int flags);
@@ -227,6 +229,7 @@ int ovl_check_setxattr(struct dentry *dentry, struct dentry *upperdentry,
int xerr);
int ovl_set_impure(struct dentry *dentry, struct dentry *upperdentry);
void ovl_set_flag(unsigned long flag, struct inode *inode);
+void ovl_clear_flag(unsigned long flag, struct inode *inode);
bool ovl_test_flag(unsigned long flag, struct inode *inode);
bool ovl_inuse_trylock(struct dentry *dentry);
void ovl_inuse_unlock(struct dentry *dentry);
@@ -254,6 +257,7 @@ extern const struct file_operations ovl_dir_operations;
int ovl_check_empty_dir(struct dentry *dentry, struct list_head *list);
void ovl_cleanup_whiteouts(struct dentry *upper, struct list_head *list);
void ovl_cache_free(struct list_head *list);
+void ovl_dir_cache_free(struct inode *inode);
int ovl_check_d_type_supported(struct path *realpath);
void ovl_workdir_cleanup(struct inode *dir, struct vfsmount *mnt,
struct dentry *dentry, int level);
@@ -270,9 +274,9 @@ int ovl_setattr(struct dentry *dentry, struct iattr *attr);
int ovl_getattr(const struct path *path, struct kstat *stat,
u32 request_mask, unsigned int flags);
int ovl_permission(struct inode *inode, int mask);
-int ovl_xattr_set(struct dentry *dentry, const char *name, const void *value,
- size_t size, int flags);
-int ovl_xattr_get(struct dentry *dentry, const char *name,
+int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name,
+ const void *value, size_t size, int flags);
+int ovl_xattr_get(struct dentry *dentry, struct inode *inode, const char *name,
void *value, size_t size);
ssize_t ovl_listxattr(struct dentry *dentry, char *list, size_t size);
struct posix_acl *ovl_get_acl(struct inode *inode, int type);
diff --git a/fs/overlayfs/readdir.c b/fs/overlayfs/readdir.c
index 0298463..62e9b22 100644
--- a/fs/overlayfs/readdir.c
+++ b/fs/overlayfs/readdir.c
@@ -15,11 +15,13 @@
#include <linux/rbtree.h>
#include <linux/security.h>
#include <linux/cred.h>
+#include <linux/ratelimit.h>
#include "overlayfs.h"
struct ovl_cache_entry {
unsigned int len;
unsigned int type;
+ u64 real_ino;
u64 ino;
struct list_head l_node;
struct rb_node node;
@@ -32,18 +34,20 @@ struct ovl_dir_cache {
long refcount;
u64 version;
struct list_head entries;
+ struct rb_root root;
};
struct ovl_readdir_data {
struct dir_context ctx;
struct dentry *dentry;
bool is_lowest;
- struct rb_root root;
+ struct rb_root *root;
struct list_head *list;
struct list_head middle;
struct ovl_cache_entry *first_maybe_whiteout;
int count;
int err;
+ bool is_upper;
bool d_type_supported;
};
@@ -58,7 +62,33 @@ struct ovl_dir_file {
static struct ovl_cache_entry *ovl_cache_entry_from_node(struct rb_node *n)
{
- return container_of(n, struct ovl_cache_entry, node);
+ return rb_entry(n, struct ovl_cache_entry, node);
+}
+
+static bool ovl_cache_entry_find_link(const char *name, int len,
+ struct rb_node ***link,
+ struct rb_node **parent)
+{
+ bool found = false;
+ struct rb_node **newp = *link;
+
+ while (!found && *newp) {
+ int cmp;
+ struct ovl_cache_entry *tmp;
+
+ *parent = *newp;
+ tmp = ovl_cache_entry_from_node(*newp);
+ cmp = strncmp(name, tmp->name, len);
+ if (cmp > 0)
+ newp = &tmp->node.rb_right;
+ else if (cmp < 0 || len < tmp->len)
+ newp = &tmp->node.rb_left;
+ else
+ found = true;
+ }
+ *link = newp;
+
+ return found;
}
static struct ovl_cache_entry *ovl_cache_entry_find(struct rb_root *root,
@@ -82,6 +112,32 @@ static struct ovl_cache_entry *ovl_cache_entry_find(struct rb_root *root,
return NULL;
}
+static bool ovl_calc_d_ino(struct ovl_readdir_data *rdd,
+ struct ovl_cache_entry *p)
+{
+ /* Don't care if not doing ovl_iter() */
+ if (!rdd->dentry)
+ return false;
+
+ /* Always recalc d_ino for parent */
+ if (strcmp(p->name, "..") == 0)
+ return true;
+
+ /* If this is lower, then native d_ino will do */
+ if (!rdd->is_upper)
+ return false;
+
+ /*
+ * Recalc d_ino for '.' and for all entries if dir is impure (contains
+ * copied up entries)
+ */
+ if ((p->name[0] == '.' && p->len == 1) ||
+ ovl_test_flag(OVL_IMPURE, d_inode(rdd->dentry)))
+ return true;
+
+ return false;
+}
+
static struct ovl_cache_entry *ovl_cache_entry_new(struct ovl_readdir_data *rdd,
const char *name, int len,
u64 ino, unsigned int d_type)
@@ -97,7 +153,11 @@ static struct ovl_cache_entry *ovl_cache_entry_new(struct ovl_readdir_data *rdd,
p->name[len] = '\0';
p->len = len;
p->type = d_type;
+ p->real_ino = ino;
p->ino = ino;
+ /* Defer setting d_ino for upper entry to ovl_iterate() */
+ if (ovl_calc_d_ino(rdd, p))
+ p->ino = 0;
p->is_whiteout = false;
if (d_type == DT_CHR) {
@@ -111,32 +171,22 @@ static int ovl_cache_entry_add_rb(struct ovl_readdir_data *rdd,
const char *name, int len, u64 ino,
unsigned int d_type)
{
- struct rb_node **newp = &rdd->root.rb_node;
+ struct rb_node **newp = &rdd->root->rb_node;
struct rb_node *parent = NULL;
struct ovl_cache_entry *p;
- while (*newp) {
- int cmp;
- struct ovl_cache_entry *tmp;
-
- parent = *newp;
- tmp = ovl_cache_entry_from_node(*newp);
- cmp = strncmp(name, tmp->name, len);
- if (cmp > 0)
- newp = &tmp->node.rb_right;
- else if (cmp < 0 || len < tmp->len)
- newp = &tmp->node.rb_left;
- else
- return 0;
- }
+ if (ovl_cache_entry_find_link(name, len, &newp, &parent))
+ return 0;
p = ovl_cache_entry_new(rdd, name, len, ino, d_type);
- if (p == NULL)
+ if (p == NULL) {
+ rdd->err = -ENOMEM;
return -ENOMEM;
+ }
list_add_tail(&p->l_node, rdd->list);
rb_link_node(&p->node, parent, newp);
- rb_insert_color(&p->node, &rdd->root);
+ rb_insert_color(&p->node, rdd->root);
return 0;
}
@@ -147,7 +197,7 @@ static int ovl_fill_lowest(struct ovl_readdir_data *rdd,
{
struct ovl_cache_entry *p;
- p = ovl_cache_entry_find(&rdd->root, name, namelen);
+ p = ovl_cache_entry_find(rdd->root, name, namelen);
if (p) {
list_move_tail(&p->l_node, &rdd->middle);
} else {
@@ -172,6 +222,16 @@ void ovl_cache_free(struct list_head *list)
INIT_LIST_HEAD(list);
}
+void ovl_dir_cache_free(struct inode *inode)
+{
+ struct ovl_dir_cache *cache = ovl_dir_cache(inode);
+
+ if (cache) {
+ ovl_cache_free(&cache->entries);
+ kfree(cache);
+ }
+}
+
static void ovl_cache_put(struct ovl_dir_file *od, struct dentry *dentry)
{
struct ovl_dir_cache *cache = od->cache;
@@ -179,8 +239,8 @@ static void ovl_cache_put(struct ovl_dir_file *od, struct dentry *dentry)
WARN_ON(cache->refcount <= 0);
cache->refcount--;
if (!cache->refcount) {
- if (ovl_dir_cache(dentry) == cache)
- ovl_set_dir_cache(dentry, NULL);
+ if (ovl_dir_cache(d_inode(dentry)) == cache)
+ ovl_set_dir_cache(d_inode(dentry), NULL);
ovl_cache_free(&cache->entries);
kfree(cache);
@@ -273,7 +333,8 @@ static void ovl_dir_reset(struct file *file)
od->is_real = false;
}
-static int ovl_dir_read_merged(struct dentry *dentry, struct list_head *list)
+static int ovl_dir_read_merged(struct dentry *dentry, struct list_head *list,
+ struct rb_root *root)
{
int err;
struct path realpath;
@@ -281,13 +342,14 @@ static int ovl_dir_read_merged(struct dentry *dentry, struct list_head *list)
.ctx.actor = ovl_fill_merge,
.dentry = dentry,
.list = list,
- .root = RB_ROOT,
+ .root = root,
.is_lowest = false,
};
int idx, next;
for (idx = 0; idx != -1; idx = next) {
next = ovl_path_next(idx, dentry, &realpath);
+ rdd.is_upper = ovl_dentry_upper(dentry) == realpath.dentry;
if (next != -1) {
err = ovl_dir_read(&realpath, &rdd);
@@ -326,12 +388,13 @@ static struct ovl_dir_cache *ovl_cache_get(struct dentry *dentry)
int res;
struct ovl_dir_cache *cache;
- cache = ovl_dir_cache(dentry);
+ cache = ovl_dir_cache(d_inode(dentry));
if (cache && ovl_dentry_version_get(dentry) == cache->version) {
+ WARN_ON(!cache->refcount);
cache->refcount++;
return cache;
}
- ovl_set_dir_cache(dentry, NULL);
+ ovl_set_dir_cache(d_inode(dentry), NULL);
cache = kzalloc(sizeof(struct ovl_dir_cache), GFP_KERNEL);
if (!cache)
@@ -339,8 +402,9 @@ static struct ovl_dir_cache *ovl_cache_get(struct dentry *dentry)
cache->refcount = 1;
INIT_LIST_HEAD(&cache->entries);
+ cache->root = RB_ROOT;
- res = ovl_dir_read_merged(dentry, &cache->entries);
+ res = ovl_dir_read_merged(dentry, &cache->entries, &cache->root);
if (res) {
ovl_cache_free(&cache->entries);
kfree(cache);
@@ -348,22 +412,266 @@ static struct ovl_dir_cache *ovl_cache_get(struct dentry *dentry)
}
cache->version = ovl_dentry_version_get(dentry);
- ovl_set_dir_cache(dentry, cache);
+ ovl_set_dir_cache(d_inode(dentry), cache);
return cache;
}
+/*
+ * Set d_ino for upper entries. Non-upper entries should always report
+ * the uppermost real inode ino and should not call this function.
+ *
+ * When not all layer are on same fs, report real ino also for upper.
+ *
+ * When all layers are on the same fs, and upper has a reference to
+ * copy up origin, call vfs_getattr() on the overlay entry to make
+ * sure that d_ino will be consistent with st_ino from stat(2).
+ */
+static int ovl_cache_update_ino(struct path *path, struct ovl_cache_entry *p)
+
+{
+ struct dentry *dir = path->dentry;
+ struct dentry *this = NULL;
+ enum ovl_path_type type;
+ u64 ino = p->real_ino;
+ int err = 0;
+
+ if (!ovl_same_sb(dir->d_sb))
+ goto out;
+
+ if (p->name[0] == '.') {
+ if (p->len == 1) {
+ this = dget(dir);
+ goto get;
+ }
+ if (p->len == 2 && p->name[1] == '.') {
+ /* we shall not be moved */
+ this = dget(dir->d_parent);
+ goto get;
+ }
+ }
+ this = lookup_one_len(p->name, dir, p->len);
+ if (IS_ERR_OR_NULL(this) || !this->d_inode) {
+ if (IS_ERR(this)) {
+ err = PTR_ERR(this);
+ this = NULL;
+ goto fail;
+ }
+ goto out;
+ }
+
+get:
+ type = ovl_path_type(this);
+ if (OVL_TYPE_ORIGIN(type)) {
+ struct kstat stat;
+ struct path statpath = *path;
+
+ statpath.dentry = this;
+ err = vfs_getattr(&statpath, &stat, STATX_INO, 0);
+ if (err)
+ goto fail;
+
+ WARN_ON_ONCE(dir->d_sb->s_dev != stat.dev);
+ ino = stat.ino;
+ }
+
+out:
+ p->ino = ino;
+ dput(this);
+ return err;
+
+fail:
+ pr_warn_ratelimited("overlay: failed to look up (%s) for ino (%i)\n",
+ p->name, err);
+ goto out;
+}
+
+static int ovl_fill_plain(struct dir_context *ctx, const char *name,
+ int namelen, loff_t offset, u64 ino,
+ unsigned int d_type)
+{
+ struct ovl_cache_entry *p;
+ struct ovl_readdir_data *rdd =
+ container_of(ctx, struct ovl_readdir_data, ctx);
+
+ rdd->count++;
+ p = ovl_cache_entry_new(rdd, name, namelen, ino, d_type);
+ if (p == NULL) {
+ rdd->err = -ENOMEM;
+ return -ENOMEM;
+ }
+ list_add_tail(&p->l_node, rdd->list);
+
+ return 0;
+}
+
+static int ovl_dir_read_impure(struct path *path, struct list_head *list,
+ struct rb_root *root)
+{
+ int err;
+ struct path realpath;
+ struct ovl_cache_entry *p, *n;
+ struct ovl_readdir_data rdd = {
+ .ctx.actor = ovl_fill_plain,
+ .list = list,
+ .root = root,
+ };
+
+ INIT_LIST_HEAD(list);
+ *root = RB_ROOT;
+ ovl_path_upper(path->dentry, &realpath);
+
+ err = ovl_dir_read(&realpath, &rdd);
+ if (err)
+ return err;
+
+ list_for_each_entry_safe(p, n, list, l_node) {
+ if (strcmp(p->name, ".") != 0 &&
+ strcmp(p->name, "..") != 0) {
+ err = ovl_cache_update_ino(path, p);
+ if (err)
+ return err;
+ }
+ if (p->ino == p->real_ino) {
+ list_del(&p->l_node);
+ kfree(p);
+ } else {
+ struct rb_node **newp = &root->rb_node;
+ struct rb_node *parent = NULL;
+
+ if (WARN_ON(ovl_cache_entry_find_link(p->name, p->len,
+ &newp, &parent)))
+ return -EIO;
+
+ rb_link_node(&p->node, parent, newp);
+ rb_insert_color(&p->node, root);
+ }
+ }
+ return 0;
+}
+
+static struct ovl_dir_cache *ovl_cache_get_impure(struct path *path)
+{
+ int res;
+ struct dentry *dentry = path->dentry;
+ struct ovl_dir_cache *cache;
+
+ cache = ovl_dir_cache(d_inode(dentry));
+ if (cache && ovl_dentry_version_get(dentry) == cache->version)
+ return cache;
+
+ /* Impure cache is not refcounted, free it here */
+ ovl_dir_cache_free(d_inode(dentry));
+ ovl_set_dir_cache(d_inode(dentry), NULL);
+
+ cache = kzalloc(sizeof(struct ovl_dir_cache), GFP_KERNEL);
+ if (!cache)
+ return ERR_PTR(-ENOMEM);
+
+ res = ovl_dir_read_impure(path, &cache->entries, &cache->root);
+ if (res) {
+ ovl_cache_free(&cache->entries);
+ kfree(cache);
+ return ERR_PTR(res);
+ }
+ if (list_empty(&cache->entries)) {
+ /* Good oportunity to get rid of an unnecessary "impure" flag */
+ ovl_do_removexattr(ovl_dentry_upper(dentry), OVL_XATTR_IMPURE);
+ ovl_clear_flag(OVL_IMPURE, d_inode(dentry));
+ kfree(cache);
+ return NULL;
+ }
+
+ cache->version = ovl_dentry_version_get(dentry);
+ ovl_set_dir_cache(d_inode(dentry), cache);
+
+ return cache;
+}
+
+struct ovl_readdir_translate {
+ struct dir_context *orig_ctx;
+ struct ovl_dir_cache *cache;
+ struct dir_context ctx;
+ u64 parent_ino;
+};
+
+static int ovl_fill_real(struct dir_context *ctx, const char *name,
+ int namelen, loff_t offset, u64 ino,
+ unsigned int d_type)
+{
+ struct ovl_readdir_translate *rdt =
+ container_of(ctx, struct ovl_readdir_translate, ctx);
+ struct dir_context *orig_ctx = rdt->orig_ctx;
+
+ if (rdt->parent_ino && strcmp(name, "..") == 0)
+ ino = rdt->parent_ino;
+ else if (rdt->cache) {
+ struct ovl_cache_entry *p;
+
+ p = ovl_cache_entry_find(&rdt->cache->root, name, namelen);
+ if (p)
+ ino = p->ino;
+ }
+
+ return orig_ctx->actor(orig_ctx, name, namelen, offset, ino, d_type);
+}
+
+static int ovl_iterate_real(struct file *file, struct dir_context *ctx)
+{
+ int err;
+ struct ovl_dir_file *od = file->private_data;
+ struct dentry *dir = file->f_path.dentry;
+ struct ovl_readdir_translate rdt = {
+ .ctx.actor = ovl_fill_real,
+ .orig_ctx = ctx,
+ };
+
+ if (OVL_TYPE_MERGE(ovl_path_type(dir->d_parent))) {
+ struct kstat stat;
+ struct path statpath = file->f_path;
+
+ statpath.dentry = dir->d_parent;
+ err = vfs_getattr(&statpath, &stat, STATX_INO, 0);
+ if (err)
+ return err;
+
+ WARN_ON_ONCE(dir->d_sb->s_dev != stat.dev);
+ rdt.parent_ino = stat.ino;
+ }
+
+ if (ovl_test_flag(OVL_IMPURE, d_inode(dir))) {
+ rdt.cache = ovl_cache_get_impure(&file->f_path);
+ if (IS_ERR(rdt.cache))
+ return PTR_ERR(rdt.cache);
+ }
+
+ return iterate_dir(od->realfile, &rdt.ctx);
+}
+
+
static int ovl_iterate(struct file *file, struct dir_context *ctx)
{
struct ovl_dir_file *od = file->private_data;
struct dentry *dentry = file->f_path.dentry;
struct ovl_cache_entry *p;
+ int err;
if (!ctx->pos)
ovl_dir_reset(file);
- if (od->is_real)
+ if (od->is_real) {
+ /*
+ * If parent is merge, then need to adjust d_ino for '..', if
+ * dir is impure then need to adjust d_ino for copied up
+ * entries.
+ */
+ if (ovl_same_sb(dentry->d_sb) &&
+ (ovl_test_flag(OVL_IMPURE, d_inode(dentry)) ||
+ OVL_TYPE_MERGE(ovl_path_type(dentry->d_parent)))) {
+ return ovl_iterate_real(file, ctx);
+ }
return iterate_dir(od->realfile, ctx);
+ }
if (!od->cache) {
struct ovl_dir_cache *cache;
@@ -378,9 +686,15 @@ static int ovl_iterate(struct file *file, struct dir_context *ctx)
while (od->cursor != &od->cache->entries) {
p = list_entry(od->cursor, struct ovl_cache_entry, l_node);
- if (!p->is_whiteout)
+ if (!p->is_whiteout) {
+ if (!p->ino) {
+ err = ovl_cache_update_ino(&file->f_path, p);
+ if (err)
+ return err;
+ }
if (!dir_emit(ctx, p->name, p->len, p->ino, p->type))
break;
+ }
od->cursor = p->l_node.next;
ctx->pos++;
}
@@ -446,14 +760,14 @@ static int ovl_dir_fsync(struct file *file, loff_t start, loff_t end,
ovl_path_upper(dentry, &upperpath);
realfile = ovl_path_open(&upperpath, O_RDONLY);
- smp_mb__before_spinlock();
+
inode_lock(inode);
if (!od->upperfile) {
if (IS_ERR(realfile)) {
inode_unlock(inode);
return PTR_ERR(realfile);
}
- od->upperfile = realfile;
+ smp_store_release(&od->upperfile, realfile);
} else {
/* somebody has beaten us to it */
if (!IS_ERR(realfile))
@@ -522,8 +836,9 @@ int ovl_check_empty_dir(struct dentry *dentry, struct list_head *list)
{
int err;
struct ovl_cache_entry *p;
+ struct rb_root root = RB_ROOT;
- err = ovl_dir_read_merged(dentry, list);
+ err = ovl_dir_read_merged(dentry, list, &root);
if (err)
return err;
@@ -612,12 +927,13 @@ static void ovl_workdir_cleanup_recurse(struct path *path, int level)
int err;
struct inode *dir = path->dentry->d_inode;
LIST_HEAD(list);
+ struct rb_root root = RB_ROOT;
struct ovl_cache_entry *p;
struct ovl_readdir_data rdd = {
.ctx.actor = ovl_fill_merge,
.dentry = NULL,
.list = &list,
- .root = RB_ROOT,
+ .root = &root,
.is_lowest = false,
};
@@ -675,12 +991,13 @@ int ovl_indexdir_cleanup(struct dentry *dentry, struct vfsmount *mnt,
struct inode *dir = dentry->d_inode;
struct path path = { .mnt = mnt, .dentry = dentry };
LIST_HEAD(list);
+ struct rb_root root = RB_ROOT;
struct ovl_cache_entry *p;
struct ovl_readdir_data rdd = {
.ctx.actor = ovl_fill_merge,
.dentry = NULL,
.list = &list,
- .root = RB_ROOT,
+ .root = &root,
.is_lowest = false,
};
@@ -703,7 +1020,10 @@ int ovl_indexdir_cleanup(struct dentry *dentry, struct vfsmount *mnt,
err = PTR_ERR(index);
break;
}
- if (ovl_verify_index(index, lowerstack, numlower)) {
+ err = ovl_verify_index(index, lowerstack, numlower);
+ if (err) {
+ if (err == -EROFS)
+ break;
err = ovl_cleanup(dir, index);
if (err)
break;
diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index 22ee2a9..fd5ea4f 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -70,20 +70,20 @@ static int ovl_check_append_only(struct inode *inode, int flag)
static struct dentry *ovl_d_real(struct dentry *dentry,
const struct inode *inode,
- unsigned int open_flags)
+ unsigned int open_flags, unsigned int flags)
{
struct dentry *real;
int err;
+ if (flags & D_REAL_UPPER)
+ return ovl_dentry_upper(dentry);
+
if (!d_is_reg(dentry)) {
if (!inode || inode == d_inode(dentry))
return dentry;
goto bug;
}
- if (d_is_negative(dentry))
- return dentry;
-
if (open_flags) {
err = ovl_open_maybe_copy_up(dentry, open_flags);
if (err)
@@ -105,7 +105,7 @@ static struct dentry *ovl_d_real(struct dentry *dentry,
goto bug;
/* Handle recursion */
- real = d_real(real, inode, open_flags);
+ real = d_real(real, inode, open_flags, 0);
if (!inode || inode == d_inode(real))
return real;
@@ -198,6 +198,7 @@ static void ovl_destroy_inode(struct inode *inode)
dput(oi->__upperdentry);
kfree(oi->redirect);
+ ovl_dir_cache_free(inode);
mutex_destroy(&oi->lock);
call_rcu(&inode->i_rcu, ovl_i_callback);
@@ -692,7 +693,7 @@ ovl_posix_acl_xattr_get(const struct xattr_handler *handler,
struct dentry *dentry, struct inode *inode,
const char *name, void *buffer, size_t size)
{
- return ovl_xattr_get(dentry, handler->name, buffer, size);
+ return ovl_xattr_get(dentry, inode, handler->name, buffer, size);
}
static int __maybe_unused
@@ -742,7 +743,7 @@ ovl_posix_acl_xattr_set(const struct xattr_handler *handler,
return err;
}
- err = ovl_xattr_set(dentry, handler->name, value, size, flags);
+ err = ovl_xattr_set(dentry, inode, handler->name, value, size, flags);
if (!err)
ovl_copyattr(ovl_inode_real(inode), inode);
@@ -772,7 +773,7 @@ static int ovl_other_xattr_get(const struct xattr_handler *handler,
struct dentry *dentry, struct inode *inode,
const char *name, void *buffer, size_t size)
{
- return ovl_xattr_get(dentry, name, buffer, size);
+ return ovl_xattr_get(dentry, inode, name, buffer, size);
}
static int ovl_other_xattr_set(const struct xattr_handler *handler,
@@ -780,7 +781,7 @@ static int ovl_other_xattr_set(const struct xattr_handler *handler,
const char *name, const void *value,
size_t size, int flags)
{
- return ovl_xattr_set(dentry, name, value, size, flags);
+ return ovl_xattr_set(dentry, inode, name, value, size, flags);
}
static const struct xattr_handler __maybe_unused
@@ -1058,10 +1059,6 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
ufs->indexdir = ovl_workdir_create(sb, ufs, workpath.dentry,
OVL_INDEXDIR_NAME, true);
- err = PTR_ERR(ufs->indexdir);
- if (IS_ERR(ufs->indexdir))
- goto out_put_lower_mnt;
-
if (ufs->indexdir) {
/* Verify upper root is index dir origin */
err = ovl_verify_origin(ufs->indexdir, ufs->upper_mnt,
@@ -1090,6 +1087,7 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
else
sb->s_d_op = &ovl_dentry_operations;
+ err = -ENOMEM;
ufs->creator_cred = cred = prepare_creds();
if (!cred)
goto out_put_indexdir;
diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c
index c492ba7..1177945 100644
--- a/fs/overlayfs/util.c
+++ b/fs/overlayfs/util.c
@@ -157,9 +157,14 @@ struct dentry *ovl_dentry_real(struct dentry *dentry)
return ovl_dentry_upper(dentry) ?: ovl_dentry_lower(dentry);
}
+struct dentry *ovl_i_dentry_upper(struct inode *inode)
+{
+ return ovl_upperdentry_dereference(OVL_I(inode));
+}
+
struct inode *ovl_inode_upper(struct inode *inode)
{
- struct dentry *upperdentry = ovl_upperdentry_dereference(OVL_I(inode));
+ struct dentry *upperdentry = ovl_i_dentry_upper(inode);
return upperdentry ? d_inode(upperdentry) : NULL;
}
@@ -175,14 +180,14 @@ struct inode *ovl_inode_real(struct inode *inode)
}
-struct ovl_dir_cache *ovl_dir_cache(struct dentry *dentry)
+struct ovl_dir_cache *ovl_dir_cache(struct inode *inode)
{
- return OVL_I(d_inode(dentry))->cache;
+ return OVL_I(inode)->cache;
}
-void ovl_set_dir_cache(struct dentry *dentry, struct ovl_dir_cache *cache)
+void ovl_set_dir_cache(struct inode *inode, struct ovl_dir_cache *cache)
{
- OVL_I(d_inode(dentry))->cache = cache;
+ OVL_I(inode)->cache = cache;
}
bool ovl_dentry_is_opaque(struct dentry *dentry)
@@ -270,12 +275,19 @@ void ovl_inode_update(struct inode *inode, struct dentry *upperdentry)
}
}
-void ovl_dentry_version_inc(struct dentry *dentry)
+void ovl_dentry_version_inc(struct dentry *dentry, bool impurity)
{
struct inode *inode = d_inode(dentry);
WARN_ON(!inode_is_locked(inode));
- OVL_I(inode)->version++;
+ /*
+ * Version is used by readdir code to keep cache consistent. For merge
+ * dirs all changes need to be noted. For non-merge dirs, cache only
+ * contains impure (ones which have been copied up and have origins)
+ * entries, so only need to note changes to impure entries.
+ */
+ if (OVL_TYPE_MERGE(ovl_path_type(dentry)) || impurity)
+ OVL_I(inode)->version++;
}
u64 ovl_dentry_version_get(struct dentry *dentry)
@@ -377,6 +389,11 @@ void ovl_set_flag(unsigned long flag, struct inode *inode)
set_bit(flag, &OVL_I(inode)->flags);
}
+void ovl_clear_flag(unsigned long flag, struct inode *inode)
+{
+ clear_bit(flag, &OVL_I(inode)->flags);
+}
+
bool ovl_test_flag(unsigned long flag, struct inode *inode)
{
return test_bit(flag, &OVL_I(inode)->flags);
OpenPOWER on IntegriCloud