diff options
Diffstat (limited to 'fs/overlayfs/inode.c')
| -rw-r--r-- | fs/overlayfs/inode.c | 505 |
1 files changed, 403 insertions, 102 deletions
diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c index 79e8994e3bc1..9e61511de7a7 100644 --- a/fs/overlayfs/inode.c +++ b/fs/overlayfs/inode.c @@ -10,17 +10,23 @@ #include <linux/xattr.h> #include <linux/posix_acl.h> #include <linux/ratelimit.h> +#include <linux/fiemap.h> +#include <linux/fileattr.h> +#include <linux/security.h> +#include <linux/namei.h> #include "overlayfs.h" -int ovl_setattr(struct dentry *dentry, struct iattr *attr) +int ovl_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { int err; + struct ovl_fs *ofs = OVL_FS(dentry->d_sb); bool full_copy_up = false; struct dentry *upperdentry; const struct cred *old_cred; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) return err; @@ -29,12 +35,6 @@ int ovl_setattr(struct dentry *dentry, struct iattr *attr) goto out; if (attr->ia_valid & ATTR_SIZE) { - struct inode *realinode = d_inode(ovl_dentry_real(dentry)); - - err = -ETXTBSY; - if (atomic_read(&realinode->i_writecount) < 0) - goto out_drop_write; - /* Truncate should trigger data copy up as well */ full_copy_up = true; } @@ -58,12 +58,30 @@ int ovl_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid & (ATTR_KILL_SUID|ATTR_KILL_SGID)) attr->ia_valid &= ~ATTR_MODE; + /* + * We might have to translate ovl file into real file object + * once use cases emerge. For now, simply don't let underlying + * filesystem rely on attr->ia_file + */ + attr->ia_valid &= ~ATTR_FILE; + + /* + * If open(O_TRUNC) is done, VFS calls ->setattr with ATTR_OPEN + * set. Overlayfs does not pass O_TRUNC flag to underlying + * filesystem during open -> do not pass ATTR_OPEN. This + * disables optimization in fuse which assumes open(O_TRUNC) + * already set file size to 0. But we never passed O_TRUNC to + * fuse. So by clearing ATTR_OPEN, fuse will be forced to send + * setattr request to server. + */ + attr->ia_valid &= ~ATTR_OPEN; + inode_lock(upperdentry->d_inode); old_cred = ovl_override_creds(dentry->d_sb); - err = notify_change(upperdentry, attr, NULL); + err = ovl_do_notify_change(ofs, upperdentry, attr); revert_creds(old_cred); if (!err) - ovl_copyattr(upperdentry->d_inode, dentry->d_inode); + ovl_copyattr(dentry->d_inode); inode_unlock(upperdentry->d_inode); if (winode) @@ -75,10 +93,11 @@ out: return err; } -static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid) +static void ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid) { bool samefs = ovl_same_fs(dentry->d_sb); unsigned int xinobits = ovl_xino_bits(dentry->d_sb); + unsigned int xinoshift = 64 - xinobits; if (samefs) { /* @@ -87,24 +106,24 @@ static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid) * which is friendly to du -x. */ stat->dev = dentry->d_sb->s_dev; - return 0; + return; } else if (xinobits) { - unsigned int shift = 64 - xinobits; /* * All inode numbers of underlying fs should not be using the * high xinobits, so we use high xinobits to partition the * overlay st_ino address space. The high bits holds the fsid - * (upper fsid is 0). This way overlay inode numbers are unique - * and all inodes use overlay st_dev. Inode numbers are also - * persistent for a given layer configuration. + * (upper fsid is 0). The lowest xinobit is reserved for mapping + * the non-persistent inode numbers range in case of overflow. + * This way all overlay inode numbers are unique and use the + * overlay st_dev. */ - if (stat->ino >> shift) { + if (likely(!(stat->ino >> xinoshift))) { + stat->ino |= ((u64)fsid) << (xinoshift + 1); + stat->dev = dentry->d_sb->s_dev; + return; + } else if (ovl_xino_warn(dentry->d_sb)) { pr_warn_ratelimited("inode number too big (%pd2, ino=%llu, xinobits=%d)\n", dentry, stat->ino, xinobits); - } else { - stat->ino |= ((u64)fsid) << shift; - stat->dev = dentry->d_sb->s_dev; - return 0; } } @@ -130,18 +149,17 @@ static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid) */ stat->dev = OVL_FS(dentry->d_sb)->fs[fsid].pseudo_dev; } - - return 0; } -int ovl_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int ovl_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; enum ovl_path_type type; struct path realpath; const struct cred *old_cred; - bool is_dir = S_ISDIR(dentry->d_inode->i_mode); + struct inode *inode = d_inode(dentry); + bool is_dir = S_ISDIR(inode->i_mode); int fsid = 0; int err; bool metacopy_blocks = false; @@ -154,6 +172,9 @@ int ovl_getattr(const struct path *path, struct kstat *stat, if (err) goto out; + /* Report the effective immutable/append-only STATX flags */ + generic_fill_statx_attr(inode, stat); + /* * For non-dir or same fs, we use st_ino of the copy up origin. * This guaranties constant st_dev/st_ino across copy up. @@ -230,9 +251,7 @@ int ovl_getattr(const struct path *path, struct kstat *stat, } } - err = ovl_map_dev_ino(dentry, stat, fsid); - if (err) - goto out; + ovl_map_dev_ino(dentry, stat, fsid); /* * It's probably not worth it to count subdirs to get the @@ -257,15 +276,18 @@ out: return err; } -int ovl_permission(struct inode *inode, int mask) +int ovl_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct inode *upperinode = ovl_inode_upper(inode); - struct inode *realinode = upperinode ?: ovl_inode_lower(inode); + struct inode *realinode; + struct path realpath; const struct cred *old_cred; int err; /* Careful in RCU walk mode */ - if (!realinode) { + ovl_i_path_real(inode, &realpath); + if (!realpath.dentry) { WARN_ON(!(mask & MAY_NOT_BLOCK)); return -ECHILD; } @@ -274,10 +296,11 @@ int ovl_permission(struct inode *inode, int mask) * Check overlay inode with the creds of task and underlying inode * with creds of mounter */ - err = generic_permission(inode, mask); + err = generic_permission(&init_user_ns, inode, mask); if (err) return err; + realinode = d_inode(realpath.dentry); old_cred = ovl_override_creds(inode->i_sb); if (!upperinode && !special_file(realinode->i_mode) && mask & MAY_WRITE) { @@ -285,7 +308,7 @@ int ovl_permission(struct inode *inode, int mask) /* Make sure mounter can read file for copy up later */ mask |= MAY_READ; } - err = inode_permission(realinode, mask); + err = inode_permission(mnt_user_ns(realpath.mnt), realinode, mask); revert_creds(old_cred); return err; @@ -307,18 +330,26 @@ static const char *ovl_get_link(struct dentry *dentry, return p; } -bool ovl_is_private_xattr(const char *name) +bool ovl_is_private_xattr(struct super_block *sb, const char *name) { - return strncmp(name, OVL_XATTR_PREFIX, - sizeof(OVL_XATTR_PREFIX) - 1) == 0; + struct ovl_fs *ofs = sb->s_fs_info; + + if (ofs->config.userxattr) + return strncmp(name, OVL_XATTR_USER_PREFIX, + sizeof(OVL_XATTR_USER_PREFIX) - 1) == 0; + else + return strncmp(name, OVL_XATTR_TRUSTED_PREFIX, + sizeof(OVL_XATTR_TRUSTED_PREFIX) - 1) == 0; } int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) { int err; + struct ovl_fs *ofs = OVL_FS(dentry->d_sb); struct dentry *upperdentry = ovl_i_dentry_upper(inode); struct dentry *realdentry = upperdentry ?: ovl_dentry_lower(dentry); + struct path realpath; const struct cred *old_cred; err = ovl_want_write(dentry); @@ -326,7 +357,10 @@ int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name, goto out; if (!value && !upperdentry) { - err = vfs_getxattr(realdentry, name, NULL, 0); + ovl_path_lower(dentry, &realpath); + old_cred = ovl_override_creds(dentry->d_sb); + err = vfs_getxattr(mnt_user_ns(realpath.mnt), realdentry, name, NULL, 0); + revert_creds(old_cred); if (err < 0) goto out_drop_write; } @@ -340,16 +374,17 @@ int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name, } old_cred = ovl_override_creds(dentry->d_sb); - if (value) - err = vfs_setxattr(realdentry, name, value, size, flags); - else { + if (value) { + err = ovl_do_setxattr(ofs, realdentry, name, value, size, + flags); + } else { WARN_ON(flags != XATTR_REPLACE); - err = vfs_removexattr(realdentry, name); + err = ovl_do_removexattr(ofs, realdentry, name); } revert_creds(old_cred); /* copy c/mtime */ - ovl_copyattr(d_inode(realdentry), inode); + ovl_copyattr(inode); out_drop_write: ovl_drop_write(dentry); @@ -362,24 +397,27 @@ int ovl_xattr_get(struct dentry *dentry, struct inode *inode, const char *name, { ssize_t res; const struct cred *old_cred; - struct dentry *realdentry = - ovl_i_dentry_upper(inode) ?: ovl_dentry_lower(dentry); + struct path realpath; + ovl_i_path_real(inode, &realpath); old_cred = ovl_override_creds(dentry->d_sb); - res = vfs_getxattr(realdentry, name, value, size); + res = vfs_getxattr(mnt_user_ns(realpath.mnt), realpath.dentry, name, value, size); revert_creds(old_cred); return res; } -static bool ovl_can_list(const char *s) +static bool ovl_can_list(struct super_block *sb, const char *s) { - /* List all non-trusted xatts */ + /* Never list private (.overlay) */ + if (ovl_is_private_xattr(sb, s)) + return false; + + /* List all non-trusted xattrs */ if (strncmp(s, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) != 0) return true; - /* Never list trusted.overlay, list other trusted for superuser only */ - return !ovl_is_private_xattr(s) && - ns_capable_noaudit(&init_user_ns, CAP_SYS_ADMIN); + /* list other trusted for superuser only */ + return ns_capable_noaudit(&init_user_ns, CAP_SYS_ADMIN); } ssize_t ovl_listxattr(struct dentry *dentry, char *list, size_t size) @@ -405,7 +443,7 @@ ssize_t ovl_listxattr(struct dentry *dentry, char *list, size_t size) return -EIO; len -= slen; - if (!ovl_can_list(s)) { + if (!ovl_can_list(dentry->d_sb, s)) { res -= slen; memmove(s, s + slen, len); } else { @@ -416,28 +454,107 @@ ssize_t ovl_listxattr(struct dentry *dentry, char *list, size_t size) return res; } -struct posix_acl *ovl_get_acl(struct inode *inode, int type) +#ifdef CONFIG_FS_POSIX_ACL +/* + * Apply the idmapping of the layer to POSIX ACLs. The caller must pass a clone + * of the POSIX ACLs retrieved from the lower layer to this function to not + * alter the POSIX ACLs for the underlying filesystem. + */ +static void ovl_idmap_posix_acl(struct inode *realinode, + struct user_namespace *mnt_userns, + struct posix_acl *acl) +{ + struct user_namespace *fs_userns = i_user_ns(realinode); + + for (unsigned int i = 0; i < acl->a_count; i++) { + vfsuid_t vfsuid; + vfsgid_t vfsgid; + + struct posix_acl_entry *e = &acl->a_entries[i]; + switch (e->e_tag) { + case ACL_USER: + vfsuid = make_vfsuid(mnt_userns, fs_userns, e->e_uid); + e->e_uid = vfsuid_into_kuid(vfsuid); + break; + case ACL_GROUP: + vfsgid = make_vfsgid(mnt_userns, fs_userns, e->e_gid); + e->e_gid = vfsgid_into_kgid(vfsgid); + break; + } + } +} + +/* + * When the relevant layer is an idmapped mount we need to take the idmapping + * of the layer into account and translate any ACL_{GROUP,USER} values + * according to the idmapped mount. + * + * We cannot alter the ACLs returned from the relevant layer as that would + * alter the cached values filesystem wide for the lower filesystem. Instead we + * can clone the ACLs and then apply the relevant idmapping of the layer. + * + * This is obviously only relevant when idmapped layers are used. + */ +struct posix_acl *ovl_get_acl(struct inode *inode, int type, bool rcu) { struct inode *realinode = ovl_inode_real(inode); - const struct cred *old_cred; - struct posix_acl *acl; + struct posix_acl *acl, *clone; + struct path realpath; - if (!IS_ENABLED(CONFIG_FS_POSIX_ACL) || !IS_POSIXACL(realinode)) + if (!IS_POSIXACL(realinode)) return NULL; - old_cred = ovl_override_creds(inode->i_sb); - acl = get_acl(realinode, type); - revert_creds(old_cred); + /* Careful in RCU walk mode */ + ovl_i_path_real(inode, &realpath); + if (!realpath.dentry) { + WARN_ON(!rcu); + return ERR_PTR(-ECHILD); + } + + if (rcu) { + acl = get_cached_acl_rcu(realinode, type); + } else { + const struct cred *old_cred; + + old_cred = ovl_override_creds(inode->i_sb); + acl = get_acl(realinode, type); + revert_creds(old_cred); + } + /* + * If there are no POSIX ACLs, or we encountered an error, + * or the layer isn't idmapped we don't need to do anything. + */ + if (!is_idmapped_mnt(realpath.mnt) || IS_ERR_OR_NULL(acl)) + return acl; + + /* + * We only get here if the layer is idmapped. So drop out of RCU path + * walk so we can clone the ACLs. There's no need to release the ACLs + * since get_cached_acl_rcu() doesn't take a reference on the ACLs. + */ + if (rcu) + return ERR_PTR(-ECHILD); - return acl; + clone = posix_acl_clone(acl, GFP_KERNEL); + if (!clone) + clone = ERR_PTR(-ENOMEM); + else + ovl_idmap_posix_acl(realinode, mnt_user_ns(realpath.mnt), clone); + /* + * Since we're not in RCU path walk we always need to release the + * original ACLs. + */ + posix_acl_release(acl); + return clone; } +#endif int ovl_update_time(struct inode *inode, struct timespec64 *ts, int flags) { if (flags & S_ATIME) { struct ovl_fs *ofs = inode->i_sb->s_fs_info; struct path upperpath = { - .mnt = ofs->upper_mnt, + .mnt = ovl_upper_mnt(ofs), .dentry = ovl_upperdentry_dereference(OVL_I(inode)), }; @@ -453,18 +570,147 @@ static int ovl_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len) { int err; - struct inode *realinode = ovl_inode_real(inode); + struct inode *realinode = ovl_inode_realdata(inode); const struct cred *old_cred; if (!realinode->i_op->fiemap) return -EOPNOTSUPP; old_cred = ovl_override_creds(inode->i_sb); + err = realinode->i_op->fiemap(realinode, fieinfo, start, len); + revert_creds(old_cred); - if (fieinfo->fi_flags & FIEMAP_FLAG_SYNC) - filemap_write_and_wait(realinode->i_mapping); + return err; +} - err = realinode->i_op->fiemap(realinode, fieinfo, start, len); +/* + * Work around the fact that security_file_ioctl() takes a file argument. + * Introducing security_inode_fileattr_get/set() hooks would solve this issue + * properly. + */ +static int ovl_security_fileattr(const struct path *realpath, struct fileattr *fa, + bool set) +{ + struct file *file; + unsigned int cmd; + int err; + + file = dentry_open(realpath, O_RDONLY, current_cred()); + if (IS_ERR(file)) + return PTR_ERR(file); + + if (set) + cmd = fa->fsx_valid ? FS_IOC_FSSETXATTR : FS_IOC_SETFLAGS; + else + cmd = fa->fsx_valid ? FS_IOC_FSGETXATTR : FS_IOC_GETFLAGS; + + err = security_file_ioctl(file, cmd, 0); + fput(file); + + return err; +} + +int ovl_real_fileattr_set(const struct path *realpath, struct fileattr *fa) +{ + int err; + + err = ovl_security_fileattr(realpath, fa, true); + if (err) + return err; + + return vfs_fileattr_set(mnt_user_ns(realpath->mnt), realpath->dentry, fa); +} + +int ovl_fileattr_set(struct user_namespace *mnt_userns, + struct dentry *dentry, struct fileattr *fa) +{ + struct inode *inode = d_inode(dentry); + struct path upperpath; + const struct cred *old_cred; + unsigned int flags; + int err; + + err = ovl_want_write(dentry); + if (err) + goto out; + + err = ovl_copy_up(dentry); + if (!err) { + ovl_path_real(dentry, &upperpath); + + old_cred = ovl_override_creds(inode->i_sb); + /* + * Store immutable/append-only flags in xattr and clear them + * in upper fileattr (in case they were set by older kernel) + * so children of "ovl-immutable" directories lower aliases of + * "ovl-immutable" hardlinks could be copied up. + * Clear xattr when flags are cleared. + */ + err = ovl_set_protattr(inode, upperpath.dentry, fa); + if (!err) + err = ovl_real_fileattr_set(&upperpath, fa); + revert_creds(old_cred); + + /* + * Merge real inode flags with inode flags read from + * overlay.protattr xattr + */ + flags = ovl_inode_real(inode)->i_flags & OVL_COPY_I_FLAGS_MASK; + + BUILD_BUG_ON(OVL_PROT_I_FLAGS_MASK & ~OVL_COPY_I_FLAGS_MASK); + flags |= inode->i_flags & OVL_PROT_I_FLAGS_MASK; + inode_set_flags(inode, flags, OVL_COPY_I_FLAGS_MASK); + + /* Update ctime */ + ovl_copyattr(inode); + } + ovl_drop_write(dentry); +out: + return err; +} + +/* Convert inode protection flags to fileattr flags */ +static void ovl_fileattr_prot_flags(struct inode *inode, struct fileattr *fa) +{ + BUILD_BUG_ON(OVL_PROT_FS_FLAGS_MASK & ~FS_COMMON_FL); + BUILD_BUG_ON(OVL_PROT_FSX_FLAGS_MASK & ~FS_XFLAG_COMMON); + + if (inode->i_flags & S_APPEND) { + fa->flags |= FS_APPEND_FL; + fa->fsx_xflags |= FS_XFLAG_APPEND; + } + if (inode->i_flags & S_IMMUTABLE) { + fa->flags |= FS_IMMUTABLE_FL; + fa->fsx_xflags |= FS_XFLAG_IMMUTABLE; + } +} + +int ovl_real_fileattr_get(const struct path *realpath, struct fileattr *fa) +{ + int err; + + err = ovl_security_fileattr(realpath, fa, false); + if (err) + return err; + + err = vfs_fileattr_get(realpath->dentry, fa); + if (err == -ENOIOCTLCMD) + err = -ENOTTY; + return err; +} + +int ovl_fileattr_get(struct dentry *dentry, struct fileattr *fa) +{ + struct inode *inode = d_inode(dentry); + struct path realpath; + const struct cred *old_cred; + int err; + + ovl_path_real(dentry, &realpath); + + old_cred = ovl_override_creds(inode->i_sb); + err = ovl_real_fileattr_get(&realpath, fa); + ovl_fileattr_prot_flags(inode, fa); revert_creds(old_cred); return err; @@ -478,6 +724,8 @@ static const struct inode_operations ovl_file_inode_operations = { .get_acl = ovl_get_acl, .update_time = ovl_update_time, .fiemap = ovl_fiemap, + .fileattr_get = ovl_fileattr_get, + .fileattr_set = ovl_fileattr_set, }; static const struct inode_operations ovl_symlink_inode_operations = { @@ -504,11 +752,11 @@ static const struct address_space_operations ovl_aops = { /* * It is possible to stack overlayfs instance on top of another - * overlayfs instance as lower layer. We need to annonate the + * overlayfs instance as lower layer. We need to annotate the * stackable i_mutex locks according to stack level of the super * block instance. An overlayfs instance can never be in stack * depth 0 (there is always a real fs below it). An overlayfs - * inode lock will use the lockdep annotaion ovl_i_mutex_key[depth]. + * inode lock will use the lockdep annotation ovl_i_mutex_key[depth]. * * For example, here is a snip from /proc/lockdep_chains after * dir_iterate of nested overlayfs: @@ -561,27 +809,76 @@ static inline void ovl_lockdep_annotate_inode_mutex_key(struct inode *inode) #endif } -static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev, - unsigned long ino, int fsid) +static void ovl_next_ino(struct inode *inode) +{ + struct ovl_fs *ofs = inode->i_sb->s_fs_info; + + inode->i_ino = atomic_long_inc_return(&ofs->last_ino); + if (unlikely(!inode->i_ino)) + inode->i_ino = atomic_long_inc_return(&ofs->last_ino); +} + +static void ovl_map_ino(struct inode *inode, unsigned long ino, int fsid) { int xinobits = ovl_xino_bits(inode->i_sb); + unsigned int xinoshift = 64 - xinobits; /* * When d_ino is consistent with st_ino (samefs or i_ino has enough * bits to encode layer), set the same value used for st_ino to i_ino, * so inode number exposed via /proc/locks and a like will be * consistent with d_ino and st_ino values. An i_ino value inconsistent - * with d_ino also causes nfsd readdirplus to fail. When called from - * ovl_new_inode(), ino arg is 0, so i_ino will be updated to real - * upper inode i_ino on ovl_inode_init() or ovl_inode_update(). + * with d_ino also causes nfsd readdirplus to fail. */ - if (ovl_same_dev(inode->i_sb)) { - inode->i_ino = ino; - if (xinobits && fsid && !(ino >> (64 - xinobits))) - inode->i_ino |= (unsigned long)fsid << (64 - xinobits); - } else { - inode->i_ino = get_next_ino(); + inode->i_ino = ino; + if (ovl_same_fs(inode->i_sb)) { + return; + } else if (xinobits && likely(!(ino >> xinoshift))) { + inode->i_ino |= (unsigned long)fsid << (xinoshift + 1); + return; + } + + /* + * For directory inodes on non-samefs with xino disabled or xino + * overflow, we allocate a non-persistent inode number, to be used for + * resolving st_ino collisions in ovl_map_dev_ino(). + * + * To avoid ino collision with legitimate xino values from upper + * layer (fsid 0), use the lowest xinobit to map the non + * persistent inode numbers to the unified st_ino address space. + */ + if (S_ISDIR(inode->i_mode)) { + ovl_next_ino(inode); + if (xinobits) { + inode->i_ino &= ~0UL >> xinobits; + inode->i_ino |= 1UL << xinoshift; + } + } +} + +void ovl_inode_init(struct inode *inode, struct ovl_inode_params *oip, + unsigned long ino, int fsid) +{ + struct inode *realinode; + struct ovl_inode *oi = OVL_I(inode); + + if (oip->upperdentry) + oi->__upperdentry = oip->upperdentry; + if (oip->lowerpath && oip->lowerpath->dentry) { + oi->lowerpath.dentry = dget(oip->lowerpath->dentry); + oi->lowerpath.layer = oip->lowerpath->layer; } + if (oip->lowerdata) + oi->lowerdata = igrab(d_inode(oip->lowerdata)); + + realinode = ovl_inode_real(inode); + ovl_copyattr(inode); + ovl_copyflags(realinode, inode); + ovl_map_ino(inode, ino, fsid); +} + +static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev) +{ inode->i_mode = mode; inode->i_flags |= S_NOCMTIME; #ifdef CONFIG_FS_POSIX_ACL @@ -625,7 +922,7 @@ static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev, * For the first, copy up case, the union nlink does not change, whether the * operation succeeds or fails, but the upper inode nlink may change. * Therefore, before copy up, we store the union nlink value relative to the - * lower inode nlink in the index inode xattr trusted.overlay.nlink. + * lower inode nlink in the index inode xattr .overlay.nlink. * * For the second, upper hardlink case, the union nlink should be incremented * or decremented IFF the operation succeeds, aligned with nlink change of the @@ -660,8 +957,8 @@ static int ovl_set_nlink_common(struct dentry *dentry, if (WARN_ON(len >= sizeof(buf))) return -EIO; - return ovl_do_setxattr(ovl_dentry_upper(dentry), - OVL_XATTR_NLINK, buf, len, 0); + return ovl_setxattr(OVL_FS(inode->i_sb), ovl_dentry_upper(dentry), + OVL_XATTR_NLINK, buf, len); } int ovl_set_nlink_upper(struct dentry *dentry) @@ -674,7 +971,7 @@ int ovl_set_nlink_lower(struct dentry *dentry) return ovl_set_nlink_common(dentry, ovl_dentry_lower(dentry), "L%+i"); } -unsigned int ovl_get_nlink(struct dentry *lowerdentry, +unsigned int ovl_get_nlink(struct ovl_fs *ofs, struct dentry *lowerdentry, struct dentry *upperdentry, unsigned int fallback) { @@ -686,7 +983,8 @@ unsigned int ovl_get_nlink(struct dentry *lowerdentry, if (!lowerdentry || !upperdentry || d_inode(lowerdentry)->i_nlink == 1) return fallback; - err = vfs_getxattr(upperdentry, OVL_XATTR_NLINK, &buf, sizeof(buf) - 1); + err = ovl_getxattr_upper(ofs, upperdentry, OVL_XATTR_NLINK, + &buf, sizeof(buf) - 1); if (err < 0) goto fail; @@ -719,7 +1017,7 @@ struct inode *ovl_new_inode(struct super_block *sb, umode_t mode, dev_t rdev) inode = new_inode(sb); if (inode) - ovl_fill_inode(inode, mode, rdev, 0, 0); + ovl_fill_inode(inode, mode, rdev); return inode; } @@ -843,7 +1141,7 @@ struct inode *ovl_get_trap_inode(struct super_block *sb, struct dentry *dir) * Does overlay inode need to be hashed by lower inode? */ static bool ovl_hash_bylower(struct super_block *sb, struct dentry *upper, - struct dentry *lower, struct dentry *index) + struct dentry *lower, bool index) { struct ovl_fs *ofs = sb->s_fs_info; @@ -856,7 +1154,7 @@ static bool ovl_hash_bylower(struct super_block *sb, struct dentry *upper, return true; /* Yes, if won't be copied up */ - if (!ofs->upper_mnt) + if (!ovl_upper_mnt(ofs)) return true; /* No, if lower hardlink is or will be broken on copy up */ @@ -884,15 +1182,20 @@ static struct inode *ovl_iget5(struct super_block *sb, struct inode *newinode, struct inode *ovl_get_inode(struct super_block *sb, struct ovl_inode_params *oip) { + struct ovl_fs *ofs = OVL_FS(sb); struct dentry *upperdentry = oip->upperdentry; struct ovl_path *lowerpath = oip->lowerpath; struct inode *realinode = upperdentry ? d_inode(upperdentry) : NULL; struct inode *inode; struct dentry *lowerdentry = lowerpath ? lowerpath->dentry : NULL; + struct path realpath = { + .dentry = upperdentry ?: lowerdentry, + .mnt = upperdentry ? ovl_upper_mnt(ofs) : lowerpath->layer->mnt, + }; bool bylower = ovl_hash_bylower(sb, upperdentry, lowerdentry, oip->index); - int fsid = bylower ? oip->lowerpath->layer->fsid : 0; - bool is_dir, metacopy = false; + int fsid = bylower ? lowerpath->layer->fsid : 0; + bool is_dir; unsigned long ino = 0; int err = oip->newinode ? -EEXIST : -ENOMEM; @@ -931,7 +1234,8 @@ struct inode *ovl_get_inode(struct super_block *sb, /* Recalculate nlink for non-dir due to indexing */ if (!is_dir) - nlink = ovl_get_nlink(lowerdentry, upperdentry, nlink); + nlink = ovl_get_nlink(ofs, lowerdentry, upperdentry, + nlink); set_nlink(inode, nlink); ino = key->i_ino; } else { @@ -941,25 +1245,18 @@ struct inode *ovl_get_inode(struct super_block *sb, err = -ENOMEM; goto out_err; } + ino = realinode->i_ino; + fsid = lowerpath->layer->fsid; } - ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev, ino, fsid); - ovl_inode_init(inode, upperdentry, lowerdentry, oip->lowerdata); + ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev); + ovl_inode_init(inode, oip, ino, fsid); - if (upperdentry && ovl_is_impuredir(upperdentry)) + if (upperdentry && ovl_is_impuredir(sb, upperdentry)) ovl_set_flag(OVL_IMPURE, inode); if (oip->index) ovl_set_flag(OVL_INDEX, inode); - if (upperdentry) { - err = ovl_check_metacopy_xattr(upperdentry); - if (err < 0) - goto out_err; - metacopy = err; - if (!metacopy) - ovl_set_flag(OVL_UPPERDATA, inode); - } - OVL_I(inode)->redirect = oip->redirect; if (bylower) @@ -968,11 +1265,15 @@ struct inode *ovl_get_inode(struct super_block *sb, /* Check for non-merge dir that may have whiteouts */ if (is_dir) { if (((upperdentry && lowerdentry) || oip->numlower > 1) || - ovl_check_origin_xattr(upperdentry ?: lowerdentry)) { + ovl_path_check_origin_xattr(ofs, &realpath)) { ovl_set_flag(OVL_WHITEOUTS, inode); } } + /* Check for immutable/append-only inode flags in xattr */ + if (upperdentry) + ovl_check_protattr(inode, upperdentry); + if (inode->i_state & I_NEW) unlock_new_inode(inode); out: |