linux/fs/overlayfs/copy_up.c

967 lines
22 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
/*
*
* Copyright (C) 2011 Novell Inc.
*/
#include <linux/module.h>
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/file.h>
#include <linux/splice.h>
#include <linux/xattr.h>
#include <linux/security.h>
#include <linux/uaccess.h>
#include <linux/sched/signal.h>
#include <linux/cred.h>
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
#include <linux/namei.h>
#include <linux/fdtable.h>
#include <linux/ratelimit.h>
#include <linux/exportfs.h>
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
#include "overlayfs.h"
#define OVL_COPY_UP_CHUNK_SIZE (1 << 20)
static int ovl_ccup_set(const char *buf, const struct kernel_param *param)
{
pr_warn("\"check_copy_up\" module option is obsolete\n");
return 0;
}
static int ovl_ccup_get(char *buf, const struct kernel_param *param)
{
return sprintf(buf, "N\n");
}
module_param_call(check_copy_up, ovl_ccup_set, ovl_ccup_get, NULL, 0644);
ovl: fix typo in MODULE_PARM_DESC Change first argument to MODULE_PARM_DESC() calls, that each of them matched the actual module parameter name. The matching results in changing (the 'parm' section from) the output of `modinfo overlay` from: parm: ovl_check_copy_up:Obsolete; does nothing parm: redirect_max:ushort parm: ovl_redirect_max:Maximum length of absolute redirect xattr value parm: redirect_dir:bool parm: ovl_redirect_dir_def:Default to on or off for the redirect_dir feature parm: redirect_always_follow:bool parm: ovl_redirect_always_follow:Follow redirects even if redirect_dir feature is turned off parm: index:bool parm: ovl_index_def:Default to on or off for the inodes index feature parm: nfs_export:bool parm: ovl_nfs_export_def:Default to on or off for the NFS export feature parm: xino_auto:bool parm: ovl_xino_auto_def:Auto enable xino feature parm: metacopy:bool parm: ovl_metacopy_def:Default to on or off for the metadata only copy up feature into: parm: check_copy_up:Obsolete; does nothing parm: redirect_max:Maximum length of absolute redirect xattr value (ushort) parm: redirect_dir:Default to on or off for the redirect_dir feature (bool) parm: redirect_always_follow:Follow redirects even if redirect_dir feature is turned off (bool) parm: index:Default to on or off for the inodes index feature (bool) parm: nfs_export:Default to on or off for the NFS export feature (bool) parm: xino_auto:Auto enable xino feature (bool) parm: metacopy:Default to on or off for the metadata only copy up feature (bool) Signed-off-by: Nicolas Schier <n.schier@avm.de> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2019-06-17 15:39:00 +08:00
MODULE_PARM_DESC(check_copy_up, "Obsolete; does nothing");
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
int ovl_copy_xattr(struct dentry *old, struct dentry *new)
{
ssize_t list_size, size, value_size = 0;
char *buf, *name, *value = NULL;
int uninitialized_var(error);
size_t slen;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
if (!(old->d_inode->i_opflags & IOP_XATTR) ||
!(new->d_inode->i_opflags & IOP_XATTR))
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
return 0;
list_size = vfs_listxattr(old, NULL, 0);
if (list_size <= 0) {
if (list_size == -EOPNOTSUPP)
return 0;
return list_size;
}
buf = kzalloc(list_size, GFP_KERNEL);
if (!buf)
return -ENOMEM;
list_size = vfs_listxattr(old, buf, list_size);
if (list_size <= 0) {
error = list_size;
goto out;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
}
for (name = buf; list_size; name += slen) {
slen = strnlen(name, list_size) + 1;
/* underlying fs providing us with an broken xattr list? */
if (WARN_ON(slen > list_size)) {
error = -EIO;
break;
}
list_size -= slen;
if (ovl_is_private_xattr(name))
continue;
retry:
size = vfs_getxattr(old, name, value, value_size);
if (size == -ERANGE)
size = vfs_getxattr(old, name, NULL, 0);
if (size < 0) {
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
error = size;
break;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
}
if (size > value_size) {
void *new;
new = krealloc(value, size, GFP_KERNEL);
if (!new) {
error = -ENOMEM;
break;
}
value = new;
value_size = size;
goto retry;
}
error = security_inode_copy_up_xattr(name);
if (error < 0 && error != -EOPNOTSUPP)
break;
if (error == 1) {
error = 0;
continue; /* Discard */
}
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
error = vfs_setxattr(new, name, value, size, 0);
if (error)
break;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
}
kfree(value);
out:
kfree(buf);
return error;
}
static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
{
struct file *old_file;
struct file *new_file;
loff_t old_pos = 0;
loff_t new_pos = 0;
loff_t cloned;
loff_t data_pos = -1;
loff_t hole_len;
bool skip_hole = false;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
int error = 0;
if (len == 0)
return 0;
old_file = ovl_path_open(old, O_LARGEFILE | O_RDONLY);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
if (IS_ERR(old_file))
return PTR_ERR(old_file);
new_file = ovl_path_open(new, O_LARGEFILE | O_WRONLY);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
if (IS_ERR(new_file)) {
error = PTR_ERR(new_file);
goto out_fput;
}
/* Try to use clone_file_range to clone up within the same fs */
cloned = do_clone_file_range(old_file, 0, new_file, 0, len, 0);
if (cloned == len)
goto out;
/* Couldn't clone, so now we try to copy the data */
/* Check if lower fs supports seek operation */
if (old_file->f_mode & FMODE_LSEEK &&
old_file->f_op->llseek)
skip_hole = true;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
while (len) {
size_t this_len = OVL_COPY_UP_CHUNK_SIZE;
long bytes;
if (len < this_len)
this_len = len;
if (signal_pending_state(TASK_KILLABLE, current)) {
error = -EINTR;
break;
}
/*
* Fill zero for hole will cost unnecessary disk space
* and meanwhile slow down the copy-up speed, so we do
* an optimization for hole during copy-up, it relies
* on SEEK_DATA implementation in lower fs so if lower
* fs does not support it, copy-up will behave as before.
*
* Detail logic of hole detection as below:
* When we detect next data position is larger than current
* position we will skip that hole, otherwise we copy
* data in the size of OVL_COPY_UP_CHUNK_SIZE. Actually,
* it may not recognize all kind of holes and sometimes
* only skips partial of hole area. However, it will be
* enough for most of the use cases.
*/
if (skip_hole && data_pos < old_pos) {
data_pos = vfs_llseek(old_file, old_pos, SEEK_DATA);
if (data_pos > old_pos) {
hole_len = data_pos - old_pos;
len -= hole_len;
old_pos = new_pos = data_pos;
continue;
} else if (data_pos == -ENXIO) {
break;
} else if (data_pos < 0) {
skip_hole = false;
}
}
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
bytes = do_splice_direct(old_file, &old_pos,
new_file, &new_pos,
this_len, SPLICE_F_MOVE);
if (bytes <= 0) {
error = bytes;
break;
}
WARN_ON(old_pos != new_pos);
len -= bytes;
}
out:
if (!error)
error = vfs_fsync(new_file, 0);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
fput(new_file);
out_fput:
fput(old_file);
return error;
}
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
static int ovl_set_size(struct dentry *upperdentry, struct kstat *stat)
{
struct iattr attr = {
.ia_valid = ATTR_SIZE,
.ia_size = stat->size,
};
return notify_change(upperdentry, &attr, NULL);
}
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
static int ovl_set_timestamps(struct dentry *upperdentry, struct kstat *stat)
{
struct iattr attr = {
.ia_valid =
ATTR_ATIME | ATTR_MTIME | ATTR_ATIME_SET | ATTR_MTIME_SET,
.ia_atime = stat->atime,
.ia_mtime = stat->mtime,
};
return notify_change(upperdentry, &attr, NULL);
}
int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat)
{
int err = 0;
if (!S_ISLNK(stat->mode)) {
struct iattr attr = {
.ia_valid = ATTR_MODE,
.ia_mode = stat->mode,
};
err = notify_change(upperdentry, &attr, NULL);
}
if (!err) {
struct iattr attr = {
.ia_valid = ATTR_UID | ATTR_GID,
.ia_uid = stat->uid,
.ia_gid = stat->gid,
};
err = notify_change(upperdentry, &attr, NULL);
}
if (!err)
ovl_set_timestamps(upperdentry, stat);
return err;
}
struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper)
{
struct ovl_fh *fh;
int fh_type, dwords;
int buflen = MAX_HANDLE_SZ;
uuid_t *uuid = &real->d_sb->s_uuid;
int err;
/* Make sure the real fid stays 32bit aligned */
BUILD_BUG_ON(OVL_FH_FID_OFFSET % 4);
BUILD_BUG_ON(MAX_HANDLE_SZ + OVL_FH_FID_OFFSET > 255);
fh = kzalloc(buflen + OVL_FH_FID_OFFSET, GFP_KERNEL);
if (!fh)
return ERR_PTR(-ENOMEM);
/*
* We encode a non-connectable file handle for non-dir, because we
* only need to find the lower inode number and we don't want to pay
* the price or reconnecting the dentry.
*/
dwords = buflen >> 2;
fh_type = exportfs_encode_fh(real, (void *)fh->fb.fid, &dwords, 0);
buflen = (dwords << 2);
err = -EIO;
if (WARN_ON(fh_type < 0) ||
WARN_ON(buflen > MAX_HANDLE_SZ) ||
WARN_ON(fh_type == FILEID_INVALID))
goto out_err;
fh->fb.version = OVL_FH_VERSION;
fh->fb.magic = OVL_FH_MAGIC;
fh->fb.type = fh_type;
fh->fb.flags = OVL_FH_FLAG_CPU_ENDIAN;
/*
* When we will want to decode an overlay dentry from this handle
* and all layers are on the same fs, if we get a disconncted real
* dentry when we decode fid, the only way to tell if we should assign
* it to upperdentry or to lowerstack is by checking this flag.
*/
if (is_upper)
fh->fb.flags |= OVL_FH_FLAG_PATH_UPPER;
fh->fb.len = sizeof(fh->fb) + buflen;
fh->fb.uuid = *uuid;
return fh;
out_err:
kfree(fh);
return ERR_PTR(err);
}
int ovl_set_origin(struct dentry *dentry, struct dentry *lower,
struct dentry *upper)
{
const struct ovl_fh *fh = NULL;
int err;
/*
* When lower layer doesn't support export operations store a 'null' fh,
* so we can use the overlay.origin xattr to distignuish between a copy
* up and a pure upper inode.
*/
if (ovl_can_decode_fh(lower->d_sb)) {
fh = ovl_encode_real_fh(lower, false);
if (IS_ERR(fh))
return PTR_ERR(fh);
}
/*
* Do not fail when upper doesn't support xattrs.
*/
err = ovl_check_setxattr(dentry, upper, OVL_XATTR_ORIGIN, fh->buf,
fh ? fh->fb.len : 0, 0);
kfree(fh);
return err;
}
/* Store file handle of @upper dir in @index dir entry */
static int ovl_set_upper_fh(struct dentry *upper, struct dentry *index)
{
const struct ovl_fh *fh;
int err;
fh = ovl_encode_real_fh(upper, true);
if (IS_ERR(fh))
return PTR_ERR(fh);
err = ovl_do_setxattr(index, OVL_XATTR_UPPER, fh->buf, fh->fb.len, 0);
kfree(fh);
return err;
}
/*
* Create and install index entry.
*
* Caller must hold i_mutex on indexdir.
*/
static int ovl_create_index(struct dentry *dentry, struct dentry *origin,
struct dentry *upper)
{
struct dentry *indexdir = ovl_indexdir(dentry->d_sb);
struct inode *dir = d_inode(indexdir);
struct dentry *index = NULL;
struct dentry *temp = NULL;
struct qstr name = { };
int err;
/*
* For now this is only used for creating index entry for directories,
* because non-dir are copied up directly to index and then hardlinked
* to upper dir.
*
* TODO: implement create index for non-dir, so we can call it when
* encoding file handle for non-dir in case index does not exist.
*/
if (WARN_ON(!d_is_dir(dentry)))
return -EIO;
/* Directory not expected to be indexed before copy up */
if (WARN_ON(ovl_test_flag(OVL_INDEX, d_inode(dentry))))
return -EIO;
err = ovl_get_index_name(origin, &name);
if (err)
return err;
temp = ovl_create_temp(indexdir, OVL_CATTR(S_IFDIR | 0));
err = PTR_ERR(temp);
if (IS_ERR(temp))
goto free_name;
err = ovl_set_upper_fh(upper, temp);
if (err)
goto out;
index = lookup_one_len(name.name, indexdir, name.len);
if (IS_ERR(index)) {
err = PTR_ERR(index);
} else {
err = ovl_do_rename(dir, temp, dir, index, 0);
dput(index);
}
out:
if (err)
ovl_cleanup(dir, temp);
dput(temp);
free_name:
kfree(name.name);
return err;
}
struct ovl_copy_up_ctx {
struct dentry *parent;
struct dentry *dentry;
struct path lowerpath;
struct kstat stat;
struct kstat pstat;
const char *link;
struct dentry *destdir;
struct qstr destname;
struct dentry *workdir;
bool origin;
bool indexed;
bool metacopy;
};
static int ovl_link_up(struct ovl_copy_up_ctx *c)
{
int err;
struct dentry *upper;
struct dentry *upperdir = ovl_dentry_upper(c->parent);
struct inode *udir = d_inode(upperdir);
/* Mark parent "impure" because it may now contain non-pure upper */
err = ovl_set_impure(c->parent, upperdir);
if (err)
return err;
err = ovl_set_nlink_lower(c->dentry);
if (err)
return err;
inode_lock_nested(udir, I_MUTEX_PARENT);
upper = lookup_one_len(c->dentry->d_name.name, upperdir,
c->dentry->d_name.len);
err = PTR_ERR(upper);
if (!IS_ERR(upper)) {
err = ovl_do_link(ovl_dentry_upper(c->dentry), udir, upper);
dput(upper);
if (!err) {
/* Restore timestamps on parent (best effort) */
ovl_set_timestamps(upperdir, &c->pstat);
ovl_dentry_set_upper_alias(c->dentry);
}
}
inode_unlock(udir);
if (err)
return err;
err = ovl_set_nlink_upper(c->dentry);
return err;
}
static int ovl_copy_up_inode(struct ovl_copy_up_ctx *c, struct dentry *temp)
{
int err;
2019-01-12 02:37:00 +08:00
/*
* Copy up data first and then xattrs. Writing data after
* xattrs will remove security.capability xattr automatically.
*/
if (S_ISREG(c->stat.mode) && !c->metacopy) {
struct path upperpath, datapath;
ovl_path_upper(c->dentry, &upperpath);
if (WARN_ON(upperpath.dentry != NULL))
return -EIO;
upperpath.dentry = temp;
ovl_path_lowerdata(c->dentry, &datapath);
err = ovl_copy_up_data(&datapath, &upperpath, c->stat.size);
if (err)
return err;
}
err = ovl_copy_xattr(c->lowerpath.dentry, temp);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
if (err)
return err;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
/*
* Store identifier of lower inode in upper inode xattr to
* allow lookup of the copy up origin inode.
*
* Don't set origin when we are breaking the association with a lower
* hard link.
*/
if (c->origin) {
err = ovl_set_origin(c->dentry, c->lowerpath.dentry, temp);
if (err)
return err;
}
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
if (c->metacopy) {
err = ovl_check_setxattr(c->dentry, temp, OVL_XATTR_METACOPY,
NULL, 0, -EOPNOTSUPP);
if (err)
return err;
}
inode_lock(temp->d_inode);
if (S_ISREG(c->stat.mode))
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
err = ovl_set_size(temp, &c->stat);
if (!err)
err = ovl_set_attr(temp, &c->stat);
inode_unlock(temp->d_inode);
return err;
}
struct ovl_cu_creds {
const struct cred *old;
struct cred *new;
};
static int ovl_prep_cu_creds(struct dentry *dentry, struct ovl_cu_creds *cc)
{
int err;
cc->old = cc->new = NULL;
err = security_inode_copy_up(dentry, &cc->new);
if (err < 0)
return err;
if (cc->new)
cc->old = override_creds(cc->new);
return 0;
}
static void ovl_revert_cu_creds(struct ovl_cu_creds *cc)
{
if (cc->new) {
revert_creds(cc->old);
put_cred(cc->new);
}
}
/*
* Copyup using workdir to prepare temp file. Used when copying up directories,
* special files or when upper fs doesn't support O_TMPFILE.
*/
static int ovl_copy_up_workdir(struct ovl_copy_up_ctx *c)
{
struct inode *inode;
struct inode *udir = d_inode(c->destdir), *wdir = d_inode(c->workdir);
struct dentry *temp, *upper;
struct ovl_cu_creds cc;
int err;
struct ovl_cattr cattr = {
/* Can't properly set mode on creation because of the umask */
.mode = c->stat.mode & S_IFMT,
.rdev = c->stat.rdev,
.link = c->link
};
err = ovl_lock_rename_workdir(c->workdir, c->destdir);
if (err)
return err;
err = ovl_prep_cu_creds(c->dentry, &cc);
if (err)
goto unlock;
temp = ovl_create_temp(c->workdir, &cattr);
ovl_revert_cu_creds(&cc);
err = PTR_ERR(temp);
if (IS_ERR(temp))
goto unlock;
err = ovl_copy_up_inode(c, temp);
if (err)
goto cleanup;
if (S_ISDIR(c->stat.mode) && c->indexed) {
err = ovl_create_index(c->dentry, c->lowerpath.dentry, temp);
if (err)
goto cleanup;
}
upper = lookup_one_len(c->destname.name, c->destdir, c->destname.len);
err = PTR_ERR(upper);
if (IS_ERR(upper))
goto cleanup;
err = ovl_do_rename(wdir, temp, udir, upper, 0);
dput(upper);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
if (err)
goto cleanup;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
if (!c->metacopy)
ovl_set_upperdata(d_inode(c->dentry));
inode = d_inode(c->dentry);
ovl_inode_update(inode, temp);
if (S_ISDIR(inode->i_mode))
ovl_set_flag(OVL_WHITEOUTS, inode);
unlock:
unlock_rename(c->workdir, c->destdir);
return err;
cleanup:
ovl_cleanup(wdir, temp);
dput(temp);
goto unlock;
}
/* Copyup using O_TMPFILE which does not require cross dir locking */
static int ovl_copy_up_tmpfile(struct ovl_copy_up_ctx *c)
{
struct inode *udir = d_inode(c->destdir);
struct dentry *temp, *upper;
struct ovl_cu_creds cc;
int err;
err = ovl_prep_cu_creds(c->dentry, &cc);
if (err)
return err;
temp = ovl_do_tmpfile(c->workdir, c->stat.mode);
ovl_revert_cu_creds(&cc);
if (IS_ERR(temp))
return PTR_ERR(temp);
err = ovl_copy_up_inode(c, temp);
if (err)
goto out_dput;
inode_lock_nested(udir, I_MUTEX_PARENT);
upper = lookup_one_len(c->destname.name, c->destdir, c->destname.len);
err = PTR_ERR(upper);
if (!IS_ERR(upper)) {
err = ovl_do_link(temp, udir, upper);
dput(upper);
}
inode_unlock(udir);
if (err)
goto out_dput;
if (!c->metacopy)
ovl_set_upperdata(d_inode(c->dentry));
ovl_inode_update(d_inode(c->dentry), temp);
return 0;
out_dput:
dput(temp);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
return err;
}
/*
* Copy up a single dentry
*
* All renames start with copy up of source if necessary. The actual
* rename will only proceed once the copy up was successful. Copy up uses
* upper parent i_mutex for exclusion. Since rename can change d_parent it
* is possible that the copy up will lock the old parent. At that point
* the file will have already been copied up anyway.
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
*/
static int ovl_do_copy_up(struct ovl_copy_up_ctx *c)
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
{
int err;
struct ovl_fs *ofs = c->dentry->d_sb->s_fs_info;
bool to_index = false;
/*
* Indexed non-dir is copied up directly to the index entry and then
* hardlinked to upper dir. Indexed dir is copied up to indexdir,
* then index entry is created and then copied up dir installed.
* Copying dir up to indexdir instead of workdir simplifies locking.
*/
if (ovl_need_index(c->dentry)) {
c->indexed = true;
if (S_ISDIR(c->stat.mode))
c->workdir = ovl_indexdir(c->dentry->d_sb);
else
to_index = true;
}
if (S_ISDIR(c->stat.mode) || c->stat.nlink == 1 || to_index)
c->origin = true;
if (to_index) {
c->destdir = ovl_indexdir(c->dentry->d_sb);
err = ovl_get_index_name(c->lowerpath.dentry, &c->destname);
if (err)
return err;
} else if (WARN_ON(!c->parent)) {
/* Disconnected dentry must be copied up to index dir */
return -EIO;
} else {
/*
* Mark parent "impure" because it may now contain non-pure
* upper
*/
err = ovl_set_impure(c->parent, c->destdir);
if (err)
return err;
}
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
/* Should we copyup with O_TMPFILE or with workdir? */
if (S_ISREG(c->stat.mode) && ofs->tmpfile)
err = ovl_copy_up_tmpfile(c);
else
err = ovl_copy_up_workdir(c);
if (err)
goto out;
if (c->indexed)
ovl_set_flag(OVL_INDEX, d_inode(c->dentry));
if (to_index) {
/* Initialize nlink for copy up of disconnected dentry */
err = ovl_set_nlink_upper(c->dentry);
} else {
struct inode *udir = d_inode(c->destdir);
/* Restore timestamps on parent (best effort) */
inode_lock(udir);
ovl_set_timestamps(c->destdir, &c->pstat);
inode_unlock(udir);
ovl_dentry_set_upper_alias(c->dentry);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
}
out:
if (to_index)
kfree(c->destname.name);
return err;
}
static bool ovl_need_meta_copy_up(struct dentry *dentry, umode_t mode,
int flags)
{
struct ovl_fs *ofs = dentry->d_sb->s_fs_info;
if (!ofs->config.metacopy)
return false;
if (!S_ISREG(mode))
return false;
if (flags && ((OPEN_FMODE(flags) & FMODE_WRITE) || (flags & O_TRUNC)))
return false;
return true;
}
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
/* Copy up data of an inode which was copied up metadata only in the past. */
static int ovl_copy_up_meta_inode_data(struct ovl_copy_up_ctx *c)
{
struct path upperpath, datapath;
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
int err;
char *capability = NULL;
ssize_t uninitialized_var(cap_size);
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
ovl_path_upper(c->dentry, &upperpath);
if (WARN_ON(upperpath.dentry == NULL))
return -EIO;
ovl_path_lowerdata(c->dentry, &datapath);
if (WARN_ON(datapath.dentry == NULL))
return -EIO;
if (c->stat.size) {
err = cap_size = ovl_getxattr(upperpath.dentry, XATTR_NAME_CAPS,
&capability, 0);
if (err < 0 && err != -ENODATA)
goto out;
}
err = ovl_copy_up_data(&datapath, &upperpath, c->stat.size);
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
if (err)
goto out_free;
/*
* Writing to upper file will clear security.capability xattr. We
* don't want that to happen for normal copy-up operation.
*/
if (capability) {
err = ovl_do_setxattr(upperpath.dentry, XATTR_NAME_CAPS,
capability, cap_size, 0);
if (err)
goto out_free;
}
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
err = vfs_removexattr(upperpath.dentry, OVL_XATTR_METACOPY);
if (err)
goto out_free;
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
ovl_set_upperdata(d_inode(c->dentry));
out_free:
kfree(capability);
out:
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
return err;
}
static int ovl_copy_up_one(struct dentry *parent, struct dentry *dentry,
int flags)
{
int err;
DEFINE_DELAYED_CALL(done);
struct path parentpath;
struct ovl_copy_up_ctx ctx = {
.parent = parent,
.dentry = dentry,
.workdir = ovl_workdir(dentry),
};
if (WARN_ON(!ctx.workdir))
return -EROFS;
ovl_path_lower(dentry, &ctx.lowerpath);
err = vfs_getattr(&ctx.lowerpath, &ctx.stat,
STATX_BASIC_STATS, AT_STATX_SYNC_AS_STAT);
if (err)
return err;
ctx.metacopy = ovl_need_meta_copy_up(dentry, ctx.stat.mode, flags);
if (parent) {
ovl_path_upper(parent, &parentpath);
ctx.destdir = parentpath.dentry;
ctx.destname = dentry->d_name;
err = vfs_getattr(&parentpath, &ctx.pstat,
STATX_ATIME | STATX_MTIME,
AT_STATX_SYNC_AS_STAT);
if (err)
return err;
}
/* maybe truncate regular file. this has no effect on dirs */
if (flags & O_TRUNC)
ctx.stat.size = 0;
if (S_ISLNK(ctx.stat.mode)) {
ctx.link = vfs_get_link(ctx.lowerpath.dentry, &done);
if (IS_ERR(ctx.link))
return PTR_ERR(ctx.link);
}
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
err = ovl_copy_up_start(dentry, flags);
/* err < 0: interrupted, err > 0: raced with another copy-up */
if (unlikely(err)) {
if (err > 0)
err = 0;
} else {
if (!ovl_dentry_upper(dentry))
err = ovl_do_copy_up(&ctx);
if (!err && parent && !ovl_dentry_has_upper_alias(dentry))
err = ovl_link_up(&ctx);
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
if (!err && ovl_dentry_needs_data_copy_up_locked(dentry, flags))
err = ovl_copy_up_meta_inode_data(&ctx);
ovl_copy_up_end(dentry);
}
do_delayed_call(&done);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
return err;
}
int ovl_copy_up_flags(struct dentry *dentry, int flags)
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
{
int err = 0;
const struct cred *old_cred = ovl_override_creds(dentry->d_sb);
bool disconnected = (dentry->d_flags & DCACHE_DISCONNECTED);
/*
* With NFS export, copy up can get called for a disconnected non-dir.
* In this case, we will copy up lower inode to index dir without
* linking it to upper dir.
*/
if (WARN_ON(disconnected && d_is_dir(dentry)))
return -EIO;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
while (!err) {
struct dentry *next;
struct dentry *parent = NULL;
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
if (ovl_already_copied_up(dentry, flags))
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
break;
next = dget(dentry);
/* find the topmost dentry not yet copied up */
for (; !disconnected;) {
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
parent = dget_parent(next);
if (ovl_dentry_upper(parent))
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
break;
dput(next);
next = parent;
}
err = ovl_copy_up_one(parent, next, flags);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
dput(parent);
dput(next);
}
revert_creds(old_cred);
overlay filesystem Overlayfs allows one, usually read-write, directory tree to be overlaid onto another, read-only directory tree. All modifications go to the upper, writable layer. This type of mechanism is most often used for live CDs but there's a wide variety of other uses. The implementation differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases. The dentry tree is duplicated from the underlying filesystems, this enables fast cached lookups without adding special support into the VFS. This uses slightly more memory than union mounts, but dentries are relatively small. Currently inodes are duplicated as well, but it is a possible optimization to share inodes for non-directories. Opening non directories results in the open forwarded to the underlying filesystem. This makes the behavior very similar to union mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file descriptors). Usage: mount -t overlayfs overlayfs -olowerdir=/lower,upperdir=/upper/upper,workdir=/upper/work /overlay The following cotributions have been folded into this patch: Neil Brown <neilb@suse.de>: - minimal remount support - use correct seek function for directories - initialise is_real before use - rename ovl_fill_cache to ovl_dir_read Felix Fietkau <nbd@openwrt.org>: - fix a deadlock in ovl_dir_read_merged - fix a deadlock in ovl_remove_whiteouts Erez Zadok <ezk@fsl.cs.sunysb.edu> - fix cleanup after WARN_ON Sedat Dilek <sedat.dilek@googlemail.com> - fix up permission to confirm to new API Robin Dong <hao.bigrat@gmail.com> - fix possible leak in ovl_new_inode - create new inode in ovl_link Andy Whitcroft <apw@canonical.com> - switch to __inode_permission() - copy up i_uid/i_gid from the underlying inode AV: - ovl_copy_up_locked() - dput(ERR_PTR(...)) on two failure exits - ovl_clear_empty() - one failure exit forgetting to do unlock_rename(), lack of check for udir being the parent of upper, dropping and regaining the lock on udir (which would require _another_ check for parent being right). - bogus d_drop() in copyup and rename [fix from your mail] - copyup/remove and copyup/rename races [fix from your mail] - ovl_dir_fsync() leaving ERR_PTR() in ->realfile - ovl_entry_free() is pointless - it's just a kfree_rcu() - fold ovl_do_lookup() into ovl_lookup() - manually assigning ->d_op is wrong. Just use ->s_d_op. [patches picked from Miklos]: * copyup/remove and copyup/rename races * bogus d_drop() in copyup and rename Also thanks to the following people for testing and reporting bugs: Jordi Pujol <jordipujolp@gmail.com> Andy Whitcroft <apw@canonical.com> Michal Suchanek <hramrach@centrum.cz> Felix Fietkau <nbd@openwrt.org> Erez Zadok <ezk@fsl.cs.sunysb.edu> Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
2014-10-24 06:14:38 +08:00
return err;
}
static bool ovl_open_need_copy_up(struct dentry *dentry, int flags)
{
/* Copy up of disconnected dentry does not set upper alias */
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
if (ovl_already_copied_up(dentry, flags))
return false;
if (special_file(d_inode(dentry)->i_mode))
return false;
ovl: A new xattr OVL_XATTR_METACOPY for file on upper Now we will have the capability to have upper inodes which might be only metadata copy up and data is still on lower inode. So add a new xattr OVL_XATTR_METACOPY to distinguish between two cases. Presence of OVL_XATTR_METACOPY reflects that file has been copied up metadata only and and data will be copied up later from lower origin. So this xattr is set when a metadata copy takes place and cleared when data copy takes place. We also use a bit in ovl_inode->flags to cache OVL_UPPERDATA which reflects whether ovl inode has data or not (as opposed to metadata only copy up). If a file is copied up metadata only and later when same file is opened for WRITE, then data copy up takes place. We copy up data, remove METACOPY xattr and then set the UPPERDATA flag in ovl_inode->flags. While all these operations happen with oi->lock held, read side of oi->flags can be lockless. That is another thread on another cpu can check if UPPERDATA flag is set or not. So this gives us an ordering requirement w.r.t UPPERDATA flag. That is, if another cpu sees UPPERDATA flag set, then it should be guaranteed that effects of data copy up and remove xattr operations are also visible. For example. CPU1 CPU2 ovl_open() acquire(oi->lock) ovl_open_maybe_copy_up() ovl_copy_up_data() open_open_need_copy_up() vfs_removexattr() ovl_already_copied_up() ovl_dentry_needs_data_copy_up() ovl_set_flag(OVL_UPPERDATA) ovl_test_flag(OVL_UPPERDATA) release(oi->lock) Say CPU2 is copying up data and in the end sets UPPERDATA flag. But if CPU1 perceives the effects of setting UPPERDATA flag but not the effects of preceding operations (ex. upper that is not fully copied up), it will be a problem. Hence this patch introduces smp_wmb() on setting UPPERDATA flag operation and smp_rmb() on UPPERDATA flag test operation. May be some other lock or barrier is already covering it. But I am not sure what that is and is it obvious enough that we will not break it in future. So hence trying to be safe here and introducing barriers explicitly for UPPERDATA flag/bit. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-05-11 23:49:28 +08:00
if (!ovl_open_flags_need_copy_up(flags))
return false;
return true;
}
int ovl_maybe_copy_up(struct dentry *dentry, int flags)
{
int err = 0;
if (ovl_open_need_copy_up(dentry, flags)) {
err = ovl_want_write(dentry);
if (!err) {
err = ovl_copy_up_flags(dentry, flags);
ovl_drop_write(dentry);
}
}
return err;
}
int ovl_copy_up_with_data(struct dentry *dentry)
{
return ovl_copy_up_flags(dentry, O_WRONLY);
}
int ovl_copy_up(struct dentry *dentry)
{
return ovl_copy_up_flags(dentry, 0);
}