Pull directory locking updates from Christian Brauner:
"This contains the work to add centralized APIs for directory locking
operations.
This series is part of a larger effort to change directory operation
locking to allow multiple concurrent operations in a directory. The
ultimate goal is to lock the target dentry(s) rather than the whole
parent directory.
To help with changing the locking protocol, this series centralizes
locking and lookup in new helper functions. The helpers establish a
pattern where it is the dentry that is being locked and unlocked
(currently the lock is held on dentry->d_parent->d_inode, but that can
change in the future).
This also changes vfs_mkdir() to unlock the parent on failure, as well
as dput()ing the dentry. This allows end_creating() to only require
the target dentry (which may be IS_ERR() after vfs_mkdir()), not the
parent"
* tag 'vfs-6.19-rc1.directory.locking' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
nfsd: fix end_creating() conversion
VFS: introduce end_creating_keep()
VFS: change vfs_mkdir() to unlock on failure.
ecryptfs: use new start_creating/start_removing APIs
Add start_renaming_two_dentries()
VFS/ovl/smb: introduce start_renaming_dentry()
VFS/nfsd/ovl: introduce start_renaming() and end_renaming()
VFS: add start_creating_killable() and start_removing_killable()
VFS: introduce start_removing_dentry()
smb/server: use end_removing_noperm for for target of smb2_create_link()
VFS: introduce start_creating_noperm() and start_removing_noperm()
VFS/nfsd/cachefiles/ovl: introduce start_removing() and end_removing()
VFS/nfsd/cachefiles/ovl: add start_creating() and end_creating()
VFS: tidy up do_unlinkat()
VFS: introduce start_dirop() and end_dirop()
debugfs: rename end_creating() to debugfs_end_creating()
vfs_mkdir() already drops the reference to the dentry on failure but it
leaves the parent locked.
This complicates end_creating() which needs to unlock the parent even
though the dentry is no longer available.
If we change vfs_mkdir() to unlock on failure as well as releasing the
dentry, we can remove the "parent" arg from end_creating() and simplify
the rules for calling it.
Note that cachefiles_get_directory() can choose to substitute an error
instead of actually calling vfs_mkdir(), for fault injection. In that
case it needs to call end_creating(), just as vfs_mkdir() now does on
error.
ovl_create_real() will now unlock on error. So the conditional
end_creating() after the call is removed, and end_creating() is called
internally on error.
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Tested-by: syzbot@syzkaller.appspotmail.com
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://patch.msgid.link/20251113002050.676694-15-neilb@ownmail.net
Signed-off-by: Christian Brauner <brauner@kernel.org>
A few callers want to lock for a rename and already have both dentries.
Also debugfs does want to perform a lookup but doesn't want permission
checking, so start_renaming_dentry() cannot be used.
This patch introduces start_renaming_two_dentries() which is given both
dentries. debugfs performs one lookup itself. As it will only continue
with a negative dentry and as those cannot be renamed or unlinked, it is
safe to do the lookup before getting the rename locks.
overlayfs uses start_renaming_two_dentries() in three places and selinux
uses it twice in sel_make_policy_nodes().
In sel_make_policy_nodes() we now lock for rename twice instead of just
once so the combined operation is no longer atomic w.r.t the parent
directory locks. As selinux_state.policy_mutex is held across the whole
operation this does not open up any interesting races.
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://patch.msgid.link/20251113002050.676694-13-neilb@ownmail.net
Signed-off-by: Christian Brauner <brauner@kernel.org>
Several callers perform a rename on a dentry they already have, and only
require lookup for the target name. This includes smb/server and a few
different places in overlayfs.
start_renaming_dentry() performs the required lookup and takes the
required lock using lock_rename_child()
It is used in three places in overlayfs and in ksmbd_vfs_rename().
In the ksmbd case, the parent of the source is not important - the
source must be renamed from wherever it is. So start_renaming_dentry()
allows rd->old_parent to be NULL and only checks it if it is non-NULL.
On success rd->old_parent will be the parent of old_dentry with an extra
reference taken. Other start_renaming function also now take the extra
reference and end_renaming() now drops this reference as well.
ovl_lookup_temp(), ovl_parent_lock(), and ovl_parent_unlock() are
all removed as they are no longer needed.
OVL_TEMPNAME_SIZE and ovl_tempname() are now declared in overlayfs.h so
that ovl_check_rename_whiteout() can access them.
ovl_copy_up_workdir() now always cleans up on error.
Reviewed-by: Namjae Jeon <linkinjeon@kernel.org>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://patch.msgid.link/20251113002050.676694-12-neilb@ownmail.net
Tested-by: syzbot@syzkaller.appspotmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
start_renaming() combines name lookup and locking to prepare for rename.
It is used when two names need to be looked up as in nfsd and overlayfs -
cases where one or both dentries are already available will be handled
separately.
__start_renaming() avoids the inode_permission check and hash
calculation and is suitable after filename_parentat() in do_renameat2().
It subsumes quite a bit of code from that function.
start_renaming() does calculate the hash and check X permission and is
suitable elsewhere:
- nfsd_rename()
- ovl_rename()
In ovl, ovl_do_rename_rd() is factored out of ovl_do_rename(), which
itself will be gone by the end of the series.
Acked-by: Chuck Lever <chuck.lever@oracle.com> (for nfsd parts)
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: NeilBrown <neil@brown.name>
--
Changes since v3:
- added missig dput() in ovl_rename when "whiteout" is not-NULL.
Changes since v2:
- in __start_renaming() some label have been renamed, and err
is always set before a "goto out_foo" rather than passing the
error in a dentry*.
- ovl_do_rename() changed to call the new ovl_do_rename_rd() rather
than keeping duplicate code
- code around ovl_cleanup() call in ovl_rename() restructured.
Link: https://patch.msgid.link/20251113002050.676694-11-neilb@ownmail.net
Tested-by: syzbot@syzkaller.appspotmail.com
Acked-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
start_removing_dentry() is similar to start_removing() but instead of
providing a name for lookup, the target dentry is given.
start_removing_dentry() checks that the dentry is still hashed and in
the parent, and if so it locks and increases the refcount so that
end_removing() can be used to finish the operation.
This is used in cachefiles, overlayfs, smb/server, and apparmor.
There will be other users including ecryptfs.
As start_removing_dentry() takes an extra reference to the dentry (to be
put by end_removing()), there is no need to explicitly take an extra
reference to stop d_delete() from using dentry_unlink_inode() to negate
the dentry - as in cachefiles_delete_object(), and ksmbd_vfs_unlink().
cachefiles_bury_object() now gets an extra ref to the victim, which is
drops. As it includes the needed end_removing() calls, the caller
doesn't need them.
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Namjae Jeon <linkinjeon@kernel.org>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://patch.msgid.link/20251113002050.676694-9-neilb@ownmail.net
Signed-off-by: Christian Brauner <brauner@kernel.org>
xfs, fuse, ipc/mqueue need variants of start_creating or start_removing
which do not check permissions.
This patch adds _noperm versions of these functions.
Note that do_mq_open() was only calling mntget() so it could call
path_put() - it didn't really need an extra reference on the mnt.
Now it doesn't call mntget() and uses end_creating() which does
the dput() half of path_put().
Also mq_unlink() previously passed
d_inode(dentry->d_parent)
as the dir inode to vfs_unlink(). This is after locking
d_inode(mnt->mnt_root)
These two inodes are the same, but normally calls use the textual
parent.
So I've changes the vfs_unlink() call to be given d_inode(mnt->mnt_root).
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: NeilBrown <neil@brown.name>
--
changes since v2:
- dir arg passed to vfs_unlink() in mq_unlink() changed to match
the dir passed to lookup_noperm()
- restore assignment to path->mnt even though the mntget() is removed.
Link: https://patch.msgid.link/20251113002050.676694-7-neilb@ownmail.net
Tested-by: syzbot@syzkaller.appspotmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
start_removing() is similar to start_creating() but will only return a
positive dentry with the expectation that it will be removed. This is
used by nfsd, cachefiles, and overlayfs. They are changed to also use
end_removing() to terminate the action begun by start_removing(). This
is a simple alias for end_dirop().
Apart from changes to the error paths, as we no longer need to unlock on
a lookup error, an effect on callers is that they don't need to test if
the found dentry is positive or negative - they can be sure it is
positive.
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://patch.msgid.link/20251113002050.676694-6-neilb@ownmail.net
Tested-by: syzbot@syzkaller.appspotmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
start_creating() is similar to simple_start_creating() but is not so
simple.
It takes a qstr for the name, includes permission checking, and does NOT
report an error if the name already exists, returning a positive dentry
instead.
This is currently used by nfsd, cachefiles, and overlayfs.
end_creating() is called after the dentry has been used.
end_creating() drops the reference to the dentry as it is generally no
longer needed. This is exactly the first section of end_creating_path()
so that function is changed to call the new end_creating()
These calls help encapsulate locking rules so that directory locking can
be changed.
Occasionally this change means that the parent lock is held for a
shorter period of time, for example in cachefiles_commit_tmpfile().
As this function now unlocks after an unlink and before the following
lookup, it is possible that the lookup could again find a positive
dentry, so a while loop is introduced there.
In overlayfs the ovl_lookup_temp() function has ovl_tempname()
split out to be used in ovl_start_creating_temp(). The other use
of ovl_lookup_temp() is preparing for a rename. When rename handling
is updated, ovl_lookup_temp() will be removed.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://patch.msgid.link/20251113002050.676694-5-neilb@ownmail.net
Tested-by: syzbot@syzkaller.appspotmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
Pull file->f_path constification from Al Viro:
"Only one thing was modifying ->f_path of an opened file - acct(2).
Massaging that away and constifying a bunch of struct path * arguments
in functions that might be given &file->f_path ends up with the
situation where we can turn ->f_path into an anon union of const
struct path f_path and struct path __f_path, the latter modified only
in a few places in fs/{file_table,open,namei}.c, all for struct file
instances that are yet to be opened"
* tag 'pull-f_path' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (23 commits)
Have cc(1) catch attempts to modify ->f_path
kernel/acct.c: saner struct file treatment
configfs:get_target() - release path as soon as we grab configfs_item reference
apparmor/af_unix: constify struct path * arguments
ovl_is_real_file: constify realpath argument
ovl_sync_file(): constify path argument
ovl_lower_dir(): constify path argument
ovl_get_verity_digest(): constify path argument
ovl_validate_verity(): constify {meta,data}path arguments
ovl_ensure_verity_loaded(): constify datapath argument
ksmbd_vfs_set_init_posix_acl(): constify path argument
ksmbd_vfs_inherit_posix_acl(): constify path argument
ksmbd_vfs_kern_path_unlock(): constify path argument
ksmbd_vfs_path_lookup_locked(): root_share_path can be const struct path *
check_export(): constify path argument
export_operations->open(): constify path argument
rqst_exp_get_by_name(): constify path argument
nfs: constify path argument of __vfs_getattr()
bpf...d_path(): constify path argument
done_path_create(): constify path argument
...
kern_path_locked() is now only used to prepare for removing an object
from the filesystem (and that is the only credible reason for wanting a
positive locked dentry). Thus it corresponds to kern_path_create() and
so should have a corresponding name.
Unfortunately the name "kern_path_create" is somewhat misleading as it
doesn't actually create anything. The recently added
simple_start_creating() provides a better pattern I believe. The
"start" can be matched with "end" to bracket the creating or removing.
So this patch changes names:
kern_path_locked -> start_removing_path
kern_path_create -> start_creating_path
user_path_create -> start_creating_user_path
user_path_locked_at -> start_removing_user_path_at
done_path_create -> end_creating_path
and also introduces end_removing_path() which is identical to
end_creating_path().
__start_removing_path (which was __kern_path_locked) is enhanced to
call mnt_want_write() for consistency with the start_creating_path().
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Christian Brauner <brauner@kernel.org>
audit_alloc_mark() and audit_get_nd() both need to perform a path
lookup getting the parent dentry (which must exist) and the final
target (following a LAST_NORM name) which sometimes doesn't need to
exist.
They don't need the parent to be locked, but use kern_path_locked() or
kern_path_locked_negative() anyway. This is somewhat misleading to the
casual reader.
This patch introduces a more targeted function, kern_path_parent(),
which returns not holding locks. On success the "path" will
be set to the parent, which must be found, and the return value is the
dentry of the target, which might be negative.
This will clear the way to rename kern_path_locked() which is
otherwise only used to prepare for removing something.
It also allows us to remove kern_path_locked_negative(), which is
transformed into the new kern_path_parent().
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Christian Brauner <brauner@kernel.org>
ovl wants a lookup which won't block on a fatal signal. It currently
uses down_write_killable() and then repeatedly calls to lookup_one()
The lock may not be needed if the name is already in the dcache and it
aids proposed future changes if the locking is kept internal to namei.c
So this patch adds lookup_one_positive_killable() which is like
lookup_one_positive() but will abort in the face of a fatal signal.
overlayfs is changed to use this.
Note that instead of always getting an exclusive lock, ovl now only gets
a shared lock, and only sometimes. The exclusive lock was never needed.
However down_read_killable() was only added in v4.15 but overlayfs started
using down_write_killable() here in v4.7.
Note that the linked list ->first_maybe_whiteout ->next_maybe_white is
local to the thread so there is no concurrency in that list which could
be threatened by removing the locking.
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Pull vfs directory lookup updates from Christian Brauner:
"This contains cleanups for the lookup_one*() family of helpers.
We expose a set of functions with names containing "lookup_one_len"
and others without the "_len". This difference has nothing to do with
"len". It's rater a historical accident that can be confusing.
The functions without "_len" take a "mnt_idmap" pointer. This is found
in the "vfsmount" and that is an important question when choosing
which to use: do you have a vfsmount, or are you "inside" the
filesystem. A related question is "is permission checking relevant
here?".
nfsd and cachefiles *do* have a vfsmount but *don't* use the non-_len
functions. They pass nop_mnt_idmap and refuse to work on filesystems
which have any other idmap.
This work changes nfsd and cachefile to use the lookup_one family of
functions and to explictily pass &nop_mnt_idmap which is consistent
with all other vfs interfaces used where &nop_mnt_idmap is explicitly
passed.
The remaining uses of the "_one" functions do not require permission
checks so these are renamed to be "_noperm" and the permission
checking is removed.
This series also changes these lookup function to take a qstr instead
of separate name and len. In many cases this simplifies the call"
* tag 'vfs-6.16-rc1.async.dir' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
VFS: change lookup_one_common and lookup_noperm_common to take a qstr
Use try_lookup_noperm() instead of d_hash_and_lookup() outside of VFS
VFS: rename lookup_one_len family to lookup_noperm and remove permission check
cachefiles: Use lookup_one() rather than lookup_one_len()
nfsd: Use lookup_one() rather than lookup_one_len()
VFS: improve interface for lookup_one functions
The audit code relies on the fact that kern_path_locked() returned a
path even for a negative dentry. If it doesn't find a valid dentry it
immediately calls:
audit_find_parent(d_backing_inode(parent_path.dentry));
which assumes that parent_path.dentry is still valid. But it isn't since
kern_path_locked() has been changed to path_put() also for a negative
dentry.
Fix this by adding a helper that implements the required audit semantics
and allows us to fix the immediate bleeding. We can find a unified
solution for this afterwards.
Link: https://lore.kernel.org/20250414-rennt-wimmeln-f186c3a780f1@brauner
Fixes: 1c3cb50b58 ("VFS: change kern_path_locked() and user_path_locked_at() to never return negative dentry")
Reported-and-tested-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
The lookup_one_len family of functions is (now) only used internally by
a filesystem on itself either
- in a context where permission checking is irrelevant such as by a
virtual filesystem populating itself, or xfs accessing its ORPHANAGE
or dquota accessing the quota file; or
- in a context where a permission check (MAY_EXEC on the parent) has just
been performed such as a network filesystem finding in "silly-rename"
file in the same directory. This is also the context after the
_parentat() functions where currently lookup_one_qstr_excl() is used.
So the permission check is pointless.
The name "one_len" is unhelpful in understanding the purpose of these
functions and should be changed. Most of the callers pass the len as
"strlen()" so using a qstr and QSTR() can simplify the code.
This patch renames these functions (include lookup_positive_unlocked()
which is part of the family despite the name) to have a name based on
"lookup_noperm". They are changed to receive a 'struct qstr' instead
of separate name and len. In a few cases the use of QSTR() results in a
new call to strlen().
try_lookup_noperm() takes a pointer to a qstr instead of the whole
qstr. This is consistent with d_hash_and_lookup() (which is nearly
identical) and useful for lookup_noperm_unlocked().
The new lookup_noperm_common() doesn't take a qstr yet. That will be
tidied up in a subsequent patch.
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://lore.kernel.org/r/20250319031545.2999807-5-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
The family of functions:
lookup_one()
lookup_one_unlocked()
lookup_one_positive_unlocked()
appear designed to be used by external clients of the filesystem rather
than by filesystems acting on themselves as the lookup_one_len family
are used.
They are used by:
btrfs/ioctl - which is a user-space interface rather than an internal
activity
exportfs - i.e. from nfsd or the open_by_handle_at interface
overlayfs - at access the underlying filesystems
smb/server - for file service
They should be used by nfsd (more than just the exportfs path) and
cachefs but aren't.
It would help if the documentation didn't claim they should "not be
called by generic code".
Also the path component name is passed as "name" and "len" which are
(confusingly?) separate by the "base". In some cases the len in simply
"strlen" and so passing a qstr using QSTR() would make the calling
clearer.
Other callers do pass separate name and len which are stored in a
struct. Sometimes these are already stored in a qstr, other times it
easily could be.
So this patch changes these three functions to receive a 'struct qstr *',
and improves the documentation.
QSTR_LEN() is added to make it easy to pass a QSTR containing a known
len.
[brauner@kernel.org: take a struct qstr pointer]
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://lore.kernel.org/r/20250319031545.2999807-2-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
The LOOKUP_ bits are not in order, which can make it awkward when adding
new bits. Two bits have recently been added to the end which makes them
look like "scoping flags", but in fact they aren't.
Also LOOKUP_PARENT is described as "internal use only" but is used in
fs/nfs/
This patch:
- Moves these three flags into the "pathwalk mode" section
- changes all bits to use the BIT(n) macro
- Allocates bits in order leaving gaps between the sections,
and documents those gaps.
Signed-off-by: NeilBrown <neilb@suse.de>
Link: https://lore.kernel.org/r/20250206054504.2950516-8-neilb@suse.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
"The definition of insanity is doing the same thing over and over
again and expecting different results”
We've tried to do this before, most recently with commit bb2314b479
("fs: Allow unprivileged linkat(..., AT_EMPTY_PATH) aka flink") about a
decade ago.
But the effort goes back even further than that, eg this thread back
from 1998 that is so old that we don't even have it archived in lore:
https://lkml.org/lkml/1998/3/10/108
which also points out some of the reasons why it's dangerous.
Or, how about then in 2003:
https://lkml.org/lkml/2003/4/6/112
where we went through some of the same arguments, just wirh different
people involved.
In particular, having access to a file descriptor does not necessarily
mean that you have access to the path that was used for lookup, and
there may be very good reasons why you absolutely must not have access
to a path to said file.
For example, if we were passed a file descriptor from the outside into
some limited environment (think chroot, but also user namespaces etc) a
'flink()' system call could now make that file visible inside a context
where it's not supposed to be visible.
In the process the user may also be able to re-open it with permissions
that the original file descriptor did not have (eg a read-only file
descriptor may be associated with an underlying file that is writable).
Another variation on this is if somebody else (typically root) opens a
file in a directory that is not accessible to others, and passes the
file descriptor on as a read-only file. Again, the access to the file
descriptor does not imply that you should have access to a path to the
file in the filesystem.
So while we have tried this several times in the past, it never works.
The last time we did this, that commit bb2314b479 quickly got reverted
again in commit f0cc6ffb8c (Revert "fs: Allow unprivileged linkat(...,
AT_EMPTY_PATH) aka flink"), with a note saying "We may re-do this once
the whole discussion about the interface is done".
Well, the discussion is long done, and didn't come to any resolution.
There's no question that 'flink()' would be a useful operation, but it's
a dangerous one.
However, it does turn out that since 2008 (commit d76b0d9b2d: "CRED:
Use creds in file structs") we have had a fairly straightforward way to
check whether the file descriptor was opened by the same credentials as
the credentials of the flink().
That allows the most common patterns that people want to use, which tend
to be to either open the source carefully (ie using the openat2()
RESOLVE_xyz flags, and/or checking ownership with fstat() before
linking), or to use O_TMPFILE and fill in the file contents before it's
exposed to the world with linkat().
But it also means that if the file descriptor was opened by somebody
else, or we've gone through a credentials change since, the operation no
longer works (unless we have CAP_DAC_READ_SEARCH capabilities in the
opener's user namespace, as before).
Note that the credential equality check is done by using pointer
equality, which means that it's not enough that you have effectively the
same user - they have to be literally identical, since our credentials
are using copy-on-write semantics.
So you can't change your credentials to something else and try to change
it back to the same ones between the open() and the linkat(). This is
not meant to be some kind of generic permission check, this is literally
meant as a "the open and link calls are 'atomic' wrt user credentials"
check.
It also means that you can't just move things between namespaces,
because the credentials aren't just a list of uid's and gid's: they
includes the pointer to the user_ns that the capabilities are relative
to.
So let's try this one more time and see if maybe this approach ends up
being workable after all.
Cc: Andrew Lutomirski <luto@kernel.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Peter Anvin <hpa@zytor.com>
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20240411001012.12513-1-torvalds@linux-foundation.org
[brauner: relax capability check to opener of the file]
Link: https://lore.kernel.org/all/20231113-undenkbar-gediegen-efde5f1c34bc@brauner
Signed-off-by: Christian Brauner <brauner@kernel.org>
SB_POSIXACL must be set when a filesystem supports POSIX ACLs, but NFSv4
also sets this flag to prevent the VFS from applying the umask on
newly-created files. NFSv4 doesn't support POSIX ACLs however, which
causes confusion when other subsystems try to test for them.
Add a new SB_I_NOUMASK flag that allows filesystems to opt-in to umask
stripping without advertising support for POSIX ACLs. Set the new flag
on NFSv4 instead of SB_POSIXACL.
Also, move mode_strip_umask to namei.h and convert init_mknod and
init_mkdir to use it.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Message-Id: <20230911-acl-fix-v3-1-b25315333f6c@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Al pointed out that ksmbd has racy issue from using ->d_parent and ->d_name
in ksmbd_vfs_unlink and smb2_vfs_rename(). and use new lock_rename_child()
to lock stable parent while underlying rename racy.
Introduce vfs_path_parent_lookup helper to avoid out of share access and
export vfs functions like the following ones to use
vfs_path_parent_lookup().
- rename __lookup_hash() to lookup_one_qstr_excl().
- export lookup_one_qstr_excl().
- export getname_kernel() and putname().
vfs_path_parent_lookup() is used for parent lookup of destination file
using absolute pathname given from FILE_RENAME_INFORMATION request.
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Pass the dentry of a source file and the dentry of a destination directory
to lock parent inodes for rename. As soon as this function returns,
->d_parent of the source file dentry is stable and inodes are properly
locked for calling vfs-rename. This helper is needed for ksmbd server.
rename request of SMB protocol has to rename an opened file, no matter
which directory it's in.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull nfsd updates from Chuck Lever:
"Two significant security enhancements are part of this release:
- NFSD's RPC header encoding and decoding, including RPCSEC GSS and
gssproxy header parsing, has been overhauled to make it more
memory-safe.
- Support for Kerberos AES-SHA2-based encryption types has been added
for both the NFS client and server. This provides a clean path for
deprecating and removing insecure encryption types based on DES and
SHA-1. AES-SHA2 is also FIPS-140 compliant, so that NFS with
Kerberos may now be used on systems with fips enabled.
In addition to these, NFSD is now able to handle crossing into an
auto-mounted mount point on an exported NFS mount. A number of fixes
have been made to NFSD's server-side copy implementation.
RPC metrics have been converted to per-CPU variables. This helps
reduce unnecessary cross-CPU and cross-node memory bus traffic, and
significantly reduces noise when KCSAN is enabled"
* tag 'nfsd-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux: (121 commits)
NFSD: Clean up nfsd_symlink()
NFSD: copy the whole verifier in nfsd_copy_write_verifier
nfsd: don't fsync nfsd_files on last close
SUNRPC: Fix occasional warning when destroying gss_krb5_enctypes
nfsd: fix courtesy client with deny mode handling in nfs4_upgrade_open
NFSD: fix problems with cleanup on errors in nfsd4_copy
nfsd: fix race to check ls_layouts
nfsd: don't hand out delegation on setuid files being opened for write
SUNRPC: Remove ->xpo_secure_port()
SUNRPC: Clean up the svc_xprt_flags() macro
nfsd: remove fs/nfsd/fault_inject.c
NFSD: fix leaked reference count of nfsd4_ssc_umount_item
nfsd: clean up potential nfsd_file refcount leaks in COPY codepath
nfsd: zero out pointers after putting nfsd_files on COPY setup error
SUNRPC: Fix whitespace damage in svcauth_unix.c
nfsd: eliminate __nfs4_get_fd
nfsd: add some kerneldoc comments for stateid preprocessing functions
nfsd: eliminate find_deleg_file_locked
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS
SUNRPC: Add encryption self-tests
...
This function is only used by NFSD to cross mount points.
If a mount point is of type auto mount, follow_down() will
not uncover it. Add LOOKUP_AUTOMOUNT to the lookup flags
to have ->d_automount() called when NFSD walks down the
mount tree.
Signed-off-by: Richard Weinberger <richard@nod.at>
Reviewed-by: Ian Kent <raven@themaw.net>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Convert to struct mnt_idmap.
Last cycle we merged the necessary infrastructure in
256c8aed2b ("fs: introduce dedicated idmap type for mounts").
This is just the conversion to struct mnt_idmap.
Currently we still pass around the plain namespace that was attached to a
mount. This is in general pretty convenient but it makes it easy to
conflate namespaces that are relevant on the filesystem with namespaces
that are relevent on the mount level. Especially for non-vfs developers
without detailed knowledge in this area this can be a potential source for
bugs.
Once the conversion to struct mnt_idmap is done all helpers down to the
really low-level helpers will take a struct mnt_idmap argument instead of
two namespace arguments. This way it becomes impossible to conflate the two
eliminating the possibility of any bugs. All of the vfs and all filesystems
only operate on struct mnt_idmap.
Acked-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Similar to the addition of lookup_one() add a version of
lookup_one_unlocked() and lookup_one_positive_unlocked() that take
idmapped mounts into account. This is required to port overlay to
support idmapped base layers.
Cc: <linux-fsdevel@vger.kernel.org>
Tested-by: Giuseppe Scrivano <gscrivan@redhat.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Various filesystems rely on the lookup_one_len() helper to lookup a
single path component relative to a well-known starting point. Allow
such filesystems to support idmapped mounts by adding a version of this
helper to take the idmap into account when calling inode_permission().
This change is a required to let btrfs (and other filesystems) support
idmapped mounts.
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Separate field in nameidata (nd->state) holding the flags that
should be internal-only - that way we both get some spare bits
in LOOKUP_... and get simpler rules for nd->root lifetime rules,
since we can set the replacement of LOOKUP_ROOT (ND_ROOT_PRESET)
at the same time we set nd->root.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
io_uring always punts opens to async context, since there's no control
over whether the lookup blocks or not. Add LOOKUP_CACHED to support
just doing the fast RCU based lookups, which we know will not block. If
we can do a cached path resolution of the filename, then we don't have
to always punt lookups for a worker.
During path resolution, we always do LOOKUP_RCU first. If that fails and
we terminate LOOKUP_RCU, then fail a LOOKUP_CACHED attempt as well.
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
->last_type values are set in 3 places: path_init() (sets to LAST_ROOT),
link_path_walk (LAST_NORM/DOT/DOTDOT) and pick_link (LAST_BIND).
The are checked in walk_component(), lookup_last() and do_last().
They also get copied to the caller by filename_parentat(). In the last
3 cases the value is what we had at the return from link_path_walk().
In case of walk_component() it's either directly downstream from
assignment in link_path_walk() or, when called by lookup_last(), the
value we have at the return from link_path_walk().
The value at the entry into link_path_walk() can survive to return only
if the pathname contains nothing but slashes. Note that pick_link()
never returns such - pure jumps are handled directly. So for the calls
of link_path_walk() for trailing symlinks it does not matter what value
had been there at the entry; the value at the return won't depend upon it.
There are 3 call chains that might have pick_link() storing LAST_BIND:
1) pick_link() from step_into() from walk_component() from
link_path_walk(). In that case we will either be parsing the next
component immediately after return into link_path_walk(), which will
overwrite the ->last_type before anyone has a chance to look at it,
or we'll fail, in which case nobody will be looking at ->last_type at all.
2) pick_link() from step_into() from walk_component() from lookup_last().
The value is never looked at due to the above; it won't affect the value
seen at return from any link_path_walk().
3) pick_link() from step_into() from do_last(). Ditto.
In other words, assignemnt in pick_link() is pointless, and so is
LAST_BIND itself; nothing ever looks at that value. Kill it off.
And make link_path_walk() _always_ assign ->last_type - in the only
case when the value at the entry might survive to the return that value
is always LAST_ROOT, inherited from path_init(). Move that assignment
from path_init() into the beginning of link_path_walk(), to consolidate
the things.
Historical note: LAST_BIND used to be used for the kludge with trailing
pure jump symlinks (extra iteration through the top-level loop).
No point keeping it anymore...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
New LOOKUP flag, telling path_lookupat() to act as path_mountpointat().
IOW, traverse mounts at the final point and skip revalidation of the
location where it ends up.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull openat2 support from Al Viro:
"This is the openat2() series from Aleksa Sarai.
I'm afraid that the rest of namei stuff will have to wait - it got
zero review the last time I'd posted #work.namei, and there had been a
leak in the posted series I'd caught only last weekend. I was going to
repost it on Monday, but the window opened and the odds of getting any
review during that... Oh, well.
Anyway, openat2 part should be ready; that _did_ get sane amount of
review and public testing, so here it comes"
From Aleksa's description of the series:
"For a very long time, extending openat(2) with new features has been
incredibly frustrating. This stems from the fact that openat(2) is
possibly the most famous counter-example to the mantra "don't silently
accept garbage from userspace" -- it doesn't check whether unknown
flags are present[1].
This means that (generally) the addition of new flags to openat(2) has
been fraught with backwards-compatibility issues (O_TMPFILE has to be
defined as __O_TMPFILE|O_DIRECTORY|[O_RDWR or O_WRONLY] to ensure old
kernels gave errors, since it's insecure to silently ignore the
flag[2]). All new security-related flags therefore have a tough road
to being added to openat(2).
Furthermore, the need for some sort of control over VFS's path
resolution (to avoid malicious paths resulting in inadvertent
breakouts) has been a very long-standing desire of many userspace
applications.
This patchset is a revival of Al Viro's old AT_NO_JUMPS[3] patchset
(which was a variant of David Drysdale's O_BENEATH patchset[4] which
was a spin-off of the Capsicum project[5]) with a few additions and
changes made based on the previous discussion within [6] as well as
others I felt were useful.
In line with the conclusions of the original discussion of
AT_NO_JUMPS, the flag has been split up into separate flags. However,
instead of being an openat(2) flag it is provided through a new
syscall openat2(2) which provides several other improvements to the
openat(2) interface (see the patch description for more details). The
following new LOOKUP_* flags are added:
LOOKUP_NO_XDEV:
Blocks all mountpoint crossings (upwards, downwards, or through
absolute links). Absolute pathnames alone in openat(2) do not
trigger this. Magic-link traversal which implies a vfsmount jump is
also blocked (though magic-link jumps on the same vfsmount are
permitted).
LOOKUP_NO_MAGICLINKS:
Blocks resolution through /proc/$pid/fd-style links. This is done
by blocking the usage of nd_jump_link() during resolution in a
filesystem. The term "magic-links" is used to match with the only
reference to these links in Documentation/, but I'm happy to change
the name.
It should be noted that this is different to the scope of
~LOOKUP_FOLLOW in that it applies to all path components. However,
you can do openat2(NO_FOLLOW|NO_MAGICLINKS) on a magic-link and it
will *not* fail (assuming that no parent component was a
magic-link), and you will have an fd for the magic-link.
In order to correctly detect magic-links, the introduction of a new
LOOKUP_MAGICLINK_JUMPED state flag was required.
LOOKUP_BENEATH:
Disallows escapes to outside the starting dirfd's
tree, using techniques such as ".." or absolute links. Absolute
paths in openat(2) are also disallowed.
Conceptually this flag is to ensure you "stay below" a certain
point in the filesystem tree -- but this requires some additional
to protect against various races that would allow escape using
"..".
Currently LOOKUP_BENEATH implies LOOKUP_NO_MAGICLINKS, because it
can trivially beam you around the filesystem (breaking the
protection). In future, there might be similar safety checks done
as in LOOKUP_IN_ROOT, but that requires more discussion.
In addition, two new flags are added that expand on the above ideas:
LOOKUP_NO_SYMLINKS:
Does what it says on the tin. No symlink resolution is allowed at
all, including magic-links. Just as with LOOKUP_NO_MAGICLINKS this
can still be used with NOFOLLOW to open an fd for the symlink as
long as no parent path had a symlink component.
LOOKUP_IN_ROOT:
This is an extension of LOOKUP_BENEATH that, rather than blocking
attempts to move past the root, forces all such movements to be
scoped to the starting point. This provides chroot(2)-like
protection but without the cost of a chroot(2) for each filesystem
operation, as well as being safe against race attacks that
chroot(2) is not.
If a race is detected (as with LOOKUP_BENEATH) then an error is
generated, and similar to LOOKUP_BENEATH it is not permitted to
cross magic-links with LOOKUP_IN_ROOT.
The primary need for this is from container runtimes, which
currently need to do symlink scoping in userspace[7] when opening
paths in a potentially malicious container.
There is a long list of CVEs that could have bene mitigated by
having RESOLVE_THIS_ROOT (such as CVE-2017-1002101,
CVE-2017-1002102, CVE-2018-15664, and CVE-2019-5736, just to name a
few).
In order to make all of the above more usable, I'm working on
libpathrs[8] which is a C-friendly library for safe path resolution.
It features a userspace-emulated backend if the kernel doesn't support
openat2(2). Hopefully we can get userspace to switch to using it, and
thus get openat2(2) support for free once it's ready.
Future work would include implementing things like
RESOLVE_NO_AUTOMOUNT and possibly a RESOLVE_NO_REMOTE (to allow
programs to be sure they don't hit DoSes though stale NFS handles)"
* 'work.openat2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
Documentation: path-lookup: include new LOOKUP flags
selftests: add openat2(2) selftests
open: introduce openat2(2) syscall
namei: LOOKUP_{IN_ROOT,BENEATH}: permit limited ".." resolution
namei: LOOKUP_IN_ROOT: chroot-like scoped resolution
namei: LOOKUP_BENEATH: O_BENEATH-like scoped resolution
namei: LOOKUP_NO_XDEV: block mountpoint crossing
namei: LOOKUP_NO_MAGICLINKS: block magic-link resolution
namei: LOOKUP_NO_SYMLINKS: block symlink resolution
namei: allow set_root() to produce errors
namei: allow nd_jump_link() to produce errors
nsfs: clean-up ns_get_path() signature to return int
namei: only return -ECHILD from follow_dotdot_rcu()
... and get rid of a bunch of bugs in it. Background:
the reason for path_mountpoint() is that umount() really doesn't
want attempts to revalidate the root of what it's trying to umount.
The thing we want to avoid actually happen from complete_walk();
solution was to do something parallel to normal path_lookupat()
and it both went overboard and got the boilerplate subtly
(and not so subtly) wrong.
A better solution is to do pretty much what the normal path_lookupat()
does, but instead of complete_walk() do unlazy_walk(). All it takes
to avoid that ->d_weak_revalidate() call... mountpoint_last() goes
away, along with everything it got wrong, and so does the magic around
LOOKUP_NO_REVAL.
Another source of bugs is that when we traverse mounts at the final
location (and we need to do that - umount . expects to get whatever's
overmounting ., if any, out of the lookup) we really ought to take
care of ->d_manage() - as it is, manual umount of autofs automount
in progress can lead to unpleasant surprises for the daemon. Easily
solved by using handle_lookup_down() instead of follow_mount().
Tested-by: Ian Kent <raven@themaw.net>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
/* Background. */
Container runtimes or other administrative management processes will
often interact with root filesystems while in the host mount namespace,
because the cost of doing a chroot(2) on every operation is too
prohibitive (especially in Go, which cannot safely use vfork). However,
a malicious program can trick the management process into doing
operations on files outside of the root filesystem through careful
crafting of symlinks.
Most programs that need this feature have attempted to make this process
safe, by doing all of the path resolution in userspace (with symlinks
being scoped to the root of the malicious root filesystem).
Unfortunately, this method is prone to foot-guns and usually such
implementations have subtle security bugs.
Thus, what userspace needs is a way to resolve a path as though it were
in a chroot(2) -- with all absolute symlinks being resolved relative to
the dirfd root (and ".." components being stuck under the dirfd root).
It is much simpler and more straight-forward to provide this
functionality in-kernel (because it can be done far more cheaply and
correctly).
More classical applications that also have this problem (which have
their own potentially buggy userspace path sanitisation code) include
web servers, archive extraction tools, network file servers, and so on.
/* Userspace API. */
LOOKUP_IN_ROOT will be exposed to userspace through openat2(2).
/* Semantics. */
Unlike most other LOOKUP flags (most notably LOOKUP_FOLLOW),
LOOKUP_IN_ROOT applies to all components of the path.
With LOOKUP_IN_ROOT, any path component which attempts to cross the
starting point of the pathname lookup (the dirfd passed to openat) will
remain at the starting point. Thus, all absolute paths and symlinks will
be scoped within the starting point.
There is a slight change in behaviour regarding pathnames -- if the
pathname is absolute then the dirfd is still used as the root of
resolution of LOOKUP_IN_ROOT is specified (this is to avoid obvious
foot-guns, at the cost of a minor API inconsistency).
As with LOOKUP_BENEATH, Jann's security concern about ".."[1] applies to
LOOKUP_IN_ROOT -- therefore ".." resolution is blocked. This restriction
will be lifted in a future patch, but requires more work to ensure that
permitting ".." is done safely.
Magic-link jumps are also blocked, because they can beam the path lookup
across the starting point. It would be possible to detect and block
only the "bad" crossings with path_is_under() checks, but it's unclear
whether it makes sense to permit magic-links at all. However, userspace
is recommended to pass LOOKUP_NO_MAGICLINKS if they want to ensure that
magic-link crossing is entirely disabled.
/* Testing. */
LOOKUP_IN_ROOT is tested as part of the openat2(2) selftests.
[1]: https://lore.kernel.org/lkml/CAG48ez1jzNvxB+bfOBnERFGp=oMM0vHWuLD6EULmne3R6xa53w@mail.gmail.com/
Cc: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
/* Background. */
There are many circumstances when userspace wants to resolve a path and
ensure that it doesn't go outside of a particular root directory during
resolution. Obvious examples include archive extraction tools, as well as
other security-conscious userspace programs. FreeBSD spun out O_BENEATH
from their Capsicum project[1,2], so it also seems reasonable to
implement similar functionality for Linux.
This is part of a refresh of Al's AT_NO_JUMPS patchset[3] (which was a
variation on David Drysdale's O_BENEATH patchset[4], which in turn was
based on the Capsicum project[5]).
/* Userspace API. */
LOOKUP_BENEATH will be exposed to userspace through openat2(2).
/* Semantics. */
Unlike most other LOOKUP flags (most notably LOOKUP_FOLLOW),
LOOKUP_BENEATH applies to all components of the path.
With LOOKUP_BENEATH, any path component which attempts to "escape" the
starting point of the filesystem lookup (the dirfd passed to openat)
will yield -EXDEV. Thus, all absolute paths and symlinks are disallowed.
Due to a security concern brought up by Jann[6], any ".." path
components are also blocked. This restriction will be lifted in a future
patch, but requires more work to ensure that permitting ".." is done
safely.
Magic-link jumps are also blocked, because they can beam the path lookup
across the starting point. It would be possible to detect and block
only the "bad" crossings with path_is_under() checks, but it's unclear
whether it makes sense to permit magic-links at all. However, userspace
is recommended to pass LOOKUP_NO_MAGICLINKS if they want to ensure that
magic-link crossing is entirely disabled.
/* Testing. */
LOOKUP_BENEATH is tested as part of the openat2(2) selftests.
[1]: https://reviews.freebsd.org/D2808
[2]: https://reviews.freebsd.org/D17547
[3]: https://lore.kernel.org/lkml/20170429220414.GT29622@ZenIV.linux.org.uk/
[4]: https://lore.kernel.org/lkml/1415094884-18349-1-git-send-email-drysdale@google.com/
[5]: https://lore.kernel.org/lkml/1404124096-21445-1-git-send-email-drysdale@google.com/
[6]: https://lore.kernel.org/lkml/CAG48ez1jzNvxB+bfOBnERFGp=oMM0vHWuLD6EULmne3R6xa53w@mail.gmail.com/
Cc: Christian Brauner <christian.brauner@ubuntu.com>
Suggested-by: David Drysdale <drysdale@google.com>
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
/* Background. */
The need to contain path operations within a mountpoint has been a
long-standing usecase that userspace has historically implemented
manually with liberal usage of stat(). find, rsync, tar and
many other programs implement these semantics -- but it'd be much
simpler to have a fool-proof way of refusing to open a path if it
crosses a mountpoint.
This is part of a refresh of Al's AT_NO_JUMPS patchset[1] (which was a
variation on David Drysdale's O_BENEATH patchset[2], which in turn was
based on the Capsicum project[3]).
/* Userspace API. */
LOOKUP_NO_XDEV will be exposed to userspace through openat2(2).
/* Semantics. */
Unlike most other LOOKUP flags (most notably LOOKUP_FOLLOW),
LOOKUP_NO_XDEV applies to all components of the path.
With LOOKUP_NO_XDEV, any path component which crosses a mount-point
during path resolution (including "..") will yield an -EXDEV. Absolute
paths, absolute symlinks, and magic-links will only yield an -EXDEV if
the jump involved changing mount-points.
/* Testing. */
LOOKUP_NO_XDEV is tested as part of the openat2(2) selftests.
[1]: https://lore.kernel.org/lkml/20170429220414.GT29622@ZenIV.linux.org.uk/
[2]: https://lore.kernel.org/lkml/1415094884-18349-1-git-send-email-drysdale@google.com/
[3]: https://lore.kernel.org/lkml/1404124096-21445-1-git-send-email-drysdale@google.com/
Cc: Christian Brauner <christian.brauner@ubuntu.com>
Suggested-by: David Drysdale <drysdale@google.com>
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
/* Background. */
There has always been a special class of symlink-like objects in procfs
(and a few other pseudo-filesystems) which allow for non-lexical
resolution of paths using nd_jump_link(). These "magic-links" do not
follow traditional mount namespace boundaries, and have been used
consistently in container escape attacks because they can be used to
trick unsuspecting privileged processes into resolving unexpected paths.
It is also non-trivial for userspace to unambiguously avoid resolving
magic-links, because they do not have a reliable indication that they
are a magic-link (in order to verify them you'd have to manually open
the path given by readlink(2) and then verify that the two file
descriptors reference the same underlying file, which is plagued with
possible race conditions or supplementary attack scenarios).
It would therefore be very helpful for userspace to be able to avoid
these symlinks easily, thus hopefully removing a tool from attackers'
toolboxes.
This is part of a refresh of Al's AT_NO_JUMPS patchset[1] (which was a
variation on David Drysdale's O_BENEATH patchset[2], which in turn was
based on the Capsicum project[3]).
/* Userspace API. */
LOOKUP_NO_MAGICLINKS will be exposed to userspace through openat2(2).
/* Semantics. */
Unlike most other LOOKUP flags (most notably LOOKUP_FOLLOW),
LOOKUP_NO_MAGICLINKS applies to all components of the path.
With LOOKUP_NO_MAGICLINKS, any magic-link path component encountered
during path resolution will yield -ELOOP. The handling of ~LOOKUP_FOLLOW
for a trailing magic-link is identical to LOOKUP_NO_SYMLINKS.
LOOKUP_NO_SYMLINKS implies LOOKUP_NO_MAGICLINKS.
/* Testing. */
LOOKUP_NO_MAGICLINKS is tested as part of the openat2(2) selftests.
[1]: https://lore.kernel.org/lkml/20170429220414.GT29622@ZenIV.linux.org.uk/
[2]: https://lore.kernel.org/lkml/1415094884-18349-1-git-send-email-drysdale@google.com/
[3]: https://lore.kernel.org/lkml/1404124096-21445-1-git-send-email-drysdale@google.com/
Cc: Christian Brauner <christian.brauner@ubuntu.com>
Suggested-by: David Drysdale <drysdale@google.com>
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
/* Background. */
Userspace cannot easily resolve a path without resolving symlinks, and
would have to manually resolve each path component with O_PATH and
O_NOFOLLOW. This is clearly inefficient, and can be fairly easy to screw
up (resulting in possible security bugs). Linus has mentioned that Git
has a particular need for this kind of flag[1]. It also resolves a
fairly long-standing perceived deficiency in O_NOFOLLOw -- that it only
blocks the opening of trailing symlinks.
This is part of a refresh of Al's AT_NO_JUMPS patchset[2] (which was a
variation on David Drysdale's O_BENEATH patchset[3], which in turn was
based on the Capsicum project[4]).
/* Userspace API. */
LOOKUP_NO_SYMLINKS will be exposed to userspace through openat2(2).
/* Semantics. */
Unlike most other LOOKUP flags (most notably LOOKUP_FOLLOW),
LOOKUP_NO_SYMLINKS applies to all components of the path.
With LOOKUP_NO_SYMLINKS, any symlink path component encountered during
path resolution will yield -ELOOP. If the trailing component is a
symlink (and no other components were symlinks), then O_PATH|O_NOFOLLOW
will not error out and will instead provide a handle to the trailing
symlink -- without resolving it.
/* Testing. */
LOOKUP_NO_SYMLINKS is tested as part of the openat2(2) selftests.
[1]: https://lore.kernel.org/lkml/CA+55aFyOKM7DW7+0sdDFKdZFXgptb5r1id9=Wvhd8AgSP7qjwQ@mail.gmail.com/
[2]: https://lore.kernel.org/lkml/20170429220414.GT29622@ZenIV.linux.org.uk/
[3]: https://lore.kernel.org/lkml/1415094884-18349-1-git-send-email-drysdale@google.com/
[4]: https://lore.kernel.org/lkml/1404124096-21445-1-git-send-email-drysdale@google.com/
Cc: Christian Brauner <christian.brauner@ubuntu.com>
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
In preparation for LOOKUP_NO_MAGICLINKS, it's necessary to add the
ability for nd_jump_link() to return an error which the corresponding
get_link() caller must propogate back up to the VFS.
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Most of the callers of lookup_one_len_unlocked() treat negatives are
ERR_PTR(-ENOENT). Provide a helper that would do just that. Note
that a pinned positive dentry remains positive - it's ->d_inode is
stable, etc.; a pinned _negative_ dentry can become positive at any
point as long as you are not holding its parent at least shared.
So using lookup_one_len_unlocked() needs to be careful;
lookup_positive_unlocked() is safer and that's what the callers
end up open-coding anyway.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
