Minki/linux
1
0
Fork 1
mirror of https://github.com/torvalds/linux.git synced 2024-11-22 04:02:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

fs.idmapped.mnt_idmap.v6.2

Browse Source 
-----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCY5by6AAKCRCRxhvAZXjc
 onblAPsFzodV8/9UoCIkKxwn0aiclbiAITTWI9ZLulmKhm0I6wD/RUOLKjt12uZJ
 m81UTfkWHopWKtQ+X3saZEcyYTNLugE=
 =AtGb
 -----END PGP SIGNATURE-----

Merge tag 'fs.idmapped.mnt_idmap.v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/idmapping

Pull idmapping updates from Christian Brauner:
 "Last cycle we've already made the interaction with idmapped mounts
  more robust and type safe by introducing the vfs{g,u}id_t type. This
  cycle we concluded the conversion and removed the legacy helpers.

  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
  filesystem developers without detailed knowledge in this area this can
  be a potential source for bugs.

  Instead of passing the plain namespace we introduce a dedicated type
  struct mnt_idmap and replace the pointer with a pointer to a struct
  mnt_idmap. There are no semantic or size changes for the mount struct
  caused by this.

  We then start converting all places aware of idmapped mounts to rely
  on struct mnt_idmap. Once the conversion is done all helpers down to
  the really low-level make_vfs{g,u}id() and from_vfs{g,u}id() will take
  a struct mnt_idmap argument instead of two namespace arguments. This
  way it becomes impossible to conflate the two removing and thus
  eliminating the possibility of any bugs. Fwiw, I fixed some issues in
  that area a while ago in ntfs3 and ksmbd in the past. Afterwards only
  low-level code can ultimately use the associated namespace for any
  permission checks. Even most of the vfs can be completely obivious
  about this ultimately and filesystems will never interact with it in
  any form in the future.

  A struct mnt_idmap currently encompasses a simple refcount and pointer
  to the relevant namespace the mount is idmapped to. If a mount isn't
  idmapped then it will point to a static nop_mnt_idmap and if it
  doesn't that it is idmapped. As usual there are no allocations or
  anything happening for non-idmapped mounts. Everthing is carefully
  written to be a nop for non-idmapped mounts as has always been the
  case.

  If an idmapped mount is created a struct mnt_idmap is allocated and a
  reference taken on the relevant namespace. Each mount that gets
  idmapped or inherits the idmap simply bumps the reference count on
  struct mnt_idmap. Just a reminder that we only allow a mount to change
  it's idmapping a single time and only if it hasn't already been
  attached to the filesystems and has no active writers.

  The actual changes are fairly straightforward but this will have huge
  benefits for maintenance and security in the long run even if it
  causes some churn.

  Note that this also makes it possible to extend struct mount_idmap in
  the future. For example, it would be possible to place the namespace
  pointer in an anonymous union together with an idmapping struct. This
  would allow us to expose an api to userspace that would let it specify
  idmappings directly instead of having to go through the detour of
  setting up namespaces at all"

* tag 'fs.idmapped.mnt_idmap.v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/idmapping:
  acl: conver higher-level helpers to rely on mnt_idmap
  fs: introduce dedicated idmap type for mounts
This commit is contained in:
Linus Torvalds 2022-12-12 19:30:18 -08:00
commit 9b93f5069f
8 changed files with 196 additions and 82 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
fs/internal.h
View File

@ -227,28 +227,28 @@ struct xattr_ctx {
};
ssize_t do_getxattr(struct user_namespace *mnt_userns,
ssize_t do_getxattr(struct mnt_idmap *idmap,
struct dentry *d,
struct xattr_ctx *ctx);
int setxattr_copy(const char __user *name, struct xattr_ctx *ctx);
int do_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry,
int do_setxattr(struct mnt_idmap *idmap, struct dentry *dentry,
struct xattr_ctx *ctx);
int may_write_xattr(struct user_namespace *mnt_userns, struct inode *inode);
#ifdef CONFIG_FS_POSIX_ACL
int do_set_acl(struct user_namespace *mnt_userns, struct dentry *dentry,
int do_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
const char *acl_name, const void *kvalue, size_t size);
ssize_t do_get_acl(struct user_namespace *mnt_userns, struct dentry *dentry,
ssize_t do_get_acl(struct mnt_idmap *idmap, struct dentry *dentry,
const char *acl_name, void *kvalue, size_t size);
#else
static inline int do_set_acl(struct user_namespace *mnt_userns,
static inline int do_set_acl(struct mnt_idmap *idmap,
struct dentry *dentry, const char *acl_name,
const void *kvalue, size_t size)
{
return -EOPNOTSUPP;
}
static inline ssize_t do_get_acl(struct user_namespace *mnt_userns,
static inline ssize_t do_get_acl(struct mnt_idmap *idmap,
struct dentry *dentry, const char *acl_name,
void *kvalue, size_t size)
{

176
fs/namespace.c
View File

@ -75,6 +75,22 @@ static DECLARE_RWSEM(namespace_sem);
static HLIST_HEAD(unmounted); /* protected by namespace_sem */
static LIST_HEAD(ex_mountpoints); /* protected by namespace_sem */
struct mnt_idmap {
struct user_namespace *owner;
refcount_t count;
};
/*
* Carries the initial idmapping of 0:0:4294967295 which is an identity
* mapping. This means that {g,u}id 0 is mapped to {g,u}id 0, {g,u}id 1 is
* mapped to {g,u}id 1, [...], {g,u}id 1000 to {g,u}id 1000, [...].
*/
struct mnt_idmap nop_mnt_idmap = {
.owner = &init_user_ns,
.count = REFCOUNT_INIT(1),
};
EXPORT_SYMBOL_GPL(nop_mnt_idmap);
struct mount_kattr {
unsigned int attr_set;
unsigned int attr_clr;
@ -82,6 +98,7 @@ struct mount_kattr {
unsigned int lookup_flags;
bool recurse;
struct user_namespace *mnt_userns;
struct mnt_idmap *mnt_idmap;
};
/* /sys/fs */
@ -193,6 +210,104 @@ int mnt_get_count(struct mount *mnt)
#endif
}
/**
* mnt_idmap_owner - retrieve owner of the mount's idmapping
* @idmap: mount idmapping
*
* This helper will go away once the conversion to use struct mnt_idmap
* everywhere has finished at which point the helper will be unexported.
*
* Only code that needs to perform permission checks based on the owner of the
* idmapping will get access to it. All other code will solely rely on
* idmappings. This will get us type safety so it's impossible to conflate
* filesystems idmappings with mount idmappings.
*
* Return: The owner of the idmapping.
*/
struct user_namespace *mnt_idmap_owner(const struct mnt_idmap *idmap)
{
return idmap->owner;
}
EXPORT_SYMBOL_GPL(mnt_idmap_owner);
/**
* mnt_user_ns - retrieve owner of an idmapped mount
* @mnt: the relevant vfsmount
*
* This helper will go away once the conversion to use struct mnt_idmap
* everywhere has finished at which point the helper will be unexported.
*
* Only code that needs to perform permission checks based on the owner of the
* idmapping will get access to it. All other code will solely rely on
* idmappings. This will get us type safety so it's impossible to conflate
* filesystems idmappings with mount idmappings.
*
* Return: The owner of the idmapped.
*/
struct user_namespace *mnt_user_ns(const struct vfsmount *mnt)
{
struct mnt_idmap *idmap = mnt_idmap(mnt);
/* Return the actual owner of the filesystem instead of the nop. */
if (idmap == &nop_mnt_idmap &&
!initial_idmapping(mnt->mnt_sb->s_user_ns))
return mnt->mnt_sb->s_user_ns;
return mnt_idmap_owner(idmap);
}
EXPORT_SYMBOL_GPL(mnt_user_ns);
/**
* alloc_mnt_idmap - allocate a new idmapping for the mount
* @mnt_userns: owning userns of the idmapping
*
* Allocate a new struct mnt_idmap which carries the idmapping of the mount.
*
* Return: On success a new idmap, on error an error pointer is returned.
*/
static struct mnt_idmap *alloc_mnt_idmap(struct user_namespace *mnt_userns)
{
struct mnt_idmap *idmap;
idmap = kzalloc(sizeof(struct mnt_idmap), GFP_KERNEL_ACCOUNT);
if (!idmap)
return ERR_PTR(-ENOMEM);
idmap->owner = get_user_ns(mnt_userns);
refcount_set(&idmap->count, 1);
return idmap;
}
/**
* mnt_idmap_get - get a reference to an idmapping
* @idmap: the idmap to bump the reference on
*
* If @idmap is not the @nop_mnt_idmap bump the reference count.
*
* Return: @idmap with reference count bumped if @not_mnt_idmap isn't passed.
*/
static inline struct mnt_idmap *mnt_idmap_get(struct mnt_idmap *idmap)
{
if (idmap != &nop_mnt_idmap)
refcount_inc(&idmap->count);
return idmap;
}
/**
* mnt_idmap_put - put a reference to an idmapping
* @idmap: the idmap to put the reference on
*
* If this is a non-initial idmapping, put the reference count when a mount is
* released and free it if we're the last user.
*/
static inline void mnt_idmap_put(struct mnt_idmap *idmap)
{
if (idmap != &nop_mnt_idmap && refcount_dec_and_test(&idmap->count)) {
put_user_ns(idmap->owner);
kfree(idmap);
}
}
static struct mount *alloc_vfsmnt(const char *name)
{
struct mount *mnt = kmem_cache_zalloc(mnt_cache, GFP_KERNEL);
@ -232,7 +347,7 @@ static struct mount *alloc_vfsmnt(const char *name)
INIT_HLIST_NODE(&mnt->mnt_mp_list);
INIT_LIST_HEAD(&mnt->mnt_umounting);
INIT_HLIST_HEAD(&mnt->mnt_stuck_children);
mnt->mnt.mnt_userns = &init_user_ns;
mnt->mnt.mnt_idmap = &nop_mnt_idmap;
}
return mnt;
@ -602,11 +717,7 @@ int sb_prepare_remount_readonly(struct super_block *sb)
static void free_vfsmnt(struct mount *mnt)
{
struct user_namespace *mnt_userns;
mnt_userns = mnt_user_ns(&mnt->mnt);
if (!initial_idmapping(mnt_userns))
put_user_ns(mnt_userns);
mnt_idmap_put(mnt_idmap(&mnt->mnt));
kfree_const(mnt->mnt_devname);
#ifdef CONFIG_SMP
free_percpu(mnt->mnt_pcp);
@ -1009,7 +1120,6 @@ static struct mount *skip_mnt_tree(struct mount *p)
struct vfsmount *vfs_create_mount(struct fs_context *fc)
{
struct mount *mnt;
struct user_namespace *fs_userns;
if (!fc->root)
return ERR_PTR(-EINVAL);
@ -1027,10 +1137,6 @@ struct vfsmount *vfs_create_mount(struct fs_context *fc)
mnt->mnt_mountpoint = mnt->mnt.mnt_root;
mnt->mnt_parent = mnt;
fs_userns = mnt->mnt.mnt_sb->s_user_ns;
if (!initial_idmapping(fs_userns))
mnt->mnt.mnt_userns = get_user_ns(fs_userns);
lock_mount_hash();
list_add_tail(&mnt->mnt_instance, &mnt->mnt.mnt_sb->s_mounts);
unlock_mount_hash();
@ -1120,9 +1226,8 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
mnt->mnt.mnt_flags &= ~(MNT_WRITE_HOLD|MNT_MARKED|MNT_INTERNAL);
atomic_inc(&sb->s_active);
mnt->mnt.mnt_userns = mnt_user_ns(&old->mnt);
if (!initial_idmapping(mnt->mnt.mnt_userns))
mnt->mnt.mnt_userns = get_user_ns(mnt->mnt.mnt_userns);
mnt->mnt.mnt_idmap = mnt_idmap_get(mnt_idmap(&old->mnt));
mnt->mnt.mnt_sb = sb;
mnt->mnt.mnt_root = dget(root);
mnt->mnt_mountpoint = mnt->mnt.mnt_root;
@ -3982,14 +4087,14 @@ static int can_idmap_mount(const struct mount_kattr *kattr, struct mount *mnt)
struct vfsmount *m = &mnt->mnt;
struct user_namespace *fs_userns = m->mnt_sb->s_user_ns;
if (!kattr->mnt_userns)
if (!kattr->mnt_idmap)
return 0;
/*
* Creating an idmapped mount with the filesystem wide idmapping
* doesn't make sense so block that. We don't allow mushy semantics.
*/
if (kattr->mnt_userns == fs_userns)
if (mnt_idmap_owner(kattr->mnt_idmap) == fs_userns)
return -EINVAL;
/*
@ -4029,7 +4134,7 @@ static inline bool mnt_allow_writers(const struct mount_kattr *kattr,
{
return (!(kattr->attr_set & MNT_READONLY) ||
(mnt->mnt.mnt_flags & MNT_READONLY)) &&
!kattr->mnt_userns;
!kattr->mnt_idmap;
}
static int mount_setattr_prepare(struct mount_kattr *kattr, struct mount *mnt)
@ -4083,27 +4188,18 @@ static int mount_setattr_prepare(struct mount_kattr *kattr, struct mount *mnt)
static void do_idmap_mount(const struct mount_kattr *kattr, struct mount *mnt)
{
struct user_namespace *mnt_userns, *old_mnt_userns;
if (!kattr->mnt_userns)
if (!kattr->mnt_idmap)
return;
/*
* We're the only ones able to change the mount's idmapping. So
* mnt->mnt.mnt_userns is stable and we can retrieve it directly.
* Pairs with smp_load_acquire() in mnt_idmap().
*
* Since we only allow a mount to change the idmapping once and
* verified this in can_idmap_mount() we know that the mount has
* @nop_mnt_idmap attached to it. So there's no need to drop any
* references.
*/
old_mnt_userns = mnt->mnt.mnt_userns;
mnt_userns = get_user_ns(kattr->mnt_userns);
/* Pairs with smp_load_acquire() in mnt_user_ns(). */
smp_store_release(&mnt->mnt.mnt_userns, mnt_userns);
/*
* If this is an idmapped filesystem drop the reference we've taken
* in vfs_create_mount() before.
*/
if (!initial_idmapping(old_mnt_userns))
put_user_ns(old_mnt_userns);
smp_store_release(&mnt->mnt.mnt_idmap, mnt_idmap_get(kattr->mnt_idmap));
}
static void mount_setattr_commit(struct mount_kattr *kattr, struct mount *mnt)
@ -4137,6 +4233,15 @@ static int do_mount_setattr(struct path *path, struct mount_kattr *kattr)
if (path->dentry != mnt->mnt.mnt_root)
return -EINVAL;
if (kattr->mnt_userns) {
struct mnt_idmap *mnt_idmap;
mnt_idmap = alloc_mnt_idmap(kattr->mnt_userns);
if (IS_ERR(mnt_idmap))
return PTR_ERR(mnt_idmap);
kattr->mnt_idmap = mnt_idmap;
}
if (kattr->propagation) {
/*
* Only take namespace_lock() if we're actually changing
@ -4324,6 +4429,9 @@ static void finish_mount_kattr(struct mount_kattr *kattr)
{
put_user_ns(kattr->mnt_userns);
kattr->mnt_userns = NULL;
if (kattr->mnt_idmap)
mnt_idmap_put(kattr->mnt_idmap);
}
SYSCALL_DEFINE5(mount_setattr, int, dfd, const char __user *, path,

16
fs/posix_acl.c
View File

@ -871,7 +871,7 @@ EXPORT_SYMBOL (posix_acl_to_xattr);
/**
* vfs_posix_acl_to_xattr - convert from kernel to userspace representation
* @mnt_userns: user namespace of the mount
* @idmap: idmap of the mount
* @inode: inode the posix acls are set on
* @acl: the posix acls as represented by the vfs
* @buffer: the buffer into which to convert @acl
@ -884,7 +884,7 @@ EXPORT_SYMBOL (posix_acl_to_xattr);
* Return: On success, the size of the stored uapi posix acls, on error a
* negative errno.
*/
static ssize_t vfs_posix_acl_to_xattr(struct user_namespace *mnt_userns,
static ssize_t vfs_posix_acl_to_xattr(struct mnt_idmap *idmap,
struct inode *inode,
const struct posix_acl *acl, void *buffer,
size_t size)
@ -893,6 +893,7 @@ static ssize_t vfs_posix_acl_to_xattr(struct user_namespace *mnt_userns,
struct posix_acl_xattr_header *ext_acl = buffer;
struct posix_acl_xattr_entry *ext_entry;
struct user_namespace *fs_userns, *caller_userns;
struct user_namespace *mnt_userns;
ssize_t real_size, n;
vfsuid_t vfsuid;
vfsgid_t vfsgid;
@ -908,6 +909,7 @@ static ssize_t vfs_posix_acl_to_xattr(struct user_namespace *mnt_userns,
fs_userns = i_user_ns(inode);
caller_userns = current_user_ns();
mnt_userns = mnt_idmap_owner(idmap);
for (n=0; n < acl->a_count; n++, ext_entry++) {
const struct posix_acl_entry *acl_e = &acl->a_entries[n];
ext_entry->e_tag = cpu_to_le16(acl_e->e_tag);
@ -1227,7 +1229,7 @@ out_inode_unlock:
}
EXPORT_SYMBOL_GPL(vfs_remove_acl);
int do_set_acl(struct user_namespace *mnt_userns, struct dentry *dentry,
int do_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
const char *acl_name, const void *kvalue, size_t size)
{
int error;
@ -1243,22 +1245,22 @@ int do_set_acl(struct user_namespace *mnt_userns, struct dentry *dentry,
return PTR_ERR(acl);
}
error = vfs_set_acl(mnt_userns, dentry, acl_name, acl);
error = vfs_set_acl(mnt_idmap_owner(idmap), dentry, acl_name, acl);
posix_acl_release(acl);
return error;
}
ssize_t do_get_acl(struct user_namespace *mnt_userns, struct dentry *dentry,
ssize_t do_get_acl(struct mnt_idmap *idmap, struct dentry *dentry,
const char *acl_name, void *kvalue, size_t size)
{
ssize_t error;
struct posix_acl *acl;
acl = vfs_get_acl(mnt_userns, dentry, acl_name);
acl = vfs_get_acl(mnt_idmap_owner(idmap), dentry, acl_name);
if (IS_ERR(acl))
return PTR_ERR(acl);
error = vfs_posix_acl_to_xattr(mnt_userns, d_inode(dentry),
error = vfs_posix_acl_to_xattr(idmap, d_inode(dentry),
acl, kvalue, size);
posix_acl_release(acl);
return error;

39
fs/xattr.c
View File

@ -597,19 +597,19 @@ int setxattr_copy(const char __user *name, struct xattr_ctx *ctx)
return error;
}
int do_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry,
int do_setxattr(struct mnt_idmap *idmap, struct dentry *dentry,
struct xattr_ctx *ctx)
{
if (is_posix_acl_xattr(ctx->kname->name))
return do_set_acl(mnt_userns, dentry, ctx->kname->name,
return do_set_acl(idmap, dentry, ctx->kname->name,
ctx->kvalue, ctx->size);
return vfs_setxattr(mnt_userns, dentry, ctx->kname->name,
return vfs_setxattr(mnt_idmap_owner(idmap), dentry, ctx->kname->name,
ctx->kvalue, ctx->size, ctx->flags);
}
static long
setxattr(struct user_namespace *mnt_userns, struct dentry *d,
setxattr(struct mnt_idmap *idmap, struct dentry *d,
const char __user *name, const void __user *value, size_t size,
int flags)
{
@ -627,7 +627,7 @@ setxattr(struct user_namespace *mnt_userns, struct dentry *d,
if (error)
return error;
error = do_setxattr(mnt_userns, d, &ctx);
error = do_setxattr(idmap, d, &ctx);
kvfree(ctx.kvalue);
return error;
@ -646,7 +646,7 @@ retry:
return error;
error = mnt_want_write(path.mnt);
if (!error) {
error = setxattr(mnt_user_ns(path.mnt), path.dentry, name,
error = setxattr(mnt_idmap(path.mnt), path.dentry, name,
value, size, flags);
mnt_drop_write(path.mnt);
}
@ -683,7 +683,7 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name,
audit_file(f.file);
error = mnt_want_write_file(f.file);
if (!error) {
error = setxattr(file_mnt_user_ns(f.file),
error = setxattr(file_mnt_idmap(f.file),
f.file->f_path.dentry, name,
value, size, flags);
mnt_drop_write_file(f.file);
@ -696,7 +696,7 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name,
* Extended attribute GET operations
*/
ssize_t
do_getxattr(struct user_namespace *mnt_userns, struct dentry *d,
do_getxattr(struct mnt_idmap *idmap, struct dentry *d,
struct xattr_ctx *ctx)
{
ssize_t error;
@ -711,9 +711,10 @@ do_getxattr(struct user_namespace *mnt_userns, struct dentry *d,
}
if (is_posix_acl_xattr(ctx->kname->name))
error = do_get_acl(mnt_userns, d, kname, ctx->kvalue, ctx->size);
error = do_get_acl(idmap, d, kname, ctx->kvalue, ctx->size);
else
error = vfs_getxattr(mnt_userns, d, kname, ctx->kvalue, ctx->size);
error = vfs_getxattr(mnt_idmap_owner(idmap), d, kname,
ctx->kvalue, ctx->size);
if (error > 0) {
if (ctx->size && copy_to_user(ctx->value, ctx->kvalue, error))
error = -EFAULT;
@ -727,7 +728,7 @@ do_getxattr(struct user_namespace *mnt_userns, struct dentry *d,
}
static ssize_t
getxattr(struct user_namespace *mnt_userns, struct dentry *d,
getxattr(struct mnt_idmap *idmap, struct dentry *d,
const char __user *name, void __user *value, size_t size)
{
ssize_t error;
@ -746,7 +747,7 @@ getxattr(struct user_namespace *mnt_userns, struct dentry *d,
if (error < 0)
return error;
error = do_getxattr(mnt_userns, d, &ctx);
error = do_getxattr(idmap, d, &ctx);
kvfree(ctx.kvalue);
return error;
@ -762,7 +763,7 @@ retry:
error = user_path_at(AT_FDCWD, pathname, lookup_flags, &path);
if (error)
return error;
error = getxattr(mnt_user_ns(path.mnt), path.dentry, name, value, size);
error = getxattr(mnt_idmap(path.mnt), path.dentry, name, value, size);
path_put(&path);
if (retry_estale(error, lookup_flags)) {
lookup_flags |= LOOKUP_REVAL;
@ -792,7 +793,7 @@ SYSCALL_DEFINE4(fgetxattr, int, fd, const char __user *, name,
if (!f.file)
return error;
audit_file(f.file);
error = getxattr(file_mnt_user_ns(f.file), f.file->f_path.dentry,
error = getxattr(file_mnt_idmap(f.file), f.file->f_path.dentry,
name, value, size);
fdput(f);
return error;
@ -877,7 +878,7 @@ SYSCALL_DEFINE3(flistxattr, int, fd, char __user *, list, size_t, size)
* Extended attribute REMOVE operations
*/
static long
removexattr(struct user_namespace *mnt_userns, struct dentry *d,
removexattr(struct mnt_idmap *idmap, struct dentry *d,
const char __user *name)
{
int error;
@ -890,9 +891,9 @@ removexattr(struct user_namespace *mnt_userns, struct dentry *d,
return error;
if (is_posix_acl_xattr(kname))
return vfs_remove_acl(mnt_userns, d, kname);
return vfs_remove_acl(mnt_idmap_owner(idmap), d, kname);
return vfs_removexattr(mnt_userns, d, kname);
return vfs_removexattr(mnt_idmap_owner(idmap), d, kname);
}
static int path_removexattr(const char __user *pathname,
@ -906,7 +907,7 @@ retry:
return error;
error = mnt_want_write(path.mnt);
if (!error) {
error = removexattr(mnt_user_ns(path.mnt), path.dentry, name);
error = removexattr(mnt_idmap(path.mnt), path.dentry, name);
mnt_drop_write(path.mnt);
}
path_put(&path);
@ -939,7 +940,7 @@ SYSCALL_DEFINE2(fremovexattr, int, fd, const char __user *, name)
audit_file(f.file);
error = mnt_want_write_file(f.file);
if (!error) {
error = removexattr(file_mnt_user_ns(f.file),
error = removexattr(file_mnt_idmap(f.file),
f.file->f_path.dentry, name);
mnt_drop_write_file(f.file);
}

10
include/linux/fs.h
View File

@ -2727,18 +2727,22 @@ static inline struct user_namespace *file_mnt_user_ns(struct file *file)
return mnt_user_ns(file->f_path.mnt);
}
static inline struct mnt_idmap *file_mnt_idmap(struct file *file)
{
return mnt_idmap(file->f_path.mnt);
}
/**
* is_idmapped_mnt - check whether a mount is mapped
* @mnt: the mount to check
*
* If @mnt has an idmapping attached different from the
* filesystem's idmapping then @mnt is mapped.
* If @mnt has an non @nop_mnt_idmap attached to it then @mnt is mapped.
*
* Return: true if mount is mapped, false if not.
*/
static inline bool is_idmapped_mnt(const struct vfsmount *mnt)
{
return mnt_user_ns(mnt) != mnt->mnt_sb->s_user_ns;
return mnt_idmap(mnt) != &nop_mnt_idmap;
}
extern long vfs_truncate(const struct path *, loff_t);

8
include/linux/mnt_idmapping.h
View File

@ -5,12 +5,10 @@
#include <linux/types.h>
#include <linux/uidgid.h>
struct mnt_idmap;
struct user_namespace;
/*
* Carries the initial idmapping of 0:0:4294967295 which is an identity
* mapping. This means that {g,u}id 0 is mapped to {g,u}id 0, {g,u}id 1 is
* mapped to {g,u}id 1, [...], {g,u}id 1000 to {g,u}id 1000, [...].
*/
extern struct mnt_idmap nop_mnt_idmap;
extern struct user_namespace init_user_ns;
typedef struct {

9
include/linux/mount.h
View File

@ -16,6 +16,7 @@
struct super_block;
struct dentry;
struct user_namespace;
struct mnt_idmap;
struct file_system_type;
struct fs_context;
struct file;
@ -70,13 +71,15 @@ struct vfsmount {
struct dentry *mnt_root; /* root of the mounted tree */
struct super_block *mnt_sb; /* pointer to superblock */
int mnt_flags;
struct user_namespace *mnt_userns;
struct mnt_idmap *mnt_idmap;
} __randomize_layout;
static inline struct user_namespace *mnt_user_ns(const struct vfsmount *mnt)
struct user_namespace *mnt_user_ns(const struct vfsmount *mnt);
struct user_namespace *mnt_idmap_owner(const struct mnt_idmap *idmap);
static inline struct mnt_idmap *mnt_idmap(const struct vfsmount *mnt)
{
/* Pairs with smp_store_release() in do_idmap_mount(). */
return smp_load_acquire(&mnt->mnt_userns);
return smp_load_acquire(&mnt->mnt_idmap);
}
extern int mnt_want_write(struct vfsmount *mnt);

8
io_uring/xattr.c
View File

@ -112,7 +112,7 @@ int io_fgetxattr(struct io_kiocb *req, unsigned int issue_flags)
if (issue_flags & IO_URING_F_NONBLOCK)
return -EAGAIN;
ret = do_getxattr(mnt_user_ns(req->file->f_path.mnt),
ret = do_getxattr(mnt_idmap(req->file->f_path.mnt),
req->file->f_path.dentry,
&ix->ctx);
@ -133,9 +133,7 @@ int io_getxattr(struct io_kiocb *req, unsigned int issue_flags)
retry:
ret = filename_lookup(AT_FDCWD, ix->filename, lookup_flags, &path, NULL);
if (!ret) {
ret = do_getxattr(mnt_user_ns(path.mnt),
path.dentry,
&ix->ctx);
ret = do_getxattr(mnt_idmap(path.mnt), path.dentry, &ix->ctx);
path_put(&path);
if (retry_estale(ret, lookup_flags)) {
@ -213,7 +211,7 @@ static int __io_setxattr(struct io_kiocb *req, unsigned int issue_flags,
ret = mnt_want_write(path->mnt);
if (!ret) {
ret = do_setxattr(mnt_user_ns(path->mnt), path->dentry, &ix->ctx);
ret = do_setxattr(mnt_idmap(path->mnt), path->dentry, &ix->ctx);
mnt_drop_write(path->mnt);
}