kernel_optimize_test/fs/afs/dynroot.c
Linus Torvalds c84ca912b0 Keyrings namespacing
-----BEGIN PGP SIGNATURE-----
 
 iQIVAwUAXRU89Pu3V2unywtrAQIdBBAAmMBsrfv+LUN4Vru/D6KdUO4zdYGcNK6m
 S56bcNfP6oIDEj6HrNNnzKkWIZpdZ61Odv1zle96+v4WZ/6rnLCTpcsdaFNTzaoO
 YT2jk7jplss0ImrMv1DSoykGqO3f0ThMIpGCxHKZADGSu0HMbjSEh+zLPV4BaMtT
 BVuF7P3eZtDRLdDtMtYcgvf5UlbdoBEY8w1FUjReQx8hKGxVopGmCo5vAeiY8W9S
 ybFSZhPS5ka33ynVrLJH2dqDo5A8pDhY8I4bdlcxmNtRhnPCYZnuvTqeAzyUKKdI
 YN9zJeDu1yHs9mi8dp45NPJiKy6xLzWmUwqH8AvR8MWEkrwzqbzNZCEHZ41j74hO
 YZWI0JXi72cboszFvOwqJERvITKxrQQyVQLPRQE2vVbG0bIZPl8i7oslFVhitsl+
 evWqHb4lXY91rI9cC6JIXR1OiUjp68zXPv7DAnxv08O+PGcioU1IeOvPivx8QSx4
 5aUeCkYIIAti/GISzv7xvcYh8mfO76kBjZSB35fX+R9DkeQpxsHmmpWe+UCykzWn
 EwhHQn86+VeBFP6RAXp8CgNCLbrwkEhjzXQl/70s1eYbwvK81VcpDAQ6+cjpf4Hb
 QUmrUJ9iE0wCNl7oqvJZoJvWVGlArvPmzpkTJk3N070X2R0T7x1WCsMlPDMJGhQ2
 fVHvA3QdgWs=
 =Push
 -----END PGP SIGNATURE-----

Merge tag 'keys-namespace-20190627' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs

Pull keyring namespacing from David Howells:
 "These patches help make keys and keyrings more namespace aware.

  Firstly some miscellaneous patches to make the process easier:

   - Simplify key index_key handling so that the word-sized chunks
     assoc_array requires don't have to be shifted about, making it
     easier to add more bits into the key.

   - Cache the hash value in the key so that we don't have to calculate
     on every key we examine during a search (it involves a bunch of
     multiplications).

   - Allow keying_search() to search non-recursively.

  Then the main patches:

   - Make it so that keyring names are per-user_namespace from the point
     of view of KEYCTL_JOIN_SESSION_KEYRING so that they're not
     accessible cross-user_namespace.

     keyctl_capabilities() shows KEYCTL_CAPS1_NS_KEYRING_NAME for this.

   - Move the user and user-session keyrings to the user_namespace
     rather than the user_struct. This prevents them propagating
     directly across user_namespaces boundaries (ie. the KEY_SPEC_*
     flags will only pick from the current user_namespace).

   - Make it possible to include the target namespace in which the key
     shall operate in the index_key. This will allow the possibility of
     multiple keys with the same description, but different target
     domains to be held in the same keyring.

     keyctl_capabilities() shows KEYCTL_CAPS1_NS_KEY_TAG for this.

   - Make it so that keys are implicitly invalidated by removal of a
     domain tag, causing them to be garbage collected.

   - Institute a network namespace domain tag that allows keys to be
     differentiated by the network namespace in which they operate. New
     keys that are of a type marked 'KEY_TYPE_NET_DOMAIN' are assigned
     the network domain in force when they are created.

   - Make it so that the desired network namespace can be handed down
     into the request_key() mechanism. This allows AFS, NFS, etc. to
     request keys specific to the network namespace of the superblock.

     This also means that the keys in the DNS record cache are
     thenceforth namespaced, provided network filesystems pass the
     appropriate network namespace down into dns_query().

     For DNS, AFS and NFS are good, whilst CIFS and Ceph are not. Other
     cache keyrings, such as idmapper keyrings, also need to set the
     domain tag - for which they need access to the network namespace of
     the superblock"

* tag 'keys-namespace-20190627' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs:
  keys: Pass the network namespace into request_key mechanism
  keys: Network namespace domain tag
  keys: Garbage collect keys for which the domain has been removed
  keys: Include target namespace in match criteria
  keys: Move the user and user-session keyrings to the user_namespace
  keys: Namespace keyring names
  keys: Add a 'recurse' flag for keyring searches
  keys: Cache the hash value to avoid lots of recalculation
  keys: Simplify key description management
2019-07-08 19:36:47 -07:00

308 lines
6.8 KiB
C

// SPDX-License-Identifier: GPL-2.0-or-later
/* AFS dynamic root handling
*
* Copyright (C) 2018 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*/
#include <linux/fs.h>
#include <linux/namei.h>
#include <linux/dns_resolver.h>
#include "internal.h"
const struct file_operations afs_dynroot_file_operations = {
.open = dcache_dir_open,
.release = dcache_dir_close,
.iterate_shared = dcache_readdir,
.llseek = dcache_dir_lseek,
};
/*
* Probe to see if a cell may exist. This prevents positive dentries from
* being created unnecessarily.
*/
static int afs_probe_cell_name(struct dentry *dentry)
{
struct afs_cell *cell;
struct afs_net *net = afs_d2net(dentry);
const char *name = dentry->d_name.name;
size_t len = dentry->d_name.len;
int ret;
/* Names prefixed with a dot are R/W mounts. */
if (name[0] == '.') {
if (len == 1)
return -EINVAL;
name++;
len--;
}
cell = afs_lookup_cell_rcu(net, name, len);
if (!IS_ERR(cell)) {
afs_put_cell(net, cell);
return 0;
}
ret = dns_query(net->net, "afsdb", name, len, "srv=1",
NULL, NULL, false);
if (ret == -ENODATA)
ret = -EDESTADDRREQ;
return ret;
}
/*
* Try to auto mount the mountpoint with pseudo directory, if the autocell
* operation is setted.
*/
struct inode *afs_try_auto_mntpt(struct dentry *dentry, struct inode *dir)
{
struct afs_vnode *vnode = AFS_FS_I(dir);
struct inode *inode;
int ret = -ENOENT;
_enter("%p{%pd}, {%llx:%llu}",
dentry, dentry, vnode->fid.vid, vnode->fid.vnode);
if (!test_bit(AFS_VNODE_AUTOCELL, &vnode->flags))
goto out;
ret = afs_probe_cell_name(dentry);
if (ret < 0)
goto out;
inode = afs_iget_pseudo_dir(dir->i_sb, false);
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
goto out;
}
_leave("= %p", inode);
return inode;
out:
_leave("= %d", ret);
return ret == -ENOENT ? NULL : ERR_PTR(ret);
}
/*
* Look up @cell in a dynroot directory. This is a substitution for the
* local cell name for the net namespace.
*/
static struct dentry *afs_lookup_atcell(struct dentry *dentry)
{
struct afs_cell *cell;
struct afs_net *net = afs_d2net(dentry);
struct dentry *ret;
unsigned int seq = 0;
char *name;
int len;
if (!net->ws_cell)
return ERR_PTR(-ENOENT);
ret = ERR_PTR(-ENOMEM);
name = kmalloc(AFS_MAXCELLNAME + 1, GFP_KERNEL);
if (!name)
goto out_p;
rcu_read_lock();
do {
read_seqbegin_or_lock(&net->cells_lock, &seq);
cell = rcu_dereference_raw(net->ws_cell);
if (cell) {
len = cell->name_len;
memcpy(name, cell->name, len + 1);
}
} while (need_seqretry(&net->cells_lock, seq));
done_seqretry(&net->cells_lock, seq);
rcu_read_unlock();
ret = ERR_PTR(-ENOENT);
if (!cell)
goto out_n;
ret = lookup_one_len(name, dentry->d_parent, len);
/* We don't want to d_add() the @cell dentry here as we don't want to
* the cached dentry to hide changes to the local cell name.
*/
out_n:
kfree(name);
out_p:
return ret;
}
/*
* Look up an entry in a dynroot directory.
*/
static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentry,
unsigned int flags)
{
_enter("%pd", dentry);
ASSERTCMP(d_inode(dentry), ==, NULL);
if (dentry->d_name.len >= AFSNAMEMAX) {
_leave(" = -ENAMETOOLONG");
return ERR_PTR(-ENAMETOOLONG);
}
if (dentry->d_name.len == 5 &&
memcmp(dentry->d_name.name, "@cell", 5) == 0)
return afs_lookup_atcell(dentry);
return d_splice_alias(afs_try_auto_mntpt(dentry, dir), dentry);
}
const struct inode_operations afs_dynroot_inode_operations = {
.lookup = afs_dynroot_lookup,
};
/*
* Dirs in the dynamic root don't need revalidation.
*/
static int afs_dynroot_d_revalidate(struct dentry *dentry, unsigned int flags)
{
return 1;
}
/*
* Allow the VFS to enquire as to whether a dentry should be unhashed (mustn't
* sleep)
* - called from dput() when d_count is going to 0.
* - return 1 to request dentry be unhashed, 0 otherwise
*/
static int afs_dynroot_d_delete(const struct dentry *dentry)
{
return d_really_is_positive(dentry);
}
const struct dentry_operations afs_dynroot_dentry_operations = {
.d_revalidate = afs_dynroot_d_revalidate,
.d_delete = afs_dynroot_d_delete,
.d_release = afs_d_release,
.d_automount = afs_d_automount,
};
/*
* Create a manually added cell mount directory.
* - The caller must hold net->proc_cells_lock
*/
int afs_dynroot_mkdir(struct afs_net *net, struct afs_cell *cell)
{
struct super_block *sb = net->dynroot_sb;
struct dentry *root, *subdir;
int ret;
if (!sb || atomic_read(&sb->s_active) == 0)
return 0;
/* Let the ->lookup op do the creation */
root = sb->s_root;
inode_lock(root->d_inode);
subdir = lookup_one_len(cell->name, root, cell->name_len);
if (IS_ERR(subdir)) {
ret = PTR_ERR(subdir);
goto unlock;
}
/* Note that we're retaining an extra ref on the dentry */
subdir->d_fsdata = (void *)1UL;
ret = 0;
unlock:
inode_unlock(root->d_inode);
return ret;
}
/*
* Remove a manually added cell mount directory.
* - The caller must hold net->proc_cells_lock
*/
void afs_dynroot_rmdir(struct afs_net *net, struct afs_cell *cell)
{
struct super_block *sb = net->dynroot_sb;
struct dentry *root, *subdir;
if (!sb || atomic_read(&sb->s_active) == 0)
return;
root = sb->s_root;
inode_lock(root->d_inode);
/* Don't want to trigger a lookup call, which will re-add the cell */
subdir = try_lookup_one_len(cell->name, root, cell->name_len);
if (IS_ERR_OR_NULL(subdir)) {
_debug("lookup %ld", PTR_ERR(subdir));
goto no_dentry;
}
_debug("rmdir %pd %u", subdir, d_count(subdir));
if (subdir->d_fsdata) {
_debug("unpin %u", d_count(subdir));
subdir->d_fsdata = NULL;
dput(subdir);
}
dput(subdir);
no_dentry:
inode_unlock(root->d_inode);
_leave("");
}
/*
* Populate a newly created dynamic root with cell names.
*/
int afs_dynroot_populate(struct super_block *sb)
{
struct afs_cell *cell;
struct afs_net *net = afs_sb2net(sb);
int ret;
mutex_lock(&net->proc_cells_lock);
net->dynroot_sb = sb;
hlist_for_each_entry(cell, &net->proc_cells, proc_link) {
ret = afs_dynroot_mkdir(net, cell);
if (ret < 0)
goto error;
}
ret = 0;
out:
mutex_unlock(&net->proc_cells_lock);
return ret;
error:
net->dynroot_sb = NULL;
goto out;
}
/*
* When a dynamic root that's in the process of being destroyed, depopulate it
* of pinned directories.
*/
void afs_dynroot_depopulate(struct super_block *sb)
{
struct afs_net *net = afs_sb2net(sb);
struct dentry *root = sb->s_root, *subdir, *tmp;
/* Prevent more subdirs from being created */
mutex_lock(&net->proc_cells_lock);
if (net->dynroot_sb == sb)
net->dynroot_sb = NULL;
mutex_unlock(&net->proc_cells_lock);
inode_lock(root->d_inode);
/* Remove all the pins for dirs created for manually added cells */
list_for_each_entry_safe(subdir, tmp, &root->d_subdirs, d_child) {
if (subdir->d_fsdata) {
subdir->d_fsdata = NULL;
dput(subdir);
}
}
inode_unlock(root->d_inode);
}