diff options
Diffstat (limited to 'fs/devpts/inode.c')
| -rw-r--r-- | fs/devpts/inode.c | 486 |
1 files changed, 230 insertions, 256 deletions
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 108df2e3602c..9f3de528c358 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -1,13 +1,10 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* -*- linux-c -*- --------------------------------------------------------- * * * linux/fs/devpts/inode.c * * Copyright 1998-2004 H. Peter Anvin -- All Rights Reserved * - * This file is part of the Linux kernel and is made available under - * the terms of the GNU General Public License, version 2, or at your - * option, any later version, incorporated herein by reference. - * * ------------------------------------------------------------------------- */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt @@ -15,6 +12,8 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/fs.h> +#include <linux/fs_context.h> +#include <linux/fs_parser.h> #include <linux/sched.h> #include <linux/namei.h> #include <linux/slab.h> @@ -24,7 +23,6 @@ #include <linux/magic.h> #include <linux/idr.h> #include <linux/devpts_fs.h> -#include <linux/parser.h> #include <linux/fsnotify.h> #include <linux/seq_file.h> @@ -46,9 +44,9 @@ static int pty_limit = NR_UNIX98_PTY_DEFAULT; static int pty_reserve = NR_UNIX98_PTY_RESERVE; static int pty_limit_min; static int pty_limit_max = INT_MAX; -static int pty_count; +static atomic_t pty_count = ATOMIC_INIT(0); -static struct ctl_table pty_table[] = { +static const struct ctl_table pty_table[] = { { .procname = "max", .maxlen = sizeof(int), @@ -72,29 +70,8 @@ static struct ctl_table pty_table[] = { .data = &pty_count, .proc_handler = proc_dointvec, }, - {} -}; - -static struct ctl_table pty_kern_table[] = { - { - .procname = "pty", - .mode = 0555, - .child = pty_table, - }, - {} -}; - -static struct ctl_table pty_root_table[] = { - { - .procname = "kernel", - .mode = 0555, - .child = pty_kern_table, - }, - {} }; -static DEFINE_MUTEX(allocated_ptys_lock); - struct pts_mount_opts { int setuid; int setgid; @@ -111,21 +88,21 @@ enum { Opt_err }; -static const match_table_t tokens = { - {Opt_uid, "uid=%u"}, - {Opt_gid, "gid=%u"}, - {Opt_mode, "mode=%o"}, - {Opt_ptmxmode, "ptmxmode=%o"}, - {Opt_newinstance, "newinstance"}, - {Opt_max, "max=%d"}, - {Opt_err, NULL} +static const struct fs_parameter_spec devpts_param_specs[] = { + fsparam_gid ("gid", Opt_gid), + fsparam_s32 ("max", Opt_max), + fsparam_u32oct ("mode", Opt_mode), + fsparam_flag ("newinstance", Opt_newinstance), + fsparam_u32oct ("ptmxmode", Opt_ptmxmode), + fsparam_uid ("uid", Opt_uid), + {} }; struct pts_fs_info { struct ida allocated_ptys; struct pts_mount_opts mount_opts; struct super_block *sb; - struct dentry *ptmx_dentry; + struct inode *ptmx_inode; // borrowed }; static inline struct pts_fs_info *DEVPTS_SB(struct super_block *sb) @@ -133,37 +110,98 @@ static inline struct pts_fs_info *DEVPTS_SB(struct super_block *sb) return sb->s_fs_info; } +static int devpts_ptmx_path(struct path *path) +{ + struct super_block *sb; + int err; + + /* Is a devpts filesystem at "pts" in the same directory? */ + err = path_pts(path); + if (err) + return err; + + /* Is the path the root of a devpts filesystem? */ + sb = path->mnt->mnt_sb; + if ((sb->s_magic != DEVPTS_SUPER_MAGIC) || + (path->mnt->mnt_root != sb->s_root)) + return -ENODEV; + + return 0; +} + +/* + * Try to find a suitable devpts filesystem. We support the following + * scenarios: + * - The ptmx device node is located in the same directory as the devpts + * mount where the pts device nodes are located. + * This is e.g. the case when calling open on the /dev/pts/ptmx device + * node when the devpts filesystem is mounted at /dev/pts. + * - The ptmx device node is located outside the devpts filesystem mount + * where the pts device nodes are located. For example, the ptmx device + * is a symlink, separate device node, or bind-mount. + * A supported scenario is bind-mounting /dev/pts/ptmx to /dev/ptmx and + * then calling open on /dev/ptmx. In this case a suitable pts + * subdirectory can be found in the common parent directory /dev of the + * devpts mount and the ptmx bind-mount, after resolving the /dev/ptmx + * bind-mount. + * If no suitable pts subdirectory can be found this function will fail. + * This is e.g. the case when bind-mounting /dev/pts/ptmx to /ptmx. + */ +struct vfsmount *devpts_mntget(struct file *filp, struct pts_fs_info *fsi) +{ + struct path path; + int err = 0; + + path = filp->f_path; + path_get(&path); + + /* Walk upward while the start point is a bind mount of + * a single file. + */ + while (path.mnt->mnt_root == path.dentry) + if (follow_up(&path) == 0) + break; + + /* devpts_ptmx_path() finds a devpts fs or returns an error. */ + if ((path.mnt->mnt_sb->s_magic != DEVPTS_SUPER_MAGIC) || + (DEVPTS_SB(path.mnt->mnt_sb) != fsi)) + err = devpts_ptmx_path(&path); + dput(path.dentry); + if (!err) { + if (DEVPTS_SB(path.mnt->mnt_sb) == fsi) + return path.mnt; + + err = -ENODEV; + } + + mntput(path.mnt); + return ERR_PTR(err); +} + struct pts_fs_info *devpts_acquire(struct file *filp) { struct pts_fs_info *result; struct path path; struct super_block *sb; - int err; path = filp->f_path; path_get(&path); /* Has the devpts filesystem already been found? */ - sb = path.mnt->mnt_sb; - if (sb->s_magic != DEVPTS_SUPER_MAGIC) { - /* Is a devpts filesystem at "pts" in the same directory? */ - err = path_pts(&path); + if (path.mnt->mnt_sb->s_magic != DEVPTS_SUPER_MAGIC) { + int err; + + err = devpts_ptmx_path(&path); if (err) { result = ERR_PTR(err); goto out; } - - /* Is the path the root of a devpts filesystem? */ - result = ERR_PTR(-ENODEV); - sb = path.mnt->mnt_sb; - if ((sb->s_magic != DEVPTS_SUPER_MAGIC) || - (path.mnt->mnt_root != sb->s_root)) - goto out; } /* * pty code needs to hold extra references in case of last /dev/tty close */ + sb = path.mnt->mnt_sb; atomic_inc(&sb->s_active); result = DEVPTS_SB(sb); @@ -177,96 +215,50 @@ void devpts_release(struct pts_fs_info *fsi) deactivate_super(fsi->sb); } -#define PARSE_MOUNT 0 -#define PARSE_REMOUNT 1 - /* - * parse_mount_options(): - * Set @opts to mount options specified in @data. If an option is not - * specified in @data, set it to its default value. - * - * Note: @data may be NULL (in which case all options are set to default). + * devpts_parse_param - Parse mount parameters */ -static int parse_mount_options(char *data, int op, struct pts_mount_opts *opts) +static int devpts_parse_param(struct fs_context *fc, struct fs_parameter *param) { - char *p; - kuid_t uid; - kgid_t gid; - - opts->setuid = 0; - opts->setgid = 0; - opts->uid = GLOBAL_ROOT_UID; - opts->gid = GLOBAL_ROOT_GID; - opts->mode = DEVPTS_DEFAULT_MODE; - opts->ptmxmode = DEVPTS_DEFAULT_PTMX_MODE; - opts->max = NR_UNIX98_PTY_MAX; - - /* Only allow instances mounted from the initial mount - * namespace to tap the reserve pool of ptys. - */ - if (op == PARSE_MOUNT) - opts->reserve = - (current->nsproxy->mnt_ns == init_task.nsproxy->mnt_ns); - - while ((p = strsep(&data, ",")) != NULL) { - substring_t args[MAX_OPT_ARGS]; - int token; - int option; - - if (!*p) - continue; - - token = match_token(p, tokens, args); - switch (token) { - case Opt_uid: - if (match_int(&args[0], &option)) - return -EINVAL; - uid = make_kuid(current_user_ns(), option); - if (!uid_valid(uid)) - return -EINVAL; - opts->uid = uid; - opts->setuid = 1; - break; - case Opt_gid: - if (match_int(&args[0], &option)) - return -EINVAL; - gid = make_kgid(current_user_ns(), option); - if (!gid_valid(gid)) - return -EINVAL; - opts->gid = gid; - opts->setgid = 1; - break; - case Opt_mode: - if (match_octal(&args[0], &option)) - return -EINVAL; - opts->mode = option & S_IALLUGO; - break; - case Opt_ptmxmode: - if (match_octal(&args[0], &option)) - return -EINVAL; - opts->ptmxmode = option & S_IALLUGO; - break; - case Opt_newinstance: - break; - case Opt_max: - if (match_int(&args[0], &option) || - option < 0 || option > NR_UNIX98_PTY_MAX) - return -EINVAL; - opts->max = option; - break; - default: - pr_err("called with bogus options\n"); - return -EINVAL; - } + struct pts_fs_info *fsi = fc->s_fs_info; + struct pts_mount_opts *opts = &fsi->mount_opts; + struct fs_parse_result result; + int opt; + + opt = fs_parse(fc, devpts_param_specs, param, &result); + if (opt < 0) + return opt; + + switch (opt) { + case Opt_uid: + opts->uid = result.uid; + opts->setuid = 1; + break; + case Opt_gid: + opts->gid = result.gid; + opts->setgid = 1; + break; + case Opt_mode: + opts->mode = result.uint_32 & S_IALLUGO; + break; + case Opt_ptmxmode: + opts->ptmxmode = result.uint_32 & S_IALLUGO; + break; + case Opt_newinstance: + break; + case Opt_max: + if (result.uint_32 > NR_UNIX98_PTY_MAX) + return invalf(fc, "max out of range"); + opts->max = result.uint_32; + break; } return 0; } -static int mknod_ptmx(struct super_block *sb) +static int mknod_ptmx(struct super_block *sb, struct fs_context *fc) { int mode; - int rc = -ENOMEM; struct dentry *dentry; struct inode *inode; struct dentry *root = sb->s_root; @@ -275,18 +267,10 @@ static int mknod_ptmx(struct super_block *sb) kuid_t ptmx_uid = current_fsuid(); kgid_t ptmx_gid = current_fsgid(); - inode_lock(d_inode(root)); - - /* If we have already created ptmx node, return */ - if (fsi->ptmx_dentry) { - rc = 0; - goto out; - } - - dentry = d_alloc_name(root, "ptmx"); - if (!dentry) { + dentry = simple_start_creating(root, "ptmx"); + if (IS_ERR(dentry)) { pr_err("Unable to alloc dentry for ptmx node\n"); - goto out; + return PTR_ERR(dentry); } /* @@ -294,44 +278,49 @@ static int mknod_ptmx(struct super_block *sb) */ inode = new_inode(sb); if (!inode) { + simple_done_creating(dentry); pr_err("Unable to alloc inode for ptmx node\n"); - dput(dentry); - goto out; + return -ENOMEM; } inode->i_ino = 2; - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); + simple_inode_init_ts(inode); mode = S_IFCHR|opts->ptmxmode; init_special_inode(inode, mode, MKDEV(TTYAUX_MAJOR, 2)); inode->i_uid = ptmx_uid; inode->i_gid = ptmx_gid; + fsi->ptmx_inode = inode; - d_add(dentry, inode); + d_make_persistent(dentry, inode); - fsi->ptmx_dentry = dentry; - rc = 0; -out: - inode_unlock(d_inode(root)); - return rc; + simple_done_creating(dentry); + + return 0; } static void update_ptmx_mode(struct pts_fs_info *fsi) { - struct inode *inode; - if (fsi->ptmx_dentry) { - inode = d_inode(fsi->ptmx_dentry); - inode->i_mode = S_IFCHR|fsi->mount_opts.ptmxmode; - } + fsi->ptmx_inode->i_mode = S_IFCHR|fsi->mount_opts.ptmxmode; } -static int devpts_remount(struct super_block *sb, int *flags, char *data) +static int devpts_reconfigure(struct fs_context *fc) { - int err; - struct pts_fs_info *fsi = DEVPTS_SB(sb); - struct pts_mount_opts *opts = &fsi->mount_opts; + struct pts_fs_info *fsi = DEVPTS_SB(fc->root->d_sb); + struct pts_fs_info *new = fc->s_fs_info; - err = parse_mount_options(data, PARSE_REMOUNT, opts); + /* Apply the revised options. We don't want to change ->reserve. + * Ideally, we'd update each option conditionally on it having been + * explicitly changed, but the default is to reset everything so that + * would break UAPI... + */ + fsi->mount_opts.setuid = new->mount_opts.setuid; + fsi->mount_opts.setgid = new->mount_opts.setgid; + fsi->mount_opts.uid = new->mount_opts.uid; + fsi->mount_opts.gid = new->mount_opts.gid; + fsi->mount_opts.mode = new->mount_opts.mode; + fsi->mount_opts.ptmxmode = new->mount_opts.ptmxmode; + fsi->mount_opts.max = new->mount_opts.max; /* * parse_mount_options() restores options to default values @@ -341,7 +330,7 @@ static int devpts_remount(struct super_block *sb, int *flags, char *data) */ update_ptmx_mode(fsi); - return err; + return 0; } static int devpts_show_options(struct seq_file *seq, struct dentry *root) @@ -365,54 +354,28 @@ static int devpts_show_options(struct seq_file *seq, struct dentry *root) static const struct super_operations devpts_sops = { .statfs = simple_statfs, - .remount_fs = devpts_remount, .show_options = devpts_show_options, }; -static void *new_pts_fs_info(struct super_block *sb) -{ - struct pts_fs_info *fsi; - - fsi = kzalloc(sizeof(struct pts_fs_info), GFP_KERNEL); - if (!fsi) - return NULL; - - ida_init(&fsi->allocated_ptys); - fsi->mount_opts.mode = DEVPTS_DEFAULT_MODE; - fsi->mount_opts.ptmxmode = DEVPTS_DEFAULT_PTMX_MODE; - fsi->sb = sb; - - return fsi; -} - -static int -devpts_fill_super(struct super_block *s, void *data, int silent) +static int devpts_fill_super(struct super_block *s, struct fs_context *fc) { + struct pts_fs_info *fsi = DEVPTS_SB(s); struct inode *inode; - int error; s->s_iflags &= ~SB_I_NODEV; s->s_blocksize = 1024; s->s_blocksize_bits = 10; s->s_magic = DEVPTS_SUPER_MAGIC; s->s_op = &devpts_sops; + s->s_d_flags = DCACHE_DONTCACHE; s->s_time_gran = 1; + fsi->sb = s; - error = -ENOMEM; - s->s_fs_info = new_pts_fs_info(s); - if (!s->s_fs_info) - goto fail; - - error = parse_mount_options(data, PARSE_MOUNT, &DEVPTS_SB(s)->mount_opts); - if (error) - goto fail; - - error = -ENOMEM; inode = new_inode(s); if (!inode) - goto fail; + return -ENOMEM; inode->i_ino = 1; - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); + simple_inode_init_ts(inode); inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; @@ -421,31 +384,60 @@ devpts_fill_super(struct super_block *s, void *data, int silent) s->s_root = d_make_root(inode); if (!s->s_root) { pr_err("get root dentry failed\n"); - goto fail; + return -ENOMEM; } - error = mknod_ptmx(s); - if (error) - goto fail_dput; - - return 0; -fail_dput: - dput(s->s_root); - s->s_root = NULL; -fail: - return error; + return mknod_ptmx(s, fc); } /* - * devpts_mount() + * devpts_get_tree() * * Mount a new (private) instance of devpts. PTYs created in this * instance are independent of the PTYs in other devpts instances. */ -static struct dentry *devpts_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data) +static int devpts_get_tree(struct fs_context *fc) { - return mount_nodev(fs_type, flags, data, devpts_fill_super); + return get_tree_nodev(fc, devpts_fill_super); +} + +static void devpts_free_fc(struct fs_context *fc) +{ + kfree(fc->s_fs_info); +} + +static const struct fs_context_operations devpts_context_ops = { + .free = devpts_free_fc, + .parse_param = devpts_parse_param, + .get_tree = devpts_get_tree, + .reconfigure = devpts_reconfigure, +}; + +/* + * Set up the filesystem mount context. + */ +static int devpts_init_fs_context(struct fs_context *fc) +{ + struct pts_fs_info *fsi; + + fsi = kzalloc(sizeof(struct pts_fs_info), GFP_KERNEL); + if (!fsi) + return -ENOMEM; + + ida_init(&fsi->allocated_ptys); + fsi->mount_opts.uid = GLOBAL_ROOT_UID; + fsi->mount_opts.gid = GLOBAL_ROOT_GID; + fsi->mount_opts.mode = DEVPTS_DEFAULT_MODE; + fsi->mount_opts.ptmxmode = DEVPTS_DEFAULT_PTMX_MODE; + fsi->mount_opts.max = NR_UNIX98_PTY_MAX; + + if (fc->purpose == FS_CONTEXT_FOR_MOUNT && + current->nsproxy->mnt_ns == init_task.nsproxy->mnt_ns) + fsi->mount_opts.reserve = true; + + fc->s_fs_info = fsi; + fc->ops = &devpts_context_ops; + return 0; } static void devpts_kill_sb(struct super_block *sb) @@ -455,12 +447,13 @@ static void devpts_kill_sb(struct super_block *sb) if (fsi) ida_destroy(&fsi->allocated_ptys); kfree(fsi); - kill_litter_super(sb); + kill_anon_super(sb); } static struct file_system_type devpts_fs_type = { .name = "devpts", - .mount = devpts_mount, + .init_fs_context = devpts_init_fs_context, + .parameters = devpts_param_specs, .kill_sb = devpts_kill_sb, .fs_flags = FS_USERNS_MOUNT, }; @@ -472,54 +465,35 @@ static struct file_system_type devpts_fs_type = { int devpts_new_index(struct pts_fs_info *fsi) { - int index; - int ida_ret; - -retry: - if (!ida_pre_get(&fsi->allocated_ptys, GFP_KERNEL)) - return -ENOMEM; + int index = -ENOSPC; - mutex_lock(&allocated_ptys_lock); - if (pty_count >= (pty_limit - - (fsi->mount_opts.reserve ? 0 : pty_reserve))) { - mutex_unlock(&allocated_ptys_lock); - return -ENOSPC; - } + if (atomic_inc_return(&pty_count) >= (pty_limit - + (fsi->mount_opts.reserve ? 0 : pty_reserve))) + goto out; - ida_ret = ida_get_new(&fsi->allocated_ptys, &index); - if (ida_ret < 0) { - mutex_unlock(&allocated_ptys_lock); - if (ida_ret == -EAGAIN) - goto retry; - return -EIO; - } + index = ida_alloc_max(&fsi->allocated_ptys, fsi->mount_opts.max - 1, + GFP_KERNEL); - if (index >= fsi->mount_opts.max) { - ida_remove(&fsi->allocated_ptys, index); - mutex_unlock(&allocated_ptys_lock); - return -ENOSPC; - } - pty_count++; - mutex_unlock(&allocated_ptys_lock); +out: + if (index < 0) + atomic_dec(&pty_count); return index; } void devpts_kill_index(struct pts_fs_info *fsi, int idx) { - mutex_lock(&allocated_ptys_lock); - ida_remove(&fsi->allocated_ptys, idx); - pty_count--; - mutex_unlock(&allocated_ptys_lock); + ida_free(&fsi->allocated_ptys, idx); + atomic_dec(&pty_count); } /** * devpts_pty_new -- create a new inode in /dev/pts/ - * @ptmx_inode: inode of the master - * @device: major+minor of the node to be created + * @fsi: Filesystem info for this instance. * @index: used as a name of the node * @priv: what's given back by devpts_get_priv * - * The created inode is returned. Remove it from /dev/pts/ by devpts_pty_kill. + * The dentry for the created inode is returned. + * Remove it from /dev/pts/ with devpts_pty_kill(). */ struct dentry *devpts_pty_new(struct pts_fs_info *fsi, int index, void *priv) { @@ -540,27 +514,26 @@ struct dentry *devpts_pty_new(struct pts_fs_info *fsi, int index, void *priv) inode->i_ino = index + 3; inode->i_uid = opts->setuid ? opts->uid : current_fsuid(); inode->i_gid = opts->setgid ? opts->gid : current_fsgid(); - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); + simple_inode_init_ts(inode); init_special_inode(inode, S_IFCHR|opts->mode, MKDEV(UNIX98_PTY_SLAVE_MAJOR, index)); sprintf(s, "%d", index); dentry = d_alloc_name(root, s); - if (dentry) { - dentry->d_fsdata = priv; - d_add(dentry, inode); - fsnotify_create(d_inode(root), dentry); - } else { + if (!dentry) { iput(inode); - dentry = ERR_PTR(-ENOMEM); + return ERR_PTR(-ENOMEM); } - - return dentry; + dentry->d_fsdata = priv; + d_make_persistent(dentry, inode); + fsnotify_create(d_inode(root), dentry); + dput(dentry); + return dentry; // borrowed } /** * devpts_get_priv -- get private data for a slave - * @pts_inode: inode of the slave + * @dentry: dentry of the slave * * Returns whatever was passed as priv in devpts_pty_new for a given inode. */ @@ -573,7 +546,7 @@ void *devpts_get_priv(struct dentry *dentry) /** * devpts_pty_kill -- remove inode form /dev/pts/ - * @inode: inode of the slave to be removed + * @dentry: dentry of the slave to be removed * * This is an inverse operation of devpts_pty_new. */ @@ -583,15 +556,16 @@ void devpts_pty_kill(struct dentry *dentry) dentry->d_fsdata = NULL; drop_nlink(dentry->d_inode); - d_delete(dentry); - dput(dentry); /* d_alloc_name() in devpts_pty_new() */ + d_drop(dentry); + fsnotify_unlink(d_inode(dentry->d_parent), dentry); + d_make_discardable(dentry); } static int __init init_devpts_fs(void) { int err = register_filesystem(&devpts_fs_type); if (!err) { - register_sysctl_table(pty_root_table); + register_sysctl("kernel/pty", pty_table); } return err; } |
