From: Ram Pai <[email protected]>
/proc/mounts in its current state fails to disambiguate bind mounts,
especially when the bind mount is subrooted. Also it does not capture
propagation state of the mounts(shared-subtree). The following patch
addresses the problem.
The patch adds '/proc/<pid>/mountinfo' which contains a superset of
information in '/proc/<pid>/mounts'. The following fields are added:
mntid -- is a unique identifier of the mount
parent -- the id of the parent mount
major:minor -- value of st_dev for files on that filesystem
dir -- the subdir in the filesystem which forms the root of this mount
propagation-type in the form of <propagation_flag>[:<mntid>][,...]
note: 'shared' flag is followed by the mntid of its peer mount
'slave' flag is followed by the mntid of its master mount
'private' flag stands by itself
'unbindable' flag stands by itself
Also mount options are split into two fileds, the first containing the
per mount flags, the second the per super block options.
Here is a sample cat /proc/mounts after execution the following commands:
mount --bind /mnt /mnt
mount --make-shared /mnt
mount --bind /mnt/1 /var
mount --make-slave /var
mount --make-shared /var
mount --bind /var/abc /tmp
mount --make-unbindable /proc
2 2 0:1 rootfs rootfs / / rw rw private
16 2 98:0 ext2 /dev/root / / rw rw private
17 16 0:3 proc /proc / /proc rw rw unbindable
18 16 0:10 devpts devpts /dev/pts / rw rw private
19 16 98:0 ext2 /dev/root /mnt /mnt rw rw shared:19
20 16 98:0 ext2 /dev/root /mnt/1 /var rw rw shared:21,slave:19
21 16 98:0 ext2 /dev/root /mnt/1/abc /tmp rw rw shared:20,slave:19
For example, the last line indicates that :
1) The mount is a shared mount.
2) Its peer mount of mount with id 20
3) It is also a slave mount of the master-mount with the id 19
4) The filesystem on device with major/minor number 98:0 and subdirectory
mnt/1/abc makes the root directory of this mount.
5) And finally the mount with id 16 is its parent.
[[email protected]]
- new file, rearrange fields
- for mount ID's use IDA (from the IDR library) instead of a 32bit
counter, which could overflow
- print canonical ID's (smallest one within the peer group) for peers
and master, this is more useful, than a random ID within the same namespace
- fix a couple of small bugs
- remove inlines
- style fixes
Signed-off-by: Ram Pai <[email protected]>
Signed-off-by: Miklos Szeredi <[email protected]>
---
Index: linux/fs/dcache.c
===================================================================
--- linux.orig/fs/dcache.c 2008-01-28 14:54:19.000000000 +0100
+++ linux/fs/dcache.c 2008-01-28 14:54:50.000000000 +0100
@@ -1890,6 +1890,60 @@ char *dynamic_dname(struct dentry *dentr
return memcpy(buffer, temp, sz);
}
+static int prepend(char **buffer, int *buflen, const char *str,
+ int namelen)
+{
+ *buflen -= namelen;
+ if (*buflen < 0)
+ return 1;
+ *buffer -= namelen;
+ memcpy(*buffer, str, namelen);
+ return 0;
+}
+
+/*
+ * Write full pathname from the root of the filesystem into the buffer.
+ */
+char *dentry_path(struct dentry *dentry, char *buf, int buflen)
+{
+ char *end = buf + buflen;
+ char *retval;
+
+ spin_lock(&dcache_lock);
+ prepend(&end, &buflen, "\0", 1);
+ if (!IS_ROOT(dentry) && d_unhashed(dentry)) {
+ if (prepend(&end, &buflen, "//deleted", 9))
+ goto Elong;
+ }
+ if (buflen < 1)
+ goto Elong;
+ /* Get '/' right */
+ retval = end-1;
+ *retval = '/';
+
+ for (;;) {
+ struct dentry *parent;
+ if (IS_ROOT(dentry))
+ break;
+
+ parent = dentry->d_parent;
+ prefetch(parent);
+
+ if (prepend(&end, &buflen, dentry->d_name.name,
+ dentry->d_name.len) ||
+ prepend(&end, &buflen, "/", 1))
+ goto Elong;
+
+ retval = end;
+ dentry = parent;
+ }
+ spin_unlock(&dcache_lock);
+ return retval;
+Elong:
+ spin_unlock(&dcache_lock);
+ return ERR_PTR(-ENAMETOOLONG);
+}
+
/*
* NOTE! The user-level library version returns a
* character pointer. The kernel system call just
Index: linux/fs/namespace.c
===================================================================
--- linux.orig/fs/namespace.c 2008-01-28 14:54:19.000000000 +0100
+++ linux/fs/namespace.c 2008-01-28 14:54:50.000000000 +0100
@@ -27,6 +27,7 @@
#include <linux/mount.h>
#include <linux/ramfs.h>
#include <linux/log2.h>
+#include <linux/idr.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
#include "pnode.h"
@@ -39,6 +40,7 @@
__cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock);
static int event;
+static DEFINE_IDA(mnt_id_ida);
static struct list_head *mount_hashtable __read_mostly;
static struct kmem_cache *mnt_cache __read_mostly;
@@ -58,10 +60,41 @@ static inline unsigned long hash(struct
#define MNT_WRITER_UNDERFLOW_LIMIT -(1<<16)
+static int mnt_alloc_id(struct vfsmount *mnt)
+{
+ int res;
+
+ retry:
+ spin_lock(&vfsmount_lock);
+ res = ida_get_new(&mnt_id_ida, &mnt->mnt_id);
+ spin_unlock(&vfsmount_lock);
+ if (res == -EAGAIN) {
+ if (ida_pre_get(&mnt_id_ida, GFP_KERNEL))
+ goto retry;
+ res = -ENOMEM;
+ }
+ return res;
+}
+
+static void mnt_free_id(struct vfsmount *mnt)
+{
+ spin_lock(&vfsmount_lock);
+ ida_remove(&mnt_id_ida, mnt->mnt_id);
+ spin_unlock(&vfsmount_lock);
+}
+
struct vfsmount *alloc_vfsmnt(const char *name)
{
struct vfsmount *mnt = kmem_cache_zalloc(mnt_cache, GFP_KERNEL);
if (mnt) {
+ int err;
+
+ err = mnt_alloc_id(mnt);
+ if (err) {
+ kmem_cache_free(mnt_cache, mnt);
+ return NULL;
+ }
+
atomic_set(&mnt->mnt_count, 1);
INIT_LIST_HEAD(&mnt->mnt_hash);
INIT_LIST_HEAD(&mnt->mnt_child);
@@ -338,6 +371,7 @@ EXPORT_SYMBOL(simple_set_mnt);
void free_vfsmnt(struct vfsmount *mnt)
{
kfree(mnt->mnt_devname);
+ mnt_free_id(mnt);
kmem_cache_free(mnt_cache, mnt);
}
@@ -640,20 +674,30 @@ static void m_stop(struct seq_file *m, v
up_read(&namespace_sem);
}
-static int show_vfsmnt(struct seq_file *m, void *v)
+struct proc_fs_info {
+ int flag;
+ const char *str;
+};
+
+static void show_sb_opts(struct seq_file *m, struct super_block *sb)
{
- struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list);
- int err = 0;
- static struct proc_fs_info {
- int flag;
- char *str;
- } fs_info[] = {
+ static const struct proc_fs_info fs_info[] = {
{ MS_SYNCHRONOUS, ",sync" },
{ MS_DIRSYNC, ",dirsync" },
{ MS_MANDLOCK, ",mand" },
{ 0, NULL }
};
- static struct proc_fs_info mnt_info[] = {
+ const struct proc_fs_info *fs_infop;
+
+ for (fs_infop = fs_info; fs_infop->flag; fs_infop++) {
+ if (sb->s_flags & fs_infop->flag)
+ seq_puts(m, fs_infop->str);
+ }
+}
+
+static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt)
+{
+ static const struct proc_fs_info mnt_info[] = {
{ MNT_NOSUID, ",nosuid" },
{ MNT_NODEV, ",nodev" },
{ MNT_NOEXEC, ",noexec" },
@@ -662,40 +706,96 @@ static int show_vfsmnt(struct seq_file *
{ MNT_RELATIME, ",relatime" },
{ 0, NULL }
};
- struct proc_fs_info *fs_infop;
+ const struct proc_fs_info *fs_infop;
+
+ for (fs_infop = mnt_info; fs_infop->flag; fs_infop++) {
+ if (mnt->mnt_flags & fs_infop->flag)
+ seq_puts(m, fs_infop->str);
+ }
+}
+
+static void show_type(struct seq_file *m, struct super_block *sb)
+{
+ mangle(m, sb->s_type->name);
+ if (sb->s_subtype && sb->s_subtype[0]) {
+ seq_putc(m, '.');
+ mangle(m, sb->s_subtype);
+ }
+}
+
+static int show_vfsmnt(struct seq_file *m, void *v)
+{
+ struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list);
+ int err = 0;
struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none");
seq_putc(m, ' ');
seq_path(m, &mnt_path, " \t\n\\");
seq_putc(m, ' ');
- mangle(m, mnt->mnt_sb->s_type->name);
- if (mnt->mnt_sb->s_subtype && mnt->mnt_sb->s_subtype[0]) {
- seq_putc(m, '.');
- mangle(m, mnt->mnt_sb->s_subtype);
- }
+ show_type(m, mnt->mnt_sb);
seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw");
- for (fs_infop = fs_info; fs_infop->flag; fs_infop++) {
- if (mnt->mnt_sb->s_flags & fs_infop->flag)
- seq_puts(m, fs_infop->str);
- }
- for (fs_infop = mnt_info; fs_infop->flag; fs_infop++) {
- if (mnt->mnt_flags & fs_infop->flag)
- seq_puts(m, fs_infop->str);
- }
+ show_sb_opts(m, mnt->mnt_sb);
+ show_mnt_opts(m, mnt);
if (mnt->mnt_sb->s_op->show_options)
err = mnt->mnt_sb->s_op->show_options(m, mnt);
seq_puts(m, " 0 0\n");
return err;
}
-struct seq_operations mounts_op = {
+const struct seq_operations mounts_op = {
.start = m_start,
.next = m_next,
.stop = m_stop,
.show = show_vfsmnt
};
+static int show_mountinfo(struct seq_file *m, void *v)
+{
+ struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list);
+ struct super_block *sb = mnt->mnt_sb;
+ struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
+ int err = 0;
+
+ seq_printf(m, "%i %i %u:%u ", mnt->mnt_id, mnt->mnt_parent->mnt_id,
+ MAJOR(sb->s_dev), MINOR(sb->s_dev));
+ show_type(m, sb);
+ seq_putc(m, ' ');
+ mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none");
+ seq_putc(m, ' ');
+ seq_dentry(m, mnt->mnt_root, " \t\n\\");
+ seq_putc(m, ' ');
+ seq_path(m, &mnt_path, " \t\n\\");
+ seq_putc(m, ' ');
+ seq_puts(m, mnt->mnt_flags & MNT_READONLY ? "ro" : "rw");
+ show_mnt_opts(m, mnt);
+ seq_putc(m, ' ');
+ seq_puts(m, sb->s_flags & MS_RDONLY ? "ro" : "rw");
+ show_sb_opts(m, sb);
+ if (sb->s_op->show_options)
+ err = sb->s_op->show_options(m, mnt);
+ if (IS_MNT_SHARED(mnt)) {
+ seq_printf(m, " shared:%i", get_peer_group_id(mnt));
+ if (IS_MNT_SLAVE(mnt))
+ seq_printf(m, ",slave:%i", get_master_id(mnt));
+ } else if (IS_MNT_SLAVE(mnt)) {
+ seq_printf(m, " slave:%i", get_master_id(mnt));
+ } else if (IS_MNT_UNBINDABLE(mnt)) {
+ seq_printf(m, " unbindable");
+ } else {
+ seq_printf(m, " private");
+ }
+ seq_putc(m, '\n');
+ return err;
+}
+
+const struct seq_operations mountinfo_op = {
+ .start = m_start,
+ .next = m_next,
+ .stop = m_stop,
+ .show = show_mountinfo,
+};
+
static int show_vfsstat(struct seq_file *m, void *v)
{
struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list);
@@ -716,7 +816,7 @@ static int show_vfsstat(struct seq_file
/* file system type */
seq_puts(m, "with fstype ");
- mangle(m, mnt->mnt_sb->s_type->name);
+ show_type(m, mnt->mnt_sb);
/* optional statistics */
if (mnt->mnt_sb->s_op->show_stats) {
@@ -728,7 +828,7 @@ static int show_vfsstat(struct seq_file
return err;
}
-struct seq_operations mountstats_op = {
+const struct seq_operations mountstats_op = {
.start = m_start,
.next = m_next,
.stop = m_stop,
Index: linux/fs/seq_file.c
===================================================================
--- linux.orig/fs/seq_file.c 2008-01-28 14:54:19.000000000 +0100
+++ linux/fs/seq_file.c 2008-01-28 14:54:50.000000000 +0100
@@ -349,28 +349,40 @@ int seq_printf(struct seq_file *m, const
}
EXPORT_SYMBOL(seq_printf);
+static char *mangle_path(char *s, char *p, char *esc)
+{
+ while (s <= p) {
+ char c = *p++;
+ if (!c) {
+ return s;
+ } else if (!strchr(esc, c)) {
+ *s++ = c;
+ } else if (s + 4 > p) {
+ break;
+ } else {
+ *s++ = '\\';
+ *s++ = '0' + ((c & 0300) >> 6);
+ *s++ = '0' + ((c & 070) >> 3);
+ *s++ = '0' + (c & 07);
+ }
+ }
+ return NULL;
+}
+
+/*
+ * return the absolute path of 'dentry' residing in mount 'mnt'.
+ */
int seq_path(struct seq_file *m, struct path *path, char *esc)
{
if (m->count < m->size) {
char *s = m->buf + m->count;
char *p = d_path(path, s, m->size - m->count);
if (!IS_ERR(p)) {
- while (s <= p) {
- char c = *p++;
- if (!c) {
- p = m->buf + m->count;
- m->count = s - m->buf;
- return s - p;
- } else if (!strchr(esc, c)) {
- *s++ = c;
- } else if (s + 4 > p) {
- break;
- } else {
- *s++ = '\\';
- *s++ = '0' + ((c & 0300) >> 6);
- *s++ = '0' + ((c & 070) >> 3);
- *s++ = '0' + (c & 07);
- }
+ s = mangle_path(s, p, esc);
+ if (s) {
+ p = m->buf + m->count;
+ m->count = s - m->buf;
+ return s - p;
}
}
}
@@ -379,6 +391,28 @@ int seq_path(struct seq_file *m, struct
}
EXPORT_SYMBOL(seq_path);
+/*
+ * returns the path of the 'dentry' from the root of its filesystem.
+ */
+int seq_dentry(struct seq_file *m, struct dentry *dentry, char *esc)
+{
+ if (m->count < m->size) {
+ char *s = m->buf + m->count;
+ char *p = dentry_path(dentry, s, m->size - m->count);
+ if (!IS_ERR(p)) {
+ s = mangle_path(s, p, esc);
+ if (s) {
+ p = m->buf + m->count;
+ m->count = s - m->buf;
+ return s - p;
+ }
+ }
+ }
+ m->count = m->size;
+ return -1;
+}
+EXPORT_SYMBOL(seq_dentry);
+
static void *single_start(struct seq_file *p, loff_t *pos)
{
return NULL + (*pos == 0);
Index: linux/include/linux/dcache.h
===================================================================
--- linux.orig/include/linux/dcache.h 2008-01-28 14:54:19.000000000 +0100
+++ linux/include/linux/dcache.h 2008-01-28 14:54:50.000000000 +0100
@@ -302,6 +302,7 @@ extern int d_validate(struct dentry *, s
extern char *dynamic_dname(struct dentry *, char *, int, const char *, ...);
extern char *d_path(struct path *, char *, int);
+extern char *dentry_path(struct dentry *, char *, int);
/* Allocation counts.. */
Index: linux/include/linux/seq_file.h
===================================================================
--- linux.orig/include/linux/seq_file.h 2008-01-28 14:54:19.000000000 +0100
+++ linux/include/linux/seq_file.h 2008-01-28 14:54:50.000000000 +0100
@@ -10,6 +10,7 @@ struct seq_operations;
struct file;
struct path;
struct inode;
+struct dentry;
struct seq_file {
char *buf;
@@ -43,6 +44,7 @@ int seq_printf(struct seq_file *, const
__attribute__ ((format (printf,2,3)));
int seq_path(struct seq_file *, struct path *, char *);
+int seq_dentry(struct seq_file *, struct dentry *, char *);
int single_open(struct file *, int (*)(struct seq_file *, void *), void *);
int single_release(struct inode *, struct file *);
Index: linux/fs/pnode.c
===================================================================
--- linux.orig/fs/pnode.c 2008-01-28 14:54:19.000000000 +0100
+++ linux/fs/pnode.c 2008-01-28 14:54:50.000000000 +0100
@@ -27,6 +27,41 @@ static inline struct vfsmount *next_slav
return list_entry(p->mnt_slave.next, struct vfsmount, mnt_slave);
}
+static int __peer_group_id(struct vfsmount *mnt)
+{
+ struct vfsmount *m;
+ int id = mnt->mnt_id;
+
+ for (m = next_peer(mnt); m != mnt; m = next_peer(m))
+ id = min(id, m->mnt_id);
+
+ return id;
+}
+
+/* return the smallest ID within the peer group */
+int get_peer_group_id(struct vfsmount *mnt)
+{
+ int id;
+
+ spin_lock(&vfsmount_lock);
+ id = __peer_group_id(mnt);
+ spin_unlock(&vfsmount_lock);
+
+ return id;
+}
+
+/* return the smallest ID within the master's peer group */
+int get_master_id(struct vfsmount *mnt)
+{
+ int id;
+
+ spin_lock(&vfsmount_lock);
+ id = __peer_group_id(mnt->mnt_master);
+ spin_unlock(&vfsmount_lock);
+
+ return id;
+}
+
static int do_make_slave(struct vfsmount *mnt)
{
struct vfsmount *peer_mnt = mnt, *master = mnt->mnt_master;
Index: linux/fs/pnode.h
===================================================================
--- linux.orig/fs/pnode.h 2008-01-28 14:54:19.000000000 +0100
+++ linux/fs/pnode.h 2008-01-28 14:54:50.000000000 +0100
@@ -35,4 +35,6 @@ int propagate_mnt(struct vfsmount *, str
struct list_head *);
int propagate_umount(struct list_head *);
int propagate_mount_busy(struct vfsmount *, int);
+int get_peer_group_id(struct vfsmount *);
+int get_master_id(struct vfsmount *);
#endif /* _LINUX_PNODE_H */
Index: linux/include/linux/mount.h
===================================================================
--- linux.orig/include/linux/mount.h 2008-01-28 14:54:19.000000000 +0100
+++ linux/include/linux/mount.h 2008-01-28 14:54:50.000000000 +0100
@@ -56,6 +56,7 @@ struct vfsmount {
struct list_head mnt_slave; /* slave list entry */
struct vfsmount *mnt_master; /* slave is on master->mnt_slave_list */
struct mnt_namespace *mnt_ns; /* containing namespace */
+ int mnt_id; /* mount identifier */
/*
* We put mnt_count & mnt_expiry_mark at the end of struct vfsmount
* to let these frequently modified fields in a separate cache line
Index: linux/fs/proc/base.c
===================================================================
--- linux.orig/fs/proc/base.c 2008-01-28 14:54:19.000000000 +0100
+++ linux/fs/proc/base.c 2008-01-28 14:54:50.000000000 +0100
@@ -435,13 +435,13 @@ static const struct inode_operations pro
.setattr = proc_setattr,
};
-extern struct seq_operations mounts_op;
struct proc_mounts {
struct seq_file m;
int event;
};
-static int mounts_open(struct inode *inode, struct file *file)
+static int mounts_open_common(struct inode *inode, struct file *file,
+ const struct seq_operations *op)
{
struct task_struct *task = get_proc_task(inode);
struct nsproxy *nsp;
@@ -467,7 +467,7 @@ static int mounts_open(struct inode *ino
p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL);
if (p) {
file->private_data = &p->m;
- ret = seq_open(file, &mounts_op);
+ ret = seq_open(file, op);
if (!ret) {
p->m.private = ns;
p->event = ns->event;
@@ -506,6 +506,11 @@ static unsigned mounts_poll(struct file
return res;
}
+static int mounts_open(struct inode *inode, struct file *file)
+{
+ return mounts_open_common(inode, file, &mounts_op);
+}
+
static const struct file_operations proc_mounts_operations = {
.open = mounts_open,
.read = seq_read,
@@ -514,38 +519,22 @@ static const struct file_operations proc
.poll = mounts_poll,
};
-extern struct seq_operations mountstats_op;
-static int mountstats_open(struct inode *inode, struct file *file)
+static int mountinfo_open(struct inode *inode, struct file *file)
{
- int ret = seq_open(file, &mountstats_op);
-
- if (!ret) {
- struct seq_file *m = file->private_data;
- struct nsproxy *nsp;
- struct mnt_namespace *mnt_ns = NULL;
- struct task_struct *task = get_proc_task(inode);
-
- if (task) {
- rcu_read_lock();
- nsp = task_nsproxy(task);
- if (nsp) {
- mnt_ns = nsp->mnt_ns;
- if (mnt_ns)
- get_mnt_ns(mnt_ns);
- }
- rcu_read_unlock();
+ return mounts_open_common(inode, file, &mountinfo_op);
+}
- put_task_struct(task);
- }
+static const struct file_operations proc_mountinfo_operations = {
+ .open = mountinfo_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = mounts_release,
+ .poll = mounts_poll,
+};
- if (mnt_ns)
- m->private = mnt_ns;
- else {
- seq_release(inode, file);
- ret = -EINVAL;
- }
- }
- return ret;
+static int mountstats_open(struct inode *inode, struct file *file)
+{
+ return mounts_open_common(inode, file, &mountstats_op);
}
static const struct file_operations proc_mountstats_operations = {
@@ -2220,6 +2209,7 @@ static const struct pid_entry tgid_base_
LNK("root", root),
LNK("exe", exe),
REG("mounts", S_IRUGO, mounts),
+ REG("mountinfo", S_IRUGO, mountinfo),
REG("mountstats", S_IRUSR, mountstats),
#ifdef CONFIG_PROC_PAGE_MONITOR
REG("clear_refs", S_IWUSR, clear_refs),
@@ -2548,6 +2538,7 @@ static const struct pid_entry tid_base_s
LNK("root", root),
LNK("exe", exe),
REG("mounts", S_IRUGO, mounts),
+ REG("mountinfo", S_IRUGO, mountinfo),
#ifdef CONFIG_PROC_PAGE_MONITOR
REG("clear_refs", S_IWUSR, clear_refs),
REG("smaps", S_IRUGO, smaps),
Index: linux/include/linux/mnt_namespace.h
===================================================================
--- linux.orig/include/linux/mnt_namespace.h 2008-01-28 14:54:19.000000000 +0100
+++ linux/include/linux/mnt_namespace.h 2008-01-28 14:54:50.000000000 +0100
@@ -37,5 +37,9 @@ static inline void get_mnt_ns(struct mnt
atomic_inc(&ns->count);
}
+extern const struct seq_operations mounts_op;
+extern const struct seq_operations mountinfo_op;
+extern const struct seq_operations mountstats_op;
+
#endif
#endif
Index: linux/Documentation/filesystems/proc.txt
===================================================================
--- linux.orig/Documentation/filesystems/proc.txt 2008-01-28 14:54:19.000000000 +0100
+++ linux/Documentation/filesystems/proc.txt 2008-01-28 14:54:50.000000000 +0100
@@ -43,6 +43,7 @@ Table of Contents
2.13 /proc/<pid>/oom_score - Display current oom-killer score
2.14 /proc/<pid>/io - Display the IO accounting fields
2.15 /proc/<pid>/coredump_filter - Core dump filtering settings
+ 2.16 /proc/<pid>/mountinfo - Information about mounts
------------------------------------------------------------------------------
Preface
@@ -2267,4 +2268,39 @@ For example:
$ echo 0x7 > /proc/self/coredump_filter
$ ./some_program
+2.16 /proc/<pid>/mountinfo - Information about mounts
+--------------------------------------------------------
+
+This file contains lines of the form:
+
+MNTID PARENT MAJOR:MINOR TYPE SOURCE ROOT MOUNTPOINT MNTOPTS SBOPTS PROPAGATION
+
+MNTID: unique identifier of the mount (may be reused after umount)
+PARENT: ID of parent (or of self for the top of the mount tree)
+MAJOR:MINOR: value of st_dev for files on filesystem
+TYPE: filesystem type
+SOURCE: source of mount (name of the device for block filesystems)
+ROOT: root of the mount within the filesystem
+MOUNTPOINT: mount point
+MNTOPTS: per mount options
+SBOPTS: per super block options
+PROPAGATION: propagation type
+
+propagation type: <propagation_flag>[:<mntid>][,...]
+ note: 'shared' flag is followed by the mntid of its peer mount
+ 'slave' flag is followed by the mntid of its master mount
+ 'private' flag stands by itself
+ 'unbindable' flag stands by itself
+
+The 'mntid' used in the propagation type is a canonical ID of the peer
+group (currently the smallest ID within the group is used for this
+purpose, but this should not be relied on). Since mounts can be added
+or removed from the peer group, this ID only guaranteed to stay the
+same on a static propagation tree.
+
+For more information see:
+
+ Documentation/filesystems/sharedsubtree.txt
+
+
------------------------------------------------------------------------------
On Tue, 29 Jan 2008 14:57:29 +0100
Miklos Szeredi <[email protected]> wrote:
> From: Ram Pai <[email protected]>
>
> /proc/mounts in its current state fails to disambiguate bind mounts,
> especially when the bind mount is subrooted. Also it does not capture
> propagation state of the mounts(shared-subtree). The following patch
> addresses the problem.
>
> The patch adds '/proc/<pid>/mountinfo' which contains a superset of
> information in '/proc/<pid>/mounts'. The following fields are added:
>
> mntid -- is a unique identifier of the mount
> parent -- the id of the parent mount
> major:minor -- value of st_dev for files on that filesystem
> dir -- the subdir in the filesystem which forms the root of this mount
> propagation-type in the form of <propagation_flag>[:<mntid>][,...]
> note: 'shared' flag is followed by the mntid of its peer mount
> 'slave' flag is followed by the mntid of its master mount
> 'private' flag stands by itself
> 'unbindable' flag stands by itself
>
> Also mount options are split into two fileds, the first containing the
> per mount flags, the second the per super block options.
>
> Here is a sample cat /proc/mounts after execution the following commands:
^^^^^^^^^^^^ /proc/<pid>/mountinfo?
>
> mount --bind /mnt /mnt
> mount --make-shared /mnt
> mount --bind /mnt/1 /var
> mount --make-slave /var
> mount --make-shared /var
> mount --bind /var/abc /tmp
> mount --make-unbindable /proc
>
> 2 2 0:1 rootfs rootfs / / rw rw private
> 16 2 98:0 ext2 /dev/root / / rw rw private
> 17 16 0:3 proc /proc / /proc rw rw unbindable
> 18 16 0:10 devpts devpts /dev/pts / rw rw private
> 19 16 98:0 ext2 /dev/root /mnt /mnt rw rw shared:19
> 20 16 98:0 ext2 /dev/root /mnt/1 /var rw rw shared:21,slave:19
> 21 16 98:0 ext2 /dev/root /mnt/1/abc /tmp rw rw shared:20,slave:19
>
> For example, the last line indicates that :
>
> 1) The mount is a shared mount.
> 2) Its peer mount of mount with id 20
> 3) It is also a slave mount of the master-mount with the id 19
> 4) The filesystem on device with major/minor number 98:0 and subdirectory
> mnt/1/abc makes the root directory of this mount.
> 5) And finally the mount with id 16 is its parent.
Boy, that's complex. It'd be nicer to have a name:value\n format, like
/proc/meminfo. But that would really imply one /proc file per mount under
/proc/<pid>/mountinfo/...
If history teaches us anything, it is that we should make this thing
extensible in ways in which we can be confident will not break existing
parsers.
Does this format ensure that? It's hard to see how.
>
> [[email protected]]
>
> - new file, rearrange fields
> - for mount ID's use IDA (from the IDR library) instead of a 32bit
> counter, which could overflow
> - print canonical ID's (smallest one within the peer group) for peers
> and master, this is more useful, than a random ID within the same namespace
> - fix a couple of small bugs
> - remove inlines
> - style fixes
>
> Signed-off-by: Ram Pai <[email protected]>
> Signed-off-by: Miklos Szeredi <[email protected]>
The patch adds a new concept of an "id" for each vfsmount (correct?).
Perhaps the semantics of that identifier should be captured in code
comments, probably in pnode.c. Things like:
- They're always unique
- They are allocated on a first-fit-from-zero basis (correct?)
- They are used for... ?
They are also signed, which seems inappropriate.
The new exported-to-everyone dentry_path() probably could do with a bit
more documentation - it's the sort of thing which people keep on wanting
and using.
How does dentry_path() differ from d_path() and why do we need both and can
we get some sharing/consolidation happening here?
Why do d_path() and dentry_path() have differing conventions for displaying
a deleted file and can we fix that?
This patch adds a lot of code which is, I guess, unused if
CONFIG_PROC_FS=n. Fixable?
> > From: Ram Pai <[email protected]>
> >
> > /proc/mounts in its current state fails to disambiguate bind mounts,
> > especially when the bind mount is subrooted. Also it does not capture
> > propagation state of the mounts(shared-subtree). The following patch
> > addresses the problem.
> >
> > The patch adds '/proc/<pid>/mountinfo' which contains a superset of
> > information in '/proc/<pid>/mounts'. The following fields are added:
> >
> > mntid -- is a unique identifier of the mount
> > parent -- the id of the parent mount
> > major:minor -- value of st_dev for files on that filesystem
> > dir -- the subdir in the filesystem which forms the root of this mount
> > propagation-type in the form of <propagation_flag>[:<mntid>][,...]
> > note: 'shared' flag is followed by the mntid of its peer mount
> > 'slave' flag is followed by the mntid of its master mount
> > 'private' flag stands by itself
> > 'unbindable' flag stands by itself
> >
> > Also mount options are split into two fileds, the first containing the
> > per mount flags, the second the per super block options.
> >
> > Here is a sample cat /proc/mounts after execution the following commands:
>
> ^^^^^^^^^^^^ /proc/<pid>/mountinfo?
> >
> > mount --bind /mnt /mnt
> > mount --make-shared /mnt
> > mount --bind /mnt/1 /var
> > mount --make-slave /var
> > mount --make-shared /var
> > mount --bind /var/abc /tmp
> > mount --make-unbindable /proc
> >
> > 2 2 0:1 rootfs rootfs / / rw rw private
> > 16 2 98:0 ext2 /dev/root / / rw rw private
> > 17 16 0:3 proc /proc / /proc rw rw unbindable
> > 18 16 0:10 devpts devpts /dev/pts / rw rw private
> > 19 16 98:0 ext2 /dev/root /mnt /mnt rw rw shared:19
> > 20 16 98:0 ext2 /dev/root /mnt/1 /var rw rw shared:21,slave:19
> > 21 16 98:0 ext2 /dev/root /mnt/1/abc /tmp rw rw shared:20,slave:19
> >
> > For example, the last line indicates that :
> >
> > 1) The mount is a shared mount.
> > 2) Its peer mount of mount with id 20
> > 3) It is also a slave mount of the master-mount with the id 19
> > 4) The filesystem on device with major/minor number 98:0 and subdirectory
> > mnt/1/abc makes the root directory of this mount.
> > 5) And finally the mount with id 16 is its parent.
>
> Boy, that's complex. It'd be nicer to have a name:value\n format, like
> /proc/meminfo. But that would really imply one /proc file per mount under
> /proc/<pid>/mountinfo/...
One disadvantage of doing it in separate files, is no good way of
indexing the mounts. The mount ID could be used, but it's not a very
meaningful thing outside the context of mount propagation.
And one advantage of doing it in a single file, is that namespace_sem
is held during the read, so that content is self-consistent (no mounts
are removed/added/moved in between).
IMO, the /proc/mounts format is quite readable, and this one doesn't
differ all that much. Maybe we could add a header line like
/proc/slabinfo. Or do the "name:value\n" thing but put all mounts
into a single file, like /proc/cpuinfo.
I don't have any strong preference, I'm OK, with the current format,
but if there are good reasons for another one, I'm willing to change
it around.
> If history teaches us anything, it is that we should make this thing
> extensible in ways in which we can be confident will not break existing
> parsers.
>
> Does this format ensure that? It's hard to see how.
Explicitly document, that adding fields at the end is OK. But I think
this is implicit in all of the multi-fields-per-line formats, and I'm
quite confident all sane parsers already ignore unknown junk at the
end of a line.
>
> >
> > [[email protected]]
> >
> > - new file, rearrange fields
> > - for mount ID's use IDA (from the IDR library) instead of a 32bit
> > counter, which could overflow
> > - print canonical ID's (smallest one within the peer group) for peers
> > and master, this is more useful, than a random ID within the same namespace
> > - fix a couple of small bugs
> > - remove inlines
> > - style fixes
> >
> > Signed-off-by: Ram Pai <[email protected]>
> > Signed-off-by: Miklos Szeredi <[email protected]>
>
> The patch adds a new concept of an "id" for each vfsmount (correct?).
> Perhaps the semantics of that identifier should be captured in code
> comments, probably in pnode.c. Things like:
>
> - They're always unique
> - They are allocated on a first-fit-from-zero basis (correct?)
Yes.
> - They are used for... ?
Just to identify mount propagation relations in this file. The
vfsmount pointer would be just as good, but we don't like to expose
things like that to userspace (and it would be hard to parse for
humans anyway).
> They are also signed, which seems inappropriate.
IDR ids are 'int' but they are always positive (AFAICT), but yeah,
maybe this is confusing.
> The new exported-to-everyone dentry_path() probably could do with a bit
> more documentation - it's the sort of thing which people keep on wanting
> and using.
OK.
> How does dentry_path() differ from d_path() and why do we need both and can
> we get some sharing/consolidation happening here?
Tried that but not easy, without removing some of the
microoptimizations in d_path(), which I'm not sure are really
important, but...
> Why do d_path() and dentry_path() have differing conventions for displaying
> a deleted file and can we fix that?
I think Ram chose a different convention in dentry_path() in order to
make sure, there was no space in the resulting path. But spaces would
be escaped anyway, so this isn't really important. So yes, this could
be fixed.
> This patch adds a lot of code which is, I guess, unused if
> CONFIG_PROC_FS=n. Fixable?
Possibly yes. A good chunk of namespace.c could be surrounded by an
#ifdef, which would save even more, than was added by this particular
patch.
Thanks,
Miklos
On Thu, 2008-01-31 at 10:17 +0100, Miklos Szeredi wrote:
> > > From: Ram Pai <[email protected]>
...snipped...
> IDR ids are 'int' but they are always positive (AFAICT), but yeah,
> maybe this is confusing.
>
> > The new exported-to-everyone dentry_path() probably could do with a bit
> > more documentation - it's the sort of thing which people keep on wanting
> > and using.
>
> OK.
>
> > How does dentry_path() differ from d_path() and why do we need both and can
> > we get some sharing/consolidation happening here?
d_path displays the path from the rootfs, whereas dentry_path displays
the path from the root of that filesystem.
>
> Tried that but not easy, without removing some of the
> microoptimizations in d_path(), which I'm not sure are really
> important, but...
this patch was intially developed with Al Viro. He preferred to keep the
two functions separate. BTW: this patch owes credits to Al Viro for his
initial set of ideas.
>
> > Why do d_path() and dentry_path() have differing conventions for displaying
> > a deleted file and can we fix that?
>
> I think Ram chose a different convention in dentry_path() in order to
> make sure, there was no space in the resulting path. But spaces would
> be escaped anyway, so this isn't really important. So yes, this could
> be fixed.
my patch was generated about a year or so back using probably the
2.6.18 code base which had the "//deleted" convention. That got copied
in my patch. But since then I see that the original code has changed to
use the " (deleted)" convention.
Yes this patch has to be changed to be consistent with the existing
code.
>
> > This patch adds a lot of code which is, I guess, unused if
> > CONFIG_PROC_FS=n. Fixable?
yes. good observation. I will send a patch with this optimization and
the above mentioned change.
RP
>
> Possibly yes. A good chunk of namespace.c could be surrounded by an
> #ifdef, which would save even more, than was added by this particular
> patch.
> Thanks,
> Miklos
1) reports deleted inode in dentry_path() consistent with that in __d_path()
2) modified __d_path() to use prepend(), reducing the size of __d_path()
3) moved all the functionality that reports mount information in /proc under
CONFIG_PROC_FS.
Could not verify if the code would work with CONFIG_PROC_FS=n, since it was
impossible to disable CONFIG_PROC_FS. Looking for ideas on how to disable
CONFIG_PROC_FS.
Signed-off-by: Ram Pai <[email protected]>
---
fs/dcache.c | 59 +++++++++++++++++++----------------------------
fs/namespace.c | 2 +
fs/seq_file.c | 2 +
include/linux/dcache.h | 3 ++
include/linux/seq_file.h | 3 ++
5 files changed, 34 insertions(+), 35 deletions(-)
Index: linux-2.6.23/fs/dcache.c
===================================================================
--- linux-2.6.23.orig/fs/dcache.c
+++ linux-2.6.23/fs/dcache.c
@@ -1747,6 +1747,17 @@ shouldnt_be_hashed:
goto shouldnt_be_hashed;
}
+static int prepend(char **buffer, int *buflen, const char *str,
+ int namelen)
+{
+ *buflen -= namelen;
+ if (*buflen < 0)
+ return 1;
+ *buffer -= namelen;
+ memcpy(*buffer, str, namelen);
+ return 0;
+}
+
/**
* d_path - return the path of a dentry
* @dentry: dentry to report
@@ -1768,17 +1779,11 @@ static char *__d_path(struct dentry *den
{
char * end = buffer+buflen;
char * retval;
- int namelen;
- *--end = '\0';
- buflen--;
- if (!IS_ROOT(dentry) && d_unhashed(dentry)) {
- buflen -= 10;
- end -= 10;
- if (buflen < 0)
+ prepend(&end, &buflen, "\0", 1);
+ if (!IS_ROOT(dentry) && d_unhashed(dentry) &&
+ prepend(&end, &buflen, " (deleted)", 10))
goto Elong;
- memcpy(end, " (deleted)", 10);
- }
if (buflen < 1)
goto Elong;
@@ -1805,13 +1810,10 @@ static char *__d_path(struct dentry *den
}
parent = dentry->d_parent;
prefetch(parent);
- namelen = dentry->d_name.len;
- buflen -= namelen + 1;
- if (buflen < 0)
+ if (prepend(&end, &buflen, dentry->d_name.name,
+ dentry->d_name.len) ||
+ prepend(&end, &buflen, "/", 1))
goto Elong;
- end -= namelen;
- memcpy(end, dentry->d_name.name, namelen);
- *--end = '/';
retval = end;
dentry = parent;
}
@@ -1819,12 +1821,9 @@ static char *__d_path(struct dentry *den
return retval;
global_root:
- namelen = dentry->d_name.len;
- buflen -= namelen;
- if (buflen < 0)
- goto Elong;
- retval -= namelen-1; /* hit the slash */
- memcpy(retval, dentry->d_name.name, namelen);
+ retval += 1; /* hit the slash */
+ if (prepend(&retval, &buflen, dentry->d_name.name, dentry->d_name.len))
+ goto Elong;
return retval;
Elong:
return ERR_PTR(-ENAMETOOLONG);
@@ -1890,17 +1889,8 @@ char *dynamic_dname(struct dentry *dentr
return memcpy(buffer, temp, sz);
}
-static int prepend(char **buffer, int *buflen, const char *str,
- int namelen)
-{
- *buflen -= namelen;
- if (*buflen < 0)
- return 1;
- *buffer -= namelen;
- memcpy(*buffer, str, namelen);
- return 0;
-}
+#ifdef CONFIG_PROC_FS
/*
* Write full pathname from the root of the filesystem into the buffer.
*/
@@ -1910,11 +1900,9 @@ char *dentry_path(struct dentry *dentry,
char *retval;
spin_lock(&dcache_lock);
- prepend(&end, &buflen, "\0", 1);
- if (!IS_ROOT(dentry) && d_unhashed(dentry)) {
- if (prepend(&end, &buflen, "//deleted", 9))
+ if (!IS_ROOT(dentry) && d_unhashed(dentry) &&
+ prepend(&end, &buflen, " (deleted)", 10))
goto Elong;
- }
if (buflen < 1)
goto Elong;
/* Get '/' right */
@@ -1943,6 +1931,7 @@ Elong:
spin_unlock(&dcache_lock);
return ERR_PTR(-ENAMETOOLONG);
}
+#endif /* CONFIG_PROC_FS */
/*
* NOTE! The user-level library version returns a
Index: linux-2.6.23/fs/namespace.c
===================================================================
--- linux-2.6.23.orig/fs/namespace.c
+++ linux-2.6.23/fs/namespace.c
@@ -609,6 +609,7 @@ void mnt_unpin(struct vfsmount *mnt)
EXPORT_SYMBOL(mnt_unpin);
+#ifdef CONFIG_PROC_FS
/* iterator */
static void *m_start(struct seq_file *m, loff_t *pos)
{
@@ -795,6 +796,7 @@ const struct seq_operations mountstats_o
.stop = m_stop,
.show = show_vfsstat,
};
+#endif /* CONFIG_PROC_FS */
/**
* may_umount_tree - check if a mount tree is busy
Index: linux-2.6.23/fs/seq_file.c
===================================================================
--- linux-2.6.23.orig/fs/seq_file.c
+++ linux-2.6.23/fs/seq_file.c
@@ -369,6 +369,7 @@ static char *mangle_path(char *s, char *
return NULL;
}
+#ifdef CONFIG_PROC_FS
/*
* return the absolute path of 'dentry' residing in mount 'mnt'.
*/
@@ -390,6 +391,7 @@ int seq_path(struct seq_file *m, struct
return -1;
}
EXPORT_SYMBOL(seq_path);
+#endif /* CONFIG_PROC_FS */
/*
* returns the path of the 'dentry' from the root of its filesystem.
Index: linux-2.6.23/include/linux/dcache.h
===================================================================
--- linux-2.6.23.orig/include/linux/dcache.h
+++ linux-2.6.23/include/linux/dcache.h
@@ -302,7 +302,10 @@ extern int d_validate(struct dentry *, s
extern char *dynamic_dname(struct dentry *, char *, int, const char *, ...);
extern char *d_path(struct path *, char *, int);
+
+#ifdef CONFIG_PROC_FS
extern char *dentry_path(struct dentry *, char *, int);
+#endif /* CONFIG_PROC_FS */
/* Allocation counts.. */
Index: linux-2.6.23/include/linux/seq_file.h
===================================================================
--- linux-2.6.23.orig/include/linux/seq_file.h
+++ linux-2.6.23/include/linux/seq_file.h
@@ -44,7 +44,10 @@ int seq_printf(struct seq_file *, const
__attribute__ ((format (printf,2,3)));
int seq_path(struct seq_file *, struct path *, char *);
+
+#ifdef CONFIG_PROC_FS
int seq_dentry(struct seq_file *, struct dentry *, char *);
+#endif /* CONFIG_PROC_FS */
int single_open(struct file *, int (*)(struct seq_file *, void *), void *);
int single_release(struct inode *, struct file *);
On Mon, 04 Feb 2008 01:15:05 -0800 Ram Pai <[email protected]> wrote:
> 1) reports deleted inode in dentry_path() consistent with that in __d_path()
> 2) modified __d_path() to use prepend(), reducing the size of __d_path()
> 3) moved all the functionality that reports mount information in /proc under
> CONFIG_PROC_FS.
>
> Could not verify if the code would work with CONFIG_PROC_FS=n, since it was
> impossible to disable CONFIG_PROC_FS. Looking for ideas on how to disable
> CONFIG_PROC_FS.
>
Do `make menuconfig', then hit '/' and search for "proc_fs".
It'll tell you that you need to set EMBEDDED=y to disable procfs.
> fs/dcache.c | 59 +++++++++++++++++++----------------------------
> fs/namespace.c | 2 +
> fs/seq_file.c | 2 +
> include/linux/dcache.h | 3 ++
> include/linux/seq_file.h | 3 ++
Please resend after testing that, thanks.
> 1) reports deleted inode in dentry_path() consistent with that in __d_path()
> 2) modified __d_path() to use prepend(), reducing the size of __d_path()
> 3) moved all the functionality that reports mount information in /proc under
> CONFIG_PROC_FS.
>
> Could not verify if the code would work with CONFIG_PROC_FS=n, since it was
> impossible to disable CONFIG_PROC_FS. Looking for ideas on how to disable
> CONFIG_PROC_FS.
>
>
>
> Signed-off-by: Ram Pai <[email protected]>
> ---
> fs/dcache.c | 59 +++++++++++++++++++----------------------------
> fs/namespace.c | 2 +
> fs/seq_file.c | 2 +
> include/linux/dcache.h | 3 ++
> include/linux/seq_file.h | 3 ++
> 5 files changed, 34 insertions(+), 35 deletions(-)
>
> Index: linux-2.6.23/fs/dcache.c
> ===================================================================
> --- linux-2.6.23.orig/fs/dcache.c
> +++ linux-2.6.23/fs/dcache.c
> @@ -1747,6 +1747,17 @@ shouldnt_be_hashed:
> goto shouldnt_be_hashed;
> }
>
> +static int prepend(char **buffer, int *buflen, const char *str,
> + int namelen)
> +{
> + *buflen -= namelen;
> + if (*buflen < 0)
> + return 1;
This is confusing. Should return -ENAMETOOLONG intead (see Chapter 16
in Documentation/CodingStyle).
> + *buffer -= namelen;
> + memcpy(*buffer, str, namelen);
> + return 0;
> +}
> +
> /**
> * d_path - return the path of a dentry
> * @dentry: dentry to report
> @@ -1768,17 +1779,11 @@ static char *__d_path(struct dentry *den
> {
> char * end = buffer+buflen;
> char * retval;
> - int namelen;
>
> - *--end = '\0';
> - buflen--;
> - if (!IS_ROOT(dentry) && d_unhashed(dentry)) {
> - buflen -= 10;
> - end -= 10;
> - if (buflen < 0)
> + prepend(&end, &buflen, "\0", 1);
> + if (!IS_ROOT(dentry) && d_unhashed(dentry) &&
> + prepend(&end, &buflen, " (deleted)", 10))
And this should test for "prepend() != 0" or "prepend() < 0" instead,
otherwise it could easily be misread as "if prepend() succeeded,
then...".
And similarly for all the later calls.
Thanks,
Miklos
On Mon, 2008-02-04 at 01:28 -0800, Andrew Morton wrote:
> On Mon, 04 Feb 2008 01:15:05 -0800 Ram Pai <[email protected]> wrote:
>
> > 1) reports deleted inode in dentry_path() consistent with that in __d_path()
> > 2) modified __d_path() to use prepend(), reducing the size of __d_path()
> > 3) moved all the functionality that reports mount information in /proc under
> > CONFIG_PROC_FS.
> >
> > Could not verify if the code would work with CONFIG_PROC_FS=n, since it was
> > impossible to disable CONFIG_PROC_FS. Looking for ideas on how to disable
> > CONFIG_PROC_FS.
this worked. thanks. There was one place in ipv4 that failed compilation
with proc_fs disabled. Fixed that. Otherwise everything compiled
cleanly.
>
> Do `make menuconfig', then hit '/' and search for "proc_fs".
>
> It'll tell you that you need to set EMBEDDED=y to disable procfs.
>
> > fs/dcache.c | 59 +++++++++++++++++++----------------------------
> > fs/namespace.c | 2 +
> > fs/seq_file.c | 2 +
> > include/linux/dcache.h | 3 ++
> > include/linux/seq_file.h | 3 ++
>
> Please resend after testing that, thanks.
with procfs disabled, the boot fails, since nash(fedora's startup
script) which links with libc.so library has dependencies on /proc. Nash
segfaults and so does init. looking for ideas.
without passing this hurdle, its hard to test the patch :(
RP