procfs: Use the proc generic infrastructure for proc/self.

I had visions at one point of splitting proc into two filesystems.  If
that had happened proc/self being the the part of proc that actually deals
with pids would have been a nice cleanup.  As it is proc/self requires
a lot of unnecessary infrastructure for a single file.

The only user visible change is that a mounted /proc for a pid namespace
that is dead now shows a broken proc symlink, instead of being completely
invisible.  I don't think anyone will notice or care.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>

5 files changed, 64 insertions, 152 deletions

diff --git a/fs/proc/Makefile b/fs/proc/Makefile
index 99349efbbc2b..981b05601931 100644
--- a/fs/proc/Makefile
+++ b/fs/proc/Makefile
@@ -21,6 +21,7 @@ proc-y	+= uptime.o
 proc-y  += version.o
 proc-y  += softirqs.o
 proc-y  += namespaces.o
+proc-y  += self.o
 proc-$(CONFIG_PROC_SYSCTL)      += proc_sysctl.o
 proc-$(CONFIG_NET)              += proc_net.o
 proc-$(CONFIG_PROC_KCORE)       += kcore.o
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 144a96732dd7..cbe454e94af8 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2237,146 +2237,6 @@ static const struct file_operations proc_coredump_filter_operations = {
 };
 #endif
 
-/*
- * /proc/self:
- */
-static int proc_self_readlink(struct dentry *dentry, char __user *buffer,
-                              int buflen)
-{
-        struct pid_namespace *ns = dentry->d_sb->s_fs_info;
-        pid_t tgid = task_tgid_nr_ns(current, ns);
-        char tmp[PROC_NUMBUF];
-        if (!tgid)
-                return -ENOENT;
-        sprintf(tmp, "%d", tgid);
-        return vfs_readlink(dentry,buffer,buflen,tmp);
-}
-
-static void *proc_self_follow_link(struct dentry *dentry, struct nameidata *nd)
-{
-        struct pid_namespace *ns = dentry->d_sb->s_fs_info;
-        pid_t tgid = task_tgid_nr_ns(current, ns);
-        char *name = ERR_PTR(-ENOENT);
-        if (tgid) {
-                /* 11 for max length of signed int in decimal + NULL term */
-                name = kmalloc(12, GFP_KERNEL);
-                if (!name)
-                        name = ERR_PTR(-ENOMEM);
-                else
-                        sprintf(name, "%d", tgid);
-        }
-        nd_set_link(nd, name);
-        return NULL;
-}
-
-static void proc_self_put_link(struct dentry *dentry, struct nameidata *nd,
-                                void *cookie)
-{
-        char *s = nd_get_link(nd);
-        if (!IS_ERR(s))
-                kfree(s);
-}
-
-static const struct inode_operations proc_self_inode_operations = {
-        .readlink       = proc_self_readlink,
-        .follow_link    = proc_self_follow_link,
-        .put_link       = proc_self_put_link,
-};
-
-/*
- * proc base
- *
- * These are the directory entries in the root directory of /proc
- * that properly belong to the /proc filesystem, as they describe
- * describe something that is process related.
- */
-static const struct pid_entry proc_base_stuff[] = {
-        NOD("self", S_IFLNK|S_IRWXUGO,
-                &proc_self_inode_operations, NULL, {}),
-};
-
-static struct dentry *proc_base_instantiate(struct inode *dir,
-        struct dentry *dentry, struct task_struct *task, const void *ptr)
-{
-        const struct pid_entry *p = ptr;
-        struct inode *inode;
-        struct proc_inode *ei;
-        struct dentry *error;
-
-        /* Allocate the inode */
-        error = ERR_PTR(-ENOMEM);
-        inode = new_inode(dir->i_sb);
-        if (!inode)
-                goto out;
-
-        /* Initialize the inode */
-        ei = PROC_I(inode);
-        inode->i_ino = get_next_ino();
-        inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
-
-        /*
-         * grab the reference to the task.
-         */
-        ei->pid = get_task_pid(task, PIDTYPE_PID);
-        if (!ei->pid)
-                goto out_iput;
-
-        inode->i_mode = p->mode;
-        if (S_ISDIR(inode->i_mode))
-                set_nlink(inode, 2);
-        if (S_ISLNK(inode->i_mode))
-                inode->i_size = 64;
-        if (p->iop)
-                inode->i_op = p->iop;
-        if (p->fop)
-                inode->i_fop = p->fop;
-        ei->op = p->op;
-        d_add(dentry, inode);
-        error = NULL;
-out:
-        return error;
-out_iput:
-        iput(inode);
-        goto out;
-}
-
-static struct dentry *proc_base_lookup(struct inode *dir, struct dentry *dentry)
-{
-        struct dentry *error;
-        struct task_struct *task = get_proc_task(dir);
-        const struct pid_entry *p, *last;
-
-        error = ERR_PTR(-ENOENT);
-
-        if (!task)
-                goto out_no_task;
-
-        /* Lookup the directory entry */
-        last = &proc_base_stuff[ARRAY_SIZE(proc_base_stuff) - 1];
-        for (p = proc_base_stuff; p <= last; p++) {
-                if (p->len != dentry->d_name.len)
-                        continue;
-                if (!memcmp(dentry->d_name.name, p->name, p->len))
-                        break;
-        }
-        if (p > last)
-                goto out;
-
-        error = proc_base_instantiate(dir, dentry, task, p);
-
-out:
-        put_task_struct(task);
-out_no_task:
-        return error;
-}
-
-static int proc_base_fill_cache(struct file *filp, void *dirent,
-        filldir_t filldir, struct task_struct *task, const struct pid_entry *p)
-{
-        return proc_fill_cache(filp, dirent, filldir, p->name, p->len,
-                                proc_base_instantiate, task, p);
-}
-
 #ifdef CONFIG_TASK_IO_ACCOUNTING
 static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 {
@@ -2767,15 +2627,11 @@ out:
 
 struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags)
 {
-        struct dentry *result;
+        struct dentry *result = NULL;
         struct task_struct *task;
         unsigned tgid;
         struct pid_namespace *ns;
 
-        result = proc_base_lookup(dir, dentry);
-        if (!IS_ERR(result) || PTR_ERR(result) != -ENOENT)
-                goto out;
-
         tgid = name_to_int(dentry);
         if (tgid == ~0U)
                 goto out;
@@ -2838,7 +2694,7 @@ retry:
         return iter;
 }
 
-#define TGID_OFFSET (FIRST_PROCESS_ENTRY + ARRAY_SIZE(proc_base_stuff))
+#define TGID_OFFSET (FIRST_PROCESS_ENTRY)
 
 static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
         struct tgid_iter iter)
@@ -2872,12 +2728,6 @@ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
         if (!reaper)
                 goto out_no_task;
 
-        for (; nr < ARRAY_SIZE(proc_base_stuff); filp->f_pos++, nr++) {
-                const struct pid_entry *p = &proc_base_stuff[nr];
-                if (proc_base_fill_cache(filp, dirent, filldir, reaper, p) < 0)
-                        goto out;
-        }
-
         ns = filp->f_dentry->d_sb->s_fs_info;
         iter.task = NULL;
         iter.tgid = filp->f_pos - TGID_OFFSET;
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 43973b084abf..252544c05207 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -15,6 +15,7 @@ struct  ctl_table_header;
 struct  mempolicy;
 
 extern struct proc_dir_entry proc_root;
+extern void proc_self_init(void);
 #ifdef CONFIG_PROC_SYSCTL
 extern int proc_sys_init(void);
 extern void sysctl_head_put(struct ctl_table_header *head);
diff --git a/fs/proc/root.c b/fs/proc/root.c
index 9889a92d2e01..5da984959edc 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -169,6 +169,7 @@ void __init proc_root_init(void)
                 return;
         }
 
+        proc_self_init();
         proc_symlink("mounts", NULL, "self/mounts");
 
         proc_net_init();
diff --git a/fs/proc/self.c b/fs/proc/self.c
new file mode 100644
index 000000000000..aa5cc3bff140
--- /dev/null
+++ b/fs/proc/self.c
@@ -0,0 +1,59 @@
+#include <linux/proc_fs.h>
+#include <linux/sched.h>
+#include <linux/namei.h>
+
+/*
+ * /proc/self:
+ */
+static int proc_self_readlink(struct dentry *dentry, char __user *buffer,
+                              int buflen)
+{
+        struct pid_namespace *ns = dentry->d_sb->s_fs_info;
+        pid_t tgid = task_tgid_nr_ns(current, ns);
+        char tmp[PROC_NUMBUF];
+        if (!tgid)
+                return -ENOENT;
+        sprintf(tmp, "%d", tgid);
+        return vfs_readlink(dentry,buffer,buflen,tmp);
+}
+
+static void *proc_self_follow_link(struct dentry *dentry, struct nameidata *nd)
+{
+        struct pid_namespace *ns = dentry->d_sb->s_fs_info;
+        pid_t tgid = task_tgid_nr_ns(current, ns);
+        char *name = ERR_PTR(-ENOENT);
+        if (tgid) {
+                /* 11 for max length of signed int in decimal + NULL term */
+                name = kmalloc(12, GFP_KERNEL);
+                if (!name)
+                        name = ERR_PTR(-ENOMEM);
+                else
+                        sprintf(name, "%d", tgid);
+        }
+        nd_set_link(nd, name);
+        return NULL;
+}
+
+static void proc_self_put_link(struct dentry *dentry, struct nameidata *nd,
+                                void *cookie)
+{
+        char *s = nd_get_link(nd);
+        if (!IS_ERR(s))
+                kfree(s);
+}
+
+static const struct inode_operations proc_self_inode_operations = {
+        .readlink       = proc_self_readlink,
+        .follow_link    = proc_self_follow_link,
+        .put_link       = proc_self_put_link,
+};
+
+void __init proc_self_init(void)
+{
+        struct proc_dir_entry *proc_self_symlink;
+        mode_t mode;
+
+        mode = S_IFLNK | S_IRWXUGO;
+        proc_self_symlink = proc_create("self", mode, NULL, NULL );
+        proc_self_symlink->proc_iops = &proc_self_inode_operations;
+}