pidfs: remove config option

Enable pidfs unconditionally. There's no real reason not do to it.
For 32bit systems we add a simple inode allocation mechanism that still
guarantees that userspace can compare processes by inode number which
they already do as I found out in [1]. If they also need the uniqueness
property that we get by default on 64bit systems they should simply
parse the contents of /proc/<pid>/fd/<nr>. On 64bit we don't have to
deal with any of this and things are nice and simple.

Link: systemd/systemd#31713 [1]
Signed-off-by: Christian Brauner <brauner@kernel.org>

fs/Kconfig

@@ -174,13 +174,6 @@ source "fs/proc/Kconfig"
 source "fs/kernfs/Kconfig"
 source "fs/sysfs/Kconfig"
 
-config FS_PID
-	bool "Pseudo filesystem for process file descriptors"
-	depends on 64BIT
-	default y
-	help
-	  Pidfs implements advanced features for process file descriptors.
-
 config TMPFS
 	bool "Tmpfs virtual memory file system support (former shm fs)"
 	depends on SHMEM

fs/pidfs.c

@@ -16,17 +16,6 @@
 
 #include "internal.h"
 
-static int pidfd_release(struct inode *inode, struct file *file)
-{
-#ifndef CONFIG_FS_PID
-	struct pid *pid = file->private_data;
-
-	file->private_data = NULL;
-	put_pid(pid);
-#endif
-	return 0;
-}
-
 #ifdef CONFIG_PROC_FS
 /**
  * pidfd_show_fdinfo - print information about a pidfd
@@ -120,7 +109,6 @@ static __poll_t pidfd_poll(struct file *file, struct poll_table_struct *pts)
 }
 
 static const struct file_operations pidfs_file_operations = {
-	.release	= pidfd_release,
 	.poll		= pidfd_poll,
 #ifdef CONFIG_PROC_FS
 	.show_fdinfo	= pidfd_show_fdinfo,
@@ -131,16 +119,47 @@ struct pid *pidfd_pid(const struct file *file)
 {
 	if (file->f_op != &pidfs_file_operations)
 		return ERR_PTR(-EBADF);
-#ifdef CONFIG_FS_PID
 	return file_inode(file)->i_private;
-#else
-	return file->private_data;
-#endif
 }
 
-#ifdef CONFIG_FS_PID
 static struct vfsmount *pidfs_mnt __ro_after_init;
 
+#if BITS_PER_LONG == 32
+/*
+ * Provide a fallback mechanism for 32-bit systems so processes remain
+ * reliably comparable by inode number even on those systems.
+ */
+static DEFINE_IDA(pidfd_inum_ida);
+
+/*
+ * Inode numbering for pidfs start at RESERVED_PIDS + 1. This avoids
+ * collisions with the root inode which is 1 for pseudo filesystems.
+ */
+static int pidfs_inum(struct pid *pid, unsigned long *ino)
+{
+	int ret;
+
+	ret = ida_alloc_range(&pidfd_inum_ida, 301, UINT_MAX, GFP_ATOMIC);
+	if (ret < 0)
+		return -ENOSPC;
+
+	*ino = ret;
+	return 0;
+}
+
+static void pidfs_free_inum(unsigned long ino)
+{
+	ida_free(&pidfd_inum_ida, ino);
+}
+#else
+static inline int pidfs_inum(struct pid *pid, unsigned long *ino)
+{
+	*ino = pid->ino;
+	return 0;
+}
+#define pidfs_free_inum(ino) ((void)(ino))
+#endif
+
 /*
  * The vfs falls back to simple_setattr() if i_op->setattr() isn't
  * implemented. Let's reject it completely until we have a clean
@@ -173,6 +192,7 @@ static void pidfs_evict_inode(struct inode *inode)
 
 	clear_inode(inode);
 	put_pid(pid);
+	pidfs_free_inum(inode->i_ino);
 }
 
 static const struct super_operations pidfs_sops = {
@@ -183,8 +203,10 @@ static const struct super_operations pidfs_sops = {
 
 static char *pidfs_dname(struct dentry *dentry, char *buffer, int buflen)
 {
-	return dynamic_dname(buffer, buflen, "pidfd:[%lu]",
-			     d_inode(dentry)->i_ino);
+	struct inode *inode = d_inode(dentry);
+	struct pid *pid = inode->i_private;
+
+	return dynamic_dname(buffer, buflen, "pidfd:[%llu]", pid->ino);
 }
 
 static const struct dentry_operations pidfs_dentry_operations = {
@@ -239,13 +261,13 @@ struct file *pidfs_alloc_file(struct pid *pid, unsigned int flags)
 	struct file *pidfd_file;
 	struct path path;
 	int ret;
+	unsigned long ino;
 
-	/*
-	* Inode numbering for pidfs start at RESERVED_PIDS + 1.
-	* This avoids collisions with the root inode which is 1
-	* for pseudo filesystems.
-	 */
-	ret = path_from_stashed(&pid->stashed, pid->ino, pidfs_mnt,
+	ret = pidfs_inum(pid, &ino);
+	if (ret < 0)
+		return ERR_PTR(ret);
+
+	ret = path_from_stashed(&pid->stashed, ino, pidfs_mnt,
 				get_pid(pid), &path);
 	if (ret < 0)
 		return ERR_PTR(ret);
@@ -261,30 +283,3 @@ void __init pidfs_init(void)
 	if (IS_ERR(pidfs_mnt))
 		panic("Failed to mount pidfs pseudo filesystem");
 }
-
-bool is_pidfs_sb(const struct super_block *sb)
-{
-	return sb == pidfs_mnt->mnt_sb;
-}
-
-#else /* !CONFIG_FS_PID */
-
-struct file *pidfs_alloc_file(struct pid *pid, unsigned int flags)
-{
-	struct file *pidfd_file;
-
-	pidfd_file = anon_inode_getfile("[pidfd]", &pidfs_file_operations, pid,
-					flags | O_RDWR);
-	if (IS_ERR(pidfd_file))
-		return pidfd_file;
-
-	get_pid(pid);
-	return pidfd_file;
-}
-
-void __init pidfs_init(void) { }
-bool is_pidfs_sb(const struct super_block *sb)
-{
-	return false;
-}
-#endif

include/linux/pid.h

@@ -55,10 +55,8 @@ struct pid
 	refcount_t count;
 	unsigned int level;
 	spinlock_t lock;
-#ifdef CONFIG_FS_PID
 	struct dentry *stashed;
-	unsigned long ino;
-#endif
+	u64 ino;
 	/* lists of tasks that use this pid */
 	struct hlist_head tasks[PIDTYPE_MAX];
 	struct hlist_head inodes;

include/linux/pidfs.h

@@ -4,6 +4,5 @@
 
 struct file *pidfs_alloc_file(struct pid *pid, unsigned int flags);
 void __init pidfs_init(void);
-bool is_pidfs_sb(const struct super_block *sb);
 
 #endif /* _LINUX_PID_FS_H */

kernel/pid.c

@@ -66,13 +66,11 @@ int pid_max = PID_MAX_DEFAULT;
 
 int pid_max_min = RESERVED_PIDS + 1;
 int pid_max_max = PID_MAX_LIMIT;
-#ifdef CONFIG_FS_PID
 /*
  * Pseudo filesystems start inode numbering after one. We use Reserved
  * PIDs as a natural offset.
  */
 static u64 pidfs_ino = RESERVED_PIDS;
-#endif
 
 /*
  * PID-map pages start out as NULL, they get allocated upon
@@ -280,10 +278,8 @@ struct pid *alloc_pid(struct pid_namespace *ns, pid_t *set_tid,
 	spin_lock_irq(&pidmap_lock);
 	if (!(ns->pid_allocated & PIDNS_ADDING))
 		goto out_unlock;
-#ifdef CONFIG_FS_PID
 	pid->stashed = NULL;
 	pid->ino = ++pidfs_ino;
-#endif
 	for ( ; upid >= pid->numbers; --upid) {
 		/* Make the PID visible to find_pid_ns. */
 		idr_replace(&upid->ns->idr, pid, upid->nr);