1 files changed, 215 insertions, 44 deletions

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index d72fc40241d9..52f0b75cbce2 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  *  linux/fs/proc/net.c
  *
@@ -7,9 +8,6 @@
  *
  *  proc net directory handling functions
  */
-
-#include <linux/uaccess.h>
-
 #include <linux/errno.h>
 #include <linux/time.h>
 #include <linux/proc_fs.h>
@@ -38,73 +36,234 @@ static struct net *get_proc_net(const struct inode *inode)
 	return maybe_get_net(PDE_NET(PDE(inode)));
 }
 
-int seq_open_net(struct inode *ino, struct file *f,
-		 const struct seq_operations *ops, int size)
+static int seq_open_net(struct inode *inode, struct file *file)
 {
-	struct net *net;
+	unsigned int state_size = PDE(inode)->state_size;
 	struct seq_net_private *p;
+	struct net *net;
 
-	BUG_ON(size < sizeof(*p));
+	WARN_ON_ONCE(state_size < sizeof(*p));
 
-	net = get_proc_net(ino);
-	if (net == NULL)
+	if (file->f_mode & FMODE_WRITE && !PDE(inode)->write)
+		return -EACCES;
+
+	net = get_proc_net(inode);
+	if (!net)
 		return -ENXIO;
 
-	p = __seq_open_private(f, ops, size);
-	if (p == NULL) {
+	p = __seq_open_private(file, PDE(inode)->seq_ops, state_size);
+	if (!p) {
 		put_net(net);
 		return -ENOMEM;
 	}
 #ifdef CONFIG_NET_NS
 	p->net = net;
+	netns_tracker_alloc(net, &p->ns_tracker, GFP_KERNEL);
 #endif
 	return 0;
 }
-EXPORT_SYMBOL_GPL(seq_open_net);
 
-int single_open_net(struct inode *inode, struct file *file,
-		int (*show)(struct seq_file *, void *))
+static void seq_file_net_put_net(struct seq_file *seq)
 {
-	int err;
-	struct net *net;
+#ifdef CONFIG_NET_NS
+	struct seq_net_private *priv = seq->private;
 
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
+	put_net_track(priv->net, &priv->ns_tracker);
+#else
+	put_net(&init_net);
+#endif
+}
 
-	err = single_open(file, show, net);
-	if (err < 0)
-		goto err_open;
+static int seq_release_net(struct inode *ino, struct file *f)
+{
+	struct seq_file *seq = f->private_data;
 
+	seq_file_net_put_net(seq);
+	seq_release_private(ino, f);
 	return 0;
+}
 
-err_open:
-	put_net(net);
-err_net:
-	return err;
+static const struct proc_ops proc_net_seq_ops = {
+	.proc_open	= seq_open_net,
+	.proc_read	= seq_read,
+	.proc_write	= proc_simple_write,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= seq_release_net,
+};
+
+int bpf_iter_init_seq_net(void *priv_data, struct bpf_iter_aux_info *aux)
+{
+#ifdef CONFIG_NET_NS
+	struct seq_net_private *p = priv_data;
+
+	p->net = get_net_track(current->nsproxy->net_ns, &p->ns_tracker,
+			       GFP_KERNEL);
+#endif
+	return 0;
 }
-EXPORT_SYMBOL_GPL(single_open_net);
 
-int seq_release_net(struct inode *ino, struct file *f)
+void bpf_iter_fini_seq_net(void *priv_data)
 {
-	struct seq_file *seq;
+#ifdef CONFIG_NET_NS
+	struct seq_net_private *p = priv_data;
 
-	seq = f->private_data;
+	put_net_track(p->net, &p->ns_tracker);
+#endif
+}
 
-	put_net(seq_file_net(seq));
-	seq_release_private(ino, f);
-	return 0;
+struct proc_dir_entry *proc_create_net_data(const char *name, umode_t mode,
+		struct proc_dir_entry *parent, const struct seq_operations *ops,
+		unsigned int state_size, void *data)
+{
+	struct proc_dir_entry *p;
+
+	p = proc_create_reg(name, mode, &parent, data);
+	if (!p)
+		return NULL;
+	pde_force_lookup(p);
+	p->proc_ops = &proc_net_seq_ops;
+	p->seq_ops = ops;
+	p->state_size = state_size;
+	return proc_register(parent, p);
+}
+EXPORT_SYMBOL_GPL(proc_create_net_data);
+
+/**
+ * proc_create_net_data_write - Create a writable net_ns-specific proc file
+ * @name: The name of the file.
+ * @mode: The file's access mode.
+ * @parent: The parent directory in which to create.
+ * @ops: The seq_file ops with which to read the file.
+ * @write: The write method with which to 'modify' the file.
+ * @state_size: The size of the per-file private state to allocate.
+ * @data: Data for retrieval by pde_data().
+ *
+ * Create a network namespaced proc file in the @parent directory with the
+ * specified @name and @mode that allows reading of a file that displays a
+ * series of elements and also provides for the file accepting writes that have
+ * some arbitrary effect.
+ *
+ * The functions in the @ops table are used to iterate over items to be
+ * presented and extract the readable content using the seq_file interface.
+ *
+ * The @write function is called with the data copied into a kernel space
+ * scratch buffer and has a NUL appended for convenience.  The buffer may be
+ * modified by the @write function.  @write should return 0 on success.
+ *
+ * The @data value is accessible from the @show and @write functions by calling
+ * pde_data() on the file inode.  The network namespace must be accessed by
+ * calling seq_file_net() on the seq_file struct.
+ */
+struct proc_dir_entry *proc_create_net_data_write(const char *name, umode_t mode,
+						  struct proc_dir_entry *parent,
+						  const struct seq_operations *ops,
+						  proc_write_t write,
+						  unsigned int state_size, void *data)
+{
+	struct proc_dir_entry *p;
+
+	p = proc_create_reg(name, mode, &parent, data);
+	if (!p)
+		return NULL;
+	pde_force_lookup(p);
+	p->proc_ops = &proc_net_seq_ops;
+	p->seq_ops = ops;
+	p->state_size = state_size;
+	p->write = write;
+	return proc_register(parent, p);
+}
+EXPORT_SYMBOL_GPL(proc_create_net_data_write);
+
+static int single_open_net(struct inode *inode, struct file *file)
+{
+	struct proc_dir_entry *de = PDE(inode);
+	struct net *net;
+	int err;
+
+	net = get_proc_net(inode);
+	if (!net)
+		return -ENXIO;
+
+	err = single_open(file, de->single_show, net);
+	if (err)
+		put_net(net);
+	return err;
 }
-EXPORT_SYMBOL_GPL(seq_release_net);
 
-int single_release_net(struct inode *ino, struct file *f)
+static int single_release_net(struct inode *ino, struct file *f)
 {
 	struct seq_file *seq = f->private_data;
 	put_net(seq->private);
 	return single_release(ino, f);
 }
-EXPORT_SYMBOL_GPL(single_release_net);
+
+static const struct proc_ops proc_net_single_ops = {
+	.proc_open	= single_open_net,
+	.proc_read	= seq_read,
+	.proc_write	= proc_simple_write,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release_net,
+};
+
+struct proc_dir_entry *proc_create_net_single(const char *name, umode_t mode,
+		struct proc_dir_entry *parent,
+		int (*show)(struct seq_file *, void *), void *data)
+{
+	struct proc_dir_entry *p;
+
+	p = proc_create_reg(name, mode, &parent, data);
+	if (!p)
+		return NULL;
+	pde_force_lookup(p);
+	p->proc_ops = &proc_net_single_ops;
+	p->single_show = show;
+	return proc_register(parent, p);
+}
+EXPORT_SYMBOL_GPL(proc_create_net_single);
+
+/**
+ * proc_create_net_single_write - Create a writable net_ns-specific proc file
+ * @name: The name of the file.
+ * @mode: The file's access mode.
+ * @parent: The parent directory in which to create.
+ * @show: The seqfile show method with which to read the file.
+ * @write: The write method with which to 'modify' the file.
+ * @data: Data for retrieval by pde_data().
+ *
+ * Create a network-namespaced proc file in the @parent directory with the
+ * specified @name and @mode that allows reading of a file that displays a
+ * single element rather than a series and also provides for the file accepting
+ * writes that have some arbitrary effect.
+ *
+ * The @show function is called to extract the readable content via the
+ * seq_file interface.
+ *
+ * The @write function is called with the data copied into a kernel space
+ * scratch buffer and has a NUL appended for convenience.  The buffer may be
+ * modified by the @write function.  @write should return 0 on success.
+ *
+ * The @data value is accessible from the @show and @write functions by calling
+ * pde_data() on the file inode.  The network namespace must be accessed by
+ * calling seq_file_single_net() on the seq_file struct.
+ */
+struct proc_dir_entry *proc_create_net_single_write(const char *name, umode_t mode,
+						    struct proc_dir_entry *parent,
+						    int (*show)(struct seq_file *, void *),
+						    proc_write_t write,
+						    void *data)
+{
+	struct proc_dir_entry *p;
+
+	p = proc_create_reg(name, mode, &parent, data);
+	if (!p)
+		return NULL;
+	pde_force_lookup(p);
+	p->proc_ops = &proc_net_single_ops;
+	p->single_show = show;
+	p->write = write;
+	return proc_register(parent, p);
+}
+EXPORT_SYMBOL_GPL(proc_create_net_single_write);
 
 static struct net *get_proc_task_net(struct inode *dir)
 {
@@ -135,13 +294,14 @@ static struct dentry *proc_tgid_net_lookup(struct inode *dir,
 	de = ERR_PTR(-ENOENT);
 	net = get_proc_task_net(dir);
 	if (net != NULL) {
-		de = proc_lookup_de(net->proc_net, dir, dentry);
+		de = proc_lookup_de(dir, dentry, net->proc_net);
 		put_net(net);
 	}
 	return de;
 }
 
-static int proc_tgid_net_getattr(const struct path *path, struct kstat *stat,
+static int proc_tgid_net_getattr(struct mnt_idmap *idmap,
+				 const struct path *path, struct kstat *stat,
 				 u32 request_mask, unsigned int query_flags)
 {
 	struct inode *inode = d_inode(path->dentry);
@@ -149,7 +309,7 @@ static int proc_tgid_net_getattr(const struct path *path, struct kstat *stat,
 
 	net = get_proc_task_net(inode);
 
-	generic_fillattr(inode, stat);
+	generic_fillattr(&nop_mnt_idmap, request_mask, inode, stat);
 
 	if (net != NULL) {
 		stat->nlink = net->proc_net->nlink;
@@ -162,6 +322,7 @@ static int proc_tgid_net_getattr(const struct path *path, struct kstat *stat,
 const struct inode_operations proc_net_inode_operations = {
 	.lookup		= proc_tgid_net_lookup,
 	.getattr	= proc_tgid_net_getattr,
+	.setattr        = proc_setattr,
 };
 
 static int proc_tgid_net_readdir(struct file *file, struct dir_context *ctx)
@@ -172,7 +333,7 @@ static int proc_tgid_net_readdir(struct file *file, struct dir_context *ctx)
 	ret = -EINVAL;
 	net = get_proc_task_net(file_inode(file));
 	if (net != NULL) {
-		ret = proc_readdir_de(net->proc_net, file, ctx);
+		ret = proc_readdir_de(file, ctx, net->proc_net);
 		put_net(net);
 	}
 	return ret;
@@ -191,8 +352,14 @@ static __net_init int proc_net_ns_init(struct net *net)
 	kgid_t gid;
 	int err;
 
+	/*
+	 * This PDE acts only as an anchor for /proc/${pid}/net hierarchy.
+	 * Corresponding inode (PDE(inode) == net->proc_net) is never
+	 * instantiated therefore blanket zeroing is fine.
+	 * net->proc_net_stat inode is instantiated normally.
+	 */
 	err = -ENOMEM;
-	netd = kzalloc(sizeof(*netd) + 4, GFP_KERNEL);
+	netd = kmem_cache_zalloc(proc_dir_entry_cache, GFP_KERNEL);
 	if (!netd)
 		goto out;
 
@@ -201,6 +368,7 @@ static __net_init int proc_net_ns_init(struct net *net)
 	netd->nlink = 2;
 	netd->namelen = 3;
 	netd->parent = &proc_root;
+	netd->name = netd->inline_name;
 	memcpy(netd->name, "net", 4);
 
 	uid = make_kuid(net->user_ns, 0);
@@ -213,6 +381,9 @@ static __net_init int proc_net_ns_init(struct net *net)
 
 	proc_set_user(netd, uid, gid);
 
+	/* Seed dentry revalidation for /proc/${pid}/net */
+	pde_force_lookup(netd);
+
 	err = -EEXIST;
 	net_statd = proc_net_mkdir(net, "stat", netd);
 	if (!net_statd)
@@ -223,7 +394,7 @@ static __net_init int proc_net_ns_init(struct net *net)
 	return 0;
 
 free_net:
-	kfree(netd);
+	pde_free(netd);
 out:
 	return err;
 }
@@ -231,7 +402,7 @@ out:
 static __net_exit void proc_net_ns_exit(struct net *net)
 {
 	remove_proc_entry("stat", net->proc_net);
-	kfree(net->proc_net);
+	pde_free(net->proc_net);
 }
 
 static struct pernet_operations __net_initdata proc_net_ns_ops = {