Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Fix error output of boot instance creation failure

   Currently if a boot instance creation fails, instead of printing out
   the name of the instance that failed, it prints "(null)". That is
   because it prints "cur_str" that had already been processed by
   strsep(). Print the saved name instead.

   While at it, print the error code of the failure.

 - Fix use-after-free for same named historgrams

   Histograms can be named so that they can be used in multiple events.
   But if the named histogram has a variable attached, the second event
   that uses the named histogram which duplicates it and needs to free
   the original after duplication leaves the old variable in place and
   still visible. If another histogram uses than variable, it will use
   the stale one which will try to reference the freed duplicate
   histogram and crash the kernel.

   Free the duplicate variables along with the duplicated histogram
   data.

 - Check return value of kthread_run() in event self test

   The events self tests uses a kthread for testing but does not check
   if it succeeded in creating a kthread. If the kthread creation were
   to fail, the code will still try to call kthread_stop() on the error
   returned.

 - Fix race between reading trace_pipe and updating subbuffer size

   If a user is reading the trace_pipe file at the same time they update
   the ring buffer sub-buffer size, can cause the trace_pipe read to
   read stale data. Add trace_access_lock() around updating the ring
   buffer sub-buffer size.

 - Fix eventfs_inode on failure path in creation of the events directory

   In the creation of the "events" directory, if after allocating the
   eventfs_inode a failure is detected, it calls cleanup_ei() which
   calls free_ei(). The free_ei() will test if eventfs_inode being freed
   has no children. It is a bug if it does. But on the failure case of
   the creation of the "events" directory, the children lists have not
   yet been initialized and the free will trigger a warning because
   list_empty() on an uninitialized list returns false.

   Move the initialization into init_ei() where it makes more sense and
   makes sure that a created eventfs_inode has its lists initialized
   upon creation.

 - Check return value of kthread_run() in ftrace direct sample code

   The sample code that shows how to use the ftrace direct calls does
   not test the return of kthread_run() to see if it succeeds. Return a
   failure if the kthread_run() doesn't succeed.

 - Clear user events state on fork in case of alloc failure

   On fork, the child gets a pointer to the parent's user events state.
   It makes a copy of it then updates the child's pointer to it. But if
   the allocation fails, the duplication function leaves the child with
   a pointer to its parent's descriptor. When the child cleans up its
   data, it will free the parent's descriptor while the parent is still
   using it.

   In the duplication function, set the child's user_event_mm to NULL
   before testing if the allocation succeeded, and when it exits it will
   not free the parent's descriptor.

 - Fix retry exhaustion in simple ring buffer reader swap

   simple_ring_buffer_swap_reader_page() starts with retry set to 8 and
   post-decrements it only after a failed link replacement. On the final
   attempt, a successful replacement leaves retry at zero, while a
   failed replacement leaves it at -1.

   But the check for success expects the retry value to be non-zero and
   exits with an error on zero. This is the opposite result. Fix it.

 - Fail nicely when the remote swap_reader_page() returns an error

   Currently, if the swap_reader_page() of a remote buffer fails, it
   triggers a WARN_ON_ONCE() and continues normally. Instead, have it
   exit with an error and a pr_warn() print instead of a full WARNING.

* tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Stop remote reader update when page swap fails
  tracing: Fix retry exhaustion in simple ring buffer reader swap
  tracing/user_events: Clear copied tracing state before fork duplication
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify
  eventfs: Initialize ei->children and ei->list in init_ei()
  tracing: Fix use-after-free in trace_pipe read on sub-buffer order change
  tracing: Fix crash passing ERR_PTR to kthread_stop()
  tracing: Fix use-after-free with same-name named triggers
  tracing: Fix logged instance name on creation failure

1 files changed, 461 insertions, 0 deletions

diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c
new file mode 100644
index 000000000..73c7b8215
--- /dev/null
+++ b/fs/ecryptfs/messaging.c
@@ -0,0 +1,461 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * eCryptfs: Linux filesystem encryption layer
+ *
+ * Copyright (C) 2004-2008 International Business Machines Corp.
+ *   Author(s): Michael A. Halcrow <mhalcrow@us.ibm.com>
+ *		Tyler Hicks <code@tyhicks.com>
+ */
+#include <linux/overflow.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/user_namespace.h>
+#include <linux/nsproxy.h>
+#include "ecryptfs_kernel.h"
+
+static LIST_HEAD(ecryptfs_msg_ctx_free_list);
+static LIST_HEAD(ecryptfs_msg_ctx_alloc_list);
+static DEFINE_MUTEX(ecryptfs_msg_ctx_lists_mux);
+
+static struct hlist_head *ecryptfs_daemon_hash;
+DEFINE_MUTEX(ecryptfs_daemon_hash_mux);
+static int ecryptfs_hash_bits;
+#define ecryptfs_current_euid_hash(uid) \
+	hash_long((unsigned long)from_kuid(&init_user_ns, current_euid()), ecryptfs_hash_bits)
+
+static u32 ecryptfs_msg_counter;
+static struct ecryptfs_msg_ctx *ecryptfs_msg_ctx_arr;
+
+/**
+ * ecryptfs_acquire_free_msg_ctx
+ * @msg_ctx: The context that was acquired from the free list
+ *
+ * Acquires a context element from the free list and locks the mutex
+ * on the context.  Sets the msg_ctx task to current.  Returns zero on
+ * success; non-zero on error or upon failure to acquire a free
+ * context element.  Must be called with ecryptfs_msg_ctx_lists_mux
+ * held.
+ */
+static int ecryptfs_acquire_free_msg_ctx(struct ecryptfs_msg_ctx **msg_ctx)
+{
+	struct list_head *p;
+	int rc;
+
+	if (list_empty(&ecryptfs_msg_ctx_free_list)) {
+		printk(KERN_WARNING "%s: The eCryptfs free "
+		       "context list is empty.  It may be helpful to "
+		       "specify the ecryptfs_message_buf_len "
+		       "parameter to be greater than the current "
+		       "value of [%d]\n", __func__, ecryptfs_message_buf_len);
+		rc = -ENOMEM;
+		goto out;
+	}
+	list_for_each(p, &ecryptfs_msg_ctx_free_list) {
+		*msg_ctx = list_entry(p, struct ecryptfs_msg_ctx, node);
+		if (mutex_trylock(&(*msg_ctx)->mux)) {
+			(*msg_ctx)->task = current;
+			rc = 0;
+			goto out;
+		}
+	}
+	rc = -ENOMEM;
+out:
+	return rc;
+}
+
+/**
+ * ecryptfs_msg_ctx_free_to_alloc
+ * @msg_ctx: The context to move from the free list to the alloc list
+ *
+ * Must be called with ecryptfs_msg_ctx_lists_mux held.
+ */
+static void ecryptfs_msg_ctx_free_to_alloc(struct ecryptfs_msg_ctx *msg_ctx)
+{
+	list_move(&msg_ctx->node, &ecryptfs_msg_ctx_alloc_list);
+	msg_ctx->state = ECRYPTFS_MSG_CTX_STATE_PENDING;
+	msg_ctx->counter = ++ecryptfs_msg_counter;
+}
+
+/**
+ * ecryptfs_msg_ctx_alloc_to_free
+ * @msg_ctx: The context to move from the alloc list to the free list
+ *
+ * Must be called with ecryptfs_msg_ctx_lists_mux held.
+ */
+void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx)
+{
+	list_move(&(msg_ctx->node), &ecryptfs_msg_ctx_free_list);
+	kfree(msg_ctx->msg);
+	msg_ctx->msg = NULL;
+	msg_ctx->state = ECRYPTFS_MSG_CTX_STATE_FREE;
+}
+
+/**
+ * ecryptfs_find_daemon_by_euid
+ * @daemon: If return value is zero, points to the desired daemon pointer
+ *
+ * Must be called with ecryptfs_daemon_hash_mux held.
+ *
+ * Search the hash list for the current effective user id.
+ *
+ * Returns zero if the user id exists in the list; non-zero otherwise.
+ */
+int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon)
+{
+	int rc;
+
+	hlist_for_each_entry(*daemon,
+			    &ecryptfs_daemon_hash[ecryptfs_current_euid_hash()],
+			    euid_chain) {
+		if (uid_eq((*daemon)->file->f_cred->euid, current_euid())) {
+			rc = 0;
+			goto out;
+		}
+	}
+	rc = -EINVAL;
+out:
+	return rc;
+}
+
+/**
+ * ecryptfs_spawn_daemon - Create and initialize a new daemon struct
+ * @daemon: Pointer to set to newly allocated daemon struct
+ * @file: File used when opening /dev/ecryptfs
+ *
+ * Must be called ceremoniously while in possession of
+ * ecryptfs_sacred_daemon_hash_mux
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+int
+ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, struct file *file)
+{
+	int rc = 0;
+
+	(*daemon) = kzalloc_obj(**daemon);
+	if (!(*daemon)) {
+		rc = -ENOMEM;
+		goto out;
+	}
+	(*daemon)->file = file;
+	mutex_init(&(*daemon)->mux);
+	INIT_LIST_HEAD(&(*daemon)->msg_ctx_out_queue);
+	init_waitqueue_head(&(*daemon)->wait);
+	(*daemon)->num_queued_msg_ctx = 0;
+	hlist_add_head(&(*daemon)->euid_chain,
+		       &ecryptfs_daemon_hash[ecryptfs_current_euid_hash()]);
+out:
+	return rc;
+}
+
+/*
+ * ecryptfs_exorcise_daemon - Destroy the daemon struct
+ *
+ * Must be called ceremoniously while in possession of
+ * ecryptfs_daemon_hash_mux and the daemon's own mux.
+ */
+int ecryptfs_exorcise_daemon(struct ecryptfs_daemon *daemon)
+{
+	struct ecryptfs_msg_ctx *msg_ctx, *msg_ctx_tmp;
+	int rc = 0;
+
+	mutex_lock(&daemon->mux);
+	if ((daemon->flags & ECRYPTFS_DAEMON_IN_READ)
+	    || (daemon->flags & ECRYPTFS_DAEMON_IN_POLL)) {
+		rc = -EBUSY;
+		mutex_unlock(&daemon->mux);
+		goto out;
+	}
+	mutex_lock(&ecryptfs_msg_ctx_lists_mux);
+	list_for_each_entry_safe(msg_ctx, msg_ctx_tmp,
+				 &daemon->msg_ctx_out_queue, daemon_out_list) {
+		list_del(&msg_ctx->daemon_out_list);
+		daemon->num_queued_msg_ctx--;
+		printk(KERN_WARNING "%s: Warning: dropping message that is in "
+		       "the out queue of a dying daemon\n", __func__);
+		ecryptfs_msg_ctx_alloc_to_free(msg_ctx);
+	}
+	mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+	hlist_del(&daemon->euid_chain);
+	mutex_unlock(&daemon->mux);
+	kfree_sensitive(daemon);
+out:
+	return rc;
+}
+
+/**
+ * ecryptfs_process_response
+ * @daemon: eCryptfs daemon object
+ * @msg: The ecryptfs message received; the caller should sanity check
+ *       msg->data_len and free the memory
+ * @seq: The sequence number of the message; must match the sequence
+ *       number for the existing message context waiting for this
+ *       response
+ *
+ * Processes a response message after sending an operation request to
+ * userspace. Some other process is awaiting this response. Before
+ * sending out its first communications, the other process allocated a
+ * msg_ctx from the ecryptfs_msg_ctx_arr at a particular index. The
+ * response message contains this index so that we can copy over the
+ * response message into the msg_ctx that the process holds a
+ * reference to. The other process is going to wake up, check to see
+ * that msg_ctx->state == ECRYPTFS_MSG_CTX_STATE_DONE, and then
+ * proceed to read off and process the response message. Returns zero
+ * upon delivery to desired context element; non-zero upon delivery
+ * failure or error.
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+int ecryptfs_process_response(struct ecryptfs_daemon *daemon,
+			      struct ecryptfs_message *msg, u32 seq)
+{
+	struct ecryptfs_msg_ctx *msg_ctx;
+	size_t msg_size;
+	int rc;
+
+	if (msg->index >= ecryptfs_message_buf_len) {
+		rc = -EINVAL;
+		printk(KERN_ERR "%s: Attempt to reference "
+		       "context buffer at index [%d]; maximum "
+		       "allowable is [%d]\n", __func__, msg->index,
+		       (ecryptfs_message_buf_len - 1));
+		goto out;
+	}
+	msg_ctx = &ecryptfs_msg_ctx_arr[msg->index];
+	mutex_lock(&msg_ctx->mux);
+	if (msg_ctx->state != ECRYPTFS_MSG_CTX_STATE_PENDING) {
+		rc = -EINVAL;
+		printk(KERN_WARNING "%s: Desired context element is not "
+		       "pending a response\n", __func__);
+		goto unlock;
+	} else if (msg_ctx->counter != seq) {
+		rc = -EINVAL;
+		printk(KERN_WARNING "%s: Invalid message sequence; "
+		       "expected [%d]; received [%d]\n", __func__,
+		       msg_ctx->counter, seq);
+		goto unlock;
+	}
+	msg_size = struct_size(msg, data, msg->data_len);
+	msg_ctx->msg = kmemdup(msg, msg_size, GFP_KERNEL);
+	if (!msg_ctx->msg) {
+		rc = -ENOMEM;
+		goto unlock;
+	}
+	msg_ctx->state = ECRYPTFS_MSG_CTX_STATE_DONE;
+	wake_up_process(msg_ctx->task);
+	rc = 0;
+unlock:
+	mutex_unlock(&msg_ctx->mux);
+out:
+	return rc;
+}
+
+/**
+ * ecryptfs_send_message_locked
+ * @data: The data to send
+ * @data_len: The length of data
+ * @msg_type: Type of message
+ * @msg_ctx: The message context allocated for the send
+ *
+ * Must be called with ecryptfs_daemon_hash_mux held.
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+static int
+ecryptfs_send_message_locked(char *data, int data_len, u8 msg_type,
+			     struct ecryptfs_msg_ctx **msg_ctx)
+{
+	struct ecryptfs_daemon *daemon;
+	int rc;
+
+	rc = ecryptfs_find_daemon_by_euid(&daemon);
+	if (rc) {
+		rc = -ENOTCONN;
+		goto out;
+	}
+	mutex_lock(&ecryptfs_msg_ctx_lists_mux);
+	rc = ecryptfs_acquire_free_msg_ctx(msg_ctx);
+	if (rc) {
+		mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+		printk(KERN_WARNING "%s: Could not claim a free "
+		       "context element\n", __func__);
+		goto out;
+	}
+	ecryptfs_msg_ctx_free_to_alloc(*msg_ctx);
+	mutex_unlock(&(*msg_ctx)->mux);
+	mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+	rc = ecryptfs_send_miscdev(data, data_len, *msg_ctx, msg_type, 0,
+				   daemon);
+	if (rc) {
+		printk(KERN_ERR "%s: Error attempting to send message to "
+		       "userspace daemon; rc = [%d]\n", __func__, rc);
+		mutex_lock(&ecryptfs_msg_ctx_lists_mux);
+		mutex_lock(&(*msg_ctx)->mux);
+		ecryptfs_msg_ctx_alloc_to_free(*msg_ctx);
+		mutex_unlock(&(*msg_ctx)->mux);
+		mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+		*msg_ctx = NULL;
+	}
+out:
+	return rc;
+}
+
+/**
+ * ecryptfs_send_message
+ * @data: The data to send
+ * @data_len: The length of data
+ * @msg_ctx: The message context allocated for the send
+ *
+ * Grabs ecryptfs_daemon_hash_mux.
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+int ecryptfs_send_message(char *data, int data_len,
+			  struct ecryptfs_msg_ctx **msg_ctx)
+{
+	int rc;
+
+	mutex_lock(&ecryptfs_daemon_hash_mux);
+	rc = ecryptfs_send_message_locked(data, data_len, ECRYPTFS_MSG_REQUEST,
+					  msg_ctx);
+	mutex_unlock(&ecryptfs_daemon_hash_mux);
+	return rc;
+}
+
+/**
+ * ecryptfs_wait_for_response
+ * @msg_ctx: The context that was assigned when sending a message
+ * @msg: The incoming message from userspace; not set if rc != 0
+ *
+ * Sleeps until awaken by ecryptfs_receive_message or until the amount
+ * of time exceeds ecryptfs_message_wait_timeout.  If zero is
+ * returned, msg will point to a valid message from userspace; a
+ * non-zero value is returned upon failure to receive a message or an
+ * error occurs. Callee must free @msg on success.
+ */
+int ecryptfs_wait_for_response(struct ecryptfs_msg_ctx *msg_ctx,
+			       struct ecryptfs_message **msg)
+{
+	signed long timeout = ecryptfs_message_wait_timeout * HZ;
+	int rc = 0;
+
+sleep:
+	timeout = schedule_timeout_interruptible(timeout);
+	mutex_lock(&ecryptfs_msg_ctx_lists_mux);
+	mutex_lock(&msg_ctx->mux);
+	if (msg_ctx->state != ECRYPTFS_MSG_CTX_STATE_DONE) {
+		if (timeout) {
+			mutex_unlock(&msg_ctx->mux);
+			mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+			goto sleep;
+		}
+		rc = -ENOMSG;
+	} else {
+		*msg = msg_ctx->msg;
+		msg_ctx->msg = NULL;
+	}
+	ecryptfs_msg_ctx_alloc_to_free(msg_ctx);
+	mutex_unlock(&msg_ctx->mux);
+	mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+	return rc;
+}
+
+int __init ecryptfs_init_messaging(void)
+{
+	int i;
+	int rc = 0;
+
+	if (ecryptfs_number_of_users > ECRYPTFS_MAX_NUM_USERS) {
+		ecryptfs_number_of_users = ECRYPTFS_MAX_NUM_USERS;
+		printk(KERN_WARNING "%s: Specified number of users is "
+		       "too large, defaulting to [%d] users\n", __func__,
+		       ecryptfs_number_of_users);
+	}
+	mutex_lock(&ecryptfs_daemon_hash_mux);
+	ecryptfs_hash_bits = 1;
+	while (ecryptfs_number_of_users >> ecryptfs_hash_bits)
+		ecryptfs_hash_bits++;
+	ecryptfs_daemon_hash = kmalloc((sizeof(struct hlist_head)
+					* (1 << ecryptfs_hash_bits)),
+				       GFP_KERNEL);
+	if (!ecryptfs_daemon_hash) {
+		rc = -ENOMEM;
+		mutex_unlock(&ecryptfs_daemon_hash_mux);
+		goto out;
+	}
+	for (i = 0; i < (1 << ecryptfs_hash_bits); i++)
+		INIT_HLIST_HEAD(&ecryptfs_daemon_hash[i]);
+	mutex_unlock(&ecryptfs_daemon_hash_mux);
+	ecryptfs_msg_ctx_arr = kmalloc((sizeof(struct ecryptfs_msg_ctx)
+					* ecryptfs_message_buf_len),
+				       GFP_KERNEL);
+	if (!ecryptfs_msg_ctx_arr) {
+		kfree(ecryptfs_daemon_hash);
+		rc = -ENOMEM;
+		goto out;
+	}
+	mutex_lock(&ecryptfs_msg_ctx_lists_mux);
+	ecryptfs_msg_counter = 0;
+	for (i = 0; i < ecryptfs_message_buf_len; i++) {
+		INIT_LIST_HEAD(&ecryptfs_msg_ctx_arr[i].node);
+		INIT_LIST_HEAD(&ecryptfs_msg_ctx_arr[i].daemon_out_list);
+		mutex_init(&ecryptfs_msg_ctx_arr[i].mux);
+		mutex_lock(&ecryptfs_msg_ctx_arr[i].mux);
+		ecryptfs_msg_ctx_arr[i].index = i;
+		ecryptfs_msg_ctx_arr[i].state = ECRYPTFS_MSG_CTX_STATE_FREE;
+		ecryptfs_msg_ctx_arr[i].counter = 0;
+		ecryptfs_msg_ctx_arr[i].task = NULL;
+		ecryptfs_msg_ctx_arr[i].msg = NULL;
+		list_add_tail(&ecryptfs_msg_ctx_arr[i].node,
+			      &ecryptfs_msg_ctx_free_list);
+		mutex_unlock(&ecryptfs_msg_ctx_arr[i].mux);
+	}
+	mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+	rc = ecryptfs_init_ecryptfs_miscdev();
+	if (rc)
+		ecryptfs_release_messaging();
+out:
+	return rc;
+}
+
+void ecryptfs_release_messaging(void)
+{
+	if (ecryptfs_msg_ctx_arr) {
+		int i;
+
+		mutex_lock(&ecryptfs_msg_ctx_lists_mux);
+		for (i = 0; i < ecryptfs_message_buf_len; i++) {
+			mutex_lock(&ecryptfs_msg_ctx_arr[i].mux);
+			kfree(ecryptfs_msg_ctx_arr[i].msg);
+			mutex_unlock(&ecryptfs_msg_ctx_arr[i].mux);
+		}
+		kfree(ecryptfs_msg_ctx_arr);
+		mutex_unlock(&ecryptfs_msg_ctx_lists_mux);
+	}
+	if (ecryptfs_daemon_hash) {
+		struct ecryptfs_daemon *daemon;
+		struct hlist_node *n;
+		int i;
+
+		mutex_lock(&ecryptfs_daemon_hash_mux);
+		for (i = 0; i < (1 << ecryptfs_hash_bits); i++) {
+			int rc;
+
+			hlist_for_each_entry_safe(daemon, n,
+						  &ecryptfs_daemon_hash[i],
+						  euid_chain) {
+				rc = ecryptfs_exorcise_daemon(daemon);
+				if (rc)
+					printk(KERN_ERR "%s: Error whilst "
+					       "attempting to destroy daemon; "
+					       "rc = [%d]. Dazed and confused, "
+					       "but trying to continue.\n",
+					       __func__, rc);
+			}
+		}
+		kfree(ecryptfs_daemon_hash);
+		mutex_unlock(&ecryptfs_daemon_hash_mux);
+	}
+	ecryptfs_destroy_ecryptfs_miscdev();
+	return;
+}