ARM: 9485/1: mm: acquire mmap write lock around show_pte() for user faultsgrafted

When CONFIG_DEBUG_USER=y, and cmdline "user_debug=31" is set,
a user fault may trigger show_pte() without any lock.
If another thread in the same process concurrently calls munmap(),
the page table pages may be freed while show_pte() is still traversing
them, causing a use-after-free in show_pte().

If CONFIG_ARM_LPAE=y, this may cause a kernel panic if the pages table
of PMD are freed when show_pte() is running.

Acquire mmap_write_lock() around show_pte() for user faults to fix the
contention.

For user faults, additionally restrict that show_pte() is called only
when the addr is a user-space address (addr < TASK_SIZE). This is because
the lock of tsk->mm only protects the virtual memory of user address space,
furthermore, dumping the page tables of a kernel-space address for user
faults is unnecessary and may have security implications.

Keep everything unchanged for kernel faults, because the kernel is
already in the "oops" state, acquiring a lock may risk a deadlock.

Co-developed-by: Qi Xi <xiqi2@huawei.com>

Fixes: 6d021b724481 ("ARM: dump pgd, pmd and pte states on unhandled data abort faults")
Link: https://lore.kernel.org/20260716014022.2823-1-xieyuanbin1@huawei.com
Acked-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
Reviewed-by: Linus Walleij <linusw@kernel.org>
Signed-off-by: Qi Xi <xiqi2@huawei.com>
Signed-off-by: Xie Yuanbin <xieyuanbin1@huawei.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>

1 files changed, 174 insertions, 0 deletions

diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c
new file mode 100644
index 000000000..3bc21d677
--- /dev/null
+++ b/fs/ecryptfs/super.c
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * eCryptfs: Linux filesystem encryption layer
+ *
+ * Copyright (C) 1997-2003 Erez Zadok
+ * Copyright (C) 2001-2003 Stony Brook University
+ * Copyright (C) 2004-2006 International Business Machines Corp.
+ *   Author(s): Michael A. Halcrow <mahalcro@us.ibm.com>
+ *              Michael C. Thompson <mcthomps@us.ibm.com>
+ */
+
+#include <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/key.h>
+#include <linux/slab.h>
+#include <linux/seq_file.h>
+#include <linux/file.h>
+#include <linux/statfs.h>
+#include <linux/magic.h>
+#include "ecryptfs_kernel.h"
+
+struct kmem_cache *ecryptfs_inode_info_cache;
+
+/**
+ * ecryptfs_alloc_inode - allocate an ecryptfs inode
+ * @sb: Pointer to the ecryptfs super block
+ *
+ * Called to bring an inode into existence.
+ *
+ * Only handle allocation, setting up structures should be done in
+ * ecryptfs_read_inode. This is because the kernel, between now and
+ * then, will 0 out the private data pointer.
+ *
+ * Returns a pointer to a newly allocated inode, NULL otherwise
+ */
+static struct inode *ecryptfs_alloc_inode(struct super_block *sb)
+{
+	struct ecryptfs_inode_info *inode_info;
+	struct inode *inode = NULL;
+
+	inode_info = alloc_inode_sb(sb, ecryptfs_inode_info_cache, GFP_KERNEL);
+	if (unlikely(!inode_info))
+		goto out;
+	ecryptfs_init_crypt_stat(&inode_info->crypt_stat);
+	mutex_init(&inode_info->lower_file_mutex);
+	atomic_set(&inode_info->lower_file_count, 0);
+	inode_info->lower_file = NULL;
+	inode = &inode_info->vfs_inode;
+out:
+	return inode;
+}
+
+static void ecryptfs_free_inode(struct inode *inode)
+{
+	struct ecryptfs_inode_info *inode_info;
+	inode_info = ecryptfs_inode_to_private(inode);
+
+	kmem_cache_free(ecryptfs_inode_info_cache, inode_info);
+}
+
+/**
+ * ecryptfs_destroy_inode
+ * @inode: The ecryptfs inode
+ *
+ * This is used during the final destruction of the inode.  All
+ * allocation of memory related to the inode, including allocated
+ * memory in the crypt_stat struct, will be released here.
+ * There should be no chance that this deallocation will be missed.
+ */
+static void ecryptfs_destroy_inode(struct inode *inode)
+{
+	struct ecryptfs_inode_info *inode_info;
+
+	inode_info = ecryptfs_inode_to_private(inode);
+	BUG_ON(inode_info->lower_file);
+	ecryptfs_destroy_crypt_stat(&inode_info->crypt_stat);
+}
+
+/**
+ * ecryptfs_statfs
+ * @dentry: The ecryptfs dentry
+ * @buf: The struct kstatfs to fill in with stats
+ *
+ * Get the filesystem statistics. Currently, we let this pass right through
+ * to the lower filesystem and take no action ourselves.
+ */
+static int ecryptfs_statfs(struct dentry *dentry, struct kstatfs *buf)
+{
+	struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
+	int rc;
+
+	if (!lower_dentry->d_sb->s_op->statfs)
+		return -ENOSYS;
+
+	rc = lower_dentry->d_sb->s_op->statfs(lower_dentry, buf);
+	if (rc)
+		return rc;
+
+	buf->f_type = ECRYPTFS_SUPER_MAGIC;
+	rc = ecryptfs_set_f_namelen(&buf->f_namelen, buf->f_namelen,
+	       &ecryptfs_superblock_to_private(dentry->d_sb)->mount_crypt_stat);
+
+	return rc;
+}
+
+/**
+ * ecryptfs_evict_inode
+ * @inode: The ecryptfs inode
+ *
+ * Called by iput() when the inode reference count reached zero
+ * and the inode is not hashed anywhere.  Used to clear anything
+ * that needs to be, before the inode is completely destroyed and put
+ * on the inode free list. We use this to drop out reference to the
+ * lower inode.
+ */
+static void ecryptfs_evict_inode(struct inode *inode)
+{
+	truncate_inode_pages_final(&inode->i_data);
+	clear_inode(inode);
+	iput(ecryptfs_inode_to_lower(inode));
+}
+
+/*
+ * ecryptfs_show_options
+ *
+ * Prints the mount options for a given superblock.
+ * Returns zero; does not fail.
+ */
+static int ecryptfs_show_options(struct seq_file *m, struct dentry *root)
+{
+	struct super_block *sb = root->d_sb;
+	struct ecryptfs_mount_crypt_stat *mount_crypt_stat =
+		&ecryptfs_superblock_to_private(sb)->mount_crypt_stat;
+	struct ecryptfs_global_auth_tok *walker;
+
+	mutex_lock(&mount_crypt_stat->global_auth_tok_list_mutex);
+	list_for_each_entry(walker,
+			    &mount_crypt_stat->global_auth_tok_list,
+			    mount_crypt_stat_list) {
+		if (walker->flags & ECRYPTFS_AUTH_TOK_FNEK)
+			seq_printf(m, ",ecryptfs_fnek_sig=%s", walker->sig);
+		else
+			seq_printf(m, ",ecryptfs_sig=%s", walker->sig);
+	}
+	mutex_unlock(&mount_crypt_stat->global_auth_tok_list_mutex);
+
+	seq_printf(m, ",ecryptfs_cipher=%s",
+		mount_crypt_stat->global_default_cipher_name);
+
+	if (mount_crypt_stat->global_default_cipher_key_size)
+		seq_printf(m, ",ecryptfs_key_bytes=%zd",
+			   mount_crypt_stat->global_default_cipher_key_size);
+	if (mount_crypt_stat->flags & ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)
+		seq_printf(m, ",ecryptfs_passthrough");
+	if (mount_crypt_stat->flags & ECRYPTFS_XATTR_METADATA_ENABLED)
+		seq_printf(m, ",ecryptfs_xattr_metadata");
+	if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED)
+		seq_printf(m, ",ecryptfs_encrypted_view");
+	if (mount_crypt_stat->flags & ECRYPTFS_UNLINK_SIGS)
+		seq_printf(m, ",ecryptfs_unlink_sigs");
+	if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_MOUNT_AUTH_TOK_ONLY)
+		seq_printf(m, ",ecryptfs_mount_auth_tok_only");
+
+	return 0;
+}
+
+const struct super_operations ecryptfs_sops = {
+	.alloc_inode = ecryptfs_alloc_inode,
+	.destroy_inode = ecryptfs_destroy_inode,
+	.free_inode = ecryptfs_free_inode,
+	.statfs = ecryptfs_statfs,
+	.evict_inode = ecryptfs_evict_inode,
+	.show_options = ecryptfs_show_options
+};