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, 114 insertions, 0 deletions

diff --git a/io_uring/sync.c b/io_uring/sync.c
new file mode 100644
index 000000000..ab7fa1cd7
--- /dev/null
+++ b/io_uring/sync.c
@@ -0,0 +1,114 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/namei.h>
+#include <linux/io_uring.h>
+#include <linux/fsnotify.h>
+
+#include <uapi/linux/io_uring.h>
+
+#include "io_uring.h"
+#include "sync.h"
+
+struct io_sync {
+	struct file			*file;
+	loff_t				len;
+	loff_t				off;
+	int				flags;
+	int				mode;
+};
+
+int io_sfr_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_sync *sync = io_kiocb_to_cmd(req, struct io_sync);
+
+	if (unlikely(sqe->addr || sqe->buf_index || sqe->splice_fd_in))
+		return -EINVAL;
+
+	sync->off = READ_ONCE(sqe->off);
+	sync->len = READ_ONCE(sqe->len);
+	sync->flags = READ_ONCE(sqe->sync_range_flags);
+	req->flags |= REQ_F_FORCE_ASYNC;
+
+	return 0;
+}
+
+int io_sync_file_range(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_sync *sync = io_kiocb_to_cmd(req, struct io_sync);
+	int ret;
+
+	/* sync_file_range always requires a blocking context */
+	WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK);
+
+	ret = sync_file_range(req->file, sync->off, sync->len, sync->flags);
+	io_req_set_res(req, ret, 0);
+	return IOU_COMPLETE;
+}
+
+int io_fsync_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_sync *sync = io_kiocb_to_cmd(req, struct io_sync);
+
+	if (unlikely(sqe->addr || sqe->buf_index || sqe->splice_fd_in))
+		return -EINVAL;
+
+	sync->flags = READ_ONCE(sqe->fsync_flags);
+	if (unlikely(sync->flags & ~IORING_FSYNC_DATASYNC))
+		return -EINVAL;
+
+	sync->off = READ_ONCE(sqe->off);
+	if (sync->off < 0)
+		return -EINVAL;
+	sync->len = READ_ONCE(sqe->len);
+	req->flags |= REQ_F_FORCE_ASYNC;
+	return 0;
+}
+
+int io_fsync(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_sync *sync = io_kiocb_to_cmd(req, struct io_sync);
+	loff_t end = sync->off + sync->len;
+	int ret;
+
+	/* fsync always requires a blocking context */
+	WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK);
+
+	ret = vfs_fsync_range(req->file, sync->off, end > 0 ? end : LLONG_MAX,
+				sync->flags & IORING_FSYNC_DATASYNC);
+	io_req_set_res(req, ret, 0);
+	return IOU_COMPLETE;
+}
+
+int io_fallocate_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_sync *sync = io_kiocb_to_cmd(req, struct io_sync);
+
+	if (sqe->buf_index || sqe->rw_flags || sqe->splice_fd_in)
+		return -EINVAL;
+
+	sync->off = READ_ONCE(sqe->off);
+	sync->len = READ_ONCE(sqe->addr);
+	sync->mode = READ_ONCE(sqe->len);
+	req->flags |= REQ_F_FORCE_ASYNC;
+	return 0;
+}
+
+int io_fallocate(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_sync *sync = io_kiocb_to_cmd(req, struct io_sync);
+	int ret;
+
+	/* fallocate always requiring blocking context */
+	WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK);
+
+	ret = vfs_fallocate(req->file, sync->mode, sync->off, sync->len);
+	if (ret >= 0)
+		fsnotify_modify(req->file);
+	io_req_set_res(req, ret, 0);
+	return IOU_COMPLETE;
+}