tty/serial: do not free trasnmit buffer page under port lock

LKP has hit yet another circular locking dependency between uart
console drivers and debugobjects [1]:

     CPU0                                    CPU1

                                            rhltable_init()
                                             __init_work()
                                              debug_object_init
     uart_shutdown()                          /* db->lock */
      /* uart_port->lock */                    debug_print_object()
       free_page()                              printk()
                                                 call_console_drivers()
        debug_check_no_obj_freed()                /* uart_port->lock */
         /* db->lock */
          debug_print_object()

So there are two dependency chains:
	uart_port->lock -> db->lock
And
	db->lock -> uart_port->lock

This particular circular locking dependency can be addressed in several
ways:

a) One way would be to move debug_print_object() out of db->lock scope
   and, thus, break the db->lock -> uart_port->lock chain.
b) Another one would be to free() transmit buffer page out of db->lock
   in UART code; which is what this patch does.

It makes sense to apply a) and b) independently: there are too many things
going on behind free(), none of which depend on uart_port->lock.

The patch fixes transmit buffer page free() in uart_shutdown() and,
additionally, in uart_port_startup() (as was suggested by Dmitry Safonov).

[1] https://lore.kernel.org/lkml/20181211091154.GL23332@shao2-debian/T/#u
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Petr Mladek <pmladek@suse.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Waiman Long <longman@redhat.com>
Cc: Dmitry Safonov <dima@arista.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

1 files changed, 16 insertions, 6 deletions

diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c
index c439a5a1e6c0..d4cca5bdaf1c 100644
--- a/drivers/tty/serial/serial_core.c
+++ b/drivers/tty/serial/serial_core.c
@@ -205,10 +205,15 @@ static int uart_port_startup(struct tty_struct *tty, struct uart_state *state,
 	if (!state->xmit.buf) {
 		state->xmit.buf = (unsigned char *) page;
 		uart_circ_clear(&state->xmit);
+		uart_port_unlock(uport, flags);
 	} else {
+		uart_port_unlock(uport, flags);
+		/*
+		 * Do not free() the page under the port lock, see
+		 * uart_shutdown().
+		 */
 		free_page(page);
 	}
-	uart_port_unlock(uport, flags);
 
 	retval = uport->ops->startup(uport);
 	if (retval == 0) {
@@ -268,6 +273,7 @@ static void uart_shutdown(struct tty_struct *tty, struct uart_state *state)
 	struct uart_port *uport = uart_port_check(state);
 	struct tty_port *port = &state->port;
 	unsigned long flags = 0;
+	char *xmit_buf = NULL;
 
 	/*
 	 * Set the TTY IO error marker
@@ -298,14 +304,18 @@ static void uart_shutdown(struct tty_struct *tty, struct uart_state *state)
 	tty_port_set_suspended(port, 0);
 
 	/*
-	 * Free the transmit buffer page.
+	 * Do not free() the transmit buffer page under the port lock since
+	 * this can create various circular locking scenarios. For instance,
+	 * console driver may need to allocate/free a debug object, which
+	 * can endup in printk() recursion.
 	 */
 	uart_port_lock(state, flags);
-	if (state->xmit.buf) {
-		free_page((unsigned long)state->xmit.buf);
-		state->xmit.buf = NULL;
-	}
+	xmit_buf = state->xmit.buf;
+	state->xmit.buf = NULL;
 	uart_port_unlock(uport, flags);
+
+	if (xmit_buf)
+		free_page((unsigned long)xmit_buf);
 }
 
 /**