[tty,v1,6/8] serial: 8250: lock port for omap8250_restore_regs()
Commit Message
omap8250_restore_regs() accesses UART_IER. This register is
modified twice by each console write (serial8250_console_write())
under the port lock. However, not all calls to omap8250_restore_regs()
are under the port lock.
Add the missing port locking around omap8250_restore_regs() calls. Add
lockdep notation to the omap8250_restore_regs().
Note that this is not fixing a real problem because the serial devices
are resumed before console printing is enabled.
However, adding this locking allows for clean locking semantics
for omap8250_restore_regs() so that lockdep can be used to identify
possible problems in the future. It also simplifies synchronization
of UART_IER in general by not needing to rely on such implementation
details.
Signed-off-by: John Ogness <john.ogness@linutronix.de>
---
drivers/tty/serial/8250/8250_omap.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
@@ -309,6 +309,9 @@ static void omap8250_restore_regs(struct uart_8250_port *up)
struct uart_8250_dma *dma = up->dma;
u8 mcr = serial8250_in_MCR(up);
+ /* Port locked to synchronize UART_IER access against the console. */
+ lockdep_assert_held_once(&up->port.lock);
+
if (dma && dma->tx_running) {
/*
* TCSANOW requests the change to occur immediately however if
@@ -1739,8 +1742,11 @@ static int omap8250_runtime_resume(struct device *dev)
if (priv->line >= 0)
up = serial8250_get_port(priv->line);
- if (up && omap8250_lost_context(up))
+ if (up && omap8250_lost_context(up)) {
+ spin_lock_irq(&up->port.lock);
omap8250_restore_regs(up);
+ spin_unlock_irq(&up->port.lock);
+ }
if (up && up->dma && up->dma->rxchan && !(priv->habit & UART_HAS_EFR2)) {
spin_lock_irq(&up->port.lock);