Skip to contentSkip to conversation

serial: 8250: Fix runtime PM for start_tx() for empty buffer

PATCH In review Tony Lindgren 3 participants 3 messages Apr 11, 2022 TTY layer and serial drivers

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 serial: 8250: Fix runtime PM for start_tx() for empty buffer

Commit 932d596378b0 ("serial: 8250: Return early in .start_tx() if there are no chars to send") caused a regression where the drivers implementing runtime PM stopped idling. This is because serial8250_rpm_put_tx() is now unbalanced on early return, it normally gets called at __stop_tx().

Fixes 932d596378b0 ("serial: 8250: Return early in .start_tx() if there are no chars to send")
Cc Steffen Trumtrar <[email protected]>
Cc Uwe Kleine-König <[email protected]>

---

Changes since v1:

- Call serial8250_rpm_get_tx() only after buffer empty test as suggested by Uwe and Johan

- Drop RS885 related patch for now, it needs to be investigated further

---
 drivers/tty/serial/8250/8250_port.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/tty/serial/8250/8250_port.c +2 -2
@@ -1675,11 +1675,11 @@ static void serial8250_start_tx(struct uart_port *port)
 	struct uart_8250_port *up = up_to_u8250p(port);
 	struct uart_8250_em485 *em485 = up->em485;
 
-	serial8250_rpm_get_tx(up);
-
 	if (!port->x_char && uart_circ_empty(&port->state->xmit))
 		return;
 
+	serial8250_rpm_get_tx(up);
+
 	if (em485 &&
 	    em485->active_timer == &em485->start_tx_timer)
 		return;
Signature · 1 line
2.35.1
Reviewed-by on v2 1/1
#1 lore raw
Quoted 29 lines · Tony Lindgren

Commit 932d596378b0 ("serial: 8250: Return early in .start_tx() if there are no chars to send") caused a regression where the drivers implementing runtime PM stopped idling. This is because serial8250_rpm_put_tx() is now unbalanced on early return, it normally gets called at __stop_tx().

Fixes 932d596378b0 ("serial: 8250: Return early in .start_tx() if there are no chars to send")
Cc Steffen Trumtrar <[email protected]>
Cc Uwe Kleine-König <[email protected]>

---

Changes since v1:

- Call serial8250_rpm_get_tx() only after buffer empty test as suggested by Uwe and Johan

- Drop RS885 related patch for now, it needs to be investigated further

---
 drivers/tty/serial/8250/8250_port.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/tty/serial/8250/8250_port.c +0 -0
@@ -1675,11 +1675,11 @@ static void serial8250_start_tx(struct uart_port *port)
 	struct uart_8250_port *up = up_to_u8250p(port);
 	struct uart_8250_em485 *em485 = up->em485;
 

drivers/tty/serial/8250/8250_port.c static void serial8250_start_tx(struct uart_port *port) · line 1685

Quoted 10 lines
drivers/tty/serial/8250/8250_port.c +2 -2
-	serial8250_rpm_get_tx(up);
-
 	if (!port->x_char && uart_circ_empty(&port->state->xmit))
 		return;
 
+	serial8250_rpm_get_tx(up);
+
 	if (em485 &&
 	    em485->active_timer == &em485->start_tx_timer)
 		return;
Reviewed-by on v2 1/1
#2 lore raw
In reply to Tony Lindgren
Quoted 9 lines · Tony Lindgren

Commit 932d596378b0 ("serial: 8250: Return early in .start_tx() if there are no chars to send") caused a regression where the drivers implementing runtime PM stopped idling. This is because serial8250_rpm_put_tx() is now unbalanced on early return, it normally gets called at __stop_tx().

Fixes 932d596378b0 ("serial: 8250: Return early in .start_tx() if there are no chars to send")
Cc Steffen Trumtrar <[email protected]>
Cc Uwe Kleine-König <[email protected]>

Thanks for cleaning up behind me,
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |