* [PATCH] tty: serial: lpuart: Add port lock for registers read in lpuart32_set_termios()
@ 2026-11-11 9:53 Sherry Sun (OSS)
2026-11-11 10:06 ` sashiko-bot
0 siblings, 1 reply; 2+ messages in thread
From: Sherry Sun (OSS) @ 2026-11-11 9:53 UTC (permalink / raw)
To: Greg Kroah-Hartman, Jiri Slaby, Jingchang Lu
Cc: linux-kernel, linux-serial, imx, sherry.sun
From: Sherry Sun <sherry.sun@nxp.com>
lpuart32_set_termios() reads UARTCTRL at function entry without holding
the port lock. lpuart32_console_write() can run concurrently on another
CPU and temporarily clears the RIE (Receiver Interrupt Enable) bit while
it polls characters out. If the unlocked read in set_termios lands inside
that window, the stale snapshot is written back, permanently clearing
RIE. Once lost, no normal code path re-enables them, so the UART stops
receiving characters and becomes unresponsive.
Add port lock for the registers read in lpuart32_set_termios() to avoid
the unlocked snapshot captured.
Please note that this only fix the normal printk path, in the
oops_in_progress path of lpuart32_console_write(), when trylock fails,
RIE is cleared and old_cr restored without the port lock held, there is
still a risk of stale snapshot, but it's acceptable for the post-oops
state and the trylock-fail window.
Fixes: 380c966c093e ("tty: serial: fsl_lpuart: add 32-bit register interface support")
Signed-off-by: Sherry Sun <sherry.sun@nxp.com>
---
drivers/tty/serial/fsl_lpuart.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c
index cd2575a3aef7..7c249b9371ae 100644
--- a/drivers/tty/serial/fsl_lpuart.c
+++ b/drivers/tty/serial/fsl_lpuart.c
@@ -2241,9 +2241,11 @@ lpuart32_set_termios(struct uart_port *port, struct ktermios *termios,
unsigned int baud;
unsigned int old_csize = old ? old->c_cflag & CSIZE : CS8;
+ uart_port_lock_irqsave(port, &flags);
ctrl = old_ctrl = lpuart32_read(port, UARTCTRL);
bd = lpuart32_read(port, UARTBAUD);
modem = lpuart32_read(port, UARTMODIR);
+ uart_port_unlock_irqrestore(port, flags);
sport->is_cs7 = false;
/*
* only support CS8 and CS7
---
base-commit: 793807ca13bf3072fc6d16ce6d90e9ab9beff2a5
change-id: 20261111-lpuart_port_lock-01f7deab451a
Best regards,
--
Sherry Sun <sherry.sun@nxp.com>
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] tty: serial: lpuart: Add port lock for registers read in lpuart32_set_termios()
2026-11-11 9:53 [PATCH] tty: serial: lpuart: Add port lock for registers read in lpuart32_set_termios() Sherry Sun (OSS)
@ 2026-11-11 10:06 ` sashiko-bot
0 siblings, 0 replies; 2+ messages in thread
From: sashiko-bot @ 2026-11-11 10:06 UTC (permalink / raw)
To: Sherry Sun (OSS); +Cc: linux-serial, imx, Frank.Li
Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Medium] tty: serial: lpuart: split port->lock race in
lpuart32_set_termios()
--
commit c6586e620953933a073532d069075858d2abff60
Author: Sherry Sun <sherry.sun@nxp.com>
tty: serial: lpuart: Add port lock for registers read in lpuart32_set_termios()
This commit wraps the initial register reads of UARTCTRL, UARTBAUD, and
UARTMODIR in lpuart32_set_termios() with uart_port_lock_irqsave() to avoid
reading a stale UARTCTRL value while lpuart32_console_write() temporarily
clears RIE.
> diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c
> index cd2575a3aef74..7c249b9371ae8 100644
> --- a/drivers/tty/serial/fsl_lpuart.c
> +++ b/drivers/tty/serial/fsl_lpuart.c
> @@ -2241,9 +2241,11 @@ lpuart32_set_termios(struct uart_port *port, struct ktermios *termios,
> unsigned int baud;
> unsigned int old_csize = old ? old->c_cflag & CSIZE : CS8;
>
> + uart_port_lock_irqsave(port, &flags);
> ctrl = old_ctrl = lpuart32_read(port, UARTCTRL);
> bd = lpuart32_read(port, UARTBAUD);
> modem = lpuart32_read(port, UARTMODIR);
> + uart_port_unlock_irqrestore(port, flags);
> sport->is_cs7 = false;
> /*
> * only support CS8 and CS7
[Severity: Medium]
Does releasing port->lock right after reading UARTCTRL, UARTBAUD, and
UARTMODIR still leave a window where ctrl and old_ctrl become stale before
they are written back to hardware?
CPU0 CPU1
lpuart32_set_termios()
uart_port_lock_irqsave()
ctrl = old_ctrl =
lpuart32_read(UARTCTRL)
(UARTCTRL_TIE is 0)
uart_port_unlock_irqrestore()
uart_write()
lpuart32_start_tx()
uart_port_lock_irqsave()
lpuart32_write(ctrl | UARTCTRL_TIE,
UARTCTRL)
uart_port_unlock_irqrestore()
uart_get_baud_rate()
lpuart_dma_rx_free()
uart_port_lock_irqsave()
...
lpuart32_write(ctrl, UARTCTRL) <-- clears UARTCTRL_TIE, stalling TX
Callers such as lpuart32_start_tx(), lpuart32_stop_tx(),
lpuart32_stop_rx(), and lpuart32_set_mctrl() update UARTCTRL under
port->lock without holding tty_port->mutex. Any modification made during
the unlocked window is overwritten when lpuart32_set_termios() re-acquires
port->lock and writes old_ctrl and ctrl back to UARTCTRL. Clearing
sport->is_cs7 outside port->lock also races with lpuart32_rxint() and
lpuart_copy_rx_to_tty().
Since neither uart_get_baud_rate() nor lpuart_dma_rx_free() uses ctrl,
old_ctrl, bd, or modem, the register reads and bit manipulations could be
moved into the existing uart_port_lock_irqsave() critical section after
lpuart_dma_rx_free() so the read-modify-write stays under a single lock
acquisition.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261111-lpuart_port_lock-v1-1-5d332c51edfe@nxp.com?part=1
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-11-11 10:06 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-11 9:53 [PATCH] tty: serial: lpuart: Add port lock for registers read in lpuart32_set_termios() Sherry Sun (OSS)
2026-11-11 10:06 ` sashiko-bot
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox