[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <X7d7HXBp01JN5It5@localhost>
Date: Fri, 20 Nov 2020 09:15:25 +0100
From: Johan Hovold <johan@...nel.org>
To: Tian Tao <tiantao6@...ilicon.com>
Cc: gregkh@...uxfoundation.org, jirislaby@...nel.org,
linux-arm-kernel@...ts.infradead.org,
linux-unisoc@...ts.infradead.org, linux-serial@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] tty: serial: rad-uart: replace spin_lock_irqsave by
spin_lock in hard IRQ
On Fri, Nov 20, 2020 at 09:26:53AM +0800, Tian Tao wrote:
> The code has been in a irq-disabled context since it is hard IRQ. There
> is no necessity to do it again.
>
> Signed-off-by: Tian Tao <tiantao6@...ilicon.com>
> ---
> drivers/tty/serial/rda-uart.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/tty/serial/rda-uart.c b/drivers/tty/serial/rda-uart.c
> index 85366e0..d6705a0 100644
> --- a/drivers/tty/serial/rda-uart.c
> +++ b/drivers/tty/serial/rda-uart.c
> @@ -406,10 +406,9 @@ static void rda_uart_receive_chars(struct uart_port *port)
> static irqreturn_t rda_interrupt(int irq, void *dev_id)
> {
> struct uart_port *port = dev_id;
> - unsigned long flags;
> u32 val, irq_mask;
>
> - spin_lock_irqsave(&port->lock, flags);
> + spin_lock(&port->lock);
This will break with forced irq threading (i.e. "threadirqs") since the
console code can still end up being called from interrupt context (which
can result in a deadlock).
>
> /* Clear IRQ cause */
> val = rda_uart_read(port, RDA_UART_IRQ_CAUSE);
> @@ -426,7 +425,7 @@ static irqreturn_t rda_interrupt(int irq, void *dev_id)
> rda_uart_send_chars(port);
> }
>
> - spin_unlock_irqrestore(&port->lock, flags);
> + spin_unlock(&port->lock);
>
> return IRQ_HANDLED;
> }
Johan
Powered by blists - more mailing lists