From: Thomas Gleixner <[email protected]>
When a serial port is used for kernel console output, then all
modifications to the UART registers which are done from other contexts,
e.g. getty, termios, are interference points for the kernel console.
So far this has been ignored and the printk output is based on the
principle of hope. The rework of the console infrastructure which aims to
support threaded and atomic consoles, requires to mark sections which
modify the UART registers as unsafe. This allows the atomic write function
to make informed decisions and eventually to restore operational state. It
also allows to prevent the regular UART code from modifying UART registers
while printk output is in progress.
All modifications of UART registers are guarded by the UART port lock,
which provides an obvious synchronization point with the console
infrastructure.
To avoid adding this functionality to all UART drivers, wrap the
spin_[un]lock*() invocations for uart_port::lock into helper functions
which just contain the spin_[un]lock*() invocations for now. In a
subsequent step these helpers will gain the console synchronization
mechanisms.
Converted with coccinelle. No functional change.
Signed-off-by: Thomas Gleixner <[email protected]>
---
drivers/tty/serial/qcom_geni_serial.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/tty/serial/qcom_geni_serial.c b/drivers/tty/serial/qcom_geni_serial.c
index b8aa4c1293ba..7e78f97e8f43 100644
--- a/drivers/tty/serial/qcom_geni_serial.c
+++ b/drivers/tty/serial/qcom_geni_serial.c
@@ -482,9 +482,9 @@ static void qcom_geni_serial_console_write(struct console *co, const char *s,
uport = &port->uport;
if (oops_in_progress)
- locked = spin_trylock_irqsave(&uport->lock, flags);
+ locked = uart_port_trylock_irqsave(uport, &flags);
else
- spin_lock_irqsave(&uport->lock, flags);
+ uart_port_lock_irqsave(uport, &flags);
geni_status = readl(uport->membase + SE_GENI_STATUS);
@@ -520,7 +520,7 @@ static void qcom_geni_serial_console_write(struct console *co, const char *s,
qcom_geni_serial_setup_tx(uport, port->tx_remaining);
if (locked)
- spin_unlock_irqrestore(&uport->lock, flags);
+ uart_port_unlock_irqrestore(uport, flags);
}
static void handle_rx_console(struct uart_port *uport, u32 bytes, bool drop)
@@ -970,7 +970,7 @@ static irqreturn_t qcom_geni_serial_isr(int isr, void *dev)
if (uport->suspended)
return IRQ_NONE;
- spin_lock(&uport->lock);
+ uart_port_lock(uport);
m_irq_status = readl(uport->membase + SE_GENI_M_IRQ_STATUS);
s_irq_status = readl(uport->membase + SE_GENI_S_IRQ_STATUS);
--
2.39.2
On Thu, Sep 14, 2023 at 08:44:06PM +0206, John Ogness wrote:
> From: Thomas Gleixner <[email protected]>
>
> When a serial port is used for kernel console output, then all
> modifications to the UART registers which are done from other contexts,
> e.g. getty, termios, are interference points for the kernel console.
>
> So far this has been ignored and the printk output is based on the
> principle of hope. The rework of the console infrastructure which aims to
> support threaded and atomic consoles, requires to mark sections which
> modify the UART registers as unsafe. This allows the atomic write function
> to make informed decisions and eventually to restore operational state. It
> also allows to prevent the regular UART code from modifying UART registers
> while printk output is in progress.
>
> All modifications of UART registers are guarded by the UART port lock,
> which provides an obvious synchronization point with the console
> infrastructure.
>
> To avoid adding this functionality to all UART drivers, wrap the
> spin_[un]lock*() invocations for uart_port::lock into helper functions
> which just contain the spin_[un]lock*() invocations for now. In a
> subsequent step these helpers will gain the console synchronization
> mechanisms.
>
> Converted with coccinelle. No functional change.
>
> Signed-off-by: Thomas Gleixner <[email protected]>
Reviewed-by: Bjorn Andersson <[email protected]>
> ---
> drivers/tty/serial/qcom_geni_serial.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/tty/serial/qcom_geni_serial.c b/drivers/tty/serial/qcom_geni_serial.c
> index b8aa4c1293ba..7e78f97e8f43 100644
> --- a/drivers/tty/serial/qcom_geni_serial.c
> +++ b/drivers/tty/serial/qcom_geni_serial.c
> @@ -482,9 +482,9 @@ static void qcom_geni_serial_console_write(struct console *co, const char *s,
>
> uport = &port->uport;
> if (oops_in_progress)
> - locked = spin_trylock_irqsave(&uport->lock, flags);
> + locked = uart_port_trylock_irqsave(uport, &flags);
> else
> - spin_lock_irqsave(&uport->lock, flags);
> + uart_port_lock_irqsave(uport, &flags);
>
> geni_status = readl(uport->membase + SE_GENI_STATUS);
>
> @@ -520,7 +520,7 @@ static void qcom_geni_serial_console_write(struct console *co, const char *s,
> qcom_geni_serial_setup_tx(uport, port->tx_remaining);
>
> if (locked)
> - spin_unlock_irqrestore(&uport->lock, flags);
> + uart_port_unlock_irqrestore(uport, flags);
> }
>
> static void handle_rx_console(struct uart_port *uport, u32 bytes, bool drop)
> @@ -970,7 +970,7 @@ static irqreturn_t qcom_geni_serial_isr(int isr, void *dev)
> if (uport->suspended)
> return IRQ_NONE;
>
> - spin_lock(&uport->lock);
> + uart_port_lock(uport);
>
Nice, now this no longer look unbalanced with the unlock in
uart_port_unlock_irqrestore().
Regards,
Bjorn
> m_irq_status = readl(uport->membase + SE_GENI_M_IRQ_STATUS);
> s_irq_status = readl(uport->membase + SE_GENI_S_IRQ_STATUS);
> --
> 2.39.2
>