On Wed, 2018-06-06 at 11:49 +0200, Giulio Benetti wrote:
> If rs485 is enabled and RTS_AFTER_SEND is set on startup need to keep
> TIOCM_RTS asserted to keep rs485 transceiver in RX when idle.
>
> Check if rs485 is on and RTS_AFTER_SEND is set and mask port->mctrl
> with
> TIOCM_RTS too and not only TIOCM_DTR.
>
This one feels wrong to be in serial_core.c. Perhaps in 8250/8250*.c.
> Signed-off-by: Giulio Benetti <giulio.bene...@micronovasrl.com>
> ---
> drivers/tty/serial/serial_core.c | 12 +++++++++++-
> 1 file changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/tty/serial/serial_core.c
> b/drivers/tty/serial/serial_core.c
> index 0466f9f08a91..06d9441f6d20 100644
> --- a/drivers/tty/serial/serial_core.c
> +++ b/drivers/tty/serial/serial_core.c
> @@ -2288,6 +2288,16 @@ uart_configure_port(struct uart_driver *drv,
> struct uart_state *state,
>
> if (port->type != PORT_UNKNOWN) {
> unsigned long flags;
> + int rs485_on = port->rs485_config &&
> + (port->rs485.flags & SER_RS485_ENABLED);
> + int RTS_after_send = !!(port->rs485.flags &
> + SER_RS485_RTS_AFTER_SEND);
> + int mctrl;
> +
> + if (rs485_on && RTS_after_send)
> + mctrl = port->mctrl & (TIOCM_DTR |
> TIOCM_RTS);
> + else
> + mctrl = port->mctrl & TIOCM_DTR;
>
> uart_report_port(drv, port);
>
> @@ -2300,7 +2310,7 @@ uart_configure_port(struct uart_driver *drv,
> struct uart_state *state,
> * We probably don't need a spinlock around this, but
> */
> spin_lock_irqsave(&port->lock, flags);
> - port->ops->set_mctrl(port, port->mctrl & TIOCM_DTR);
> + port->ops->set_mctrl(port, mctrl);
> spin_unlock_irqrestore(&port->lock, flags);
>
> /*
--
Andy Shevchenko <andriy.shevche...@linux.intel.com>
Intel Finland Oy
