On Mon 2018-05-14 12:15:05, Uwe Kleine-König wrote:
> The rx trigger fires when data is pushed to the ldisc by the driver. This
> is a bit later than the actual receiving of data but has the nice benefit
> that it doesn't need adaption for each driver and isn't in the hot path.
>
> Similarly the tx trigger fires when data was copied from userspace and is
> given to the ldisc.
>
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
> ---
> drivers/tty/Kconfig | 7 +++++++
> drivers/tty/tty_buffer.c | 2 ++
> drivers/tty/tty_io.c | 3 +++
> drivers/tty/tty_port.c | 32 ++++++++++++++++++++++++++++++--
> include/linux/tty.h | 22 ++++++++++++++++++++++
> 5 files changed, 64 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/tty/Kconfig b/drivers/tty/Kconfig
> index 0840d27381ea..b119c0fa1f5a 100644
> --- a/drivers/tty/Kconfig
> +++ b/drivers/tty/Kconfig
> @@ -41,6 +41,13 @@ config VT
> If unsure, say Y, or else you won't be able to do much with your new
> shiny Linux system :-)
>
> +config TTY_LEDS_TRIGGERS
> + bool "Enable support for TTY actions making LEDs blink"
> + depends on LEDS_TRIGGERS
> + ---help---
> + This enable support for tty triggers. It provides two LED triggers
> + (rx and tx) for each TTY.
> +
> config CONSOLE_TRANSLATIONS
> depends on VT
> default y
> diff --git a/drivers/tty/tty_buffer.c b/drivers/tty/tty_buffer.c
> index c996b6859c5e..364080ce8e91 100644
> --- a/drivers/tty/tty_buffer.c
> +++ b/drivers/tty/tty_buffer.c
> @@ -521,6 +521,8 @@ static void flush_to_ldisc(struct work_struct *work)
> continue;
> }
>
> + tty_led_trigger_rx(port);
> +
> count = receive_buf(port, head, count);
> if (!count)
> break;
> diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c
> index 7c838b90a31d..8ef597dc0c3d 100644
> --- a/drivers/tty/tty_io.c
> +++ b/drivers/tty/tty_io.c
> @@ -955,6 +955,9 @@ static inline ssize_t do_tty_write(
> ret = -EFAULT;
> if (copy_from_user(tty->write_buf, buf, size))
> break;
> +
> + tty_led_trigger_tx(tty->port);
> +
> ret = write(tty, file, tty->write_buf, size);
> if (ret <= 0)
> break;
> diff --git a/drivers/tty/tty_port.c b/drivers/tty/tty_port.c
> index 25d736880013..72dcde234a7d 100644
> --- a/drivers/tty/tty_port.c
> +++ b/drivers/tty/tty_port.c
> @@ -37,6 +37,8 @@ static int tty_port_default_receive_buf(struct tty_port *port,
>
> ret = tty_ldisc_receive_buf(disc, p, (char *)f, count);
>
> + tty_led_trigger_rx(port);
> +
> tty_ldisc_deref(disc);
>
> return ret;
> @@ -163,8 +165,31 @@ struct device *tty_port_register_device_attr_serdev(struct tty_port *port,
> return dev;
> }
>
> - return tty_register_device_attr(driver, index, device, drvdata,
> - attr_grp);
> + if (IS_ENABLED(CONFIG_TTY_LEDS_TRIGGERS)) {
> + int ret;
> +
> + ret = led_trigger_register_format(&port->led_trigger_rx,
> + "%s%d-rx", driver->name, index);
> + if (ret < 0)
> + pr_warn("Failed to register rx trigger for %s%d (%d)\n",
> + driver->name, index, ret);
> +
> + ret = led_trigger_register_format(&port->led_trigger_tx,
> + "%s%d-tx", driver->name, index);
> + if (ret < 0)
> + pr_warn("Failed to register tx trigger for %s%d (%d)\n",
> + driver->name, index, ret);
> + }
There are many ttys and you'll eat quite a bit of kernel memory with this (besides other
effects).
Could we get trigger which takes tty name and rx/tx as a parameter? That should save a bit of
RAM and some headaches with .../triggers file being too big.
Thanks,
Pavel
--
To unsubscribe from this list: send the line "unsubscribe linux-serial" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
