On 21/06/2017 11:30, Uwe Kleine-König wrote:
> Most functions in the atmel serial driver check for pdata being non-NULL before
> making use of it, atmel_init_property and atmel_init_rs485() don't.
> Probably this isn't a big problem because devices that are not probed
> from device tree relied on this before
>
> 5fbe46b67680 ("tty/serial: atmel_serial: add device tree support")
>
> (contained in v3.2-rc1) and I guess since then the platform devices
> didn't change much. Make this consistent and safe anyhow.
>
> I didn't see (or try to find) a platform crashing here, only noticed
> this inconsistency while working on this driver.
>
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
> ---
> drivers/tty/serial/atmel_serial.c | 16 ++++++++++++----
> 1 file changed, 12 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
> index 906e2b91bf57..b753d2020068 100644
> --- a/drivers/tty/serial/atmel_serial.c
> +++ b/drivers/tty/serial/atmel_serial.c
> @@ -1686,8 +1686,13 @@ static void atmel_init_property(struct atmel_uart_port *atmel_port,
> }
>
> } else {
> - atmel_port->use_pdc_rx = pdata->use_dma_rx;
> - atmel_port->use_pdc_tx = pdata->use_dma_tx;
> + if (pdata) {
> + atmel_port->use_pdc_rx = pdata->use_dma_rx;
> + atmel_port->use_pdc_tx = pdata->use_dma_tx;
> + } else {
> + atmel_port->use_pdc_rx = false;
> + atmel_port->use_pdc_tx = false;
> + }
> atmel_port->use_dma_rx = false;
> atmel_port->use_dma_tx = false;
> }
> @@ -1717,10 +1722,13 @@ static void atmel_init_rs485(struct uart_port *port,
> if (of_get_property(np, "linux,rs485-enabled-at-boot-time",
> NULL))
> rs485conf->flags |= SER_RS485_ENABLED;
> + } else if (pdata) {
> + port->rs485 = pdata->rs485;
> } else {
> - port->rs485 = pdata->rs485;
> - }
> + struct serial_rs485 default_rs485 = { 0, };
>
> + port->rs485 = default_rs485;
> + }
> }
>
> static void atmel_set_ops(struct uart_port *port)
>
In patch https://www.mail-archive.com/linux-kernel@xxxxxxxxxxxxxxx/msg1420769.html
Alexandre got rid of pdata altogether, so I guess the inconsistency will gone with pdata :)
Richard.
--
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
