On Thu, Dec 01, 2016 at 03:20:35PM +0100, Gil Weber wrote:
> Hello,
> I am unable to communicate in RS485 since I upgrade to the latest kernel
> version (v4.9-rc7). Previously I was in 3.12.
>
> It seems to be caused by commit "0058f0871efe7b01c6f2b3046c68196ab73e96da
> tty/serial: atmel: fix RS485 half duplex with DMA".
> RX is disabled during TX, but in my case it is not re-enabled after.
> If I revert this patch, everything seems to work again.
Any reason you didn't cc: the author of that patch? :)
> Currently enable RX is done in atmel_tx_dma but it is not called if there are
> still data to transfer... problem in that it is never done later.
> So I think we should also enable RX in atmel_complete_tx_dma.
>
> I have made this fix, and it seems to work now:
>
>
> diff --git a/drivers/tty/serial/atmel_serial.c
> b/drivers/tty/serial/atmel_serial.c
> index 74e97dc..4190012 100644
> --- a/drivers/tty/serial/atmel_serial.c
> +++ b/drivers/tty/serial/atmel_serial.c
> @@ -806,6 +806,10 @@ static void atmel_complete_tx_dma(void *arg)
> */
> if (!uart_circ_empty(xmit))
> atmel_tasklet_schedule(atmel_port, &atmel_port->tasklet_tx);
> + else if (port->rs485.flags & SER_RS485_ENABLED) {
> + /* DMA done, start RX for RS485 */
> + atmel_start_rx(port);
> + }
>
> spin_unlock_irqrestore(&port->lock, flags);
>
>
>
>
> Is it correct? or maybe there is a better way to fix it?
> Thanks,
> Gil
Alexandre, any ideas?
--
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
