> On Aug 2, 2015, at 6:37 AM, Peter Hurley <peter@xxxxxxxxxxxxxxxxxx> wrote:
>
> On Sat, Aug 1, 2015 at 11:26 AM, Joe Nall <joe@xxxxxxxx> wrote:
> Is there a mechanism to open a serial port for writing without impacting the control lines?
>
> Sorry, no.
Why would RTS be asserted when there is no reader? Does O_WRONLY have no meaning for a tty?
> open("/dev/ttyS0", O_WRONLY|O_NONBLOCK|O_NOCTTY)
>
> asserts RTS when there is no listener and I would like to avoid that.
>
> It is possible to open, set baud to B0, and close; subsequent open()'s would not
> raise RTS. However, RTS would be raised when restoring the baud rate prior to writing.
>
> What's the purpose of not raising RTS?
We have a couple of use cases.
The first is an application that receives and transmits messages on a serial port. The user interface allows transmit and receive to be enabled separately. Right now, the application starts in a disabled receive state by explicitly opening the device and then deasserting DTR/RTS. When the user want to enable reception, we raise DTR/RTS. This has a brief window where RTS is asserted incorrectly that I would like to be able to remove. There has also been some concern that manually raising RTS might impact subsequent automatic flow control. I had hoped I could rewrite the code to use separate read only and write only fds and get rid of the application fiddling with DTR/RTS.
The second is an application that statuses the control lines on software. The code is raising DTR/RTS as a side effect of opening the device. Setting the baud rate to B0 won't work in this case either.
thanks,
joe
--
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
