A small update in case that someone is interested. It turned out to be two issues. Both aren't omap related : The first one is that "serial8250_em485_init" in "8250_port.c" doesn't drive the RTS line if it finds that "em485" isn't NULL. rs485 settings are kept between close/open calls, so enabling rs485 again after opening doesn't drive the RTS line (em485 isn't NULL). It seems that something happening after/on the open call messes up the rs485 state. ioctl still reports the port as having rs485 enabled. The other issue is that "serial_core.c" is not setting the RTS when rs485 is enabled on uart_open. I have a pair of patches, both very simple. One resets the timer and drive the RTS line any time that "serial8250_em485_init" is called on (8250_port.c) and the other one calls "uart_port->rs485_config" on "uart_open" (serial_core.c). Both seem workarounds more than fixes. With some guidance I could try to submit something with required level of quality.-- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
