uplcom(4) problem

[Matthew Dillon]

    Ok, I *think* I fixed uplcom/ucom.  It turns out that when a TIOC 
    is sent to simply flush the rx data, the ucom driver was in fact
    halting the receive pipe entirely.

    I slipped the DragonFly_Stable tag for the ucom.c fix.  Please try it
    and tell me if everything is dandy now!

    --

    I'm working on the pppd run loop bug now.
    
					-Matt
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>