On Sun, 18 Aug 1996, Ervin D. Walter wrote:
> Mark Phillips writes: > > > > You may recall I posted, complaining that I had to kill gpm before I > > could get X to run. I solved the problem by changing the mouse device > > over to /dev/ttyS0 (whereas before it was /dev/cua0). Now things work > > fine! > > > > Am I right in thinking that the cua devices are now obsolete and that > > ttyS devices are the way to go? > > > > Mark. > > This is sort of right. /dev/cuaxxx are in most cases the wrong choice > because they implement kernel level locking. What was happening was > that once one process (gpm) opens /dev/cua0, the kernel does not allow > another process (X) to use the same device. The /dev/ttySxx are > specifically not designed this way. They are designed with the idea > that intelligent programs will take care of their own locking (which > gpm and X do nicely) so more than one program can share the same > devices. > > Erv > > -- Supposdly that should be true, but it seems for some reason, on my system, ttyS's also get locked and the only way I can get gpm and X to work is to run gpm and gpm -R and point X to /dev/gpmdata and call it a MouseSystems mouse. Shaya -- Shaya Potter [EMAIL PROTECTED]
