> Adding a flag to the socket structure, set one way or the other depending
> on which emulation opened the socket is probably okay to start, but need
> to think about what happens when passing the file to another process via
> exec, etc.
 
Umm... If the emulation changes, we need to review all open sockets on
execve(), and modify their flags according to the new emulation. Is this
reasonable? I got no idea of the overhead this would introduce. 

The other question is what field to use? A free bit in the so_options,
or a new field? I would use a bit in so_options, defined like this:

#define SO_BSDASYNCIO 0x1000      /* use BSD style asynchronous I/O */

Opinions?

-- 
Emmanuel Dreyfus.  
p99dreyf@criens.u-psud.fr