[zeromq-dev] patch: non-blocking tcp accept

[Martin Sustrik]

>> Anyway, it seems that checking for both errors everywhere we are checking
>> for one of the two at the moment would yield a sane result.
> 
> I believe that to be an acceptable solution (and common-place).

Ok, fixed in SVN rev. 1406

Still there are some strange but benign behaviour to be expected. Say 
QNX defines error codes from "accept" function this way:

EAGAIN
     Insufficient resources to create the new socket.
EWOULDBLOCK
     The socket is marked nonblocking and no connections are present to 
be accepted.

Martin