home *** CD-ROM | disk | FTP | other *** search
- > From: Ted Pederson <tjp@twg.com>
- > Date: Fri, 25 Mar 1994 10:49:56 -0500
-
- Hi Ted,
-
- > I have 2 questions on the 1.1 winsock spec regarding FD_WRITE.
- > First, if a winsock client calls WSAAsyncSelect a second time, will
- > an FD_WRITE message be sent to the client (assuming readiness for
- > writing)?
-
- It's supposed to happen that way (assuming you've asked for FD_WRITE
- notification in your WSAAsyncSelect() call, of course).
-
- Some reports have indicated that calling WSAAsyncSelect() either
- 'too often' and/or 'in a superfluous manner', _seems_ to cause
- problems with some WinSock.DLLs.
-
- > Second, why are send() and sentto() listed as re-enabling functions
- > for FD_WRITE (on p. 90)? Each time send() is called the winsock
- > library is not supposed to post an FD_WRITE message, right? Only
- > after a send() or sentto() fails,
-
- ...with WSAEWOULDBLOCK only...
-
- > and then buffer space becomes available, was my understanding.
-
- Correct.
-
- > Just wanted to make sure...
-
- I think you might be just being a bit overly literal on this...
- I think it's adequately explained that the send()/sendto() re-enablers
- are qualified by the WSAEWOULDBLOCK/free buffer space cycle.
-
- But, then again, I'm just trying to make sure I understand it too!
-
- Still hazy after all these years,
-
- BobN
- ________________________________________________________________
- Bob Natale | American Computer | 301-258-9850 [v]
- Director | 209 Perry Pkwy | 301-921-0434 [f]
- Network Mgmt Products | Gaithersburg MD 20877 | natale@acec.com
-
-
-