Mailing-List:	contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe:	<mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive:	<http://sources.redhat.com/ml/cygwin/>
List-Post:	<mailto:cygwin AT cygwin DOT com>
List-Help:	<mailto:cygwin-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender:	cygwin-owner AT cygwin DOT com
Mail-Followup-To:	cygwin AT cygwin DOT com
Delivered-To:	mailing list cygwin AT cygwin DOT com
Message-Id:	<200302251718.h1PHI4N00913@head-cfa.cfa.harvard.edu>
To:	cygwin AT cygwin DOT com
Subject:	Re: connect() not interrupted by sigalrm?
cc:	eric AT head-cfa DOT cfa DOT harvard DOT edu
References:	<200302241757 DOT h1OHvQN16044 AT head-cfa DOT cfa DOT harvard DOT edu>
Date:	Tue, 25 Feb 2003 12:18:04 -0500
From:	Eric Mandel <eric AT head-cfa DOT cfa DOT harvard DOT edu>

Dear Corinna,

Thanks for the quick clarification on plans for interruptible connect().

> Btw., switching to non-blocking for the connect/select pair isn't
> actually difficult so I wonder what your problem is.

Cf. Stevens "Unix Network Programming", section 15.3:

    "There are many portability problems with nonblocking connnects
    that we mention in the examples that follow."

and, after a lengthy discussion of portability issues:

     "Unfortunately, nonblocking connects are one of the most nonportable
     areas of network programming."

Since my code must run on several flavors of Unix, I hesitate to use
this technique. I suppose I could use it solely for Cygwin and retain
the more portable alarm()/connect() technique for the other platforms,
and that is what I am considering now.

Regards,

Eric


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

- Raw text -