From owner-cvs-all  Tue Sep 25 22: 8:12 2001
Delivered-To: cvs-all@freebsd.org
Received: from mailsrv.otenet.gr (mailsrv.otenet.gr [195.170.0.5])
	by hub.freebsd.org (Postfix) with ESMTP
	id C83E437B413; Tue, 25 Sep 2001 22:08:02 -0700 (PDT)
Received: from hades.hell.gr (patr530-a028.otenet.gr [212.205.215.28])
	by mailsrv.otenet.gr (8.11.5/8.11.5) with ESMTP id f8Q57NN04593;
	Wed, 26 Sep 2001 08:07:28 +0300 (EEST)
Received: (from charon@localhost)
	by hades.hell.gr (8.11.6/8.11.6) id f8Q2uSf22152;
	Wed, 26 Sep 2001 05:56:29 +0300 (EEST)
	(envelope-from charon@labs.gr)
Date: Wed, 26 Sep 2001 05:56:26 +0300
From: Giorgos Keramidas <charon@labs.gr>
To: Matt Dillon <dillon@earth.backplane.com>
Cc: Ruslan Ermilov <ru@FreeBSD.ORG>, cvs-committers@FreeBSD.ORG,
	cvs-all@FreeBSD.ORG
Subject: Re: cvs commit: src/bin/cat cat.1 cat.c
Message-ID: <20010926055625.A21817@hades.hell.gr>
References: <200109150039.f8F0dFZ41705@freefall.freebsd.org> <20010925211639.C57333@sunbay.com> <200109251909.f8PJ9eJ03402@earth.backplane.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <200109251909.f8PJ9eJ03402@earth.backplane.com>
User-Agent: Mutt/1.3.22.1i
X-GPG-Fingerprint: C1EB 0653 DB8B A557 3829  00F9 D60F 941A 3186 03B6
X-URL: http://labs.gr/~charon/
Sender: owner-cvs-all@FreeBSD.ORG
Precedence: bulk
List-ID: <cvs-all.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20cvs-all>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20cvs-all>
X-Loop: FreeBSD.ORG

Matt Dillon <dillon@earth.backplane.com> wrote:
> 
> :
> :How about testing with stat(2) that an argument is a socket
> :thus eliminating the need to open(2) sockets and connect(2)
> :to non-sockets?
> 
>     I don't see any particular reason why.  An open() failure is as fast
>     or faster then stat().  If it fails and we get the appropriate errno,
>     we try a connect().  Either connect() is supported or it isn't.  If it
>     is a case of a missing file we don't try the connect() because we get
>     a different errno.  I see no particular need to stat the path, no ill
>     effects will occur if the connect() fails.

And there is always a possibility of a race condition between a stat() and an
open().  So relying on stat() to control what open() and/or connect() will do
is kind of dangerous in my book :-/

-giorgos


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message