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
Date:	Tue, 28 Jan 2003 08:44:51 +0200
Message-Id:	<200301280644.h0S6ipG21454@hki-web-1.sunpoint.net>
To:	cygwin AT cygwin DOT com
Subject:	bug: select() for sockets returned by accept() returns always ready for write (sel.always_ready 1)
From:	Risto Virkkala <risto DOT virkkala AT sunpoint DOT net>

Hi,

I am using non-blocking sockets in a small server software. Select() seems to indicate that sockets returned by accept() are always ready for write. This causes 99% cpu usage since the socket gets always selected as writeable and after that write returns EAGAIN and again we go to the select().

After looking select.cc fhandler.cc fhandler_socket.cc I think that the reason could be that in fhandler_socket::accept the socket is not set to CONNECTED state and in fhandler_socket::select_write sets write_ready if socket is in unconnected state.

(uname -a
CYGWIN_NT-5.0 O1400031 1.3.19(0.71/3/2) 2003-01-23 21:31 i686 unknown unknown Cygwin)

- Risto
-
Sunpoint tiedottaa:

Puhelintuki numerossa 0600-10 900 1,83E/min+pvm

Tutustu myös Etätukeen tästä
http://www.sunpoint.net/SunAds/click.htm?mode=footer&id=70&jump=http%3A%2F%2Fwww.pilotvastauspalvelu.fi%2Fdefault.asp%3Fsivu%3D3

--
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 -