Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 30 Nov 2000 15:06:11 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 30 Nov 2000 15:06:01 -0500 Received: from 4dyn129.com21.casema.net ([212.64.95.129]:38925 "HELO home.ds9a.nl") by vger.kernel.org with SMTP id ; Thu, 30 Nov 2000 14:51:32 -0500 Date: Thu, 30 Nov 2000 20:20:43 +0100 From: bert hubert To: linux-kernel@vger.kernel.org Cc: aeb@cwi.nl Subject: SO_RCVTIMEO and SO_SNDTIMEO implemented? manpage says no, kernel yes? Message-ID: <20001130202042.A19106@home.ds9a.nl> Mail-Followup-To: linux-kernel@vger.kernel.org, aeb@cwi.nl Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, I was reading the works of Stevens, and saw the very neat SO_RCVTIMEO and SO_SNDTIMEO features mentioned. Socket.7 told me that these socket options aren't implemented: SO_RCVTIMEO and SO_SNDTIMEO Specify the sending or receiving timeouts until reporting an error. They are fixed to a protocol specific setting in Linux and cannot be read or written. Their functionality can be emulated using alarm(2) or setitimer(2). However, reading 2.4.0test10 net/core/sock.c appears to indicate that the kernel at least does some of the work. Does anybody know if these socket options work as they should under Linux, and if so, which versions? I might even whip up a better entry for the manpage if given enough data. Regards, bert hubert -- PowerDNS Versatile DNS Services Trilab The Technology People 'SYN! .. SYN|ACK! .. ACK!' - the mating call of the internet - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/