From owner-freebsd-stable@FreeBSD.ORG  Wed Feb  1 11:44:12 2006
Return-Path: <owner-freebsd-stable@FreeBSD.ORG>
X-Original-To: freebsd-stable@freebsd.org
Delivered-To: freebsd-stable@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id BDABD16A420
	for <freebsd-stable@freebsd.org>; Wed,  1 Feb 2006 11:44:12 +0000 (GMT)
	(envelope-from petefrench@ticketswitch.com)
Received: from mail.ticketswitch.com (mail.ticketswitch.com [194.200.93.188])
	by mx1.FreeBSD.org (Postfix) with ESMTP id 66EA643D53
	for <freebsd-stable@freebsd.org>; Wed,  1 Feb 2006 11:44:12 +0000 (GMT)
	(envelope-from petefrench@ticketswitch.com)
Received: from [172.16.1.6] (helo=dilbert.firstcallgroup.co.uk)
	by mail.ticketswitch.com with esmtp (Exim 4.52 (FreeBSD))
	id 1F4GPA-000EoJ-QQ
	for freebsd-stable@freebsd.org; Wed, 01 Feb 2006 11:44:08 +0000
Received: from petefrench by dilbert.firstcallgroup.co.uk with local (Exim
	4.52 (FreeBSD)) id 1F4GPA-000CBe-Du
	for freebsd-stable@freebsd.org; Wed, 01 Feb 2006 11:44:08 +0000
To: freebsd-stable@freebsd.org
Message-Id: <E1F4GPA-000CBe-Du@dilbert.firstcallgroup.co.uk>
From: Pete French <petefrench@ticketswitch.com>
Date: Wed, 01 Feb 2006 11:44:08 +0000
Subject: Restartable system call behaviour
X-BeenThere: freebsd-stable@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Production branch of FreeBSD source code <freebsd-stable.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>, 
	<mailto:freebsd-stable-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-stable>
List-Post: <mailto:freebsd-stable@freebsd.org>
List-Help: <mailto:freebsd-stable-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>,
	<mailto:freebsd-stable-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2006 11:44:12 -0000

I have a piece of coode which does some networking, in which I see read
and write calls failing with 'Interrupted system call' from time to time.

The reason it puzzles me is that I am explicitly catching every signal
that could possibly be causing this, with the SA_RESTART flag set. So
surely I should never see this ?

This only started happening when we moved away from 4.11 - it was happening
the whole time I was running on 5.4 and still happens on 6.0. I thought
that the default was for restartable system calls - I only started setting
the SA_RESTART flag explicitly in the last 24 hours in an effort to fix
this, but it is still happening.

Have I missed something fundamental about signals and system calls ? I am
about to go in and adapt my code so that it retries the system call if it
is interrupted - but I am uneasy about doing this without knowing *why*
the system calls arent't restarting automaticly.

-pcf.