From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265661AbUBPQxh (ORCPT ); Mon, 16 Feb 2004 11:53:37 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265700AbUBPQxh (ORCPT ); Mon, 16 Feb 2004 11:53:37 -0500 Received: from mail.shareable.org ([81.29.64.88]:17284 "EHLO mail.shareable.org") by vger.kernel.org with ESMTP id S265661AbUBPQxg (ORCPT ); Mon, 16 Feb 2004 11:53:36 -0500 Date: Mon, 16 Feb 2004 16:53:29 +0000 From: Jamie Lokier To: Rusty Russell Cc: Christophe Saout , LKML , pavel@suse.cz Subject: Re: kthread, signals and PF_FREEZE (suspend) Message-ID: <20040216165329.GB17323@mail.shareable.org> References: <1076890731.5525.31.camel@leto.cs.pocnet.net> <20040216034251.0912E2C0F8@lists.samba.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040216034251.0912E2C0F8@lists.samba.org> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Rusty Russell wrote: > > That means that signal_pending() will return true for that process which > > will make kthread stop the thread. > > Yes, the way they are currently coded. I had assumed that spurious > signals do not occur. Yowch. Does suspend mean this warning in futex_wait is wrong? /* A spurious wakeup should never happen. */ WARN_ON(!signal_pending(current)); return -EINTR; -- Jamie