From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on dcvr.yhbt.net X-Spam-Level: X-Spam-ASN: AS15169 209.85.128.0/17 X-Spam-Status: No, score=-2.4 required=3.0 tests=AWL,BAYES_00,FREEMAIL_FROM, RCVD_IN_DNSWL_LOW shortcircuit=no autolearn=unavailable version=3.3.2 X-Original-To: cmogstored-public@bogomips.org Received: from mail-la0-f47.google.com (mail-la0-f47.google.com [209.85.215.47]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by dcvr.yhbt.net (Postfix) with ESMTPS id B90791F6A8 for ; Tue, 10 Mar 2015 19:23:16 +0000 (UTC) Received: by labgd6 with SMTP id gd6so4020495lab.6 for ; Tue, 10 Mar 2015 12:23:15 -0700 (PDT) X-Received: by 10.112.185.101 with SMTP id fb5mr32407733lbc.12.1426015394879; Tue, 10 Mar 2015 12:23:14 -0700 (PDT) Received: from localhost ([91.225.202.255]) by mx.google.com with ESMTPSA id j6sm291945laf.5.2015.03.10.12.23.13 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 10 Mar 2015 12:23:14 -0700 (PDT) Sender: Mykola Golub Date: Tue, 10 Mar 2015 21:23:12 +0200 From: Mykola Golub To: Eric Wong Cc: cmogstored-public@bogomips.org Subject: Re: FreeBSD: sleeper using ppoll does not sleep forever Message-ID: <20150310192311.GA5161@gmail.com> References: <20150309151851.GC2195@gmail.com> <20150309203543.GA18791@dcvr.yhbt.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150309203543.GA18791@dcvr.yhbt.net> User-Agent: Mutt/1.5.23 (2014-03-12) List-Id: On Mon, Mar 09, 2015 at 08:35:43PM +0000, Eric Wong wrote: > Thanks, there's at least two problems in cmogstored here: > > 1) The assertion could trigger on any OS due to the lack of retry > on EINTR. > > 2) ppoll is non-POSIX, and is a cancellation point in GNU libc, > but perhaps not the FreeBSD libc based on your bug report. > We'll switch to pselect which POSIX as a cancellation point. > > I suggest FreeBSD developers make ppoll a thread cancellation > point for GNU/Linux compatibility. > > So two patches coming in reply to this email... Cool! Works here. Thanks. -- Mykola Golub