From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 91621 invoked by alias); 21 Aug 2018 15:51:19 -0000 Mailing-List: contact libc-help-help@sourceware.org; run by ezmlm Precedence: bulk List-Subscribe: List-Post: List-Help: , Sender: libc-help-owner@sourceware.org Received: (qmail 91605 invoked by uid 89); 21 Aug 2018 15:51:18 -0000 Authentication-Results: sourceware.org; auth=none X-Spam-SWARE-Status: No, score=-1.9 required=5.0 tests=AWL,BAYES_00,SPF_HELO_PASS autolearn=ham version=3.3.2 spammy=reduced, our X-HELO: mx1.redhat.com Received: from mx3-rdu2.redhat.com (HELO mx1.redhat.com) (66.187.233.73) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Tue, 21 Aug 2018 15:51:17 +0000 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.rdu2.redhat.com [10.11.54.5]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 4932140251DD; Tue, 21 Aug 2018 15:51:16 +0000 (UTC) Received: from ovpn-117-166.ams2.redhat.com (ovpn-117-166.ams2.redhat.com [10.36.117.166]) by smtp.corp.redhat.com (Postfix) with ESMTPS id E9D1A9464B; Tue, 21 Aug 2018 15:51:12 +0000 (UTC) Message-ID: <1534866671.4893.75.camel@redhat.com> Subject: Re: Possible hang in pthread_cond_broadcast() From: Torvald Riegel To: Carlos O'Donell Cc: "Kuchta, Martin" , "libc-help@sourceware.org" Date: Tue, 21 Aug 2018 15:51:00 -0000 In-Reply-To: <6de6e911-b248-9e31-800e-2f029e4cb63f@redhat.com> References: <6de6e911-b248-9e31-800e-2f029e4cb63f@redhat.com> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit X-SW-Source: 2018-08/txt/msg00035.txt.bz2 On Wed, 2018-08-15 at 22:01 -0400, Carlos O'Donell wrote: > On 08/15/2018 06:13 PM, Kuchta, Martin wrote: > > Ideally, we'd like to have a standalone reproducible test, but our > > efforts so far to develop one have been unsuccessful. The problem > > only seems to show itself in our application very rarely (once per > > 15-24 hours) and only under very heavy load. > > Right, the ideal is a simple reproducer. Even if there's no reproducer that triggers the problem reliably, I'd like to see the code that uses the condvar (not the full application, but something reduced). I'll try to find some time to review this bug report and your analysis.