Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932093AbVJMRFu (ORCPT ); Thu, 13 Oct 2005 13:05:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932111AbVJMRFu (ORCPT ); Thu, 13 Oct 2005 13:05:50 -0400 Received: from mail.tv-sign.ru ([213.234.233.51]:38122 "EHLO several.ru") by vger.kernel.org with ESMTP id S932093AbVJMRFu (ORCPT ); Thu, 13 Oct 2005 13:05:50 -0400 Message-ID: <434E96FA.48CBB6E7@tv-sign.ru> Date: Thu, 13 Oct 2005 21:18:50 +0400 From: Oleg Nesterov X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.20 i686) X-Accept-Language: en MIME-Version: 1.0 To: Michael Kerrisk Cc: dan@debian.org, roland@redhat.com, mingo@elte.hu, linux-kernel@vger.kernel.org, michael.kerrisk@gmx.net Subject: Re: Process with many NPTL threads terminates slowly on core dump signal References: <434D48FA.FD0439AA@tv-sign.ru> <18436.1129192898@www29.gmx.net> Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 743 Lines: 19 Michael Kerrisk wrote: > > Thanks. I've applied it to 2.6.14-rc4: this patch does fix the > specific behaviour that my program demonstrates. > > What remains to be solved? The problem remains that all threads spin with TIF_SIGPENDING flag after the coredumping thread sets SIGNAL_GROUP_EXIT in do_coredump() and before it kills them in zap_threads(). If one of these threads is SCHED_FIFO we have a deadlock. Also, the coredumping thread could be preempted, placed in ->expired array ... Oleg. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/