From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 17030 invoked by alias); 27 Feb 2014 14:38:30 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Received: (qmail 17020 invoked by uid 89); 27 Feb 2014 14:38:29 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.8 required=5.0 tests=AWL,BAYES_00,RP_MATCHES_RCVD,SPF_HELO_PASS,SPF_PASS autolearn=ham version=3.3.2 X-HELO: mx1.redhat.com Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Thu, 27 Feb 2014 14:38:28 +0000 Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id s1REcLph014833 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 27 Feb 2014 09:38:22 -0500 Received: from [127.0.0.1] (ovpn01.gateway.prod.ext.ams2.redhat.com [10.39.146.11]) by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with ESMTP id s1REcJDQ029290; Thu, 27 Feb 2014 09:38:20 -0500 Message-ID: <530F4DDB.3010803@redhat.com> Date: Thu, 27 Feb 2014 14:38:00 -0000 From: Pedro Alves User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7 MIME-Version: 1.0 To: Pedro Alves CC: Hui Zhu , gdb-patches ml , "Gustavo, Luis" Subject: Re: [PATCH] Fix PR 12702 - gdb can hang waiting for thread group leader (gdbserver) References: <51F01C13.3080402@redhat.com> <52E12075.3040601@redhat.com> In-Reply-To: <52E12075.3040601@redhat.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-SW-Source: 2014-02/txt/msg00830.txt.bz2 On 01/23/2014 02:00 PM, Pedro Alves wrote: > This is main gripe with this patch. I'd really like to avoid > bringing in more of this broken use of waitpid(PID) into gdbserver > (I realize most of this is copied from GDB), and this duplication > of low level wait status handling. I think we can do > better in gdbserver. There's really no need for wait_for_sigstop > to wait for each LWP in turn. I think that with some > refactoring, we can make it reuse linux_wait_for_event, and > only end up with having to handle pending statuses in one place. > > So I've been working on this most of this week, and I got this > almost done but I'm afraid I'll need to give attention to other > work, so I'm not sure I'll be able to be back at this before > February. Anyway, I'll post a WIP series in response to this email. I've now rebased this on current mainline, wrote ChangeLogs, etc. and pushed it. https://sourceware.org/ml/gdb-patches/2014-02/msg00826.html > In addition, it goes a step forward and starts adding support for > TARGET_WAITKIND_NO_RESUMED to the RSP as well, which is needed for > fully fixing no-unwaited-for-left.exp. I didn't get to finish that one yet, so I've left it out for now. -- Pedro Alves