From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from conssluserg-04.nifty.com (conssluserg-04.nifty.com [210.131.2.83]) by sourceware.org (Postfix) with ESMTPS id 6240D39F1485 for ; Fri, 5 Feb 2021 15:18:28 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org 6240D39F1485 Received: from Express5800-S70 (x067108.dynamic.ppp.asahi-net.or.jp [122.249.67.108]) (authenticated) by conssluserg-04.nifty.com with ESMTP id 115FI6C5010634 for ; Sat, 6 Feb 2021 00:18:06 +0900 DKIM-Filter: OpenDKIM Filter v2.10.3 conssluserg-04.nifty.com 115FI6C5010634 X-Nifty-SrcIP: [122.249.67.108] Date: Sat, 6 Feb 2021 00:18:09 +0900 From: Takashi Yano To: cygwin@cygwin.com Subject: Re: Cygwin GDB won't start or hangs at startup Message-Id: <20210206001809.f2d27fddcf3cfec75e348d71@nifty.ne.jp> In-Reply-To: References: X-Mailer: Sylpheed 3.7.0 (GTK+ 2.24.30; i686-pc-mingw32) Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Spam-Status: No, score=-4.6 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, NICE_REPLY_A, RCVD_IN_DNSWL_NONE, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org X-BeenThere: cygwin@cygwin.com X-Mailman-Version: 2.1.29 Precedence: list List-Id: General Cygwin discussions and problem reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 05 Feb 2021 15:18:32 -0000 On Fri, 5 Feb 2021 10:07:46 +0100 Marco Atzeri wrote: > On 05.02.2021 08:55, Daniel Röhrborn via Cygwin wrote: > >> On 04.02.2021 13:13, Daniel Röhrborn via Cygwin wrote: > > > > > Hi Marco, > > it seems stacked and when blindly typing "quit" it responds with exiting. So it seems like the output is invisible. > > Tested it with Win7_32bit, GDB version 9.2-1 with Cygwin/bin in PATH (since the beginning). > > > > Regards > > Daniel > > I can not fully replicate on my W10, > > but I notice that running from CMD there is a small difference > between the working instance > > > Marco@LAPTOP-82F08ILC /proc/3072/fd > i686 $ ls -l > total 0 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 0 -> /dev/cons1 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 1 -> /dev/cons1 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 10 -> pipe:[21474850180] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 2 -> /dev/cons1 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 3 -> pipe:[8589948292] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 4 -> pipe:[8589948292] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 5 -> pipe:[12884915588] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 6 -> pipe:[12884915588] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 7 -> pipe:[17179882884] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 8 -> pipe:[17179882884] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:59 9 -> pipe:[21474850180] > > and a not working one > > Marco@LAPTOP-82F08ILC /proc/3055/fd > i686 $ ls -l > total 0 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 0 -> /dev/cons0 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 1 -> pipe:[4294988268] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 10 -> pipe:[17179890156] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 11 -> pipe:[21474857452] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 12 -> pipe:[21474857452] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 2 -> pipe:[4294988268] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 3 -> pipe:[8589955564] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 4 -> pipe:[8589955564] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 5 -> /dev/cons0 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 6 -> /dev/cons0 > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 7 -> pipe:[12884922860] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 8 -> pipe:[12884922860] > lrwxrwxrwx 1 Marco Kein 0 Feb 5 09:57 9 -> pipe:[17179890156] > > it is like the file descriptor of outputs are misaligned, > and I see two more of them > > Maybe Takashi or Jon can provide hints on the matter. I looked into this problem and found this issue causes if iconv cannot be found in the search path. Installing libiconv package should solve the issue, I guess. It seems that the GDB package should set a dependency on libiconv. -- Takashi Yano