Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-developers-owner AT sourceware DOT cygnus DOT com Delivered-To: mailing list cygwin-developers AT sourceware DOT cygnus DOT com To: cygwin-developers AT sourceware DOT cygnus DOT com Subject: Re: Yet another hanging issue. References: <20000314103313 DOT A15910 AT cygnus DOT com> <20000314124712 DOT A19647 AT cygnus DOT com> Mime-Version: 1.0 (generated by tm-edit 7.106) Content-Type: text/plain; charset=US-ASCII From: Kazuhiro Fujieda Date: 16 Mar 2000 03:29:52 +0900 In-Reply-To: Chris Faylor's message of Tue, 14 Mar 2000 12:47:12 -0500 Message-ID: Lines: 28 X-Mailer: Gnus v5.3/Emacs 19.34 >>> On Tue, 14 Mar 2000 12:47:12 -0500 >>> Chris Faylor said: > >In an old snapshot such as cygwin1-19990115.dll, typing `^Z' > >makes bash return to its prompt and doesn't abort notepad. > > Ok. I restored this behavior. It wasn't as hard as I thought. The latest fixes stop bash consuming a cpu, but typing `^Z' can't wake up its prompt. I tried tracking down this and found the following problem, but I couldn't fix it. In that case, SIGTSTP caused by bash wakes the "sig" thread in the win32 process creating the notepad process. This process can't properly handle the signal because sync_proc_subproc has been released by proc_terminate() in spawn_guts(). In the concrete, proc_subproc() can't clean waiting threads so the "main" thread in the process can't start to execute the signal handler. Is my understandings right? ____ | AIST Kazuhiro Fujieda | HOKURIKU School of Information Science o_/ 1990 Japan Advanced Institute of Science and Technology