From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp-out3.iol.cz ([194.228.2.91]) by speech.braille.uwo.ca with esmtp (Exim 3.36 #1 (Debian)) id 1GfQFc-0005oa-00 for ; Wed, 01 Nov 2006 19:16:08 -0500 Received: from antivir3.iol.cz (antivir3.iplanet.iol.cz [192.168.30.206]) by smtp-out3.iol.cz (Postfix) with ESMTP id 0257B6486C for ; Thu, 2 Nov 2006 01:15:38 +0100 (CET) Received: from localhost (antivir3.iol.cz [127.0.0.1]) by antivir3.iol.cz (Postfix) with ESMTP id EAF9042000A for ; Thu, 2 Nov 2006 01:15:37 +0100 (CET) Received: from smtp-out3.iol.cz (smtp-out-3.iplanet.iol.cz [192.168.30.28]) by antivir3.iol.cz (Postfix) with ESMTP id D8D9A420006 for ; Thu, 2 Nov 2006 01:15:37 +0100 (CET) Received: from [192.168.1.36] (77.93.broadband2.iol.cz [83.208.93.77]) by smtp-out3.iol.cz (Postfix) with ESMTP id 0609557E10 for ; Thu, 2 Nov 2006 01:15:36 +0100 (CET) Subject: Re: modprobe speakupmain hangs From: Hynek Hanke To: "Speakup is a screen review system for Linux." In-Reply-To: References: <1162392434.3281.50.camel@chopin> Content-Type: text/plain; charset=UTF-8 Date: Thu, 02 Nov 2006 01:14:26 +0100 Message-Id: <1162426466.3304.7.camel@chopin> Mime-Version: 1.0 X-Mailer: Evolution 2.6.2 X-Virus-Scanned: amavisd-new at iol.cz Content-Transfer-Encoding: quoted-printable X-BeenThere: speakup@braille.uwo.ca X-Mailman-Version: 2.1.9 Precedence: list Reply-To: "Speakup is a screen review system for Linux." List-Id: "Speakup is a screen review system for Linux." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 02 Nov 2006 00:16:08 -0000 luke p=C3=AD=C5=A1e v St 01. 11. 2006 v 16:53 -0600: > I reported this problem 3 or 4 weeks ago. Gregory Nowak was able to=20 > confirm that it was happening. In that case, I'm sorry for the duplicity. > While this is untested in regards to modprobe, it would be better to us= e: > kill -9 $! Yes, you are completely right that the quick hack I sent has possible side effects. Actually, it is pretty broken with the sleeps and all :) However, I think the method you described is not guaranteed to work, because in the time of the crash, modprobe is running a child process and you would risk that this child process ends up being a zombie and block the thing. I didn't test the exact behavior, but I wouldn't rely on it. I agree that in other cases what you propose would be a much cleaner solution. With regards, Hynek Hanke