From mboxrd@z Thu Jan 1 00:00:00 1970 From: Johannes Sixt Subject: Re: [PATCH 1/3] usage: refactor die-recursion checks Date: Tue, 16 Apr 2013 16:13:56 +0200 Message-ID: <516D5CA4.7000500@viscovery.net> References: <20130415230651.GA16670@sigill.intra.peff.net> <20130415230802.GA11267@sigill.intra.peff.net> <20130416004228.GA14995@sigill.intra.peff.net> <20130416025024.GA20932@sigill.intra.peff.net> <516CFB56.1090805@viscovery.net> <20130416130154.GA17976@sigill.intra.peff.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: Brandon Casey , "git@vger.kernel.org" To: Jeff King X-From: git-owner@vger.kernel.org Tue Apr 16 16:14:08 2013 Return-path: Envelope-to: gcvg-git-2@plane.gmane.org Received: from vger.kernel.org ([209.132.180.67]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1US6e8-0001yO-DR for gcvg-git-2@plane.gmane.org; Tue, 16 Apr 2013 16:14:08 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S935486Ab3DPOOD (ORCPT ); Tue, 16 Apr 2013 10:14:03 -0400 Received: from so.liwest.at ([212.33.55.14]:36525 "EHLO so.liwest.at" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S935444Ab3DPOOA (ORCPT ); Tue, 16 Apr 2013 10:14:00 -0400 Received: from [81.10.228.254] (helo=theia.linz.viscovery) by so.liwest.at with esmtpa (Exim 4.77) (envelope-from ) id 1US6dx-0005Yw-59; Tue, 16 Apr 2013 16:13:57 +0200 Received: from [192.168.1.95] (J6T.linz.viscovery [192.168.1.95]) by theia.linz.viscovery (Postfix) with ESMTP id DB5D21660F; Tue, 16 Apr 2013 16:13:56 +0200 (CEST) User-Agent: Mozilla/5.0 (Windows NT 5.1; rv:17.0) Gecko/20130328 Thunderbird/17.0.5 In-Reply-To: <20130416130154.GA17976@sigill.intra.peff.net> X-Enigmail-Version: 1.5.1 X-Spam-Score: -1.0 (-) Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: Am 4/16/2013 15:01, schrieb Jeff King: > On Tue, Apr 16, 2013 at 09:18:46AM +0200, Johannes Sixt wrote: > >>> Yeah, that seems sane; my biggest worry was that it would create >>> headaches for Windows folks, who would have to emulate pthread_key. But >>> it seems like we already added support in 9ba604a. >> >> pthread_key is not a problem, but pthread_once is. It's certainly >> solvable, but do we really have to? > > I'm not clear on what you are suggesting. That we protect only the main > thread from recursion, or that we drop the check entirely? Or that we > implement thread-local storage for this case without using pthread_once? Anything(*) that does not require pthread_once. A pthread_once implementation on Windows would be tricky and voluminous and and on top of it very likely to be done differently for gcc and MSVC. I don't like to go there if we can avoid it. (*) That includes doing nothing, but does not include ripping out the recursion check, as it protects us from crashes. -- Hannes