From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff King Subject: Re: upgrading GIT from version 1.7.0.4 to 1.7.6.1. Date: Mon, 19 Sep 2011 18:44:49 -0400 Message-ID: <20110919224449.GC4056@sigill.intra.peff.net> References: <1316435430491-6808156.post@n2.nabble.com> <4E773A71.6020905@op5.se> <20110919190023.GC26115@sigill.intra.peff.net> <7vwrd48bq8.fsf@alter.siamese.dyndns.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Cc: Andreas Ericsson , =?utf-8?Q?Mika=C3=ABl?= , git@vger.kernel.org To: Junio C Hamano X-From: git-owner@vger.kernel.org Tue Sep 20 00:44:57 2011 Return-path: Envelope-to: gcvg-git-2@lo.gmane.org Received: from vger.kernel.org ([209.132.180.67]) by lo.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1R5ma8-000193-E7 for gcvg-git-2@lo.gmane.org; Tue, 20 Sep 2011 00:44:56 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750806Ab1ISWow (ORCPT ); Mon, 19 Sep 2011 18:44:52 -0400 Received: from 99-108-226-0.lightspeed.iplsin.sbcglobal.net ([99.108.226.0]:37774 "EHLO peff.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750698Ab1ISWov (ORCPT ); Mon, 19 Sep 2011 18:44:51 -0400 Received: (qmail 10840 invoked by uid 107); 19 Sep 2011 22:49:49 -0000 Received: from sigill.intra.peff.net (HELO sigill.intra.peff.net) (10.0.0.7) (smtp-auth username relayok, mechanism cram-md5) by peff.net (qpsmtpd/0.84) with ESMTPA; Mon, 19 Sep 2011 18:49:49 -0400 Received: by sigill.intra.peff.net (sSMTP sendmail emulation); Mon, 19 Sep 2011 18:44:49 -0400 Content-Disposition: inline In-Reply-To: <7vwrd48bq8.fsf@alter.siamese.dyndns.org> Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: On Mon, Sep 19, 2011 at 02:26:55PM -0700, Junio C Hamano wrote: > Jeff King writes: > > > [2] I suspect a similar thing happened with turning on packed refs > > (around the v1.4.4 era?), but I didn't dig around for details. > > Also when bisecting down to really ancient versions of git, you may get > complaints from the configuration reader. E.g. > > $ rungit v1.0.0 ls-files > fatal: bad config file line 78 in .git/config > $ sed -n 78p .git/config > [remote "git-gui"] Ugh, yeah. There are similar problems with some specific config options (one I run into a lot these days is that "pager.foo" used to complain if the value was non-boolean, but now you can put in an arbitrary command). But those only bite you if you start using the new config options (which you obviously are not doing until after you upgrade). But your example will bite anyone because git generates remote config like that by default. Anyway, most of that is only applicable if you are crossing several major version boundaries, or are using ancient versions of git. I'm pretty sure that v1.7.0.x -> v1.7.6.x is pretty safe. -Peff