From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 11683 invoked by alias); 5 Jun 2004 13:33:29 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 11676 invoked from network); 5 Jun 2004 13:33:29 -0000 Received: from unknown (HELO mx1.redhat.com) (66.187.233.31) by sourceware.org with SMTP; 5 Jun 2004 13:33:29 -0000 Received: from int-mx1.corp.redhat.com (int-mx1.corp.redhat.com [172.16.52.254]) by mx1.redhat.com (8.12.10/8.12.10) with ESMTP id i55DXSi7016626 for ; Sat, 5 Jun 2004 09:33:29 -0400 Received: from localhost.redhat.com (porkchop.devel.redhat.com [172.16.58.2]) by int-mx1.corp.redhat.com (8.11.6/8.11.6) with ESMTP id i55DXH003565; Sat, 5 Jun 2004 09:33:18 -0400 Received: from gnu.org (localhost [127.0.0.1]) by localhost.redhat.com (Postfix) with ESMTP id 09A152B9F; Sat, 5 Jun 2004 09:33:03 -0400 (EDT) Message-ID: <40C1CB8E.7080108@gnu.org> Date: Sat, 05 Jun 2004 13:33:00 -0000 From: Andrew Cagney User-Agent: Mozilla/5.0 (X11; U; NetBSD macppc; en-GB; rv:1.4.1) Gecko/20040217 MIME-Version: 1.0 To: Eli Zaretskii Cc: mec.gnu@mindspring.com, brobecker@gnat.com, hilfingr@gnat.com, gdb-patches@sources.redhat.com Subject: Re: [PATCH]: Updates to Ada sources, part 1 (longish) References: <20040603051228.E269F4B104@berman.michael-chastain.com> <2914-Fri04Jun2004144147+0300-eliz@gnu.org> <40C0C01D.7080504@gnu.org> <9743-Sat05Jun2004130832+0300-eliz@gnu.org> In-Reply-To: <9743-Sat05Jun2004130832+0300-eliz@gnu.org> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-SW-Source: 2004-06/txt/msg00098.txt.bz2 >>> Date: Fri, 04 Jun 2004 14:31:57 -0400 >>> From: Andrew Cagney >> >>>> > >>>> > That'd be fine with me, but I'd also suggest to have a pointer to >>>> > ChangeLog.GNAT in gdb/ChangeLog, right where the import of Ada changes >>>> > is recorded. Something like "See ChangeLog.GNAT for details of the >>>> > changes." >> >>> >>> Please don't do this. ChangeLog.XXX is good for branches (as with even >>> readline/ChangeLog.gdb) but not for what is ment to be the mainline. > > > I agree with the principle, but I don't think it is such a sacred one > that would justify asking the GNAT people to invest such a large > effort. > > We need the information to be there, and the suggested compromises > achieve that with a reasonable effort. We've on a number of occasions seen large through to extreemly large merges (HP comes to mind) and on each occason the contributor, unprompted, ensure that a correct ChangeLog was included. >>> The ChangeLog entry should provide a summary of what was added/changed >>> at this point in time - stuff like listing the new functions and summary >>> of changed functions. Can we do that? > > > How many man-hours would you say is reasonable for such an effort? > 1? 10? 100? 1000? Where, if at all, do we draw the line? We're talking an hour or perhaphs two, and something we're _all_ expected to do willingly. The way to do this is to ignore the history and just look at the final diff. A bit of sed'n'sort will in a matter of minutes give the functions added/deleted bit, leaving just terse verbage of the other functions changed. Can we do that? >>> Much of the stuff in that GNAT ChangeLog will no longer be relevant. > > > I understand that the problem is to translate the irrelevant suff into > something that is relevant. I suggested that at first, but the GNAT > people tell that it will take a lot of work. > > >>> I'm told Diego took rougly a day to prepare his tree-SSA ChangeLog entry. > > > Who is Diego and how is the tree-SSA stuff relevant to our case? Diego did a very very large (read 4 years of work x many people) merge of GCC's mainline. When committed it included a very large (but terse) ChangeLog entry. Andrew