From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham autolearn_force=no version=3.4.4 X-Google-Thread: 103376,32bee71a8464bfc2 X-Google-Attributes: gid103376,public X-Google-Language: ENGLISH,ASCII-7-bit Path: g2news1.google.com!news2.google.com!proxad.net!newsfeed1.ip.tiscali.net!tiscali!transit1.news.tiscali.nl!dreader2.news.tiscali.nl!not-for-mail Newsgroups: comp.lang.ada Subject: Re: A bug in gnat/gcc 3.3.3? References: <1104854323.582074.155390@z14g2000cwz.googlegroups.com> <1105114956.835835.209620@f14g2000cwb.googlegroups.com> From: Ludovic Brenta Date: Fri, 07 Jan 2005 19:13:25 +0100 Message-ID: <87y8f5ksze.fsf@insalien.org> User-Agent: Gnus/5.1007 (Gnus v5.10.7) Emacs/21.3 (gnu/linux) Cancel-Lock: sha1:IuYtECgMdHmHo92IRooNt6X6LDQ= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Organization: Tiscali bv NNTP-Posting-Date: 07 Jan 2005 19:13:25 CET NNTP-Posting-Host: 83.134.239.204 X-Trace: 1105121605 dreader2.news.tiscali.nl 44109 83.134.239.204:38536 X-Complaints-To: abuse@tiscali.nl Xref: g2news1.google.com comp.lang.ada:7540 Date: 2005-01-07T19:13:25+01:00 List-Id: Adrian Hoe writes: > Why 3.15p? Why 3.3.x? Why 3.4? Why PolyORB? Why this? Why that? Why > can't we just have one single standard? If gnat is going to advance > into 3.3.x, 3.4 and beyond, then make it work. Why still keep 3.15p? > And you can't bootstrap 3.15p from 3.3.3, for example. Switch to Debian. Seriously. It provides the single standard that you are seeking. The standard is documented in the Debian Ada Policy, and I have no trouble enforcing it, since I'm the one doing all the work :) Why keep 3.15p? Because 3.4 has a serious regression (tasking on powerpc), and because of ASIS and GLADE. Now, Martin has done some effort porting them both to GCC 3.4, and I am willing to consider using them as time permits. This will be done after Sarge is released, because it is something of a big bang, as all packages must be recompiled. > I wonder what's happening at gnu/gcc. Why hadn't they tested the > compiler before rolling into public? I can't even find my problem in > "Known Issues" in gnat/gcc documentation. Yes, they did test gnat-3.15p like they did every "p" release. But they did not catch your particular bug. Is this bug known in the Debian BTS, or in GCC's bugzilla? > The chaotic compiler standards will break the spirit of a newbie! That is true. Newbies should use a binary distribution that supports Ada well. > I think the Ada community should take the initiative to standardize > builds if gnu/gcc cannot manage this alone. This initiative will > make Ada more viable in serious development and as well as making > learning Ada more comfortable and hassle-free. I do exactly that on Debian, on three architectures. My goal when I set out doing this was precisely to attract more newbies to Ada. Debian is a distribution that encourages this kind of initiative, as do others like Gentoo and FreeBSD. Each of these distros has an Ada maintainer who decides on the policy so that all binary packages are compatible. The reason why I chose Debian was because it is primarily a binary distribution, whereas Gentoo and FreeBSD are primarily source-based. Choose whichever you prefer. > We all have a day time job with limited evening hours. If everyone > would contribute an hour a day, there will be a lot of man-hour per > day. I am willing to contriubte. Great. Would you move to Debian and help package ASIS and GLADE for gnat-3.4? Or PolyORB? Help is wanted and welcome. You could also try to provide packages for SuSE, but I'm not sure that the SuSE people would let you take charge :) -- Ludovic Brenta.