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-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,2cb54c327a92d5fd X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 2003-04-28 08:44:17 PST Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!cyclone.bc.net!skynet.be!skynet.be!213.51.129.3.MISMATCH!newshub1.home.nl!home.nl!news.telebyte.nl!nntpfeed-01.ops.asmr-01.energis-idc.net!nntpfeed-01.ops.asmr-01.energis-idc.net!newsfeed.kabelfoon.nl!195.129.110.21.MISMATCH!bnewsfeed00.bru.ops.eu.uu.net!bnewsifeed00.bru.ops.eu.uu.net!bnewspost00.bru.ops.eu.uu.net!emea.uu.net!read.news.de.uu.net!not-for-mail Sender: Andrew Lynch Message-ID: <3ead4c4a$0$4091$4d4ebb8e@read.news.de.uu.net> From: ALynch@slb.com Subject: Re: Which GCC compiler for Linux-GNAT Newsgroups: comp.lang.ada References: <3EAAC338.1030109@wanadoo.fr> User-Agent: tin/1.4.5-20010409 ("One More Nightmare") (UNIX) (Linux/2.4.18-14 (i686)) Date: 28 Apr 2003 15:44:10 GMT NNTP-Posting-Host: zerberus.cci.de X-Trace: 1051544650 read.news.de.uu.net 4091 193.103.165.100 X-Complaints-To: abuse@de.uu.net Xref: archiver1.google.com comp.lang.ada:36688 Date: 2003-04-28T15:44:10+00:00 List-Id: Preben Randhol wrote: > Bernard WEISSER wrote: [...] >> With the arrival of the new LINUX Slackware 9.0 release, it appeared to >> me that now a full bunch of compilers is provided (and ready to use), >> namely : C, C++, F77 and GNAT 3.15. All these front-ends are based on >> the unique GCC 3.2.2 compiler version. Using this new GNAT version, I >> succeded to compile a rather complex Ada application, but it turned out >> that the binary run quickly ended with a CONSTRAINT_ERROR. Same here with Red Hat 8.0. It appears to be the math functions that are causing the trouble. There are at least two problem reports in the bug database at http://gcc.gnu.org/ (PR6911 and PR9033). The first one originated in the Debian bug tracking system and the discussion there points to a more general problem with the support for assembly code includes. > The Gnat-gcc 3.x series are not up to speed yet, so wait with upgrading > until they have finished porting them over to the Gcc 3.x backend. Unfortunately the upgrading is done for many of us, due to the featuritis that has infested the Linux distributors. Just like MS they seem to think that everyone always wants the latest and greatest. We are now using RH 8.0 because that is what the client wants - a few months ago their baseline was still RH 7.1. "Luckily" this project is in C++, but there are some modules which will be reimplementations of existing Ada code. I have several test harnesses which I have been using to compare output from the original Ada code with that of the C++ implementation. None of them work on the new system, so I have to keep switching between old and new environment. What also worries me is the potential damage done to GNAT's reputation. The major distributions advertise the fact that they have an Ada compiler, unfortunately for the current releases it is almost useless. A number of potential Ada newbies could be in for a disappointment. Andrew.