From owner-freebsd-chat Fri May 29 18:19:17 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id SAA18458 for freebsd-chat-outgoing; Fri, 29 May 1998 18:19:17 -0700 (PDT) (envelope-from owner-freebsd-chat@FreeBSD.ORG) Received: from smtp04.primenet.com (root@smtp04.primenet.com [206.165.6.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA18448 for ; Fri, 29 May 1998 18:19:05 -0700 (PDT) (envelope-from tlambert@usr04.primenet.com) Received: (from daemon@localhost) by smtp04.primenet.com (8.8.8/8.8.8) id PAA13979; Fri, 29 May 1998 15:15:38 -0700 (MST) Received: from usr04.primenet.com(206.165.6.204) via SMTP by smtp04.primenet.com, id smtpd026640; Fri May 29 14:34:57 1998 Received: (from tlambert@localhost) by usr04.primenet.com (8.8.5/8.8.5) id OAA15936; Fri, 29 May 1998 14:34:55 -0700 (MST) From: Terry Lambert Message-Id: <199805292134.OAA15936@usr04.primenet.com> Subject: Re: Fix for undefined "__error" and discussion of shared object versioning To: alk@pobox.com Date: Fri, 29 May 1998 21:34:54 +0000 (GMT) Cc: chat@FreeBSD.ORG, nate@mt.sri.com In-Reply-To: <199805282249.RAA03682@pobox.com> from "Tony Kimball" at May 28, 98 05:49:19 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Various points: > > 1 Non-portable != broken. Bah humbug. That's *exactly* the definition of "broken". Code that fails to compile under GCC could be termed "non-portable to GCC" otherwise. > 2 I'm amazed to see this coming from Nate "Java" Williams! > Gcc is the *platform*. Gcc is what makes code portable, > not ANSI, which is merely an idea, and not a platform. So compile FreeBSD up for SPARC. I'd like a copy... > 3 People use gccisms because then they don't have to write asm, > which is much *less* portable. Using gcc'isms to get assembly features is not architecturally transferrable. It is better to seperate machine dependent and independent code into seperate modules. This may include the use of machine-variant header files to get this, but that's irrelevent. > 4 Are you planning on fixing them? Egad, the number of ports! If the patches are submitted back, I'm sure if the code is actively maintained, the maintainers will be interested in increased code portability. I know I would be. The larger the audience for the code, the better. > > Better/faster/less buggy compiler with a much less restrictive Copyright > > seems like a win overall to me. > > Remains to be seen, as far as I am concerned. Various points: > > 1 TenDRA should be compared to GCC 2.8.1, not 2.7.2.1. Agreed. > 2 GPL should not be a restriction in practice for anyone > except someone who wants to retain source for a commercialization > of added-value back-end code generation under FreeBSD, since there > are plenty of source-to-source compilers suitable as front-ends to > GCC. Since the amount of money to be made doing this is a negative > number, it's not a restriction in practice. This should be allowed. GPL disallows this. Note that private implementation languages would all have to be built as translaters for your plan to be viable, and that's not a reasonable restriction. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message