From owner-freebsd-current Thu May 28 16:07:11 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA19492 for freebsd-current-outgoing; Thu, 28 May 1998 16:07:11 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from ns1.yes.no (ns1.yes.no [195.119.24.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA19482 for ; Thu, 28 May 1998 16:07:06 -0700 (PDT) (envelope-from eivind@bitbox.follo.net) Received: from bitbox.follo.net (bitbox.follo.net [195.204.143.218]) by ns1.yes.no (8.8.7/8.8.7) with ESMTP id XAA14692; Thu, 28 May 1998 23:06:52 GMT Received: (from eivind@localhost) by bitbox.follo.net (8.8.8/8.8.6) id BAA23613; Fri, 29 May 1998 01:06:40 +0200 (MET DST) Message-ID: <19980529010639.54255@follo.net> Date: Fri, 29 May 1998 01:06:39 +0200 From: Eivind Eklund To: joelh@gnu.org Cc: rnordier@nordier.com, current@FreeBSD.ORG Subject: Re: Fix for undefined "__error" and discussion of shared object versioning References: <199805271551.RAA11565@ceia.nordier.com> <199805281829.NAA01253@detlev.UUCP> <19980528212713.63593@follo.net> <199805282015.PAA01671@detlev.UUCP> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89.1i In-Reply-To: <199805282015.PAA01671@detlev.UUCP>; from Joel Ray Holveck on Thu, May 28, 1998 at 03:15:44PM -0500 Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Thu, May 28, 1998 at 03:15:44PM -0500, Joel Ray Holveck wrote: > >> My own concern would be the amount of code in third-party programs > >> that uses gccisms. I guess I don't see why we're looking to change. > > Well, here are a couple of points: > > * License > > I'll assume that TenDRA and XANDF are both BSD-license? BSD-style, yes. Not an exact BSD license. > > * General quality of system (GCC is written under the paradigm 'learn > > writing compilers as we go') > > I can't comment on that one, not having examined the source code to > the various compilers. But are the others written by professional > compiler writers? As far as I know: Yes. They've been contracted for the development, and I've seen there come a lot of papers from their research. > > * Possibilities for exploiting the cross-CPU nature of XANDF > > How are XANDF's cross-cpu capabilities more powerful than gcc's? XANDF is an intermediate format designed for being compiled to different processors at install-time. This means that we can do only the last phase of a cross-compilation, instead of doing the entire compilation, which we'd have to if we used gcc. > > * Better error checking/control > > How do you mean? Look at the TenDRA docs. It let you work with the defintion of APIs, instead of with whatever gunk ends up in the header files. It also can do a lot of checking against the ANSI/ISO standard. > > * Choice (by being able to be compiled with more than one compiler) > > I'm definately in agreement on this point. I don't like to be boxed > in to *any* software, be it M$, GNU, or BSD. > > Is the OS compilable with pcc? Not AFAIK. > > They might not be enough to be persuative, but they are at least in > > favour. > > Okay. I'm not familiar enough with the compilers' specs, features, and > limitations to make a judgement. So for now I'll gather information > before I cast a vote. So am I. I don't know if we should switch - but the only thing I really see in favour of GCC on a long-term basis is the size of the userbase. The actual design is OK, and the license (for FreeBSD purposes) suck. Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message