From owner-freebsd-current Wed Jun 3 01:16:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA26438 for freebsd-current-outgoing; Wed, 3 Jun 1998 01:16:43 -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 BAA26429 for ; Wed, 3 Jun 1998 01:16:40 -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 IAA19940; Wed, 3 Jun 1998 08:16:22 GMT Received: (from eivind@localhost) by bitbox.follo.net (8.8.8/8.8.6) id KAA29184; Wed, 3 Jun 1998 10:15:57 +0200 (MET DST) Message-ID: <19980603101552.35895@follo.net> Date: Wed, 3 Jun 1998 10:15:52 +0200 From: Eivind Eklund To: joelh@gnu.org, tlambert@primenet.com Cc: rnordier@nordier.com, current@FreeBSD.ORG Subject: Re: Fix for undefined "__error" and discussion of shared object versioning References: <199805292120.OAA14978@usr04.primenet.com> <199805300312.WAA02058@detlev.UUCP> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89.1i In-Reply-To: <199805300312.WAA02058@detlev.UUCP>; from Joel Ray Holveck on Fri, May 29, 1998 at 10:12:28PM -0500 Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Fri, May 29, 1998 at 10:12:28PM -0500, Joel Ray Holveck wrote: > >>> * Possibilities for exploiting the cross-CPU nature of XANDF > >> How are XANDF's cross-cpu capabilities more powerful than gcc's? > > You can distribute "binaries" and localize them to an architecture at > > install time. > > This means you can distribute commercial code that will run on x86, > > Alpha, MIPS, PPC, 68k, VAX, SPARC, etc., etc.. > > Does it have problems with endianness, et al? That is, if a program, > at compile time, needs to know its endianness (or another > architecture-dependant detail), does it still work? Yes. You use if() instead of #ifdef, and TenDRA do dead code elmination at install time. BTW: I just found out that TenDRA _do_ support __asm(). This is supported to a limited degree in the C frontend (it can only send a pure string to the backend, which will typically insert it in the asm output). I'm not certain if this is enough for our needs. > >>> * Better error checking/control > >> How do you mean? > > Full mapping of the error checking and warning space. GCC only maps > > the parts that they thought were important, and then it's done pretty > > haphazardly. > > Mapping, you mean, to diagnostics? Yes. Another neat part of this is that TenDRA indicate what you're in violation of by ISO section - always nice when you have to hit some Linuxer over the head with his violation ;-) Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message