Path: csiph.com!usenet.pasdenom.info!aioe.org!.POSTED!not-for-mail From: Bill Richards Newsgroups: comp.lang.forth Subject: Re: jonesforth: taking up the challenge Date: Wed, 12 Jun 2013 15:05:43 +0000 (UTC) Organization: Aioe.org NNTP Server Lines: 29 Message-ID: References: <51ae2e6d$0$26867$e4fe514c@dreader37.news.xs4all.nl> <2013Jun11.165956@mips.complang.tuwien.ac.at> <51b8831b$0$6073$e4fe514c@dreader36.news.xs4all.nl> NNTP-Posting-Host: /AW4nOTmvgbPCaVUBGuY2w.user.speranza.aioe.org X-Complaints-To: abuse@aioe.org X-Notice: Filtered by postfilter v. 0.8.2 Xref: csiph.com comp.lang.forth:23493 On 2013-06-12, Albert van der Horst wrote: > In article , > Andrew Haley wrote: >>Bill Richards wrote: >>> >>> 32 bit Linux as it was first presented is probably still the best choice >>> although I think most people prefer the human-looking Intel syntax over the >>> machine readable "gas" syntax. >> >>Eh? Heavens, no! And BTW, it's AT&T UNIX syntax; nothing to do with >>GNU as, which implements it for compatibility. > > Intel deviates from the convention in assemblers with its destination > source order in operands. There is no such convention. And as you would expect a correct and useful assembler exposes the ISA it was designed to run on. Intel's encoding is reflected in the choice of op dest, src. Why anyone would intentionally bass-ackwards an assembler is hard to figure. For once Microsoft seems to have done the right thing as well ;-) > It is the first time I hear some one refer to that as human. If you want to be isolated from the chip you're working on then by all means, the AT&T/gnu "compatability" syntax helps quite a bit. Otherwise Intel ought to know what they're doing when it comes to writing assemblers and assembly language for their chips don't you think? Bill