X-Authentication-Warning: kendall.sfbr.org: jeffw set sender to jeffw AT darwin DOT sfbr DOT org using -f Date: Fri, 2 Mar 2001 08:20:21 -0600 From: JT Williams To: djgpp-workers AT delorie DOT com Subject: Re: djasm patch #5 Message-ID: <20010302082021.C9404@kendall.sfbr.org> Mail-Followup-To: djgpp-workers AT delorie DOT com References: <20010301154709 DOT B8889 AT kendall DOT sfbr DOT org> <2593-Fri02Mar2001121352+0200-eliz AT is DOT elta DOT co DOT il> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <2593-Fri02Mar2001121352+0200-eliz@is.elta.co.il>; from eliz@is.elta.co.il on Fri, Mar 02, 2001 at 12:13:53PM +0200 Reply-To: djgpp-workers AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk -: > In this patch, the current double-precision shift instructions -: > are removed, and `sh[lr]d' is implemented to be consistent with -: > other mnemonics. I don't think this breaks any djasm code -: > currently in CVS; I couldn't find any `*.asm' files that used -: > the double-precision `shld' or `shrd' instructions. -: -: Is it possible to add a compatibility option which would revert -: sh[lr]d to their old meaning? DJGPP sources are not the only users of -: djasm, although that's what djasm was originally invented for. I -: think we should make a good-faith effort not to break -: back-compatibility without leaving some escape hatch. DJ, can you recall if you implemented the double-precision shift in djasm because you actually used it in a specific application? My *impression* from studying the source is that when you implemented sh[lr]d, you realized there was a name conflict and so you didn't go any further. You also didn't worry about it because you had already implemented the instruction set you needed. But that's just my hypothetical reconstruction of events, and it may have little to do with reality [:-/