home *** CD-ROM | disk | FTP | other *** search
- Path: news.eunet.fi!fipnet!kone!jsaarinen
- Newsgroups: comp.sys.amiga.programmer
- X-NewsReader: IntuiNews 1.2b (31.7.94)
- References: <38232191@kone.fipnet.fi> <551.6602T1231T2646@ifi.uio.no> <38232238@kone.fipnet.fi> <1316.6605T735T2597@ifi.uio.no>
- From: "Jyrki Saarinen" <jsaarinen@kone.fipnet.fi>
- Date: Sat, 3 Feb 96 09:00:07 UT
- Comments: Illegal date header - new date added by quicknews
- X-Original-Date: Fri, 02 Feb 96 18:41:14
- MIME-Version: 1.0
- Content-Type: text/plain; charset=iso-8859-1
- Content-Transfer-Encoding: binary
- Subject: Re: PPC compilers
- Message-ID: <38232280@kone.fipnet.fi>
-
-
- > Well, I know on thing for sure, and that is:
- >
- > > moveq #0,d0
- > > add.l d1,d0
- >
- > ...*ALLWAYS* sets the x-flag to 0 for any value d1!
- >
- > So *YOU* allways clear the x-flag too. :))
-
- Yep.. ;)
-
- > Now, I'm thinking about something that /can/ be wrong with the addx
- > routines.
- >
- > The X-flag is not added to d2 at the correct time, I think.
- > Its delayed with 1 run throw the loop:
- >
- > So here is a suggestion:
- >
- > sub.w d1,d2 ; This routine will only affect the hi-word
- > of
- > add.l d1,d2 ; d2 and set the xflag right.
- >
- > This routine will solve the the x-flag *and* the delay problem.
- > (If you are using 16:16)
- >
- > It looked cool, I think I will try that in my routines! :)
-
- Yep, the problem is, that the decimals are delayed. So this
- will fix it? How about with two addx? What then?
-
- -- _
- a Stellar programmer _ //
- "Amiga - back for the future" \X/
-