home *** CD-ROM | disk | FTP | other *** search
- Path: informatik.tu-muenchen.de!fischerj
- From: fischerj@informatik.tu-muenchen.de (Juergen "Rally" Fischer)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: TMapping again!
- Date: 7 Mar 1996 16:00:56 GMT
- Organization: Technische Universitaet Muenchen, Germany
- Distribution: world
- Message-ID: <4hn17o$j3t@sunsystem5.informatik.tu-muenchen.de>
- References: <38232763@kone.fipnet.fi> <4hb5n1$r52@maureen.teleport.com> <38232824@kone.fipnet.fi>
- NNTP-Posting-Host: hphalle5.informatik.tu-muenchen.de
- Originator: fischerj@hphalle5.informatik.tu-muenchen.de
-
-
- In article <38232824@kone.fipnet.fi>, "Jyrki Saarinen" <jsaarinen@kone.fipnet.fi> writes:
- |>
- |> > : Hmm, I am not sure if add.l d0,d1 is as fast add.l #$12345678,d1
- |> > : on the 68030. Propably not.
- |> >
- |> > Its not, I knew that it is on the 040 and I assumed the same on the
- |> > 030... but imediate cost you extra cycle.
- |>
- |> It seemed that at least on the 68020 add.w #imm,dn was 4 cycles
- |> vs. 2 cycles register add, and add.l #imm,dn was 6 cycles.
- |> Assuming in cache ofcourse..
-
- yep, confirm :)
- but (I already mentioned I think) add.w #imm,An is 6 cycles.
- use lea d(An),An instead to get 4 cycles Adressregister add.
-
- don't know about and/or immedeate but assume they are same like add.
- hmm 6 cycles to do and.l #$f0f0f0f0,Dn. any fake idea ? :)
-
- |>
- |> -- _
- |> a Stellar programmer _ //
- |> "Amiga - back for the future" \X/
- ------------------------------------------------------------------------
- fischerj@Informatik.TU-Muenchen.DE (Juergen "Rally" Fischer) =:)
-
-