NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/sys/amiga/programmer/5302

Internet Message Format | 1996-08-05 | 3.3 KB

Path: informatik.tu-muenchen.de!fischerj
From: [e-mail] (Juergen "Rally" Fischer)
Newsgroups: comp.sys.amiga.programmer
Subject: Re: need help on A1200/030 board programming
Date: 12 Mar 1996 18:00:21 GMT
Organization: Technische Universitaet Muenchen, Germany
Distribution: world
Message-ID: <4i4e3l$[e-mail]>
References: <4gb9ao$[e-mail]>   <[e-mail]> <[e-mail]>  <[e-mail]> <4h4ek7$[e-mail]> <[e-mail]>
NNTP-Posting-Host: hphalle10.informatik.tu-muenchen.de
Originator: [e-mail]


In article <[e-mail]>, [e-mail] (Damon Davies) writes:
|> Organization: PlaNet (NZ) Manawatu, Palmerston Nth, New Zealand [phone]
|> Lines: 41
|> Message-ID: <[e-mail]>
|> References: <4gb9ao$[e-mail]>
|>      <[e-mail]> <[e-mail]>
|>     <[e-mail]> <4h4ek7$[e-mail]>
|> NNTP-Posting-Host: kurg.manawatu.gen.nz
|> Mime-Version: 1.0
|> Content-Type: text/plain; charset=iso-8859-1
|> Content-Transfer-Encoding: 8bit
|> X-Newsreader: THOR 2.21 (Amiga;TCP/IP)
|>
|> Juergen writes:
|>
|> >In article <[e-mail]>, [e-mail] (Damon
|> >Davies) writes:
|> >|> >You dont have to "recognise" FAST-ram, just dont force stuff
|> >|> >into CHIP-ram which does not need to be there, ie. only
|> >|> >gfx data, samples, display memory etc.
|> >|>
|> >|> >--                               _
|> >|> >a Stellar programmer          _ //
|> >|> >"Amiga - back for the future" \X/
|> >|>
|> >|> Just an ellaboration:
|> >|>
|> >|> This is how you do it in assembler "Section TheCode,code" will put the
|> >|> code  in fast ram preferably, but in chip if there is no fast."Section
|> >|> TheCode,code_f" will put the code in fast only! and fail if there is none.
|>
|> >Forcing code to fast is imho nonsense.
|>
|> >I only use alloc(fast) one time:
|>
|> >  p=alloc(fast)
|> >  if fail (p=alloc(chip); chip=true)
|>
|> >  so my code will know if it has to copy to chip if I use a 0-cpu-pass
|> >  routine.
|>
|> >can't see any other reason for using the memf_fast flag...
|>
|> mmm, you should learn too read Juergen, what I was pointing out is that
|> there was an assembly directive that put code in fast PREFERABLY and in
|> chip OTHERWISE.

don't know what you said, but I refered to (look up)
"TheCode,code_f will put the code in fast only!".

A directive that will "put code in fast PREFERABLY and in chip OTHERWISE"
is not needed, this is system default.

|> You can do the same thing with the exec Alloc() and AllocVec() commands,
|> instead of specifying MEMF_CHIP or MEMF_FAST in the flags, you may put
|> MEMF_ANY and it will do as I describe above. So the code you describe is a
      ^--- i.e. 0.

|> bit silly really, if you now how to use exec.... :^|

why silly. what would "now how to use exec" do in the case I described ?
imho mine is most transparen, shortest, best way :)

|>
|> Damon.
|>
|>
------------------------------------------------------------------------
   [e-mail] (Juergen "Rally" Fischer)   =:)