home *** CD-ROM | disk | FTP | other *** search
- Path: nntp.teleport.com!sschaem
- From: sschaem@teleport.com (Stephan Schaem)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: C2P
- Date: 7 Jan 1996 05:39:19 GMT
- Organization: Teleport - Portland's Public Access (503) 220-1016
- Distribution: world
- Message-ID: <4cnma7$jjo@maureen.teleport.com>
- References: <4cj57f$7qe@irz210.inf.tu-dresden.de> <4cjl2o$4im@maureen.teleport.com> <38231877@kone.fipnet.fi>
- NNTP-Posting-Host: julie.teleport.com
- X-Newsreader: TIN [version 1.2 PL2]
-
- Jyrki Saarinen (jsaarinen@kone.fipnet.fi) wrote:
-
- : > I personaly use the 5 pass methode.
- : > swap word
- : > swap byte
- : > swap nyble
- : > swap bitpair
- : > swap bit
-
- : Oh, word swapping is the 5th pass. What is your opinion,
- : I think 030/50 should be able to do c2p free. Perhaps
- : three passes with CPU and the last one with the blitter.
-
- You can do 13 inst betwen chip write on a 50mhz 030, right?
- Thats 52 inst. Alomst yea. (4bit )
- Actually my table methode use 50inst to do the 5pass, including
- load/store.32 inst without the last pass, but on a 030 you
- cant do fast read with a chip write.
- basicly:
- mw (a0)+,d0
- ml (a1,d0*4),d1
- mw (a0)+,d0
- add.l (a2,d0*4),d1
- .. 3 more time (Do pass 1,2,4)
- pass 16+pass8 is 30 inst ... easy to make it overlap with
- the 4 chipram write. (You write the 4 previous converted
- long, this mean you have to unroll the loop and have 2 set of
- register) But there is enought register & the all thing fit in the
- cache. So the blitter is not needed...
-
-
- Stephan
-