home *** CD-ROM | disk | FTP | other *** search
- Path: tomtec.abg.sub.org!judas
- From: judas@tomtec.abg.sub.org (Th.Huber)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: 680X0 -> PPC translator?
- Distribution: world
- Message-ID: <judas.0j7o@tomtec.abg.sub.org>
- References: <volker.0g32@vb.franken.de> <315C198B.49C2@netvision.net.il> <volker.0g5w@vb.franken.de> <1996Apr2.230841.8275@scala.scala.com> <volker.0g8b@vb.franken.de>
- Date: 5 Apr 96 21:19:45 MET
-
- In article <volker.0g8b@vb.franken.de> volker@vb.franken.de (Volker Barthelmann) writes:
- >Dave Haynie (dave.haynie@scala.com) wrote:
-
- >But that's not what we were talking about! We were talking about a static
- >and complete translation of 68k to PPC before it is executed.
-
- Which is defenitly NOT possible. You`re not able to decide, what part
- of a 68k-programm is used for data and what`s used for c0de.
- As the c0de will differ in size after the translation, you run
- into big big trouble.
-
- >: The AmigaOS and the 680x0 don't support self-modifying code.
-
- But real dirty c00der can do it anyway .. =)
-
- >I was not talking of self-modifying code there. Only the address of the
- >code is modified and copied around, but a static translator would have
- >to know where the jump goes to in order to decide what's code.
-
- Yep, that`s your prob. There`s lot of data within a code-hunk.
- A translater won`t be able to decide, whether a longword is used
- as a counter, or as a table for jumps, or even c0de itself.
-
-
- >: So basically, stop the arguments, I'm offering an existance proof.
- >
- >Not for static translators.
-
- A proof must be a 100%-translation, not kind of a "neuronal network",
- that makes the best guess.....
-