Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!ucbvax!UOTTAWA.BITNET!451061
From: 451061@UOTTAWA.BITNET (Valentin Pepelea)
Newsgroups: comp.sys.amiga.tech
Subject: Re: Safe storage of task context
Message-ID: <8907270423.AA11556@jade.berkeley.edu>
Date: 27 Jul 89 04:15:11 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Lines: 24

Steve Faiwisewski <intern@lilink.uucp> writes in Message-ID: <791@lilink.UUCP>

> According to the RKM, upon a context switch all the CPU's registers
> get saved on the task's (user) stack.  Unfortunately, the RKM does not
> specify the order of the saved registers.  Using MetaScope to look at
> a process' stack, I found that it conflicted with MetaScope's
> Registers display for that process.  So what gives?  I must be missing
> something.

To be sure, you should check Markus Wandel's disassembly of the Exec, but I
would assume that the registers get stored onto the stack with the

      movem.l  A0-A7/D0-D7,-(sp)

instruction. To be specific, look at the movem definition in the 680x0 User's
Manual to find out the order in which the registers are stored.

Valentin
_________________________________________________________________________
"An  operating  system  without         Name:   Valentin Pepelea
 virtual memory is an operating         Phonet: (613) 231-7476 (New!)
 system without virtue."                Bitnet: 451061@Uottawa.bitnet
                                        Usenet: Use cunyvm.cuny.edu gate
         - Ancient Inca Proverb         Planet: 451061@acadvm1.UOttawa.CA