SB-PRAM simulator

(C) 1994 by Michael Bosch (hirbli@cs.uni-sb.de)
and Stefan Franziskus (stefran@cs.uni-sb.de)

Permission to use, copy, modify, and distribute this software and its
documentation for any purpose and without fee is hereby granted, provided
that the above copyright notice appear in all copies.



compile: ./configure and make

First do a './configure --help' to see all options:
If you want GNU--readline support just add a '--with-readline' to the 
command line. If you want to use lazy--malloc you must configure
with a '--with-pram-file'. '---enable-autonice' seems not to work 
properly for all machines. All other options are only usefull to
develop low--level routines for the SB-PRAM hardware.

You can specify the install--path with --prefix=PREFIX. For example
if PREFIX is /usr/local, the pramsim is installed in /usr/local/bin.

After that just do a 'make pramsim' and a 'make install'.


With this version you need the SB-PRAM coff-loader which is added to
the package. Make shure that the loader is installed in your 
search-path. You also neeed a .ldrc in the path from which you run
the pramsim. In this file you can specify defaults for stack, heap
and so on. The structure of this file depends on the used compiler
and startup--code. Please don't change the structure of this file !!!
If you want to write your own compiler and startup--code you may need
further information about this file. Feel free to contact me under
stefran@cs.uni-sb.de. 
If you want to have a default loaderrc, you can specify path and name
of this file with the environment variable "SBP_LOADERRC".


help: pramsim --help / info-file

Frequently used command line options can also be specified in a file
'.pramsimrc', which should be in the directory from where you run the
pramsim.

Example files for '.ldrc' and '.pramsimrc' are enclosed.

pramsim has been tested on:

i486-*-linux1.1.51
sparc-sun-sunos4.1.3
mips-sgi-irix5.2
sparc-sun-solaris2.3

pramsim has only been tested with gcc.

We rely on the compiler supporting:
  ANSI-C
  + optimisations:
     constant folding
     if(0) elimination
     automatic register allocation
  + extensions:
     inline functions
     64 bit wide integers (mul, mlhs and mlhu simulation)
  + float = 32bit ieee754
