Re: 1-pass Assembler Design

kochenbu@khe.scn.de (Andreas Kochenburger)
19 Aug 1998 16:16:41 -0400

          From comp.compilers

Related articles
1-pass Assembler Design marvl@fusionmm.com (Marv Luse) (1998-08-13)
Re: 1-pass Assembler Design leichter@smarts.com (Jerry Leichter) (1998-08-16)
Re: 1-pass Assembler Design gkasten@auco.com (Glenn Kasten) (1998-08-16)
Re: 1-pass Assembler Design nr@labrador.cs.virginia.edu (Norman Ramsey) (1998-08-16)
Re: 1-pass Assembler Design ian@cygnus.com (1998-08-16)
Re: 1-pass Assembler Design adrian@dcs.rhbnc.ac.uk (1998-08-17)
Re: 1-pass Assembler Design meissner@cygnus.com (Michael Meissner) (1998-08-17)
Re: 1-pass Assembler Design kochenbu@khe.scn.de (1998-08-19)
Re: 1-pass Assembler Design jcrens@magicnet.net (Jack W. Crenshaw) (1998-08-19)
Re: 1-pass Assembler Design nr@labrador.cs.virginia.edu (Norman Ramsey) (1998-08-20)
| List of all articles for this month |

From: kochenbu@khe.scn.de (Andreas Kochenburger)
Newsgroups: comp.compilers
Date: 19 Aug 1998 16:16:41 -0400
Organization: Compilers Central
References: 98-08-096
Keywords: assembler, comment

If you can restrict your assembler to structured control flows you can
introduce a control flow stack CFS of label addresses. In principle


begin \ compile a jump, push here to CFS
...
until \ compile a conditional jump, pop old address from CFS and
calculate distance


Structures like IF..ELSE..ENDIF or counted loops all work the same
way. Unstructured jumps to LABELs are possible if you put in some
restrictions by checking the CFS depth.


The fastest way towards productivity however is not to reinvent the
wheel but to use Forth. See the FAQ in comp.lang.forth.


Andreas
[A standard trick is to chain the list of backpatches through the
references to the symbol in the output file. -John]
--


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.