Eventually I'd like to add some intelligence to the disassembler, so it can automatically find and mark more data tables, for example by looking for LDA $xxxx,y type instructions, and things like vector tables. Currently my code tracer won't find these so could miss a lot of valid code. Steve From: Marko Mäkelä <msmakela@gmail.com> To: cbm-hackers@musoftware.de Sent: Thursday, May 18, 2017 4:05 PM Subject: Re: CBM-Transfer 1.02 On Wed, May 17, 2017 at 06:50:45PM +0000, Steve Gray wrote: >What many people don't realize is it includes a full-featured >interactive symbolic 6502 disassembler. The latest release is out and I >wanted to let everyone know it now supports code tracing / flow >tracing. This will "run" the 6502 code using entry points you define >and follows all jmp, jsr, and branches to mark the parts that are code. >Anything not marked is "data" and is listed for you. This sounds much like d65 which I implemented back in 1994. OK, I now see that the last update to that was almost ten years ago, at Cameron Kaiser’s web page: http://www.floodgap.com/retrotech/xa/#dxa >It also comes with pre-defined "platforms" for most commodore machines >and also a special platform for identifying unknown code. If I remember correctly, d65 allowed you to specify the 6502 dialect (documented opcodes only, some variants with some undocumented opcodes, and CMOS 6502 opcodes). But as far as I remember, it did not allow you to specify which external addresses are code and which ones are definitely not (for example, so that it would detect that $20 $d3 $ff must be data; you would not typically have jsr $ffd3 in a Commodore program, but instead, jsr $ffd2).    Marko    Message was sent through the cbm-hackers mailing list Message was sent through the cbm-hackers mailing listReceived on 2017-05-19 19:00:02
Archive generated by hypermail 2.2.0.