flat assembler
Message board for the users of flat assembler.

Index > Macroinstructions > Macro translation woes - NASM <> FASM

Author
Thread Post new topic Reply to topic
Tarkin



Joined: 27 Aug 2010
Posts: 22
Tarkin
Hello all,
I've roamed around a bit, and found pieces of whta I need to do, but have so far
been unable to fit them all together.

Basically, I am trying to convert jonesforth
to fasm syntax.

After much weeping and gnashing of teeth, I translated jonesforth.S to NASM syntax.

My trouble is, of course macros, which are different across gAS, NASM, and fasm.

Here is an example, in gAS (original), and my translated NASM eqiuvalent:

gAS:
Code:
.macro defcode name, namelen, flags=0, label
    .section .rodata
    .align 4
    .globl name_\label
name_\label :
   .int link               // link
     .set link,name_\label
      .byte \flags+\namelen // flags + length byte
      .ascii "\name"               // the name
 .align 4                // padding to next 4 byte boundary
  .globl \label
\label :
     .int code_\label       // codeword
 .text
       //.align 4
  .globl code_\label
code_\label :                       // assembler code follows
   .endm
    


NASM:
Code:
%macro defcode 4
  SECTION .rodata
  ALIGN 4
  GLOBAL name_%4
name_%4:
  dd link                   ; link
  %define link name_%4      ; update link
  db %2 + %3                ; flags + length byte
  db %1                     ; the name
  ALIGN 4                   ; padding to next 4 byte boundary
  GLOBAL %4
%4: dd code_%4              ; codeword
  SECTION .text
  ALIGN 4
  GLOBAL code_%4
code_%4:
; assembler code follows
%endmacro
    


TIA,
Tarkin[/code]
Post 29 Aug 2010, 16:32
View user's profile Send private message Reply with quote
bitRAKE



Joined: 21 Jul 2003
Posts: 3055
Location: vpcmipstrm
bitRAKE
The length can be determined from the name - not sure why it's passed as a parameter to the macro? IIRC, FASM cannot do any section combining. In FASM I imagine an array of parameters could be collected to build the .rodata section later. Alternately, the combining could be left to the linker - grabbing pieces of multiple object files.
Post 29 Aug 2010, 21:07
View user's profile Send private message Visit poster's website Reply with quote
bitRAKE



Joined: 21 Jul 2003
Posts: 3055
Location: vpcmipstrm
bitRAKE
Code:
defcodeArr equ
link equ 0

macro defcode nm,fl,lb {
  match any, defcodeArr \{ defcodeArr equ defcodeArr,nm,fl,lb \}
  match , defcodeArr \{ defcodeArr equ nm,fl,lb \}

  align 4
  label code_#lb
}
macro rodata [nm,fl,lb] {
 local temp,length
   display `nm,13,10
   name_#lb dd link
    link equ name_#lb
   db fl or length
     temp:
   db nm
       length = $-temp
     align 4
     ; NOTE: leading underscores for reserved word labels
    __#lb dd code_#lb
}
macro NEXT {
    lodsq
       jmp rax
}

; testing...

defcode "DROP",0,DROP ; drop top of stack
  pop rax
NEXT

defcode "SWAP",0,SWAP ; swap top two elements on stack
  pop rax
  pop rbx
  push rax
  push rbx
NEXT

defcode "DUP",0,DUP ; duplicate top of stack
  push qword [rsp]
NEXT

 match def, defcodeArr {
 rodata def
 }    
...this seems to almost work.
Post 29 Aug 2010, 22:57
View user's profile Send private message Visit poster's website Reply with quote
Tarkin



Joined: 27 Aug 2010
Posts: 22
Tarkin
I found it !!
I've been poring through fasm-related posts in comp.lang.forth, and I found
my answer here: http://groups.google.com/group/comp.lang.forth/msg/f7cf0479bcadeaaf?

bitRAKE:
I thought being able to update sections/segments as you went along was one of the
primary goals in the docs for fasm- though I may have misuderstood the docs...

But I am not trying to 'combine' sections, just update entries in the respective sections/segments
as I go along, preserving my 'programming context'; i.e. I don't want (and should have) to
have two monlithic sections in my source, and have to constantly scroll back and forth between
them to update related code/data in the sections. Also, this is rather elf-specific...

As for passing the length as a parameter, rather than processing it in the macro itself,
well, it was that way when I found it...

Without further ado, here is the 'defcode' macro in fasm syntax-
cheking it with FASMPRE.EXE (under wine, as I develop mostly on GNU/Linux)
shows that fasm is generating what I want...
Code:
macro defcode _name,len,flags,_label {
align 4
section '.rodata'
name_#_label:
public name_#_label
  dd link
link = name_#_label
  db flags + len
  db _name
align 4
_label:
public _label
  dd code_#_label
section '.text'
align 4
code_#_label:
public code_#_label
; assebler code follows
}
    


I got really mixed up at some point, trying to use match, macros to generate
the label names, all sorts of blind alleys...I will report back if the sections give me
any trouble.

Thanks,
Tarkin
Post 29 Aug 2010, 23:08
View user's profile Send private message Reply with quote
Display posts from previous:
Post new topic Reply to topic

Jump to:  


< Last Thread | Next Thread >
Forum Rules:
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum


Copyright © 1999-2020, Tomasz Grysztar. Also on GitHub, YouTube, Twitter.

Website powered by rwasa.