From: Ralph Ronnquist <ralph.ronnquist@gmail.com>
Date: Thu, 13 May 2021 02:25:13 +0000 (+1000)
Subject: baseline
X-Git-Url: https://git.rrq.au/?a=commitdiff_plain;h=9c582d87bca4bce0aa889cad0bd200ee85512f78;p=rrq%2Frrqforth.git

baseline
---

diff --git a/Makefile b/Makefile
index 39e68f2..1e629b9 100644
--- a/Makefile
+++ b/Makefile
@@ -2,9 +2,13 @@ BINARY = rrqforth
 
 default: rrqforth
 
+VFMT := "RRQ Forth version %h at %aI"
+VERSION := $(shell git log -1 --pretty=format:'$(VFMT)' main.fasm)
+
+main.fasm: machine.fasm stdio.fasm
+
 version: main.fasm
-	git log --pretty=format:"msg db 'RRQ Forth version %h at %aI'" $< > $@
-	echo "\ndb 10" >> $@
+	@echo "db '$(VERSION)'\ndb 10" > $@
 
 rrqforth: main.fasm | version
 	fasm $^ $@
diff --git a/machine.fasm b/machine.fasm
new file mode 100644
index 0000000..92f51a1
--- /dev/null
+++ b/machine.fasm
@@ -0,0 +1,101 @@
+;;; This file define/describes the "machine"
+;;;
+;;; Abstract Machine:
+;;; https://en.wikibooks.org/wiki/X86_Assembly/X86_Architecture
+;;; 
+;;; General Purpose Registers ( * marks those used in syscalls )
+;;;  *rax = ( -, eax = ( -, ax = ( ah, al) )) "accumulator"
+;;;   rbx = ( -, ebx = ( -, bx = ( bh, bl) )) "base"
+;;;  *rcx = ( -, ecx = ( -, cx = ( ch, cl) )) "counter"
+;;;  *rdx = ( -, edx = ( -, dx = ( dh, dl) )) "data"
+;;;   rsp = ( -, esp = ( -, sp = ( -, spl) )) "stack pointer"
+;;;   rbp = ( -, ebp = ( -, bp = ( -, bpl) )) "stack base pointer"
+;;;  *rsi = ( -, esi = ( -, si = ( -, sil) )) "source"
+;;;  *rdi = ( -, edi = ( -, di = ( -, dil) )) "destination"
+;;;  *r8
+;;;  *r9
+;;;   r10
+;;;  *r11
+;;;   r12
+;;;   r13
+;;;   r14
+;;;   r15
+;;; clobbers rdi rsi rdx rcx r8 r9 r11
+;;; rax = syscall id
+;;;
+;;; Segment Registers
+;;;   SS "Stack Segment"
+;;;   CS "Code Segment"
+;;;   DS "Data Segment"
+;;;   ES "Extra Segment"
+;;;   FS "more Extra Segment"
+;;;   GS "more more Extra Segment"
+;;;
+;;; EFLAGS Register
+;;;   0,0,0,0,0,0,0,0,0,0,ID,VIP,VIF,AC,VM,RF,
+;;;   0,NT,[IOPL,IOPL],OF,DF,IF,TF,SF,ZF,0,AF,0,PF,1,CF
+;;; 
+;;; Instruction pointer
+;;;   EIP
+;;; 
+;;; Syscall allocations
+;;; clobbers rdi rsi rdx rcx r8 r9 r11
+;;; rax = syscall id
+;;;
+;;; function calling
+
+;;; FORTH model
+;;; rsp = data stack pointer
+;;; rbp = frame pointer
+;;; rdi = frame stack
+;;; rsi = instruction pointer
+
+;;; ========================================
+;;; The next macro "moves" execution to the next FORTH instruction,
+;;; using rsi as instruction pointer.
+
+	macro next {
+	lodsq			; mov rax, qword [rsi]
+				; add rsi,8
+	jmp qword [rax]		; goto code of that FORTH word
+	}
+
+;;; ========================================
+;;; The pushr macro pushes x onto the return stack
+;;; The popr macro pops x from the return stack
+	macro pushr x {
+	sub rbp, 8
+	mov qword [rbp], x
+	}
+
+	macro popr x {
+	mov x, [rbp]
+	add rbp, 8
+	}
+
+;;; ========================================
+;;; 
+
+	previous_word = 0
+
+	;; Macro WORD starts a FORTH word definition in this code
+	macro WORD label, name, doer, flags {
+label#_tfa:
+	;; TFA
+	dq previous_word
+	;; PFA
+label#_word:
+	previous_word = label#_word
+	db flags + 0
+	db label - $ - 1
+	db name
+	;; CFA = pointer to "interpreter"
+label:
+	dq doer
+	}
+
+	;; Macro WORD_assembler begins an assembler implementation
+	macro WORD_assembler label, name, flags {
+	WORD label, name, label#_code, flags
+label#_code:
+	}
diff --git a/main.fasm b/main.fasm
index dd4f30d..25e48b4 100644
--- a/main.fasm
+++ b/main.fasm
@@ -1,20 +1,79 @@
-; This is a program
+; This is a forth interpreter
 	format elf64 executable
-	entry main
+	entry main_code
 
-	segment readable executable
-	include 'version'
-	length = $ - msg
+;;; ############################################################
+;;; The FORTH words
+	segment readable writable executable
 
-main:
-	lea rsi,[msg]		; address of message
-	mov edx,length		; length od trdting
+include 'machine.fasm'
+	
+	;; PROGRAM_VERSION is the program version string
+	WORD program_version, 'PROGRAM_VERSION', marker
+	db length
+program_version_string:	
+	include 'version'
+	length = $ - program_version_string
+	
+	;; MAIN is the program entry point
+	;; ( -- )
+	WORD_assembler main, "MAIN"
+	mov rsi,program_version_string ; address of string
+	mov edx,length		; length of string (cheating)
 	mov edi,1 		; stdout
 	mov eax,1		; sys_write
 	syscall
+	jmp terminate0_code
 
-fini:
+	;; TERMINATE0 terminates the program with code 0
+	;; ( -- )
+	WORD_assembler terminate0, 'TERMINATE0'
 	xor edi,edi
 	mov eax,60
 	syscall
+
+	;; EXIT ends a forth code defintion, returning to caller
+	;; ( -- )
+	WORD_assembler exit, 'EXIT'
+	popr rsi
+	next
 	
+	;; MARKER is a word that pushes the address after itself, then exits
+	;; ( -- p )
+	WORD_assembler marker, 'MARKER'
+	push qword rsi
+	jmp exit_code
+
+	;; MARKER@ is a word that pushes a the value after itself, then exits
+	;; ( -- v )
+	WORD_assembler marker_get, 'MARKER@'
+	push qword [rsi]
+	jmp exit_code
+
+	;; DOFORTH begins a FORTH defintion
+	WORD_assembler doforth, 'DOFORTH'
+	pushr rsi
+	lea rsi, [rax + 8]
+	next
+
+	;; LIT is a word that pushes a the value after itself, then continues
+	;; ( -- v )
+	WORD_assembler lit, 'LIT'
+	push qword [rsi]
+	add rsi, 8
+	next
+
+	;; HERE is a variable pointing to the free heap
+	WORD here, 'HERE', marker
+	dq heap_start		; initialise to first "free" data
+
+	;; WORDS is the list of words
+	WORD words, 'WORDS', marker
+	dq forth_tfa		; initialise to last forth word
+
+	;; FORTH is the last word of the VOCABULARY 
+	WORD forth, 'FORTH', marker_get
+	dq forth_tfa
+	
+heap_start:
+