optimize print_string, fix boot and kernel entry

config.s

@@ -1,5 +1,5 @@
-NUMSEG equ 1 ; TODO build system should calculate this value
+NUMSEG equ 4; TODO build system should calculate this value
 
 ; might be good to move one of these into the high memory area
-K_ADDR    equ 0x500  ; kernel load address
+KERNEL_SEG equ 0x50
 STACK_SEG  equ 0x1050

src/boot.s

@@ -5,7 +5,7 @@
 [org 0x7C00]
 [bits 16]
 
-mov ax, K_ADDR >> 4
+mov ax, KERNEL_SEG
 push ax ; setup for ds
 mov es, ax
 mov gs, ax
@@ -13,7 +13,7 @@ mov fs, ax
 
 xor cx, cx
 mov ds, cx
-mov word [0x21*4], 2 ; see kernel_entry
+mov word [0x21*4], 2 ; see kjmp in kernel.s
 mov word [0x21*4+2], ax
 
 mov ax, 0x200 | NUMSEG
@@ -34,7 +34,7 @@ push ds
 push ax
 retf
 
-; TODO we have lots of space for init code here
+; TODO filesystem and reserve space for mbr partition table
 
 times 510 - ($-$$) db 0
 dw 0xAA55

src/int21.s

@@ -45,7 +45,13 @@ tmp: dw 0
 retcode: dw 0
 verify: db 0
 
-rdin_echo:; ah=01h
+; AH = 01h
+; read a character from stdin and print it to stdout
+; inputs:
+; 	none
+; outputs:
+; 	AL: character read from stdin
+rdin_echo:
 	push ax
 	mov ah, 8
 	int 0x21
@@ -133,11 +139,11 @@ setint:
 	shl ax, 2
 	mov di, ax
 	xor ax, ax
-	push ds
-	mov ds, ax
-	mov word [di], dx
-	mov word [di+2], ds
-	pop ds
+	push es
+	mov es, ax
+	mov word es:[di], dx
+	mov word es:[di+2], ds
+	pop es
 	popa
 	ret
 

src/kernel.s

@@ -1,6 +1,10 @@
 %include "config.s"
 
-kernel_entry: ; MUST be a short jump due to loader config
+kjmp: ; MUST be a short jump due to loader config
+	jmp kernel_entry
+%include "int21.s"
+
+kernel_entry:
 	mov si, hello_string
 	mov bl, 0x70
 	call print_string
@@ -8,5 +12,4 @@ kernel_entry: ; MUST be a short jump due to loader config
 
 hello_string: db "hello world!", 13, 10, '$'
 
-%include "int21.s"
 %include "vga.s"

src/vga.s

@@ -5,11 +5,7 @@
 ; outputs:
 ; 	none
 print_string:
-	push ax
-	push bx
-	push cx
-	push dx
-	push bp
+	pusha
 	push es
 
 	; get cursor position in DX
@@ -20,27 +16,22 @@ print_string:
 	; calculate string length by iterating over it until we reach '$'
 	; this sucks but we have to do it because the bios expects to be passed the string length directly
 	xor cx, cx
-	push si
+	mov bp, si
+	jmp .start_loop
 .size_loop:
 	inc cx
 	inc si
+.start_loop:
 	cmp byte [si], '$'
 	jnz .size_loop
-	pop si
-	mov bp, si
 
 	; print string and update cursor position
 	mov ax, ds
 	mov es, ax
-	mov ah, 0x13
-	mov al, 0x01
+	mov ax, 0x1301
 	mov bh, 0x00
 	int 0x10
 
 	pop es
-	pop bp
-	pop dx
-	pop cx
-	pop bx
-	pop ax
+	popa
 	ret