diff --git a/stages/hex0/helloworld.h1 b/stages/hex0/helloworld.h1
new file mode 100644
index 0000000..b014837
--- /dev/null
+++ b/stages/hex0/helloworld.h1
@@ -0,0 +1,90 @@
+; ELF header
+; e_ident: u128
+; Magic number: 0x7f454c46 = "\x7fELF"
+7f 45 4c 46
+; EI_CLASS: u8 = ELFCLASS64
+02
+; EI_DATA: u8 = ELFDATA2LSB
+01
+; EI_VERSION: u8 = EV_CURRENT
+01
+; EI_OSABI: u8 = ELFOSABI_NONE
+00
+; EI_ABIVERSION: u8 = 0
+00
+; EI_PAD: [u8; 7]
+00 00 00 00 00 00 00
+; e_type: u16 = E_EXEC
+02 00
+; e_machine: u16 = EM_X86_64
+3e 00
+; e_version: u32 = EV_CURRENT
+01 00 00 00
+; e_entry: u64 = 0x400084
+84 00 40 00 00 00 00 00
+; e_phoff: u64 = 0x40
+40 00 00 00 00 00 00 00
+; e_shoff: u64 = 0x0
+00 00 00 00 00 00 00 00
+; e_flags: u32 = 0
+00 00 00 00
+; e_ehsize: u16 = 64
+40 00
+; e_phentsize: u16 = 56
+38 00
+; e_phnum: u16 = 1
+01 00
+; e_shentsize: u16 = 0
+00 00
+; e_shnum: u16 = 0
+00 00
+; e_shstrndx: u16 = 0
+00 00
+; End of ELF header
+;
+; Program header
+; p_type: u32 = PT_LOAD
+01 00 00 00
+; p_flags: u32 = PF_R | PF_X
+05 00 00 00
+; p_offset: u64 = 0
+00 00 00 00 00 00 00 00
+; p_vaddr: u64 = 0x400000
+00 00 40 00 00 00 00 00
+; p_paddr: u64 = 0x0
+00 00 00 00 00 00 00 00
+; p_filesz: u64 
+ae 00 00 00 00 00 00 00
+; p_memsz: u64 
+ae 00 00 00 00 00 00 00
+; p_align: u64 = 0x1000
+00 10 00 00 00 00 00 00
+; End of program header
+;
+; Ehdr size = 64 = 0x40
+; Phdr size = 56 = 0x38
+; .text size = 54 = 0x36
+; total 0xae
+;
+; Code segment
+;
+; _start
+@hello_world:
+68 65 6c 6c 6f 20 77 6f 72 6c 64 21
+@_start:
+; mov rax, 1
+48 c7 c0 01 00 00 00
+; mov rdi, 1
+48 c7 c7 01 00 00 00
+; mov rsi, [@hello_world]
+48 8d 35 'hello_world
+; mov rdx, 12
+48 c7 c2 0c 00 00 00
+; syscall
+0f 05
+; mov rax, 60 (sys_exit)
+48 c7 c0 3c 00 00 00
+; xor rdi, rdi
+48 31 ff
+; syscall
+0f 05
\ No newline at end of file
diff --git a/stages/hex0/hex1.asm b/stages/hex0/hex1.asm
index 7e1acfe..8a1e1a3 100644
--- a/stages/hex0/hex1.asm
+++ b/stages/hex0/hex1.asm
@@ -199,59 +199,6 @@ get_next_token:
     call skip_comment
     jmp get_next_token
 
-    ;; read the label into scratch, call push_lbl
-label:
-    cmp al, '@'
-    jne .not_label
-    xor rsi, rsi            ; rsi = length of label
-    .label_loop:
-    push rsi
-    call rd_buf
-    pop rsi
-    lea rdi, [r15 + 0x6004] ; rdi = pointer to scratch
-    cmp al, ':'
-    je .label_ret
-    mov byte [rdi + rsi], al
-    inc rsi
-    jmp .label_loop
-
-    .label_ret:
-    lea rdi, [r15 + 0x6004]     ; rdi = pointer to scratch
-    call push_lbl
-    .not_label:
-    ret
-
-label_ref:
-    cmp al, "'"
-    jne .not_label_ref
-    call rd_buf
-    xor rsi, rsi                ; rsi = length of label
-    .label_ref_loop:
-    push rsi
-    call rd_buf
-    pop rsi
-    lea rdi, [r15 + 0x6004] ; rdi = pointer to scratch
-    cmp al, ' '
-    je .label_ref_ret
-    cmp al, `\t`
-    je .label_ref_ret
-    cmp al, `\n`
-    je .label_ref_ret
-    mov byte [rdi + rsi], al
-    inc rsi
-    jmp .label_ref_loop
-
-    .label_ref_ret:
-    lea rdi, [r15 + 0x6004]
-    call find_lbl
-    add dword [r15 + 0x500c], 4 ; advance the file offset by 4
-    sub eax, dword [r15 + 0x500c]       ; add current file offset
-    mov edi, dword [r15 + 0x5004]       ; get current write cursor
-    mov dword [r15 + 0x1000 + rdi], eax ; write the opcode for jmp rel32
-    add dword [r15 + 0x5004], 4 ; advance the write cursor by 4
-    .not_label_ref:
-    ret
-
 char_to_u4:
     cmp al, '0'
     jb .invalid_hex
@@ -296,6 +243,62 @@ octet:
     add rsp, 8
     ret
 
+label_ref:
+    cmp al, "'"
+    jne .not_label_ref
+    xor rsi, rsi                ; rsi = length of label
+    .label_ref_loop:
+    push rsi
+    call rd_buf
+    pop rsi
+    lea rdi, [r15 + 0x6004] ; rdi = pointer to scratch
+    cmp al, ' '
+    je .label_ref_ret
+    cmp al, `\t`
+    je .label_ref_ret
+    cmp al, `\n`
+    je .label_ref_ret
+    mov byte [rdi + rsi], al
+    inc rsi
+    jmp .label_ref_loop
+
+    .label_ref_ret:
+    lea rdi, [r15 + 0x6004]
+    call find_lbl
+    add dword [r15 + 0x500c], 4 ; advance the file offset by 4
+    sub eax, dword [r15 + 0x500c]       ; add current file offset
+    mov edi, dword [r15 + 0x5004]       ; get current write cursor
+    mov dword [r15 + 0x1000 + rdi], eax ; write the opcode for jmp rel32
+    add dword [r15 + 0x5004], 4 ; advance the write cursor by 4
+    ret
+    .not_label_ref:
+    call octet
+    ret
+
+    ;; read the label into scratch, call push_lbl
+label:
+    cmp al, '@'
+    jne .not_label
+    xor rsi, rsi            ; rsi = length of label
+    .label_loop:
+    push rsi
+    call rd_buf
+    pop rsi
+    lea rdi, [r15 + 0x6004] ; rdi = pointer to scratch
+    cmp al, ':'
+    je .label_ret
+    mov byte [rdi + rsi], al
+    inc rsi
+    jmp .label_loop
+
+    .label_ret:
+    lea rdi, [r15 + 0x6004]     ; rdi = pointer to scratch
+    call push_lbl
+    ret
+    .not_label:
+    call label_ref
+    ret
+
     global _start
 _start:
     push rbp
@@ -312,8 +315,4 @@ _start:
     call rd_buf
     call get_next_token
     call label
-    call label_ref
-    call octet
     jmp .loop
-    
-