From a972f38bb636fdcb41a2bb7c856f7f546076b397 Mon Sep 17 00:00:00 2001
From: andromeda <andromeda@lenovo>
Date: Mon, 9 Mar 2026 23:00:39 +0100
Subject: [PATCH] add get_direct_addressing_ModRM and test_*, fix a couple
 bugs/typos, add 'reg value' as register metadata

---
 twasm/README.md     | 16 +++++----
 twasm/asm/main.asm  | 84 ++++++++++++++++++++++++++++++++++++++-------
 twasm/asm/tests.asm | 53 ++++++++++++++++++++++++++--
 3 files changed, 132 insertions(+), 21 deletions(-)

diff --git a/twasm/README.md b/twasm/README.md
index e6f386e..ce6beec 100644
--- a/twasm/README.md
+++ b/twasm/README.md
@@ -127,13 +127,15 @@ type metadata for the different types is as follows:
 ```
 
 ```
-+------------------+
-| type 0x2         |
-+----------+-------+
-| 31    26 | 25 24 |
-+----------+-------+
-| reserved | width |
-+----------+-------+
++------------------------------+
+| type 0x2                     |
++----------+-----------+-------+
+| 31    29 | 28     26 | 25 24 |
++----------+-----------+-------+
+| reserved | reg value | width |
++----------+-----------+-------+
+
+; reg is the value that cooresponds to the register in the ModR/M byte
 
 ; width:
 00b ; 8 bit
diff --git a/twasm/asm/main.asm b/twasm/asm/main.asm
index 57af625..1dfad3b 100644
--- a/twasm/asm/main.asm
+++ b/twasm/asm/main.asm
@@ -187,6 +187,63 @@ get_tte_typed_metadata:
     mov al, [3 + tokens.by_id + rax * 4]
     ret
 
+; ------------------------------------------------------------------------------
+; get_direct_addressing_ModRM
+;
+; description:
+; given 2 register tokens, returns the ModR/M byte in direct addressing
+; (mod = 11b) mode
+;
+; parameters:
+; di = token table entry `reg`
+; si = token table entry `R/M`
+;
+; returned:
+; al = ModR/M byte
+; ------------------------------------------------------------------------------
+
+get_direct_addressing_ModRM:
+  ; TODO something is backwards in this function but I don't see it. If the test
+  ; suite fails, it's too far gone; rewrite it.
+  push rdi
+  push rsi
+  ; get metadata of reg
+  call get_tte_typed_metadata
+  ; al = typed metadata of reg
+  pop rsi
+  pop rdi
+
+  mov bl, al ; bl = metadata of reg
+
+  push rdi
+  push rsi
+  push rbx
+
+  ; get metadata of R/M
+  mov di, si
+  call get_tte_typed_metadata
+  ; al = typed metadata of R/M
+
+  pop rbx
+  pop rsi
+  pop rdi
+
+  mov dl, al
+
+  shr dl, 2
+  and dl, 111b ; mask
+
+  shr bl, 2
+  and bl, 111b ; mask
+  shl bl, 3
+
+  xor eax, eax
+  or al, 11b << 6 ; mod bits
+  or al, dl ; reg bits
+  or al, bl ; R/M bits
+  and rax, 0xFF ; mask for byte
+  ret
+
 ; ------------------------------------------------------------------------------
 ; tokenising
 ; ------------------------------------------------------------------------------
@@ -872,30 +929,33 @@ tokens:
     dw 0x0059
   .by_name_5:
   .by_id:
+    dw 0x0010    ; eax
+    db 0x02      ; type: register
+    db 00000010b ; reg: 000b
+                 ; width: 10b (32 bits)
+
+    dw 0x0000    ; rax
+    db 0x02      ; type: register
+    db 00000011b ; reg: 000b
+                 ; width: 11b (64 bits)
+
+    dw 0x0003    ; rdx
+    db 0x02      ; type: register
+    db 00001011b ; reg: 010b
+                 ; width: 11b (64 bits)
+
     dw 0x0053 ; xor
     db 0x01   ; type: operator
     db 0x02   ; # operands
 
-    dw 0x0010 ; eax
-    db 0x02   ; type: register
-    db 0x02   ; width: 32 bit
-
     dw 0x0054 ; inc
     db 0x01   ; type: operator
     db 0x01   ; # operands
 
-    dw 0x0000 ; rax
-    db 0x02   ; type: register
-    db 0x03   ; width: 64 bit
-
     dw 0x0056 ; mov
     db 0x01   ; type: operator
     db 0x02   ; # operands
 
-    dw 0x0003 ; rdx
-    db 0x02   ; type: register
-    db 0x03   ; width: 64 bit
-
     dw 0x004F ; hlt
     db 0x01   ; type: operator
     db 0x00   ; # operands
diff --git a/twasm/asm/tests.asm b/twasm/asm/tests.asm
index f48d8bf..8480554 100644
--- a/twasm/asm/tests.asm
+++ b/twasm/asm/tests.asm
@@ -34,6 +34,9 @@ run_tests:
   call clear_test_arena
   call test_get_tte_typed_metadata
 
+  call clear_test_arena
+  call test_get_direct_addressing_ModRM
+
   ret
   .msg db "running test suite...", 0x0A, 0x00
 
@@ -441,7 +444,8 @@ test_get_tte_typed_metadata:
 
   mov di, 0x0003              ; rdx
   call get_tte_typed_metadata
-  cmp al, 0x03                ; width: 64 bits
+  cmp al, 00001011b           ; reg: 010b
+                              ; width: 11b (64 bits)
   jne .fail
 
   mov di, 0x0056              ; mov
@@ -462,7 +466,52 @@ test_get_tte_typed_metadata:
     mov rsi, msg_fail
     call print
     ret
-  .msg db "test_get_tte_type...", 0x00
+  .msg db "test_get_tte_typed_metadata...", 0x00
+
+; ------------------------------------------------------------------------------
+; test_get_direct_addressing_ModRM
+;
+; description:
+; tests get_direct_addressing_ModRM described functionality
+; ------------------------------------------------------------------------------
+
+test_get_direct_addressing_ModRM:
+  mov rsi, .msg
+  call print
+
+  mov di, 0x0000 ; rax
+  mov si, 0x0000 ; rax
+  call get_direct_addressing_ModRM
+  cmp al, 11000000b ; Mod Reg R/M: 11b 000b 000b
+  jne .fail
+
+  mov di, 0x0000 ; rax
+  mov si, 0x0003 ; rdx
+  call get_direct_addressing_ModRM
+  cmp al, 11000010b ; Mod Reg R/M: 11b 000b 010b
+  jne .fail
+
+  mov di, 0x0003 ; rdx
+  mov si, 0x0000 ; rax
+  call get_direct_addressing_ModRM
+  cmp al, 11010000b ; Mod Reg R/M: 11b 010b 000b
+  jne .fail
+
+  mov di, 0x0003 ; rdx
+  mov si, 0x0003 ; rdx
+  call get_direct_addressing_ModRM
+  cmp al, 11010010b ; Mod Reg R/M 11b 010b 010b
+  jne .fail
+
+  .pass:
+    mov rsi, msg_pass
+    call print
+    ret
+  .fail:
+    mov rsi, msg_fail
+    call print
+    ret
+  .msg db "test_get_direct_addressing_ModRM...", 0x00
 
 msg_pass:
   db 0x0A