From 466a2b0dfbe9838d4aa97c2b294b2aed34a41017 Mon Sep 17 00:00:00 2001
From: Alvin Cheng <eventide1029@gmail.com>
Date: Sat, 4 Jan 2025 08:52:50 +1100
Subject: [PATCH] Added support for MOVSX (#32)

This pull has added complete support for the MOVSX (Move with sign
extension) instruction, since this instruction exhibits very similar
encodings to the previously implemented MOVSX instruction, many encoding
functionality and encoder tables can be *shared* (Such as pre-processors
and opcods)

Although this pull is listed as only adding support for the MOVSX
instruction, this pull has also renamed the `pre_movzx` pre-proces-
or function `pre_small_operands` to better encapsulate its behavior
and allowing cross-usage to other instruction enocder tables.
---
 TODO.txt                     | 2 +-
 libjas/include/instruction.h | 1 +
 libjas/instruction.c         | 7 ++++---
 3 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/TODO.txt b/TODO.txt
index bfc5462..5699cd3 100644
--- a/TODO.txt
+++ b/TODO.txt
@@ -31,7 +31,7 @@ https://docs.google.com/document/d/1bm9lxVyAk2qogOF9ujERyRcSwXkpKxLW8IWVU8FuMyE/
 More instructions:
 Data Movement Instructions
 XCHG: Exchange data between registers or memory.
-MOVSX/MOVZX: Move with sign/zero extension.
+MOVSX/MOVZX: Move with sign/zero extension. - ✅
 BSWAP: Byte swap (useful for endian conversion).
 CMOVcc: Conditional move based on flags (e.g., CMOVZ, CMOVNZ).
 Arithmetic and Logic Instructions
diff --git a/libjas/include/instruction.h b/libjas/include/instruction.h
index d801e41..4fa5b66 100644
--- a/libjas/include/instruction.h
+++ b/libjas/include/instruction.h
@@ -83,6 +83,7 @@ enum instructions {
   INSTR_INT,
   INSTR_SYSCALL,
   INSTR_MOVZX,
+  INSTR_MOVSX,
 
   INSTR_DUMMY,
 
diff --git a/libjas/instruction.c b/libjas/instruction.c
index 0eed573..19a9881 100644
--- a/libjas/instruction.c
+++ b/libjas/instruction.c
@@ -176,12 +176,13 @@ DEFINE_TAB(syscall) = {
     INSTR_TERMINATOR,
 };
 
-static void pre_movzx(operand_t *op_arr, buffer_t *buf, instr_encode_table_t *instr_ref, enum modes mode) {
+static void pre_small_operands(operand_t *op_arr, buffer_t *buf, instr_encode_table_t *instr_ref, enum modes mode) {
   if (op_sizeof(op_arr[1].type) < 16)
     err("Invalid operand size for MOVZX instruction");
 }
 
-DEFINE_TAB(movzx) = {{ENC_RM, NULL, {0x0F, 0xB7}, MODE_SUPPORT_ALL, {0x0F, 0xB6}, 2, &pre_movzx, true}, INSTR_TERMINATOR};
+DEFINE_TAB(movzx) = {{ENC_RM, NULL, {0x0F, 0xB7}, MODE_SUPPORT_ALL, {0x0F, 0xB6}, 2, &pre_small_operands, true}, INSTR_TERMINATOR};
+DEFINE_TAB(movsx) = {{ENC_RM, NULL, {0x0F, 0xBF}, MODE_SUPPORT_ALL, {0x0F, 0xBE}, 2, &pre_small_operands, true}, INSTR_TERMINATOR};
 
 // clang-format off
 
@@ -190,7 +191,7 @@ instr_encode_table_t *instr_table[] =
         mov, lea, add, sub, mul, div, and, or, xor, _not, inc,
         dec, jmp, je, jne, jz, jnz, call, ret, cmp, push, pop,
         in, out, clc, stc, cli, sti, nop, hlt, _int, syscall, 
-        movzx,
+        movzx, movsx,
     };