Lua source code reading 4-Lua VM Command System

Source: Internet
Author: User

This article mainly discusses the Lua Command System (the file lopcodes. c involved ).

 

In Lua, a 32-bit unsigned int type is used to represent a command operation code. The 32-bit value contains a 6-bit operation code and a 26-bit command field.

All instructions have an opcode in the first 6 bits.
Instructions can have the following fields:
'A': 8 bits
'B': 9 bits
'C': 9 bits
'Bx ': 18 bits (' B 'and 'C' together)
'Sbx': signed BX

 

 

 

Enum opmode {IABC, iabx, iasbx};/* Basic Instruction format */

According to the enumerated values above, there are three types of script codes: IABC, iabx, and iasbx. The structure is as follows:

 

 

For command fields a, B, c, there are three types of value sources:

/*
** R (x)-register
** Kst (x)-constant (in constant table)
** Rk (x) = If ISK (x) Then kst (indexK (x) else R (X)
*/
Here, R is used to represent the register value, kst is used to represent the constant table, and rk may be a constant table or register.

 

 

 

In lua5.14, there are a total of 38 command operation codes. The operation code name is as follows:

Const char * const luap_opnames [num_opcodes + 1] ={< br/> "move", <br/> "loadk", <br/> "loadbool ", <br/> "loadnil", <br/> "getupval", <br/> "getglobal", <br/> "gettable", <br/> "setglobal ", <br/> "setupval", <br/> "settable", <br/> "newtable", <br/> "self", <br/> "add ", <br/> "sub", <br/> "Mul", <br/> "Div", <br/> "Mod", <br/> "pow ", <br/> "UNM", <br/> "not", <br/> "len", <br/> "Concat", <br/> "JMP ", <br/> "EQ", <br/> "LT", <br/> "Le", <br/> "test", <br/> "testset ", <br/> "call", <br/> "tailcall", <br/> "return", <br/> "forloop", <br/> "forprep ", <br/> "tforloop", <br/> "setlist", <br/> "close", <br/> "closure", <br/> "vararg ", <br/> null <br/>}; <br/>

 

 

The enumerated value of the operation code, which corresponds to the name of the preceding operation code for convenient calling:

Typedef Enum {<br/>/* example <br/> nameargsdescription <br/> example */<br/> op_move,/* a Br (): = R (B) */<br/> op_loadk,/* A bxr (a): = kst (BX) */<br/> op_loadbool, /* a B Cr (a): = (bool) B; If (c) PC ++ */<br/> op_loadnil,/* a Br (): =...: = R (B): = nil */<br/> op_getupval,/* a Br (): = Upvalue [B] */</P> <p> op_getglobal,/* A bxr (a): = GBL [kst (BX)] */<br/> op_gettable,/* a B Cr (a): = R (B) [rk (c)] */</P> <p> op_setglobal, /* A bxgbl [kst (BX)]: = R (A) */<br/> op_setupval,/* A bupvalue [B]: = R () */<br/> op_settable,/* a B Cr (a) [rk (B)]: = rk (c) */</P> <p> op_newtable, /* a B Cr (a) :{} (size = B, c) */</P> <p> op_self,/* a B Cr (a + 1): = R (B); R (A): = R (B) [rk (c)] */</P> <p> op_add, /* a B Cr (a): = rk (B) + rk (c )* /<Br/> op_sub,/* a B Cr (a): = rk (B)-rk (c) */<br/> op_mul, /* a B Cr (a): = rk (B) * rk (c) */<br/> op_div,/* a B Cr (): = rk (B)/rk (c) */<br/> op_mod,/* a B Cr (a): = rk (B) % rk (c) */<br/> op_pow,/* a B Cr (a): = rk (B) ^ rk (c) */<br/> op_unm, /* a Br (a): =-R (B) */<br/> op_not,/* a Br (a): = Not r (B) */<br/> op_len,/* a Br (a): = length of R (B) */</P> <p> op_concat, /* a B Cr (a): = R (B )....... R (c) */</P> <p> op_jmp,/* sbxpc + = Sbx */</P> <p> op_eq,/* a B cif (rk (B) = rk (c ))~ = A) Then PC ++ */<br/> op_lt,/* a B cif (rk (B) <rk (c ))~ = A) Then PC ++ */<br/> op_le,/* a B cif (rk (B) <= rk (c ))~ = A) Then PC ++ */</P> <p> op_test,/* a CIF not (R (A) <=> C) then PC ++ */<br/> op_testset,/* a B cif (R (B) <=> C) then R (A): = R (B) else PC ++ */</P> <p> op_call,/* a B Cr (),..., R (A + C-2): = R (A) (R (A + 1 ),..., R (A + B-1) */<br/> op_tailcall,/* a B creturn R (A) (R (A + 1 ),..., R (A + B-1) */<br/> op_return,/* A breturn R (),..., R (A + B-2) (see note) */</P> <p> op_forloop,/* A sbxr (A) + = R (A + 2 ); <br/> If R (A) <? = R (A + 1) Then {PC + = sbx; R (A + 3) = R (A)} */<br/> op_forprep, /* A sbxr (a)-= R (A + 2); PC + = sbx */</P> <p> op_tforloop, /* a Cr (a + 3 ),..., R (A + 2 + C): = R (A) (R (A + 1), R (A + 2 )); <br/> If R (A + 3 )~ = Nil then R (A + 2) = R (A + 3) else PC + + + */<br/> op_setlist,/* a B Cr (a) [(C-1) * fpf + I]: = R (A + I), 1 <= I <= B */</P> <p> op_close, /* a close all variables in the stack up to (> =) R (A) */<br/> op_closure,/* A bxr (): = closure (kproto [BX], R (),..., R (A + n) */</P> <p> op_vararg/* a Br (A), R (A + 1 ),..., R (A + B-1) = vararg */<br/>} opcode; <br/>

 

 

 

The operation mode corresponding to 38 operation codes is as follows:

 

 # Define opmode (T, A, B, C, M) (t) <7) | (a) <6) | (B) <4) | (c) <2) | (m )) </P> <p> const lu_byte luap_opmodes [num_opcodes] ={< br/>/* t a B C mode opcode */<br/> opmode (0, 1, opargr, opargn, IABC)/* op_move */<br/>, opmode (0, 1, opargk, opargn, iabx)/* op_loadk */<br/>, opmode (0, 1, opargu, opargu, IABC)/* op_loadbool */<br/>, opmode (0, 1, opargr, opargn, IABC) /* op_loadnil */<br/>, opmode (0, 1, opargu, opargn, IABC)/* op_getupval */<br/>, opmode (0, 1, opargk, opargn, iabx)/* op_getglobal */<br/>, opmode (0, 1, opargr, opargk, IABC)/* op_gettable */<br/>, opmode (0, 0, opargk, opargn, iabx)/* op_setglobal */<br/>, opmode (0, 0, opargu, opargn, IABC)/* op_setupval */<br/>, opmode (0, 0, opargk, opargk, IABC)/* op_settable */<br/>, opmode (0, 1, opargu, opargu, IABC) /* op_newtable */<br/>, opmode (0, 1, opargr, opargk, IABC)/* op_self */<br/>, opmode (0, 1, opargk, opargk, IABC)/* op_add */<br/>, opmode (0, 1, opargk, opargk, IABC)/* op_sub */<br/>, opmode (0, 1, opargk, opargk, IABC)/* op_mul */<br/>, opmode (0, 1, opargk, opargk, IABC)/* op_div */<br/>, opmode (0, 1, opargk, opargk, IABC)/* op_mod */<br/>, opmode (0, 1, opargk, opargk, IABC) /* op_pow */<br/>, opmode (0, 1, opargr, opargn, IABC)/* op_unm */<br/>, opmode (0, 1, opargr, opargn, IABC)/* op_not */<br/>, opmode (0, 1, opargr, opargn, IABC)/* op_len */<br/>, opmode (0, 1, opargr, opargr, IABC)/* op_concat */<br/>, opmode (0, 0, opargr, opargn, iasbx)/* op_jmp */<br/>, opmode (1, 0, opargk, opargk, IABC)/* op_eq */<br/>, opmode (1, 0, opargk, opargk, IABC) /* op_lt */<br/>, opmode (1, 0, opargk, opargk, IABC)/* op_le */<br/>, opmode (1, 1, opargr, opargu, IABC)/* op_test */<br/>, opmode (1, 1, opargr, opargu, IABC)/* op_testset */<br/>, opmode (0, 1, opargu, opargu, IABC)/* op_call */<br/>, opmode (0, 1, opargu, opargu, IABC)/* op_tailcall */<br/>, opmode (0, 0, opargu, opargn, IABC)/* op_return */<br/>, opmode (0, 1, opargr, opargn, iasbx) /* op_forloop */<br/>, opmode (0, 1, opargr, opargn, iasbx)/* op_forprep */<br/>, opmode (1, 0, opargn, opargu, IABC)/* op_tforloop */<br/>, opmode (0, 0, opargu, opargu, IABC)/* op_setlist */<br/>, opmode (0, 0, opargn, opargn, IABC)/* op_close */<br/>, opmode (0, 1, opargu, opargn, iabx)/* op_closure */<br/>, opmode (0, 1, opargu, opargn, IABC)/* op_vararg */<br/>}; </P> <p>

 

 

The formula for calculating the operation mode is as follows:

# Define opmode (T, A, B, C, M) (t) <7) | (a) <6) | (B) <4) | (c) <2) | (m ))

Represented by an 8-bit value

/*
** Masks for instruction properties. The format is:
** Bits 0-1: op Mode
** Bits 2-3: C Arg Mode
** Bits 4-5: B Arg Mode
** Bit 6: Instruction Set register
** Bit 7: operator is a test
*/

From low to high in turn indicates the following meanings:

The first and second bits indicate the operation code mode. Value: Enum opmode {IABC, iabx, iasbx};/* Basic Instruction format */

The second and third bits indicate the C mode parameter. The value is set to Enum opargmask {
Opargn,/* argument is not used */
Opargu,/* argument is used */
Opargr,/* argument is a register or a jump offset */
Opargk/* argument is a constant or register/constant */
};

Opargn indicates unused parameters, opargu indicates available parameters, opargr indicates register parameters or jump instructions, and opargk indicates constant table or register parameters,

The fourth and fifth digits indicate the B mode parameter. The value range is the same as that of C.

The value 6th indicates whether register parameter A is set.

The value 7th indicates whether to perform a test.

 

 

 

 

 

 

 

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.