ViewVC Help
View File | Revision Log | Show Annotations | Revision Graph | Root Listing
root/cebix/BasiliskII/src/uae_cpu/compiler/codegen_x86.h
Revision: 1.3
Committed: 2003-03-17T17:18:24Z (21 years, 8 months ago) by gbeauche
Content type: text/plain
Branch: MAIN
Changes since 1.2: +335 -68 lines
Log Message:
Add some SSE/SSE2 instructions

File Contents

# User Rev Content
1 gbeauche 1.2 /******************** -*- mode: C; tab-width: 8 -*- ********************
2 gbeauche 1.1 *
3     * Run-time assembler for i386 and x86-64
4     *
5     ***********************************************************************/
6    
7    
8     /***********************************************************************
9     *
10     * This file is derived from GNU lightning.
11     *
12     * Copyright 1999, 2000, 2001, 2002, 2003 Ian Piumarta
13     *
14     * Adaptations and enhancements for x86-64 support, Copyright 2003
15     * Gwenole Beauchesne
16     *
17     * Basilisk II (C) 1997-2003 Christian Bauer
18     *
19     * This program is free software; you can redistribute it and/or modify
20     * it under the terms of the GNU General Public License as published by
21     * the Free Software Foundation; either version 2 of the License, or
22     * (at your option) any later version.
23     *
24     * This program is distributed in the hope that it will be useful,
25     * but WITHOUT ANY WARRANTY; without even the implied warranty of
26     * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
27     * GNU General Public License for more details.
28     *
29     * You should have received a copy of the GNU General Public License
30     * along with this program; if not, write to the Free Software
31     * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
32     *
33     ***********************************************************************/
34    
35     #ifndef X86_RTASM_H
36     #define X86_RTASM_H
37    
38     /* NOTES
39     *
40     * o Best viewed on a 1024x768 screen with fixed-6x10 font ;-)
41     *
42     * TODO
43     *
44     * o Fix FIXMEs
45     * o Conditional moves
46     * o i387 FPU instructions
47     * o SSE instructions
48     * o Add notes about RIP addressing mode for x86-64
49     * o Optimize for cases where register numbers are not integral constants
50     */
51    
52     /* --- Configuration ------------------------------------------------------- */
53    
54     /* Define to settle a "flat" register set, i.e. different regno for
55     each size variant. */
56     #ifndef X86_FLAT_REGISTERS
57     #define X86_FLAT_REGISTERS 1
58     #endif
59    
60     /* Define to generate x86-64 code. */
61     #ifndef X86_TARGET_64BIT
62     #define X86_TARGET_64BIT 0
63     #endif
64    
65     /* Define to optimize ALU instructions. */
66     #ifndef X86_OPTIMIZE_ALU
67     #define X86_OPTIMIZE_ALU 1
68     #endif
69    
70     /* Define to optimize rotate/shift instructions. */
71     #ifndef X86_OPTIMIZE_ROTSHI
72     #define X86_OPTIMIZE_ROTSHI 1
73     #endif
74    
75    
76     /* --- Macros -------------------------------------------------------------- */
77    
78     /* Functions used to emit code.
79     *
80     * x86_emit_byte(B)
81     * x86_emit_word(W)
82     * x86_emit_long(L)
83     */
84    
85     /* Get pointer to current code
86     *
87     * x86_get_target()
88     */
89    
90     /* Abort assembler, fatal failure.
91     *
92     * x86_emit_failure(MSG)
93     */
94    
95    
96     /* --- Register set -------------------------------------------------------- */
97    
98 gbeauche 1.2 enum {
99 gbeauche 1.1 #if X86_FLAT_REGISTERS
100 gbeauche 1.3 X86_NOREG = 0,
101     X86_Reg8L_Base = 0x10,
102     X86_Reg8H_Base = 0x20,
103     X86_Reg16_Base = 0x30,
104     X86_Reg32_Base = 0x40,
105     X86_Reg64_Base = 0x50,
106     X86_RegMMX_Base = 0x60,
107     X86_RegXMM_Base = 0x70,
108 gbeauche 1.1 #else
109 gbeauche 1.3 X86_NOREG = -1,
110     X86_Reg8L_Base = 0,
111     X86_Reg8H_Base = 16,
112     X86_Reg16_Base = 0,
113     X86_Reg32_Base = 0,
114     X86_Reg64_Base = 0,
115     X86_RegMMX_Base = 0,
116     X86_RegXMM_Base = 0,
117 gbeauche 1.1 #endif
118 gbeauche 1.2 };
119 gbeauche 1.1
120 gbeauche 1.2 enum {
121 gbeauche 1.1 X86_AL = X86_Reg8L_Base,
122     X86_CL, X86_DL, X86_BL,
123     X86_AH, X86_CH, X86_DH, X86_BH,
124     X86_R8B, X86_R9B, X86_R10B, X86_R11B,
125     X86_R12B, X86_R13B, X86_R14B, X86_R15B,
126     X86_SPL = X86_Reg8H_Base + 4,
127     X86_BPL, X86_SIL, X86_DIL
128 gbeauche 1.2 };
129 gbeauche 1.1
130 gbeauche 1.2 enum {
131 gbeauche 1.1 X86_AX = X86_Reg16_Base,
132     X86_CX, X86_DX, X86_BX,
133     X86_SP, X86_BP, X86_SI, X86_DI,
134     X86_R8W, X86_R9W, X86_R10W, X86_R11W,
135     X86_R12W, X86_R13W, X86_R14W, X86_R15W
136 gbeauche 1.2 };
137 gbeauche 1.1
138 gbeauche 1.2 enum {
139 gbeauche 1.1 X86_EAX = X86_Reg32_Base,
140     X86_ECX, X86_EDX, X86_EBX,
141     X86_ESP, X86_EBP, X86_ESI, X86_EDI,
142     X86_R8D, X86_R9D, X86_R10D, X86_R11D,
143     X86_R12D, X86_R13D, X86_R14D, X86_R15D
144 gbeauche 1.2 };
145 gbeauche 1.1
146 gbeauche 1.2 enum {
147 gbeauche 1.1 X86_RAX = X86_Reg64_Base,
148     X86_RCX, X86_RDX, X86_RBX,
149     X86_RSP, X86_RBP, X86_RSI, X86_RDI,
150     X86_R8, X86_R9, X86_R10, X86_R11,
151     X86_R12, X86_R13, X86_R14, X86_R15
152 gbeauche 1.2 };
153 gbeauche 1.1
154 gbeauche 1.3 enum {
155     X86_MM0 = X86_RegMMX_Base,
156     X86_MM1, X86_MM2, X86_MM3,
157     X86_MM4, X86_MM5, X86_MM6, X86_MM7,
158     };
159    
160     enum {
161     X86_XMM0 = X86_RegXMM_Base,
162     X86_XMM1, X86_XMM2, X86_XMM3,
163     X86_XMM4, X86_XMM5, X86_XMM6, X86_XMM7,
164     X86_XMM8, X86_XMM9, X86_XMM10, X86_XMM11,
165     X86_XMM12, X86_XMM13, X86_XMM14, X86_XMM15
166     };
167    
168 gbeauche 1.1 /* Register control and access
169     *
170 gbeauche 1.3 * _r0P(R) Null register?
171     * _rXP(R) Extended register?
172 gbeauche 1.1 *
173 gbeauche 1.3 * _rC(R) Class of register (only valid if X86_FLAT_REGISTERS)
174 gbeauche 1.1 * _rR(R) Full register number
175     * _rN(R) Short register number for encoding
176     *
177     * _r1(R) 8-bit register ID
178     * _r2(R) 16-bit register ID
179     * _r4(R) 32-bit register ID
180     * _r8(R) 64-bit register ID
181 gbeauche 1.3 * _rM(R) MMX register ID
182     * _rX(R) XMM register ID
183     * _rA(R) Address register ID used for EA calculation
184 gbeauche 1.1 */
185    
186 gbeauche 1.3 #define _r0P(R) ((R) == X86_NOREG)
187 gbeauche 1.1
188 gbeauche 1.3 #define _rC(R) ((R) & 0xf0)
189     #define _rR(R) ((R) & 0x0f)
190     #define _rN(R) ((R) & 0x07)
191     #define _rXP(R) (_rR(R) > 7)
192 gbeauche 1.1
193 gbeauche 1.3 #if !defined(_ASM_SAFETY) || ! X86_FLAT_REGISTERS
194 gbeauche 1.1 #define _r1(R) _rN(R)
195     #define _r2(R) _rN(R)
196     #define _r4(R) _rN(R)
197     #define _r8(R) _rN(R)
198 gbeauche 1.3 #define _rA(R) _rN(R)
199     #define _rM(R) _rN(R)
200     #define _rX(R) _rN(R)
201 gbeauche 1.1 #else
202 gbeauche 1.3 #define _r1(R) ( ((_rC(R) & (X86_Reg8L_Base | X86_Reg8H_Base)) != 0) ? _rN(R) : x86_emit_failure( "8-bit register required"))
203     #define _r2(R) ( (_rC(R) == X86_Reg16_Base) ? _rN(R) : x86_emit_failure("16-bit register required"))
204     #define _r4(R) ( (_rC(R) == X86_Reg32_Base) ? _rN(R) : x86_emit_failure("32-bit register required"))
205     #define _r8(R) ( (_rC(R) == X86_Reg64_Base) ? _rN(R) : x86_emit_failure("64-bit register required"))
206     #define _rA(R) ( X86_TARGET_64BIT ? \
207     ( (_rC(R) == X86_Reg64_Base) ? _rN(R) : x86_emit_failure("not a valid 64-bit base/index expression")) : \
208     ( (_rC(R) == X86_Reg32_Base) ? _rN(R) : x86_emit_failure("not a valid 32-bit base/index expression")) )
209     #define _rM(R) ( (_rC(R) == X86_RegMMX_Base) ? _rN(R) : x86_emit_failure("MMX register required"))
210     #define _rX(R) ( (_rC(R) == X86_RegXMM_Base) ? _rN(R) : x86_emit_failure("SSE register required"))
211 gbeauche 1.1 #endif
212    
213 gbeauche 1.3 #define _rSP() (X86_TARGET_64BIT ? X86_RSP : X86_ESP)
214     #define _rbpP(R) (_rR(R) == _rR(X86_RBP))
215     #define _rspP(R) (_rR(R) == _rR(X86_RSP))
216     #define _rsp12P(R) (_rN(R) == _rN(X86_RSP))
217 gbeauche 1.1
218    
219     /* ========================================================================= */
220     /* --- UTILITY ------------------------------------------------------------- */
221     /* ========================================================================= */
222    
223     typedef char _sc;
224     typedef unsigned char _uc;
225     typedef unsigned short _us;
226     typedef int _sl;
227     typedef unsigned int _ul;
228    
229     #define _UC(X) ((_uc )(X))
230     #define _US(X) ((_us )(X))
231     #define _SL(X) ((_sl )(X))
232     #define _UL(X) ((_ul )(X))
233    
234     # define _PUC(X) ((_uc *)(X))
235     # define _PUS(X) ((_us *)(X))
236     # define _PSL(X) ((_sl *)(X))
237     # define _PUL(X) ((_ul *)(X))
238    
239     #define _B(B) x86_emit_byte((B))
240     #define _W(W) x86_emit_word((W))
241     #define _L(L) x86_emit_long((L))
242    
243     #define _MASK(N) ((unsigned)((1<<(N)))-1)
244     #define _siP(N,I) (!((((unsigned)(I))^(((unsigned)(I))<<1))&~_MASK(N)))
245     #define _uiP(N,I) (!(((unsigned)(I))&~_MASK(N)))
246     #define _suiP(N,I) (_siP(N,I) | _uiP(N,I))
247    
248     #ifndef _ASM_SAFETY
249     #define _ck_s(W,I) (_UL(I) & _MASK(W))
250     #define _ck_u(W,I) (_UL(I) & _MASK(W))
251     #define _ck_su(W,I) (_UL(I) & _MASK(W))
252     #define _ck_d(W,I) (_UL(I) & _MASK(W))
253     #else
254     #define _ck_s(W,I) (_siP(W,I) ? (_UL(I) & _MASK(W)) : x86_emit_failure( "signed integer `"#I"' too large for "#W"-bit field"))
255     #define _ck_u(W,I) (_uiP(W,I) ? (_UL(I) & _MASK(W)) : x86_emit_failure("unsigned integer `"#I"' too large for "#W"-bit field"))
256     #define _ck_su(W,I) (_suiP(W,I) ? (_UL(I) & _MASK(W)) : x86_emit_failure( "integer `"#I"' too large for "#W"-bit field"))
257     #define _ck_d(W,I) (_siP(W,I) ? (_UL(I) & _MASK(W)) : x86_emit_failure( "displacement `"#I"' too large for "#W"-bit field"))
258     #endif
259    
260     #define _s0P(I) ((I)==0)
261     #define _s8P(I) _siP(8,I)
262     #define _s16P(I) _siP(16,I)
263     #define _u8P(I) _uiP(8,I)
264     #define _u16P(I) _uiP(16,I)
265    
266     #define _su8(I) _ck_su(8,I)
267     #define _su16(I) _ck_su(16,I)
268    
269     #define _s1(I) _ck_s( 1,I)
270     #define _s2(I) _ck_s( 2,I)
271     #define _s3(I) _ck_s( 3,I)
272     #define _s4(I) _ck_s( 4,I)
273     #define _s5(I) _ck_s( 5,I)
274     #define _s6(I) _ck_s( 6,I)
275     #define _s7(I) _ck_s( 7,I)
276     #define _s8(I) _ck_s( 8,I)
277     #define _s9(I) _ck_s( 9,I)
278     #define _s10(I) _ck_s(10,I)
279     #define _s11(I) _ck_s(11,I)
280     #define _s12(I) _ck_s(12,I)
281     #define _s13(I) _ck_s(13,I)
282     #define _s14(I) _ck_s(14,I)
283     #define _s15(I) _ck_s(15,I)
284     #define _s16(I) _ck_s(16,I)
285     #define _s17(I) _ck_s(17,I)
286     #define _s18(I) _ck_s(18,I)
287     #define _s19(I) _ck_s(19,I)
288     #define _s20(I) _ck_s(20,I)
289     #define _s21(I) _ck_s(21,I)
290     #define _s22(I) _ck_s(22,I)
291     #define _s23(I) _ck_s(23,I)
292     #define _s24(I) _ck_s(24,I)
293     #define _s25(I) _ck_s(25,I)
294     #define _s26(I) _ck_s(26,I)
295     #define _s27(I) _ck_s(27,I)
296     #define _s28(I) _ck_s(28,I)
297     #define _s29(I) _ck_s(29,I)
298     #define _s30(I) _ck_s(30,I)
299     #define _s31(I) _ck_s(31,I)
300     #define _u1(I) _ck_u( 1,I)
301     #define _u2(I) _ck_u( 2,I)
302     #define _u3(I) _ck_u( 3,I)
303     #define _u4(I) _ck_u( 4,I)
304     #define _u5(I) _ck_u( 5,I)
305     #define _u6(I) _ck_u( 6,I)
306     #define _u7(I) _ck_u( 7,I)
307     #define _u8(I) _ck_u( 8,I)
308     #define _u9(I) _ck_u( 9,I)
309     #define _u10(I) _ck_u(10,I)
310     #define _u11(I) _ck_u(11,I)
311     #define _u12(I) _ck_u(12,I)
312     #define _u13(I) _ck_u(13,I)
313     #define _u14(I) _ck_u(14,I)
314     #define _u15(I) _ck_u(15,I)
315     #define _u16(I) _ck_u(16,I)
316     #define _u17(I) _ck_u(17,I)
317     #define _u18(I) _ck_u(18,I)
318     #define _u19(I) _ck_u(19,I)
319     #define _u20(I) _ck_u(20,I)
320     #define _u21(I) _ck_u(21,I)
321     #define _u22(I) _ck_u(22,I)
322     #define _u23(I) _ck_u(23,I)
323     #define _u24(I) _ck_u(24,I)
324     #define _u25(I) _ck_u(25,I)
325     #define _u26(I) _ck_u(26,I)
326     #define _u27(I) _ck_u(27,I)
327     #define _u28(I) _ck_u(28,I)
328     #define _u29(I) _ck_u(29,I)
329     #define _u30(I) _ck_u(30,I)
330     #define _u31(I) _ck_u(31,I)
331    
332     /* ========================================================================= */
333     /* --- ASSEMBLER ----------------------------------------------------------- */
334     /* ========================================================================= */
335    
336     #define _b00 0
337     #define _b01 1
338     #define _b10 2
339     #define _b11 3
340    
341     #define _b000 0
342     #define _b001 1
343     #define _b010 2
344     #define _b011 3
345     #define _b100 4
346     #define _b101 5
347     #define _b110 6
348     #define _b111 7
349    
350     #define _OFF4(D) (_UL(D) - _UL(x86_get_target()))
351     #define _CKD8(D) _ck_d(8, ((_uc) _OFF4(D)) )
352    
353     #define _D8(D) (_B(0), ((*(_PUC(x86_get_target())-1))= _CKD8(D)))
354     #define _D32(D) (_L(0), ((*(_PUL(x86_get_target())-1))= _OFF4(D)))
355    
356     #ifndef _ASM_SAFETY
357     # define _M(M) (M)
358     # define _r(R) (R)
359     # define _m(M) (M)
360     # define _s(S) (S)
361     # define _i(I) (I)
362     # define _b(B) (B)
363     #else
364     # define _M(M) (((M)>3) ? x86_emit_failure("internal error: mod = " #M) : (M))
365     # define _r(R) (((R)>7) ? x86_emit_failure("internal error: reg = " #R) : (R))
366     # define _m(M) (((M)>7) ? x86_emit_failure("internal error: r/m = " #M) : (M))
367     # define _s(S) (((S)>3) ? x86_emit_failure("internal error: memory scale = " #S) : (S))
368     # define _i(I) (((I)>7) ? x86_emit_failure("internal error: memory index = " #I) : (I))
369     # define _b(B) (((B)>7) ? x86_emit_failure("internal error: memory base = " #B) : (B))
370     #endif
371    
372     #define _Mrm(Md,R,M) _B((_M(Md)<<6)|(_r(R)<<3)|_m(M))
373     #define _SIB(Sc,I, B) _B((_s(Sc)<<6)|(_i(I)<<3)|_b(B))
374    
375     #define _SCL(S) ((((S)==1) ? _b00 : \
376     (((S)==2) ? _b01 : \
377     (((S)==4) ? _b10 : \
378     (((S)==8) ? _b11 : x86_emit_failure("illegal scale: " #S))))))
379    
380    
381     /* --- Memory subformats - urgh! ------------------------------------------- */
382    
383 gbeauche 1.3 #define _r_D( R, D ) (_Mrm(_b00,_rN(R),_b101 ) ,_L((long)(D)))
384     #define _r_0B( R, B ) (_Mrm(_b00,_rN(R),_rA(B)) )
385     #define _r_0BIS(R, B,I,S) (_Mrm(_b00,_rN(R),_b100 ),_SIB(_SCL(S),_rA(I),_rA(B)) )
386     #define _r_1B( R, D,B ) (_Mrm(_b01,_rN(R),_rA(B)) ,_B((long)(D)))
387     #define _r_1BIS(R, D,B,I,S) (_Mrm(_b01,_rN(R),_b100 ),_SIB(_SCL(S),_rA(I),_rA(B)),_B((long)(D)))
388     #define _r_4B( R, D,B ) (_Mrm(_b10,_rN(R),_rA(B)) ,_L((long)(D)))
389     #define _r_4IS( R, D,I,S) (_Mrm(_b00,_rN(R),_b100 ),_SIB(_SCL(S),_rA(I),_b101 ),_L((long)(D)))
390     #define _r_4BIS(R, D,B,I,S) (_Mrm(_b10,_rN(R),_b100 ),_SIB(_SCL(S),_rA(I),_rA(B)),_L((long)(D)))
391    
392     #define _r_DB( R, D,B ) ((_s0P(D) && (!_rbpP(B)) ? _r_0B (R, B ) : (_s8P(D) ? _r_1B( R,D,B ) : _r_4B( R,D,B ))))
393     #define _r_DBIS(R, D,B,I,S) ((_s0P(D) ? _r_0BIS(R, B,I,S) : (_s8P(D) ? _r_1BIS(R,D,B,I,S) : _r_4BIS(R,D,B,I,S))))
394    
395     #define _r_X( R, D,B,I,S) (_r0P(I) ? (_r0P(B) ? _r_D (R,D ) : \
396     (_rsp12P(B) ? _r_DBIS(R,D,_rSP(),_rSP(),1) : \
397     _r_DB (R,D, B ))) : \
398     (_r0P(B) ? _r_4IS (R,D, I,S) : \
399     (!_rspP(I) ? _r_DBIS(R,D, B, I,S) : \
400     x86_emit_failure("illegal index register: %esp"))))
401 gbeauche 1.1
402    
403     /* --- Instruction formats ------------------------------------------------- */
404    
405     #define _m32only(X) (! X86_TARGET_64BIT ? X : x86_emit_failure("invalid instruction in 64-bit mode"))
406     #define _m64only(X) ( X86_TARGET_64BIT ? X : x86_emit_failure("invalid instruction in 32-bit mode"))
407     #define _m64(X) ( X86_TARGET_64BIT ? X : ((void)0) )
408    
409     /* _format Opcd ModR/M dN(rB,rI,Sc) imm... */
410    
411     #define _d16() ( _B(0x66 ) )
412     #define _O( OP ) ( _B( OP ) )
413     #define _Or( OP,R ) ( _B( (OP)|_r(R)) )
414     #define _OO( OP ) ( _B((OP)>>8), _B( (OP) ) )
415     #define _OOr( OP,R ) ( _B((OP)>>8), _B( (OP)|_r(R)) )
416     #define _Os( OP,B ) ( _s8P(B) ? _B(((OP)|_b10)) : _B(OP) )
417     #define _sW( W ) ( _s8P(W) ? _B(W):_W(W) )
418     #define _sL( L ) ( _s8P(L) ? _B(L):_L(L) )
419     #define _O_B( OP ,B ) ( _O ( OP ) ,_B(B) )
420     #define _O_W( OP ,W ) ( _O ( OP ) ,_W(W) )
421     #define _O_L( OP ,L ) ( _O ( OP ) ,_L(L) )
422     #define _O_D8( OP ,D ) ( _O ( OP ) ,_D8(D) )
423     #define _O_D32( OP ,D ) ( _O ( OP ) ,_D32(D) )
424     #define _OO_D32( OP ,D ) ( _OO ( OP ) ,_D32(D) )
425     #define _Os_sW( OP ,W ) ( _Os ( OP,W) ,_sW(W) )
426     #define _Os_sL( OP ,L ) ( _Os ( OP,L) ,_sL(L) )
427     #define _O_W_B( OP ,W,B) ( _O ( OP ) ,_W(W),_B(B))
428     #define _Or_B( OP,R ,B ) ( _Or ( OP,R) ,_B(B) )
429     #define _Or_W( OP,R ,W ) ( _Or ( OP,R) ,_W(W) )
430     #define _Or_L( OP,R ,L ) ( _Or ( OP,R) ,_L(L) )
431     #define _O_Mrm( OP ,MO,R,M ) ( _O ( OP ),_Mrm(MO,R,M ) )
432     #define _OO_Mrm( OP ,MO,R,M ) ( _OO ( OP ),_Mrm(MO,R,M ) )
433     #define _O_Mrm_B( OP ,MO,R,M ,B ) ( _O ( OP ),_Mrm(MO,R,M ) ,_B(B) )
434     #define _O_Mrm_W( OP ,MO,R,M ,W ) ( _O ( OP ),_Mrm(MO,R,M ) ,_W(W) )
435     #define _O_Mrm_L( OP ,MO,R,M ,L ) ( _O ( OP ),_Mrm(MO,R,M ) ,_L(L) )
436     #define _OO_Mrm_B( OP ,MO,R,M ,B ) ( _OO ( OP ),_Mrm(MO,R,M ) ,_B(B) )
437     #define _Os_Mrm_sW(OP ,MO,R,M ,W ) ( _Os ( OP,W),_Mrm(MO,R,M ),_sW(W) )
438     #define _Os_Mrm_sL(OP ,MO,R,M ,L ) ( _Os ( OP,L),_Mrm(MO,R,M ),_sL(L) )
439     #define _O_r_X( OP ,R ,MD,MB,MI,MS ) ( _O ( OP ),_r_X( R ,MD,MB,MI,MS) )
440     #define _OO_r_X( OP ,R ,MD,MB,MI,MS ) ( _OO ( OP ),_r_X( R ,MD,MB,MI,MS) )
441     #define _O_r_X_B( OP ,R ,MD,MB,MI,MS,B ) ( _O ( OP ),_r_X( R ,MD,MB,MI,MS) ,_B(B) )
442     #define _O_r_X_W( OP ,R ,MD,MB,MI,MS,W ) ( _O ( OP ),_r_X( R ,MD,MB,MI,MS) ,_W(W) )
443     #define _O_r_X_L( OP ,R ,MD,MB,MI,MS,L ) ( _O ( OP ),_r_X( R ,MD,MB,MI,MS) ,_L(L) )
444     #define _OO_r_X_B( OP ,R ,MD,MB,MI,MS,B ) ( _OO ( OP ),_r_X( R ,MD,MB,MI,MS) ,_B(B) )
445     #define _Os_r_X_sW(OP ,R ,MD,MB,MI,MS,W ) ( _Os ( OP,W),_r_X( R ,MD,MB,MI,MS),_sW(W) )
446     #define _Os_r_X_sL(OP ,R ,MD,MB,MI,MS,L ) ( _Os ( OP,L),_r_X( R ,MD,MB,MI,MS),_sL(L) )
447     #define _O_X_B( OP ,MD,MB,MI,MS,B ) ( _O_r_X_B( OP ,0 ,MD,MB,MI,MS ,B) )
448     #define _O_X_W( OP ,MD,MB,MI,MS,W ) ( _O_r_X_W( OP ,0 ,MD,MB,MI,MS ,W) )
449     #define _O_X_L( OP ,MD,MB,MI,MS,L ) ( _O_r_X_L( OP ,0 ,MD,MB,MI,MS ,L) )
450    
451    
452     /* --- REX prefixes -------------------------------------------------------- */
453    
454     #define _VOID() ((void)0)
455     #define _BIT(X) (!!(X))
456     #define _d64(W,R,X,B) (_B(0x40|(W)<<3|(R)<<2|(X)<<1|(B)))
457    
458     #define __REXwrxb(L,W,R,X,B) ((W|R|X|B) || (L) ? _d64(W,R,X,B) : _VOID())
459     #define __REXwrx_(L,W,R,X,MR) (__REXwrxb(L,W,R,X,_BIT(_rXP(MR))))
460     #define __REXw_x_(L,W,R,X,MR) (__REXwrx_(L,W,_BIT(_rXP(R)),X,MR))
461    
462     // FIXME: can't mix new (SPL,BPL,SIL,DIL) with (AH,BH,CH,DH)
463     #define _REXBrr(RR,MR) _m64(__REXw_x_(((RR)|(MR))>=X86_SPL,0,RR,0,MR))
464     #define _REXBmr(MB,MI,RD) _m64(__REXw_x_(((RR)|(MR))>=X86_SPL,0,RD,_BIT(_rXP(MI)),MB))
465     #define _REXBrm(RS,MB,MI) _REXBmr(MB,MI,RS)
466    
467     #define _REXLrr(RR,MR) _m64(__REXw_x_(0,0,RR,0,MR))
468     #define _REXLmr(MB,MI,RD) _m64(__REXw_x_(0,0,RD,_BIT(_rXP(MI)),MB))
469     #define _REXLrm(RS,MB,MI) _REXLmr(MB,MI,RS)
470    
471     #define _REXQrr(RR,MR) _m64only(__REXw_x_(0,1,RR,0,MR))
472     #define _REXQmr(MB,MI,RD) _m64only(__REXw_x_(0,1,RD,_BIT(_rXP(MI)),MB))
473     #define _REXQrm(RS,MB,MI) _REXQmr(MB,MI,RS)
474    
475    
476     /* ========================================================================= */
477     /* --- Fully-qualified intrinsic instructions ------------------------------ */
478     /* ========================================================================= */
479    
480     /* OPCODE + i = immediate operand
481     * + r = register operand
482     * + m = memory operand (disp,base,index,scale)
483     * + sr/sm = a star preceding a register or memory
484 gbeauche 1.2 * + 0 = top of stack register (for FPU instructions)
485 gbeauche 1.1 */
486    
487     /* --- ALU instructions ---------------------------------------------------- */
488    
489 gbeauche 1.2 enum {
490 gbeauche 1.1 X86_ADD = 0,
491     X86_OR = 1,
492     X86_ADC = 2,
493     X86_SBB = 3,
494     X86_AND = 4,
495     X86_SUB = 5,
496     X86_XOR = 6,
497     X86_CMP = 7,
498 gbeauche 1.2 };
499 gbeauche 1.1
500     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
501    
502     #define _ALUBrr(OP,RS, RD) (_REXBrr(RS, RD), _O_Mrm (((OP) << 3) ,_b11,_r1(RS),_r1(RD) ))
503     #define _ALUBmr(OP, MD, MB, MI, MS, RD) (_REXBmr(MB, MI, RD), _O_r_X (((OP) << 3) + 2,_r1(RD) ,MD,MB,MI,MS ))
504     #define _ALUBrm(OP, RS, MD, MB, MI, MS) (_REXBrm(RS, MB, MI), _O_r_X (((OP) << 3) , ,_r1(RS) ,MD,MB,MI,MS ))
505     #define _ALUBir(OP, IM, RD) (X86_OPTIMIZE_ALU && ((RD) == X86_AL) ? \
506     (_REXBrr(0, RD), _O_B (((OP) << 3) + 4 ,_su8(IM))) : \
507     (_REXBrr(0, RD), _O_Mrm_B (0x80 ,_b11,OP ,_r1(RD) ,_su8(IM))) )
508     #define _ALUBim(OP, IM, MD, MB, MI, MS) (_REXBrm(0, MB, MI), _O_r_X_B (0x80 ,OP ,MD,MB,MI,MS ,_su8(IM)))
509    
510     #define _ALUWrr(OP, RS, RD) (_d16(), _REXLrr(RS, RD), _O_Mrm (((OP) << 3) + 1,_b11,_r2(RS),_r2(RD) ))
511     #define _ALUWmr(OP, MD, MB, MI, MS, RD) (_d16(), _REXLmr(MD, MI, RD), _O_r_X (((OP) << 3) + 3 ,_r2(RD) ,MD,MB,MI,MS ))
512     #define _ALUWrm(OP, RS, MD, MB, MI, MS) (_d16(), _REXLrm(RS, MD, MI), _O_r_X (((OP) << 3) + 1 ,_r2(RS) ,MD,MB,MI,MS ))
513     #define _ALUWir(OP, IM, RD) (X86_OPTIMIZE_ALU && ((RD) == X86_AX) ? \
514     (_d16(), _REXLrr(0, RD), _O_W (((OP) << 3) + 5 ,_su16(IM))) : \
515     (_d16(), _REXLrr(0, RD), _Os_Mrm_sW (0x81 ,_b11,OP ,_r2(RD) ,_su16(IM))) )
516     #define _ALUWim(OP, IM, MD, MB, MI, MS) (_d16(), _REXLrm(0, MD, MI), _Os_r_X_sW (0x81 ,OP ,MD,MB,MI,MS ,_su16(IM)))
517    
518     #define _ALULrr(OP, RS, RD) (_REXLrr(RS, RD), _O_Mrm (((OP) << 3) + 1,_b11,_r4(RS),_r4(RD) ))
519     #define _ALULmr(OP, MD, MB, MI, MS, RD) (_REXLmr(MB, MI, RD), _O_r_X (((OP) << 3) + 3 ,_r4(RD) ,MD,MB,MI,MS ))
520     #define _ALULrm(OP, RS, MD, MB, MI, MS) (_REXLrm(RS, MB, MI), _O_r_X (((OP) << 3) + 1 ,_r4(RS) ,MD,MB,MI,MS ))
521     #define _ALULir(OP, IM, RD) (X86_OPTIMIZE_ALU && ((RD) == X86_EAX) ? \
522     (_REXLrr(0, RD), _O_L (((OP) << 3) + 5 ,IM )) : \
523     (_REXLrr(0, RD), _Os_Mrm_sL (0x81 ,_b11,OP ,_r4(RD) ,IM )) )
524     #define _ALULim(OP, IM, MD, MB, MI, MS) (_REXLrm(0, MB, MI), _Os_r_X_sL (0x81 ,OP ,MD,MB,MI,MS ,IM ))
525    
526     #define _ALUQrr(OP, RS, RD) (_REXQrr(RS, RD), _O_Mrm (((OP) << 3) + 1,_b11,_r8(RS),_r8(RD) ))
527     #define _ALUQmr(OP, MD, MB, MI, MS, RD) (_REXQmr(MB, MI, RD), _O_r_X (((OP) << 3) + 3 ,_r8(RD) ,MD,MB,MI,MS ))
528     #define _ALUQrm(OP, RS, MD, MB, MI, MS) (_REXQrm(RS, MB, MI), _O_r_X (((OP) << 3) + 1 ,_r8(RS) ,MD,MB,MI,MS ))
529     #define _ALUQir(OP, IM, RD) (X86_OPTIMIZE_ALU && ((RD) == X86_RAX) ? \
530     (_REXQrr(0, RD), _O_L (((OP) << 3) + 5 ,IM )) : \
531     (_REXQrr(0, RD), _Os_Mrm_sL (0x81 ,_b11,OP ,_r8(RD) ,IM )) )
532     #define _ALUQim(OP, IM, MD, MB, MI, MS) (_REXQrm(0, MB, MI), _Os_r_X_sL (0x81 ,OP ,MD,MB,MI,MS ,IM ))
533    
534     #define ADCBrr(RS, RD) _ALUBrr(X86_ADC, RS, RD)
535     #define ADCBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_ADC, MD, MB, MI, MS, RD)
536     #define ADCBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_ADC, RS, MD, MB, MI, MS)
537     #define ADCBir(IM, RD) _ALUBir(X86_ADC, IM, RD)
538     #define ADCBim(IM, MD, MB, MI, MS) _ALUBim(X86_ADC, IM, MD, MB, MI, MS)
539    
540     #define ADCWrr(RS, RD) _ALUWrr(X86_ADC, RS, RD)
541     #define ADCWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_ADC, MD, MB, MI, MS, RD)
542     #define ADCWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_ADC, RS, MD, MB, MI, MS)
543     #define ADCWir(IM, RD) _ALUWir(X86_ADC, IM, RD)
544     #define ADCWim(IM, MD, MB, MI, MS) _ALUWim(X86_ADC, IM, MD, MB, MI, MS)
545    
546     #define ADCLrr(RS, RD) _ALULrr(X86_ADC, RS, RD)
547     #define ADCLmr(MD, MB, MI, MS, RD) _ALULmr(X86_ADC, MD, MB, MI, MS, RD)
548     #define ADCLrm(RS, MD, MB, MI, MS) _ALULrm(X86_ADC, RS, MD, MB, MI, MS)
549     #define ADCLir(IM, RD) _ALULir(X86_ADC, IM, RD)
550     #define ADCLim(IM, MD, MB, MI, MS) _ALULim(X86_ADC, IM, MD, MB, MI, MS)
551    
552     #define ADCQrr(RS, RD) _ALUQrr(X86_ADC, RS, RD)
553     #define ADCQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_ADC, MD, MB, MI, MS, RD)
554     #define ADCQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_ADC, RS, MD, MB, MI, MS)
555     #define ADCQir(IM, RD) _ALUQir(X86_ADC, IM, RD)
556     #define ADCQim(IM, MD, MB, MI, MS) _ALUQim(X86_ADC, IM, MD, MB, MI, MS)
557    
558     #define ADDBrr(RS, RD) _ALUBrr(X86_ADD, RS, RD)
559     #define ADDBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_ADD, MD, MB, MI, MS, RD)
560     #define ADDBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_ADD, RS, MD, MB, MI, MS)
561     #define ADDBir(IM, RD) _ALUBir(X86_ADD, IM, RD)
562     #define ADDBim(IM, MD, MB, MI, MS) _ALUBim(X86_ADD, IM, MD, MB, MI, MS)
563    
564     #define ADDWrr(RS, RD) _ALUWrr(X86_ADD, RS, RD)
565     #define ADDWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_ADD, MD, MB, MI, MS, RD)
566     #define ADDWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_ADD, RS, MD, MB, MI, MS)
567     #define ADDWir(IM, RD) _ALUWir(X86_ADD, IM, RD)
568     #define ADDWim(IM, MD, MB, MI, MS) _ALUWim(X86_ADD, IM, MD, MB, MI, MS)
569    
570     #define ADDLrr(RS, RD) _ALULrr(X86_ADD, RS, RD)
571     #define ADDLmr(MD, MB, MI, MS, RD) _ALULmr(X86_ADD, MD, MB, MI, MS, RD)
572     #define ADDLrm(RS, MD, MB, MI, MS) _ALULrm(X86_ADD, RS, MD, MB, MI, MS)
573     #define ADDLir(IM, RD) _ALULir(X86_ADD, IM, RD)
574     #define ADDLim(IM, MD, MB, MI, MS) _ALULim(X86_ADD, IM, MD, MB, MI, MS)
575    
576     #define ADDQrr(RS, RD) _ALUQrr(X86_ADD, RS, RD)
577     #define ADDQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_ADD, MD, MB, MI, MS, RD)
578     #define ADDQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_ADD, RS, MD, MB, MI, MS)
579     #define ADDQir(IM, RD) _ALUQir(X86_ADD, IM, RD)
580     #define ADDQim(IM, MD, MB, MI, MS) _ALUQim(X86_ADD, IM, MD, MB, MI, MS)
581    
582     #define ANDBrr(RS, RD) _ALUBrr(X86_AND, RS, RD)
583     #define ANDBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_AND, MD, MB, MI, MS, RD)
584     #define ANDBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_AND, RS, MD, MB, MI, MS)
585     #define ANDBir(IM, RD) _ALUBir(X86_AND, IM, RD)
586     #define ANDBim(IM, MD, MB, MI, MS) _ALUBim(X86_AND, IM, MD, MB, MI, MS)
587    
588     #define ANDWrr(RS, RD) _ALUWrr(X86_AND, RS, RD)
589     #define ANDWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_AND, MD, MB, MI, MS, RD)
590     #define ANDWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_AND, RS, MD, MB, MI, MS)
591     #define ANDWir(IM, RD) _ALUWir(X86_AND, IM, RD)
592     #define ANDWim(IM, MD, MB, MI, MS) _ALUWim(X86_AND, IM, MD, MB, MI, MS)
593    
594     #define ANDLrr(RS, RD) _ALULrr(X86_AND, RS, RD)
595     #define ANDLmr(MD, MB, MI, MS, RD) _ALULmr(X86_AND, MD, MB, MI, MS, RD)
596     #define ANDLrm(RS, MD, MB, MI, MS) _ALULrm(X86_AND, RS, MD, MB, MI, MS)
597     #define ANDLir(IM, RD) _ALULir(X86_AND, IM, RD)
598     #define ANDLim(IM, MD, MB, MI, MS) _ALULim(X86_AND, IM, MD, MB, MI, MS)
599    
600     #define ANDQrr(RS, RD) _ALUQrr(X86_AND, RS, RD)
601     #define ANDQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_AND, MD, MB, MI, MS, RD)
602     #define ANDQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_AND, RS, MD, MB, MI, MS)
603     #define ANDQir(IM, RD) _ALUQir(X86_AND, IM, RD)
604     #define ANDQim(IM, MD, MB, MI, MS) _ALUQim(X86_AND, IM, MD, MB, MI, MS)
605    
606     #define CMPBrr(RS, RD) _ALUBrr(X86_CMP, RS, RD)
607     #define CMPBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_CMP, MD, MB, MI, MS, RD)
608     #define CMPBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_CMP, RS, MD, MB, MI, MS)
609     #define CMPBir(IM, RD) _ALUBir(X86_CMP, IM, RD)
610     #define CMPBim(IM, MD, MB, MI, MS) _ALUBim(X86_CMP, IM, MD, MB, MI, MS)
611    
612     #define CMPWrr(RS, RD) _ALUWrr(X86_CMP, RS, RD)
613     #define CMPWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_CMP, MD, MB, MI, MS, RD)
614     #define CMPWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_CMP, RS, MD, MB, MI, MS)
615     #define CMPWir(IM, RD) _ALUWir(X86_CMP, IM, RD)
616     #define CMPWim(IM, MD, MB, MI, MS) _ALUWim(X86_CMP, IM, MD, MB, MI, MS)
617    
618     #define CMPLrr(RS, RD) _ALULrr(X86_CMP, RS, RD)
619     #define CMPLmr(MD, MB, MI, MS, RD) _ALULmr(X86_CMP, MD, MB, MI, MS, RD)
620     #define CMPLrm(RS, MD, MB, MI, MS) _ALULrm(X86_CMP, RS, MD, MB, MI, MS)
621     #define CMPLir(IM, RD) _ALULir(X86_CMP, IM, RD)
622     #define CMPLim(IM, MD, MB, MI, MS) _ALULim(X86_CMP, IM, MD, MB, MI, MS)
623    
624     #define CMPQrr(RS, RD) _ALUQrr(X86_CMP, RS, RD)
625     #define CMPQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_CMP, MD, MB, MI, MS, RD)
626     #define CMPQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_CMP, RS, MD, MB, MI, MS)
627     #define CMPQir(IM, RD) _ALUQir(X86_CMP, IM, RD)
628     #define CMPQim(IM, MD, MB, MI, MS) _ALUQim(X86_CMP, IM, MD, MB, MI, MS)
629    
630     #define ORBrr(RS, RD) _ALUBrr(X86_OR, RS, RD)
631     #define ORBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_OR, MD, MB, MI, MS, RD)
632     #define ORBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_OR, RS, MD, MB, MI, MS)
633     #define ORBir(IM, RD) _ALUBir(X86_OR, IM, RD)
634     #define ORBim(IM, MD, MB, MI, MS) _ALUBim(X86_OR, IM, MD, MB, MI, MS)
635    
636     #define ORWrr(RS, RD) _ALUWrr(X86_OR, RS, RD)
637     #define ORWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_OR, MD, MB, MI, MS, RD)
638     #define ORWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_OR, RS, MD, MB, MI, MS)
639     #define ORWir(IM, RD) _ALUWir(X86_OR, IM, RD)
640     #define ORWim(IM, MD, MB, MI, MS) _ALUWim(X86_OR, IM, MD, MB, MI, MS)
641    
642     #define ORLrr(RS, RD) _ALULrr(X86_OR, RS, RD)
643     #define ORLmr(MD, MB, MI, MS, RD) _ALULmr(X86_OR, MD, MB, MI, MS, RD)
644     #define ORLrm(RS, MD, MB, MI, MS) _ALULrm(X86_OR, RS, MD, MB, MI, MS)
645     #define ORLir(IM, RD) _ALULir(X86_OR, IM, RD)
646     #define ORLim(IM, MD, MB, MI, MS) _ALULim(X86_OR, IM, MD, MB, MI, MS)
647    
648     #define ORQrr(RS, RD) _ALUQrr(X86_OR, RS, RD)
649     #define ORQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_OR, MD, MB, MI, MS, RD)
650     #define ORQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_OR, RS, MD, MB, MI, MS)
651     #define ORQir(IM, RD) _ALUQir(X86_OR, IM, RD)
652     #define ORQim(IM, MD, MB, MI, MS) _ALUQim(X86_OR, IM, MD, MB, MI, MS)
653    
654     #define SBBBrr(RS, RD) _ALUBrr(X86_SBB, RS, RD)
655     #define SBBBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_SBB, MD, MB, MI, MS, RD)
656     #define SBBBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_SBB, RS, MD, MB, MI, MS)
657     #define SBBBir(IM, RD) _ALUBir(X86_SBB, IM, RD)
658     #define SBBBim(IM, MD, MB, MI, MS) _ALUBim(X86_SBB, IM, MD, MB, MI, MS)
659    
660     #define SBBWrr(RS, RD) _ALUWrr(X86_SBB, RS, RD)
661     #define SBBWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_SBB, MD, MB, MI, MS, RD)
662     #define SBBWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_SBB, RS, MD, MB, MI, MS)
663     #define SBBWir(IM, RD) _ALUWir(X86_SBB, IM, RD)
664     #define SBBWim(IM, MD, MB, MI, MS) _ALUWim(X86_SBB, IM, MD, MB, MI, MS)
665    
666     #define SBBLrr(RS, RD) _ALULrr(X86_SBB, RS, RD)
667     #define SBBLmr(MD, MB, MI, MS, RD) _ALULmr(X86_SBB, MD, MB, MI, MS, RD)
668     #define SBBLrm(RS, MD, MB, MI, MS) _ALULrm(X86_SBB, RS, MD, MB, MI, MS)
669     #define SBBLir(IM, RD) _ALULir(X86_SBB, IM, RD)
670     #define SBBLim(IM, MD, MB, MI, MS) _ALULim(X86_SBB, IM, MD, MB, MI, MS)
671    
672     #define SBBQrr(RS, RD) _ALUQrr(X86_SBB, RS, RD)
673     #define SBBQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_SBB, MD, MB, MI, MS, RD)
674     #define SBBQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_SBB, RS, MD, MB, MI, MS)
675     #define SBBQir(IM, RD) _ALUQir(X86_SBB, IM, RD)
676     #define SBBQim(IM, MD, MB, MI, MS) _ALUQim(X86_SBB, IM, MD, MB, MI, MS)
677    
678     #define SUBBrr(RS, RD) _ALUBrr(X86_SUB, RS, RD)
679     #define SUBBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_SUB, MD, MB, MI, MS, RD)
680     #define SUBBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_SUB, RS, MD, MB, MI, MS)
681     #define SUBBir(IM, RD) _ALUBir(X86_SUB, IM, RD)
682     #define SUBBim(IM, MD, MB, MI, MS) _ALUBim(X86_SUB, IM, MD, MB, MI, MS)
683    
684     #define SUBWrr(RS, RD) _ALUWrr(X86_SUB, RS, RD)
685     #define SUBWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_SUB, MD, MB, MI, MS, RD)
686     #define SUBWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_SUB, RS, MD, MB, MI, MS)
687     #define SUBWir(IM, RD) _ALUWir(X86_SUB, IM, RD)
688     #define SUBWim(IM, MD, MB, MI, MS) _ALUWim(X86_SUB, IM, MD, MB, MI, MS)
689    
690     #define SUBLrr(RS, RD) _ALULrr(X86_SUB, RS, RD)
691     #define SUBLmr(MD, MB, MI, MS, RD) _ALULmr(X86_SUB, MD, MB, MI, MS, RD)
692     #define SUBLrm(RS, MD, MB, MI, MS) _ALULrm(X86_SUB, RS, MD, MB, MI, MS)
693     #define SUBLir(IM, RD) _ALULir(X86_SUB, IM, RD)
694     #define SUBLim(IM, MD, MB, MI, MS) _ALULim(X86_SUB, IM, MD, MB, MI, MS)
695    
696     #define SUBQrr(RS, RD) _ALUQrr(X86_SUB, RS, RD)
697     #define SUBQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_SUB, MD, MB, MI, MS, RD)
698     #define SUBQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_SUB, RS, MD, MB, MI, MS)
699     #define SUBQir(IM, RD) _ALUQir(X86_SUB, IM, RD)
700     #define SUBQim(IM, MD, MB, MI, MS) _ALUQim(X86_SUB, IM, MD, MB, MI, MS)
701    
702     #define XORBrr(RS, RD) _ALUBrr(X86_XOR, RS, RD)
703     #define XORBmr(MD, MB, MI, MS, RD) _ALUBmr(X86_XOR, MD, MB, MI, MS, RD)
704     #define XORBrm(RS, MD, MB, MI, MS) _ALUBrm(X86_XOR, RS, MD, MB, MI, MS)
705     #define XORBir(IM, RD) _ALUBir(X86_XOR, IM, RD)
706     #define XORBim(IM, MD, MB, MI, MS) _ALUBim(X86_XOR, IM, MD, MB, MI, MS)
707    
708     #define XORWrr(RS, RD) _ALUWrr(X86_XOR, RS, RD)
709     #define XORWmr(MD, MB, MI, MS, RD) _ALUWmr(X86_XOR, MD, MB, MI, MS, RD)
710     #define XORWrm(RS, MD, MB, MI, MS) _ALUWrm(X86_XOR, RS, MD, MB, MI, MS)
711     #define XORWir(IM, RD) _ALUWir(X86_XOR, IM, RD)
712     #define XORWim(IM, MD, MB, MI, MS) _ALUWim(X86_XOR, IM, MD, MB, MI, MS)
713    
714     #define XORLrr(RS, RD) _ALULrr(X86_XOR, RS, RD)
715     #define XORLmr(MD, MB, MI, MS, RD) _ALULmr(X86_XOR, MD, MB, MI, MS, RD)
716     #define XORLrm(RS, MD, MB, MI, MS) _ALULrm(X86_XOR, RS, MD, MB, MI, MS)
717     #define XORLir(IM, RD) _ALULir(X86_XOR, IM, RD)
718     #define XORLim(IM, MD, MB, MI, MS) _ALULim(X86_XOR, IM, MD, MB, MI, MS)
719    
720     #define XORQrr(RS, RD) _ALUQrr(X86_XOR, RS, RD)
721     #define XORQmr(MD, MB, MI, MS, RD) _ALUQmr(X86_XOR, MD, MB, MI, MS, RD)
722     #define XORQrm(RS, MD, MB, MI, MS) _ALUQrm(X86_XOR, RS, MD, MB, MI, MS)
723     #define XORQir(IM, RD) _ALUQir(X86_XOR, IM, RD)
724     #define XORQim(IM, MD, MB, MI, MS) _ALUQim(X86_XOR, IM, MD, MB, MI, MS)
725    
726    
727     /* --- Shift/Rotate instructions ------------------------------------------- */
728    
729 gbeauche 1.2 enum {
730 gbeauche 1.1 X86_ROL = 0,
731     X86_ROR = 1,
732     X86_RCL = 2,
733     X86_RCR = 3,
734     X86_SHL = 4,
735     X86_SHR = 5,
736     X86_SAR = 7,
737 gbeauche 1.2 };
738 gbeauche 1.1
739     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
740    
741     #define _ROTSHIBir(OP,IM,RD) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
742     (_REXBrr(0, RD), _O_Mrm (0xd0 ,_b11,OP,_r1(RD) )) : \
743     (_REXBrr(0, RD), _O_Mrm_B (0xc0 ,_b11,OP,_r1(RD) ,_u8(IM))) )
744     #define _ROTSHIBim(OP,IM,MD,MB,MI,MS) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
745     (_REXBrm(0, MB, MI), _O_r_X (0xd0 ,OP ,MD,MB,MI,MS )) : \
746     (_REXBrm(0, MB, MI), _O_r_X_B (0xc0 ,OP ,MD,MB,MI,MS ,_u8(IM))) )
747     #define _ROTSHIBrr(OP,RS,RD) (((RS) == X86_CL) ? \
748     (_REXBrr(RS, RD), _O_Mrm (0xd2 ,_b11,OP,_r1(RD) )) : \
749     x86_emit_failure("source register must be CL" ) )
750     #define _ROTSHIBrm(OP,RS,MD,MB,MI,MS) (((RS) == X86_CL) ? \
751     (_REXBrm(RS, MB, MI), _O_r_X (0xd2 ,OP ,MD,MB,MI,MS )) : \
752     x86_emit_failure("source register must be CL" ) )
753    
754     #define _ROTSHIWir(OP,IM,RD) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
755     (_d16(), _REXLrr(0, RD), _O_Mrm (0xd1 ,_b11,OP,_r2(RD) )) : \
756     (_d16(), _REXLrr(0, RD), _O_Mrm_B (0xc1 ,_b11,OP,_r2(RD) ,_u8(IM))) )
757     #define _ROTSHIWim(OP,IM,MD,MB,MI,MS) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
758     (_d16(), _REXLrm(0, MB, MI), _O_r_X (0xd1 ,OP ,MD,MB,MI,MS )) : \
759     (_d16(), _REXLrm(0, MB, MI), _O_r_X_B (0xc1 ,OP ,MD,MB,MI,MS ,_u8(IM))) )
760     #define _ROTSHIWrr(OP,RS,RD) (((RS) == X86_CL) ? \
761     (_d16(), _REXLrr(RS, RD), _O_Mrm (0xd3 ,_b11,OP,_r2(RD) )) : \
762     x86_emit_failure("source register must be CL" ) )
763     #define _ROTSHIWrm(OP,RS,MD,MB,MI,MS) (((RS) == X86_CL) ? \
764     (_d16(), _REXLrm(RS, MB, MI), _O_r_X (0xd3 ,OP ,MD,MB,MI,MS )) : \
765     x86_emit_failure("source register must be CL" ) )
766    
767     #define _ROTSHILir(OP,IM,RD) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
768     (_REXLrr(0, RD), _O_Mrm (0xd1 ,_b11,OP,_r4(RD) )) : \
769     (_REXLrr(0, RD), _O_Mrm_B (0xc1 ,_b11,OP,_r4(RD) ,_u8(IM))) )
770     #define _ROTSHILim(OP,IM,MD,MB,MI,MS) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
771     (_REXLrm(0, MB, MI), _O_r_X (0xd1 ,OP ,MD,MB,MI,MS )) : \
772     (_REXLrm(0, MB, MI), _O_r_X_B (0xc1 ,OP ,MD,MB,MI,MS ,_u8(IM))) )
773     #define _ROTSHILrr(OP,RS,RD) (((RS) == X86_CL) ? \
774     (_REXLrr(RS, RD), _O_Mrm (0xd3 ,_b11,OP,_r4(RD) )) : \
775     x86_emit_failure("source register must be CL" ) )
776     #define _ROTSHILrm(OP,RS,MD,MB,MI,MS) (((RS) == X86_CL) ? \
777     (_REXLrm(RS, MB, MI), _O_r_X (0xd3 ,OP ,MD,MB,MI,MS )) : \
778     x86_emit_failure("source register must be CL" ) )
779    
780     #define _ROTSHIQir(OP,IM,RD) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
781     (_REXQrr(0, RD), _O_Mrm (0xd1 ,_b11,OP,_r8(RD) )) : \
782     (_REXQrr(0, RD), _O_Mrm_B (0xc1 ,_b11,OP,_r8(RD) ,_u8(IM))) )
783     #define _ROTSHIQim(OP,IM,MD,MB,MI,MS) (X86_OPTIMIZE_ROTSHI && ((IM) == 1) ? \
784     (_REXQrm(0, MB, MI), _O_r_X (0xd1 ,OP ,MD,MB,MI,MS )) : \
785     (_REXQrm(0, MB, MI), _O_r_X_B (0xc1 ,OP ,MD,MB,MI,MS ,_u8(IM))) )
786     #define _ROTSHIQrr(OP,RS,RD) (((RS) == X86_CL) ? \
787     (_REXQrr(RS, RD), _O_Mrm (0xd3 ,_b11,OP,_r8(RD) )) : \
788     x86_emit_failure("source register must be CL" ) )
789     #define _ROTSHIQrm(OP,RS,MD,MB,MI,MS) (((RS) == X86_CL) ? \
790     (_REXQrm(RS, MB, MI), _O_r_X (0xd3 ,OP ,MD,MB,MI,MS )) : \
791     x86_emit_failure("source register must be CL" ) )
792    
793     #define ROLBir(IM, RD) _ROTSHIBir(X86_ROL, IM, RD)
794     #define ROLBim(IM, MD, MB, MI, MS) _ROTSHIBim(X86_ROL, IM, MD, MB, MI, MS)
795     #define ROLBrr(RS, RD) _ROTSHIBrr(X86_ROL, RS, RD)
796     #define ROLBrm(RS, MD, MB, MI, MS) _ROTSHIBrm(X86_ROL, RS, MD, MB, MI, MS)
797    
798     #define ROLWir(IM, RD) _ROTSHIWir(X86_ROL, IM, RD)
799     #define ROLWim(IM, MD, MB, MI, MS) _ROTSHIWim(X86_ROL, IM, MD, MB, MI, MS)
800     #define ROLWrr(RS, RD) _ROTSHIWrr(X86_ROL, RS, RD)
801     #define ROLWrm(RS, MD, MB, MI, MS) _ROTSHIWrm(X86_ROL, RS, MD, MB, MI, MS)
802    
803     #define ROLLir(IM, RD) _ROTSHILir(X86_ROL, IM, RD)
804     #define ROLLim(IM, MD, MB, MI, MS) _ROTSHILim(X86_ROL, IM, MD, MB, MI, MS)
805     #define ROLLrr(RS, RD) _ROTSHILrr(X86_ROL, RS, RD)
806     #define ROLLrm(RS, MD, MB, MI, MS) _ROTSHILrm(X86_ROL, RS, MD, MB, MI, MS)
807    
808     #define ROLQir(IM, RD) _ROTSHIQir(X86_ROL, IM, RD)
809     #define ROLQim(IM, MD, MB, MI, MS) _ROTSHIQim(X86_ROL, IM, MD, MB, MI, MS)
810     #define ROLQrr(RS, RD) _ROTSHIQrr(X86_ROL, RS, RD)
811     #define ROLQrm(RS, MD, MB, MI, MS) _ROTSHIQrm(X86_ROL, RS, MD, MB, MI, MS)
812    
813     #define RORBir(IM, RD) _ROTSHIBir(X86_ROR, IM, RD)
814     #define RORBim(IM, MD, MB, MI, MS) _ROTSHIBim(X86_ROR, IM, MD, MB, MI, MS)
815     #define RORBrr(RS, RD) _ROTSHIBrr(X86_ROR, RS, RD)
816     #define RORBrm(RS, MD, MB, MI, MS) _ROTSHIBrm(X86_ROR, RS, MD, MB, MI, MS)
817    
818     #define RORWir(IM, RD) _ROTSHIWir(X86_ROR, IM, RD)
819     #define RORWim(IM, MD, MB, MI, MS) _ROTSHIWim(X86_ROR, IM, MD, MB, MI, MS)
820     #define RORWrr(RS, RD) _ROTSHIWrr(X86_ROR, RS, RD)
821     #define RORWrm(RS, MD, MB, MI, MS) _ROTSHIWrm(X86_ROR, RS, MD, MB, MI, MS)
822    
823     #define RORLir(IM, RD) _ROTSHILir(X86_ROR, IM, RD)
824     #define RORLim(IM, MD, MB, MI, MS) _ROTSHILim(X86_ROR, IM, MD, MB, MI, MS)
825     #define RORLrr(RS, RD) _ROTSHILrr(X86_ROR, RS, RD)
826     #define RORLrm(RS, MD, MB, MI, MS) _ROTSHILrm(X86_ROR, RS, MD, MB, MI, MS)
827    
828     #define RORQir(IM, RD) _ROTSHIQir(X86_ROR, IM, RD)
829     #define RORQim(IM, MD, MB, MI, MS) _ROTSHIQim(X86_ROR, IM, MD, MB, MI, MS)
830     #define RORQrr(RS, RD) _ROTSHIQrr(X86_ROR, RS, RD)
831     #define RORQrm(RS, MD, MB, MI, MS) _ROTSHIQrm(X86_ROR, RS, MD, MB, MI, MS)
832    
833     #define RCLBir(IM, RD) _ROTSHIBir(X86_RCL, IM, RD)
834     #define RCLBim(IM, MD, MB, MI, MS) _ROTSHIBim(X86_RCL, IM, MD, MB, MI, MS)
835     #define RCLBrr(RS, RD) _ROTSHIBrr(X86_RCL, RS, RD)
836     #define RCLBrm(RS, MD, MB, MI, MS) _ROTSHIBrm(X86_RCL, RS, MD, MB, MI, MS)
837    
838     #define RCLWir(IM, RD) _ROTSHIWir(X86_RCL, IM, RD)
839     #define RCLWim(IM, MD, MB, MI, MS) _ROTSHIWim(X86_RCL, IM, MD, MB, MI, MS)
840     #define RCLWrr(RS, RD) _ROTSHIWrr(X86_RCL, RS, RD)
841     #define RCLWrm(RS, MD, MB, MI, MS) _ROTSHIWrm(X86_RCL, RS, MD, MB, MI, MS)
842    
843     #define RCLLir(IM, RD) _ROTSHILir(X86_RCL, IM, RD)
844     #define RCLLim(IM, MD, MB, MI, MS) _ROTSHILim(X86_RCL, IM, MD, MB, MI, MS)
845     #define RCLLrr(RS, RD) _ROTSHILrr(X86_RCL, RS, RD)
846     #define RCLLrm(RS, MD, MB, MI, MS) _ROTSHILrm(X86_RCL, RS, MD, MB, MI, MS)
847    
848     #define RCLQir(IM, RD) _ROTSHIQir(X86_RCL, IM, RD)
849     #define RCLQim(IM, MD, MB, MI, MS) _ROTSHIQim(X86_RCL, IM, MD, MB, MI, MS)
850     #define RCLQrr(RS, RD) _ROTSHIQrr(X86_RCL, RS, RD)
851     #define RCLQrm(RS, MD, MB, MI, MS) _ROTSHIQrm(X86_RCL, RS, MD, MB, MI, MS)
852    
853     #define RCRBir(IM, RD) _ROTSHIBir(X86_RCR, IM, RD)
854     #define RCRBim(IM, MD, MB, MI, MS) _ROTSHIBim(X86_RCR, IM, MD, MB, MI, MS)
855     #define RCRBrr(RS, RD) _ROTSHIBrr(X86_RCR, RS, RD)
856     #define RCRBrm(RS, MD, MB, MI, MS) _ROTSHIBrm(X86_RCR, RS, MD, MB, MI, MS)
857    
858     #define RCRWir(IM, RD) _ROTSHIWir(X86_RCR, IM, RD)
859     #define RCRWim(IM, MD, MB, MI, MS) _ROTSHIWim(X86_RCR, IM, MD, MB, MI, MS)
860     #define RCRWrr(RS, RD) _ROTSHIWrr(X86_RCR, RS, RD)
861     #define RCRWrm(RS, MD, MB, MI, MS) _ROTSHIWrm(X86_RCR, RS, MD, MB, MI, MS)
862    
863     #define RCRLir(IM, RD) _ROTSHILir(X86_RCR, IM, RD)
864     #define RCRLim(IM, MD, MB, MI, MS) _ROTSHILim(X86_RCR, IM, MD, MB, MI, MS)
865     #define RCRLrr(RS, RD) _ROTSHILrr(X86_RCR, RS, RD)
866     #define RCRLrm(RS, MD, MB, MI, MS) _ROTSHILrm(X86_RCR, RS, MD, MB, MI, MS)
867    
868     #define RCRQir(IM, RD) _ROTSHIQir(X86_RCR, IM, RD)
869     #define RCRQim(IM, MD, MB, MI, MS) _ROTSHIQim(X86_RCR, IM, MD, MB, MI, MS)
870     #define RCRQrr(RS, RD) _ROTSHIQrr(X86_RCR, RS, RD)
871     #define RCRQrm(RS, MD, MB, MI, MS) _ROTSHIQrm(X86_RCR, RS, MD, MB, MI, MS)
872    
873     #define SHLBir(IM, RD) _ROTSHIBir(X86_SHL, IM, RD)
874     #define SHLBim(IM, MD, MB, MI, MS) _ROTSHIBim(X86_SHL, IM, MD, MB, MI, MS)
875     #define SHLBrr(RS, RD) _ROTSHIBrr(X86_SHL, RS, RD)
876     #define SHLBrm(RS, MD, MB, MI, MS) _ROTSHIBrm(X86_SHL, RS, MD, MB, MI, MS)
877    
878     #define SHLWir(IM, RD) _ROTSHIWir(X86_SHL, IM, RD)
879     #define SHLWim(IM, MD, MB, MI, MS) _ROTSHIWim(X86_SHL, IM, MD, MB, MI, MS)
880     #define SHLWrr(RS, RD) _ROTSHIWrr(X86_SHL, RS, RD)
881     #define SHLWrm(RS, MD, MB, MI, MS) _ROTSHIWrm(X86_SHL, RS, MD, MB, MI, MS)
882    
883     #define SHLLir(IM, RD) _ROTSHILir(X86_SHL, IM, RD)
884     #define SHLLim(IM, MD, MB, MI, MS) _ROTSHILim(X86_SHL, IM, MD, MB, MI, MS)
885     #define SHLLrr(RS, RD) _ROTSHILrr(X86_SHL, RS, RD)
886     #define SHLLrm(RS, MD, MB, MI, MS) _ROTSHILrm(X86_SHL, RS, MD, MB, MI, MS)
887    
888     #define SHLQir(IM, RD) _ROTSHIQir(X86_SHL, IM, RD)
889     #define SHLQim(IM, MD, MB, MI, MS) _ROTSHIQim(X86_SHL, IM, MD, MB, MI, MS)
890     #define SHLQrr(RS, RD) _ROTSHIQrr(X86_SHL, RS, RD)
891     #define SHLQrm(RS, MD, MB, MI, MS) _ROTSHIQrm(X86_SHL, RS, MD, MB, MI, MS)
892    
893     #define SHRBir(IM, RD) _ROTSHIBir(X86_SHR, IM, RD)
894     #define SHRBim(IM, MD, MB, MI, MS) _ROTSHIBim(X86_SHR, IM, MD, MB, MI, MS)
895     #define SHRBrr(RS, RD) _ROTSHIBrr(X86_SHR, RS, RD)
896     #define SHRBrm(RS, MD, MB, MI, MS) _ROTSHIBrm(X86_SHR, RS, MD, MB, MI, MS)
897    
898     #define SHRWir(IM, RD) _ROTSHIWir(X86_SHR, IM, RD)
899     #define SHRWim(IM, MD, MB, MI, MS) _ROTSHIWim(X86_SHR, IM, MD, MB, MI, MS)
900     #define SHRWrr(RS, RD) _ROTSHIWrr(X86_SHR, RS, RD)
901     #define SHRWrm(RS, MD, MB, MI, MS) _ROTSHIWrm(X86_SHR, RS, MD, MB, MI, MS)
902    
903     #define SHRLir(IM, RD) _ROTSHILir(X86_SHR, IM, RD)
904     #define SHRLim(IM, MD, MB, MI, MS) _ROTSHILim(X86_SHR, IM, MD, MB, MI, MS)
905     #define SHRLrr(RS, RD) _ROTSHILrr(X86_SHR, RS, RD)
906     #define SHRLrm(RS, MD, MB, MI, MS) _ROTSHILrm(X86_SHR, RS, MD, MB, MI, MS)
907    
908     #define SHRQir(IM, RD) _ROTSHIQir(X86_SHR, IM, RD)
909     #define SHRQim(IM, MD, MB, MI, MS) _ROTSHIQim(X86_SHR, IM, MD, MB, MI, MS)
910     #define SHRQrr(RS, RD) _ROTSHIQrr(X86_SHR, RS, RD)
911     #define SHRQrm(RS, MD, MB, MI, MS) _ROTSHIQrm(X86_SHR, RS, MD, MB, MI, MS)
912    
913     #define SALBir SHLBir
914     #define SALBim SHLBim
915     #define SALBrr SHLBrr
916     #define SALBrm SHLBrm
917    
918     #define SALWir SHLWir
919     #define SALWim SHLWim
920     #define SALWrr SHLWrr
921     #define SALWrm SHLWrm
922    
923     #define SALLir SHLLir
924     #define SALLim SHLLim
925     #define SALLrr SHLLrr
926     #define SALLrm SHLLrm
927    
928     #define SALQir SHLQir
929     #define SALQim SHLQim
930     #define SALQrr SHLQrr
931     #define SALQrm SHLQrm
932    
933     #define SARBir(IM, RD) _ROTSHIBir(X86_SAR, IM, RD)
934     #define SARBim(IM, MD, MB, MI, MS) _ROTSHIBim(X86_SAR, IM, MD, MB, MI, MS)
935     #define SARBrr(RS, RD) _ROTSHIBrr(X86_SAR, RS, RD)
936     #define SARBrm(RS, MD, MB, MI, MS) _ROTSHIBrm(X86_SAR, RS, MD, MB, MI, MS)
937    
938     #define SARWir(IM, RD) _ROTSHIWir(X86_SAR, IM, RD)
939     #define SARWim(IM, MD, MB, MI, MS) _ROTSHIWim(X86_SAR, IM, MD, MB, MI, MS)
940     #define SARWrr(RS, RD) _ROTSHIWrr(X86_SAR, RS, RD)
941     #define SARWrm(RS, MD, MB, MI, MS) _ROTSHIWrm(X86_SAR, RS, MD, MB, MI, MS)
942    
943     #define SARLir(IM, RD) _ROTSHILir(X86_SAR, IM, RD)
944     #define SARLim(IM, MD, MB, MI, MS) _ROTSHILim(X86_SAR, IM, MD, MB, MI, MS)
945     #define SARLrr(RS, RD) _ROTSHILrr(X86_SAR, RS, RD)
946     #define SARLrm(RS, MD, MB, MI, MS) _ROTSHILrm(X86_SAR, RS, MD, MB, MI, MS)
947    
948     #define SARQir(IM, RD) _ROTSHIQir(X86_SAR, IM, RD)
949     #define SARQim(IM, MD, MB, MI, MS) _ROTSHIQim(X86_SAR, IM, MD, MB, MI, MS)
950     #define SARQrr(RS, RD) _ROTSHIQrr(X86_SAR, RS, RD)
951     #define SARQrm(RS, MD, MB, MI, MS) _ROTSHIQrm(X86_SAR, RS, MD, MB, MI, MS)
952    
953    
954     /* --- Bit test instructions ----------------------------------------------- */
955    
956 gbeauche 1.2 enum {
957 gbeauche 1.1 X86_BT = 4,
958     X86_BTS = 5,
959     X86_BTR = 6,
960     X86_BTC = 7,
961 gbeauche 1.2 };
962 gbeauche 1.1
963     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
964    
965     #define _BTWir(OP, IM, RD) (_d16(), _REXLrr(0, RD), _OO_Mrm_B (0x0fba ,_b11,OP ,_r2(RD) ,_u8(IM)))
966     #define _BTWim(OP, IM, MD, MB, MI, MS) (_d16(), _REXLrm(0, MB, MI), _OO_r_X_B (0x0fba ,OP ,MD,MB,MI,MS ,_u8(IM)))
967     #define _BTWrr(OP, RS, RD) (_d16(), _REXLrr(RS, RD), _OO_Mrm (0x0f83|((OP)<<3),_b11,_r2(RS),_r2(RD) ))
968     #define _BTWrm(OP, RS, MD, MB, MI, MS) (_d16(), _REXLrm(RS, MB, MI), _OO_r_X (0x0f83|((OP)<<3) ,_r2(RS) ,MD,MB,MI,MS ))
969    
970     #define _BTLir(OP, IM, RD) (_REXLrr(0, RD), _OO_Mrm_B (0x0fba ,_b11,OP ,_r4(RD) ,_u8(IM)))
971     #define _BTLim(OP, IM, MD, MB, MI, MS) (_REXLrm(0, MB, MI), _OO_r_X_B (0x0fba ,OP ,MD,MB,MI,MS ,_u8(IM)))
972     #define _BTLrr(OP, RS, RD) (_REXLrr(RS, RD), _OO_Mrm (0x0f83|((OP)<<3),_b11,_r4(RS),_r4(RD) ))
973     #define _BTLrm(OP, RS, MD, MB, MI, MS) (_REXLrm(RS, MB, MI), _OO_r_X (0x0f83|((OP)<<3) ,_r4(RS) ,MD,MB,MI,MS ))
974    
975     #define _BTQir(OP, IM, RD) (_REXQrr(0, RD), _OO_Mrm_B (0x0fba ,_b11,OP ,_r8(RD) ,_u8(IM)))
976     #define _BTQim(OP, IM, MD, MB, MI, MS) (_REXQrm(0, MB, MI), _OO_r_X_B (0x0fba ,OP ,MD,MB,MI,MS ,_u8(IM)))
977     #define _BTQrr(OP, RS, RD) (_REXQrr(RS, RD), _OO_Mrm (0x0f83|((OP)<<3),_b11,_r8(RS),_r8(RD) ))
978     #define _BTQrm(OP, RS, MD, MB, MI, MS) (_REXQrm(RS, MB, MI), _OO_r_X (0x0f83|((OP)<<3) ,_r8(RS) ,MD,MB,MI,MS ))
979    
980     #define BTWir(IM, RD) _BTWir(X86_BT, IM, RD)
981     #define BTWim(IM, MD, MB, MI, MS) _BTWim(X86_BT, IM, MD, MI, MS)
982     #define BTWrr(RS, RD) _BTWrr(X86_BT, RS, RD)
983     #define BTWrm(RS, MD, MB, MI, MS) _BTWrm(X86_BT, RS, MD, MB, MI, MS)
984    
985     #define BTLir(IM, RD) _BTLir(X86_BT, IM, RD)
986     #define BTLim(IM, MD, MB, MI, MS) _BTLim(X86_BT, IM, MD, MB, MI, MS)
987     #define BTLrr(RS, RD) _BTLrr(X86_BT, RS, RD)
988     #define BTLrm(RS, MD, MB, MI, MS) _BTLrm(X86_BT, RS, MD, MB, MI, MS)
989    
990     #define BTQir(IM, RD) _BTQir(X86_BT, IM, RD)
991     #define BTQim(IM, MD, MB, MI, MS) _BTQim(X86_BT, IM, MD, MB, MI, MS)
992     #define BTQrr(RS, RD) _BTQrr(X86_BT, RS, RD)
993     #define BTQrm(RS, MD, MB, MI, MS) _BTQrm(X86_BT, RS, MD, MB, MI, MS)
994    
995     #define BTCWir(IM, RD) _BTWir(X86_BTC, IM, RD)
996     #define BTCWim(IM, MD, MB, MI, MS) _BTWim(X86_BTC, IM, MD, MI, MS)
997     #define BTCWrr(RS, RD) _BTWrr(X86_BTC, RS, RD)
998     #define BTCWrm(RS, MD, MB, MI, MS) _BTWrm(X86_BTC, RS, MD, MB, MI, MS)
999    
1000     #define BTCLir(IM, RD) _BTLir(X86_BTC, IM, RD)
1001     #define BTCLim(IM, MD, MB, MI, MS) _BTLim(X86_BTC, IM, MD, MB, MI, MS)
1002     #define BTCLrr(RS, RD) _BTLrr(X86_BTC, RS, RD)
1003     #define BTCLrm(RS, MD, MB, MI, MS) _BTLrm(X86_BTC, RS, MD, MB, MI, MS)
1004    
1005     #define BTCQir(IM, RD) _BTQir(X86_BTC, IM, RD)
1006     #define BTCQim(IM, MD, MB, MI, MS) _BTQim(X86_BTC, IM, MD, MB, MI, MS)
1007     #define BTCQrr(RS, RD) _BTQrr(X86_BTC, RS, RD)
1008     #define BTCQrm(RS, MD, MB, MI, MS) _BTQrm(X86_BTC, RS, MD, MB, MI, MS)
1009    
1010     #define BTRWir(IM, RD) _BTWir(X86_BTR, IM, RD)
1011     #define BTRWim(IM, MD, MB, MI, MS) _BTWim(X86_BTR, IM, MD, MI, MS)
1012     #define BTRWrr(RS, RD) _BTWrr(X86_BTR, RS, RD)
1013     #define BTRWrm(RS, MD, MB, MI, MS) _BTWrm(X86_BTR, RS, MD, MB, MI, MS)
1014    
1015     #define BTRLir(IM, RD) _BTLir(X86_BTR, IM, RD)
1016     #define BTRLim(IM, MD, MB, MI, MS) _BTLim(X86_BTR, IM, MD, MB, MI, MS)
1017     #define BTRLrr(RS, RD) _BTLrr(X86_BTR, RS, RD)
1018     #define BTRLrm(RS, MD, MB, MI, MS) _BTLrm(X86_BTR, RS, MD, MB, MI, MS)
1019    
1020     #define BTRQir(IM, RD) _BTQir(X86_BTR, IM, RD)
1021     #define BTRQim(IM, MD, MB, MI, MS) _BTQim(X86_BTR, IM, MD, MB, MI, MS)
1022     #define BTRQrr(RS, RD) _BTQrr(X86_BTR, RS, RD)
1023     #define BTRQrm(RS, MD, MB, MI, MS) _BTQrm(X86_BTR, RS, MD, MB, MI, MS)
1024    
1025     #define BTSWir(IM, RD) _BTWir(X86_BTS, IM, RD)
1026     #define BTSWim(IM, MD, MB, MI, MS) _BTWim(X86_BTS, IM, MD, MI, MS)
1027     #define BTSWrr(RS, RD) _BTWrr(X86_BTS, RS, RD)
1028     #define BTSWrm(RS, MD, MB, MI, MS) _BTWrm(X86_BTS, RS, MD, MB, MI, MS)
1029    
1030     #define BTSLir(IM, RD) _BTLir(X86_BTS, IM, RD)
1031     #define BTSLim(IM, MD, MB, MI, MS) _BTLim(X86_BTS, IM, MD, MB, MI, MS)
1032     #define BTSLrr(RS, RD) _BTLrr(X86_BTS, RS, RD)
1033     #define BTSLrm(RS, MD, MB, MI, MS) _BTLrm(X86_BTS, RS, MD, MB, MI, MS)
1034    
1035     #define BTSQir(IM, RD) _BTQir(X86_BTS, IM, RD)
1036     #define BTSQim(IM, MD, MB, MI, MS) _BTQim(X86_BTS, IM, MD, MB, MI, MS)
1037     #define BTSQrr(RS, RD) _BTQrr(X86_BTS, RS, RD)
1038     #define BTSQrm(RS, MD, MB, MI, MS) _BTQrm(X86_BTS, RS, MD, MB, MI, MS)
1039    
1040    
1041     /* --- Move instructions --------------------------------------------------- */
1042    
1043     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1044    
1045     #define MOVBrr(RS, RD) (_REXBrr(RS, RD), _O_Mrm (0x80 ,_b11,_r1(RS),_r1(RD) ))
1046     #define MOVBmr(MD, MB, MI, MS, RD) (_REXBmr(MB, MI, RD), _O_r_X (0x8a ,_r1(RD) ,MD,MB,MI,MS ))
1047     #define MOVBrm(RS, MD, MB, MI, MS) (_REXBrm(RS, MB, MI), _O_r_X (0x88 ,_r1(RS) ,MD,MB,MI,MS ))
1048     #define MOVBir(IM, R) (_REXBrr(0, R), _Or_B (0xb0,_r1(R) ,_su8(IM)))
1049     #define MOVBim(IM, MD, MB, MI, MS) (_REXBrm(0, MB, MI), _O_X_B (0xc6 ,MD,MB,MI,MS ,_su8(IM)))
1050    
1051     #define MOVWrr(RS, RD) (_d16(), _REXLrr(RS, RD), _O_Mrm (0x89 ,_b11,_r2(RS),_r2(RD) ))
1052     #define MOVWmr(MD, MB, MI, MS, RD) (_d16(), _REXLmr(MD, MI, RD), _O_r_X (0x8b ,_r2(RD) ,MD,MB,MI,MS ))
1053     #define MOVWrm(RS, MD, MB, MI, MS) (_d16(), _REXLrm(RS, MD, MI), _O_r_X (0x89 ,_r2(RS) ,MD,MB,MI,MS ))
1054     #define MOVWir(IM, R) (_d16(), _REXLrr(0, R), _Or_W (0xb8,_r2(R) ,_su16(IM)))
1055     #define MOVWim(IM, MD, MB, MI, MS) (_d16(), _REXLrm(0, MD, MI), _O_X_W (0xc7 ,MD,MB,MI,MS ,_su16(IM)))
1056    
1057     #define MOVLrr(RS, RD) (_REXLrr(RS, RD), _O_Mrm (0x89 ,_b11,_r4(RS),_r4(RD) ))
1058     #define MOVLmr(MD, MB, MI, MS, RD) (_REXLmr(MB, MI, RD), _O_r_X (0x8b ,_r4(RD) ,MD,MB,MI,MS ))
1059     #define MOVLrm(RS, MD, MB, MI, MS) (_REXLrm(RS, MB, MI), _O_r_X (0x89 ,_r4(RS) ,MD,MB,MI,MS ))
1060     #define MOVLir(IM, R) (_REXLrr(0, R), _Or_L (0xb8,_r4(R) ,IM ))
1061     #define MOVLim(IM, MD, MB, MI, MS) (_REXLrm(0, MB, MI), _O_X_L (0xc7 ,MD,MB,MI,MS ,IM ))
1062    
1063     #define MOVQrr(RS, RD) (_REXQrr(RS, RD), _O_Mrm (0x89 ,_b11,_r8(RS),_r8(RD) ))
1064     #define MOVQmr(MD, MB, MI, MS, RD) (_REXQmr(MB, MI, RD), _O_r_X (0x8b ,_r8(RD) ,MD,MB,MI,MS ))
1065     #define MOVQrm(RS, MD, MB, MI, MS) (_REXQrm(RS, MB, MI), _O_r_X (0x89 ,_r8(RS) ,MD,MB,MI,MS ))
1066     #define MOVQir(IM, R) (_REXQrr(0, R), _Or_L (0xb8,_r8(R) ,IM ))
1067     #define MOVQim(IM, MD, MB, MI, MS) (_REXQrm(0, MB, MI), _O_X_L (0xc7 ,MD,MB,MI,MS ,IM ))
1068    
1069    
1070     /* --- Unary and Multiply/Divide instructions ------------------------------ */
1071    
1072 gbeauche 1.2 enum {
1073 gbeauche 1.1 X86_NOT = 2,
1074     X86_NEG = 3,
1075     X86_MUL = 4,
1076     X86_IMUL = 5,
1077     X86_DIV = 6,
1078     X86_IDIV = 7,
1079 gbeauche 1.2 };
1080 gbeauche 1.1
1081     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1082    
1083     #define _UNARYBr(OP, RS) (_REXBrr(0, RS), _O_Mrm (0xf6 ,_b11,OP ,_r1(RS) ))
1084     #define _UNARYBm(OP, MD, MB, MI, MS) (_REXBrm(0, MB, MI), _O_r_X (0xf6 ,OP ,MD,MB,MI,MS ))
1085     #define _UNARYWr(OP, RS) (_d16(), _REXLrr(0, RS), _O_Mrm (0xf7 ,_b11,OP ,_r2(RS) ))
1086     #define _UNARYWm(OP, MD, MB, MI, MS) (_d16(), _REXLmr(MB, MI, 0), _O_r_X (0xf7 ,OP ,MD,MB,MI,MS ))
1087     #define _UNARYLr(OP, RS) (_REXLrr(0, RS), _O_Mrm (0xf7 ,_b11,OP ,_r4(RS) ))
1088     #define _UNARYLm(OP, MD, MB, MI, MS) (_REXLmr(MB, MI, 0), _O_r_X (0xf7 ,OP ,MD,MB,MI,MS ))
1089     #define _UNARYQr(OP, RS) (_REXQrr(0, RS), _O_Mrm (0xf7 ,_b11,OP ,_r8(RS) ))
1090     #define _UNARYQm(OP, MD, MB, MI, MS) (_REXQmr(MB, MI, 0), _O_r_X (0xf7 ,OP ,MD,MB,MI,MS ))
1091    
1092     #define NOTBr(RS) _UNARYBr(X86_NOT, RS)
1093     #define NOTBm(MD, MB, MI, MS) _UNARYBm(X86_NOT, MD, MB, MI, MS)
1094     #define NOTWr(RS) _UNARYWr(X86_NOT, RS)
1095     #define NOTWm(MD, MB, MI, MS) _UNARYWm(X86_NOT, MD, MB, MI, MS)
1096     #define NOTLr(RS) _UNARYLr(X86_NOT, RS)
1097     #define NOTLm(MD, MB, MI, MS) _UNARYLm(X86_NOT, MD, MB, MI, MS)
1098     #define NOTQr(RS) _UNARYQr(X86_NOT, RS)
1099     #define NOTQm(MD, MB, MI, MS) _UNARYQm(X86_NOT, MD, MB, MI, MS)
1100    
1101     #define NEGBr(RS) _UNARYBr(X86_NEG, RS)
1102     #define NEGBm(MD, MB, MI, MS) _UNARYBm(X86_NEG, MD, MB, MI, MS)
1103     #define NEGWr(RS) _UNARYWr(X86_NEG, RS)
1104     #define NEGWm(MD, MB, MI, MS) _UNARYWm(X86_NEG, MD, MB, MI, MS)
1105     #define NEGLr(RS) _UNARYLr(X86_NEG, RS)
1106     #define NEGLm(MD, MB, MI, MS) _UNARYLm(X86_NEG, MD, MB, MI, MS)
1107     #define NEGQr(RS) _UNARYQr(X86_NEG, RS)
1108     #define NEGQm(MD, MB, MI, MS) _UNARYQm(X86_NEG, MD, MB, MI, MS)
1109    
1110     #define MULBr(RS) _UNARYBr(X86_MUL, RS)
1111     #define MULBm(MD, MB, MI, MS) _UNARYBm(X86_MUL, MD, MB, MI, MS)
1112     #define MULWr(RS) _UNARYWr(X86_MUL, RS)
1113     #define MULWm(MD, MB, MI, MS) _UNARYWm(X86_MUL, MD, MB, MI, MS)
1114     #define MULLr(RS) _UNARYLr(X86_MUL, RS)
1115     #define MULLm(MD, MB, MI, MS) _UNARYLm(X86_MUL, MD, MB, MI, MS)
1116     #define MULQr(RS) _UNARYQr(X86_MUL, RS)
1117     #define MULQm(MD, MB, MI, MS) _UNARYQm(X86_MUL, MD, MB, MI, MS)
1118    
1119     #define IMULBr(RS) _UNARYBr(X86_IMUL, RS)
1120     #define IMULBm(MD, MB, MI, MS) _UNARYBm(X86_IMUL, MD, MB, MI, MS)
1121     #define IMULWr(RS) _UNARYWr(X86_IMUL, RS)
1122     #define IMULWm(MD, MB, MI, MS) _UNARYWm(X86_IMUL, MD, MB, MI, MS)
1123     #define IMULLr(RS) _UNARYLr(X86_IMUL, RS)
1124     #define IMULLm(MD, MB, MI, MS) _UNARYLm(X86_IMUL, MD, MB, MI, MS)
1125     #define IMULQr(RS) _UNARYQr(X86_IMUL, RS)
1126     #define IMULQm(MD, MB, MI, MS) _UNARYQm(X86_IMUL, MD, MB, MI, MS)
1127    
1128     #define DIVBr(RS) _UNARYBr(X86_DIV, RS)
1129     #define DIVBm(MD, MB, MI, MS) _UNARYBm(X86_DIV, MD, MB, MI, MS)
1130     #define DIVWr(RS) _UNARYWr(X86_DIV, RS)
1131     #define DIVWm(MD, MB, MI, MS) _UNARYWm(X86_DIV, MD, MB, MI, MS)
1132     #define DIVLr(RS) _UNARYLr(X86_DIV, RS)
1133     #define DIVLm(MD, MB, MI, MS) _UNARYLm(X86_DIV, MD, MB, MI, MS)
1134     #define DIVQr(RS) _UNARYQr(X86_DIV, RS)
1135     #define DIVQm(MD, MB, MI, MS) _UNARYQm(X86_DIV, MD, MB, MI, MS)
1136    
1137     #define IDIVBr(RS) _UNARYBr(X86_IDIV, RS)
1138     #define IDIVBm(MD, MB, MI, MS) _UNARYBm(X86_IDIV, MD, MB, MI, MS)
1139     #define IDIVWr(RS) _UNARYWr(X86_IDIV, RS)
1140     #define IDIVWm(MD, MB, MI, MS) _UNARYWm(X86_IDIV, MD, MB, MI, MS)
1141     #define IDIVLr(RS) _UNARYLr(X86_IDIV, RS)
1142     #define IDIVLm(MD, MB, MI, MS) _UNARYLm(X86_IDIV, MD, MB, MI, MS)
1143     #define IDIVQr(RS) _UNARYQr(X86_IDIV, RS)
1144     #define IDIVQm(MD, MB, MI, MS) _UNARYQm(X86_IDIV, MD, MB, MI, MS)
1145    
1146     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1147    
1148     #define IMULWrr(RS, RD) (_d16(), _REXLrr(RS, RD), _OO_Mrm (0x0faf ,_b11,_r2(RS),_r2(RD) ))
1149     #define IMULWmr(MD, MB, MI, MS, RD) (_d16(), _REXLmr(MB, MI, RD), _OO_r_X (0x0faf ,_r2(RD) ,MD,MB,MI,MS ))
1150    
1151     #define IMULWirr(IM,RS,RD) (_d16(), _REXLrr(RS, RD), _Os_Mrm_sW (0x69 ,_b11,_r2(RS),_r2(RD) ,_su16(IM) ))
1152     #define IMULWimr(IM,MD,MB,MI,MS,RD) (_d16(), _REXLmr(MB, MI, RD), _Os_r_X_sW (0x69 ,_r2(RD) ,MD,MB,MI,MS ,_su16(IM) ))
1153    
1154     #define IMULLir(IM, RD) (_REXLrr(0, RD), _Os_Mrm_sL (0x69 ,_b11,_r4(RD),_r4(RD) ,IM ))
1155     #define IMULLrr(RS, RD) (_REXLrr(RS, RD), _OO_Mrm (0x0faf ,_b11,_r4(RD),_r4(RS) ))
1156     #define IMULLmr(MD, MB, MI, MS, RD) (_REXLmr(MB, MI, RD), _OO_r_X (0x0faf ,_r4(RD) ,MD,MB,MI,MS ))
1157    
1158     #define IMULQir(IM, RD) (_REXQrr(0, RD), _Os_Mrm_sL (0x69 ,_b11,_r8(RD),_r8(RD) ,IM ))
1159     #define IMULQrr(RS, RD) (_REXQrr(RS, RD), _OO_Mrm (0x0faf ,_b11,_r8(RD),_r8(RS) ))
1160     #define IMULQmr(MD, MB, MI, MS, RD) (_REXQmr(MB, MI, RD), _OO_r_X (0x0faf ,_r8(RD) ,MD,MB,MI,MS ))
1161    
1162     #define IMULLirr(IM,RS,RD) (_REXLrr(RS, RD), _Os_Mrm_sL (0x69 ,_b11,_r4(RS),_r4(RD) ,IM ))
1163     #define IMULLimr(IM,MD,MB,MI,MS,RD) (_REXLmr(MB, MI, RD), _Os_r_X_sL (0x69 ,_r4(RD) ,MD,MB,MI,MS ,IM ))
1164    
1165     #define IMULQirr(IM,RS,RD) (_REXQrr(RS, RD), _Os_Mrm_sL (0x69 ,_b11,_r8(RS),_r8(RD) ,IM ))
1166     #define IMULQimr(IM,MD,MB,MI,MS,RD) (_REXQmr(MB, MI, RD), _Os_r_X_sL (0x69 ,_r8(RD) ,MD,MB,MI,MS ,IM ))
1167    
1168    
1169     /* --- Control Flow related instructions ----------------------------------- */
1170    
1171     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1172    
1173     // FIXME: no prefix is availble to encode a 32-bit operand size in 64-bit mode
1174     #define CALLm(M) _O_D32 (0xe8 ,(int)(M) )
1175     #define CALLsr(R) (_REXLrr(0, R), _O_Mrm (0xff ,_b11,_b010,_r4(R) ))
1176     #define CALLQsr(R) (_REXQrr(0, R), _O_Mrm (0xff ,_b11,_b010,_r8(R) ))
1177     #define CALLsm(D,B,I,S) (_REXLrm(0, B, I), _O_r_X (0xff ,_b010 ,(int)(D),B,I,S ))
1178    
1179     // FIXME: no prefix is availble to encode a 32-bit operand size in 64-bit mode
1180     #define JMPSm(M) _O_D8 (0xeb ,(int)(D) )
1181     #define JMPm(M) _O_D32 (0xe9 ,(int)(D) )
1182     #define JMPsr(R) (_REXLrr(0, R), _O_Mrm (0xff ,_b11,_b100,_r4(R) ))
1183     #define JMPQsr(R) (_REXQrr(0, R), _O_Mrm (0xff ,_b11,_b100,_r8(R) ))
1184     #define JMPsm(D,B,I,S) (_REXLrm(0, B, I), _O_r_X (0xff ,_b100 ,(int)(D),B,I,S ))
1185    
1186     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1187     #define JCCSim(CC, D) _O_D8 (0x70|(CC) ,(int)(D) )
1188     #define JOSm(D) JCCSim(0x0, D)
1189     #define JNOSm(D) JCCSim(0x1, D)
1190     #define JBSm(D) JCCSim(0x2, D)
1191     #define JNAESm(D) JCCSim(0x2, D)
1192     #define JNBSm(D) JCCSim(0x3, D)
1193     #define JAESm(D) JCCSim(0x3, D)
1194     #define JESm(D) JCCSim(0x4, D)
1195     #define JZSm(D) JCCSim(0x4, D)
1196     #define JNESm(D) JCCSim(0x5, D)
1197     #define JNZSm(D) JCCSim(0x5, D)
1198     #define JBESm(D) JCCSim(0x6, D)
1199     #define JNASm(D) JCCSim(0x6, D)
1200     #define JNBESm(D) JCCSim(0x7, D)
1201     #define JASm(D) JCCSim(0x7, D)
1202     #define JSSm(D) JCCSim(0x8, D)
1203     #define JNSSm(D) JCCSim(0x9, D)
1204     #define JPSm(D) JCCSim(0xa, D)
1205     #define JPESm(D) JCCSim(0xa, D)
1206     #define JNPSm(D) JCCSim(0xb, D)
1207     #define JPOSm(D) JCCSim(0xb, D)
1208     #define JLSm(D) JCCSim(0xc, D)
1209     #define JNGESm(D) JCCSim(0xc, D)
1210     #define JNLSm(D) JCCSim(0xd, D)
1211     #define JGESm(D) JCCSim(0xd, D)
1212     #define JLESm(D) JCCSim(0xe, D)
1213     #define JNGSm(D) JCCSim(0xe, D)
1214     #define JNLESm(D) JCCSim(0xf, D)
1215     #define JGSm(D) JCCSim(0xf, D)
1216    
1217     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1218     #define JCCim(CC, D) _OO_D32 (0x0f80|(CC) ,(int)(D) )
1219     #define JOm(D) JCCim(0x0, D)
1220     #define JNOm(D) JCCim(0x1, D)
1221     #define JBm(D) JCCim(0x2, D)
1222     #define JNAEm(D) JCCim(0x2, D)
1223     #define JNBm(D) JCCim(0x3, D)
1224     #define JAEm(D) JCCim(0x3, D)
1225     #define JEm(D) JCCim(0x4, D)
1226     #define JZm(D) JCCim(0x4, D)
1227     #define JNEm(D) JCCim(0x5, D)
1228     #define JNZm(D) JCCim(0x5, D)
1229     #define JBEm(D) JCCim(0x6, D)
1230     #define JNAm(D) JCCim(0x6, D)
1231     #define JNBEm(D) JCCim(0x7, D)
1232     #define JAm(D) JCCim(0x7, D)
1233     #define JSm(D) JCCim(0x8, D)
1234     #define JNSm(D) JCCim(0x9, D)
1235     #define JPm(D) JCCim(0xa, D)
1236     #define JPEm(D) JCCim(0xa, D)
1237     #define JNPm(D) JCCim(0xb, D)
1238     #define JPOm(D) JCCim(0xb, D)
1239     #define JLm(D) JCCim(0xc, D)
1240     #define JNGEm(D) JCCim(0xc, D)
1241     #define JNLm(D) JCCim(0xd, D)
1242     #define JGEm(D) JCCim(0xd, D)
1243     #define JLEm(D) JCCim(0xe, D)
1244     #define JNGm(D) JCCim(0xe, D)
1245     #define JNLEm(D) JCCim(0xf, D)
1246     #define JGm(D) JCCim(0xf, D)
1247    
1248     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1249     #define SETCCir(CC, RD) (_REXBrr(0, RD), _OO_Mrm (0x0f90|(CC) ,_b11,_b000,_r1(RD) ))
1250     #define SETOr(RD) SETCCir(0x0,RD)
1251     #define SETNOr(RD) SETCCir(0x1,RD)
1252     #define SETBr(RD) SETCCir(0x2,RD)
1253     #define SETNAEr(RD) SETCCir(0x2,RD)
1254     #define SETNBr(RD) SETCCir(0x3,RD)
1255     #define SETAEr(RD) SETCCir(0x3,RD)
1256     #define SETEr(RD) SETCCir(0x4,RD)
1257     #define SETZr(RD) SETCCir(0x4,RD)
1258     #define SETNEr(RD) SETCCir(0x5,RD)
1259     #define SETNZr(RD) SETCCir(0x5,RD)
1260     #define SETBEr(RD) SETCCir(0x6,RD)
1261     #define SETNAr(RD) SETCCir(0x6,RD)
1262     #define SETNBEr(RD) SETCCir(0x7,RD)
1263     #define SETAr(RD) SETCCir(0x7,RD)
1264     #define SETSr(RD) SETCCir(0x8,RD)
1265     #define SETNSr(RD) SETCCir(0x9,RD)
1266     #define SETPr(RD) SETCCir(0xa,RD)
1267     #define SETPEr(RD) SETCCir(0xa,RD)
1268     #define SETNPr(RD) SETCCir(0xb,RD)
1269     #define SETPOr(RD) SETCCir(0xb,RD)
1270     #define SETLr(RD) SETCCir(0xc,RD)
1271     #define SETNGEr(RD) SETCCir(0xc,RD)
1272     #define SETNLr(RD) SETCCir(0xd,RD)
1273     #define SETGEr(RD) SETCCir(0xd,RD)
1274     #define SETLEr(RD) SETCCir(0xe,RD)
1275     #define SETNGr(RD) SETCCir(0xe,RD)
1276     #define SETNLEr(RD) SETCCir(0xf,RD)
1277     #define SETGr(RD) SETCCir(0xf,RD)
1278    
1279     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1280     #define SETCCim(CC,MD,MB,MI,MS) (_REXBrm(0, MB, MI), _OO_r_X (0x0f90|(CC) ,_b000 ,MD,MB,MI,MS ))
1281     #define SETOm(D, B, I, S) SETCCim(0x0, D, B, I, S)
1282     #define SETNOm(D, B, I, S) SETCCim(0x1, D, B, I, S)
1283     #define SETBm(D, B, I, S) SETCCim(0x2, D, B, I, S)
1284     #define SETNAEm(D, B, I, S) SETCCim(0x2, D, B, I, S)
1285     #define SETNBm(D, B, I, S) SETCCim(0x3, D, B, I, S)
1286     #define SETAEm(D, B, I, S) SETCCim(0x3, D, B, I, S)
1287     #define SETEm(D, B, I, S) SETCCim(0x4, D, B, I, S)
1288     #define SETZm(D, B, I, S) SETCCim(0x4, D, B, I, S)
1289     #define SETNEm(D, B, I, S) SETCCim(0x5, D, B, I, S)
1290     #define SETNZm(D, B, I, S) SETCCim(0x5, D, B, I, S)
1291     #define SETBEm(D, B, I, S) SETCCim(0x6, D, B, I, S)
1292     #define SETNAm(D, B, I, S) SETCCim(0x6, D, B, I, S)
1293     #define SETNBEm(D, B, I, S) SETCCim(0x7, D, B, I, S)
1294     #define SETAm(D, B, I, S) SETCCim(0x7, D, B, I, S)
1295     #define SETSm(D, B, I, S) SETCCim(0x8, D, B, I, S)
1296     #define SETNSm(D, B, I, S) SETCCim(0x9, D, B, I, S)
1297     #define SETPm(D, B, I, S) SETCCim(0xa, D, B, I, S)
1298     #define SETPEm(D, B, I, S) SETCCim(0xa, D, B, I, S)
1299     #define SETNPm(D, B, I, S) SETCCim(0xb, D, B, I, S)
1300     #define SETPOm(D, B, I, S) SETCCim(0xb, D, B, I, S)
1301     #define SETLm(D, B, I, S) SETCCim(0xc, D, B, I, S)
1302     #define SETNGEm(D, B, I, S) SETCCim(0xc, D, B, I, S)
1303     #define SETNLm(D, B, I, S) SETCCim(0xd, D, B, I, S)
1304     #define SETGEm(D, B, I, S) SETCCim(0xd, D, B, I, S)
1305     #define SETLEm(D, B, I, S) SETCCim(0xe, D, B, I, S)
1306     #define SETNGm(D, B, I, S) SETCCim(0xe, D, B, I, S)
1307     #define SETNLEm(D, B, I, S) SETCCim(0xf, D, B, I, S)
1308     #define SETGm(D, B, I, S) SETCCim(0xf, D, B, I, S)
1309    
1310    
1311     /* --- Push/Pop instructions ----------------------------------------------- */
1312    
1313     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1314    
1315     #define POPWr(RD) _m32only((_d16(), _Or (0x58,_r2(RD) )))
1316     #define POPWm(MD, MB, MI, MS) _m32only((_d16(), _O_r_X (0x8f ,_b000 ,MD,MB,MI,MS )))
1317    
1318     #define POPLr(RD) _m32only( _Or (0x58,_r4(RD) ))
1319     #define POPLm(MD, MB, MI, MS) _m32only( _O_r_X (0x8f ,_b000 ,MD,MB,MI,MS ))
1320    
1321     #define POPQr(RD) _m64only( _Or (0x58,_r8(RD) ))
1322     #define POPQm(MD, MB, MI, MS) _m64only( _O_r_X (0x8f ,_b000 ,MD,MB,MI,MS ))
1323    
1324     #define PUSHWr(RS) _m32only((_d16(), _Or (0x50,_r2(RS) )))
1325     #define PUSHWm(MD, MB, MI, MS) _m32only((_d16(), _O_r_X (0xff, ,_b110 ,MD,MB,MI,MS )))
1326     #define PUSHWi(IM) _m32only((_d16(), _Os_sW (0x68 ,IM )))
1327    
1328     #define PUSHLr(RS) _m32only( _Or (0x50,_r4(RS) ))
1329     #define PUSHLm(MD, MB, MI, MS) _m32only( _O_r_X (0xff ,_b110 ,MD,MB,MI,MS ))
1330     #define PUSHLi(IM) _m32only( _Os_sL (0x68 ,IM ))
1331    
1332     #define PUSHQr(RS) _m64only( _Or (0x50,_r8(RS) ))
1333     #define PUSHQm(MD, MB, MI, MS) _m64only( _O_r_X (0xff ,_b110 ,MD,MB,MI,MS ))
1334     #define PUSHQi(IM) _m64only( _Os_sL (0x68 ,IM ))
1335    
1336     #define POPA() (_d16(), _O (0x61 ))
1337     #define POPAD() _O (0x61 )
1338    
1339     #define PUSHA() (_d16(), _O (0x60 ))
1340     #define PUSHAD() _O (0x60 )
1341    
1342     #define POPF() (_d16(), _O (0x9d ))
1343     #define POPFD() _O (0x9d )
1344    
1345     #define PUSHF() _O (0x9c )
1346     #define PUSHFD() (_d16(), _O (0x9c ))
1347    
1348    
1349     /* --- Test instructions --------------------------------------------------- */
1350    
1351     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1352    
1353     #define TESTBrr(RS, RD) (_REXBrr(RS, RD), _O_Mrm (0x84 ,_b11,_r1(RS),_r1(RD) ))
1354     #define TESTBrm(RS, MD, MB, MI, MS) (_REXBrm(RS, MB, MI), _O_r_X (0x84 ,_r1(RS) ,MD,MB,MI,MS ))
1355     #define TESTBir(IM, RD) (_REXBrr(0, RD), _O_Mrm_B (0xf6 ,_b11,_b000 ,_r1(RD) ,_u8(IM)))
1356     #define TESTBim(IM, MD, MB, MI, MS) (_REXBrm(0, MB, MI), _O_r_X_B (0xf6 ,_b000 ,MD,MB,MI,MS ,_u8(IM)))
1357    
1358     #define TESTWrr(RS, RD) (_d16(), _REXLrr(RS, RD), _O_Mrm (0x85 ,_b11,_r2(RS),_r2(RD) ))
1359     #define TESTWrm(RS, MD, MB, MI, MS) (_d16(), _REXLrm(RS, MD, MI), _O_r_X (0x85 ,_r2(RS) ,MD,MB,MI,MS ))
1360     #define TESTWir(IM, RD) (_d16(), _REXLrr(0, RD), _O_Mrm_W (0xf7 ,_b11,_b000 ,_r2(RD) ,_u16(IM)))
1361     #define TESTWim(IM, MD, MB, MI, MS) (_d16(), _REXLrm(0, MD, MI), _O_r_X_W (0xf7 ,_b000 ,MD,MB,MI,MS ,_u16(IM)))
1362    
1363     #define TESTLrr(RS, RD) (_REXLrr(RS, RD), _O_Mrm (0x85 ,_b11,_r4(RS),_r4(RD) ))
1364     #define TESTLrm(RS, MD, MB, MI, MS) (_REXLrm(RS, MB, MI), _O_r_X (0x85 ,_r4(RS) ,MD,MB,MI,MS ))
1365     #define TESTLir(IM, RD) (_REXLrr(0, RD), _O_Mrm_L (0xf7 ,_b11,_b000 ,_r4(RD) ,IM ))
1366     #define TESTLim(IM, MD, MB, MI, MS) (_REXLrm(0, MB, MI), _O_r_X_L (0xf7 ,_b000 ,MD,MB,MI,MS ,IM ))
1367    
1368     #define TESTQrr(RS, RD) (_REXQrr(RS, RD), _O_Mrm (0x85 ,_b11,_r8(RS),_r8(RD) ))
1369     #define TESTQrm(RS, MD, MB, MI, MS) (_REXQrm(RS, MB, MI), _O_r_X (0x85 ,_r8(RS) ,MD,MB,MI,MS ))
1370     #define TESTQir(IM, RD) (_REXQrr(0, RD), _O_Mrm_L (0xf7 ,_b11,_b000 ,_r8(RD) ,IM ))
1371     #define TESTQim(IM, MD, MB, MI, MS) (_REXQrm(0, MB, MI), _O_r_X_L (0xf7 ,_b000 ,MD,MB,MI,MS ,IM ))
1372    
1373    
1374     /* --- Exchange instructions ----------------------------------------------- */
1375    
1376     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1377    
1378     #define CMPXCHGBrr(RS, RD) (_REXBrr(RS, RD), _OO_Mrm (0x0fb0 ,_b11,_r1(RS),_r1(RD) ))
1379     #define CMPXCHGBrm(RS, MD, MB, MI, MS) (_REXBrm(RS, MB, MI), _OO_r_X (0x0fb0 ,_r1(RS) ,MD,MB,MI,MS ))
1380    
1381     #define CMPXCHGWrr(RS, RD) (_d16(), _REXLrr(RS, RD), _OO_Mrm (0x0fb1 ,_b11,_r2(RS),_r2(RD) ))
1382     #define CMPXCHGWrm(RS, MD, MB, MI, MS) (_d16(), _REXLrm(RS, MB, MI), _OO_r_X (0x0fb1 ,_r2(RS) ,MD,MB,MI,MS ))
1383    
1384     #define CMPXCHGLrr(RS, RD) (_REXLrr(RS, RD), _OO_Mrm (0x0fb1 ,_b11,_r4(RS),_r4(RD) ))
1385     #define CMPXCHGLrm(RS, MD, MB, MI, MS) (_REXLrm(RS, MB, MI), _OO_r_X (0x0fb1 ,_r4(RS) ,MD,MB,MI,MS ))
1386    
1387     #define CMPXCHGQrr(RS, RD) (_REXQrr(RS, RD), _OO_Mrm (0x0fb1 ,_b11,_r8(RS),_r8(RD) ))
1388     #define CMPXCHGQrm(RS, MD, MB, MI, MS) (_REXQrm(RS, MB, MI), _OO_r_X (0x0fb1 ,_r8(RS) ,MD,MB,MI,MS ))
1389    
1390     #define XADDBrr(RS, RD) (_REXBrr(RS, RD), _OO_Mrm (0x0fc0 ,_b11,_r1(RS),_r1(RD) ))
1391     #define XADDBrm(RS, MD, MB, MI, MS) (_REXBrm(RS, MB, MI), _OO_r_X (0x0fc0 ,_r1(RS) ,MD,MB,MI,MS ))
1392    
1393     #define XADDWrr(RS, RD) (_d16(), _REXLrr(RS, RD), _OO_Mrm (0x0fc1 ,_b11,_r2(RS),_r2(RD) ))
1394     #define XADDWrm(RS, MD, MB, MI, MS) (_d16(), _REXLrm(RS, MB, MI), _OO_r_X (0x0fc1 ,_r2(RS) ,MD,MB,MI,MS ))
1395    
1396     #define XADDLrr(RS, RD) (_REXLrr(RS, RD), _OO_Mrm (0x0fc1 ,_b11,_r4(RS),_r4(RD) ))
1397     #define XADDLrm(RS, MD, MB, MI, MS) (_REXLrm(RS, MB, MI), _OO_r_X (0x0fc1 ,_r4(RS) ,MD,MB,MI,MS ))
1398    
1399     #define XADDQrr(RS, RD) (_REXQrr(RS, RD), _OO_Mrm (0x0fc1 ,_b11,_r8(RS),_r8(RD) ))
1400     #define XADDQrm(RS, MD, MB, MI, MS) (_REXQrm(RS, MB, MI), _OO_r_X (0x0fc1 ,_r8(RS) ,MD,MB,MI,MS ))
1401    
1402     #define XCHGBrr(RS, RD) (_REXBrr(RS, RD), _O_Mrm (0x86 ,_b11,_r1(RS),_r1(RD) ))
1403     #define XCHGBrm(RS, MD, MB, MI, MS) (_REXBrm(RS, MB, MI), _O_r_X (0x86 ,_r1(RS) ,MD,MB,MI,MS ))
1404    
1405     #define XCHGWrr(RS, RD) (_d16(), _REXLrr(RS, RD), _O_Mrm (0x87 ,_b11,_r2(RS),_r2(RD) ))
1406     #define XCHGWrm(RS, MD, MB, MI, MS) (_d16(), _REXLrm(RS, MB, MI), _O_r_X (0x87 ,_r2(RS) ,MD,MB,MI,MS ))
1407    
1408     #define XCHGLrr(RS, RD) (_REXLrr(RS, RD), _O_Mrm (0x87 ,_b11,_r4(RS),_r4(RD) ))
1409     #define XCHGLrm(RS, MD, MB, MI, MS) (_REXLrm(RS, MB, MI), _O_r_X (0x87 ,_r4(RS) ,MD,MB,MI,MS ))
1410    
1411     #define XCHGQrr(RS, RD) (_REXQrr(RS, RD), _O_Mrm (0x87 ,_b11,_r8(RS),_r8(RD) ))
1412     #define XCHGQrm(RS, MD, MB, MI, MS) (_REXQrm(RS, MB, MI), _O_r_X (0x87 ,_r8(RS) ,MD,MB,MI,MS ))
1413    
1414    
1415     /* --- Increment/Decrement instructions ------------------------------------ */
1416    
1417     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1418    
1419     #define DECBm(MD, MB, MI, MS) (_REXBrm(0, MB, MI), _O_r_X (0xfe ,_b001 ,MD,MB,MI,MS ))
1420     #define DECBr(RD) (_REXBrr(0, RD), _O_Mrm (0xfe ,_b11,_b001 ,_r1(RD) ))
1421    
1422     #define DECWm(MD, MB, MI, MS) (_d16(), _REXLrm(0, MB, MI), _O_r_X (0xff ,_b001 ,MD,MB,MI,MS ))
1423     #define DECWr(RD) (! X86_TARGET_64BIT ? (_d16(), _Or (0x48,_r2(RD) )) : \
1424     (_d16(), _REXLrr(0, RD), _O_Mrm (0xff ,_b11,_b001 ,_r2(RD) )))
1425    
1426     #define DECLm(MD, MB, MI, MS) (_REXLrm(0, MB, MI), _O_r_X (0xff ,_b001 ,MD,MB,MI,MS ))
1427     #define DECLr(RD) (! X86_TARGET_64BIT ? _Or (0x48,_r4(RD) ) : \
1428     (_REXLrr(0, RD), _O_Mrm (0xff ,_b11,_b001 ,_r4(RD) )))
1429    
1430     #define DECQm(MD, MB, MI, MS) (_REXQrm(0, MB, MI), _O_r_X (0xff ,_b001 ,MD,MB,MI,MS ))
1431     #define DECQr(RD) (_REXQrr(0, RD), _O_Mrm (0xff ,_b11,_b001 ,_r8(RD) ))
1432    
1433     #define INCBm(MD, MB, MI, MS) (_REXBrm(0, MB, MI), _O_r_X (0xfe ,_b000 ,MD,MB,MI,MS ))
1434     #define INCBr(RD) (_REXBrr(0, RD), _O_Mrm (0xfe ,_b11,_b000 ,_r1(RD) ))
1435    
1436     #define INCWm(MD, MB, MI, MS) (_d16(), _REXLrm(0, MB, MI), _O_r_X (0xff ,_b000 ,MD,MB,MI,MS ))
1437     #define INCWr(RD) (! X86_TARGET_64BIT ? (_d16(), _Or (0x40,_r2(RD) )) : \
1438     (_d16(), _REXLrr(0, RD), _O_Mrm (0xff ,_b11,_b000 ,_r2(RD) )) )
1439    
1440     #define INCLm(MD, MB, MI, MS) (_REXLrm(0, MB, MI), _O_r_X (0xff ,_b000 ,MD,MB,MI,MS ))
1441     #define INCLr(RD) (! X86_TARGET_64BIT ? _Or (0x40,_r4(RD) ) : \
1442     (_REXLrr(0, RD), _O_Mrm (0xff ,_b11,_b000 ,_r4(RD) )))
1443    
1444     #define INCQm(MD, MB, MI, MS) (_REXQrm(0, MB, MI), _O_r_X (0xff ,_b000 ,MD,MB,MI,MS ))
1445     #define INCQr(RD) (_REXQrr(0, RD), _O_Mrm (0xff ,_b11,_b000 ,_r8(RD) ))
1446    
1447    
1448     /* --- Misc/Garbage instructions ------------------------------------------- */
1449    
1450     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1451    
1452     #define LEALmr(MD, MB, MI, MS, RD) (_REXLmr(MB, MI, RD), _O_r_X (0x8d ,_r4(RD) ,MD,MB,MI,MS ))
1453    
1454     #define BSWAPLr(R) (_REXLrr(0, R), _OOr (0x0fc8,_r4(R) ))
1455     #define BSWAPQr(R) (_REXQrr(0, R), _OOr (0x0fc8,_r8(R) ))
1456    
1457     #define CLC() _O (0xf8 )
1458     #define STC() _O (0xf9 )
1459    
1460     #define CMC() _O (0xf5 )
1461     #define CLD() _O (0xfc )
1462     #define STD() _O (0xfd )
1463    
1464     #define CBTW() (_d16(), _O (0x98 ))
1465     #define CWTL() _O (0x98 )
1466     #define CLTQ() _m64only(_REXQrr(0, 0), _O (0x98 ))
1467    
1468     #define CBW CBTW
1469     #define CWDE CWTL
1470     #define CDQE CLTQ
1471    
1472     #define CWTD() (_d16(), _O (0x99 ))
1473     #define CLTD() _O (0x99 )
1474     #define CQTO() _m64only(_REXQrr(0, 0), _O (0x99 ))
1475    
1476     #define CWD CWTD
1477     #define CDQ CLTD
1478     #define CQO CQTO
1479    
1480     #define LAHF() _m32only( _O (0x9f ))
1481     #define SAHF() _m32only( _O (0x9e ))
1482    
1483 gbeauche 1.2 /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1484    
1485 gbeauche 1.1 #define RDTSC() _OO (0xff31 )
1486    
1487     #define ENTERii(W, B) _O_W_B (0xc8 ,_su16(W),_su8(B))
1488    
1489     #define LEAVE() _O (0xc9 )
1490     #define RET() _O (0xc3 )
1491     #define RETi(IM) _O_W (0xc2 ,_su16(IM))
1492    
1493     #define NOP() _O (0x90 )
1494 gbeauche 1.3
1495    
1496     /* --- Media 128-bit instructions ------------------------------------------ */
1497    
1498     enum {
1499     X86_SSE_CVTIS = 0x2a,
1500     X86_SSE_CVTSI = 0x2d,
1501     X86_SSE_UCOMI = 0x2e,
1502     X86_SSE_COMI = 0x2f,
1503     X86_SSE_SQRT = 0x51,
1504     X86_SSE_RSQRT = 0x52,
1505     X86_SSE_RCP = 0x53,
1506     X86_SSE_AND = 0x54,
1507     X86_SSE_ANDN = 0x55,
1508     X86_SSE_OR = 0x56,
1509     X86_SSE_XOR = 0x57,
1510     X86_SSE_ADD = 0x58,
1511     X86_SSE_MUL = 0x59,
1512     X86_SSE_CVTSD = 0x5a,
1513     X86_SSE_CVTDT = 0x5b,
1514     X86_SSE_SUB = 0x5c,
1515     X86_SSE_MIN = 0x5d,
1516     X86_SSE_DIV = 0x5e,
1517     X86_SSE_MAX = 0x5f,
1518     };
1519    
1520     /* _format Opcd ,Mod ,r ,m ,mem=dsp+sib ,imm... */
1521    
1522     #define __SSELrr(OP,RS,RSA,RD,RDA) (_REXLrr(RD, RS), _OO_Mrm (0x0f00|(OP) ,_b11,RDA(RD),RSA(RS) ))
1523     #define __SSELmr(OP,MD,MB,MI,MS,RD,RDA) (_REXLmr(MB, MI, RD), _OO_r_X (0x0f00|(OP) ,RDA(RD) ,MD,MB,MI,MS ))
1524     #define __SSELrm(OP,RS,RSA,MD,MB,MI,MS) (_REXLrm(RS, MB, MI), _OO_r_X (0x0f00|(OP) ,RSA(RS) ,MD,MB,MI,MS ))
1525    
1526     #define __SSEQrr(OP,RS,RSA,RD,RDA) (_REXQrr(RD, RS), _OO_Mrm (0x0f00|(OP) ,_b11,RDA(RD),RSA(RS) ))
1527     #define __SSEQmr(OP,MD,MB,MI,MS,RD,RDA) (_REXQmr(MB, MI, RD), _OO_r_X (0x0f00|(OP) ,RDA(RD) ,MD,MB,MI,MS ))
1528     #define __SSEQrm(OP,RS,RSA,MD,MB,MI,MS) (_REXQrm(RS, MB, MI), _OO_r_X (0x0f00|(OP) ,RSA(RS) ,MD,MB,MI,MS ))
1529    
1530     #define _SSELrr(PX,OP,RS,RSA,RD,RDA) (_B(PX), __SSELrr(OP, RS, RSA, RD, RDA))
1531     #define _SSELmr(PX,OP,MD,MB,MI,MS,RD,RDA) (_B(PX), __SSELmr(OP, MD, MB, MI, MS, RD, RDA))
1532     #define _SSELrm(PX,OP,RS,RSA,MD,MB,MI,MS) (_B(PX), __SSELrm(OP, RS, RSA, MD, MB, MI, MS))
1533    
1534     #define _SSEQrr(PX,OP,RS,RSA,RD,RDA) (_B(PX), __SSEQrr(OP, RS, RSA, RD, RDA))
1535     #define _SSEQmr(PX,OP,MD,MB,MI,MS,RD,RDA) (_B(PX), __SSEQmr(OP, MD, MB, MI, MS, RD, RDA))
1536     #define _SSEQrm(PX,OP,RS,RSA,MD,MB,MI,MS) (_B(PX), __SSEQrm(OP, RS, RSA, MD, MB, MI, MS))
1537    
1538     #define _SSEPSrr(OP,RS,RD) __SSELrr( OP, RS,_rX, RD,_rX)
1539     #define _SSEPSmr(OP,MD,MB,MI,MS,RD) __SSELmr( OP, MD, MB, MI, MS, RD,_rX)
1540     #define _SSEPSrm(OP,RS,MD,MB,MI,MS) __SSELrm( OP, RS,_rX, MD, MB, MI, MS)
1541    
1542     #define _SSEPDrr(OP,RS,RD) _SSELrr(0x66, OP, RS,_rX, RD,_rX)
1543     #define _SSEPDmr(OP,MD,MB,MI,MS,RD) _SSELmr(0x66, OP, MD, MB, MI, MS, RD,_rX)
1544     #define _SSEPDrm(OP,RS,MD,MB,MI,MS) _SSELrm(0x66, OP, RS,_rX, MD, MB, MI, MS)
1545    
1546     #define _SSESSrr(OP,RS,RD) _SSELrr(0xf3, OP, RS,_rX, RD,_rX)
1547     #define _SSESSmr(OP,MD,MB,MI,MS,RD) _SSELmr(0xf3, OP, MD, MB, MI, MS, RD,_rX)
1548     #define _SSESSrm(OP,RS,MD,MB,MI,MS) _SSELrm(0xf3, OP, RS,_rX, MD, MB, MI, MS)
1549    
1550     #define _SSESDrr(OP,RS,RD) _SSELrr(0xf2, OP, RS,_rX, RD,_rX)
1551     #define _SSESDmr(OP,MD,MB,MI,MS,RD) _SSELmr(0xf2, OP, MD, MB, MI, MS, RD,_rX)
1552     #define _SSESDrm(OP,RS,MD,MB,MI,MS) _SSELrm(0xf2, OP, RS,_rX, MD, MB, MI, MS)
1553    
1554     #define ADDPSrr(RS, RD) _SSEPSrr(X86_SSE_ADD, RS, RD)
1555     #define ADDPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_ADD, MD, MB, MI, MS, RD)
1556     #define ADDPDrr(RS, RD) _SSEPDrr(X86_SSE_ADD, RS, RD)
1557     #define ADDPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_ADD, MD, MB, MI, MS, RD)
1558    
1559     #define ADDSSrr(RS, RD) _SSESSrr(X86_SSE_ADD, RS, RD)
1560     #define ADDSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_ADD, MD, MB, MI, MS, RD)
1561     #define ADDSDrr(RS, RD) _SSESDrr(X86_SSE_ADD, RS, RD)
1562     #define ADDSDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_ADD, MD, MB, MI, MS, RD)
1563    
1564     #define ANDNPSrr(RS, RD) _SSEPSrr(X86_SSE_ANDN, RS, RD)
1565     #define ANDNPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_ANDN, MD, MB, MI, MS, RD)
1566     #define ANDNPDrr(RS, RD) _SSEPDrr(X86_SSE_ANDN, RS, RD)
1567     #define ANDNPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_ANDN, MD, MB, MI, MS, RD)
1568    
1569     #define ANDPSrr(RS, RD) _SSEPSrr(X86_SSE_AND, RS, RD)
1570     #define ANDPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_AND, MD, MB, MI, MS, RD)
1571     #define ANDPDrr(RS, RD) _SSEPDrr(X86_SSE_AND, RS, RD)
1572     #define ANDPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_AND, MD, MB, MI, MS, RD)
1573    
1574     #define DIVPSrr(RS, RD) _SSEPSrr(X86_SSE_DIV, RS, RD)
1575     #define DIVPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_DIV, MD, MB, MI, MS, RD)
1576     #define DIVPDrr(RS, RD) _SSEPDrr(X86_SSE_DIV, RS, RD)
1577     #define DIVPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_DIV, MD, MB, MI, MS, RD)
1578    
1579     #define DIVSSrr(RS, RD) _SSESSrr(X86_SSE_DIV, RS, RD)
1580     #define DIVSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_DIV, MD, MB, MI, MS, RD)
1581     #define DIVSDrr(RS, RD) _SSESDrr(X86_SSE_DIV, RS, RD)
1582     #define DIVSDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_DIV, MD, MB, MI, MS, RD)
1583    
1584     #define MAXPSrr(RS, RD) _SSEPSrr(X86_SSE_MAX, RS, RD)
1585     #define MAXPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_MAX, MD, MB, MI, MS, RD)
1586     #define MAXPDrr(RS, RD) _SSEPDrr(X86_SSE_MAX, RS, RD)
1587     #define MAXPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_MAX, MD, MB, MI, MS, RD)
1588    
1589     #define MAXSSrr(RS, RD) _SSESSrr(X86_SSE_MAX, RS, RD)
1590     #define MAXSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_MAX, MD, MB, MI, MS, RD)
1591     #define MAXSDrr(RS, RD) _SSESDrr(X86_SSE_MAX, RS, RD)
1592     #define MAXSDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_MAX, MD, MB, MI, MS, RD)
1593    
1594     #define MINPSrr(RS, RD) _SSEPSrr(X86_SSE_MIN, RS, RD)
1595     #define MINPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_MIN, MD, MB, MI, MS, RD)
1596     #define MINPDrr(RS, RD) _SSEPDrr(X86_SSE_MIN, RS, RD)
1597     #define MINPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_MIN, MD, MB, MI, MS, RD)
1598    
1599     #define MINSSrr(RS, RD) _SSESSrr(X86_SSE_MIN, RS, RD)
1600     #define MINSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_MIN, MD, MB, MI, MS, RD)
1601     #define MINSDrr(RS, RD) _SSESDrr(X86_SSE_MIN, RS, RD)
1602     #define MINSDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_MIN, MD, MB, MI, MS, RD)
1603    
1604     #define MULPSrr(RS, RD) _SSEPSrr(X86_SSE_MUL, RS, RD)
1605     #define MULPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_MUL, MD, MB, MI, MS, RD)
1606     #define MULPDrr(RS, RD) _SSEPDrr(X86_SSE_MUL, RS, RD)
1607     #define MULPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_MUL, MD, MB, MI, MS, RD)
1608    
1609     #define MULSSrr(RS, RD) _SSESSrr(X86_SSE_MUL, RS, RD)
1610     #define MULSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_MUL, MD, MB, MI, MS, RD)
1611     #define MULSDrr(RS, RD) _SSESDrr(X86_SSE_MUL, RS, RD)
1612     #define MULSDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_MUL, MD, MB, MI, MS, RD)
1613    
1614     #define ORPSrr(RS, RD) _SSEPSrr(X86_SSE_OR, RS, RD)
1615     #define ORPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_OR, MD, MB, MI, MS, RD)
1616     #define ORPDrr(RS, RD) _SSEPDrr(X86_SSE_OR, RS, RD)
1617     #define ORPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_OR, MD, MB, MI, MS, RD)
1618    
1619     #define RCPPSrr(RS, RD) _SSEPSrr(X86_SSE_RCP, RS, RD)
1620     #define RCPPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_RCP, MD, MB, MI, MS, RD)
1621     #define RCPSSrr(RS, RD) _SSESSrr(X86_SSE_RCP, RS, RD)
1622     #define RCPSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_RCP, MD, MB, MI, MS, RD)
1623    
1624     #define RSQRTPSrr(RS, RD) _SSEPSrr(X86_SSE_RSQRT, RS, RD)
1625     #define RSQRTPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_RSQRT, MD, MB, MI, MS, RD)
1626     #define RSQRTSSrr(RS, RD) _SSESSrr(X86_SSE_RSQRT, RS, RD)
1627     #define RSQRTSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_RSQRT, MD, MB, MI, MS, RD)
1628    
1629     #define SQRTPSrr(RS, RD) _SSEPSrr(X86_SSE_SQRT, RS, RD)
1630     #define SQRTPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_SQRT, MD, MB, MI, MS, RD)
1631     #define SQRTPDrr(RS, RD) _SSEPDrr(X86_SSE_SQRT, RS, RD)
1632     #define SQRTPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_SQRT, MD, MB, MI, MS, RD)
1633    
1634     #define SQRTSSrr(RS, RD) _SSESSrr(X86_SSE_SQRT, RS, RD)
1635     #define SQRTSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_SQRT, MD, MB, MI, MS, RD)
1636     #define SQRTSDrr(RS, RD) _SSESDrr(X86_SSE_SQRT, RS, RD)
1637     #define SQRTSDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_SQRT, MD, MB, MI, MS, RD)
1638    
1639     #define SUBPSrr(RS, RD) _SSEPSrr(X86_SSE_SUB, RS, RD)
1640     #define SUBPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_SUB, MD, MB, MI, MS, RD)
1641     #define SUBPDrr(RS, RD) _SSEPDrr(X86_SSE_SUB, RS, RD)
1642     #define SUBPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_SUB, MD, MB, MI, MS, RD)
1643    
1644     #define SUBSSrr(RS, RD) _SSESSrr(X86_SSE_SUB, RS, RD)
1645     #define SUBSSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_SUB, MD, MB, MI, MS, RD)
1646     #define SUBSDrr(RS, RD) _SSESDrr(X86_SSE_SUB, RS, RD)
1647     #define SUBSDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_SUB, MD, MB, MI, MS, RD)
1648    
1649     #define XORPSrr(RS, RD) _SSEPSrr(X86_SSE_XOR, RS, RD)
1650     #define XORPSmr(MD, MB, MI, MS, RD) _SSEPSmr(X86_SSE_XOR, MD, MB, MI, MS, RD)
1651     #define XORPDrr(RS, RD) _SSEPDrr(X86_SSE_XOR, RS, RD)
1652     #define XORPDmr(MD, MB, MI, MS, RD) _SSEPDmr(X86_SSE_XOR, MD, MB, MI, MS, RD)
1653    
1654     #define COMISSrr(RS, RD) _SSESSrr(X86_SSE_COMI, RS, RD)
1655     #define COMISSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_COMI, MD, MB, MI, MS, RD)
1656     #define COMISDrr(RS, RD) _SSESDrr(X86_SSE_COMI, RS, RD)
1657     #define COMISDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_COMI, MD, MB, MI, MS, RD)
1658    
1659     #define UCOMISSrr(RS, RD) _SSESSrr(X86_SSE_UCOMI, RS, RD)
1660     #define UCOMISSmr(MD, MB, MI, MS, RD) _SSESSmr(X86_SSE_UCOMI, MD, MB, MI, MS, RD)
1661     #define UCOMISDrr(RS, RD) _SSESDrr(X86_SSE_UCOMI, RS, RD)
1662     #define UCOMISDmr(MD, MB, MI, MS, RD) _SSESDmr(X86_SSE_UCOMI, MD, MB, MI, MS, RD)
1663    
1664     #define MOVAPSrr(RS, RD) _SSEPSrr(0x28, RS, RD)
1665     #define MOVAPSmr(MD, MB, MI, MS, RD) _SSEPSmr(0x28, MD, MB, MI, MS, RD)
1666     #define MOVAPSrm(RS, MD, MB, MI, MS) _SSEPSrm(0x29, RS, MD, MB, MI, MS)
1667    
1668     #define MOVAPDrr(RS, RD) _SSEPDrr(0x28, RS, RD)
1669     #define MOVAPDmr(MD, MB, MI, MS, RD) _SSEPDmr(0x28, MD, MB, MI, MS, RD)
1670     #define MOVAPDrm(RS, MD, MB, MI, MS) _SSEPDrm(0x29, RS, MD, MB, MI, MS)
1671    
1672     #define CVTPS2PIrr(RS, RD) __SSELrr( X86_SSE_CVTSI, RS,_rX, RD,_rM)
1673     #define CVTPS2PImr(MD, MB, MI, MS, RD) __SSELmr( X86_SSE_CVTSI, MD, MB, MI, MS, RD,_rM)
1674     #define CVTPD2PIrr(RS, RD) _SSELrr(0x66, X86_SSE_CVTSI, RS,_rX, RD,_rM)
1675     #define CVTPD2PImr(MD, MB, MI, MS, RD) _SSELmr(0x66, X86_SSE_CVTSI, MD, MB, MI, MS, RD,_rM)
1676    
1677     #define CVTPI2PSrr(RS, RD) __SSELrr( X86_SSE_CVTIS, RS,_rM, RD,_rX)
1678     #define CVTPI2PSmr(MD, MB, MI, MS, RD) __SSELmr( X86_SSE_CVTIS, MD, MB, MI, MS, RD,_rX)
1679     #define CVTPI2PDrr(RS, RD) _SSELrr(0x66, X86_SSE_CVTIS, RS,_rM, RD,_rX)
1680     #define CVTPI2PDmr(MD, MB, MI, MS, RD) _SSELmr(0x66, X86_SSE_CVTIS, MD, MB, MI, MS, RD,_rX)
1681    
1682     #define CVTPS2PDrr(RS, RD) __SSELrr( X86_SSE_CVTSD, RS,_rX, RD,_rX)
1683     #define CVTPS2PDmr(MD, MB, MI, MS, RD) __SSELmr( X86_SSE_CVTSD, MD, MB, MI, MS, RD,_rX)
1684     #define CVTPD2PSrr(RS, RD) _SSELrr(0x66, X86_SSE_CVTSD, RS,_rX, RD,_rX)
1685     #define CVTPD2PSmr(MD, MB, MI, MS, RD) _SSELmr(0x66, X86_SSE_CVTSD, MD, MB, MI, MS, RD,_rX)
1686    
1687     #define CVTSS2SDrr(RS, RD) _SSELrr(0xf3, X86_SSE_CVTSD, RS,_rX, RD,_rX)
1688     #define CVTSS2SDmr(MD, MB, MI, MS, RD) _SSELmr(0xf3, X86_SSE_CVTSD, MD, MB, MI, MS, RD,_rX)
1689     #define CVTSD2SSrr(RS, RD) _SSELrr(0xf2, X86_SSE_CVTSD, RS,_rX, RD,_rX)
1690     #define CVTSD2SSmr(MD, MB, MI, MS, RD) _SSELmr(0xf2, X86_SSE_CVTSD, MD, MB, MI, MS, RD,_rX)
1691    
1692     #define CVTSS2SILrr(RS, RD) _SSELrr(0xf3, X86_SSE_CVTSI, RS,_rX, RD,_r4)
1693     #define CVTSS2SILmr(MD, MB, MI, MS, RD) _SSELmr(0xf3, X86_SSE_CVTSI, MD, MB, MI, MS, RD,_r4)
1694     #define CVTSD2SILrr(RS, RD) _SSELrr(0xf2, X86_SSE_CVTSI, RS,_rX, RD,_r4)
1695     #define CVTSD2SILmr(MD, MB, MI, MS, RD) _SSELmr(0xf2, X86_SSE_CVTSI, MD, MB, MI, MS, RD,_r4)
1696    
1697     #define CVTSI2SSLrr(RS, RD) _SSELrr(0xf3, X86_SSE_CVTIS, RS,_r4, RD,_rX)
1698     #define CVTSI2SSLmr(MD, MB, MI, MS, RD) _SSELmr(0xf3, X86_SSE_CVTIS, MD, MB, MI, MS, RD,_rX)
1699     #define CVTSI2SDLrr(RS, RD) _SSELrr(0xf2, X86_SSE_CVTIS, RS,_r4, RD,_rX)
1700     #define CVTSI2SDLmr(MD, MB, MI, MS, RD) _SSELmr(0xf2, X86_SSE_CVTIS, MD, MB, MI, MS, RD,_rX)
1701    
1702     #define CVTSS2SIQrr(RS, RD) _SSEQrr(0xf3, X86_SSE_CVTSI, RS,_rX, RD,_r8)
1703     #define CVTSS2SIQmr(MD, MB, MI, MS, RD) _SSEQmr(0xf3, X86_SSE_CVTSI, MD, MB, MI, MS, RD,_r8)
1704     #define CVTSD2SIQrr(RS, RD) _SSEQrr(0xf2, X86_SSE_CVTSI, RS,_rX, RD,_r8)
1705     #define CVTSD2SIQmr(MD, MB, MI, MS, RD) _SSEQmr(0xf2, X86_SSE_CVTSI, MD, MB, MI, MS, RD,_r8)
1706    
1707     #define CVTSI2SSQrr(RS, RD) _SSEQrr(0xf3, X86_SSE_CVTIS, RS,_r8, RD,_rX)
1708     #define CVTSI2SSQmr(MD, MB, MI, MS, RD) _SSEQmr(0xf3, X86_SSE_CVTIS, MD, MB, MI, MS, RD,_rX)
1709     #define CVTSI2SDQrr(RS, RD) _SSEQrr(0xf2, X86_SSE_CVTIS, RS,_r8, RD,_rX)
1710     #define CVTSI2SDQmr(MD, MB, MI, MS, RD) _SSEQmr(0xf2, X86_SSE_CVTIS, MD, MB, MI, MS, RD,_rX)
1711    
1712     #define MOVDLXrr(RS, RD) _SSELrr(0x66, 0x6e, RS,_r4, RD,_rX)
1713     #define MOVDLXmr(MD, MB, MI, MS, RD) _SSELmr(0x66, 0x6e, MD, MB, MI, MS, RD,_rX)
1714     #define MOVDQXrr(RS, RD) _SSEQrr(0x66, 0x6e, RS,_r8, RD,_rX)
1715     #define MOVDQXmr(MD, MB, MI, MS, RD) _SSEQmr(0x66, 0x6e, MD, MB, MI, MS, RD,_rX)
1716    
1717     #define MOVDXLrr(RS, RD) _SSELrr(0x66, 0x7e, RS,_rX, RD,_r4)
1718     #define MOVDXLrm(RS, MD, MB, MI, MS) _SSELrm(0x66, 0x7e, RS,_rX, MD, MB, MI, MS)
1719     #define MOVDXQrr(RS, RD) _SSEQrr(0x66, 0x7e, RS,_rX, RD,_r8)
1720     #define MOVDXQrm(RS, MD, MB, MI, MS) _SSEQrm(0x66, 0x7e, RS,_rX, MD, MB, MI, MS)
1721    
1722     #define MOVDLMrr(RS, RD) __SSELrr( 0x6e, RS,_r4, RD,_rM)
1723     #define MOVDLMmr(MD, MB, MI, MS, RD) __SSELmr( 0x6e, MD, MB, MI, MS, RD,_rM)
1724     #define MOVDQMrr(RS, RD) __SSEQrr( 0x6e, RS,_r8, RD,_rM)
1725     #define MOVDQMmr(MD, MB, MI, MS, RD) __SSEQmr( 0x6e, MD, MB, MI, MS, RD,_rM)
1726    
1727     #define MOVDMLrr(RS, RD) __SSELrr( 0x7e, RS,_rM, RD,_r4)
1728     #define MOVDMLrm(RS, MD, MB, MI, MS) __SSELrm( 0x7e, RS,_rM, MD, MB, MI, MS)
1729     #define MOVDMQrr(RS, RD) __SSEQrr( 0x7e, RS,_rM, RD,_r8)
1730     #define MOVDMQrm(RS, MD, MB, MI, MS) __SSEQrm( 0x7e, RS,_rM, MD, MB, MI, MS)
1731    
1732     #define MOVDQ2Qrr(RS, RD) _SSELrr(0xf2, 0xd6, RS,_rX, RD,_rM)
1733     #define MOVHLPSrr(RS, RD) __SSELrr( 0x12, RS,_rX, RD,_rX)
1734     #define MOVLHPSrr(RS, RD) __SSELrr( 0x16, RS,_rX, RD,_rX)
1735    
1736     #define MOVDQArr(RS, RD) _SSELrr(0x66, 0x6f, RS,_rX, RD,_rX)
1737     #define MOVDQAmr(MD, MB, MI, MS, RD) _SSELmr(0x66, 0x6f, MD, MB, MI, MS, RD,_rX)
1738     #define MOVDQArm(RS, MD, MB, MI, MS) _SSELrm(0x66, 0x7f, RS,_rX, MD, MB, MI, MS)
1739    
1740     #define MOVDQUrr(RS, RD) _SSELrr(0xf3, 0x6f, RS,_rX, RD,_rX)
1741     #define MOVDQUmr(MD, MB, MI, MS, RD) _SSELmr(0xf3, 0x6f, MD, MB, MI, MS, RD,_rX)
1742     #define MOVDQUrm(RS, MD, MB, MI, MS) _SSELrm(0xf3, 0x7f, RS,_rX, MD, MB, MI, MS)
1743    
1744     #define MOVHPDmr(MD, MB, MI, MS, RD) _SSELmr(0x66, 0x16, MD, MB, MI, MS, RD,_rX)
1745     #define MOVHPDrm(RS, MD, MB, MI, MS) _SSELrm(0x66, 0x17, RS,_rX, MD, MB, MI, MS)
1746     #define MOVHPSmr(MD, MB, MI, MS, RD) __SSELmr( 0x16, MD, MB, MI, MS, RD,_rX)
1747     #define MOVHPSrm(RS, MD, MB, MI, MS) __SSELrm( 0x17, RS,_rX, MD, MB, MI, MS)
1748    
1749     #define MOVLPDmr(MD, MB, MI, MS, RD) _SSELmr(0x66, 0x12, MD, MB, MI, MS, RD,_rX)
1750     #define MOVLPDrm(RS, MD, MB, MI, MS) _SSELrm(0x66, 0x13, RS,_rX, MD, MB, MI, MS)
1751     #define MOVLPSmr(MD, MB, MI, MS, RD) __SSELmr( 0x12, MD, MB, MI, MS, RD,_rX)
1752     #define MOVLPSrm(RS, MD, MB, MI, MS) __SSELrm( 0x13, RS,_rX, MD, MB, MI, MS)
1753 gbeauche 1.2
1754    
1755     /* --- FLoating-Point instructions ----------------------------------------- */
1756    
1757     #define _ESCmi(D,B,I,S,OP) (_REXLrm(0,B,I), _O_r_X(0xd8|(OP & 7), (OP >> 3), D,B,I,S))
1758    
1759     #define FLDr(R) _OOr(0xd9c0,_rN(R))
1760     #define FLDLm(D,B,I,S) _ESCmi(D,B,I,S,005)
1761     #define FLDSm(D,B,I,S) _ESCmi(D,B,I,S,001)
1762     #define FLDTm(D,B,I,S) _ESCmi(D,B,I,S,053)
1763    
1764     #define FSTr(R) _OOr(0xddd0,_rN(R))
1765     #define FSTSm(D,B,I,S) _ESCmi(D,B,I,S,021)
1766     #define FSTLm(D,B,I,S) _ESCmi(D,B,I,S,025)
1767    
1768     #define FSTPr(R) _OOr(0xddd8,_rN(R))
1769     #define FSTPSm(D,B,I,S) _ESCmi(D,B,I,S,031)
1770     #define FSTPLm(D,B,I,S) _ESCmi(D,B,I,S,035)
1771     #define FSTPTm(D,B,I,S) _ESCmi(D,B,I,S,073)
1772    
1773     #define FADDr0(R) _OOr(0xd8c0,_rN(R))
1774     #define FADD0r(R) _OOr(0xdcc0,_rN(R))
1775     #define FADDP0r(R) _OOr(0xdec0,_rN(R))
1776     #define FADDSm(D,B,I,S) _ESCmi(D,B,I,S,000)
1777     #define FADDLm(D,B,I,S) _ESCmi(D,B,I,S,004)
1778    
1779     #define FSUBSm(D,B,I,S) _ESCmi(D,B,I,S,040)
1780     #define FSUBLm(D,B,I,S) _ESCmi(D,B,I,S,044)
1781     #define FSUBr0(R) _OOr(0xd8e0,_rN(R))
1782     #define FSUB0r(R) _OOr(0xdce8,_rN(R))
1783     #define FSUBP0r(R) _OOr(0xdee8,_rN(R))
1784    
1785     #define FSUBRr0(R) _OOr(0xd8e8,_rN(R))
1786     #define FSUBR0r(R) _OOr(0xdce0,_rN(R))
1787     #define FSUBRP0r(R) _OOr(0xdee0,_rN(R))
1788     #define FSUBRSm(D,B,I,S) _ESCmi(D,B,I,S,050)
1789     #define FSUBRLm(D,B,I,S) _ESCmi(D,B,I,S,054)
1790    
1791     #define FMULr0(R) _OOr(0xd8c8,_rN(R))
1792     #define FMUL0r(R) _OOr(0xdcc8,_rN(R))
1793     #define FMULP0r(R) _OOr(0xdec8,_rN(R))
1794     #define FMULSm(D,B,I,S) _ESCmi(D,B,I,S,010)
1795     #define FMULLm(D,B,I,S) _ESCmi(D,B,I,S,014)
1796    
1797     #define FDIVr0(R) _OOr(0xd8f0,_rN(R))
1798     #define FDIV0r(R) _OOr(0xdcf8,_rN(R))
1799     #define FDIVP0r(R) _OOr(0xdef8,_rN(R))
1800     #define FDIVSm(D,B,I,S) _ESCmi(D,B,I,S,060)
1801     #define FDIVLm(D,B,I,S) _ESCmi(D,B,I,S,064)
1802    
1803     #define FDIVRr0(R) _OOr(0xd8f8,_rN(R))
1804     #define FDIVR0r(R) _OOr(0xdcf0,_rN(R))
1805     #define FDIVRP0r(R) _OOr(0xdef0,_rN(R))
1806     #define FDIVRSm(D,B,I,S) _ESCmi(D,B,I,S,070)
1807     #define FDIVRLm(D,B,I,S) _ESCmi(D,B,I,S,074)
1808    
1809     #define FCMOVBr0(R) _OOr(0xdac0,_rN(R))
1810     #define FCMOVBEr0(R) _OOr(0xdad0,_rN(R))
1811     #define FCMOVEr0(R) _OOr(0xdac8,_rN(R))
1812     #define FCMOVNBr0(R) _OOr(0xdbc0,_rN(R))
1813     #define FCMOVNBEr0(R) _OOr(0xdbd0,_rN(R))
1814     #define FCMOVNEr0(R) _OOr(0xdbc8,_rN(R))
1815     #define FCMOVNUr0(R) _OOr(0xdbd8,_rN(R))
1816     #define FCMOVUr0(R) _OOr(0xdad8,_rN(R))
1817     #define FCOMIr0(R) _OOr(0xdbf0,_rN(R))
1818     #define FCOMIPr0(R) _OOr(0xdff0,_rN(R))
1819    
1820     #define FCOMr(R) _OOr(0xd8d0,_rN(R))
1821     #define FCOMSm(D,B,I,S) _ESCmi(D,B,I,S,020)
1822     #define FCOMLm(D,B,I,S) _ESCmi(D,B,I,S,024)
1823    
1824     #define FCOMPr(R) _OOr(0xd8d8,_rN(R))
1825     #define FCOMPSm(D,B,I,S) _ESCmi(D,B,I,S,030)
1826     #define FCOMPLm(D,B,I,S) _ESCmi(D,B,I,S,034)
1827    
1828     #define FUCOMIr0(R) _OOr(0xdbe8,_rN(R))
1829     #define FUCOMIPr0(R) _OOr(0xdfe8,_rN(R))
1830     #define FUCOMPr(R) _OOr(0xdde8,_rN(R))
1831     #define FUCOMr(R) _OOr(0xdde0,_rN(R))
1832    
1833     #define FIADDLm(D,B,I,S) _ESCmi(D,B,I,S,002)
1834     #define FICOMLm(D,B,I,S) _ESCmi(D,B,I,S,022)
1835     #define FICOMPLm(D,B,I,S) _ESCmi(D,B,I,S,032)
1836     #define FIDIVLm(D,B,I,S) _ESCmi(D,B,I,S,062)
1837     #define FIDIVRLm(D,B,I,S) _ESCmi(D,B,I,S,072)
1838     #define FILDLm(D,B,I,S) _ESCmi(D,B,I,S,003)
1839     #define FILDQm(D,B,I,S) _ESCmi(D,B,I,S,057)
1840     #define FIMULLm(D,B,I,S) _ESCmi(D,B,I,S,012)
1841     #define FISTLm(D,B,I,S) _ESCmi(D,B,I,S,023)
1842     #define FISTPLm(D,B,I,S) _ESCmi(D,B,I,S,033)
1843     #define FISTPQm(D,B,I,S) _ESCmi(D,B,I,S,077)
1844     #define FISUBLm(D,B,I,S) _ESCmi(D,B,I,S,042)
1845     #define FISUBRLm(D,B,I,S) _ESCmi(D,B,I,S,052)
1846    
1847     #define FREEr(R) _OOr(0xddc0,_rN(R))
1848     #define FXCHr(R) _OOr(0xd9c8,_rN(R))
1849 gbeauche 1.1
1850     #endif /* X86_RTASM_H */