1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
|
#include <stdio.h>
#include <stdlib.h>
#include <stddef.h>
#include <errno.h>
#include "debug.h"
#include "box64context.h"
#include "dynarec.h"
#include "emu/x64emu_private.h"
#include "emu/x64run_private.h"
#include "x64run.h"
#include "x64emu.h"
#include "box64stack.h"
#include "callback.h"
#include "emu/x64run_private.h"
#include "x64trace.h"
#include "dynarec_native.h"
#include "custommem.h"
#include "rv64_printer.h"
#include "dynarec_rv64_private.h"
#include "dynarec_rv64_helper.h"
#include "dynarec_rv64_functions.h"
uintptr_t dynarec64_64(dynarec_rv64_t* dyn, uintptr_t addr, uintptr_t ip, int ninst, rex_t rex, int rep, int seg, int* ok, int* need_epilog)
{
(void)ip; (void)rep; (void)need_epilog;
uint8_t opcode = F8;
uint8_t nextop;
uint8_t u8;
uint8_t gd, ed, eb1, eb2, gb1, gb2;
uint8_t gback, wback, wb1, wb2, wb;
int64_t i64, j64;
uint64_t u64;
int v0, v1;
int q0;
int d0;
int64_t fixedaddress, gdoffset;
int unscaled;
MAYUSE(eb1);
MAYUSE(eb2);
MAYUSE(wb1);
MAYUSE(wb2);
MAYUSE(gb1);
MAYUSE(gb2);
MAYUSE(j64);
MAYUSE(d0);
MAYUSE(q0);
MAYUSE(v0);
MAYUSE(v1);
while((opcode==0xF2) || (opcode==0xF3)) {
rep = opcode-0xF1;
opcode = F8;
}
GETREX();
switch(opcode) {
case 0x03:
INST_NAME("ADD Gd, Seg:Ed");
SETFLAGS(X_ALL, SF_SET_PENDING);
grab_segdata(dyn, addr, ninst, x4, seg);
nextop = F8;
GETGD;
GETEDO(x4, 0, x5);
emit_add32(dyn, ninst, rex, gd, ed, x3, x4, x5);
break;
case 0x0F:
opcode = F8;
switch(opcode) {
case 0x11:
switch(rep) {
case 0:
INST_NAME("MOVUPS Ex,Gx");
nextop = F8;
GETGX();
GETEX(x2, 0);
if(!MODREG) {
grab_segdata(dyn, addr, ninst, x4, seg);
ADD(x4, x4, wback);
wback = x4;
}
LD(x3, gback, gdoffset+0);
LD(x5, gback, gdoffset+8);
SD(x3, wback, fixedaddress+0);
SD(x5, wback, fixedaddress+8);
if(!MODREG)
SMWRITE2();
break;
case 1:
INST_NAME("MOVSD Ex, Gx");
nextop = F8;
GETG;
v0 = sse_get_reg(dyn, ninst, x1, gd, 0);
if(MODREG) {
ed = (nextop&7)+ (rex.b<<3);
d0 = sse_get_reg(dyn, ninst, x1, ed, 0);
FMVD(d0, v0);
} else {
grab_segdata(dyn, addr, ninst, x4, seg);
addr = geted(dyn, addr, ninst, nextop, &ed, x1, x2, &fixedaddress, rex, NULL, 1, 0);
ADD(x4, x4, ed);
ed = x4;
FSD(v0, ed, fixedaddress);
SMWRITE2();
}
break;
case 2:
INST_NAME("MOVSS Ex, Gx");
nextop = F8;
GETG;
v0 = sse_get_reg(dyn, ninst, x1, gd, 1);
if(MODREG) {
q0 = sse_get_reg(dyn, ninst, x1, (nextop&7) + (rex.b<<3), 1);
FMVS(q0, v0);
} else {
grab_segdata(dyn, addr, ninst, x4, seg);
addr = geted(dyn, addr, ninst, nextop, &ed, x1, x2, &fixedaddress, rex, NULL, 1, 0);
ADD(x4, x4, ed);
ed = x4;
FSW(v0, ed, fixedaddress);
SMWRITE2();
}
break;
default:
DEFAULT;
}
break;
default:
DEFAULT;
}
break;
case 0x2B:
INST_NAME("SUB Gd, Seg:Ed");
SETFLAGS(X_ALL, SF_SET_PENDING);
grab_segdata(dyn, addr, ninst, x4, seg);
nextop = F8;
GETGD;
GETEDO(x4, 0, x5);
emit_sub32(dyn, ninst, rex, gd, ed, x3, x4, x5);
break;
case 0x33:
INST_NAME("XOR Gd, Seg:Ed");
SETFLAGS(X_ALL, SF_SET_PENDING);
grab_segdata(dyn, addr, ninst, x4, seg);
nextop = F8;
GETGD;
GETEDO(x4, 0, x5);
emit_xor32(dyn, ninst, rex, gd, ed, x3, x4);
break;
case 0x66:
addr = dynarec64_6664(dyn, addr, ip, ninst, rex, seg, ok, need_epilog);
break;
case 0x80:
nextop = F8;
switch((nextop>>3)&7) {
case 0: // ADD
INST_NAME("ADD Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
emit_add8c(dyn, ninst, x1, u8, x2, x4, x5);
EBBACK(x5, 0);
break;
case 1: // OR
INST_NAME("OR Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
emit_or8c(dyn, ninst, x1, u8, x2, x4, x5);
EBBACK(x5, 0);
break;
case 2: // ADC
INST_NAME("ADC Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
READFLAGS(X_CF);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
emit_adc8c(dyn, ninst, x1, u8, x2, x4, x5, x6);
EBBACK(x5, 0);
break;
case 3: // SBB
INST_NAME("SBB Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
READFLAGS(X_CF);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
emit_sbb8c(dyn, ninst, x1, u8, x2, x4, x5, x6);
EBBACK(x5, 0);
break;
case 4: // AND
INST_NAME("AND Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
emit_and8c(dyn, ninst, x1, u8, x2, x4);
EBBACK(x5, 0);
break;
case 5: // SUB
INST_NAME("SUB Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
emit_sub8c(dyn, ninst, x1, u8, x2, x4, x5, x6);
EBBACK(x5, 0);
break;
case 6: // XOR
INST_NAME("XOR Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
emit_xor8c(dyn, ninst, x1, u8, x2, x4);
EBBACK(x5, 0);
break;
case 7: // CMP
INST_NAME("CMP Eb, Ib");
grab_segdata(dyn, addr, ninst, x1, seg);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEBO(x1, 1);
u8 = F8;
if(u8) {
MOV32w(x2, u8);
emit_cmp8(dyn, ninst, x1, x2, x3, x4, x5, x6);
} else {
emit_cmp8_0(dyn, ninst, x1, x3, x4);
}
break;
default:
DEFAULT;
}
break;
case 0x81:
case 0x83:
nextop = F8;
grab_segdata(dyn, addr, ninst, x6, seg);
switch((nextop>>3)&7) {
case 0: // ADD
if(opcode==0x81) {INST_NAME("ADD Ed, Id");} else {INST_NAME("ADD Ed, Ib");}
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
emit_add32c(dyn, ninst, rex, ed, i64, x3, x4, x5, x9);
WBACKO(x6);
break;
case 1: // OR
if(opcode==0x81) {INST_NAME("OR Ed, Id");} else {INST_NAME("OR Ed, Ib");}
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
emit_or32c(dyn, ninst, rex, ed, i64, x3, x4);
WBACKO(x6);
break;
case 2: // ADC
if(opcode==0x81) {INST_NAME("ADC Ed, Id");} else {INST_NAME("ADC Ed, Ib");}
READFLAGS(X_CF);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
MOV64xw(x5, i64);
SD(x6, xEmu, offsetof(x64emu_t, scratch));
emit_adc32(dyn, ninst, rex, ed, x5, x3, x4, x6, x9);
LD(x6, xEmu, offsetof(x64emu_t, scratch));
WBACKO(x6);
break;
case 3: // SBB
if(opcode==0x81) {INST_NAME("SBB Ed, Id");} else {INST_NAME("SBB Ed, Ib");}
READFLAGS(X_CF);
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
MOV64xw(x5, i64);
emit_sbb32(dyn, ninst, rex, ed, x5, x3, x4, x9);
WBACKO(x6);
break;
case 4: // AND
if(opcode==0x81) {INST_NAME("AND Ed, Id");} else {INST_NAME("AND Ed, Ib");}
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
emit_and32c(dyn, ninst, rex, ed, i64, x3, x4);
WBACKO(x6);
break;
case 5: // SUB
if(opcode==0x81) {INST_NAME("SUB Ed, Id");} else {INST_NAME("SUB Ed, Ib");}
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
emit_sub32c(dyn, ninst, rex, ed, i64, x3, x4, x5, x9);
WBACKO(x6);
break;
case 6: // XOR
if(opcode==0x81) {INST_NAME("XOR Ed, Id");} else {INST_NAME("XOR Ed, Ib");}
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
emit_xor32c(dyn, ninst, rex, ed, i64, x3, x4);
WBACKO(x6);
break;
case 7: // CMP
if(opcode==0x81) {INST_NAME("CMP Ed, Id");} else {INST_NAME("CMP Ed, Ib");}
SETFLAGS(X_ALL, SF_SET_PENDING);
GETEDO(x6, (opcode==0x81)?4:1, x5);
if(opcode==0x81) i64 = F32S; else i64 = F8S;
if(i64) {
MOV64xw(x2, i64);
emit_cmp32(dyn, ninst, rex, ed, x2, x3, x4, x5, x6);
} else
emit_cmp32_0(dyn, ninst, rex, ed, x3, x4);
break;
}
break;
case 0x88:
INST_NAME("MOV Seg:Eb, Gb");
grab_segdata(dyn, addr, ninst, x4, seg);
nextop=F8;
gd = ((nextop&0x38)>>3)+(rex.r<<3);
if(rex.rex) {
gb2 = 0;
gb1 = xRAX + gd;
} else {
gb2 = ((gd&4)>>2);
gb1 = xRAX+(gd&3);
}
gd = x5;
if(gb2) {
SRLI(x5, gb1, 8);
gb1 = x5;
}
if(MODREG) {
ed = (nextop&7) + (rex.b<<3);
if(rex.rex) {
eb1 = xRAX+ed;
eb2 = 0;
} else {
eb1 = xRAX+(ed&3); // Ax, Cx, Dx or Bx
eb2 = ((ed&4)>>2); // L or H
}
ANDI(gd, gb1, 0xff);
if(eb2) {
MOV64x(x1, 0xffffffffffff00ffLL);
ANDI(x1, eb1, x1);
SLLI(gd, gd, 8);
OR(eb1, x1, gd);
} else {
ANDI(x1, eb1, ~0xff);
OR(eb1, x1, gd);
}
} else {
addr = geted(dyn, addr, ninst, nextop, &ed, x2, x1, &fixedaddress, rex, NULL, 1, 0);
ADD(x4, ed, x4);
SB(gb1, x4, fixedaddress);
SMWRITE2();
}
break;
case 0x89:
INST_NAME("MOV Seg:Ed, Gd");
grab_segdata(dyn, addr, ninst, x4, seg);
nextop=F8;
GETGD;
if(MODREG) { // reg <= reg
MVxw(xRAX+(nextop&7)+(rex.b<<3), gd);
} else { // mem <= reg
addr = geted(dyn, addr, ninst, nextop, &ed, x2, x1, &fixedaddress, rex, NULL, 1, 0);
ADD(x4, ed, x4);
SDxw(gd, x4, fixedaddress);
SMWRITE2();
}
break;
case 0x8B:
INST_NAME("MOV Gd, Seg:Ed");
grab_segdata(dyn, addr, ninst, x4, seg);
nextop=F8;
GETGD;
if(MODREG) { // reg <= reg
MVxw(gd, xRAX+(nextop&7)+(rex.b<<3));
} else { // mem <= reg
SMREAD();
addr = geted(dyn, addr, ninst, nextop, &ed, x2, x1, &fixedaddress, rex, NULL, 1, 0);
ADD(x4, ed, x4);
LDxw(gd, x4, fixedaddress);
}
break;
case 0x8E:
INST_NAME("MOV Seg,Ew");
grab_segdata(dyn, addr, ninst, x4, seg);
nextop = F8;
u8 = (nextop&0x38) >> 3;
if ((nextop&0xC0) == 0xC0) {
ed = xRAX+(nextop&7) + (rex.b<<3);
} else {
SMREAD();
addr = geted(dyn, addr, ninst, nextop, &wback, x2, x1, &fixedaddress, rex, NULL, 0, 0);
ADD(x4, wback, x4);
LHU(x1, x4, 0);
ed = x1;
}
SH(ed, xEmu, offsetof(x64emu_t, segs[u8]));
SW(xZR, xEmu, offsetof(x64emu_t, segs_serial[u8]));
break;
case 0x8F:
INST_NAME("POP FS:Ed");
grab_segdata(dyn, addr, ninst, x4, seg);
nextop = F8;
if (MODREG) {
POP1z(xRAX + (nextop & 7) + (rex.b << 3));
} else {
POP1z(x3); // so this can handle POP [ESP] and maybe some variant too
addr = geted(dyn, addr, ninst, nextop, &ed, x2, x1, &fixedaddress, rex, NULL, 0, 0);
if (ed == xRSP) {
ADD(x4, ed, x4);
SDz(x3, x4, 0);
} else {
// complicated to just allow a segfault that can be recovered correctly
ADDIz(xRSP, xRSP, rex.is32bits ? -4 : -8);
ADD(x4, ed, x4);
SDz(x3, x4, 0);
ADDIz(xRSP, xRSP, rex.is32bits ? 4 : 8);
}
}
break;
case 0xA1:
INST_NAME("MOV EAX,FS:Od");
grab_segdata(dyn, addr, ninst, x4, seg);
if(rex.is32bits)
u64 = F32;
else
u64 = F64;
// TODO: could be optimized.
MOV64z(x1, u64);
ADD(x1, x1, x4);
LDxw(xRAX, x1, 0);
break;
case 0xA3:
INST_NAME("MOV FS:Od,EAX");
grab_segdata(dyn, addr, ninst, x4, seg);
if(rex.is32bits)
u64 = F32;
else
u64 = F64;
// TODO: could be optimized.
MOV64z(x1, u64);
ADD(x1, x1, x4);
SDxw(xRAX, x1, 0);
SMWRITE2();
break;
case 0xC6:
INST_NAME("MOV Seg:Eb, Ib");
grab_segdata(dyn, addr, ninst, x4, seg);
nextop=F8;
if(MODREG) { // reg <= u8
u8 = F8;
if(!rex.rex) {
ed = (nextop&7);
eb1 = xRAX+(ed&3); // Ax, Cx, Dx or Bx
eb2 = (ed&4)>>2; // L or H
} else {
eb1 = xRAX+(nextop&7)+(rex.b<<3);
eb2 = 0;
}
if (eb2) {
// load a mask to x3 (ffffffffffff00ff)
LUI(x3, 0xffff0);
ORI(x3, x3, 0xff);
// apply mask
AND(eb1, eb1, x3);
if(u8) {
if((u8<<8)<2048) {
ADDI(x4, xZR, u8<<8);
} else {
ADDI(x4, xZR, u8);
SLLI(x4, x4, 8);
}
OR(eb1, eb1, x4);
}
} else {
ANDI(eb1, eb1, 0xf00); // mask ffffffffffffff00
ORI(eb1, eb1, u8);
}
} else { // mem <= u8
addr = geted(dyn, addr, ninst, nextop, &wback, x2, x1, &fixedaddress, rex, NULL, 1, 1);
u8 = F8;
if(u8) {
ADDI(x3, xZR, u8);
ed = x3;
} else
ed = xZR;
ADD(x4, wback, x4);
SB(ed, x4, fixedaddress);
SMWRITE2();
}
break;
case 0xC7:
INST_NAME("MOV Seg:Ed, Id");
grab_segdata(dyn, addr, ninst, x4, seg);
nextop=F8;
if(MODREG) { // reg <= i32
i64 = F32S;
ed = xRAX+(nextop&7)+(rex.b<<3);
MOV64xw(ed, i64);
} else { // mem <= i32
addr = geted(dyn, addr, ninst, nextop, &wback, x2, x1, &fixedaddress, rex, NULL, 1, 4);
i64 = F32S;
if(i64) {
MOV64xw(x3, i64);
ed = x3;
} else
ed = xZR;
ADD(x4, wback, x4);
SDxw(ed, x4, fixedaddress);
SMWRITE2();
}
break;
case 0xFF:
nextop = F8;
grab_segdata(dyn, addr, ninst, x6, seg);
switch((nextop>>3)&7) {
case 6: // Push Ed
INST_NAME("PUSH Ed");
GETEDOz(x6, 0, x3);
PUSH1z(ed);
break;
default:
DEFAULT;
}
break;
default:
DEFAULT;
}
return addr;
}
|