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
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
|
/*
* Tiny Code Generator for QEMU
*
* Copyright (c) 2008 Fabrice Bellard
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*/
#include "../tcg-ldst.c.inc"
#include "../tcg-pool.c.inc"
#ifdef CONFIG_DEBUG_TCG
static const char * const tcg_target_reg_names[TCG_TARGET_NB_REGS] = {
#if TCG_TARGET_REG_BITS == 64
"%rax", "%rcx", "%rdx", "%rbx", "%rsp", "%rbp", "%rsi", "%rdi",
#else
"%eax", "%ecx", "%edx", "%ebx", "%esp", "%ebp", "%esi", "%edi",
#endif
"%r8", "%r9", "%r10", "%r11", "%r12", "%r13", "%r14", "%r15",
"%xmm0", "%xmm1", "%xmm2", "%xmm3", "%xmm4", "%xmm5", "%xmm6", "%xmm7",
#if TCG_TARGET_REG_BITS == 64
"%xmm8", "%xmm9", "%xmm10", "%xmm11",
"%xmm12", "%xmm13", "%xmm14", "%xmm15",
#endif
};
#endif
static const int tcg_target_reg_alloc_order[] = {
#if TCG_TARGET_REG_BITS == 64
TCG_REG_RBP,
TCG_REG_RBX,
TCG_REG_R12,
TCG_REG_R13,
TCG_REG_R14,
TCG_REG_R15,
TCG_REG_R10,
TCG_REG_R11,
TCG_REG_R9,
TCG_REG_R8,
TCG_REG_RCX,
TCG_REG_RDX,
TCG_REG_RSI,
TCG_REG_RDI,
TCG_REG_RAX,
#else
TCG_REG_EBX,
TCG_REG_ESI,
TCG_REG_EDI,
TCG_REG_EBP,
TCG_REG_ECX,
TCG_REG_EDX,
TCG_REG_EAX,
#endif
TCG_REG_XMM0,
TCG_REG_XMM1,
TCG_REG_XMM2,
TCG_REG_XMM3,
TCG_REG_XMM4,
TCG_REG_XMM5,
#ifndef _WIN64
/* The Win64 ABI has xmm6-xmm15 as caller-saves, and we do not save
any of them. Therefore only allow xmm0-xmm5 to be allocated. */
TCG_REG_XMM6,
TCG_REG_XMM7,
#if TCG_TARGET_REG_BITS == 64
TCG_REG_XMM8,
TCG_REG_XMM9,
TCG_REG_XMM10,
TCG_REG_XMM11,
TCG_REG_XMM12,
TCG_REG_XMM13,
TCG_REG_XMM14,
TCG_REG_XMM15,
#endif
#endif
};
static const int tcg_target_call_iarg_regs[] = {
#if TCG_TARGET_REG_BITS == 64
#if defined(_WIN64)
TCG_REG_RCX,
TCG_REG_RDX,
#else
TCG_REG_RDI,
TCG_REG_RSI,
TCG_REG_RDX,
TCG_REG_RCX,
#endif
TCG_REG_R8,
TCG_REG_R9,
#else
/* 32 bit mode uses stack based calling convention (GCC default). */
#endif
};
static TCGReg tcg_target_call_oarg_reg(TCGCallReturnKind kind, int slot)
{
switch (kind) {
case TCG_CALL_RET_NORMAL:
tcg_debug_assert(slot >= 0 && slot <= 1);
return slot ? TCG_REG_EDX : TCG_REG_EAX;
#ifdef _WIN64
case TCG_CALL_RET_BY_VEC:
tcg_debug_assert(slot == 0);
return TCG_REG_XMM0;
#endif
default:
g_assert_not_reached();
}
}
/* Constants we accept. */
#define TCG_CT_CONST_S32 0x100
#define TCG_CT_CONST_U32 0x200
#define TCG_CT_CONST_I32 0x400
#define TCG_CT_CONST_WSZ 0x800
/* Registers used with L constraint, which are the first argument
registers on x86_64, and two random call clobbered registers on
i386. */
#if TCG_TARGET_REG_BITS == 64
# define TCG_REG_L0 tcg_target_call_iarg_regs[0]
# define TCG_REG_L1 tcg_target_call_iarg_regs[1]
#else
# define TCG_REG_L0 TCG_REG_EAX
# define TCG_REG_L1 TCG_REG_EDX
#endif
#define ALL_BYTEH_REGS 0x0000000fu
#if TCG_TARGET_REG_BITS == 64
# define ALL_GENERAL_REGS 0x0000ffffu
# define ALL_VECTOR_REGS 0xffff0000u
# define ALL_BYTEL_REGS ALL_GENERAL_REGS
#else
# define ALL_GENERAL_REGS 0x000000ffu
# define ALL_VECTOR_REGS 0x00ff0000u
# define ALL_BYTEL_REGS ALL_BYTEH_REGS
#endif
#ifdef CONFIG_SOFTMMU
# define SOFTMMU_RESERVE_REGS ((1 << TCG_REG_L0) | (1 << TCG_REG_L1))
#else
# define SOFTMMU_RESERVE_REGS 0
#endif
/* The host compiler should supply <cpuid.h> to enable runtime features
detection, as we're not going to go so far as our own inline assembly.
If not available, default values will be assumed. */
#if defined(CONFIG_CPUID_H)
#include "qemu/cpuid.h"
#endif
/* For 64-bit, we always know that CMOV is available. */
#if TCG_TARGET_REG_BITS == 64
# define have_cmov 1
#elif defined(CONFIG_CPUID_H)
static bool have_cmov;
#else
# define have_cmov 0
#endif
/* We need these symbols in tcg-target.h, and we can't properly conditionalize
it there. Therefore we always define the variable. */
bool have_bmi1;
bool have_popcnt;
bool have_avx1;
bool have_avx2;
bool have_avx512bw;
bool have_avx512dq;
bool have_avx512vbmi2;
bool have_avx512vl;
bool have_movbe;
#ifdef CONFIG_CPUID_H
static bool have_bmi2;
static bool have_lzcnt;
#else
# define have_bmi2 0
# define have_lzcnt 0
#endif
static const tcg_insn_unit *tb_ret_addr;
static bool patch_reloc(tcg_insn_unit *code_ptr, int type,
intptr_t value, intptr_t addend)
{
value += addend;
switch(type) {
case R_386_PC32:
value -= (uintptr_t)tcg_splitwx_to_rx(code_ptr);
if (value != (int32_t)value) {
return false;
}
/* FALLTHRU */
case R_386_32:
tcg_patch32(code_ptr, value);
break;
case R_386_PC8:
value -= (uintptr_t)tcg_splitwx_to_rx(code_ptr);
if (value != (int8_t)value) {
return false;
}
tcg_patch8(code_ptr, value);
break;
default:
g_assert_not_reached();
}
return true;
}
/* test if a constant matches the constraint */
static bool tcg_target_const_match(int64_t val, TCGType type, int ct)
{
if (ct & TCG_CT_CONST) {
return 1;
}
if (type == TCG_TYPE_I32) {
if (ct & (TCG_CT_CONST_S32 | TCG_CT_CONST_U32 | TCG_CT_CONST_I32)) {
return 1;
}
} else {
if ((ct & TCG_CT_CONST_S32) && val == (int32_t)val) {
return 1;
}
if ((ct & TCG_CT_CONST_U32) && val == (uint32_t)val) {
return 1;
}
if ((ct & TCG_CT_CONST_I32) && ~val == (int32_t)~val) {
return 1;
}
}
if ((ct & TCG_CT_CONST_WSZ) && val == (type == TCG_TYPE_I32 ? 32 : 64)) {
return 1;
}
return 0;
}
# define LOWREGMASK(x) ((x) & 7)
#define P_EXT 0x100 /* 0x0f opcode prefix */
#define P_EXT38 0x200 /* 0x0f 0x38 opcode prefix */
#define P_DATA16 0x400 /* 0x66 opcode prefix */
#define P_VEXW 0x1000 /* Set VEX.W = 1 */
#if TCG_TARGET_REG_BITS == 64
# define P_REXW P_VEXW /* Set REX.W = 1; match VEXW */
# define P_REXB_R 0x2000 /* REG field as byte register */
# define P_REXB_RM 0x4000 /* R/M field as byte register */
# define P_GS 0x8000 /* gs segment override */
#else
# define P_REXW 0
# define P_REXB_R 0
# define P_REXB_RM 0
# define P_GS 0
#endif
#define P_EXT3A 0x10000 /* 0x0f 0x3a opcode prefix */
#define P_SIMDF3 0x20000 /* 0xf3 opcode prefix */
#define P_SIMDF2 0x40000 /* 0xf2 opcode prefix */
#define P_VEXL 0x80000 /* Set VEX.L = 1 */
#define P_EVEX 0x100000 /* Requires EVEX encoding */
#define OPC_ARITH_EvIz (0x81)
#define OPC_ARITH_EvIb (0x83)
#define OPC_ARITH_GvEv (0x03) /* ... plus (ARITH_FOO << 3) */
#define OPC_ANDN (0xf2 | P_EXT38)
#define OPC_ADD_GvEv (OPC_ARITH_GvEv | (ARITH_ADD << 3))
#define OPC_AND_GvEv (OPC_ARITH_GvEv | (ARITH_AND << 3))
#define OPC_BLENDPS (0x0c | P_EXT3A | P_DATA16)
#define OPC_BSF (0xbc | P_EXT)
#define OPC_BSR (0xbd | P_EXT)
#define OPC_BSWAP (0xc8 | P_EXT)
#define OPC_CALL_Jz (0xe8)
#define OPC_CMOVCC (0x40 | P_EXT) /* ... plus condition code */
#define OPC_CMP_GvEv (OPC_ARITH_GvEv | (ARITH_CMP << 3))
#define OPC_DEC_r32 (0x48)
#define OPC_IMUL_GvEv (0xaf | P_EXT)
#define OPC_IMUL_GvEvIb (0x6b)
#define OPC_IMUL_GvEvIz (0x69)
#define OPC_INC_r32 (0x40)
#define OPC_JCC_long (0x80 | P_EXT) /* ... plus condition code */
#define OPC_JCC_short (0x70) /* ... plus condition code */
#define OPC_JMP_long (0xe9)
#define OPC_JMP_short (0xeb)
#define OPC_LEA (0x8d)
#define OPC_LZCNT (0xbd | P_EXT | P_SIMDF3)
#define OPC_MOVB_EvGv (0x88) /* stores, more or less */
#define OPC_MOVL_EvGv (0x89) /* stores, more or less */
#define OPC_MOVL_GvEv (0x8b) /* loads, more or less */
#define OPC_MOVB_EvIz (0xc6)
#define OPC_MOVL_EvIz (0xc7)
#define OPC_MOVL_Iv (0xb8)
#define OPC_MOVBE_GyMy (0xf0 | P_EXT38)
#define OPC_MOVBE_MyGy (0xf1 | P_EXT38)
#define OPC_MOVD_VyEy (0x6e | P_EXT | P_DATA16)
#define OPC_MOVD_EyVy (0x7e | P_EXT | P_DATA16)
#define OPC_MOVDDUP (0x12 | P_EXT | P_SIMDF2)
#define OPC_MOVDQA_VxWx (0x6f | P_EXT | P_DATA16)
#define OPC_MOVDQA_WxVx (0x7f | P_EXT | P_DATA16)
#define OPC_MOVDQU_VxWx (0x6f | P_EXT | P_SIMDF3)
#define OPC_MOVDQU_WxVx (0x7f | P_EXT | P_SIMDF3)
#define OPC_MOVQ_VqWq (0x7e | P_EXT | P_SIMDF3)
#define OPC_MOVQ_WqVq (0xd6 | P_EXT | P_DATA16)
#define OPC_MOVSBL (0xbe | P_EXT)
#define OPC_MOVSWL (0xbf | P_EXT)
#define OPC_MOVSLQ (0x63 | P_REXW)
#define OPC_MOVZBL (0xb6 | P_EXT)
#define OPC_MOVZWL (0xb7 | P_EXT)
#define OPC_PABSB (0x1c | P_EXT38 | P_DATA16)
#define OPC_PABSW (0x1d | P_EXT38 | P_DATA16)
#define OPC_PABSD (0x1e | P_EXT38 | P_DATA16)
#define OPC_VPABSQ (0x1f | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_PACKSSDW (0x6b | P_EXT | P_DATA16)
#define OPC_PACKSSWB (0x63 | P_EXT | P_DATA16)
#define OPC_PACKUSDW (0x2b | P_EXT38 | P_DATA16)
#define OPC_PACKUSWB (0x67 | P_EXT | P_DATA16)
#define OPC_PADDB (0xfc | P_EXT | P_DATA16)
#define OPC_PADDW (0xfd | P_EXT | P_DATA16)
#define OPC_PADDD (0xfe | P_EXT | P_DATA16)
#define OPC_PADDQ (0xd4 | P_EXT | P_DATA16)
#define OPC_PADDSB (0xec | P_EXT | P_DATA16)
#define OPC_PADDSW (0xed | P_EXT | P_DATA16)
#define OPC_PADDUB (0xdc | P_EXT | P_DATA16)
#define OPC_PADDUW (0xdd | P_EXT | P_DATA16)
#define OPC_PAND (0xdb | P_EXT | P_DATA16)
#define OPC_PANDN (0xdf | P_EXT | P_DATA16)
#define OPC_PBLENDW (0x0e | P_EXT3A | P_DATA16)
#define OPC_PCMPEQB (0x74 | P_EXT | P_DATA16)
#define OPC_PCMPEQW (0x75 | P_EXT | P_DATA16)
#define OPC_PCMPEQD (0x76 | P_EXT | P_DATA16)
#define OPC_PCMPEQQ (0x29 | P_EXT38 | P_DATA16)
#define OPC_PCMPGTB (0x64 | P_EXT | P_DATA16)
#define OPC_PCMPGTW (0x65 | P_EXT | P_DATA16)
#define OPC_PCMPGTD (0x66 | P_EXT | P_DATA16)
#define OPC_PCMPGTQ (0x37 | P_EXT38 | P_DATA16)
#define OPC_PMAXSB (0x3c | P_EXT38 | P_DATA16)
#define OPC_PMAXSW (0xee | P_EXT | P_DATA16)
#define OPC_PMAXSD (0x3d | P_EXT38 | P_DATA16)
#define OPC_VPMAXSQ (0x3d | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_PMAXUB (0xde | P_EXT | P_DATA16)
#define OPC_PMAXUW (0x3e | P_EXT38 | P_DATA16)
#define OPC_PMAXUD (0x3f | P_EXT38 | P_DATA16)
#define OPC_VPMAXUQ (0x3f | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_PMINSB (0x38 | P_EXT38 | P_DATA16)
#define OPC_PMINSW (0xea | P_EXT | P_DATA16)
#define OPC_PMINSD (0x39 | P_EXT38 | P_DATA16)
#define OPC_VPMINSQ (0x39 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_PMINUB (0xda | P_EXT | P_DATA16)
#define OPC_PMINUW (0x3a | P_EXT38 | P_DATA16)
#define OPC_PMINUD (0x3b | P_EXT38 | P_DATA16)
#define OPC_VPMINUQ (0x3b | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_PMOVSXBW (0x20 | P_EXT38 | P_DATA16)
#define OPC_PMOVSXWD (0x23 | P_EXT38 | P_DATA16)
#define OPC_PMOVSXDQ (0x25 | P_EXT38 | P_DATA16)
#define OPC_PMOVZXBW (0x30 | P_EXT38 | P_DATA16)
#define OPC_PMOVZXWD (0x33 | P_EXT38 | P_DATA16)
#define OPC_PMOVZXDQ (0x35 | P_EXT38 | P_DATA16)
#define OPC_PMULLW (0xd5 | P_EXT | P_DATA16)
#define OPC_PMULLD (0x40 | P_EXT38 | P_DATA16)
#define OPC_VPMULLQ (0x40 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_POR (0xeb | P_EXT | P_DATA16)
#define OPC_PSHUFB (0x00 | P_EXT38 | P_DATA16)
#define OPC_PSHUFD (0x70 | P_EXT | P_DATA16)
#define OPC_PSHUFLW (0x70 | P_EXT | P_SIMDF2)
#define OPC_PSHUFHW (0x70 | P_EXT | P_SIMDF3)
#define OPC_PSHIFTW_Ib (0x71 | P_EXT | P_DATA16) /* /2 /6 /4 */
#define OPC_PSHIFTD_Ib (0x72 | P_EXT | P_DATA16) /* /1 /2 /6 /4 */
#define OPC_PSHIFTQ_Ib (0x73 | P_EXT | P_DATA16) /* /2 /6 /4 */
#define OPC_PSLLW (0xf1 | P_EXT | P_DATA16)
#define OPC_PSLLD (0xf2 | P_EXT | P_DATA16)
#define OPC_PSLLQ (0xf3 | P_EXT | P_DATA16)
#define OPC_PSRAW (0xe1 | P_EXT | P_DATA16)
#define OPC_PSRAD (0xe2 | P_EXT | P_DATA16)
#define OPC_VPSRAQ (0xe2 | P_EXT | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_PSRLW (0xd1 | P_EXT | P_DATA16)
#define OPC_PSRLD (0xd2 | P_EXT | P_DATA16)
#define OPC_PSRLQ (0xd3 | P_EXT | P_DATA16)
#define OPC_PSUBB (0xf8 | P_EXT | P_DATA16)
#define OPC_PSUBW (0xf9 | P_EXT | P_DATA16)
#define OPC_PSUBD (0xfa | P_EXT | P_DATA16)
#define OPC_PSUBQ (0xfb | P_EXT | P_DATA16)
#define OPC_PSUBSB (0xe8 | P_EXT | P_DATA16)
#define OPC_PSUBSW (0xe9 | P_EXT | P_DATA16)
#define OPC_PSUBUB (0xd8 | P_EXT | P_DATA16)
#define OPC_PSUBUW (0xd9 | P_EXT | P_DATA16)
#define OPC_PUNPCKLBW (0x60 | P_EXT | P_DATA16)
#define OPC_PUNPCKLWD (0x61 | P_EXT | P_DATA16)
#define OPC_PUNPCKLDQ (0x62 | P_EXT | P_DATA16)
#define OPC_PUNPCKLQDQ (0x6c | P_EXT | P_DATA16)
#define OPC_PUNPCKHBW (0x68 | P_EXT | P_DATA16)
#define OPC_PUNPCKHWD (0x69 | P_EXT | P_DATA16)
#define OPC_PUNPCKHDQ (0x6a | P_EXT | P_DATA16)
#define OPC_PUNPCKHQDQ (0x6d | P_EXT | P_DATA16)
#define OPC_PXOR (0xef | P_EXT | P_DATA16)
#define OPC_POP_r32 (0x58)
#define OPC_POPCNT (0xb8 | P_EXT | P_SIMDF3)
#define OPC_PUSH_r32 (0x50)
#define OPC_PUSH_Iv (0x68)
#define OPC_PUSH_Ib (0x6a)
#define OPC_RET (0xc3)
#define OPC_SETCC (0x90 | P_EXT | P_REXB_RM) /* ... plus cc */
#define OPC_SHIFT_1 (0xd1)
#define OPC_SHIFT_Ib (0xc1)
#define OPC_SHIFT_cl (0xd3)
#define OPC_SARX (0xf7 | P_EXT38 | P_SIMDF3)
#define OPC_SHUFPS (0xc6 | P_EXT)
#define OPC_SHLX (0xf7 | P_EXT38 | P_DATA16)
#define OPC_SHRX (0xf7 | P_EXT38 | P_SIMDF2)
#define OPC_SHRD_Ib (0xac | P_EXT)
#define OPC_TESTL (0x85)
#define OPC_TZCNT (0xbc | P_EXT | P_SIMDF3)
#define OPC_UD2 (0x0b | P_EXT)
#define OPC_VPBLENDD (0x02 | P_EXT3A | P_DATA16)
#define OPC_VPBLENDVB (0x4c | P_EXT3A | P_DATA16)
#define OPC_VPINSRB (0x20 | P_EXT3A | P_DATA16)
#define OPC_VPINSRW (0xc4 | P_EXT | P_DATA16)
#define OPC_VBROADCASTSS (0x18 | P_EXT38 | P_DATA16)
#define OPC_VBROADCASTSD (0x19 | P_EXT38 | P_DATA16)
#define OPC_VPBROADCASTB (0x78 | P_EXT38 | P_DATA16)
#define OPC_VPBROADCASTW (0x79 | P_EXT38 | P_DATA16)
#define OPC_VPBROADCASTD (0x58 | P_EXT38 | P_DATA16)
#define OPC_VPBROADCASTQ (0x59 | P_EXT38 | P_DATA16)
#define OPC_VPERMQ (0x00 | P_EXT3A | P_DATA16 | P_VEXW)
#define OPC_VPERM2I128 (0x46 | P_EXT3A | P_DATA16 | P_VEXL)
#define OPC_VPROLVD (0x15 | P_EXT38 | P_DATA16 | P_EVEX)
#define OPC_VPROLVQ (0x15 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPRORVD (0x14 | P_EXT38 | P_DATA16 | P_EVEX)
#define OPC_VPRORVQ (0x14 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSHLDW (0x70 | P_EXT3A | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSHLDD (0x71 | P_EXT3A | P_DATA16 | P_EVEX)
#define OPC_VPSHLDQ (0x71 | P_EXT3A | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSHLDVW (0x70 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSHLDVD (0x71 | P_EXT38 | P_DATA16 | P_EVEX)
#define OPC_VPSHLDVQ (0x71 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSHRDVW (0x72 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSHRDVD (0x73 | P_EXT38 | P_DATA16 | P_EVEX)
#define OPC_VPSHRDVQ (0x73 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSLLVW (0x12 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSLLVD (0x47 | P_EXT38 | P_DATA16)
#define OPC_VPSLLVQ (0x47 | P_EXT38 | P_DATA16 | P_VEXW)
#define OPC_VPSRAVW (0x11 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSRAVD (0x46 | P_EXT38 | P_DATA16)
#define OPC_VPSRAVQ (0x46 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSRLVW (0x10 | P_EXT38 | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VPSRLVD (0x45 | P_EXT38 | P_DATA16)
#define OPC_VPSRLVQ (0x45 | P_EXT38 | P_DATA16 | P_VEXW)
#define OPC_VPTERNLOGQ (0x25 | P_EXT3A | P_DATA16 | P_VEXW | P_EVEX)
#define OPC_VZEROUPPER (0x77 | P_EXT)
#define OPC_XCHG_ax_r32 (0x90)
#define OPC_GRP3_Eb (0xf6)
#define OPC_GRP3_Ev (0xf7)
#define OPC_GRP5 (0xff)
#define OPC_GRP14 (0x73 | P_EXT | P_DATA16)
/* Group 1 opcode extensions for 0x80-0x83.
These are also used as modifiers for OPC_ARITH. */
#define ARITH_ADD 0
#define ARITH_OR 1
#define ARITH_ADC 2
#define ARITH_SBB 3
#define ARITH_AND 4
#define ARITH_SUB 5
#define ARITH_XOR 6
#define ARITH_CMP 7
/* Group 2 opcode extensions for 0xc0, 0xc1, 0xd0-0xd3. */
#define SHIFT_ROL 0
#define SHIFT_ROR 1
#define SHIFT_SHL 4
#define SHIFT_SHR 5
#define SHIFT_SAR 7
/* Group 3 opcode extensions for 0xf6, 0xf7. To be used with OPC_GRP3. */
#define EXT3_TESTi 0
#define EXT3_NOT 2
#define EXT3_NEG 3
#define EXT3_MUL 4
#define EXT3_IMUL 5
#define EXT3_DIV 6
#define EXT3_IDIV 7
/* Group 5 opcode extensions for 0xff. To be used with OPC_GRP5. */
#define EXT5_INC_Ev 0
#define EXT5_DEC_Ev 1
#define EXT5_CALLN_Ev 2
#define EXT5_JMPN_Ev 4
/* Condition codes to be added to OPC_JCC_{long,short}. */
#define JCC_JMP (-1)
#define JCC_JO 0x0
#define JCC_JNO 0x1
#define JCC_JB 0x2
#define JCC_JAE 0x3
#define JCC_JE 0x4
#define JCC_JNE 0x5
#define JCC_JBE 0x6
#define JCC_JA 0x7
#define JCC_JS 0x8
#define JCC_JNS 0x9
#define JCC_JP 0xa
#define JCC_JNP 0xb
#define JCC_JL 0xc
#define JCC_JGE 0xd
#define JCC_JLE 0xe
#define JCC_JG 0xf
static const uint8_t tcg_cond_to_jcc[] = {
[TCG_COND_EQ] = JCC_JE,
[TCG_COND_NE] = JCC_JNE,
[TCG_COND_LT] = JCC_JL,
[TCG_COND_GE] = JCC_JGE,
[TCG_COND_LE] = JCC_JLE,
[TCG_COND_GT] = JCC_JG,
[TCG_COND_LTU] = JCC_JB,
[TCG_COND_GEU] = JCC_JAE,
[TCG_COND_LEU] = JCC_JBE,
[TCG_COND_GTU] = JCC_JA,
};
#if TCG_TARGET_REG_BITS == 64
static void tcg_out_opc(TCGContext *s, int opc, int r, int rm, int x)
{
int rex;
if (opc & P_GS) {
tcg_out8(s, 0x65);
}
if (opc & P_DATA16) {
/* We should never be asking for both 16 and 64-bit operation. */
tcg_debug_assert((opc & P_REXW) == 0);
tcg_out8(s, 0x66);
}
if (opc & P_SIMDF3) {
tcg_out8(s, 0xf3);
} else if (opc & P_SIMDF2) {
tcg_out8(s, 0xf2);
}
rex = 0;
rex |= (opc & P_REXW) ? 0x8 : 0x0; /* REX.W */
rex |= (r & 8) >> 1; /* REX.R */
rex |= (x & 8) >> 2; /* REX.X */
rex |= (rm & 8) >> 3; /* REX.B */
/* P_REXB_{R,RM} indicates that the given register is the low byte.
For %[abcd]l we need no REX prefix, but for %{si,di,bp,sp}l we do,
as otherwise the encoding indicates %[abcd]h. Note that the values
that are ORed in merely indicate that the REX byte must be present;
those bits get discarded in output. */
rex |= opc & (r >= 4 ? P_REXB_R : 0);
rex |= opc & (rm >= 4 ? P_REXB_RM : 0);
if (rex) {
tcg_out8(s, (uint8_t)(rex | 0x40));
}
if (opc & (P_EXT | P_EXT38 | P_EXT3A)) {
tcg_out8(s, 0x0f);
if (opc & P_EXT38) {
tcg_out8(s, 0x38);
} else if (opc & P_EXT3A) {
tcg_out8(s, 0x3a);
}
}
tcg_out8(s, opc);
}
#else
static void tcg_out_opc(TCGContext *s, int opc)
{
if (opc & P_DATA16) {
tcg_out8(s, 0x66);
}
if (opc & P_SIMDF3) {
tcg_out8(s, 0xf3);
} else if (opc & P_SIMDF2) {
tcg_out8(s, 0xf2);
}
if (opc & (P_EXT | P_EXT38 | P_EXT3A)) {
tcg_out8(s, 0x0f);
if (opc & P_EXT38) {
tcg_out8(s, 0x38);
} else if (opc & P_EXT3A) {
tcg_out8(s, 0x3a);
}
}
tcg_out8(s, opc);
}
/* Discard the register arguments to tcg_out_opc early, so as not to penalize
the 32-bit compilation paths. This method works with all versions of gcc,
whereas relying on optimization may not be able to exclude them. */
#define tcg_out_opc(s, opc, r, rm, x) (tcg_out_opc)(s, opc)
#endif
static void tcg_out_modrm(TCGContext *s, int opc, int r, int rm)
{
tcg_out_opc(s, opc, r, rm, 0);
tcg_out8(s, 0xc0 | (LOWREGMASK(r) << 3) | LOWREGMASK(rm));
}
static void tcg_out_vex_opc(TCGContext *s, int opc, int r, int v,
int rm, int index)
{
int tmp;
/* Use the two byte form if possible, which cannot encode
VEX.W, VEX.B, VEX.X, or an m-mmmm field other than P_EXT. */
if ((opc & (P_EXT | P_EXT38 | P_EXT3A | P_VEXW)) == P_EXT
&& ((rm | index) & 8) == 0) {
/* Two byte VEX prefix. */
tcg_out8(s, 0xc5);
tmp = (r & 8 ? 0 : 0x80); /* VEX.R */
} else {
/* Three byte VEX prefix. */
tcg_out8(s, 0xc4);
/* VEX.m-mmmm */
if (opc & P_EXT3A) {
tmp = 3;
} else if (opc & P_EXT38) {
tmp = 2;
} else if (opc & P_EXT) {
tmp = 1;
} else {
g_assert_not_reached();
}
tmp |= (r & 8 ? 0 : 0x80); /* VEX.R */
tmp |= (index & 8 ? 0 : 0x40); /* VEX.X */
tmp |= (rm & 8 ? 0 : 0x20); /* VEX.B */
tcg_out8(s, tmp);
tmp = (opc & P_VEXW ? 0x80 : 0); /* VEX.W */
}
tmp |= (opc & P_VEXL ? 0x04 : 0); /* VEX.L */
/* VEX.pp */
if (opc & P_DATA16) {
tmp |= 1; /* 0x66 */
} else if (opc & P_SIMDF3) {
tmp |= 2; /* 0xf3 */
} else if (opc & P_SIMDF2) {
tmp |= 3; /* 0xf2 */
}
tmp |= (~v & 15) << 3; /* VEX.vvvv */
tcg_out8(s, tmp);
tcg_out8(s, opc);
}
static void tcg_out_evex_opc(TCGContext *s, int opc, int r, int v,
int rm, int index)
{
/* The entire 4-byte evex prefix; with R' and V' set. */
uint32_t p = 0x08041062;
int mm, pp;
tcg_debug_assert(have_avx512vl);
/* EVEX.mm */
if (opc & P_EXT3A) {
mm = 3;
} else if (opc & P_EXT38) {
mm = 2;
} else if (opc & P_EXT) {
mm = 1;
} else {
g_assert_not_reached();
}
/* EVEX.pp */
if (opc & P_DATA16) {
pp = 1; /* 0x66 */
} else if (opc & P_SIMDF3) {
pp = 2; /* 0xf3 */
} else if (opc & P_SIMDF2) {
pp = 3; /* 0xf2 */
} else {
pp = 0;
}
p = deposit32(p, 8, 2, mm);
p = deposit32(p, 13, 1, (rm & 8) == 0); /* EVEX.RXB.B */
p = deposit32(p, 14, 1, (index & 8) == 0); /* EVEX.RXB.X */
p = deposit32(p, 15, 1, (r & 8) == 0); /* EVEX.RXB.R */
p = deposit32(p, 16, 2, pp);
p = deposit32(p, 19, 4, ~v);
p = deposit32(p, 23, 1, (opc & P_VEXW) != 0);
p = deposit32(p, 29, 2, (opc & P_VEXL) != 0);
tcg_out32(s, p);
tcg_out8(s, opc);
}
static void tcg_out_vex_modrm(TCGContext *s, int opc, int r, int v, int rm)
{
if (opc & P_EVEX) {
tcg_out_evex_opc(s, opc, r, v, rm, 0);
} else {
tcg_out_vex_opc(s, opc, r, v, rm, 0);
}
tcg_out8(s, 0xc0 | (LOWREGMASK(r) << 3) | LOWREGMASK(rm));
}
/* Output an opcode with a full "rm + (index<<shift) + offset" address mode.
We handle either RM and INDEX missing with a negative value. In 64-bit
mode for absolute addresses, ~RM is the size of the immediate operand
that will follow the instruction. */
static void tcg_out_sib_offset(TCGContext *s, int r, int rm, int index,
int shift, intptr_t offset)
{
int mod, len;
if (index < 0 && rm < 0) {
if (TCG_TARGET_REG_BITS == 64) {
/* Try for a rip-relative addressing mode. This has replaced
the 32-bit-mode absolute addressing encoding. */
intptr_t pc = (intptr_t)s->code_ptr + 5 + ~rm;
intptr_t disp = offset - pc;
if (disp == (int32_t)disp) {
tcg_out8(s, (LOWREGMASK(r) << 3) | 5);
tcg_out32(s, disp);
return;
}
/* Try for an absolute address encoding. This requires the
use of the MODRM+SIB encoding and is therefore larger than
rip-relative addressing. */
if (offset == (int32_t)offset) {
tcg_out8(s, (LOWREGMASK(r) << 3) | 4);
tcg_out8(s, (4 << 3) | 5);
tcg_out32(s, offset);
return;
}
/* ??? The memory isn't directly addressable. */
g_assert_not_reached();
} else {
/* Absolute address. */
tcg_out8(s, (r << 3) | 5);
tcg_out32(s, offset);
return;
}
}
/* Find the length of the immediate addend. Note that the encoding
that would be used for (%ebp) indicates absolute addressing. */
if (rm < 0) {
mod = 0, len = 4, rm = 5;
} else if (offset == 0 && LOWREGMASK(rm) != TCG_REG_EBP) {
mod = 0, len = 0;
} else if (offset == (int8_t)offset) {
mod = 0x40, len = 1;
} else {
mod = 0x80, len = 4;
}
/* Use a single byte MODRM format if possible. Note that the encoding
that would be used for %esp is the escape to the two byte form. */
if (index < 0 && LOWREGMASK(rm) != TCG_REG_ESP) {
/* Single byte MODRM format. */
tcg_out8(s, mod | (LOWREGMASK(r) << 3) | LOWREGMASK(rm));
} else {
/* Two byte MODRM+SIB format. */
/* Note that the encoding that would place %esp into the index
field indicates no index register. In 64-bit mode, the REX.X
bit counts, so %r12 can be used as the index. */
if (index < 0) {
index = 4;
} else {
tcg_debug_assert(index != TCG_REG_ESP);
}
tcg_out8(s, mod | (LOWREGMASK(r) << 3) | 4);
tcg_out8(s, (shift << 6) | (LOWREGMASK(index) << 3) | LOWREGMASK(rm));
}
if (len == 1) {
tcg_out8(s, offset);
} else if (len == 4) {
tcg_out32(s, offset);
}
}
static void tcg_out_modrm_sib_offset(TCGContext *s, int opc, int r, int rm,
int index, int shift, intptr_t offset)
{
tcg_out_opc(s, opc, r, rm < 0 ? 0 : rm, index < 0 ? 0 : index);
tcg_out_sib_offset(s, r, rm, index, shift, offset);
}
static void tcg_out_vex_modrm_sib_offset(TCGContext *s, int opc, int r, int v,
int rm, int index, int shift,
intptr_t offset)
{
tcg_out_vex_opc(s, opc, r, v, rm < 0 ? 0 : rm, index < 0 ? 0 : index);
tcg_out_sib_offset(s, r, rm, index, shift, offset);
}
/* A simplification of the above with no index or shift. */
static inline void tcg_out_modrm_offset(TCGContext *s, int opc, int r,
int rm, intptr_t offset)
{
tcg_out_modrm_sib_offset(s, opc, r, rm, -1, 0, offset);
}
static inline void tcg_out_vex_modrm_offset(TCGContext *s, int opc, int r,
int v, int rm, intptr_t offset)
{
tcg_out_vex_modrm_sib_offset(s, opc, r, v, rm, -1, 0, offset);
}
/* Output an opcode with an expected reference to the constant pool. */
static inline void tcg_out_modrm_pool(TCGContext *s, int opc, int r)
{
tcg_out_opc(s, opc, r, 0, 0);
/* Absolute for 32-bit, pc-relative for 64-bit. */
tcg_out8(s, LOWREGMASK(r) << 3 | 5);
tcg_out32(s, 0);
}
/* Output an opcode with an expected reference to the constant pool. */
static inline void tcg_out_vex_modrm_pool(TCGContext *s, int opc, int r)
{
tcg_out_vex_opc(s, opc, r, 0, 0, 0);
/* Absolute for 32-bit, pc-relative for 64-bit. */
tcg_out8(s, LOWREGMASK(r) << 3 | 5);
tcg_out32(s, 0);
}
/* Generate dest op= src. Uses the same ARITH_* codes as tgen_arithi. */
static inline void tgen_arithr(TCGContext *s, int subop, int dest, int src)
{
/* Propagate an opcode prefix, such as P_REXW. */
int ext = subop & ~0x7;
subop &= 0x7;
tcg_out_modrm(s, OPC_ARITH_GvEv + (subop << 3) + ext, dest, src);
}
static bool tcg_out_mov(TCGContext *s, TCGType type, TCGReg ret, TCGReg arg)
{
int rexw = 0;
if (arg == ret) {
return true;
}
switch (type) {
case TCG_TYPE_I64:
rexw = P_REXW;
/* fallthru */
case TCG_TYPE_I32:
if (ret < 16) {
if (arg < 16) {
tcg_out_modrm(s, OPC_MOVL_GvEv + rexw, ret, arg);
} else {
tcg_out_vex_modrm(s, OPC_MOVD_EyVy + rexw, arg, 0, ret);
}
} else {
if (arg < 16) {
tcg_out_vex_modrm(s, OPC_MOVD_VyEy + rexw, ret, 0, arg);
} else {
tcg_out_vex_modrm(s, OPC_MOVQ_VqWq, ret, 0, arg);
}
}
break;
case TCG_TYPE_V64:
tcg_debug_assert(ret >= 16 && arg >= 16);
tcg_out_vex_modrm(s, OPC_MOVQ_VqWq, ret, 0, arg);
break;
case TCG_TYPE_V128:
tcg_debug_assert(ret >= 16 && arg >= 16);
tcg_out_vex_modrm(s, OPC_MOVDQA_VxWx, ret, 0, arg);
break;
case TCG_TYPE_V256:
tcg_debug_assert(ret >= 16 && arg >= 16);
tcg_out_vex_modrm(s, OPC_MOVDQA_VxWx | P_VEXL, ret, 0, arg);
break;
default:
g_assert_not_reached();
}
return true;
}
static const int avx2_dup_insn[4] = {
OPC_VPBROADCASTB, OPC_VPBROADCASTW,
OPC_VPBROADCASTD, OPC_VPBROADCASTQ,
};
static bool tcg_out_dup_vec(TCGContext *s, TCGType type, unsigned vece,
TCGReg r, TCGReg a)
{
if (have_avx2) {
int vex_l = (type == TCG_TYPE_V256 ? P_VEXL : 0);
tcg_out_vex_modrm(s, avx2_dup_insn[vece] + vex_l, r, 0, a);
} else {
switch (vece) {
case MO_8:
/* ??? With zero in a register, use PSHUFB. */
tcg_out_vex_modrm(s, OPC_PUNPCKLBW, r, a, a);
a = r;
/* FALLTHRU */
case MO_16:
tcg_out_vex_modrm(s, OPC_PUNPCKLWD, r, a, a);
a = r;
/* FALLTHRU */
case MO_32:
tcg_out_vex_modrm(s, OPC_PSHUFD, r, 0, a);
/* imm8 operand: all output lanes selected from input lane 0. */
tcg_out8(s, 0);
break;
case MO_64:
tcg_out_vex_modrm(s, OPC_PUNPCKLQDQ, r, a, a);
break;
default:
g_assert_not_reached();
}
}
return true;
}
static bool tcg_out_dupm_vec(TCGContext *s, TCGType type, unsigned vece,
TCGReg r, TCGReg base, intptr_t offset)
{
if (have_avx2) {
int vex_l = (type == TCG_TYPE_V256 ? P_VEXL : 0);
tcg_out_vex_modrm_offset(s, avx2_dup_insn[vece] + vex_l,
r, 0, base, offset);
} else {
switch (vece) {
case MO_64:
tcg_out_vex_modrm_offset(s, OPC_MOVDDUP, r, 0, base, offset);
break;
case MO_32:
tcg_out_vex_modrm_offset(s, OPC_VBROADCASTSS, r, 0, base, offset);
break;
case MO_16:
tcg_out_vex_modrm_offset(s, OPC_VPINSRW, r, r, base, offset);
tcg_out8(s, 0); /* imm8 */
tcg_out_dup_vec(s, type, vece, r, r);
break;
case MO_8:
tcg_out_vex_modrm_offset(s, OPC_VPINSRB, r, r, base, offset);
tcg_out8(s, 0); /* imm8 */
tcg_out_dup_vec(s, type, vece, r, r);
break;
default:
g_assert_not_reached();
}
}
return true;
}
static void tcg_out_dupi_vec(TCGContext *s, TCGType type, unsigned vece,
TCGReg ret, int64_t arg)
{
int vex_l = (type == TCG_TYPE_V256 ? P_VEXL : 0);
if (arg == 0) {
tcg_out_vex_modrm(s, OPC_PXOR, ret, ret, ret);
return;
}
if (arg == -1) {
tcg_out_vex_modrm(s, OPC_PCMPEQB + vex_l, ret, ret, ret);
return;
}
if (TCG_TARGET_REG_BITS == 32 && vece < MO_64) {
if (have_avx2) {
tcg_out_vex_modrm_pool(s, OPC_VPBROADCASTD + vex_l, ret);
} else {
tcg_out_vex_modrm_pool(s, OPC_VBROADCASTSS, ret);
}
new_pool_label(s, arg, R_386_32, s->code_ptr - 4, 0);
} else {
if (type == TCG_TYPE_V64) {
tcg_out_vex_modrm_pool(s, OPC_MOVQ_VqWq, ret);
} else if (have_avx2) {
tcg_out_vex_modrm_pool(s, OPC_VPBROADCASTQ + vex_l, ret);
} else {
tcg_out_vex_modrm_pool(s, OPC_MOVDDUP, ret);
}
if (TCG_TARGET_REG_BITS == 64) {
new_pool_label(s, arg, R_386_PC32, s->code_ptr - 4, -4);
} else {
new_pool_l2(s, R_386_32, s->code_ptr - 4, 0, arg, arg >> 32);
}
}
}
static void tcg_out_movi_vec(TCGContext *s, TCGType type,
TCGReg ret, tcg_target_long arg)
{
if (arg == 0) {
tcg_out_vex_modrm(s, OPC_PXOR, ret, ret, ret);
return;
}
if (arg == -1) {
tcg_out_vex_modrm(s, OPC_PCMPEQB, ret, ret, ret);
return;
}
int rexw = (type == TCG_TYPE_I32 ? 0 : P_REXW);
tcg_out_vex_modrm_pool(s, OPC_MOVD_VyEy + rexw, ret);
if (TCG_TARGET_REG_BITS == 64) {
new_pool_label(s, arg, R_386_PC32, s->code_ptr - 4, -4);
} else {
new_pool_label(s, arg, R_386_32, s->code_ptr - 4, 0);
}
}
static void tcg_out_movi_int(TCGContext *s, TCGType type,
TCGReg ret, tcg_target_long arg)
{
tcg_target_long diff;
if (arg == 0) {
tgen_arithr(s, ARITH_XOR, ret, ret);
return;
}
if (arg == (uint32_t)arg || type == TCG_TYPE_I32) {
tcg_out_opc(s, OPC_MOVL_Iv + LOWREGMASK(ret), 0, ret, 0);
tcg_out32(s, arg);
return;
}
if (arg == (int32_t)arg) {
tcg_out_modrm(s, OPC_MOVL_EvIz + P_REXW, 0, ret);
tcg_out32(s, arg);
return;
}
/* Try a 7 byte pc-relative lea before the 10 byte movq. */
diff = tcg_pcrel_diff(s, (const void *)arg) - 7;
if (diff == (int32_t)diff) {
tcg_out_opc(s, OPC_LEA | P_REXW, ret, 0, 0);
tcg_out8(s, (LOWREGMASK(ret) << 3) | 5);
tcg_out32(s, diff);
return;
}
tcg_out_opc(s, OPC_MOVL_Iv + P_REXW + LOWREGMASK(ret), 0, ret, 0);
tcg_out64(s, arg);
}
static void tcg_out_movi(TCGContext *s, TCGType type,
TCGReg ret, tcg_target_long arg)
{
switch (type) {
case TCG_TYPE_I32:
#if TCG_TARGET_REG_BITS == 64
case TCG_TYPE_I64:
#endif
if (ret < 16) {
tcg_out_movi_int(s, type, ret, arg);
} else {
tcg_out_movi_vec(s, type, ret, arg);
}
break;
default:
g_assert_not_reached();
}
}
static void tcg_out_addi_ptr(TCGContext *s, TCGReg rd, TCGReg rs,
tcg_target_long imm)
{
/* This function is only used for passing structs by reference. */
tcg_debug_assert(imm == (int32_t)imm);
tcg_out_modrm_offset(s, OPC_LEA, rd, rs, imm);
}
static inline void tcg_out_pushi(TCGContext *s, tcg_target_long val)
{
if (val == (int8_t)val) {
tcg_out_opc(s, OPC_PUSH_Ib, 0, 0, 0);
tcg_out8(s, val);
} else if (val == (int32_t)val) {
tcg_out_opc(s, OPC_PUSH_Iv, 0, 0, 0);
tcg_out32(s, val);
} else {
g_assert_not_reached();
}
}
static inline void tcg_out_mb(TCGContext *s, TCGArg a0)
{
/* Given the strength of x86 memory ordering, we only need care for
store-load ordering. Experimentally, "lock orl $0,0(%esp)" is
faster than "mfence", so don't bother with the sse insn. */
if (a0 & TCG_MO_ST_LD) {
tcg_out8(s, 0xf0);
tcg_out_modrm_offset(s, OPC_ARITH_EvIb, ARITH_OR, TCG_REG_ESP, 0);
tcg_out8(s, 0);
}
}
static inline void tcg_out_push(TCGContext *s, int reg)
{
tcg_out_opc(s, OPC_PUSH_r32 + LOWREGMASK(reg), 0, reg, 0);
}
static inline void tcg_out_pop(TCGContext *s, int reg)
{
tcg_out_opc(s, OPC_POP_r32 + LOWREGMASK(reg), 0, reg, 0);
}
static void tcg_out_ld(TCGContext *s, TCGType type, TCGReg ret,
TCGReg arg1, intptr_t arg2)
{
switch (type) {
case TCG_TYPE_I32:
if (ret < 16) {
tcg_out_modrm_offset(s, OPC_MOVL_GvEv, ret, arg1, arg2);
} else {
tcg_out_vex_modrm_offset(s, OPC_MOVD_VyEy, ret, 0, arg1, arg2);
}
break;
case TCG_TYPE_I64:
if (ret < 16) {
tcg_out_modrm_offset(s, OPC_MOVL_GvEv | P_REXW, ret, arg1, arg2);
break;
}
/* FALLTHRU */
case TCG_TYPE_V64:
/* There is no instruction that can validate 8-byte alignment. */
tcg_debug_assert(ret >= 16);
tcg_out_vex_modrm_offset(s, OPC_MOVQ_VqWq, ret, 0, arg1, arg2);
break;
case TCG_TYPE_V128:
/*
* The gvec infrastructure is asserts that v128 vector loads
* and stores use a 16-byte aligned offset. Validate that the
* final pointer is aligned by using an insn that will SIGSEGV.
*/
tcg_debug_assert(ret >= 16);
tcg_out_vex_modrm_offset(s, OPC_MOVDQA_VxWx, ret, 0, arg1, arg2);
break;
case TCG_TYPE_V256:
/*
* The gvec infrastructure only requires 16-byte alignment,
* so here we must use an unaligned load.
*/
tcg_debug_assert(ret >= 16);
tcg_out_vex_modrm_offset(s, OPC_MOVDQU_VxWx | P_VEXL,
ret, 0, arg1, arg2);
break;
default:
g_assert_not_reached();
}
}
static void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg,
TCGReg arg1, intptr_t arg2)
{
switch (type) {
case TCG_TYPE_I32:
if (arg < 16) {
tcg_out_modrm_offset(s, OPC_MOVL_EvGv, arg, arg1, arg2);
} else {
tcg_out_vex_modrm_offset(s, OPC_MOVD_EyVy, arg, 0, arg1, arg2);
}
break;
case TCG_TYPE_I64:
if (arg < 16) {
tcg_out_modrm_offset(s, OPC_MOVL_EvGv | P_REXW, arg, arg1, arg2);
break;
}
/* FALLTHRU */
case TCG_TYPE_V64:
/* There is no instruction that can validate 8-byte alignment. */
tcg_debug_assert(arg >= 16);
tcg_out_vex_modrm_offset(s, OPC_MOVQ_WqVq, arg, 0, arg1, arg2);
break;
case TCG_TYPE_V128:
/*
* The gvec infrastructure is asserts that v128 vector loads
* and stores use a 16-byte aligned offset. Validate that the
* final pointer is aligned by using an insn that will SIGSEGV.
*
* This specific instance is also used by TCG_CALL_RET_BY_VEC,
* for _WIN64, which must have SSE2 but may not have AVX.
*/
tcg_debug_assert(arg >= 16);
if (have_avx1) {
tcg_out_vex_modrm_offset(s, OPC_MOVDQA_WxVx, arg, 0, arg1, arg2);
} else {
tcg_out_modrm_offset(s, OPC_MOVDQA_WxVx, arg, arg1, arg2);
}
break;
case TCG_TYPE_V256:
/*
* The gvec infrastructure only requires 16-byte alignment,
* so here we must use an unaligned store.
*/
tcg_debug_assert(arg >= 16);
tcg_out_vex_modrm_offset(s, OPC_MOVDQU_WxVx | P_VEXL,
arg, 0, arg1, arg2);
break;
default:
g_assert_not_reached();
}
}
static bool tcg_out_sti(TCGContext *s, TCGType type, TCGArg val,
TCGReg base, intptr_t ofs)
{
int rexw = 0;
if (TCG_TARGET_REG_BITS == 64 && type == TCG_TYPE_I64) {
if (val != (int32_t)val) {
return false;
}
rexw = P_REXW;
} else if (type != TCG_TYPE_I32) {
return false;
}
tcg_out_modrm_offset(s, OPC_MOVL_EvIz | rexw, 0, base, ofs);
tcg_out32(s, val);
return true;
}
static void tcg_out_shifti(TCGContext *s, int subopc, int reg, int count)
{
/* Propagate an opcode prefix, such as P_DATA16. */
int ext = subopc & ~0x7;
subopc &= 0x7;
if (count == 1) {
tcg_out_modrm(s, OPC_SHIFT_1 + ext, subopc, reg);
} else {
tcg_out_modrm(s, OPC_SHIFT_Ib + ext, subopc, reg);
tcg_out8(s, count);
}
}
static inline void tcg_out_bswap32(TCGContext *s, int reg)
{
tcg_out_opc(s, OPC_BSWAP + LOWREGMASK(reg), 0, reg, 0);
}
static inline void tcg_out_rolw_8(TCGContext *s, int reg)
{
tcg_out_shifti(s, SHIFT_ROL + P_DATA16, reg, 8);
}
static void tcg_out_ext8u(TCGContext *s, TCGReg dest, TCGReg src)
{
/* movzbl */
tcg_debug_assert(src < 4 || TCG_TARGET_REG_BITS == 64);
tcg_out_modrm(s, OPC_MOVZBL + P_REXB_RM, dest, src);
}
static void tcg_out_ext8s(TCGContext *s, TCGType type, TCGReg dest, TCGReg src)
{
int rexw = type == TCG_TYPE_I32 ? 0 : P_REXW;
/* movsbl */
tcg_debug_assert(src < 4 || TCG_TARGET_REG_BITS == 64);
tcg_out_modrm(s, OPC_MOVSBL + P_REXB_RM + rexw, dest, src);
}
static inline void tcg_out_ext16u(TCGContext *s, int dest, int src)
{
/* movzwl */
tcg_out_modrm(s, OPC_MOVZWL, dest, src);
}
static inline void tcg_out_ext16s(TCGContext *s, int dest, int src, int rexw)
{
/* movsw[lq] */
tcg_out_modrm(s, OPC_MOVSWL + rexw, dest, src);
}
static inline void tcg_out_ext32u(TCGContext *s, int dest, int src)
{
/* 32-bit mov zero extends. */
tcg_out_modrm(s, OPC_MOVL_GvEv, dest, src);
}
static inline void tcg_out_ext32s(TCGContext *s, int dest, int src)
{
tcg_out_modrm(s, OPC_MOVSLQ, dest, src);
}
static inline void tcg_out_bswap64(TCGContext *s, int reg)
{
tcg_out_opc(s, OPC_BSWAP + P_REXW + LOWREGMASK(reg), 0, reg, 0);
}
static void tgen_arithi(TCGContext *s, int c, int r0,
tcg_target_long val, int cf)
{
int rexw = 0;
if (TCG_TARGET_REG_BITS == 64) {
rexw = c & -8;
c &= 7;
}
/* ??? While INC is 2 bytes shorter than ADDL $1, they also induce
partial flags update stalls on Pentium4 and are not recommended
by current Intel optimization manuals. */
if (!cf && (c == ARITH_ADD || c == ARITH_SUB) && (val == 1 || val == -1)) {
int is_inc = (c == ARITH_ADD) ^ (val < 0);
if (TCG_TARGET_REG_BITS == 64) {
/* The single-byte increment encodings are re-tasked as the
REX prefixes. Use the MODRM encoding. */
tcg_out_modrm(s, OPC_GRP5 + rexw,
(is_inc ? EXT5_INC_Ev : EXT5_DEC_Ev), r0);
} else {
tcg_out8(s, (is_inc ? OPC_INC_r32 : OPC_DEC_r32) + r0);
}
return;
}
if (c == ARITH_AND) {
if (TCG_TARGET_REG_BITS == 64) {
if (val == 0xffffffffu) {
tcg_out_ext32u(s, r0, r0);
return;
}
if (val == (uint32_t)val) {
/* AND with no high bits set can use a 32-bit operation. */
rexw = 0;
}
}
if (val == 0xffu && (r0 < 4 || TCG_TARGET_REG_BITS == 64)) {
tcg_out_ext8u(s, r0, r0);
return;
}
if (val == 0xffffu) {
tcg_out_ext16u(s, r0, r0);
return;
}
}
if (val == (int8_t)val) {
tcg_out_modrm(s, OPC_ARITH_EvIb + rexw, c, r0);
tcg_out8(s, val);
return;
}
if (rexw == 0 || val == (int32_t)val) {
tcg_out_modrm(s, OPC_ARITH_EvIz + rexw, c, r0);
tcg_out32(s, val);
return;
}
g_assert_not_reached();
}
static void tcg_out_addi(TCGContext *s, int reg, tcg_target_long val)
{
if (val != 0) {
tgen_arithi(s, ARITH_ADD + P_REXW, reg, val, 0);
}
}
/* Set SMALL to force a short forward branch. */
static void tcg_out_jxx(TCGContext *s, int opc, TCGLabel *l, bool small)
{
int32_t val, val1;
if (l->has_value) {
val = tcg_pcrel_diff(s, l->u.value_ptr);
val1 = val - 2;
if ((int8_t)val1 == val1) {
if (opc == -1) {
tcg_out8(s, OPC_JMP_short);
} else {
tcg_out8(s, OPC_JCC_short + opc);
}
tcg_out8(s, val1);
} else {
tcg_debug_assert(!small);
if (opc == -1) {
tcg_out8(s, OPC_JMP_long);
tcg_out32(s, val - 5);
} else {
tcg_out_opc(s, OPC_JCC_long + opc, 0, 0, 0);
tcg_out32(s, val - 6);
}
}
} else if (small) {
if (opc == -1) {
tcg_out8(s, OPC_JMP_short);
} else {
tcg_out8(s, OPC_JCC_short + opc);
}
tcg_out_reloc(s, s->code_ptr, R_386_PC8, l, -1);
s->code_ptr += 1;
} else {
if (opc == -1) {
tcg_out8(s, OPC_JMP_long);
} else {
tcg_out_opc(s, OPC_JCC_long + opc, 0, 0, 0);
}
tcg_out_reloc(s, s->code_ptr, R_386_PC32, l, -4);
s->code_ptr += 4;
}
}
static void tcg_out_cmp(TCGContext *s, TCGArg arg1, TCGArg arg2,
int const_arg2, int rexw)
{
if (const_arg2) {
if (arg2 == 0) {
/* test r, r */
tcg_out_modrm(s, OPC_TESTL + rexw, arg1, arg1);
} else {
tgen_arithi(s, ARITH_CMP + rexw, arg1, arg2, 0);
}
} else {
tgen_arithr(s, ARITH_CMP + rexw, arg1, arg2);
}
}
static void tcg_out_brcond32(TCGContext *s, TCGCond cond,
TCGArg arg1, TCGArg arg2, int const_arg2,
TCGLabel *label, int small)
{
tcg_out_cmp(s, arg1, arg2, const_arg2, 0);
tcg_out_jxx(s, tcg_cond_to_jcc[cond], label, small);
}
#if TCG_TARGET_REG_BITS == 64
static void tcg_out_brcond64(TCGContext *s, TCGCond cond,
TCGArg arg1, TCGArg arg2, int const_arg2,
TCGLabel *label, int small)
{
tcg_out_cmp(s, arg1, arg2, const_arg2, P_REXW);
tcg_out_jxx(s, tcg_cond_to_jcc[cond], label, small);
}
#else
/* XXX: we implement it at the target level to avoid having to
handle cross basic blocks temporaries */
static void tcg_out_brcond2(TCGContext *s, const TCGArg *args,
const int *const_args, int small)
{
TCGLabel *label_next = gen_new_label();
TCGLabel *label_this = arg_label(args[5]);
switch(args[4]) {
case TCG_COND_EQ:
tcg_out_brcond32(s, TCG_COND_NE, args[0], args[2], const_args[2],
label_next, 1);
tcg_out_brcond32(s, TCG_COND_EQ, args[1], args[3], const_args[3],
label_this, small);
break;
case TCG_COND_NE:
tcg_out_brcond32(s, TCG_COND_NE, args[0], args[2], const_args[2],
label_this, small);
tcg_out_brcond32(s, TCG_COND_NE, args[1], args[3], const_args[3],
label_this, small);
break;
case TCG_COND_LT:
tcg_out_brcond32(s, TCG_COND_LT, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_LTU, args[0], args[2], const_args[2],
label_this, small);
break;
case TCG_COND_LE:
tcg_out_brcond32(s, TCG_COND_LT, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_LEU, args[0], args[2], const_args[2],
label_this, small);
break;
case TCG_COND_GT:
tcg_out_brcond32(s, TCG_COND_GT, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_GTU, args[0], args[2], const_args[2],
label_this, small);
break;
case TCG_COND_GE:
tcg_out_brcond32(s, TCG_COND_GT, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_GEU, args[0], args[2], const_args[2],
label_this, small);
break;
case TCG_COND_LTU:
tcg_out_brcond32(s, TCG_COND_LTU, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_LTU, args[0], args[2], const_args[2],
label_this, small);
break;
case TCG_COND_LEU:
tcg_out_brcond32(s, TCG_COND_LTU, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_LEU, args[0], args[2], const_args[2],
label_this, small);
break;
case TCG_COND_GTU:
tcg_out_brcond32(s, TCG_COND_GTU, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_GTU, args[0], args[2], const_args[2],
label_this, small);
break;
case TCG_COND_GEU:
tcg_out_brcond32(s, TCG_COND_GTU, args[1], args[3], const_args[3],
label_this, small);
tcg_out_jxx(s, JCC_JNE, label_next, 1);
tcg_out_brcond32(s, TCG_COND_GEU, args[0], args[2], const_args[2],
label_this, small);
break;
default:
g_assert_not_reached();
}
tcg_out_label(s, label_next);
}
#endif
static void tcg_out_setcond32(TCGContext *s, TCGCond cond, TCGArg dest,
TCGArg arg1, TCGArg arg2, int const_arg2)
{
tcg_out_cmp(s, arg1, arg2, const_arg2, 0);
tcg_out_modrm(s, OPC_SETCC | tcg_cond_to_jcc[cond], 0, dest);
tcg_out_ext8u(s, dest, dest);
}
#if TCG_TARGET_REG_BITS == 64
static void tcg_out_setcond64(TCGContext *s, TCGCond cond, TCGArg dest,
TCGArg arg1, TCGArg arg2, int const_arg2)
{
tcg_out_cmp(s, arg1, arg2, const_arg2, P_REXW);
tcg_out_modrm(s, OPC_SETCC | tcg_cond_to_jcc[cond], 0, dest);
tcg_out_ext8u(s, dest, dest);
}
#else
static void tcg_out_setcond2(TCGContext *s, const TCGArg *args,
const int *const_args)
{
TCGArg new_args[6];
TCGLabel *label_true, *label_over;
memcpy(new_args, args+1, 5*sizeof(TCGArg));
if (args[0] == args[1] || args[0] == args[2]
|| (!const_args[3] && args[0] == args[3])
|| (!const_args[4] && args[0] == args[4])) {
/* When the destination overlaps with one of the argument
registers, don't do anything tricky. */
label_true = gen_new_label();
label_over = gen_new_label();
new_args[5] = label_arg(label_true);
tcg_out_brcond2(s, new_args, const_args+1, 1);
tcg_out_movi(s, TCG_TYPE_I32, args[0], 0);
tcg_out_jxx(s, JCC_JMP, label_over, 1);
tcg_out_label(s, label_true);
tcg_out_movi(s, TCG_TYPE_I32, args[0], 1);
tcg_out_label(s, label_over);
} else {
/* When the destination does not overlap one of the arguments,
clear the destination first, jump if cond false, and emit an
increment in the true case. This results in smaller code. */
tcg_out_movi(s, TCG_TYPE_I32, args[0], 0);
label_over = gen_new_label();
new_args[4] = tcg_invert_cond(new_args[4]);
new_args[5] = label_arg(label_over);
tcg_out_brcond2(s, new_args, const_args+1, 1);
tgen_arithi(s, ARITH_ADD, args[0], 1, 0);
tcg_out_label(s, label_over);
}
}
#endif
static void tcg_out_cmov(TCGContext *s, TCGCond cond, int rexw,
TCGReg dest, TCGReg v1)
{
if (have_cmov) {
tcg_out_modrm(s, OPC_CMOVCC | tcg_cond_to_jcc[cond] | rexw, dest, v1);
} else {
TCGLabel *over = gen_new_label();
tcg_out_jxx(s, tcg_cond_to_jcc[tcg_invert_cond(cond)], over, 1);
tcg_out_mov(s, TCG_TYPE_I32, dest, v1);
tcg_out_label(s, over);
}
}
static void tcg_out_movcond32(TCGContext *s, TCGCond cond, TCGReg dest,
TCGReg c1, TCGArg c2, int const_c2,
TCGReg v1)
{
tcg_out_cmp(s, c1, c2, const_c2, 0);
tcg_out_cmov(s, cond, 0, dest, v1);
}
#if TCG_TARGET_REG_BITS == 64
static void tcg_out_movcond64(TCGContext *s, TCGCond cond, TCGReg dest,
TCGReg c1, TCGArg c2, int const_c2,
TCGReg v1)
{
tcg_out_cmp(s, c1, c2, const_c2, P_REXW);
tcg_out_cmov(s, cond, P_REXW, dest, v1);
}
#endif
static void tcg_out_ctz(TCGContext *s, int rexw, TCGReg dest, TCGReg arg1,
TCGArg arg2, bool const_a2)
{
if (have_bmi1) {
tcg_out_modrm(s, OPC_TZCNT + rexw, dest, arg1);
if (const_a2) {
tcg_debug_assert(arg2 == (rexw ? 64 : 32));
} else {
tcg_debug_assert(dest != arg2);
tcg_out_cmov(s, TCG_COND_LTU, rexw, dest, arg2);
}
} else {
tcg_debug_assert(dest != arg2);
tcg_out_modrm(s, OPC_BSF + rexw, dest, arg1);
tcg_out_cmov(s, TCG_COND_EQ, rexw, dest, arg2);
}
}
static void tcg_out_clz(TCGContext *s, int rexw, TCGReg dest, TCGReg arg1,
TCGArg arg2, bool const_a2)
{
if (have_lzcnt) {
tcg_out_modrm(s, OPC_LZCNT + rexw, dest, arg1);
if (const_a2) {
tcg_debug_assert(arg2 == (rexw ? 64 : 32));
} else {
tcg_debug_assert(dest != arg2);
tcg_out_cmov(s, TCG_COND_LTU, rexw, dest, arg2);
}
} else {
tcg_debug_assert(!const_a2);
tcg_debug_assert(dest != arg1);
tcg_debug_assert(dest != arg2);
/* Recall that the output of BSR is the index not the count. */
tcg_out_modrm(s, OPC_BSR + rexw, dest, arg1);
tgen_arithi(s, ARITH_XOR + rexw, dest, rexw ? 63 : 31, 0);
/* Since we have destroyed the flags from BSR, we have to re-test. */
tcg_out_cmp(s, arg1, 0, 1, rexw);
tcg_out_cmov(s, TCG_COND_EQ, rexw, dest, arg2);
}
}
static void tcg_out_branch(TCGContext *s, int call, const tcg_insn_unit *dest)
{
intptr_t disp = tcg_pcrel_diff(s, dest) - 5;
if (disp == (int32_t)disp) {
tcg_out_opc(s, call ? OPC_CALL_Jz : OPC_JMP_long, 0, 0, 0);
tcg_out32(s, disp);
} else {
/* rip-relative addressing into the constant pool.
This is 6 + 8 = 14 bytes, as compared to using an
immediate load 10 + 6 = 16 bytes, plus we may
be able to re-use the pool constant for more calls. */
tcg_out_opc(s, OPC_GRP5, 0, 0, 0);
tcg_out8(s, (call ? EXT5_CALLN_Ev : EXT5_JMPN_Ev) << 3 | 5);
new_pool_label(s, (uintptr_t)dest, R_386_PC32, s->code_ptr, -4);
tcg_out32(s, 0);
}
}
static void tcg_out_call(TCGContext *s, const tcg_insn_unit *dest,
const TCGHelperInfo *info)
{
tcg_out_branch(s, 1, dest);
#ifndef _WIN32
if (TCG_TARGET_REG_BITS == 32 && info->out_kind == TCG_CALL_RET_BY_REF) {
/*
* The sysv i386 abi for struct return places a reference as the
* first argument of the stack, and pops that argument with the
* return statement. Since we want to retain the aligned stack
* pointer for the callee, we do not want to actually push that
* argument before the call but rely on the normal store to the
* stack slot. But we do need to compensate for the pop in order
* to reset our correct stack pointer value.
* Pushing a garbage value back onto the stack is quickest.
*/
tcg_out_push(s, TCG_REG_EAX);
}
#endif
}
static void tcg_out_jmp(TCGContext *s, const tcg_insn_unit *dest)
{
tcg_out_branch(s, 0, dest);
}
static void tcg_out_nopn(TCGContext *s, int n)
{
int i;
/* Emit 1 or 2 operand size prefixes for the standard one byte nop,
* "xchg %eax,%eax", forming "xchg %ax,%ax". All cores accept the
* duplicate prefix, and all of the interesting recent cores can
* decode and discard the duplicates in a single cycle.
*/
tcg_debug_assert(n >= 1);
for (i = 1; i < n; ++i) {
tcg_out8(s, 0x66);
}
tcg_out8(s, 0x90);
}
#if defined(CONFIG_SOFTMMU)
/* helper signature: helper_ret_ld_mmu(CPUState *env, target_ulong addr,
* int mmu_idx, uintptr_t ra)
*/
static void * const qemu_ld_helpers[(MO_SIZE | MO_BSWAP) + 1] = {
[MO_UB] = helper_ret_ldub_mmu,
[MO_LEUW] = helper_le_lduw_mmu,
[MO_LEUL] = helper_le_ldul_mmu,
[MO_LEUQ] = helper_le_ldq_mmu,
[MO_BEUW] = helper_be_lduw_mmu,
[MO_BEUL] = helper_be_ldul_mmu,
[MO_BEUQ] = helper_be_ldq_mmu,
};
/* helper signature: helper_ret_st_mmu(CPUState *env, target_ulong addr,
* uintxx_t val, int mmu_idx, uintptr_t ra)
*/
static void * const qemu_st_helpers[(MO_SIZE | MO_BSWAP) + 1] = {
[MO_UB] = helper_ret_stb_mmu,
[MO_LEUW] = helper_le_stw_mmu,
[MO_LEUL] = helper_le_stl_mmu,
[MO_LEUQ] = helper_le_stq_mmu,
[MO_BEUW] = helper_be_stw_mmu,
[MO_BEUL] = helper_be_stl_mmu,
[MO_BEUQ] = helper_be_stq_mmu,
};
/* Perform the TLB load and compare.
Inputs:
ADDRLO and ADDRHI contain the low and high part of the address.
MEM_INDEX and S_BITS are the memory context and log2 size of the load.
WHICH is the offset into the CPUTLBEntry structure of the slot to read.
This should be offsetof addr_read or addr_write.
Outputs:
LABEL_PTRS is filled with 1 (32-bit addresses) or 2 (64-bit addresses)
positions of the displacements of forward jumps to the TLB miss case.
Second argument register is loaded with the low part of the address.
In the TLB hit case, it has been adjusted as indicated by the TLB
and so is a host address. In the TLB miss case, it continues to
hold a guest address.
First argument register is clobbered. */
static inline void tcg_out_tlb_load(TCGContext *s, TCGReg addrlo, TCGReg addrhi,
int mem_index, MemOp opc,
tcg_insn_unit **label_ptr, int which)
{
const TCGReg r0 = TCG_REG_L0;
const TCGReg r1 = TCG_REG_L1;
TCGType ttype = TCG_TYPE_I32;
TCGType tlbtype = TCG_TYPE_I32;
int trexw = 0, hrexw = 0, tlbrexw = 0;
unsigned a_bits = get_alignment_bits(opc);
unsigned s_bits = opc & MO_SIZE;
unsigned a_mask = (1 << a_bits) - 1;
unsigned s_mask = (1 << s_bits) - 1;
target_ulong tlb_mask;
if (TCG_TARGET_REG_BITS == 64) {
if (TARGET_LONG_BITS == 64) {
ttype = TCG_TYPE_I64;
trexw = P_REXW;
}
if (TCG_TYPE_PTR == TCG_TYPE_I64) {
hrexw = P_REXW;
if (TARGET_PAGE_BITS + CPU_TLB_DYN_MAX_BITS > 32) {
tlbtype = TCG_TYPE_I64;
tlbrexw = P_REXW;
}
}
}
tcg_out_mov(s, tlbtype, r0, addrlo);
tcg_out_shifti(s, SHIFT_SHR + tlbrexw, r0,
TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS);
tcg_out_modrm_offset(s, OPC_AND_GvEv + trexw, r0, TCG_AREG0,
TLB_MASK_TABLE_OFS(mem_index) +
offsetof(CPUTLBDescFast, mask));
tcg_out_modrm_offset(s, OPC_ADD_GvEv + hrexw, r0, TCG_AREG0,
TLB_MASK_TABLE_OFS(mem_index) +
offsetof(CPUTLBDescFast, table));
/* If the required alignment is at least as large as the access, simply
copy the address and mask. For lesser alignments, check that we don't
cross pages for the complete access. */
if (a_bits >= s_bits) {
tcg_out_mov(s, ttype, r1, addrlo);
} else {
tcg_out_modrm_offset(s, OPC_LEA + trexw, r1, addrlo, s_mask - a_mask);
}
tlb_mask = (target_ulong)TARGET_PAGE_MASK | a_mask;
tgen_arithi(s, ARITH_AND + trexw, r1, tlb_mask, 0);
/* cmp 0(r0), r1 */
tcg_out_modrm_offset(s, OPC_CMP_GvEv + trexw, r1, r0, which);
/* Prepare for both the fast path add of the tlb addend, and the slow
path function argument setup. */
tcg_out_mov(s, ttype, r1, addrlo);
/* jne slow_path */
tcg_out_opc(s, OPC_JCC_long + JCC_JNE, 0, 0, 0);
label_ptr[0] = s->code_ptr;
s->code_ptr += 4;
if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) {
/* cmp 4(r0), addrhi */
tcg_out_modrm_offset(s, OPC_CMP_GvEv, addrhi, r0, which + 4);
/* jne slow_path */
tcg_out_opc(s, OPC_JCC_long + JCC_JNE, 0, 0, 0);
label_ptr[1] = s->code_ptr;
s->code_ptr += 4;
}
/* TLB Hit. */
/* add addend(r0), r1 */
tcg_out_modrm_offset(s, OPC_ADD_GvEv + hrexw, r1, r0,
offsetof(CPUTLBEntry, addend));
}
/*
* Record the context of a call to the out of line helper code for the slow path
* for a load or store, so that we can later generate the correct helper code
*/
static void add_qemu_ldst_label(TCGContext *s, bool is_ld, bool is_64,
MemOpIdx oi,
TCGReg datalo, TCGReg datahi,
TCGReg addrlo, TCGReg addrhi,
tcg_insn_unit *raddr,
tcg_insn_unit **label_ptr)
{
TCGLabelQemuLdst *label = new_ldst_label(s);
label->is_ld = is_ld;
label->oi = oi;
label->type = is_64 ? TCG_TYPE_I64 : TCG_TYPE_I32;
label->datalo_reg = datalo;
label->datahi_reg = datahi;
label->addrlo_reg = addrlo;
label->addrhi_reg = addrhi;
label->raddr = tcg_splitwx_to_rx(raddr);
label->label_ptr[0] = label_ptr[0];
if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) {
label->label_ptr[1] = label_ptr[1];
}
}
/*
* Generate code for the slow path for a load at the end of block
*/
static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *l)
{
MemOpIdx oi = l->oi;
MemOp opc = get_memop(oi);
TCGReg data_reg;
tcg_insn_unit **label_ptr = &l->label_ptr[0];
int rexw = (l->type == TCG_TYPE_I64 ? P_REXW : 0);
/* resolve label address */
tcg_patch32(label_ptr[0], s->code_ptr - label_ptr[0] - 4);
if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) {
tcg_patch32(label_ptr[1], s->code_ptr - label_ptr[1] - 4);
}
if (TCG_TARGET_REG_BITS == 32) {
int ofs = 0;
tcg_out_st(s, TCG_TYPE_PTR, TCG_AREG0, TCG_REG_ESP, ofs);
ofs += 4;
tcg_out_st(s, TCG_TYPE_I32, l->addrlo_reg, TCG_REG_ESP, ofs);
ofs += 4;
if (TARGET_LONG_BITS == 64) {
tcg_out_st(s, TCG_TYPE_I32, l->addrhi_reg, TCG_REG_ESP, ofs);
ofs += 4;
}
tcg_out_sti(s, TCG_TYPE_I32, oi, TCG_REG_ESP, ofs);
ofs += 4;
tcg_out_sti(s, TCG_TYPE_PTR, (uintptr_t)l->raddr, TCG_REG_ESP, ofs);
} else {
tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[0], TCG_AREG0);
/* The second argument is already loaded with addrlo. */
tcg_out_movi(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[2], oi);
tcg_out_movi(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[3],
(uintptr_t)l->raddr);
}
tcg_out_branch(s, 1, qemu_ld_helpers[opc & (MO_BSWAP | MO_SIZE)]);
data_reg = l->datalo_reg;
switch (opc & MO_SSIZE) {
case MO_SB:
tcg_out_ext8s(s, l->type, data_reg, TCG_REG_EAX);
break;
case MO_SW:
tcg_out_ext16s(s, data_reg, TCG_REG_EAX, rexw);
break;
#if TCG_TARGET_REG_BITS == 64
case MO_SL:
tcg_out_ext32s(s, data_reg, TCG_REG_EAX);
break;
#endif
case MO_UB:
case MO_UW:
/* Note that the helpers have zero-extended to tcg_target_long. */
case MO_UL:
tcg_out_mov(s, TCG_TYPE_I32, data_reg, TCG_REG_EAX);
break;
case MO_UQ:
if (TCG_TARGET_REG_BITS == 64) {
tcg_out_mov(s, TCG_TYPE_I64, data_reg, TCG_REG_RAX);
} else if (data_reg == TCG_REG_EDX) {
/* xchg %edx, %eax */
tcg_out_opc(s, OPC_XCHG_ax_r32 + TCG_REG_EDX, 0, 0, 0);
tcg_out_mov(s, TCG_TYPE_I32, l->datahi_reg, TCG_REG_EAX);
} else {
tcg_out_mov(s, TCG_TYPE_I32, data_reg, TCG_REG_EAX);
tcg_out_mov(s, TCG_TYPE_I32, l->datahi_reg, TCG_REG_EDX);
}
break;
default:
g_assert_not_reached();
}
/* Jump to the code corresponding to next IR of qemu_st */
tcg_out_jmp(s, l->raddr);
return true;
}
/*
* Generate code for the slow path for a store at the end of block
*/
static bool tcg_out_qemu_st_slow_path(TCGContext *s, TCGLabelQemuLdst *l)
{
MemOpIdx oi = l->oi;
MemOp opc = get_memop(oi);
MemOp s_bits = opc & MO_SIZE;
tcg_insn_unit **label_ptr = &l->label_ptr[0];
TCGReg retaddr;
/* resolve label address */
tcg_patch32(label_ptr[0], s->code_ptr - label_ptr[0] - 4);
if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) {
tcg_patch32(label_ptr[1], s->code_ptr - label_ptr[1] - 4);
}
if (TCG_TARGET_REG_BITS == 32) {
int ofs = 0;
tcg_out_st(s, TCG_TYPE_PTR, TCG_AREG0, TCG_REG_ESP, ofs);
ofs += 4;
tcg_out_st(s, TCG_TYPE_I32, l->addrlo_reg, TCG_REG_ESP, ofs);
ofs += 4;
if (TARGET_LONG_BITS == 64) {
tcg_out_st(s, TCG_TYPE_I32, l->addrhi_reg, TCG_REG_ESP, ofs);
ofs += 4;
}
tcg_out_st(s, TCG_TYPE_I32, l->datalo_reg, TCG_REG_ESP, ofs);
ofs += 4;
if (s_bits == MO_64) {
tcg_out_st(s, TCG_TYPE_I32, l->datahi_reg, TCG_REG_ESP, ofs);
ofs += 4;
}
tcg_out_sti(s, TCG_TYPE_I32, oi, TCG_REG_ESP, ofs);
ofs += 4;
retaddr = TCG_REG_EAX;
tcg_out_movi(s, TCG_TYPE_PTR, retaddr, (uintptr_t)l->raddr);
tcg_out_st(s, TCG_TYPE_PTR, retaddr, TCG_REG_ESP, ofs);
} else {
tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[0], TCG_AREG0);
/* The second argument is already loaded with addrlo. */
tcg_out_mov(s, (s_bits == MO_64 ? TCG_TYPE_I64 : TCG_TYPE_I32),
tcg_target_call_iarg_regs[2], l->datalo_reg);
tcg_out_movi(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[3], oi);
if (ARRAY_SIZE(tcg_target_call_iarg_regs) > 4) {
retaddr = tcg_target_call_iarg_regs[4];
tcg_out_movi(s, TCG_TYPE_PTR, retaddr, (uintptr_t)l->raddr);
} else {
retaddr = TCG_REG_RAX;
tcg_out_movi(s, TCG_TYPE_PTR, retaddr, (uintptr_t)l->raddr);
tcg_out_st(s, TCG_TYPE_PTR, retaddr, TCG_REG_ESP,
TCG_TARGET_CALL_STACK_OFFSET);
}
}
/* "Tail call" to the helper, with the return address back inline. */
tcg_out_push(s, retaddr);
tcg_out_jmp(s, qemu_st_helpers[opc & (MO_BSWAP | MO_SIZE)]);
return true;
}
#else
static void tcg_out_test_alignment(TCGContext *s, bool is_ld, TCGReg addrlo,
TCGReg addrhi, unsigned a_bits)
{
unsigned a_mask = (1 << a_bits) - 1;
TCGLabelQemuLdst *label;
/*
* We are expecting a_bits to max out at 7, so we can usually use testb.
* For i686, we have to use testl for %esi/%edi.
*/
if (a_mask <= 0xff && (TCG_TARGET_REG_BITS == 64 || addrlo < 4)) {
tcg_out_modrm(s, OPC_GRP3_Eb | P_REXB_RM, EXT3_TESTi, addrlo);
tcg_out8(s, a_mask);
} else {
tcg_out_modrm(s, OPC_GRP3_Ev, EXT3_TESTi, addrlo);
tcg_out32(s, a_mask);
}
/* jne slow_path */
tcg_out_opc(s, OPC_JCC_long + JCC_JNE, 0, 0, 0);
label = new_ldst_label(s);
label->is_ld = is_ld;
label->addrlo_reg = addrlo;
label->addrhi_reg = addrhi;
label->raddr = tcg_splitwx_to_rx(s->code_ptr + 4);
label->label_ptr[0] = s->code_ptr;
s->code_ptr += 4;
}
static bool tcg_out_fail_alignment(TCGContext *s, TCGLabelQemuLdst *l)
{
/* resolve label address */
tcg_patch32(l->label_ptr[0], s->code_ptr - l->label_ptr[0] - 4);
if (TCG_TARGET_REG_BITS == 32) {
int ofs = 0;
tcg_out_st(s, TCG_TYPE_PTR, TCG_AREG0, TCG_REG_ESP, ofs);
ofs += 4;
tcg_out_st(s, TCG_TYPE_I32, l->addrlo_reg, TCG_REG_ESP, ofs);
ofs += 4;
if (TARGET_LONG_BITS == 64) {
tcg_out_st(s, TCG_TYPE_I32, l->addrhi_reg, TCG_REG_ESP, ofs);
ofs += 4;
}
tcg_out_pushi(s, (uintptr_t)l->raddr);
} else {
tcg_out_mov(s, TCG_TYPE_TL, tcg_target_call_iarg_regs[1],
l->addrlo_reg);
tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[0], TCG_AREG0);
tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_RAX, (uintptr_t)l->raddr);
tcg_out_push(s, TCG_REG_RAX);
}
/* "Tail call" to the helper, with the return address back inline. */
tcg_out_jmp(s, (const void *)(l->is_ld ? helper_unaligned_ld
: helper_unaligned_st));
return true;
}
static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *l)
{
return tcg_out_fail_alignment(s, l);
}
static bool tcg_out_qemu_st_slow_path(TCGContext *s, TCGLabelQemuLdst *l)
{
return tcg_out_fail_alignment(s, l);
}
#if TCG_TARGET_REG_BITS == 32
# define x86_guest_base_seg 0
# define x86_guest_base_index -1
# define x86_guest_base_offset guest_base
#else
static int x86_guest_base_seg;
static int x86_guest_base_index = -1;
static int32_t x86_guest_base_offset;
# if defined(__x86_64__) && defined(__linux__)
# include <asm/prctl.h>
# include <sys/prctl.h>
int arch_prctl(int code, unsigned long addr);
static inline int setup_guest_base_seg(void)
{
if (arch_prctl(ARCH_SET_GS, guest_base) == 0) {
return P_GS;
}
return 0;
}
# elif defined (__FreeBSD__) || defined (__FreeBSD_kernel__)
# include <machine/sysarch.h>
static inline int setup_guest_base_seg(void)
{
if (sysarch(AMD64_SET_GSBASE, &guest_base) == 0) {
return P_GS;
}
return 0;
}
# else
static inline int setup_guest_base_seg(void)
{
return 0;
}
# endif
#endif
#endif /* SOFTMMU */
static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg datalo, TCGReg datahi,
TCGReg base, int index, intptr_t ofs,
int seg, bool is64, MemOp memop)
{
bool use_movbe = false;
int rexw = is64 * P_REXW;
int movop = OPC_MOVL_GvEv;
/* Do big-endian loads with movbe. */
if (memop & MO_BSWAP) {
tcg_debug_assert(have_movbe);
use_movbe = true;
movop = OPC_MOVBE_GyMy;
}
switch (memop & MO_SSIZE) {
case MO_UB:
tcg_out_modrm_sib_offset(s, OPC_MOVZBL + seg, datalo,
base, index, 0, ofs);
break;
case MO_SB:
tcg_out_modrm_sib_offset(s, OPC_MOVSBL + rexw + seg, datalo,
base, index, 0, ofs);
break;
case MO_UW:
if (use_movbe) {
/* There is no extending movbe; only low 16-bits are modified. */
if (datalo != base && datalo != index) {
/* XOR breaks dependency chains. */
tgen_arithr(s, ARITH_XOR, datalo, datalo);
tcg_out_modrm_sib_offset(s, OPC_MOVBE_GyMy + P_DATA16 + seg,
datalo, base, index, 0, ofs);
} else {
tcg_out_modrm_sib_offset(s, OPC_MOVBE_GyMy + P_DATA16 + seg,
datalo, base, index, 0, ofs);
tcg_out_ext16u(s, datalo, datalo);
}
} else {
tcg_out_modrm_sib_offset(s, OPC_MOVZWL + seg, datalo,
base, index, 0, ofs);
}
break;
case MO_SW:
if (use_movbe) {
tcg_out_modrm_sib_offset(s, OPC_MOVBE_GyMy + P_DATA16 + seg,
datalo, base, index, 0, ofs);
tcg_out_ext16s(s, datalo, datalo, rexw);
} else {
tcg_out_modrm_sib_offset(s, OPC_MOVSWL + rexw + seg,
datalo, base, index, 0, ofs);
}
break;
case MO_UL:
tcg_out_modrm_sib_offset(s, movop + seg, datalo, base, index, 0, ofs);
break;
#if TCG_TARGET_REG_BITS == 64
case MO_SL:
if (use_movbe) {
tcg_out_modrm_sib_offset(s, OPC_MOVBE_GyMy + seg, datalo,
base, index, 0, ofs);
tcg_out_ext32s(s, datalo, datalo);
} else {
tcg_out_modrm_sib_offset(s, OPC_MOVSLQ + seg, datalo,
base, index, 0, ofs);
}
break;
#endif
case MO_UQ:
if (TCG_TARGET_REG_BITS == 64) {
tcg_out_modrm_sib_offset(s, movop + P_REXW + seg, datalo,
base, index, 0, ofs);
} else {
if (use_movbe) {
TCGReg t = datalo;
datalo = datahi;
datahi = t;
}
if (base != datalo) {
tcg_out_modrm_sib_offset(s, movop + seg, datalo,
base, index, 0, ofs);
tcg_out_modrm_sib_offset(s, movop + seg, datahi,
base, index, 0, ofs + 4);
} else {
tcg_out_modrm_sib_offset(s, movop + seg, datahi,
base, index, 0, ofs + 4);
tcg_out_modrm_sib_offset(s, movop + seg, datalo,
base, index, 0, ofs);
}
}
break;
default:
g_assert_not_reached();
}
}
/* XXX: qemu_ld and qemu_st could be modified to clobber only EDX and
EAX. It will be useful once fixed registers globals are less
common. */
static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, bool is64)
{
TCGReg datalo, datahi, addrlo;
TCGReg addrhi __attribute__((unused));
MemOpIdx oi;
MemOp opc;
#if defined(CONFIG_SOFTMMU)
int mem_index;
tcg_insn_unit *label_ptr[2];
#else
unsigned a_bits;
#endif
datalo = *args++;
datahi = (TCG_TARGET_REG_BITS == 32 && is64 ? *args++ : 0);
addrlo = *args++;
addrhi = (TARGET_LONG_BITS > TCG_TARGET_REG_BITS ? *args++ : 0);
oi = *args++;
opc = get_memop(oi);
#if defined(CONFIG_SOFTMMU)
mem_index = get_mmuidx(oi);
tcg_out_tlb_load(s, addrlo, addrhi, mem_index, opc,
label_ptr, offsetof(CPUTLBEntry, addr_read));
/* TLB Hit. */
tcg_out_qemu_ld_direct(s, datalo, datahi, TCG_REG_L1, -1, 0, 0, is64, opc);
/* Record the current context of a load into ldst label */
add_qemu_ldst_label(s, true, is64, oi, datalo, datahi, addrlo, addrhi,
s->code_ptr, label_ptr);
#else
a_bits = get_alignment_bits(opc);
if (a_bits) {
tcg_out_test_alignment(s, true, addrlo, addrhi, a_bits);
}
tcg_out_qemu_ld_direct(s, datalo, datahi, addrlo, x86_guest_base_index,
x86_guest_base_offset, x86_guest_base_seg,
is64, opc);
#endif
}
static void tcg_out_qemu_st_direct(TCGContext *s, TCGReg datalo, TCGReg datahi,
TCGReg base, int index, intptr_t ofs,
int seg, MemOp memop)
{
bool use_movbe = false;
int movop = OPC_MOVL_EvGv;
/*
* Do big-endian stores with movbe or softmmu.
* User-only without movbe will have its swapping done generically.
*/
if (memop & MO_BSWAP) {
tcg_debug_assert(have_movbe);
use_movbe = true;
movop = OPC_MOVBE_MyGy;
}
switch (memop & MO_SIZE) {
case MO_8:
/* This is handled with constraints on INDEX_op_qemu_st8_i32. */
tcg_debug_assert(TCG_TARGET_REG_BITS == 64 || datalo < 4);
tcg_out_modrm_sib_offset(s, OPC_MOVB_EvGv + P_REXB_R + seg,
datalo, base, index, 0, ofs);
break;
case MO_16:
tcg_out_modrm_sib_offset(s, movop + P_DATA16 + seg, datalo,
base, index, 0, ofs);
break;
case MO_32:
tcg_out_modrm_sib_offset(s, movop + seg, datalo, base, index, 0, ofs);
break;
case MO_64:
if (TCG_TARGET_REG_BITS == 64) {
tcg_out_modrm_sib_offset(s, movop + P_REXW + seg, datalo,
base, index, 0, ofs);
} else {
if (use_movbe) {
TCGReg t = datalo;
datalo = datahi;
datahi = t;
}
tcg_out_modrm_sib_offset(s, movop + seg, datalo,
base, index, 0, ofs);
tcg_out_modrm_sib_offset(s, movop + seg, datahi,
base, index, 0, ofs + 4);
}
break;
default:
g_assert_not_reached();
}
}
static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, bool is64)
{
TCGReg datalo, datahi, addrlo;
TCGReg addrhi __attribute__((unused));
MemOpIdx oi;
MemOp opc;
#if defined(CONFIG_SOFTMMU)
int mem_index;
tcg_insn_unit *label_ptr[2];
#else
unsigned a_bits;
#endif
datalo = *args++;
datahi = (TCG_TARGET_REG_BITS == 32 && is64 ? *args++ : 0);
addrlo = *args++;
addrhi = (TARGET_LONG_BITS > TCG_TARGET_REG_BITS ? *args++ : 0);
oi = *args++;
opc = get_memop(oi);
#if defined(CONFIG_SOFTMMU)
mem_index = get_mmuidx(oi);
tcg_out_tlb_load(s, addrlo, addrhi, mem_index, opc,
label_ptr, offsetof(CPUTLBEntry, addr_write));
/* TLB Hit. */
tcg_out_qemu_st_direct(s, datalo, datahi, TCG_REG_L1, -1, 0, 0, opc);
/* Record the current context of a store into ldst label */
add_qemu_ldst_label(s, false, is64, oi, datalo, datahi, addrlo, addrhi,
s->code_ptr, label_ptr);
#else
a_bits = get_alignment_bits(opc);
if (a_bits) {
tcg_out_test_alignment(s, false, addrlo, addrhi, a_bits);
}
tcg_out_qemu_st_direct(s, datalo, datahi, addrlo, x86_guest_base_index,
x86_guest_base_offset, x86_guest_base_seg, opc);
#endif
}
static void tcg_out_exit_tb(TCGContext *s, uintptr_t a0)
{
/* Reuse the zeroing that exists for goto_ptr. */
if (a0 == 0) {
tcg_out_jmp(s, tcg_code_gen_epilogue);
} else {
tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_EAX, a0);
tcg_out_jmp(s, tb_ret_addr);
}
}
static void tcg_out_goto_tb(TCGContext *s, int which)
{
/*
* Jump displacement must be aligned for atomic patching;
* see if we need to add extra nops before jump
*/
int gap = QEMU_ALIGN_PTR_UP(s->code_ptr + 1, 4) - s->code_ptr;
if (gap != 1) {
tcg_out_nopn(s, gap - 1);
}
tcg_out8(s, OPC_JMP_long); /* jmp im */
set_jmp_insn_offset(s, which);
tcg_out32(s, 0);
set_jmp_reset_offset(s, which);
}
void tb_target_set_jmp_target(const TranslationBlock *tb, int n,
uintptr_t jmp_rx, uintptr_t jmp_rw)
{
/* patch the branch destination */
uintptr_t addr = tb->jmp_target_addr[n];
qatomic_set((int32_t *)jmp_rw, addr - (jmp_rx + 4));
/* no need to flush icache explicitly */
}
static inline void tcg_out_op(TCGContext *s, TCGOpcode opc,
const TCGArg args[TCG_MAX_OP_ARGS],
const int const_args[TCG_MAX_OP_ARGS])
{
TCGArg a0, a1, a2;
int c, const_a2, vexop, rexw = 0;
#if TCG_TARGET_REG_BITS == 64
# define OP_32_64(x) \
case glue(glue(INDEX_op_, x), _i64): \
rexw = P_REXW; /* FALLTHRU */ \
case glue(glue(INDEX_op_, x), _i32)
#else
# define OP_32_64(x) \
case glue(glue(INDEX_op_, x), _i32)
#endif
/* Hoist the loads of the most common arguments. */
a0 = args[0];
a1 = args[1];
a2 = args[2];
const_a2 = const_args[2];
switch (opc) {
case INDEX_op_goto_ptr:
/* jmp to the given host address (could be epilogue) */
tcg_out_modrm(s, OPC_GRP5, EXT5_JMPN_Ev, a0);
break;
case INDEX_op_br:
tcg_out_jxx(s, JCC_JMP, arg_label(a0), 0);
break;
OP_32_64(ld8u):
/* Note that we can ignore REXW for the zero-extend to 64-bit. */
tcg_out_modrm_offset(s, OPC_MOVZBL, a0, a1, a2);
break;
OP_32_64(ld8s):
tcg_out_modrm_offset(s, OPC_MOVSBL + rexw, a0, a1, a2);
break;
OP_32_64(ld16u):
/* Note that we can ignore REXW for the zero-extend to 64-bit. */
tcg_out_modrm_offset(s, OPC_MOVZWL, a0, a1, a2);
break;
OP_32_64(ld16s):
tcg_out_modrm_offset(s, OPC_MOVSWL + rexw, a0, a1, a2);
break;
#if TCG_TARGET_REG_BITS == 64
case INDEX_op_ld32u_i64:
#endif
case INDEX_op_ld_i32:
tcg_out_ld(s, TCG_TYPE_I32, a0, a1, a2);
break;
OP_32_64(st8):
if (const_args[0]) {
tcg_out_modrm_offset(s, OPC_MOVB_EvIz, 0, a1, a2);
tcg_out8(s, a0);
} else {
tcg_out_modrm_offset(s, OPC_MOVB_EvGv | P_REXB_R, a0, a1, a2);
}
break;
OP_32_64(st16):
if (const_args[0]) {
tcg_out_modrm_offset(s, OPC_MOVL_EvIz | P_DATA16, 0, a1, a2);
tcg_out16(s, a0);
} else {
tcg_out_modrm_offset(s, OPC_MOVL_EvGv | P_DATA16, a0, a1, a2);
}
break;
#if TCG_TARGET_REG_BITS == 64
case INDEX_op_st32_i64:
#endif
case INDEX_op_st_i32:
if (const_args[0]) {
tcg_out_modrm_offset(s, OPC_MOVL_EvIz, 0, a1, a2);
tcg_out32(s, a0);
} else {
tcg_out_st(s, TCG_TYPE_I32, a0, a1, a2);
}
break;
OP_32_64(add):
/* For 3-operand addition, use LEA. */
if (a0 != a1) {
TCGArg c3 = 0;
if (const_a2) {
c3 = a2, a2 = -1;
} else if (a0 == a2) {
/* Watch out for dest = src + dest, since we've removed
the matching constraint on the add. */
tgen_arithr(s, ARITH_ADD + rexw, a0, a1);
break;
}
tcg_out_modrm_sib_offset(s, OPC_LEA + rexw, a0, a1, a2, 0, c3);
break;
}
c = ARITH_ADD;
goto gen_arith;
OP_32_64(sub):
c = ARITH_SUB;
goto gen_arith;
OP_32_64(and):
c = ARITH_AND;
goto gen_arith;
OP_32_64(or):
c = ARITH_OR;
goto gen_arith;
OP_32_64(xor):
c = ARITH_XOR;
goto gen_arith;
gen_arith:
if (const_a2) {
tgen_arithi(s, c + rexw, a0, a2, 0);
} else {
tgen_arithr(s, c + rexw, a0, a2);
}
break;
OP_32_64(andc):
if (const_a2) {
tcg_out_mov(s, rexw ? TCG_TYPE_I64 : TCG_TYPE_I32, a0, a1);
tgen_arithi(s, ARITH_AND + rexw, a0, ~a2, 0);
} else {
tcg_out_vex_modrm(s, OPC_ANDN + rexw, a0, a2, a1);
}
break;
OP_32_64(mul):
if (const_a2) {
int32_t val;
val = a2;
if (val == (int8_t)val) {
tcg_out_modrm(s, OPC_IMUL_GvEvIb + rexw, a0, a0);
tcg_out8(s, val);
} else {
tcg_out_modrm(s, OPC_IMUL_GvEvIz + rexw, a0, a0);
tcg_out32(s, val);
}
} else {
tcg_out_modrm(s, OPC_IMUL_GvEv + rexw, a0, a2);
}
break;
OP_32_64(div2):
tcg_out_modrm(s, OPC_GRP3_Ev + rexw, EXT3_IDIV, args[4]);
break;
OP_32_64(divu2):
tcg_out_modrm(s, OPC_GRP3_Ev + rexw, EXT3_DIV, args[4]);
break;
OP_32_64(shl):
/* For small constant 3-operand shift, use LEA. */
if (const_a2 && a0 != a1 && (a2 - 1) < 3) {
if (a2 - 1 == 0) {
/* shl $1,a1,a0 -> lea (a1,a1),a0 */
tcg_out_modrm_sib_offset(s, OPC_LEA + rexw, a0, a1, a1, 0, 0);
} else {
/* shl $n,a1,a0 -> lea 0(,a1,n),a0 */
tcg_out_modrm_sib_offset(s, OPC_LEA + rexw, a0, -1, a1, a2, 0);
}
break;
}
c = SHIFT_SHL;
vexop = OPC_SHLX;
goto gen_shift_maybe_vex;
OP_32_64(shr):
c = SHIFT_SHR;
vexop = OPC_SHRX;
goto gen_shift_maybe_vex;
OP_32_64(sar):
c = SHIFT_SAR;
vexop = OPC_SARX;
goto gen_shift_maybe_vex;
OP_32_64(rotl):
c = SHIFT_ROL;
goto gen_shift;
OP_32_64(rotr):
c = SHIFT_ROR;
goto gen_shift;
gen_shift_maybe_vex:
if (have_bmi2) {
if (!const_a2) {
tcg_out_vex_modrm(s, vexop + rexw, a0, a2, a1);
break;
}
tcg_out_mov(s, rexw ? TCG_TYPE_I64 : TCG_TYPE_I32, a0, a1);
}
/* FALLTHRU */
gen_shift:
if (const_a2) {
tcg_out_shifti(s, c + rexw, a0, a2);
} else {
tcg_out_modrm(s, OPC_SHIFT_cl + rexw, c, a0);
}
break;
OP_32_64(ctz):
tcg_out_ctz(s, rexw, args[0], args[1], args[2], const_args[2]);
break;
OP_32_64(clz):
tcg_out_clz(s, rexw, args[0], args[1], args[2], const_args[2]);
break;
OP_32_64(ctpop):
tcg_out_modrm(s, OPC_POPCNT + rexw, a0, a1);
break;
case INDEX_op_brcond_i32:
tcg_out_brcond32(s, a2, a0, a1, const_args[1], arg_label(args[3]), 0);
break;
case INDEX_op_setcond_i32:
tcg_out_setcond32(s, args[3], a0, a1, a2, const_a2);
break;
case INDEX_op_movcond_i32:
tcg_out_movcond32(s, args[5], a0, a1, a2, const_a2, args[3]);
break;
OP_32_64(bswap16):
if (a2 & TCG_BSWAP_OS) {
/* Output must be sign-extended. */
if (rexw) {
tcg_out_bswap64(s, a0);
tcg_out_shifti(s, SHIFT_SAR + rexw, a0, 48);
} else {
tcg_out_bswap32(s, a0);
tcg_out_shifti(s, SHIFT_SAR, a0, 16);
}
} else if ((a2 & (TCG_BSWAP_IZ | TCG_BSWAP_OZ)) == TCG_BSWAP_OZ) {
/* Output must be zero-extended, but input isn't. */
tcg_out_bswap32(s, a0);
tcg_out_shifti(s, SHIFT_SHR, a0, 16);
} else {
tcg_out_rolw_8(s, a0);
}
break;
OP_32_64(bswap32):
tcg_out_bswap32(s, a0);
if (rexw && (a2 & TCG_BSWAP_OS)) {
tcg_out_ext32s(s, a0, a0);
}
break;
OP_32_64(neg):
tcg_out_modrm(s, OPC_GRP3_Ev + rexw, EXT3_NEG, a0);
break;
OP_32_64(not):
tcg_out_modrm(s, OPC_GRP3_Ev + rexw, EXT3_NOT, a0);
break;
OP_32_64(ext16s):
tcg_out_ext16s(s, a0, a1, rexw);
break;
OP_32_64(ext16u):
tcg_out_ext16u(s, a0, a1);
break;
case INDEX_op_qemu_ld_i32:
tcg_out_qemu_ld(s, args, 0);
break;
case INDEX_op_qemu_ld_i64:
tcg_out_qemu_ld(s, args, 1);
break;
case INDEX_op_qemu_st_i32:
case INDEX_op_qemu_st8_i32:
tcg_out_qemu_st(s, args, 0);
break;
case INDEX_op_qemu_st_i64:
tcg_out_qemu_st(s, args, 1);
break;
OP_32_64(mulu2):
tcg_out_modrm(s, OPC_GRP3_Ev + rexw, EXT3_MUL, args[3]);
break;
OP_32_64(muls2):
tcg_out_modrm(s, OPC_GRP3_Ev + rexw, EXT3_IMUL, args[3]);
break;
OP_32_64(add2):
if (const_args[4]) {
tgen_arithi(s, ARITH_ADD + rexw, a0, args[4], 1);
} else {
tgen_arithr(s, ARITH_ADD + rexw, a0, args[4]);
}
if (const_args[5]) {
tgen_arithi(s, ARITH_ADC + rexw, a1, args[5], 1);
} else {
tgen_arithr(s, ARITH_ADC + rexw, a1, args[5]);
}
break;
OP_32_64(sub2):
if (const_args[4]) {
tgen_arithi(s, ARITH_SUB + rexw, a0, args[4], 1);
} else {
tgen_arithr(s, ARITH_SUB + rexw, a0, args[4]);
}
if (const_args[5]) {
tgen_arithi(s, ARITH_SBB + rexw, a1, args[5], 1);
} else {
tgen_arithr(s, ARITH_SBB + rexw, a1, args[5]);
}
break;
#if TCG_TARGET_REG_BITS == 32
case INDEX_op_brcond2_i32:
tcg_out_brcond2(s, args, const_args, 0);
break;
case INDEX_op_setcond2_i32:
tcg_out_setcond2(s, args, const_args);
break;
#else /* TCG_TARGET_REG_BITS == 64 */
case INDEX_op_ld32s_i64:
tcg_out_modrm_offset(s, OPC_MOVSLQ, a0, a1, a2);
break;
case INDEX_op_ld_i64:
tcg_out_ld(s, TCG_TYPE_I64, a0, a1, a2);
break;
case INDEX_op_st_i64:
if (const_args[0]) {
tcg_out_modrm_offset(s, OPC_MOVL_EvIz | P_REXW, 0, a1, a2);
tcg_out32(s, a0);
} else {
tcg_out_st(s, TCG_TYPE_I64, a0, a1, a2);
}
break;
case INDEX_op_brcond_i64:
tcg_out_brcond64(s, a2, a0, a1, const_args[1], arg_label(args[3]), 0);
break;
case INDEX_op_setcond_i64:
tcg_out_setcond64(s, args[3], a0, a1, a2, const_a2);
break;
case INDEX_op_movcond_i64:
tcg_out_movcond64(s, args[5], a0, a1, a2, const_a2, args[3]);
break;
case INDEX_op_bswap64_i64:
tcg_out_bswap64(s, a0);
break;
case INDEX_op_extu_i32_i64:
case INDEX_op_ext32u_i64:
case INDEX_op_extrl_i64_i32:
tcg_out_ext32u(s, a0, a1);
break;
case INDEX_op_ext_i32_i64:
case INDEX_op_ext32s_i64:
tcg_out_ext32s(s, a0, a1);
break;
case INDEX_op_extrh_i64_i32:
tcg_out_shifti(s, SHIFT_SHR + P_REXW, a0, 32);
break;
#endif
OP_32_64(deposit):
if (args[3] == 0 && args[4] == 8) {
/* load bits 0..7 */
tcg_out_modrm(s, OPC_MOVB_EvGv | P_REXB_R | P_REXB_RM, a2, a0);
} else if (args[3] == 8 && args[4] == 8) {
/* load bits 8..15 */
tcg_out_modrm(s, OPC_MOVB_EvGv, a2, a0 + 4);
} else if (args[3] == 0 && args[4] == 16) {
/* load bits 0..15 */
tcg_out_modrm(s, OPC_MOVL_EvGv | P_DATA16, a2, a0);
} else {
g_assert_not_reached();
}
break;
case INDEX_op_extract_i64:
if (a2 + args[3] == 32) {
/* This is a 32-bit zero-extending right shift. */
tcg_out_mov(s, TCG_TYPE_I32, a0, a1);
tcg_out_shifti(s, SHIFT_SHR, a0, a2);
break;
}
/* FALLTHRU */
case INDEX_op_extract_i32:
/* On the off-chance that we can use the high-byte registers.
Otherwise we emit the same ext16 + shift pattern that we
would have gotten from the normal tcg-op.c expansion. */
tcg_debug_assert(a2 == 8 && args[3] == 8);
if (a1 < 4 && a0 < 8) {
tcg_out_modrm(s, OPC_MOVZBL, a0, a1 + 4);
} else {
tcg_out_ext16u(s, a0, a1);
tcg_out_shifti(s, SHIFT_SHR, a0, 8);
}
break;
case INDEX_op_sextract_i32:
/* We don't implement sextract_i64, as we cannot sign-extend to
64-bits without using the REX prefix that explicitly excludes
access to the high-byte registers. */
tcg_debug_assert(a2 == 8 && args[3] == 8);
if (a1 < 4 && a0 < 8) {
tcg_out_modrm(s, OPC_MOVSBL, a0, a1 + 4);
} else {
tcg_out_ext16s(s, a0, a1, 0);
tcg_out_shifti(s, SHIFT_SAR, a0, 8);
}
break;
OP_32_64(extract2):
/* Note that SHRD outputs to the r/m operand. */
tcg_out_modrm(s, OPC_SHRD_Ib + rexw, a2, a0);
tcg_out8(s, args[3]);
break;
case INDEX_op_mb:
tcg_out_mb(s, a0);
break;
case INDEX_op_mov_i32: /* Always emitted via tcg_out_mov. */
case INDEX_op_mov_i64:
case INDEX_op_call: /* Always emitted via tcg_out_call. */
case INDEX_op_exit_tb: /* Always emitted via tcg_out_exit_tb. */
case INDEX_op_goto_tb: /* Always emitted via tcg_out_goto_tb. */
case INDEX_op_ext8s_i32: /* Always emitted via tcg_reg_alloc_op. */
case INDEX_op_ext8s_i64:
case INDEX_op_ext8u_i32:
case INDEX_op_ext8u_i64:
default:
g_assert_not_reached();
}
#undef OP_32_64
}
static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc,
unsigned vecl, unsigned vece,
const TCGArg args[TCG_MAX_OP_ARGS],
const int const_args[TCG_MAX_OP_ARGS])
{
static int const add_insn[4] = {
OPC_PADDB, OPC_PADDW, OPC_PADDD, OPC_PADDQ
};
static int const ssadd_insn[4] = {
OPC_PADDSB, OPC_PADDSW, OPC_UD2, OPC_UD2
};
static int const usadd_insn[4] = {
OPC_PADDUB, OPC_PADDUW, OPC_UD2, OPC_UD2
};
static int const sub_insn[4] = {
OPC_PSUBB, OPC_PSUBW, OPC_PSUBD, OPC_PSUBQ
};
static int const sssub_insn[4] = {
OPC_PSUBSB, OPC_PSUBSW, OPC_UD2, OPC_UD2
};
static int const ussub_insn[4] = {
OPC_PSUBUB, OPC_PSUBUW, OPC_UD2, OPC_UD2
};
static int const mul_insn[4] = {
OPC_UD2, OPC_PMULLW, OPC_PMULLD, OPC_VPMULLQ
};
static int const shift_imm_insn[4] = {
OPC_UD2, OPC_PSHIFTW_Ib, OPC_PSHIFTD_Ib, OPC_PSHIFTQ_Ib
};
static int const cmpeq_insn[4] = {
OPC_PCMPEQB, OPC_PCMPEQW, OPC_PCMPEQD, OPC_PCMPEQQ
};
static int const cmpgt_insn[4] = {
OPC_PCMPGTB, OPC_PCMPGTW, OPC_PCMPGTD, OPC_PCMPGTQ
};
static int const punpckl_insn[4] = {
OPC_PUNPCKLBW, OPC_PUNPCKLWD, OPC_PUNPCKLDQ, OPC_PUNPCKLQDQ
};
static int const punpckh_insn[4] = {
OPC_PUNPCKHBW, OPC_PUNPCKHWD, OPC_PUNPCKHDQ, OPC_PUNPCKHQDQ
};
static int const packss_insn[4] = {
OPC_PACKSSWB, OPC_PACKSSDW, OPC_UD2, OPC_UD2
};
static int const packus_insn[4] = {
OPC_PACKUSWB, OPC_PACKUSDW, OPC_UD2, OPC_UD2
};
static int const smin_insn[4] = {
OPC_PMINSB, OPC_PMINSW, OPC_PMINSD, OPC_VPMINSQ
};
static int const smax_insn[4] = {
OPC_PMAXSB, OPC_PMAXSW, OPC_PMAXSD, OPC_VPMAXSQ
};
static int const umin_insn[4] = {
OPC_PMINUB, OPC_PMINUW, OPC_PMINUD, OPC_VPMINUQ
};
static int const umax_insn[4] = {
OPC_PMAXUB, OPC_PMAXUW, OPC_PMAXUD, OPC_VPMAXUQ
};
static int const rotlv_insn[4] = {
OPC_UD2, OPC_UD2, OPC_VPROLVD, OPC_VPROLVQ
};
static int const rotrv_insn[4] = {
OPC_UD2, OPC_UD2, OPC_VPRORVD, OPC_VPRORVQ
};
static int const shlv_insn[4] = {
OPC_UD2, OPC_VPSLLVW, OPC_VPSLLVD, OPC_VPSLLVQ
};
static int const shrv_insn[4] = {
OPC_UD2, OPC_VPSRLVW, OPC_VPSRLVD, OPC_VPSRLVQ
};
static int const sarv_insn[4] = {
OPC_UD2, OPC_VPSRAVW, OPC_VPSRAVD, OPC_VPSRAVQ
};
static int const shls_insn[4] = {
OPC_UD2, OPC_PSLLW, OPC_PSLLD, OPC_PSLLQ
};
static int const shrs_insn[4] = {
OPC_UD2, OPC_PSRLW, OPC_PSRLD, OPC_PSRLQ
};
static int const sars_insn[4] = {
OPC_UD2, OPC_PSRAW, OPC_PSRAD, OPC_VPSRAQ
};
static int const vpshldi_insn[4] = {
OPC_UD2, OPC_VPSHLDW, OPC_VPSHLDD, OPC_VPSHLDQ
};
static int const vpshldv_insn[4] = {
OPC_UD2, OPC_VPSHLDVW, OPC_VPSHLDVD, OPC_VPSHLDVQ
};
static int const vpshrdv_insn[4] = {
OPC_UD2, OPC_VPSHRDVW, OPC_VPSHRDVD, OPC_VPSHRDVQ
};
static int const abs_insn[4] = {
OPC_PABSB, OPC_PABSW, OPC_PABSD, OPC_VPABSQ
};
TCGType type = vecl + TCG_TYPE_V64;
int insn, sub;
TCGArg a0, a1, a2, a3;
a0 = args[0];
a1 = args[1];
a2 = args[2];
switch (opc) {
case INDEX_op_add_vec:
insn = add_insn[vece];
goto gen_simd;
case INDEX_op_ssadd_vec:
insn = ssadd_insn[vece];
goto gen_simd;
case INDEX_op_usadd_vec:
insn = usadd_insn[vece];
goto gen_simd;
case INDEX_op_sub_vec:
insn = sub_insn[vece];
goto gen_simd;
case INDEX_op_sssub_vec:
insn = sssub_insn[vece];
goto gen_simd;
case INDEX_op_ussub_vec:
insn = ussub_insn[vece];
goto gen_simd;
case INDEX_op_mul_vec:
insn = mul_insn[vece];
goto gen_simd;
case INDEX_op_and_vec:
insn = OPC_PAND;
goto gen_simd;
case INDEX_op_or_vec:
insn = OPC_POR;
goto gen_simd;
case INDEX_op_xor_vec:
insn = OPC_PXOR;
goto gen_simd;
case INDEX_op_smin_vec:
insn = smin_insn[vece];
goto gen_simd;
case INDEX_op_umin_vec:
insn = umin_insn[vece];
goto gen_simd;
case INDEX_op_smax_vec:
insn = smax_insn[vece];
goto gen_simd;
case INDEX_op_umax_vec:
insn = umax_insn[vece];
goto gen_simd;
case INDEX_op_shlv_vec:
insn = shlv_insn[vece];
goto gen_simd;
case INDEX_op_shrv_vec:
insn = shrv_insn[vece];
goto gen_simd;
case INDEX_op_sarv_vec:
insn = sarv_insn[vece];
goto gen_simd;
case INDEX_op_rotlv_vec:
insn = rotlv_insn[vece];
goto gen_simd;
case INDEX_op_rotrv_vec:
insn = rotrv_insn[vece];
goto gen_simd;
case INDEX_op_shls_vec:
insn = shls_insn[vece];
goto gen_simd;
case INDEX_op_shrs_vec:
insn = shrs_insn[vece];
goto gen_simd;
case INDEX_op_sars_vec:
insn = sars_insn[vece];
goto gen_simd;
case INDEX_op_x86_punpckl_vec:
insn = punpckl_insn[vece];
goto gen_simd;
case INDEX_op_x86_punpckh_vec:
insn = punpckh_insn[vece];
goto gen_simd;
case INDEX_op_x86_packss_vec:
insn = packss_insn[vece];
goto gen_simd;
case INDEX_op_x86_packus_vec:
insn = packus_insn[vece];
goto gen_simd;
case INDEX_op_x86_vpshldv_vec:
insn = vpshldv_insn[vece];
a1 = a2;
a2 = args[3];
goto gen_simd;
case INDEX_op_x86_vpshrdv_vec:
insn = vpshrdv_insn[vece];
a1 = a2;
a2 = args[3];
goto gen_simd;
#if TCG_TARGET_REG_BITS == 32
case INDEX_op_dup2_vec:
/* First merge the two 32-bit inputs to a single 64-bit element. */
tcg_out_vex_modrm(s, OPC_PUNPCKLDQ, a0, a1, a2);
/* Then replicate the 64-bit elements across the rest of the vector. */
if (type != TCG_TYPE_V64) {
tcg_out_dup_vec(s, type, MO_64, a0, a0);
}
break;
#endif
case INDEX_op_abs_vec:
insn = abs_insn[vece];
a2 = a1;
a1 = 0;
goto gen_simd;
gen_simd:
tcg_debug_assert(insn != OPC_UD2);
if (type == TCG_TYPE_V256) {
insn |= P_VEXL;
}
tcg_out_vex_modrm(s, insn, a0, a1, a2);
break;
case INDEX_op_cmp_vec:
sub = args[3];
if (sub == TCG_COND_EQ) {
insn = cmpeq_insn[vece];
} else if (sub == TCG_COND_GT) {
insn = cmpgt_insn[vece];
} else {
g_assert_not_reached();
}
goto gen_simd;
case INDEX_op_andc_vec:
insn = OPC_PANDN;
if (type == TCG_TYPE_V256) {
insn |= P_VEXL;
}
tcg_out_vex_modrm(s, insn, a0, a2, a1);
break;
case INDEX_op_shli_vec:
insn = shift_imm_insn[vece];
sub = 6;
goto gen_shift;
case INDEX_op_shri_vec:
insn = shift_imm_insn[vece];
sub = 2;
goto gen_shift;
case INDEX_op_sari_vec:
if (vece == MO_64) {
insn = OPC_PSHIFTD_Ib | P_VEXW | P_EVEX;
} else {
insn = shift_imm_insn[vece];
}
sub = 4;
goto gen_shift;
case INDEX_op_rotli_vec:
insn = OPC_PSHIFTD_Ib | P_EVEX; /* VPROL[DQ] */
if (vece == MO_64) {
insn |= P_VEXW;
}
sub = 1;
goto gen_shift;
gen_shift:
tcg_debug_assert(vece != MO_8);
if (type == TCG_TYPE_V256) {
insn |= P_VEXL;
}
tcg_out_vex_modrm(s, insn, sub, a0, a1);
tcg_out8(s, a2);
break;
case INDEX_op_ld_vec:
tcg_out_ld(s, type, a0, a1, a2);
break;
case INDEX_op_st_vec:
tcg_out_st(s, type, a0, a1, a2);
break;
case INDEX_op_dupm_vec:
tcg_out_dupm_vec(s, type, vece, a0, a1, a2);
break;
case INDEX_op_x86_shufps_vec:
insn = OPC_SHUFPS;
sub = args[3];
goto gen_simd_imm8;
case INDEX_op_x86_blend_vec:
if (vece == MO_16) {
insn = OPC_PBLENDW;
} else if (vece == MO_32) {
insn = (have_avx2 ? OPC_VPBLENDD : OPC_BLENDPS);
} else {
g_assert_not_reached();
}
sub = args[3];
goto gen_simd_imm8;
case INDEX_op_x86_vperm2i128_vec:
insn = OPC_VPERM2I128;
sub = args[3];
goto gen_simd_imm8;
case INDEX_op_x86_vpshldi_vec:
insn = vpshldi_insn[vece];
sub = args[3];
goto gen_simd_imm8;
case INDEX_op_not_vec:
insn = OPC_VPTERNLOGQ;
a2 = a1;
sub = 0x33; /* !B */
goto gen_simd_imm8;
case INDEX_op_nor_vec:
insn = OPC_VPTERNLOGQ;
sub = 0x11; /* norCB */
goto gen_simd_imm8;
case INDEX_op_nand_vec:
insn = OPC_VPTERNLOGQ;
sub = 0x77; /* nandCB */
goto gen_simd_imm8;
case INDEX_op_eqv_vec:
insn = OPC_VPTERNLOGQ;
sub = 0x99; /* xnorCB */
goto gen_simd_imm8;
case INDEX_op_orc_vec:
insn = OPC_VPTERNLOGQ;
sub = 0xdd; /* orB!C */
goto gen_simd_imm8;
case INDEX_op_bitsel_vec:
insn = OPC_VPTERNLOGQ;
a3 = args[3];
if (a0 == a1) {
a1 = a2;
a2 = a3;
sub = 0xca; /* A?B:C */
} else if (a0 == a2) {
a2 = a3;
sub = 0xe2; /* B?A:C */
} else {
tcg_out_mov(s, type, a0, a3);
sub = 0xb8; /* B?C:A */
}
goto gen_simd_imm8;
gen_simd_imm8:
tcg_debug_assert(insn != OPC_UD2);
if (type == TCG_TYPE_V256) {
insn |= P_VEXL;
}
tcg_out_vex_modrm(s, insn, a0, a1, a2);
tcg_out8(s, sub);
break;
case INDEX_op_x86_vpblendvb_vec:
insn = OPC_VPBLENDVB;
if (type == TCG_TYPE_V256) {
insn |= P_VEXL;
}
tcg_out_vex_modrm(s, insn, a0, a1, a2);
tcg_out8(s, args[3] << 4);
break;
case INDEX_op_x86_psrldq_vec:
tcg_out_vex_modrm(s, OPC_GRP14, 3, a0, a1);
tcg_out8(s, a2);
break;
case INDEX_op_mov_vec: /* Always emitted via tcg_out_mov. */
case INDEX_op_dup_vec: /* Always emitted via tcg_out_dup_vec. */
default:
g_assert_not_reached();
}
}
static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode op)
{
switch (op) {
case INDEX_op_goto_ptr:
return C_O0_I1(r);
case INDEX_op_ld8u_i32:
case INDEX_op_ld8u_i64:
case INDEX_op_ld8s_i32:
case INDEX_op_ld8s_i64:
case INDEX_op_ld16u_i32:
case INDEX_op_ld16u_i64:
case INDEX_op_ld16s_i32:
case INDEX_op_ld16s_i64:
case INDEX_op_ld_i32:
case INDEX_op_ld32u_i64:
case INDEX_op_ld32s_i64:
case INDEX_op_ld_i64:
return C_O1_I1(r, r);
case INDEX_op_st8_i32:
case INDEX_op_st8_i64:
return C_O0_I2(qi, r);
case INDEX_op_st16_i32:
case INDEX_op_st16_i64:
case INDEX_op_st_i32:
case INDEX_op_st32_i64:
return C_O0_I2(ri, r);
case INDEX_op_st_i64:
return C_O0_I2(re, r);
case INDEX_op_add_i32:
case INDEX_op_add_i64:
return C_O1_I2(r, r, re);
case INDEX_op_sub_i32:
case INDEX_op_sub_i64:
case INDEX_op_mul_i32:
case INDEX_op_mul_i64:
case INDEX_op_or_i32:
case INDEX_op_or_i64:
case INDEX_op_xor_i32:
case INDEX_op_xor_i64:
return C_O1_I2(r, 0, re);
case INDEX_op_and_i32:
case INDEX_op_and_i64:
return C_O1_I2(r, 0, reZ);
case INDEX_op_andc_i32:
case INDEX_op_andc_i64:
return C_O1_I2(r, r, rI);
case INDEX_op_shl_i32:
case INDEX_op_shl_i64:
case INDEX_op_shr_i32:
case INDEX_op_shr_i64:
case INDEX_op_sar_i32:
case INDEX_op_sar_i64:
return have_bmi2 ? C_O1_I2(r, r, ri) : C_O1_I2(r, 0, ci);
case INDEX_op_rotl_i32:
case INDEX_op_rotl_i64:
case INDEX_op_rotr_i32:
case INDEX_op_rotr_i64:
return C_O1_I2(r, 0, ci);
case INDEX_op_brcond_i32:
case INDEX_op_brcond_i64:
return C_O0_I2(r, re);
case INDEX_op_bswap16_i32:
case INDEX_op_bswap16_i64:
case INDEX_op_bswap32_i32:
case INDEX_op_bswap32_i64:
case INDEX_op_bswap64_i64:
case INDEX_op_neg_i32:
case INDEX_op_neg_i64:
case INDEX_op_not_i32:
case INDEX_op_not_i64:
case INDEX_op_extrh_i64_i32:
return C_O1_I1(r, 0);
case INDEX_op_ext8s_i32:
case INDEX_op_ext8s_i64:
case INDEX_op_ext8u_i32:
case INDEX_op_ext8u_i64:
return C_O1_I1(r, q);
case INDEX_op_ext16s_i32:
case INDEX_op_ext16s_i64:
case INDEX_op_ext16u_i32:
case INDEX_op_ext16u_i64:
case INDEX_op_ext32s_i64:
case INDEX_op_ext32u_i64:
case INDEX_op_ext_i32_i64:
case INDEX_op_extu_i32_i64:
case INDEX_op_extrl_i64_i32:
case INDEX_op_extract_i32:
case INDEX_op_extract_i64:
case INDEX_op_sextract_i32:
case INDEX_op_ctpop_i32:
case INDEX_op_ctpop_i64:
return C_O1_I1(r, r);
case INDEX_op_extract2_i32:
case INDEX_op_extract2_i64:
return C_O1_I2(r, 0, r);
case INDEX_op_deposit_i32:
case INDEX_op_deposit_i64:
return C_O1_I2(Q, 0, Q);
case INDEX_op_setcond_i32:
case INDEX_op_setcond_i64:
return C_O1_I2(q, r, re);
case INDEX_op_movcond_i32:
case INDEX_op_movcond_i64:
return C_O1_I4(r, r, re, r, 0);
case INDEX_op_div2_i32:
case INDEX_op_div2_i64:
case INDEX_op_divu2_i32:
case INDEX_op_divu2_i64:
return C_O2_I3(a, d, 0, 1, r);
case INDEX_op_mulu2_i32:
case INDEX_op_mulu2_i64:
case INDEX_op_muls2_i32:
case INDEX_op_muls2_i64:
return C_O2_I2(a, d, a, r);
case INDEX_op_add2_i32:
case INDEX_op_add2_i64:
case INDEX_op_sub2_i32:
case INDEX_op_sub2_i64:
return C_O2_I4(r, r, 0, 1, re, re);
case INDEX_op_ctz_i32:
case INDEX_op_ctz_i64:
return have_bmi1 ? C_N1_I2(r, r, rW) : C_N1_I2(r, r, r);
case INDEX_op_clz_i32:
case INDEX_op_clz_i64:
return have_lzcnt ? C_N1_I2(r, r, rW) : C_N1_I2(r, r, r);
case INDEX_op_qemu_ld_i32:
return (TARGET_LONG_BITS <= TCG_TARGET_REG_BITS
? C_O1_I1(r, L) : C_O1_I2(r, L, L));
case INDEX_op_qemu_st_i32:
return (TARGET_LONG_BITS <= TCG_TARGET_REG_BITS
? C_O0_I2(L, L) : C_O0_I3(L, L, L));
case INDEX_op_qemu_st8_i32:
return (TARGET_LONG_BITS <= TCG_TARGET_REG_BITS
? C_O0_I2(s, L) : C_O0_I3(s, L, L));
case INDEX_op_qemu_ld_i64:
return (TCG_TARGET_REG_BITS == 64 ? C_O1_I1(r, L)
: TARGET_LONG_BITS <= TCG_TARGET_REG_BITS ? C_O2_I1(r, r, L)
: C_O2_I2(r, r, L, L));
case INDEX_op_qemu_st_i64:
return (TCG_TARGET_REG_BITS == 64 ? C_O0_I2(L, L)
: TARGET_LONG_BITS <= TCG_TARGET_REG_BITS ? C_O0_I3(L, L, L)
: C_O0_I4(L, L, L, L));
case INDEX_op_brcond2_i32:
return C_O0_I4(r, r, ri, ri);
case INDEX_op_setcond2_i32:
return C_O1_I4(r, r, r, ri, ri);
case INDEX_op_ld_vec:
case INDEX_op_dupm_vec:
return C_O1_I1(x, r);
case INDEX_op_st_vec:
return C_O0_I2(x, r);
case INDEX_op_add_vec:
case INDEX_op_sub_vec:
case INDEX_op_mul_vec:
case INDEX_op_and_vec:
case INDEX_op_or_vec:
case INDEX_op_xor_vec:
case INDEX_op_andc_vec:
case INDEX_op_orc_vec:
case INDEX_op_nand_vec:
case INDEX_op_nor_vec:
case INDEX_op_eqv_vec:
case INDEX_op_ssadd_vec:
case INDEX_op_usadd_vec:
case INDEX_op_sssub_vec:
case INDEX_op_ussub_vec:
case INDEX_op_smin_vec:
case INDEX_op_umin_vec:
case INDEX_op_smax_vec:
case INDEX_op_umax_vec:
case INDEX_op_shlv_vec:
case INDEX_op_shrv_vec:
case INDEX_op_sarv_vec:
case INDEX_op_rotlv_vec:
case INDEX_op_rotrv_vec:
case INDEX_op_shls_vec:
case INDEX_op_shrs_vec:
case INDEX_op_sars_vec:
case INDEX_op_cmp_vec:
case INDEX_op_x86_shufps_vec:
case INDEX_op_x86_blend_vec:
case INDEX_op_x86_packss_vec:
case INDEX_op_x86_packus_vec:
case INDEX_op_x86_vperm2i128_vec:
case INDEX_op_x86_punpckl_vec:
case INDEX_op_x86_punpckh_vec:
case INDEX_op_x86_vpshldi_vec:
#if TCG_TARGET_REG_BITS == 32
case INDEX_op_dup2_vec:
#endif
return C_O1_I2(x, x, x);
case INDEX_op_abs_vec:
case INDEX_op_dup_vec:
case INDEX_op_not_vec:
case INDEX_op_shli_vec:
case INDEX_op_shri_vec:
case INDEX_op_sari_vec:
case INDEX_op_rotli_vec:
case INDEX_op_x86_psrldq_vec:
return C_O1_I1(x, x);
case INDEX_op_x86_vpshldv_vec:
case INDEX_op_x86_vpshrdv_vec:
return C_O1_I3(x, 0, x, x);
case INDEX_op_bitsel_vec:
case INDEX_op_x86_vpblendvb_vec:
return C_O1_I3(x, x, x, x);
default:
g_assert_not_reached();
}
}
int tcg_can_emit_vec_op(TCGOpcode opc, TCGType type, unsigned vece)
{
switch (opc) {
case INDEX_op_add_vec:
case INDEX_op_sub_vec:
case INDEX_op_and_vec:
case INDEX_op_or_vec:
case INDEX_op_xor_vec:
case INDEX_op_andc_vec:
case INDEX_op_orc_vec:
case INDEX_op_nand_vec:
case INDEX_op_nor_vec:
case INDEX_op_eqv_vec:
case INDEX_op_not_vec:
case INDEX_op_bitsel_vec:
return 1;
case INDEX_op_cmp_vec:
case INDEX_op_cmpsel_vec:
return -1;
case INDEX_op_rotli_vec:
return have_avx512vl && vece >= MO_32 ? 1 : -1;
case INDEX_op_shli_vec:
case INDEX_op_shri_vec:
/* We must expand the operation for MO_8. */
return vece == MO_8 ? -1 : 1;
case INDEX_op_sari_vec:
switch (vece) {
case MO_8:
return -1;
case MO_16:
case MO_32:
return 1;
case MO_64:
if (have_avx512vl) {
return 1;
}
/*
* We can emulate this for MO_64, but it does not pay off
* unless we're producing at least 4 values.
*/
return type >= TCG_TYPE_V256 ? -1 : 0;
}
return 0;
case INDEX_op_shls_vec:
case INDEX_op_shrs_vec:
return vece >= MO_16;
case INDEX_op_sars_vec:
switch (vece) {
case MO_16:
case MO_32:
return 1;
case MO_64:
return have_avx512vl;
}
return 0;
case INDEX_op_rotls_vec:
return vece >= MO_16 ? -1 : 0;
case INDEX_op_shlv_vec:
case INDEX_op_shrv_vec:
switch (vece) {
case MO_16:
return have_avx512bw;
case MO_32:
case MO_64:
return have_avx2;
}
return 0;
case INDEX_op_sarv_vec:
switch (vece) {
case MO_16:
return have_avx512bw;
case MO_32:
return have_avx2;
case MO_64:
return have_avx512vl;
}
return 0;
case INDEX_op_rotlv_vec:
case INDEX_op_rotrv_vec:
switch (vece) {
case MO_16:
return have_avx512vbmi2 ? -1 : 0;
case MO_32:
case MO_64:
return have_avx512vl ? 1 : have_avx2 ? -1 : 0;
}
return 0;
case INDEX_op_mul_vec:
switch (vece) {
case MO_8:
return -1;
case MO_64:
return have_avx512dq;
}
return 1;
case INDEX_op_ssadd_vec:
case INDEX_op_usadd_vec:
case INDEX_op_sssub_vec:
case INDEX_op_ussub_vec:
return vece <= MO_16;
case INDEX_op_smin_vec:
case INDEX_op_smax_vec:
case INDEX_op_umin_vec:
case INDEX_op_umax_vec:
case INDEX_op_abs_vec:
return vece <= MO_32 || have_avx512vl;
default:
return 0;
}
}
static void expand_vec_shi(TCGType type, unsigned vece, TCGOpcode opc,
TCGv_vec v0, TCGv_vec v1, TCGArg imm)
{
TCGv_vec t1, t2;
tcg_debug_assert(vece == MO_8);
t1 = tcg_temp_new_vec(type);
t2 = tcg_temp_new_vec(type);
/*
* Unpack to W, shift, and repack. Tricky bits:
* (1) Use punpck*bw x,x to produce DDCCBBAA,
* i.e. duplicate in other half of the 16-bit lane.
* (2) For right-shift, add 8 so that the high half of the lane
* becomes zero. For left-shift, and left-rotate, we must
* shift up and down again.
* (3) Step 2 leaves high half zero such that PACKUSWB
* (pack with unsigned saturation) does not modify
* the quantity.
*/
vec_gen_3(INDEX_op_x86_punpckl_vec, type, MO_8,
tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(v1));
vec_gen_3(INDEX_op_x86_punpckh_vec, type, MO_8,
tcgv_vec_arg(t2), tcgv_vec_arg(v1), tcgv_vec_arg(v1));
if (opc != INDEX_op_rotli_vec) {
imm += 8;
}
if (opc == INDEX_op_shri_vec) {
tcg_gen_shri_vec(MO_16, t1, t1, imm);
tcg_gen_shri_vec(MO_16, t2, t2, imm);
} else {
tcg_gen_shli_vec(MO_16, t1, t1, imm);
tcg_gen_shli_vec(MO_16, t2, t2, imm);
tcg_gen_shri_vec(MO_16, t1, t1, 8);
tcg_gen_shri_vec(MO_16, t2, t2, 8);
}
vec_gen_3(INDEX_op_x86_packus_vec, type, MO_8,
tcgv_vec_arg(v0), tcgv_vec_arg(t1), tcgv_vec_arg(t2));
tcg_temp_free_vec(t1);
tcg_temp_free_vec(t2);
}
static void expand_vec_sari(TCGType type, unsigned vece,
TCGv_vec v0, TCGv_vec v1, TCGArg imm)
{
TCGv_vec t1, t2;
switch (vece) {
case MO_8:
/* Unpack to W, shift, and repack, as in expand_vec_shi. */
t1 = tcg_temp_new_vec(type);
t2 = tcg_temp_new_vec(type);
vec_gen_3(INDEX_op_x86_punpckl_vec, type, MO_8,
tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(v1));
vec_gen_3(INDEX_op_x86_punpckh_vec, type, MO_8,
tcgv_vec_arg(t2), tcgv_vec_arg(v1), tcgv_vec_arg(v1));
tcg_gen_sari_vec(MO_16, t1, t1, imm + 8);
tcg_gen_sari_vec(MO_16, t2, t2, imm + 8);
vec_gen_3(INDEX_op_x86_packss_vec, type, MO_8,
tcgv_vec_arg(v0), tcgv_vec_arg(t1), tcgv_vec_arg(t2));
tcg_temp_free_vec(t1);
tcg_temp_free_vec(t2);
break;
case MO_64:
t1 = tcg_temp_new_vec(type);
if (imm <= 32) {
/*
* We can emulate a small sign extend by performing an arithmetic
* 32-bit shift and overwriting the high half of a 64-bit logical
* shift. Note that the ISA says shift of 32 is valid, but TCG
* does not, so we have to bound the smaller shift -- we get the
* same result in the high half either way.
*/
tcg_gen_sari_vec(MO_32, t1, v1, MIN(imm, 31));
tcg_gen_shri_vec(MO_64, v0, v1, imm);
vec_gen_4(INDEX_op_x86_blend_vec, type, MO_32,
tcgv_vec_arg(v0), tcgv_vec_arg(v0),
tcgv_vec_arg(t1), 0xaa);
} else {
/* Otherwise we will need to use a compare vs 0 to produce
* the sign-extend, shift and merge.
*/
tcg_gen_cmp_vec(TCG_COND_GT, MO_64, t1,
tcg_constant_vec(type, MO_64, 0), v1);
tcg_gen_shri_vec(MO_64, v0, v1, imm);
tcg_gen_shli_vec(MO_64, t1, t1, 64 - imm);
tcg_gen_or_vec(MO_64, v0, v0, t1);
}
tcg_temp_free_vec(t1);
break;
default:
g_assert_not_reached();
}
}
static void expand_vec_rotli(TCGType type, unsigned vece,
TCGv_vec v0, TCGv_vec v1, TCGArg imm)
{
TCGv_vec t;
if (vece == MO_8) {
expand_vec_shi(type, vece, INDEX_op_rotli_vec, v0, v1, imm);
return;
}
if (have_avx512vbmi2) {
vec_gen_4(INDEX_op_x86_vpshldi_vec, type, vece,
tcgv_vec_arg(v0), tcgv_vec_arg(v1), tcgv_vec_arg(v1), imm);
return;
}
t = tcg_temp_new_vec(type);
tcg_gen_shli_vec(vece, t, v1, imm);
tcg_gen_shri_vec(vece, v0, v1, (8 << vece) - imm);
tcg_gen_or_vec(vece, v0, v0, t);
tcg_temp_free_vec(t);
}
static void expand_vec_rotv(TCGType type, unsigned vece, TCGv_vec v0,
TCGv_vec v1, TCGv_vec sh, bool right)
{
TCGv_vec t;
if (have_avx512vbmi2) {
vec_gen_4(right ? INDEX_op_x86_vpshrdv_vec : INDEX_op_x86_vpshldv_vec,
type, vece, tcgv_vec_arg(v0), tcgv_vec_arg(v1),
tcgv_vec_arg(v1), tcgv_vec_arg(sh));
return;
}
t = tcg_temp_new_vec(type);
tcg_gen_dupi_vec(vece, t, 8 << vece);
tcg_gen_sub_vec(vece, t, t, sh);
if (right) {
tcg_gen_shlv_vec(vece, t, v1, t);
tcg_gen_shrv_vec(vece, v0, v1, sh);
} else {
tcg_gen_shrv_vec(vece, t, v1, t);
tcg_gen_shlv_vec(vece, v0, v1, sh);
}
tcg_gen_or_vec(vece, v0, v0, t);
tcg_temp_free_vec(t);
}
static void expand_vec_rotls(TCGType type, unsigned vece,
TCGv_vec v0, TCGv_vec v1, TCGv_i32 lsh)
{
TCGv_vec t = tcg_temp_new_vec(type);
tcg_debug_assert(vece != MO_8);
if (vece >= MO_32 ? have_avx512vl : have_avx512vbmi2) {
tcg_gen_dup_i32_vec(vece, t, lsh);
if (vece >= MO_32) {
tcg_gen_rotlv_vec(vece, v0, v1, t);
} else {
expand_vec_rotv(type, vece, v0, v1, t, false);
}
} else {
TCGv_i32 rsh = tcg_temp_new_i32();
tcg_gen_neg_i32(rsh, lsh);
tcg_gen_andi_i32(rsh, rsh, (8 << vece) - 1);
tcg_gen_shls_vec(vece, t, v1, lsh);
tcg_gen_shrs_vec(vece, v0, v1, rsh);
tcg_gen_or_vec(vece, v0, v0, t);
tcg_temp_free_i32(rsh);
}
tcg_temp_free_vec(t);
}
static void expand_vec_mul(TCGType type, unsigned vece,
TCGv_vec v0, TCGv_vec v1, TCGv_vec v2)
{
TCGv_vec t1, t2, t3, t4, zero;
tcg_debug_assert(vece == MO_8);
/*
* Unpack v1 bytes to words, 0 | x.
* Unpack v2 bytes to words, y | 0.
* This leaves the 8-bit result, x * y, with 8 bits of right padding.
* Shift logical right by 8 bits to clear the high 8 bytes before
* using an unsigned saturated pack.
*
* The difference between the V64, V128 and V256 cases is merely how
* we distribute the expansion between temporaries.
*/
switch (type) {
case TCG_TYPE_V64:
t1 = tcg_temp_new_vec(TCG_TYPE_V128);
t2 = tcg_temp_new_vec(TCG_TYPE_V128);
zero = tcg_constant_vec(TCG_TYPE_V128, MO_8, 0);
vec_gen_3(INDEX_op_x86_punpckl_vec, TCG_TYPE_V128, MO_8,
tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(zero));
vec_gen_3(INDEX_op_x86_punpckl_vec, TCG_TYPE_V128, MO_8,
tcgv_vec_arg(t2), tcgv_vec_arg(zero), tcgv_vec_arg(v2));
tcg_gen_mul_vec(MO_16, t1, t1, t2);
tcg_gen_shri_vec(MO_16, t1, t1, 8);
vec_gen_3(INDEX_op_x86_packus_vec, TCG_TYPE_V128, MO_8,
tcgv_vec_arg(v0), tcgv_vec_arg(t1), tcgv_vec_arg(t1));
tcg_temp_free_vec(t1);
tcg_temp_free_vec(t2);
break;
case TCG_TYPE_V128:
case TCG_TYPE_V256:
t1 = tcg_temp_new_vec(type);
t2 = tcg_temp_new_vec(type);
t3 = tcg_temp_new_vec(type);
t4 = tcg_temp_new_vec(type);
zero = tcg_constant_vec(TCG_TYPE_V128, MO_8, 0);
vec_gen_3(INDEX_op_x86_punpckl_vec, type, MO_8,
tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(zero));
vec_gen_3(INDEX_op_x86_punpckl_vec, type, MO_8,
tcgv_vec_arg(t2), tcgv_vec_arg(zero), tcgv_vec_arg(v2));
vec_gen_3(INDEX_op_x86_punpckh_vec, type, MO_8,
tcgv_vec_arg(t3), tcgv_vec_arg(v1), tcgv_vec_arg(zero));
vec_gen_3(INDEX_op_x86_punpckh_vec, type, MO_8,
tcgv_vec_arg(t4), tcgv_vec_arg(zero), tcgv_vec_arg(v2));
tcg_gen_mul_vec(MO_16, t1, t1, t2);
tcg_gen_mul_vec(MO_16, t3, t3, t4);
tcg_gen_shri_vec(MO_16, t1, t1, 8);
tcg_gen_shri_vec(MO_16, t3, t3, 8);
vec_gen_3(INDEX_op_x86_packus_vec, type, MO_8,
tcgv_vec_arg(v0), tcgv_vec_arg(t1), tcgv_vec_arg(t3));
tcg_temp_free_vec(t1);
tcg_temp_free_vec(t2);
tcg_temp_free_vec(t3);
tcg_temp_free_vec(t4);
break;
default:
g_assert_not_reached();
}
}
static bool expand_vec_cmp_noinv(TCGType type, unsigned vece, TCGv_vec v0,
TCGv_vec v1, TCGv_vec v2, TCGCond cond)
{
enum {
NEED_INV = 1,
NEED_SWAP = 2,
NEED_BIAS = 4,
NEED_UMIN = 8,
NEED_UMAX = 16,
};
TCGv_vec t1, t2, t3;
uint8_t fixup;
switch (cond) {
case TCG_COND_EQ:
case TCG_COND_GT:
fixup = 0;
break;
case TCG_COND_NE:
case TCG_COND_LE:
fixup = NEED_INV;
break;
case TCG_COND_LT:
fixup = NEED_SWAP;
break;
case TCG_COND_GE:
fixup = NEED_SWAP | NEED_INV;
break;
case TCG_COND_LEU:
if (tcg_can_emit_vec_op(INDEX_op_umin_vec, type, vece)) {
fixup = NEED_UMIN;
} else {
fixup = NEED_BIAS | NEED_INV;
}
break;
case TCG_COND_GTU:
if (tcg_can_emit_vec_op(INDEX_op_umin_vec, type, vece)) {
fixup = NEED_UMIN | NEED_INV;
} else {
fixup = NEED_BIAS;
}
break;
case TCG_COND_GEU:
if (tcg_can_emit_vec_op(INDEX_op_umax_vec, type, vece)) {
fixup = NEED_UMAX;
} else {
fixup = NEED_BIAS | NEED_SWAP | NEED_INV;
}
break;
case TCG_COND_LTU:
if (tcg_can_emit_vec_op(INDEX_op_umax_vec, type, vece)) {
fixup = NEED_UMAX | NEED_INV;
} else {
fixup = NEED_BIAS | NEED_SWAP;
}
break;
default:
g_assert_not_reached();
}
if (fixup & NEED_INV) {
cond = tcg_invert_cond(cond);
}
if (fixup & NEED_SWAP) {
t1 = v1, v1 = v2, v2 = t1;
cond = tcg_swap_cond(cond);
}
t1 = t2 = NULL;
if (fixup & (NEED_UMIN | NEED_UMAX)) {
t1 = tcg_temp_new_vec(type);
if (fixup & NEED_UMIN) {
tcg_gen_umin_vec(vece, t1, v1, v2);
} else {
tcg_gen_umax_vec(vece, t1, v1, v2);
}
v2 = t1;
cond = TCG_COND_EQ;
} else if (fixup & NEED_BIAS) {
t1 = tcg_temp_new_vec(type);
t2 = tcg_temp_new_vec(type);
t3 = tcg_constant_vec(type, vece, 1ull << ((8 << vece) - 1));
tcg_gen_sub_vec(vece, t1, v1, t3);
tcg_gen_sub_vec(vece, t2, v2, t3);
v1 = t1;
v2 = t2;
cond = tcg_signed_cond(cond);
}
tcg_debug_assert(cond == TCG_COND_EQ || cond == TCG_COND_GT);
/* Expand directly; do not recurse. */
vec_gen_4(INDEX_op_cmp_vec, type, vece,
tcgv_vec_arg(v0), tcgv_vec_arg(v1), tcgv_vec_arg(v2), cond);
if (t1) {
tcg_temp_free_vec(t1);
if (t2) {
tcg_temp_free_vec(t2);
}
}
return fixup & NEED_INV;
}
static void expand_vec_cmp(TCGType type, unsigned vece, TCGv_vec v0,
TCGv_vec v1, TCGv_vec v2, TCGCond cond)
{
if (expand_vec_cmp_noinv(type, vece, v0, v1, v2, cond)) {
tcg_gen_not_vec(vece, v0, v0);
}
}
static void expand_vec_cmpsel(TCGType type, unsigned vece, TCGv_vec v0,
TCGv_vec c1, TCGv_vec c2,
TCGv_vec v3, TCGv_vec v4, TCGCond cond)
{
TCGv_vec t = tcg_temp_new_vec(type);
if (expand_vec_cmp_noinv(type, vece, t, c1, c2, cond)) {
/* Invert the sense of the compare by swapping arguments. */
TCGv_vec x;
x = v3, v3 = v4, v4 = x;
}
vec_gen_4(INDEX_op_x86_vpblendvb_vec, type, vece,
tcgv_vec_arg(v0), tcgv_vec_arg(v4),
tcgv_vec_arg(v3), tcgv_vec_arg(t));
tcg_temp_free_vec(t);
}
void tcg_expand_vec_op(TCGOpcode opc, TCGType type, unsigned vece,
TCGArg a0, ...)
{
va_list va;
TCGArg a2;
TCGv_vec v0, v1, v2, v3, v4;
va_start(va, a0);
v0 = temp_tcgv_vec(arg_temp(a0));
v1 = temp_tcgv_vec(arg_temp(va_arg(va, TCGArg)));
a2 = va_arg(va, TCGArg);
switch (opc) {
case INDEX_op_shli_vec:
case INDEX_op_shri_vec:
expand_vec_shi(type, vece, opc, v0, v1, a2);
break;
case INDEX_op_sari_vec:
expand_vec_sari(type, vece, v0, v1, a2);
break;
case INDEX_op_rotli_vec:
expand_vec_rotli(type, vece, v0, v1, a2);
break;
case INDEX_op_rotls_vec:
expand_vec_rotls(type, vece, v0, v1, temp_tcgv_i32(arg_temp(a2)));
break;
case INDEX_op_rotlv_vec:
v2 = temp_tcgv_vec(arg_temp(a2));
expand_vec_rotv(type, vece, v0, v1, v2, false);
break;
case INDEX_op_rotrv_vec:
v2 = temp_tcgv_vec(arg_temp(a2));
expand_vec_rotv(type, vece, v0, v1, v2, true);
break;
case INDEX_op_mul_vec:
v2 = temp_tcgv_vec(arg_temp(a2));
expand_vec_mul(type, vece, v0, v1, v2);
break;
case INDEX_op_cmp_vec:
v2 = temp_tcgv_vec(arg_temp(a2));
expand_vec_cmp(type, vece, v0, v1, v2, va_arg(va, TCGArg));
break;
case INDEX_op_cmpsel_vec:
v2 = temp_tcgv_vec(arg_temp(a2));
v3 = temp_tcgv_vec(arg_temp(va_arg(va, TCGArg)));
v4 = temp_tcgv_vec(arg_temp(va_arg(va, TCGArg)));
expand_vec_cmpsel(type, vece, v0, v1, v2, v3, v4, va_arg(va, TCGArg));
break;
default:
break;
}
va_end(va);
}
static const int tcg_target_callee_save_regs[] = {
#if TCG_TARGET_REG_BITS == 64
TCG_REG_RBP,
TCG_REG_RBX,
#if defined(_WIN64)
TCG_REG_RDI,
TCG_REG_RSI,
#endif
TCG_REG_R12,
TCG_REG_R13,
TCG_REG_R14, /* Currently used for the global env. */
TCG_REG_R15,
#else
TCG_REG_EBP, /* Currently used for the global env. */
TCG_REG_EBX,
TCG_REG_ESI,
TCG_REG_EDI,
#endif
};
/* Compute frame size via macros, to share between tcg_target_qemu_prologue
and tcg_register_jit. */
#define PUSH_SIZE \
((1 + ARRAY_SIZE(tcg_target_callee_save_regs)) \
* (TCG_TARGET_REG_BITS / 8))
#define FRAME_SIZE \
((PUSH_SIZE \
+ TCG_STATIC_CALL_ARGS_SIZE \
+ CPU_TEMP_BUF_NLONGS * sizeof(long) \
+ TCG_TARGET_STACK_ALIGN - 1) \
& ~(TCG_TARGET_STACK_ALIGN - 1))
/* Generate global QEMU prologue and epilogue code */
static void tcg_target_qemu_prologue(TCGContext *s)
{
int i, stack_addend;
/* TB prologue */
/* Reserve some stack space, also for TCG temps. */
stack_addend = FRAME_SIZE - PUSH_SIZE;
tcg_set_frame(s, TCG_REG_CALL_STACK, TCG_STATIC_CALL_ARGS_SIZE,
CPU_TEMP_BUF_NLONGS * sizeof(long));
/* Save all callee saved registers. */
for (i = 0; i < ARRAY_SIZE(tcg_target_callee_save_regs); i++) {
tcg_out_push(s, tcg_target_callee_save_regs[i]);
}
#if TCG_TARGET_REG_BITS == 32
tcg_out_ld(s, TCG_TYPE_PTR, TCG_AREG0, TCG_REG_ESP,
(ARRAY_SIZE(tcg_target_callee_save_regs) + 1) * 4);
tcg_out_addi(s, TCG_REG_ESP, -stack_addend);
/* jmp *tb. */
tcg_out_modrm_offset(s, OPC_GRP5, EXT5_JMPN_Ev, TCG_REG_ESP,
(ARRAY_SIZE(tcg_target_callee_save_regs) + 2) * 4
+ stack_addend);
#else
# if !defined(CONFIG_SOFTMMU) && TCG_TARGET_REG_BITS == 64
if (guest_base) {
int seg = setup_guest_base_seg();
if (seg != 0) {
x86_guest_base_seg = seg;
} else if (guest_base == (int32_t)guest_base) {
x86_guest_base_offset = guest_base;
} else {
/* Choose R12 because, as a base, it requires a SIB byte. */
x86_guest_base_index = TCG_REG_R12;
tcg_out_movi(s, TCG_TYPE_PTR, x86_guest_base_index, guest_base);
tcg_regset_set_reg(s->reserved_regs, x86_guest_base_index);
}
}
# endif
tcg_out_mov(s, TCG_TYPE_PTR, TCG_AREG0, tcg_target_call_iarg_regs[0]);
tcg_out_addi(s, TCG_REG_ESP, -stack_addend);
/* jmp *tb. */
tcg_out_modrm(s, OPC_GRP5, EXT5_JMPN_Ev, tcg_target_call_iarg_regs[1]);
#endif
/*
* Return path for goto_ptr. Set return value to 0, a-la exit_tb,
* and fall through to the rest of the epilogue.
*/
tcg_code_gen_epilogue = tcg_splitwx_to_rx(s->code_ptr);
tcg_out_movi(s, TCG_TYPE_REG, TCG_REG_EAX, 0);
/* TB epilogue */
tb_ret_addr = tcg_splitwx_to_rx(s->code_ptr);
tcg_out_addi(s, TCG_REG_CALL_STACK, stack_addend);
if (have_avx2) {
tcg_out_vex_opc(s, OPC_VZEROUPPER, 0, 0, 0, 0);
}
for (i = ARRAY_SIZE(tcg_target_callee_save_regs) - 1; i >= 0; i--) {
tcg_out_pop(s, tcg_target_callee_save_regs[i]);
}
tcg_out_opc(s, OPC_RET, 0, 0, 0);
}
static void tcg_out_nop_fill(tcg_insn_unit *p, int count)
{
memset(p, 0x90, count);
}
static void tcg_target_init(TCGContext *s)
{
#ifdef CONFIG_CPUID_H
unsigned a, b, c, d, b7 = 0, c7 = 0;
unsigned max = __get_cpuid_max(0, 0);
if (max >= 7) {
/* BMI1 is available on AMD Piledriver and Intel Haswell CPUs. */
__cpuid_count(7, 0, a, b7, c7, d);
have_bmi1 = (b7 & bit_BMI) != 0;
have_bmi2 = (b7 & bit_BMI2) != 0;
}
if (max >= 1) {
__cpuid(1, a, b, c, d);
#ifndef have_cmov
/* For 32-bit, 99% certainty that we're running on hardware that
supports cmov, but we still need to check. In case cmov is not
available, we'll use a small forward branch. */
have_cmov = (d & bit_CMOV) != 0;
#endif
/* MOVBE is only available on Intel Atom and Haswell CPUs, so we
need to probe for it. */
have_movbe = (c & bit_MOVBE) != 0;
have_popcnt = (c & bit_POPCNT) != 0;
/* There are a number of things we must check before we can be
sure of not hitting invalid opcode. */
if (c & bit_OSXSAVE) {
unsigned bv = xgetbv_low(0);
if ((bv & 6) == 6) {
have_avx1 = (c & bit_AVX) != 0;
have_avx2 = (b7 & bit_AVX2) != 0;
/*
* There are interesting instructions in AVX512, so long
* as we have AVX512VL, which indicates support for EVEX
* on sizes smaller than 512 bits. We are required to
* check that OPMASK and all extended ZMM state are enabled
* even if we're not using them -- the insns will fault.
*/
if ((bv & 0xe0) == 0xe0
&& (b7 & bit_AVX512F)
&& (b7 & bit_AVX512VL)) {
have_avx512vl = true;
have_avx512bw = (b7 & bit_AVX512BW) != 0;
have_avx512dq = (b7 & bit_AVX512DQ) != 0;
have_avx512vbmi2 = (c7 & bit_AVX512VBMI2) != 0;
}
}
}
}
max = __get_cpuid_max(0x8000000, 0);
if (max >= 1) {
__cpuid(0x80000001, a, b, c, d);
/* LZCNT was introduced with AMD Barcelona and Intel Haswell CPUs. */
have_lzcnt = (c & bit_LZCNT) != 0;
}
#endif /* CONFIG_CPUID_H */
tcg_target_available_regs[TCG_TYPE_I32] = ALL_GENERAL_REGS;
if (TCG_TARGET_REG_BITS == 64) {
tcg_target_available_regs[TCG_TYPE_I64] = ALL_GENERAL_REGS;
}
if (have_avx1) {
tcg_target_available_regs[TCG_TYPE_V64] = ALL_VECTOR_REGS;
tcg_target_available_regs[TCG_TYPE_V128] = ALL_VECTOR_REGS;
}
if (have_avx2) {
tcg_target_available_regs[TCG_TYPE_V256] = ALL_VECTOR_REGS;
}
tcg_target_call_clobber_regs = ALL_VECTOR_REGS;
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_EAX);
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_EDX);
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_ECX);
if (TCG_TARGET_REG_BITS == 64) {
#if !defined(_WIN64)
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_RDI);
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_RSI);
#endif
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R8);
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R9);
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R10);
tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R11);
}
s->reserved_regs = 0;
tcg_regset_set_reg(s->reserved_regs, TCG_REG_CALL_STACK);
#ifdef _WIN64
/* These are call saved, and we don't save them, so don't use them. */
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM6);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM7);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM8);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM9);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM10);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM11);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM12);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM13);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM14);
tcg_regset_set_reg(s->reserved_regs, TCG_REG_XMM15);
#endif
}
typedef struct {
DebugFrameHeader h;
uint8_t fde_def_cfa[4];
uint8_t fde_reg_ofs[14];
} DebugFrame;
/* We're expecting a 2 byte uleb128 encoded value. */
QEMU_BUILD_BUG_ON(FRAME_SIZE >= (1 << 14));
#if !defined(__ELF__)
/* Host machine without ELF. */
#elif TCG_TARGET_REG_BITS == 64
#define ELF_HOST_MACHINE EM_X86_64
static const DebugFrame debug_frame = {
.h.cie.len = sizeof(DebugFrameCIE)-4, /* length after .len member */
.h.cie.id = -1,
.h.cie.version = 1,
.h.cie.code_align = 1,
.h.cie.data_align = 0x78, /* sleb128 -8 */
.h.cie.return_column = 16,
/* Total FDE size does not include the "len" member. */
.h.fde.len = sizeof(DebugFrame) - offsetof(DebugFrame, h.fde.cie_offset),
.fde_def_cfa = {
12, 7, /* DW_CFA_def_cfa %rsp, ... */
(FRAME_SIZE & 0x7f) | 0x80, /* ... uleb128 FRAME_SIZE */
(FRAME_SIZE >> 7)
},
.fde_reg_ofs = {
0x90, 1, /* DW_CFA_offset, %rip, -8 */
/* The following ordering must match tcg_target_callee_save_regs. */
0x86, 2, /* DW_CFA_offset, %rbp, -16 */
0x83, 3, /* DW_CFA_offset, %rbx, -24 */
0x8c, 4, /* DW_CFA_offset, %r12, -32 */
0x8d, 5, /* DW_CFA_offset, %r13, -40 */
0x8e, 6, /* DW_CFA_offset, %r14, -48 */
0x8f, 7, /* DW_CFA_offset, %r15, -56 */
}
};
#else
#define ELF_HOST_MACHINE EM_386
static const DebugFrame debug_frame = {
.h.cie.len = sizeof(DebugFrameCIE)-4, /* length after .len member */
.h.cie.id = -1,
.h.cie.version = 1,
.h.cie.code_align = 1,
.h.cie.data_align = 0x7c, /* sleb128 -4 */
.h.cie.return_column = 8,
/* Total FDE size does not include the "len" member. */
.h.fde.len = sizeof(DebugFrame) - offsetof(DebugFrame, h.fde.cie_offset),
.fde_def_cfa = {
12, 4, /* DW_CFA_def_cfa %esp, ... */
(FRAME_SIZE & 0x7f) | 0x80, /* ... uleb128 FRAME_SIZE */
(FRAME_SIZE >> 7)
},
.fde_reg_ofs = {
0x88, 1, /* DW_CFA_offset, %eip, -4 */
/* The following ordering must match tcg_target_callee_save_regs. */
0x85, 2, /* DW_CFA_offset, %ebp, -8 */
0x83, 3, /* DW_CFA_offset, %ebx, -12 */
0x86, 4, /* DW_CFA_offset, %esi, -16 */
0x87, 5, /* DW_CFA_offset, %edi, -20 */
}
};
#endif
#if defined(ELF_HOST_MACHINE)
void tcg_register_jit(const void *buf, size_t buf_size)
{
tcg_register_jit_int(buf, buf_size, &debug_frame, sizeof(debug_frame));
}
#endif
|