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
|
/*
* Copyright 2015 Intel Corporation
*
* 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 (including the next
* paragraph) 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 <assert.h>
#include <stdarg.h>
#include <stdio.h>
#include "genxml/genX_bits.h"
#include "isl.h"
#include "isl_gen4.h"
#include "isl_gen6.h"
#include "isl_gen7.h"
#include "isl_gen8.h"
#include "isl_gen9.h"
#include "isl_priv.h"
void PRINTFLIKE(3, 4) UNUSED
__isl_finishme(const char *file, int line, const char *fmt, ...)
{
va_list ap;
char buf[512];
va_start(ap, fmt);
vsnprintf(buf, sizeof(buf), fmt, ap);
va_end(ap);
fprintf(stderr, "%s:%d: FINISHME: %s\n", file, line, buf);
}
void
isl_device_init(struct isl_device *dev,
const struct gen_device_info *info,
bool has_bit6_swizzling)
{
dev->info = info;
dev->use_separate_stencil = ISL_DEV_GEN(dev) >= 6;
dev->has_bit6_swizzling = has_bit6_swizzling;
/* The ISL_DEV macros may be defined in the CFLAGS, thus hardcoding some
* device properties at buildtime. Verify that the macros with the device
* properties chosen during runtime.
*/
ISL_DEV_GEN_SANITIZE(dev);
ISL_DEV_USE_SEPARATE_STENCIL_SANITIZE(dev);
/* Did we break hiz or stencil? */
if (ISL_DEV_USE_SEPARATE_STENCIL(dev))
assert(info->has_hiz_and_separate_stencil);
if (info->must_use_separate_stencil)
assert(ISL_DEV_USE_SEPARATE_STENCIL(dev));
dev->ss.size = RENDER_SURFACE_STATE_length(info) * 4;
dev->ss.align = isl_align(dev->ss.size, 32);
assert(RENDER_SURFACE_STATE_SurfaceBaseAddress_start(info) % 8 == 0);
dev->ss.addr_offset =
RENDER_SURFACE_STATE_SurfaceBaseAddress_start(info) / 8;
/* The "Auxiliary Surface Base Address" field starts a bit higher up
* because the bottom 12 bits are used for other things. Round down to
* the nearest dword before.
*/
dev->ss.aux_addr_offset =
(RENDER_SURFACE_STATE_AuxiliarySurfaceBaseAddress_start(info) & ~31) / 8;
}
/**
* @brief Query the set of multisamples supported by the device.
*
* This function always returns non-zero, as ISL_SAMPLE_COUNT_1_BIT is always
* supported.
*/
isl_sample_count_mask_t ATTRIBUTE_CONST
isl_device_get_sample_counts(struct isl_device *dev)
{
if (ISL_DEV_GEN(dev) >= 9) {
return ISL_SAMPLE_COUNT_1_BIT |
ISL_SAMPLE_COUNT_2_BIT |
ISL_SAMPLE_COUNT_4_BIT |
ISL_SAMPLE_COUNT_8_BIT |
ISL_SAMPLE_COUNT_16_BIT;
} else if (ISL_DEV_GEN(dev) >= 8) {
return ISL_SAMPLE_COUNT_1_BIT |
ISL_SAMPLE_COUNT_2_BIT |
ISL_SAMPLE_COUNT_4_BIT |
ISL_SAMPLE_COUNT_8_BIT;
} else if (ISL_DEV_GEN(dev) >= 7) {
return ISL_SAMPLE_COUNT_1_BIT |
ISL_SAMPLE_COUNT_4_BIT |
ISL_SAMPLE_COUNT_8_BIT;
} else if (ISL_DEV_GEN(dev) >= 6) {
return ISL_SAMPLE_COUNT_1_BIT |
ISL_SAMPLE_COUNT_4_BIT;
} else {
return ISL_SAMPLE_COUNT_1_BIT;
}
}
/**
* @param[out] info is written only on success
*/
static bool
isl_tiling_get_info(const struct isl_device *dev,
enum isl_tiling tiling,
uint32_t format_bpb,
struct isl_tile_info *tile_info)
{
const uint32_t bs = format_bpb / 8;
struct isl_extent2d logical_el, phys_B;
if (tiling != ISL_TILING_LINEAR && !isl_is_pow2(format_bpb)) {
/* It is possible to have non-power-of-two formats in a tiled buffer.
* The easiest way to handle this is to treat the tile as if it is three
* times as wide. This way no pixel will ever cross a tile boundary.
* This really only works on legacy X and Y tiling formats.
*/
assert(tiling == ISL_TILING_X || tiling == ISL_TILING_Y0);
assert(bs % 3 == 0 && isl_is_pow2(format_bpb / 3));
return isl_tiling_get_info(dev, tiling, format_bpb / 3, tile_info);
}
switch (tiling) {
case ISL_TILING_LINEAR:
assert(bs > 0);
logical_el = isl_extent2d(1, 1);
phys_B = isl_extent2d(bs, 1);
break;
case ISL_TILING_X:
assert(bs > 0);
logical_el = isl_extent2d(512 / bs, 8);
phys_B = isl_extent2d(512, 8);
break;
case ISL_TILING_Y0:
assert(bs > 0);
logical_el = isl_extent2d(128 / bs, 32);
phys_B = isl_extent2d(128, 32);
break;
case ISL_TILING_W:
assert(bs == 1);
logical_el = isl_extent2d(64, 64);
/* From the Broadwell PRM Vol 2d, RENDER_SURFACE_STATE::SurfacePitch:
*
* "If the surface is a stencil buffer (and thus has Tile Mode set
* to TILEMODE_WMAJOR), the pitch must be set to 2x the value
* computed based on width, as the stencil buffer is stored with two
* rows interleaved."
*
* This, together with the fact that stencil buffers are referred to as
* being Y-tiled in the PRMs for older hardware implies that the
* physical size of a W-tile is actually the same as for a Y-tile.
*/
phys_B = isl_extent2d(128, 32);
break;
case ISL_TILING_Yf:
case ISL_TILING_Ys: {
if (ISL_DEV_GEN(dev) < 9)
return false;
if (!isl_is_pow2(bs))
return false;
bool is_Ys = tiling == ISL_TILING_Ys;
assert(bs > 0);
unsigned width = 1 << (6 + (ffs(bs) / 2) + (2 * is_Ys));
unsigned height = 1 << (6 - (ffs(bs) / 2) + (2 * is_Ys));
logical_el = isl_extent2d(width / bs, height);
phys_B = isl_extent2d(width, height);
break;
}
case ISL_TILING_HIZ:
/* HiZ buffers are required to have ISL_FORMAT_HIZ which is an 8x4
* 128bpb format. The tiling has the same physical dimensions as
* Y-tiling but actually has two HiZ columns per Y-tiled column.
*/
assert(bs == 16);
logical_el = isl_extent2d(16, 16);
phys_B = isl_extent2d(128, 32);
break;
case ISL_TILING_CCS:
/* CCS surfaces are required to have one of the GENX_CCS_* formats which
* have a block size of 1 or 2 bits per block and each CCS element
* corresponds to one cache-line pair in the main surface. From the Sky
* Lake PRM Vol. 12 in the section on planes:
*
* "The Color Control Surface (CCS) contains the compression status
* of the cache-line pairs. The compression state of the cache-line
* pair is specified by 2 bits in the CCS. Each CCS cache-line
* represents an area on the main surface of 16x16 sets of 128 byte
* Y-tiled cache-line-pairs. CCS is always Y tiled."
*
* The CCS being Y-tiled implies that it's an 8x8 grid of cache-lines.
* Since each cache line corresponds to a 16x16 set of cache-line pairs,
* that yields total tile area of 128x128 cache-line pairs or CCS
* elements. On older hardware, each CCS element is 1 bit and the tile
* is 128x256 elements.
*/
assert(format_bpb == 1 || format_bpb == 2);
logical_el = isl_extent2d(128, 256 / format_bpb);
phys_B = isl_extent2d(128, 32);
break;
default:
unreachable("not reached");
} /* end switch */
*tile_info = (struct isl_tile_info) {
.tiling = tiling,
.format_bpb = format_bpb,
.logical_extent_el = logical_el,
.phys_extent_B = phys_B,
};
return true;
}
/**
* @param[out] tiling is set only on success
*/
static bool
isl_surf_choose_tiling(const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
enum isl_tiling *tiling)
{
isl_tiling_flags_t tiling_flags = info->tiling_flags;
/* HiZ surfaces always use the HiZ tiling */
if (info->usage & ISL_SURF_USAGE_HIZ_BIT) {
assert(info->format == ISL_FORMAT_HIZ);
assert(tiling_flags == ISL_TILING_HIZ_BIT);
*tiling = ISL_TILING_HIZ;
return true;
}
/* CCS surfaces always use the CCS tiling */
if (info->usage & ISL_SURF_USAGE_CCS_BIT) {
assert(isl_format_get_layout(info->format)->txc == ISL_TXC_CCS);
assert(tiling_flags == ISL_TILING_CCS_BIT);
*tiling = ISL_TILING_CCS;
return true;
}
if (ISL_DEV_GEN(dev) >= 6) {
isl_gen6_filter_tiling(dev, info, &tiling_flags);
} else {
isl_finishme("%s: gen%u", __func__, ISL_DEV_GEN(dev));
isl_gen6_filter_tiling(dev, info, &tiling_flags);
}
#define CHOOSE(__tiling) \
do { \
if (tiling_flags & (1u << (__tiling))) { \
*tiling = (__tiling); \
return true; \
} \
} while (0)
/* Of the tiling modes remaining, choose the one that offers the best
* performance.
*/
if (info->dim == ISL_SURF_DIM_1D) {
/* Prefer linear for 1D surfaces because they do not benefit from
* tiling. To the contrary, tiling leads to wasted memory and poor
* memory locality due to the swizzling and alignment restrictions
* required in tiled surfaces.
*/
CHOOSE(ISL_TILING_LINEAR);
}
CHOOSE(ISL_TILING_Ys);
CHOOSE(ISL_TILING_Yf);
CHOOSE(ISL_TILING_Y0);
CHOOSE(ISL_TILING_X);
CHOOSE(ISL_TILING_W);
CHOOSE(ISL_TILING_LINEAR);
#undef CHOOSE
/* No tiling mode accomodates the inputs. */
return false;
}
static bool
isl_choose_msaa_layout(const struct isl_device *dev,
const struct isl_surf_init_info *info,
enum isl_tiling tiling,
enum isl_msaa_layout *msaa_layout)
{
if (ISL_DEV_GEN(dev) >= 8) {
return isl_gen8_choose_msaa_layout(dev, info, tiling, msaa_layout);
} else if (ISL_DEV_GEN(dev) >= 7) {
return isl_gen7_choose_msaa_layout(dev, info, tiling, msaa_layout);
} else if (ISL_DEV_GEN(dev) >= 6) {
return isl_gen6_choose_msaa_layout(dev, info, tiling, msaa_layout);
} else {
return isl_gen4_choose_msaa_layout(dev, info, tiling, msaa_layout);
}
}
struct isl_extent2d
isl_get_interleaved_msaa_px_size_sa(uint32_t samples)
{
assert(isl_is_pow2(samples));
/* From the Broadwell PRM >> Volume 5: Memory Views >> Computing Mip Level
* Sizes (p133):
*
* If the surface is multisampled and it is a depth or stencil surface
* or Multisampled Surface StorageFormat in SURFACE_STATE is
* MSFMT_DEPTH_STENCIL, W_L and H_L must be adjusted as follows before
* proceeding: [...]
*/
return (struct isl_extent2d) {
.width = 1 << ((ffs(samples) - 0) / 2),
.height = 1 << ((ffs(samples) - 1) / 2),
};
}
static void
isl_msaa_interleaved_scale_px_to_sa(uint32_t samples,
uint32_t *width, uint32_t *height)
{
const struct isl_extent2d px_size_sa =
isl_get_interleaved_msaa_px_size_sa(samples);
if (width)
*width = isl_align(*width, 2) * px_size_sa.width;
if (height)
*height = isl_align(*height, 2) * px_size_sa.height;
}
static enum isl_array_pitch_span
isl_choose_array_pitch_span(const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
enum isl_dim_layout dim_layout,
const struct isl_extent4d *phys_level0_sa)
{
switch (dim_layout) {
case ISL_DIM_LAYOUT_GEN9_1D:
case ISL_DIM_LAYOUT_GEN4_2D:
if (ISL_DEV_GEN(dev) >= 8) {
/* QPitch becomes programmable in Broadwell. So choose the
* most compact QPitch possible in order to conserve memory.
*
* From the Broadwell PRM >> Volume 2d: Command Reference: Structures
* >> RENDER_SURFACE_STATE Surface QPitch (p325):
*
* - Software must ensure that this field is set to a value
* sufficiently large such that the array slices in the surface
* do not overlap. Refer to the Memory Data Formats section for
* information on how surfaces are stored in memory.
*
* - This field specifies the distance in rows between array
* slices. It is used only in the following cases:
*
* - Surface Array is enabled OR
* - Number of Mulitsamples is not NUMSAMPLES_1 and
* Multisampled Surface Storage Format set to MSFMT_MSS OR
* - Surface Type is SURFTYPE_CUBE
*/
return ISL_ARRAY_PITCH_SPAN_COMPACT;
} else if (ISL_DEV_GEN(dev) >= 7) {
/* Note that Ivybridge introduces
* RENDER_SURFACE_STATE.SurfaceArraySpacing, which provides the
* driver more control over the QPitch.
*/
if (phys_level0_sa->array_len == 1) {
/* The hardware will never use the QPitch. So choose the most
* compact QPitch possible in order to conserve memory.
*/
return ISL_ARRAY_PITCH_SPAN_COMPACT;
}
if (isl_surf_usage_is_depth_or_stencil(info->usage) ||
(info->usage & ISL_SURF_USAGE_HIZ_BIT)) {
/* From the Ivybridge PRM >> Volume 1 Part 1: Graphics Core >>
* Section 6.18.4.7: Surface Arrays (p112):
*
* If Surface Array Spacing is set to ARYSPC_FULL (note that
* the depth buffer and stencil buffer have an implied value of
* ARYSPC_FULL):
*/
return ISL_ARRAY_PITCH_SPAN_FULL;
}
if (info->levels == 1) {
/* We are able to set RENDER_SURFACE_STATE.SurfaceArraySpacing
* to ARYSPC_LOD0.
*/
return ISL_ARRAY_PITCH_SPAN_COMPACT;
}
return ISL_ARRAY_PITCH_SPAN_FULL;
} else if ((ISL_DEV_GEN(dev) == 5 || ISL_DEV_GEN(dev) == 6) &&
ISL_DEV_USE_SEPARATE_STENCIL(dev) &&
isl_surf_usage_is_stencil(info->usage)) {
/* [ILK-SNB] Errata from the Sandy Bridge PRM >> Volume 4 Part 1:
* Graphics Core >> Section 7.18.3.7: Surface Arrays:
*
* The separate stencil buffer does not support mip mapping, thus
* the storage for LODs other than LOD 0 is not needed.
*/
assert(info->levels == 1);
assert(phys_level0_sa->array_len == 1);
return ISL_ARRAY_PITCH_SPAN_COMPACT;
} else {
if ((ISL_DEV_GEN(dev) == 5 || ISL_DEV_GEN(dev) == 6) &&
ISL_DEV_USE_SEPARATE_STENCIL(dev) &&
isl_surf_usage_is_stencil(info->usage)) {
/* [ILK-SNB] Errata from the Sandy Bridge PRM >> Volume 4 Part 1:
* Graphics Core >> Section 7.18.3.7: Surface Arrays:
*
* The separate stencil buffer does not support mip mapping,
* thus the storage for LODs other than LOD 0 is not needed.
*/
assert(info->levels == 1);
assert(phys_level0_sa->array_len == 1);
return ISL_ARRAY_PITCH_SPAN_COMPACT;
}
if (phys_level0_sa->array_len == 1) {
/* The hardware will never use the QPitch. So choose the most
* compact QPitch possible in order to conserve memory.
*/
return ISL_ARRAY_PITCH_SPAN_COMPACT;
}
return ISL_ARRAY_PITCH_SPAN_FULL;
}
case ISL_DIM_LAYOUT_GEN4_3D:
/* The hardware will never use the QPitch. So choose the most
* compact QPitch possible in order to conserve memory.
*/
return ISL_ARRAY_PITCH_SPAN_COMPACT;
}
unreachable("bad isl_dim_layout");
return ISL_ARRAY_PITCH_SPAN_FULL;
}
static void
isl_choose_image_alignment_el(const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
enum isl_tiling tiling,
enum isl_dim_layout dim_layout,
enum isl_msaa_layout msaa_layout,
struct isl_extent3d *image_align_el)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
if (fmtl->txc == ISL_TXC_MCS) {
assert(tiling == ISL_TILING_Y0);
/*
* IvyBrigde PRM Vol 2, Part 1, "11.7 MCS Buffer for Render Target(s)":
*
* Height, width, and layout of MCS buffer in this case must match with
* Render Target height, width, and layout. MCS buffer is tiledY.
*
* To avoid wasting memory, choose the smallest alignment possible:
* HALIGN_4 and VALIGN_4.
*/
*image_align_el = isl_extent3d(4, 4, 1);
return;
} else if (info->format == ISL_FORMAT_HIZ) {
assert(ISL_DEV_GEN(dev) >= 6);
/* HiZ surfaces are always aligned to 16x8 pixels in the primary surface
* which works out to 2x2 HiZ elments.
*/
*image_align_el = isl_extent3d(2, 2, 1);
return;
}
if (ISL_DEV_GEN(dev) >= 9) {
isl_gen9_choose_image_alignment_el(dev, info, tiling, dim_layout,
msaa_layout, image_align_el);
} else if (ISL_DEV_GEN(dev) >= 8) {
isl_gen8_choose_image_alignment_el(dev, info, tiling, dim_layout,
msaa_layout, image_align_el);
} else if (ISL_DEV_GEN(dev) >= 7) {
isl_gen7_choose_image_alignment_el(dev, info, tiling, dim_layout,
msaa_layout, image_align_el);
} else if (ISL_DEV_GEN(dev) >= 6) {
isl_gen6_choose_image_alignment_el(dev, info, tiling, dim_layout,
msaa_layout, image_align_el);
} else {
isl_gen4_choose_image_alignment_el(dev, info, tiling, dim_layout,
msaa_layout, image_align_el);
}
}
static enum isl_dim_layout
isl_surf_choose_dim_layout(const struct isl_device *dev,
enum isl_surf_dim logical_dim,
enum isl_tiling tiling)
{
if (ISL_DEV_GEN(dev) >= 9) {
switch (logical_dim) {
case ISL_SURF_DIM_1D:
/* From the Sky Lake PRM Vol. 5, "1D Surfaces":
*
* One-dimensional surfaces use a tiling mode of linear.
* Technically, they are not tiled resources, but the Tiled
* Resource Mode field in RENDER_SURFACE_STATE is still used to
* indicate the alignment requirements for this linear surface
* (See 1D Alignment requirements for how 4K and 64KB Tiled
* Resource Modes impact alignment). Alternatively, a 1D surface
* can be defined as a 2D tiled surface (e.g. TileY or TileX) with
* a height of 0.
*
* In other words, ISL_DIM_LAYOUT_GEN9_1D is only used for linear
* surfaces and, for tiled surfaces, ISL_DIM_LAYOUT_GEN4_2D is used.
*/
if (tiling == ISL_TILING_LINEAR)
return ISL_DIM_LAYOUT_GEN9_1D;
else
return ISL_DIM_LAYOUT_GEN4_2D;
case ISL_SURF_DIM_2D:
case ISL_SURF_DIM_3D:
return ISL_DIM_LAYOUT_GEN4_2D;
}
} else {
switch (logical_dim) {
case ISL_SURF_DIM_1D:
case ISL_SURF_DIM_2D:
return ISL_DIM_LAYOUT_GEN4_2D;
case ISL_SURF_DIM_3D:
return ISL_DIM_LAYOUT_GEN4_3D;
}
}
unreachable("bad isl_surf_dim");
return ISL_DIM_LAYOUT_GEN4_2D;
}
/**
* Calculate the physical extent of the surface's first level, in units of
* surface samples. The result is aligned to the format's compression block.
*/
static void
isl_calc_phys_level0_extent_sa(const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
enum isl_dim_layout dim_layout,
enum isl_tiling tiling,
enum isl_msaa_layout msaa_layout,
struct isl_extent4d *phys_level0_sa)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
if (isl_format_is_yuv(info->format))
isl_finishme("%s:%s: YUV format", __FILE__, __func__);
switch (info->dim) {
case ISL_SURF_DIM_1D:
assert(info->height == 1);
assert(info->depth == 1);
assert(info->samples == 1);
switch (dim_layout) {
case ISL_DIM_LAYOUT_GEN4_3D:
unreachable("bad isl_dim_layout");
case ISL_DIM_LAYOUT_GEN9_1D:
case ISL_DIM_LAYOUT_GEN4_2D:
*phys_level0_sa = (struct isl_extent4d) {
.w = isl_align_npot(info->width, fmtl->bw),
.h = fmtl->bh,
.d = 1,
.a = info->array_len,
};
break;
}
break;
case ISL_SURF_DIM_2D:
assert(dim_layout == ISL_DIM_LAYOUT_GEN4_2D);
if (tiling == ISL_TILING_Ys && info->samples > 1)
isl_finishme("%s:%s: multisample TileYs layout", __FILE__, __func__);
switch (msaa_layout) {
case ISL_MSAA_LAYOUT_NONE:
assert(info->depth == 1);
assert(info->samples == 1);
*phys_level0_sa = (struct isl_extent4d) {
.w = isl_align_npot(info->width, fmtl->bw),
.h = isl_align_npot(info->height, fmtl->bh),
.d = 1,
.a = info->array_len,
};
break;
case ISL_MSAA_LAYOUT_ARRAY:
assert(info->depth == 1);
assert(info->levels == 1);
assert(isl_format_supports_multisampling(dev->info, info->format));
assert(fmtl->bw == 1 && fmtl->bh == 1);
*phys_level0_sa = (struct isl_extent4d) {
.w = info->width,
.h = info->height,
.d = 1,
.a = info->array_len * info->samples,
};
break;
case ISL_MSAA_LAYOUT_INTERLEAVED:
assert(info->depth == 1);
assert(info->levels == 1);
assert(isl_format_supports_multisampling(dev->info, info->format));
*phys_level0_sa = (struct isl_extent4d) {
.w = info->width,
.h = info->height,
.d = 1,
.a = info->array_len,
};
isl_msaa_interleaved_scale_px_to_sa(info->samples,
&phys_level0_sa->w,
&phys_level0_sa->h);
phys_level0_sa->w = isl_align(phys_level0_sa->w, fmtl->bw);
phys_level0_sa->h = isl_align(phys_level0_sa->h, fmtl->bh);
break;
}
break;
case ISL_SURF_DIM_3D:
assert(info->array_len == 1);
assert(info->samples == 1);
if (fmtl->bd > 1) {
isl_finishme("%s:%s: compression block with depth > 1",
__FILE__, __func__);
}
switch (dim_layout) {
case ISL_DIM_LAYOUT_GEN9_1D:
unreachable("bad isl_dim_layout");
case ISL_DIM_LAYOUT_GEN4_2D:
assert(ISL_DEV_GEN(dev) >= 9);
*phys_level0_sa = (struct isl_extent4d) {
.w = isl_align_npot(info->width, fmtl->bw),
.h = isl_align_npot(info->height, fmtl->bh),
.d = 1,
.a = info->depth,
};
break;
case ISL_DIM_LAYOUT_GEN4_3D:
assert(ISL_DEV_GEN(dev) < 9);
*phys_level0_sa = (struct isl_extent4d) {
.w = isl_align(info->width, fmtl->bw),
.h = isl_align(info->height, fmtl->bh),
.d = info->depth,
.a = 1,
};
break;
}
break;
}
}
/**
* A variant of isl_calc_phys_slice0_extent_sa() specific to
* ISL_DIM_LAYOUT_GEN4_2D.
*/
static void
isl_calc_phys_slice0_extent_sa_gen4_2d(
const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
enum isl_msaa_layout msaa_layout,
const struct isl_extent3d *image_align_sa,
const struct isl_extent4d *phys_level0_sa,
struct isl_extent2d *phys_slice0_sa)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
assert(phys_level0_sa->depth == 1);
if (info->levels == 1) {
/* Do not pad the surface to the image alignment. Instead, pad it only
* to the pixel format's block alignment.
*
* For tiled surfaces, using a reduced alignment here avoids wasting CPU
* cycles on the below mipmap layout caluclations. Reducing the
* alignment here is safe because we later align the row pitch and array
* pitch to the tile boundary. It is safe even for
* ISL_MSAA_LAYOUT_INTERLEAVED, because phys_level0_sa is already scaled
* to accomodate the interleaved samples.
*
* For linear surfaces, reducing the alignment here permits us to later
* choose an arbitrary, non-aligned row pitch. If the surface backs
* a VkBuffer, then an arbitrary pitch may be needed to accomodate
* VkBufferImageCopy::bufferRowLength.
*/
*phys_slice0_sa = (struct isl_extent2d) {
.w = isl_align_npot(phys_level0_sa->w, fmtl->bw),
.h = isl_align_npot(phys_level0_sa->h, fmtl->bh),
};
return;
}
uint32_t slice_top_w = 0;
uint32_t slice_bottom_w = 0;
uint32_t slice_left_h = 0;
uint32_t slice_right_h = 0;
uint32_t W0 = phys_level0_sa->w;
uint32_t H0 = phys_level0_sa->h;
for (uint32_t l = 0; l < info->levels; ++l) {
uint32_t W = isl_minify(W0, l);
uint32_t H = isl_minify(H0, l);
uint32_t w = isl_align_npot(W, image_align_sa->w);
uint32_t h = isl_align_npot(H, image_align_sa->h);
if (l == 0) {
slice_top_w = w;
slice_left_h = h;
slice_right_h = h;
} else if (l == 1) {
slice_bottom_w = w;
slice_left_h += h;
} else if (l == 2) {
slice_bottom_w += w;
slice_right_h += h;
} else {
slice_right_h += h;
}
}
*phys_slice0_sa = (struct isl_extent2d) {
.w = MAX(slice_top_w, slice_bottom_w),
.h = MAX(slice_left_h, slice_right_h),
};
}
/**
* A variant of isl_calc_phys_slice0_extent_sa() specific to
* ISL_DIM_LAYOUT_GEN4_3D.
*/
static void
isl_calc_phys_slice0_extent_sa_gen4_3d(
const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
const struct isl_extent3d *image_align_sa,
const struct isl_extent4d *phys_level0_sa,
struct isl_extent2d *phys_slice0_sa)
{
assert(info->samples == 1);
assert(phys_level0_sa->array_len == 1);
uint32_t slice_w = 0;
uint32_t slice_h = 0;
uint32_t W0 = phys_level0_sa->w;
uint32_t H0 = phys_level0_sa->h;
uint32_t D0 = phys_level0_sa->d;
for (uint32_t l = 0; l < info->levels; ++l) {
uint32_t level_w = isl_align_npot(isl_minify(W0, l), image_align_sa->w);
uint32_t level_h = isl_align_npot(isl_minify(H0, l), image_align_sa->h);
uint32_t level_d = isl_align_npot(isl_minify(D0, l), image_align_sa->d);
uint32_t max_layers_horiz = MIN(level_d, 1u << l);
uint32_t max_layers_vert = isl_align(level_d, 1u << l) / (1u << l);
slice_w = MAX(slice_w, level_w * max_layers_horiz);
slice_h += level_h * max_layers_vert;
}
*phys_slice0_sa = (struct isl_extent2d) {
.w = slice_w,
.h = slice_h,
};
}
/**
* A variant of isl_calc_phys_slice0_extent_sa() specific to
* ISL_DIM_LAYOUT_GEN9_1D.
*/
static void
isl_calc_phys_slice0_extent_sa_gen9_1d(
const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
const struct isl_extent3d *image_align_sa,
const struct isl_extent4d *phys_level0_sa,
struct isl_extent2d *phys_slice0_sa)
{
MAYBE_UNUSED const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
assert(phys_level0_sa->height == 1);
assert(phys_level0_sa->depth == 1);
assert(info->samples == 1);
assert(image_align_sa->w >= fmtl->bw);
uint32_t slice_w = 0;
const uint32_t W0 = phys_level0_sa->w;
for (uint32_t l = 0; l < info->levels; ++l) {
uint32_t W = isl_minify(W0, l);
uint32_t w = isl_align_npot(W, image_align_sa->w);
slice_w += w;
}
*phys_slice0_sa = isl_extent2d(slice_w, 1);
}
/**
* Calculate the physical extent of the surface's first array slice, in units
* of surface samples. If the surface is multi-leveled, then the result will
* be aligned to \a image_align_sa.
*/
static void
isl_calc_phys_slice0_extent_sa(const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
enum isl_dim_layout dim_layout,
enum isl_msaa_layout msaa_layout,
const struct isl_extent3d *image_align_sa,
const struct isl_extent4d *phys_level0_sa,
struct isl_extent2d *phys_slice0_sa)
{
switch (dim_layout) {
case ISL_DIM_LAYOUT_GEN9_1D:
isl_calc_phys_slice0_extent_sa_gen9_1d(dev, info,
image_align_sa, phys_level0_sa,
phys_slice0_sa);
return;
case ISL_DIM_LAYOUT_GEN4_2D:
isl_calc_phys_slice0_extent_sa_gen4_2d(dev, info, msaa_layout,
image_align_sa, phys_level0_sa,
phys_slice0_sa);
return;
case ISL_DIM_LAYOUT_GEN4_3D:
isl_calc_phys_slice0_extent_sa_gen4_3d(dev, info, image_align_sa,
phys_level0_sa, phys_slice0_sa);
return;
}
}
/**
* Calculate the pitch between physical array slices, in units of rows of
* surface elements.
*/
static uint32_t
isl_calc_array_pitch_el_rows(const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
const struct isl_tile_info *tile_info,
enum isl_dim_layout dim_layout,
enum isl_array_pitch_span array_pitch_span,
const struct isl_extent3d *image_align_sa,
const struct isl_extent4d *phys_level0_sa,
const struct isl_extent2d *phys_slice0_sa)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
uint32_t pitch_sa_rows = 0;
switch (dim_layout) {
case ISL_DIM_LAYOUT_GEN9_1D:
/* Each row is an array slice */
pitch_sa_rows = 1;
break;
case ISL_DIM_LAYOUT_GEN4_2D:
switch (array_pitch_span) {
case ISL_ARRAY_PITCH_SPAN_COMPACT:
pitch_sa_rows = isl_align_npot(phys_slice0_sa->h, image_align_sa->h);
break;
case ISL_ARRAY_PITCH_SPAN_FULL: {
/* The QPitch equation is found in the Broadwell PRM >> Volume 5:
* Memory Views >> Common Surface Formats >> Surface Layout >> 2D
* Surfaces >> Surface Arrays.
*/
uint32_t H0_sa = phys_level0_sa->h;
uint32_t H1_sa = isl_minify(H0_sa, 1);
uint32_t h0_sa = isl_align_npot(H0_sa, image_align_sa->h);
uint32_t h1_sa = isl_align_npot(H1_sa, image_align_sa->h);
uint32_t m;
if (ISL_DEV_GEN(dev) >= 7) {
/* The QPitch equation changed slightly in Ivybridge. */
m = 12;
} else {
m = 11;
}
pitch_sa_rows = h0_sa + h1_sa + (m * image_align_sa->h);
if (ISL_DEV_GEN(dev) == 6 && info->samples > 1 &&
(info->height % 4 == 1)) {
/* [SNB] Errata from the Sandy Bridge PRM >> Volume 4 Part 1:
* Graphics Core >> Section 7.18.3.7: Surface Arrays:
*
* [SNB] Errata: Sampler MSAA Qpitch will be 4 greater than
* the value calculated in the equation above , for every
* other odd Surface Height starting from 1 i.e. 1,5,9,13.
*
* XXX(chadv): Is the errata natural corollary of the physical
* layout of interleaved samples?
*/
pitch_sa_rows += 4;
}
pitch_sa_rows = isl_align_npot(pitch_sa_rows, fmtl->bh);
} /* end case */
break;
}
break;
case ISL_DIM_LAYOUT_GEN4_3D:
assert(array_pitch_span == ISL_ARRAY_PITCH_SPAN_COMPACT);
pitch_sa_rows = isl_align_npot(phys_slice0_sa->h, image_align_sa->h);
break;
default:
unreachable("bad isl_dim_layout");
break;
}
assert(pitch_sa_rows % fmtl->bh == 0);
uint32_t pitch_el_rows = pitch_sa_rows / fmtl->bh;
if (ISL_DEV_GEN(dev) >= 9 && fmtl->txc == ISL_TXC_CCS) {
/*
* From the Sky Lake PRM Vol 7, "MCS Buffer for Render Target(s)" (p. 632):
*
* "Mip-mapped and arrayed surfaces are supported with MCS buffer
* layout with these alignments in the RT space: Horizontal
* Alignment = 128 and Vertical Alignment = 64."
*
* From the Sky Lake PRM Vol. 2d, "RENDER_SURFACE_STATE" (p. 435):
*
* "For non-multisampled render target's CCS auxiliary surface,
* QPitch must be computed with Horizontal Alignment = 128 and
* Surface Vertical Alignment = 256. These alignments are only for
* CCS buffer and not for associated render target."
*
* The first restriction is already handled by isl_choose_image_alignment_el
* but the second restriction, which is an extension of the first, only
* applies to qpitch and must be applied here.
*/
assert(fmtl->bh == 4);
pitch_el_rows = isl_align(pitch_el_rows, 256 / 4);
}
if (ISL_DEV_GEN(dev) >= 9 &&
info->dim == ISL_SURF_DIM_3D &&
tile_info->tiling != ISL_TILING_LINEAR) {
/* From the Skylake BSpec >> RENDER_SURFACE_STATE >> Surface QPitch:
*
* Tile Mode != Linear: This field must be set to an integer multiple
* of the tile height
*/
pitch_el_rows = isl_align(pitch_el_rows, tile_info->logical_extent_el.height);
}
return pitch_el_rows;
}
static uint32_t
isl_calc_row_pitch_alignment(const struct isl_surf_init_info *surf_info,
const struct isl_tile_info *tile_info)
{
if (tile_info->tiling != ISL_TILING_LINEAR)
return tile_info->phys_extent_B.width;
/* From the Broadwel PRM >> Volume 2d: Command Reference: Structures >>
* RENDER_SURFACE_STATE Surface Pitch (p349):
*
* - For linear render target surfaces and surfaces accessed with the
* typed data port messages, the pitch must be a multiple of the
* element size for non-YUV surface formats. Pitch must be
* a multiple of 2 * element size for YUV surface formats.
*
* - [Requirements for SURFTYPE_BUFFER and SURFTYPE_STRBUF, which we
* ignore because isl doesn't do buffers.]
*
* - For other linear surfaces, the pitch can be any multiple of
* bytes.
*/
const struct isl_format_layout *fmtl = isl_format_get_layout(surf_info->format);
const uint32_t bs = fmtl->bpb / 8;
if (surf_info->usage & ISL_SURF_USAGE_RENDER_TARGET_BIT) {
if (isl_format_is_yuv(surf_info->format)) {
return 2 * bs;
} else {
return bs;
}
}
return 1;
}
static uint32_t
isl_calc_linear_min_row_pitch(const struct isl_device *dev,
const struct isl_surf_init_info *info,
const struct isl_extent2d *phys_slice0_sa,
uint32_t alignment)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
const uint32_t bs = fmtl->bpb / 8;
assert(phys_slice0_sa->w % fmtl->bw == 0);
return isl_align_npot(bs * (phys_slice0_sa->w / fmtl->bw), alignment);
}
static uint32_t
isl_calc_tiled_min_row_pitch(const struct isl_device *dev,
const struct isl_surf_init_info *surf_info,
const struct isl_tile_info *tile_info,
const struct isl_extent2d *phys_slice0_sa,
uint32_t alignment)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(surf_info->format);
assert(fmtl->bpb % tile_info->format_bpb == 0);
assert(phys_slice0_sa->w % fmtl->bw == 0);
const uint32_t tile_el_scale = fmtl->bpb / tile_info->format_bpb;
const uint32_t total_w_el = phys_slice0_sa->width / fmtl->bw;
const uint32_t total_w_tl =
isl_align_div(total_w_el * tile_el_scale,
tile_info->logical_extent_el.width);
assert(alignment == tile_info->phys_extent_B.width);
return total_w_tl * tile_info->phys_extent_B.width;
}
static uint32_t
isl_calc_min_row_pitch(const struct isl_device *dev,
const struct isl_surf_init_info *surf_info,
const struct isl_tile_info *tile_info,
const struct isl_extent2d *phys_slice0_sa,
uint32_t alignment)
{
if (tile_info->tiling == ISL_TILING_LINEAR) {
return isl_calc_linear_min_row_pitch(dev, surf_info, phys_slice0_sa,
alignment);
} else {
return isl_calc_tiled_min_row_pitch(dev, surf_info, tile_info,
phys_slice0_sa, alignment);
}
}
/**
* Is `pitch` in the valid range for a hardware bitfield, if the bitfield's
* size is `bits` bits?
*
* Hardware pitch fields are offset by 1. For example, if the size of
* RENDER_SURFACE_STATE::SurfacePitch is B bits, then the range of valid
* pitches is [1, 2^b] inclusive. If the surface pitch is N, then
* RENDER_SURFACE_STATE::SurfacePitch must be set to N-1.
*/
static bool
pitch_in_range(uint32_t n, uint32_t bits)
{
assert(n != 0);
return likely(bits != 0 && 1 <= n && n <= (1 << bits));
}
static bool
isl_calc_row_pitch(const struct isl_device *dev,
const struct isl_surf_init_info *surf_info,
const struct isl_tile_info *tile_info,
enum isl_dim_layout dim_layout,
const struct isl_extent2d *phys_slice0_sa,
uint32_t *out_row_pitch)
{
const uint32_t alignment =
isl_calc_row_pitch_alignment(surf_info, tile_info);
const uint32_t min_row_pitch =
isl_calc_min_row_pitch(dev, surf_info, tile_info, phys_slice0_sa,
alignment);
uint32_t row_pitch = min_row_pitch;
if (surf_info->row_pitch != 0) {
row_pitch = surf_info->row_pitch;
if (row_pitch < min_row_pitch)
return false;
if (row_pitch % alignment != 0)
return false;
}
const uint32_t row_pitch_tiles = row_pitch / tile_info->phys_extent_B.width;
if (row_pitch == 0)
return false;
if (dim_layout == ISL_DIM_LAYOUT_GEN9_1D) {
/* SurfacePitch is ignored for this layout. */
goto done;
}
if ((surf_info->usage & (ISL_SURF_USAGE_RENDER_TARGET_BIT |
ISL_SURF_USAGE_TEXTURE_BIT |
ISL_SURF_USAGE_STORAGE_BIT)) &&
!pitch_in_range(row_pitch, RENDER_SURFACE_STATE_SurfacePitch_bits(dev->info)))
return false;
if ((surf_info->usage & (ISL_SURF_USAGE_CCS_BIT |
ISL_SURF_USAGE_MCS_BIT)) &&
!pitch_in_range(row_pitch_tiles, RENDER_SURFACE_STATE_AuxiliarySurfacePitch_bits(dev->info)))
return false;
if ((surf_info->usage & ISL_SURF_USAGE_DEPTH_BIT) &&
!pitch_in_range(row_pitch, _3DSTATE_DEPTH_BUFFER_SurfacePitch_bits(dev->info)))
return false;
if ((surf_info->usage & ISL_SURF_USAGE_HIZ_BIT) &&
!pitch_in_range(row_pitch, _3DSTATE_HIER_DEPTH_BUFFER_SurfacePitch_bits(dev->info)))
return false;
if (surf_info->usage & ISL_SURF_USAGE_STENCIL_BIT)
isl_finishme("validate row pitch of stencil surfaces");
done:
*out_row_pitch = row_pitch;
return true;
}
/**
* Calculate and apply any padding required for the surface.
*
* @param[inout] total_h_el is updated with the new height
* @param[out] pad_bytes is overwritten with additional padding requirements.
*/
static void
isl_apply_surface_padding(const struct isl_device *dev,
const struct isl_surf_init_info *restrict info,
const struct isl_tile_info *tile_info,
uint32_t *total_h_el,
uint32_t *pad_bytes)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
*pad_bytes = 0;
/* From the Broadwell PRM >> Volume 5: Memory Views >> Common Surface
* Formats >> Surface Padding Requirements >> Render Target and Media
* Surfaces:
*
* The data port accesses data (pixels) outside of the surface if they
* are contained in the same cache request as pixels that are within the
* surface. These pixels will not be returned by the requesting message,
* however if these pixels lie outside of defined pages in the GTT,
* a GTT error will result when the cache request is processed. In
* order to avoid these GTT errors, “padding” at the bottom of the
* surface is sometimes necessary.
*
* From the Broadwell PRM >> Volume 5: Memory Views >> Common Surface
* Formats >> Surface Padding Requirements >> Sampling Engine Surfaces:
*
* ... Lots of padding requirements, all listed separately below.
*/
/* We can safely ignore the first padding requirement, quoted below,
* because isl doesn't do buffers.
*
* - [pre-BDW] For buffers, which have no inherent “height,” padding
* requirements are different. A buffer must be padded to the next
* multiple of 256 array elements, with an additional 16 bytes added
* beyond that to account for the L1 cache line.
*/
/*
* - For compressed textures [...], padding at the bottom of the surface
* is to an even compressed row.
*/
if (isl_format_is_compressed(info->format))
*total_h_el = isl_align(*total_h_el, 2);
/*
* - For cube surfaces, an additional two rows of padding are required
* at the bottom of the surface.
*/
if (info->usage & ISL_SURF_USAGE_CUBE_BIT)
*total_h_el += 2;
/*
* - For packed YUV, 96 bpt, 48 bpt, and 24 bpt surface formats,
* additional padding is required. These surfaces require an extra row
* plus 16 bytes of padding at the bottom in addition to the general
* padding requirements.
*/
if (isl_format_is_yuv(info->format) &&
(fmtl->bpb == 96 || fmtl->bpb == 48|| fmtl->bpb == 24)) {
*total_h_el += 1;
*pad_bytes += 16;
}
/*
* - For linear surfaces, additional padding of 64 bytes is required at
* the bottom of the surface. This is in addition to the padding
* required above.
*/
if (tile_info->tiling == ISL_TILING_LINEAR)
*pad_bytes += 64;
/* The below text weakens, not strengthens, the padding requirements for
* linear surfaces. Therefore we can safely ignore it.
*
* - [BDW+] For SURFTYPE_BUFFER, SURFTYPE_1D, and SURFTYPE_2D non-array,
* non-MSAA, non-mip-mapped surfaces in linear memory, the only
* padding requirement is to the next aligned 64-byte boundary beyond
* the end of the surface. The rest of the padding requirements
* documented above do not apply to these surfaces.
*/
/*
* - [SKL+] For SURFTYPE_2D and SURFTYPE_3D with linear mode and
* height % 4 != 0, the surface must be padded with
* 4-(height % 4)*Surface Pitch # of bytes.
*/
if (ISL_DEV_GEN(dev) >= 9 &&
tile_info->tiling == ISL_TILING_LINEAR &&
(info->dim == ISL_SURF_DIM_2D || info->dim == ISL_SURF_DIM_3D)) {
*total_h_el = isl_align(*total_h_el, 4);
}
/*
* - [SKL+] For SURFTYPE_1D with linear mode, the surface must be padded
* to 4 times the Surface Pitch # of bytes
*/
if (ISL_DEV_GEN(dev) >= 9 &&
tile_info->tiling == ISL_TILING_LINEAR &&
info->dim == ISL_SURF_DIM_1D) {
*total_h_el += 4;
}
}
bool
isl_surf_init_s(const struct isl_device *dev,
struct isl_surf *surf,
const struct isl_surf_init_info *restrict info)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(info->format);
const struct isl_extent4d logical_level0_px = {
.w = info->width,
.h = info->height,
.d = info->depth,
.a = info->array_len,
};
enum isl_tiling tiling;
if (!isl_surf_choose_tiling(dev, info, &tiling))
return false;
struct isl_tile_info tile_info;
if (!isl_tiling_get_info(dev, tiling, fmtl->bpb, &tile_info))
return false;
const enum isl_dim_layout dim_layout =
isl_surf_choose_dim_layout(dev, info->dim, tiling);
enum isl_msaa_layout msaa_layout;
if (!isl_choose_msaa_layout(dev, info, tiling, &msaa_layout))
return false;
struct isl_extent3d image_align_el;
isl_choose_image_alignment_el(dev, info, tiling, dim_layout, msaa_layout,
&image_align_el);
struct isl_extent3d image_align_sa =
isl_extent3d_el_to_sa(info->format, image_align_el);
struct isl_extent4d phys_level0_sa;
isl_calc_phys_level0_extent_sa(dev, info, dim_layout, tiling, msaa_layout,
&phys_level0_sa);
assert(phys_level0_sa.w % fmtl->bw == 0);
assert(phys_level0_sa.h % fmtl->bh == 0);
enum isl_array_pitch_span array_pitch_span =
isl_choose_array_pitch_span(dev, info, dim_layout, &phys_level0_sa);
struct isl_extent2d phys_slice0_sa;
isl_calc_phys_slice0_extent_sa(dev, info, dim_layout, msaa_layout,
&image_align_sa, &phys_level0_sa,
&phys_slice0_sa);
assert(phys_slice0_sa.w % fmtl->bw == 0);
assert(phys_slice0_sa.h % fmtl->bh == 0);
const uint32_t array_pitch_el_rows =
isl_calc_array_pitch_el_rows(dev, info, &tile_info, dim_layout,
array_pitch_span, &image_align_sa,
&phys_level0_sa, &phys_slice0_sa);
uint32_t total_h_el = phys_level0_sa.array_len * array_pitch_el_rows;
uint32_t pad_bytes;
isl_apply_surface_padding(dev, info, &tile_info, &total_h_el, &pad_bytes);
uint32_t row_pitch;
if (!isl_calc_row_pitch(dev, info, &tile_info, dim_layout,
&phys_slice0_sa, &row_pitch))
return false;
uint32_t size, base_alignment;
if (tiling == ISL_TILING_LINEAR) {
size = row_pitch * total_h_el + pad_bytes;
/* From the Broadwell PRM Vol 2d, RENDER_SURFACE_STATE::SurfaceBaseAddress:
*
* "The Base Address for linear render target surfaces and surfaces
* accessed with the typed surface read/write data port messages must
* be element-size aligned, for non-YUV surface formats, or a
* multiple of 2 element-sizes for YUV surface formats. Other linear
* surfaces have no alignment requirements (byte alignment is
* sufficient.)"
*/
base_alignment = MAX(1, info->min_alignment);
if (info->usage & ISL_SURF_USAGE_RENDER_TARGET_BIT) {
if (isl_format_is_yuv(info->format)) {
base_alignment = MAX(base_alignment, fmtl->bpb / 4);
} else {
base_alignment = MAX(base_alignment, fmtl->bpb / 8);
}
}
base_alignment = isl_round_up_to_power_of_two(base_alignment);
} else {
total_h_el += isl_align_div_npot(pad_bytes, row_pitch);
const uint32_t total_h_tl =
isl_align_div(total_h_el, tile_info.logical_extent_el.height);
size = total_h_tl * tile_info.phys_extent_B.height * row_pitch;
const uint32_t tile_size = tile_info.phys_extent_B.width *
tile_info.phys_extent_B.height;
assert(isl_is_pow2(info->min_alignment) && isl_is_pow2(tile_size));
base_alignment = MAX(info->min_alignment, tile_size);
}
*surf = (struct isl_surf) {
.dim = info->dim,
.dim_layout = dim_layout,
.msaa_layout = msaa_layout,
.tiling = tiling,
.format = info->format,
.levels = info->levels,
.samples = info->samples,
.image_alignment_el = image_align_el,
.logical_level0_px = logical_level0_px,
.phys_level0_sa = phys_level0_sa,
.size = size,
.alignment = base_alignment,
.row_pitch = row_pitch,
.array_pitch_el_rows = array_pitch_el_rows,
.array_pitch_span = array_pitch_span,
.usage = info->usage,
};
return true;
}
void
isl_surf_get_tile_info(const struct isl_device *dev,
const struct isl_surf *surf,
struct isl_tile_info *tile_info)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(surf->format);
isl_tiling_get_info(dev, surf->tiling, fmtl->bpb, tile_info);
}
bool
isl_surf_get_hiz_surf(const struct isl_device *dev,
const struct isl_surf *surf,
struct isl_surf *hiz_surf)
{
assert(ISL_DEV_GEN(dev) >= 5 && ISL_DEV_USE_SEPARATE_STENCIL(dev));
/* Multisampled depth is always interleaved */
assert(surf->msaa_layout == ISL_MSAA_LAYOUT_NONE ||
surf->msaa_layout == ISL_MSAA_LAYOUT_INTERLEAVED);
/* From the Broadwell PRM Vol. 7, "Hierarchical Depth Buffer":
*
* "The Surface Type, Height, Width, Depth, Minimum Array Element, Render
* Target View Extent, and Depth Coordinate Offset X/Y of the
* hierarchical depth buffer are inherited from the depth buffer. The
* height and width of the hierarchical depth buffer that must be
* allocated are computed by the following formulas, where HZ is the
* hierarchical depth buffer and Z is the depth buffer. The Z_Height,
* Z_Width, and Z_Depth values given in these formulas are those present
* in 3DSTATE_DEPTH_BUFFER incremented by one.
*
* "The value of Z_Height and Z_Width must each be multiplied by 2 before
* being applied to the table below if Number of Multisamples is set to
* NUMSAMPLES_4. The value of Z_Height must be multiplied by 2 and
* Z_Width must be multiplied by 4 before being applied to the table
* below if Number of Multisamples is set to NUMSAMPLES_8."
*
* In the Sky Lake PRM, the second paragraph is replaced with this:
*
* "The Z_Height and Z_Width values must equal those present in
* 3DSTATE_DEPTH_BUFFER incremented by one."
*
* In other words, on Sandy Bridge through Broadwell, each 128-bit HiZ
* block corresponds to a region of 8x4 samples in the primary depth
* surface. On Sky Lake, on the other hand, each HiZ block corresponds to
* a region of 8x4 pixels in the primary depth surface regardless of the
* number of samples. The dimensions of a HiZ block in both pixels and
* samples are given in the table below:
*
* | SNB - BDW | SKL+
* ------+-----------+-------------
* 1x | 8 x 4 sa | 8 x 4 sa
* MSAA | 8 x 4 px | 8 x 4 px
* ------+-----------+-------------
* 2x | 8 x 4 sa | 16 x 4 sa
* MSAA | 4 x 4 px | 8 x 4 px
* ------+-----------+-------------
* 4x | 8 x 4 sa | 16 x 8 sa
* MSAA | 4 x 2 px | 8 x 4 px
* ------+-----------+-------------
* 8x | 8 x 4 sa | 32 x 8 sa
* MSAA | 2 x 2 px | 8 x 4 px
* ------+-----------+-------------
* 16x | N/A | 32 x 16 sa
* MSAA | N/A | 8 x 4 px
* ------+-----------+-------------
*
* There are a number of different ways that this discrepency could be
* handled. The way we have chosen is to simply make MSAA HiZ have the
* same number of samples as the parent surface pre-Sky Lake and always be
* single-sampled on Sky Lake and above. Since the block sizes of
* compressed formats are given in samples, this neatly handles everything
* without the need for additional HiZ formats with different block sizes
* on SKL+.
*/
const unsigned samples = ISL_DEV_GEN(dev) >= 9 ? 1 : surf->samples;
return isl_surf_init(dev, hiz_surf,
.dim = surf->dim,
.format = ISL_FORMAT_HIZ,
.width = surf->logical_level0_px.width,
.height = surf->logical_level0_px.height,
.depth = surf->logical_level0_px.depth,
.levels = surf->levels,
.array_len = surf->logical_level0_px.array_len,
.samples = samples,
.usage = ISL_SURF_USAGE_HIZ_BIT,
.tiling_flags = ISL_TILING_HIZ_BIT);
}
bool
isl_surf_get_mcs_surf(const struct isl_device *dev,
const struct isl_surf *surf,
struct isl_surf *mcs_surf)
{
/* It must be multisampled with an array layout */
assert(surf->samples > 1 && surf->msaa_layout == ISL_MSAA_LAYOUT_ARRAY);
/* The following are true of all multisampled surfaces */
assert(surf->dim == ISL_SURF_DIM_2D);
assert(surf->levels == 1);
assert(surf->logical_level0_px.depth == 1);
/* The "Auxiliary Surface Pitch" field in RENDER_SURFACE_STATE is only 9
* bits which means the maximum pitch of a compression surface is 512
* tiles or 64KB (since MCS is always Y-tiled). Since a 16x MCS buffer is
* 64bpp, this gives us a maximum width of 8192 pixels. We can create
* larger multisampled surfaces, we just can't compress them. For 2x, 4x,
* and 8x, we have enough room for the full 16k supported by the hardware.
*/
if (surf->samples == 16 && surf->logical_level0_px.width > 8192)
return false;
enum isl_format mcs_format;
switch (surf->samples) {
case 2: mcs_format = ISL_FORMAT_MCS_2X; break;
case 4: mcs_format = ISL_FORMAT_MCS_4X; break;
case 8: mcs_format = ISL_FORMAT_MCS_8X; break;
case 16: mcs_format = ISL_FORMAT_MCS_16X; break;
default:
unreachable("Invalid sample count");
}
return isl_surf_init(dev, mcs_surf,
.dim = ISL_SURF_DIM_2D,
.format = mcs_format,
.width = surf->logical_level0_px.width,
.height = surf->logical_level0_px.height,
.depth = 1,
.levels = 1,
.array_len = surf->logical_level0_px.array_len,
.samples = 1, /* MCS surfaces are really single-sampled */
.usage = ISL_SURF_USAGE_MCS_BIT,
.tiling_flags = ISL_TILING_Y0_BIT);
}
bool
isl_surf_get_ccs_surf(const struct isl_device *dev,
const struct isl_surf *surf,
struct isl_surf *ccs_surf)
{
assert(surf->samples == 1 && surf->msaa_layout == ISL_MSAA_LAYOUT_NONE);
assert(ISL_DEV_GEN(dev) >= 7);
if (surf->usage & ISL_SURF_USAGE_DISABLE_AUX_BIT)
return false;
if (ISL_DEV_GEN(dev) <= 8 && surf->dim != ISL_SURF_DIM_2D)
return false;
if (isl_format_is_compressed(surf->format))
return false;
/* TODO: More conditions where it can fail. */
enum isl_format ccs_format;
if (ISL_DEV_GEN(dev) >= 9) {
if (!isl_tiling_is_any_y(surf->tiling))
return false;
switch (isl_format_get_layout(surf->format)->bpb) {
case 32: ccs_format = ISL_FORMAT_GEN9_CCS_32BPP; break;
case 64: ccs_format = ISL_FORMAT_GEN9_CCS_64BPP; break;
case 128: ccs_format = ISL_FORMAT_GEN9_CCS_128BPP; break;
default:
return false;
}
} else if (surf->tiling == ISL_TILING_Y0) {
switch (isl_format_get_layout(surf->format)->bpb) {
case 32: ccs_format = ISL_FORMAT_GEN7_CCS_32BPP_Y; break;
case 64: ccs_format = ISL_FORMAT_GEN7_CCS_64BPP_Y; break;
case 128: ccs_format = ISL_FORMAT_GEN7_CCS_128BPP_Y; break;
default:
return false;
}
} else if (surf->tiling == ISL_TILING_X) {
switch (isl_format_get_layout(surf->format)->bpb) {
case 32: ccs_format = ISL_FORMAT_GEN7_CCS_32BPP_X; break;
case 64: ccs_format = ISL_FORMAT_GEN7_CCS_64BPP_X; break;
case 128: ccs_format = ISL_FORMAT_GEN7_CCS_128BPP_X; break;
default:
return false;
}
} else {
return false;
}
return isl_surf_init(dev, ccs_surf,
.dim = surf->dim,
.format = ccs_format,
.width = surf->logical_level0_px.width,
.height = surf->logical_level0_px.height,
.depth = surf->logical_level0_px.depth,
.levels = surf->levels,
.array_len = surf->logical_level0_px.array_len,
.samples = 1,
.usage = ISL_SURF_USAGE_CCS_BIT,
.tiling_flags = ISL_TILING_CCS_BIT);
}
void
isl_surf_fill_state_s(const struct isl_device *dev, void *state,
const struct isl_surf_fill_state_info *restrict info)
{
#ifndef NDEBUG
isl_surf_usage_flags_t _base_usage =
info->view->usage & (ISL_SURF_USAGE_RENDER_TARGET_BIT |
ISL_SURF_USAGE_TEXTURE_BIT |
ISL_SURF_USAGE_STORAGE_BIT);
/* They may only specify one of the above bits at a time */
assert(__builtin_popcount(_base_usage) == 1);
/* The only other allowed bit is ISL_SURF_USAGE_CUBE_BIT */
assert((info->view->usage & ~ISL_SURF_USAGE_CUBE_BIT) == _base_usage);
#endif
if (info->surf->dim == ISL_SURF_DIM_3D) {
assert(info->view->base_array_layer + info->view->array_len <=
info->surf->logical_level0_px.depth);
} else {
assert(info->view->base_array_layer + info->view->array_len <=
info->surf->logical_level0_px.array_len);
}
switch (ISL_DEV_GEN(dev)) {
case 4:
if (ISL_DEV_IS_G4X(dev)) {
/* G45 surface state is the same as gen5 */
isl_gen5_surf_fill_state_s(dev, state, info);
} else {
isl_gen4_surf_fill_state_s(dev, state, info);
}
break;
case 5:
isl_gen5_surf_fill_state_s(dev, state, info);
break;
case 6:
isl_gen6_surf_fill_state_s(dev, state, info);
break;
case 7:
if (ISL_DEV_IS_HASWELL(dev)) {
isl_gen75_surf_fill_state_s(dev, state, info);
} else {
isl_gen7_surf_fill_state_s(dev, state, info);
}
break;
case 8:
isl_gen8_surf_fill_state_s(dev, state, info);
break;
case 9:
isl_gen9_surf_fill_state_s(dev, state, info);
break;
default:
assert(!"Cannot fill surface state for this gen");
}
}
void
isl_buffer_fill_state_s(const struct isl_device *dev, void *state,
const struct isl_buffer_fill_state_info *restrict info)
{
switch (ISL_DEV_GEN(dev)) {
case 4:
case 5:
/* Gen 4-5 are all the same when it comes to buffer surfaces */
isl_gen5_buffer_fill_state_s(state, info);
break;
case 6:
isl_gen6_buffer_fill_state_s(state, info);
break;
case 7:
if (ISL_DEV_IS_HASWELL(dev)) {
isl_gen75_buffer_fill_state_s(state, info);
} else {
isl_gen7_buffer_fill_state_s(state, info);
}
break;
case 8:
isl_gen8_buffer_fill_state_s(state, info);
break;
case 9:
isl_gen9_buffer_fill_state_s(state, info);
break;
default:
assert(!"Cannot fill surface state for this gen");
}
}
/**
* A variant of isl_surf_get_image_offset_sa() specific to
* ISL_DIM_LAYOUT_GEN4_2D.
*/
static void
get_image_offset_sa_gen4_2d(const struct isl_surf *surf,
uint32_t level, uint32_t logical_array_layer,
uint32_t *x_offset_sa,
uint32_t *y_offset_sa)
{
assert(level < surf->levels);
if (surf->dim == ISL_SURF_DIM_3D)
assert(logical_array_layer < surf->logical_level0_px.depth);
else
assert(logical_array_layer < surf->logical_level0_px.array_len);
const struct isl_extent3d image_align_sa =
isl_surf_get_image_alignment_sa(surf);
const uint32_t W0 = surf->phys_level0_sa.width;
const uint32_t H0 = surf->phys_level0_sa.height;
const uint32_t phys_layer = logical_array_layer *
(surf->msaa_layout == ISL_MSAA_LAYOUT_ARRAY ? surf->samples : 1);
uint32_t x = 0;
uint32_t y = phys_layer * isl_surf_get_array_pitch_sa_rows(surf);
for (uint32_t l = 0; l < level; ++l) {
if (l == 1) {
uint32_t W = isl_minify(W0, l);
x += isl_align_npot(W, image_align_sa.w);
} else {
uint32_t H = isl_minify(H0, l);
y += isl_align_npot(H, image_align_sa.h);
}
}
*x_offset_sa = x;
*y_offset_sa = y;
}
/**
* A variant of isl_surf_get_image_offset_sa() specific to
* ISL_DIM_LAYOUT_GEN4_3D.
*/
static void
get_image_offset_sa_gen4_3d(const struct isl_surf *surf,
uint32_t level, uint32_t logical_z_offset_px,
uint32_t *x_offset_sa,
uint32_t *y_offset_sa)
{
assert(level < surf->levels);
assert(logical_z_offset_px < isl_minify(surf->phys_level0_sa.depth, level));
assert(surf->phys_level0_sa.array_len == 1);
const struct isl_extent3d image_align_sa =
isl_surf_get_image_alignment_sa(surf);
const uint32_t W0 = surf->phys_level0_sa.width;
const uint32_t H0 = surf->phys_level0_sa.height;
const uint32_t D0 = surf->phys_level0_sa.depth;
uint32_t x = 0;
uint32_t y = 0;
for (uint32_t l = 0; l < level; ++l) {
const uint32_t level_h = isl_align_npot(isl_minify(H0, l), image_align_sa.h);
const uint32_t level_d = isl_align_npot(isl_minify(D0, l), image_align_sa.d);
const uint32_t max_layers_vert = isl_align(level_d, 1u << l) / (1u << l);
y += level_h * max_layers_vert;
}
const uint32_t level_w = isl_align_npot(isl_minify(W0, level), image_align_sa.w);
const uint32_t level_h = isl_align_npot(isl_minify(H0, level), image_align_sa.h);
const uint32_t level_d = isl_align_npot(isl_minify(D0, level), image_align_sa.d);
const uint32_t max_layers_horiz = MIN(level_d, 1u << level);
x += level_w * (logical_z_offset_px % max_layers_horiz);
y += level_h * (logical_z_offset_px / max_layers_horiz);
*x_offset_sa = x;
*y_offset_sa = y;
}
/**
* A variant of isl_surf_get_image_offset_sa() specific to
* ISL_DIM_LAYOUT_GEN9_1D.
*/
static void
get_image_offset_sa_gen9_1d(const struct isl_surf *surf,
uint32_t level, uint32_t layer,
uint32_t *x_offset_sa,
uint32_t *y_offset_sa)
{
assert(level < surf->levels);
assert(layer < surf->phys_level0_sa.array_len);
assert(surf->phys_level0_sa.height == 1);
assert(surf->phys_level0_sa.depth == 1);
assert(surf->samples == 1);
const uint32_t W0 = surf->phys_level0_sa.width;
const struct isl_extent3d image_align_sa =
isl_surf_get_image_alignment_sa(surf);
uint32_t x = 0;
for (uint32_t l = 0; l < level; ++l) {
uint32_t W = isl_minify(W0, l);
uint32_t w = isl_align_npot(W, image_align_sa.w);
x += w;
}
*x_offset_sa = x;
*y_offset_sa = layer * isl_surf_get_array_pitch_sa_rows(surf);
}
/**
* Calculate the offset, in units of surface samples, to a subimage in the
* surface.
*
* @invariant level < surface levels
* @invariant logical_array_layer < logical array length of surface
* @invariant logical_z_offset_px < logical depth of surface at level
*/
void
isl_surf_get_image_offset_sa(const struct isl_surf *surf,
uint32_t level,
uint32_t logical_array_layer,
uint32_t logical_z_offset_px,
uint32_t *x_offset_sa,
uint32_t *y_offset_sa)
{
assert(level < surf->levels);
assert(logical_array_layer < surf->logical_level0_px.array_len);
assert(logical_z_offset_px
< isl_minify(surf->logical_level0_px.depth, level));
switch (surf->dim_layout) {
case ISL_DIM_LAYOUT_GEN9_1D:
get_image_offset_sa_gen9_1d(surf, level, logical_array_layer,
x_offset_sa, y_offset_sa);
break;
case ISL_DIM_LAYOUT_GEN4_2D:
get_image_offset_sa_gen4_2d(surf, level, logical_array_layer
+ logical_z_offset_px,
x_offset_sa, y_offset_sa);
break;
case ISL_DIM_LAYOUT_GEN4_3D:
get_image_offset_sa_gen4_3d(surf, level, logical_z_offset_px,
x_offset_sa, y_offset_sa);
break;
default:
unreachable("not reached");
}
}
void
isl_surf_get_image_offset_el(const struct isl_surf *surf,
uint32_t level,
uint32_t logical_array_layer,
uint32_t logical_z_offset_px,
uint32_t *x_offset_el,
uint32_t *y_offset_el)
{
const struct isl_format_layout *fmtl = isl_format_get_layout(surf->format);
assert(level < surf->levels);
assert(logical_array_layer < surf->logical_level0_px.array_len);
assert(logical_z_offset_px
< isl_minify(surf->logical_level0_px.depth, level));
uint32_t x_offset_sa, y_offset_sa;
isl_surf_get_image_offset_sa(surf, level,
logical_array_layer,
logical_z_offset_px,
&x_offset_sa,
&y_offset_sa);
*x_offset_el = x_offset_sa / fmtl->bw;
*y_offset_el = y_offset_sa / fmtl->bh;
}
void
isl_tiling_get_intratile_offset_el(const struct isl_device *dev,
enum isl_tiling tiling,
uint8_t bs,
uint32_t row_pitch,
uint32_t total_x_offset_el,
uint32_t total_y_offset_el,
uint32_t *base_address_offset,
uint32_t *x_offset_el,
uint32_t *y_offset_el)
{
if (tiling == ISL_TILING_LINEAR) {
*base_address_offset = total_y_offset_el * row_pitch +
total_x_offset_el * bs;
*x_offset_el = 0;
*y_offset_el = 0;
return;
}
const uint32_t bpb = bs * 8;
struct isl_tile_info tile_info;
isl_tiling_get_info(dev, tiling, bpb, &tile_info);
assert(row_pitch % tile_info.phys_extent_B.width == 0);
/* For non-power-of-two formats, we need the address to be both tile and
* element-aligned. The easiest way to achieve this is to work with a tile
* that is three times as wide as the regular tile.
*
* The tile info returned by get_tile_info has a logical size that is an
* integer number of tile_info.format_bpb size elements. To scale the
* tile, we scale up the physical width and then treat the logical tile
* size as if it has bpb size elements.
*/
const uint32_t tile_el_scale = bpb / tile_info.format_bpb;
tile_info.phys_extent_B.width *= tile_el_scale;
/* Compute the offset into the tile */
*x_offset_el = total_x_offset_el % tile_info.logical_extent_el.w;
*y_offset_el = total_y_offset_el % tile_info.logical_extent_el.h;
/* Compute the offset of the tile in units of whole tiles */
uint32_t x_offset_tl = total_x_offset_el / tile_info.logical_extent_el.w;
uint32_t y_offset_tl = total_y_offset_el / tile_info.logical_extent_el.h;
*base_address_offset =
y_offset_tl * tile_info.phys_extent_B.h * row_pitch +
x_offset_tl * tile_info.phys_extent_B.h * tile_info.phys_extent_B.w;
}
uint32_t
isl_surf_get_depth_format(const struct isl_device *dev,
const struct isl_surf *surf)
{
/* Support for separate stencil buffers began in gen5. Support for
* interleaved depthstencil buffers ceased in gen7. The intermediate gens,
* those that supported separate and interleaved stencil, were gen5 and
* gen6.
*
* For a list of all available formats, see the Sandybridge PRM >> Volume
* 2 Part 1: 3D/Media - 3D Pipeline >> 3DSTATE_DEPTH_BUFFER >> Surface
* Format (p321).
*/
bool has_stencil = surf->usage & ISL_SURF_USAGE_STENCIL_BIT;
assert(surf->usage & ISL_SURF_USAGE_DEPTH_BIT);
if (has_stencil)
assert(ISL_DEV_GEN(dev) < 7);
switch (surf->format) {
default:
unreachable("bad isl depth format");
case ISL_FORMAT_R32_FLOAT_X8X24_TYPELESS:
assert(ISL_DEV_GEN(dev) < 7);
return 0; /* D32_FLOAT_S8X24_UINT */
case ISL_FORMAT_R32_FLOAT:
assert(!has_stencil);
return 1; /* D32_FLOAT */
case ISL_FORMAT_R24_UNORM_X8_TYPELESS:
if (has_stencil) {
assert(ISL_DEV_GEN(dev) < 7);
return 2; /* D24_UNORM_S8_UINT */
} else {
assert(ISL_DEV_GEN(dev) >= 5);
return 3; /* D24_UNORM_X8_UINT */
}
case ISL_FORMAT_R16_UNORM:
assert(!has_stencil);
return 5; /* D16_UNORM */
}
}
|