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
|
; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
; RUN: llc -global-isel -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1010 -stop-after=instruction-select -verify-machineinstrs -o - %s | FileCheck -check-prefix=GFX10_GFX11 %s
; RUN: llc -global-isel -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1100 -stop-after=instruction-select -verify-machineinstrs -o - %s | FileCheck -check-prefix=GFX10_GFX11 %s
; RUN: llc -global-isel -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1200 -stop-after=instruction-select -verify-machineinstrs -o - %s | FileCheck -check-prefix=GFX12 %s
; Natural mapping
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 0)
ret void
}
; Natural mapping
define amdgpu_ps void @raw_tbuffer_store_v2f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset(<2 x float> %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_v2f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1, $vgpr2
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY2]], %subreg.sub0, [[COPY3]], %subreg.sub1, [[COPY4]], %subreg.sub2, [[COPY5]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_XY_OFFEN_exact [[REG_SEQUENCE]], [[COPY6]], [[REG_SEQUENCE1]], [[COPY7]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (<2 x s32>), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_v2f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1, $vgpr2
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY2]], %subreg.sub0, [[COPY3]], %subreg.sub1, [[COPY4]], %subreg.sub2, [[COPY5]], %subreg.sub3
; GFX12-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[COPY7:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_XY_VBUFFER_OFFEN_exact [[REG_SEQUENCE]], [[COPY6]], [[REG_SEQUENCE1]], [[COPY7]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (<2 x s32>), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.v2f32(<2 x float> %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 0)
ret void
}
; Natural mapping
define amdgpu_ps void @raw_tbuffer_store_v3f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset(<3 x float> %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_v3f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1, $vgpr2, $vgpr3
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_96 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1, [[COPY2]], %subreg.sub2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY3]], %subreg.sub0, [[COPY4]], %subreg.sub1, [[COPY5]], %subreg.sub2, [[COPY6]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX10_GFX11-NEXT: [[COPY8:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_XYZ_OFFEN_exact [[REG_SEQUENCE]], [[COPY7]], [[REG_SEQUENCE1]], [[COPY8]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (<3 x s32>), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_v3f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1, $vgpr2, $vgpr3
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_96 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1, [[COPY2]], %subreg.sub2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY3]], %subreg.sub0, [[COPY4]], %subreg.sub1, [[COPY5]], %subreg.sub2, [[COPY6]], %subreg.sub3
; GFX12-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX12-NEXT: [[COPY8:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_XYZ_VBUFFER_OFFEN_exact [[REG_SEQUENCE]], [[COPY7]], [[REG_SEQUENCE1]], [[COPY8]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (<3 x s32>), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.v3f32(<3 x float> %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 0)
ret void
}
; Natural mapping
define amdgpu_ps void @raw_tbuffer_store_v4f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset(<4 x float> %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_v4f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1, [[COPY2]], %subreg.sub2, [[COPY3]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY4]], %subreg.sub0, [[COPY5]], %subreg.sub1, [[COPY6]], %subreg.sub2, [[COPY7]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX10_GFX11-NEXT: [[COPY9:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_XYZW_OFFEN_exact [[REG_SEQUENCE]], [[COPY8]], [[REG_SEQUENCE1]], [[COPY9]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (<4 x s32>), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_v4f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1, [[COPY2]], %subreg.sub2, [[COPY3]], %subreg.sub3
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY7:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY4]], %subreg.sub0, [[COPY5]], %subreg.sub1, [[COPY6]], %subreg.sub2, [[COPY7]], %subreg.sub3
; GFX12-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX12-NEXT: [[COPY9:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_XYZW_VBUFFER_OFFEN_exact [[REG_SEQUENCE]], [[COPY8]], [[REG_SEQUENCE1]], [[COPY9]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (<4 x s32>), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.v4f32(<4 x float> %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 0)
ret void
}
; Copies for VGPR arguments
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__sgpr_voffset__sgpr_soffset(float %val, <4 x i32> inreg %rsrc, i32 inreg %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__sgpr_voffset__sgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $vgpr0
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr7
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]]
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY7]], [[REG_SEQUENCE]], [[COPY6]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__sgpr_voffset__sgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $vgpr0
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr7
; GFX12-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]]
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY7]], [[REG_SEQUENCE]], [[COPY6]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
; Waterfall for rsrc
define amdgpu_ps void @raw_tbuffer_store_f32__vgpr_rsrc__vgpr_voffset__sgpr_soffset(float %val, <4 x i32> %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__vgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: successors: %bb.2(0x80000000)
; GFX10_GFX11-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.2:
; GFX10_GFX11-NEXT: successors: %bb.3(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX10_GFX11-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX10_GFX11-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.3:
; GFX10_GFX11-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[COPY6]], 0, 94, 1, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX10_GFX11-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.4:
; GFX10_GFX11-NEXT: successors: %bb.5(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_]]
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.5:
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__vgpr_rsrc__vgpr_voffset__sgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: successors: %bb.2(0x80000000)
; GFX12-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX12-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.2:
; GFX12-NEXT: successors: %bb.3(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX12-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX12-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX12-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX12-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX12-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX12-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.3:
; GFX12-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[COPY6]], 0, 94, 1, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX12-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.4:
; GFX12-NEXT: successors: %bb.5(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_]]
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.5:
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 1)
ret void
}
; Waterfall for rsrc and soffset
define amdgpu_ps void @raw_tbuffer_store_f32__vgpr_rsrc__vgpr_voffset__vgpr_soffset(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__vgpr_rsrc__vgpr_voffset__vgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: successors: %bb.2(0x80000000)
; GFX10_GFX11-NEXT: liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr6
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.2:
; GFX10_GFX11-NEXT: successors: %bb.3(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX10_GFX11-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX10_GFX11-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec
; GFX10_GFX11-NEXT: [[S_AND_B32_1:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[S_AND_B32_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_1]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.3:
; GFX10_GFX11-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[V_READFIRSTLANE_B32_4]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX10_GFX11-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.4:
; GFX10_GFX11-NEXT: successors: %bb.5(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_]]
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.5:
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__vgpr_rsrc__vgpr_voffset__vgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: successors: %bb.2(0x80000000)
; GFX12-NEXT: liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX12-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX12-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr6
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.2:
; GFX12-NEXT: successors: %bb.3(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX12-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX12-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX12-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX12-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX12-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX12-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec
; GFX12-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec
; GFX12-NEXT: [[S_AND_B32_1:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[S_AND_B32_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc
; GFX12-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_1]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.3:
; GFX12-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[V_READFIRSTLANE_B32_4]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX12-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.4:
; GFX12-NEXT: successors: %bb.5(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_]]
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.5:
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
; Waterfall for rsrc and soffset, copy for voffset
define amdgpu_ps void @raw_tbuffer_store_f32__vgpr_rsrc__sgpr_voffset__vgpr_soffset(float %val, <4 x i32> %rsrc, i32 inreg %voffset, i32 %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__vgpr_rsrc__sgpr_voffset__vgpr_soffset
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: successors: %bb.2(0x80000000)
; GFX10_GFX11-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]]
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.2:
; GFX10_GFX11-NEXT: successors: %bb.3(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY9:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX10_GFX11-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY11:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY11]], [[COPY9]], implicit $exec
; GFX10_GFX11-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec
; GFX10_GFX11-NEXT: [[S_AND_B32_1:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[S_AND_B32_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_1]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.3:
; GFX10_GFX11-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY7]], [[REG_SEQUENCE1]], [[V_READFIRSTLANE_B32_4]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX10_GFX11-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.4:
; GFX10_GFX11-NEXT: successors: %bb.5(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_]]
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.5:
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__vgpr_rsrc__sgpr_voffset__vgpr_soffset
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: successors: %bb.2(0x80000000)
; GFX12-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX12-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX12-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]]
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.2:
; GFX12-NEXT: successors: %bb.3(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX12-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX12-NEXT: [[COPY9:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX12-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX12-NEXT: [[COPY11:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY11]], [[COPY9]], implicit $exec
; GFX12-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX12-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec
; GFX12-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec
; GFX12-NEXT: [[S_AND_B32_1:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[S_AND_B32_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc
; GFX12-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_1]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.3:
; GFX12-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY7]], [[REG_SEQUENCE1]], [[V_READFIRSTLANE_B32_4]], 0, 78, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX12-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.4:
; GFX12-NEXT: successors: %bb.5(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_]]
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.5:
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 0)
ret void
}
; Natural mapping + glc
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_glc(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_glc
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 1, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_glc
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 1, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 1)
ret void
}
; Natural mapping + slc
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_slc(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_slc
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 2, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_slc
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 2, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 2)
ret void
}
; Natural mapping + glc + slc
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_slc_glc(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_slc_glc
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 3, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_slc_glc
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 3, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 3)
ret void
}
; Natural mapping + dlc
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_dlc(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_dlc
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 4, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_dlc
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 0, 78, 4, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 78, i32 4)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vdpr_voffset__sgpr_soffset__voffset0(float %val, <4 x i32> inreg %rsrc, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vdpr_voffset__sgpr_soffset__voffset0
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFSET_exact [[COPY]], [[REG_SEQUENCE]], [[COPY5]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vdpr_voffset__sgpr_soffset__voffset0
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFSET_exact [[COPY]], [[REG_SEQUENCE]], [[COPY5]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 0, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset4095(float %val, <4 x i32> inreg %rsrc, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset4095
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFSET_exact [[COPY]], [[REG_SEQUENCE]], [[COPY5]], 4095, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset4095
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFSET_exact [[COPY]], [[REG_SEQUENCE]], [[COPY5]], 4095, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 4095, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset4096(float %val, <4 x i32> inreg %rsrc, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset4096
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4096
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY6]], [[REG_SEQUENCE]], [[COPY5]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset4096
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFSET_exact [[COPY]], [[REG_SEQUENCE]], [[COPY5]], 4096, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 4096, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_voffset_add16(float %val, <4 x i32> inreg %rsrc, i32 %voffset.base, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_voffset_add16
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 16, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_voffset_add16
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 16, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
%voffset = add i32 %voffset.base, 16
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset_add4095(float %val, <4 x i32> inreg %rsrc, i32 %voffset.base, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset_add4095
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 4095, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset_add4095
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 4095, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
%voffset = add i32 %voffset.base, 4095
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset_add4096(float %val, <4 x i32> inreg %rsrc, i32 %voffset.base, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset_add4096
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4096
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
; GFX10_GFX11-NEXT: [[V_ADD_U32_e64_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e64 [[COPY5]], [[COPY7]], 0, implicit $exec
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[V_ADD_U32_e64_]], [[REG_SEQUENCE]], [[COPY6]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset__voffset_add4096
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[COPY6]], 4096, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
%voffset = add i32 %voffset.base, 4096
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset4095(float %val, <4 x i32> inreg %rsrc, i32 %voffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset4095
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4095
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_MOV_B32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset4095
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4095
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_MOV_B32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 4095, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset4096(float %val, <4 x i32> inreg %rsrc, i32 %voffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset4096
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4096
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_MOV_B32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset4096
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4096
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_MOV_B32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 4096, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add16(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset.base) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add16
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 16
; GFX10_GFX11-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add16
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 16
; GFX12-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
%soffset = add i32 %soffset.base, 16
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add4095(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset.base) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add4095
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4095
; GFX10_GFX11-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add4095
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4095
; GFX12-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
%soffset = add i32 %soffset.base, 4095
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add4096(float %val, <4 x i32> inreg %rsrc, i32 %voffset, i32 inreg %soffset.base) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add4096
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4096
; GFX10_GFX11-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add4096
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: liveins: $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $vgpr0, $vgpr1
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY $sgpr3
; GFX12-NEXT: [[COPY3:%[0-9]+]]:sreg_32 = COPY $sgpr4
; GFX12-NEXT: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr5
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr6
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4096
; GFX12-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: S_ENDPGM 0
%soffset = add i32 %soffset.base, 4096
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
; An add of the offset is necessary, with a waterfall loop. Make sure the add is done outside of the waterfall loop.
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add5000(float %val, <4 x i32> %rsrc, i32 %voffset, i32 inreg %soffset.base) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add5000
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: successors: %bb.2(0x80000000)
; GFX10_GFX11-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 5000
; GFX10_GFX11-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.2:
; GFX10_GFX11-NEXT: successors: %bb.3(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX10_GFX11-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX10_GFX11-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.3:
; GFX10_GFX11-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX10_GFX11-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.4:
; GFX10_GFX11-NEXT: successors: %bb.5(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_1]]
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.5:
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_soffset_add5000
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: successors: %bb.2(0x80000000)
; GFX12-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX12-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 5000
; GFX12-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[COPY6]], [[S_MOV_B32_]], implicit-def dead $scc
; GFX12-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.2:
; GFX12-NEXT: successors: %bb.3(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX12-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX12-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX12-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX12-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX12-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX12-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.3:
; GFX12-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[S_ADD_I32_]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX12-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.4:
; GFX12-NEXT: successors: %bb.5(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_1]]
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.5:
; GFX12-NEXT: S_ENDPGM 0
%soffset = add i32 %soffset.base, 5000
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
; An add of the offset is necessary, with a waterfall loop. Make sure the add is done outside of the waterfall loop.
define amdgpu_ps void @raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_voffset_add5000(float %val, <4 x i32> %rsrc, i32 %voffset.base, i32 inreg %soffset) {
; GFX10_GFX11-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_voffset_add5000
; GFX10_GFX11: bb.1 (%ir-block.0):
; GFX10_GFX11-NEXT: successors: %bb.2(0x80000000)
; GFX10_GFX11-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10_GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10_GFX11-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10_GFX11-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX10_GFX11-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX10_GFX11-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX10_GFX11-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX10_GFX11-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 4096
; GFX10_GFX11-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
; GFX10_GFX11-NEXT: [[V_ADD_U32_e64_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e64 [[COPY5]], [[COPY7]], 0, implicit $exec
; GFX10_GFX11-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.2:
; GFX10_GFX11-NEXT: successors: %bb.3(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX10_GFX11-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX10_GFX11-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX10_GFX11-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY9:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX10_GFX11-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX10_GFX11-NEXT: [[COPY11:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX10_GFX11-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY11]], [[COPY9]], implicit $exec
; GFX10_GFX11-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX10_GFX11-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.3:
; GFX10_GFX11-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: TBUFFER_STORE_FORMAT_X_OFFEN_exact [[COPY]], [[V_ADD_U32_e64_]], [[REG_SEQUENCE1]], [[COPY6]], 904, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX10_GFX11-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX10_GFX11-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.4:
; GFX10_GFX11-NEXT: successors: %bb.5(0x80000000)
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_1]]
; GFX10_GFX11-NEXT: {{ $}}
; GFX10_GFX11-NEXT: bb.5:
; GFX10_GFX11-NEXT: S_ENDPGM 0
;
; GFX12-LABEL: name: raw_tbuffer_store_f32__sgpr_rsrc__vgpr_voffset__sgpr_soffset_voffset_add5000
; GFX12: bb.1 (%ir-block.0):
; GFX12-NEXT: successors: %bb.2(0x80000000)
; GFX12-NEXT: liveins: $sgpr2, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX12-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX12-NEXT: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX12-NEXT: [[COPY3:%[0-9]+]]:vgpr_32 = COPY $vgpr3
; GFX12-NEXT: [[COPY4:%[0-9]+]]:vgpr_32 = COPY $vgpr4
; GFX12-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3
; GFX12-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5
; GFX12-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GFX12-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_MOV_B32 $exec_lo
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.2:
; GFX12-NEXT: successors: %bb.3(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec
; GFX12-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec
; GFX12-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3
; GFX12-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1
; GFX12-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3
; GFX12-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1
; GFX12-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec
; GFX12-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec
; GFX12-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_B32 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc
; GFX12-NEXT: [[S_AND_SAVEEXEC_B32_:%[0-9]+]]:sreg_32_xm0_xexec = S_AND_SAVEEXEC_B32 killed [[S_AND_B32_]], implicit-def $exec, implicit-def $scc, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.3:
; GFX12-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: TBUFFER_STORE_FORMAT_X_VBUFFER_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[COPY6]], 5000, 94, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
; GFX12-NEXT: $exec_lo = S_XOR_B32_term $exec_lo, [[S_AND_SAVEEXEC_B32_]], implicit-def $scc
; GFX12-NEXT: SI_WATERFALL_LOOP %bb.2, implicit $exec
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.4:
; GFX12-NEXT: successors: %bb.5(0x80000000)
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: $exec_lo = S_MOV_B32_term [[S_MOV_B32_]]
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: bb.5:
; GFX12-NEXT: S_ENDPGM 0
%voffset = add i32 %voffset.base, 5000
call void @llvm.amdgcn.raw.tbuffer.store.f32(float %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 94, i32 0)
ret void
}
declare void @llvm.amdgcn.raw.tbuffer.store.f32(float, <4 x i32>, i32, i32, i32 immarg, i32 immarg)
declare void @llvm.amdgcn.raw.tbuffer.store.v2f32(<2 x float>, <4 x i32>, i32, i32, i32 immarg, i32 immarg)
declare void @llvm.amdgcn.raw.tbuffer.store.v3f32(<3 x float>, <4 x i32>, i32, i32, i32 immarg, i32 immarg)
declare void @llvm.amdgcn.raw.tbuffer.store.v4f32(<4 x float>, <4 x i32>, i32, i32, i32 immarg, i32 immarg)
|