공재호 / JJS_Project1

Go to a project
Toggle navigation Toggle navigation pinning
Switch branch/tag
vec-strict-fptoint-512.ll 41.7 KB
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 
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx512f,avx512vl -O3 | FileCheck %s --check-prefixes=CHECK,AVX512VL,AVX512VL-32
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,avx512vl -O3 | FileCheck %s --check-prefixes=CHECK,AVX512VL,AVX512VL-64
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=avx512f,avx512dq -O3 | FileCheck %s --check-prefixes=CHECK,AVX512DQ,AVX512DQ-32
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx512f,avx512dq -O3 | FileCheck %s --check-prefixes=CHECK,AVX512DQ,AVX512DQ-64

declare <8 x i64> @llvm.experimental.constrained.fptosi.v8i64.v8f64(<8 x double>, metadata)
declare <8 x i64> @llvm.experimental.constrained.fptoui.v8i64.v8f64(<8 x double>, metadata)
declare <8 x i64> @llvm.experimental.constrained.fptosi.v8i64.v8f32(<8 x float>, metadata)
declare <8 x i64> @llvm.experimental.constrained.fptoui.v8i64.v8f32(<8 x float>, metadata)
declare <8 x i32> @llvm.experimental.constrained.fptosi.v8i32.v8f64(<8 x double>, metadata)
declare <8 x i32> @llvm.experimental.constrained.fptoui.v8i32.v8f64(<8 x double>, metadata)
declare <8 x i16> @llvm.experimental.constrained.fptosi.v8i16.v8f64(<8 x double>, metadata)
declare <8 x i16> @llvm.experimental.constrained.fptoui.v8i16.v8f64(<8 x double>, metadata)
declare <8 x i8> @llvm.experimental.constrained.fptosi.v8i8.v8f64(<8 x double>, metadata)
declare <8 x i8> @llvm.experimental.constrained.fptoui.v8i8.v8f64(<8 x double>, metadata)
declare <8 x i1> @llvm.experimental.constrained.fptosi.v8i1.v8f64(<8 x double>, metadata)
declare <8 x i1> @llvm.experimental.constrained.fptoui.v8i1.v8f64(<8 x double>, metadata)
declare <8 x i1> @llvm.experimental.constrained.fptosi.v8i1.v8f32(<8 x float>, metadata)
declare <8 x i1> @llvm.experimental.constrained.fptoui.v8i1.v8f32(<8 x float>, metadata)

declare <16 x i32> @llvm.experimental.constrained.fptosi.v16i32.v16f32(<16 x float>, metadata)
declare <16 x i32> @llvm.experimental.constrained.fptoui.v16i32.v16f32(<16 x float>, metadata)
declare <16 x i16> @llvm.experimental.constrained.fptosi.v16i16.v16f32(<16 x float>, metadata)
declare <16 x i16> @llvm.experimental.constrained.fptoui.v16i16.v16f32(<16 x float>, metadata)
declare <16 x i8> @llvm.experimental.constrained.fptosi.v16i8.v16f32(<16 x float>, metadata)
declare <16 x i8> @llvm.experimental.constrained.fptoui.v16i8.v16f32(<16 x float>, metadata)
declare <16 x i1> @llvm.experimental.constrained.fptosi.v16i1.v16f32(<16 x float>, metadata)
declare <16 x i1> @llvm.experimental.constrained.fptoui.v16i1.v16f32(<16 x float>, metadata)

define <8 x i64> @strict_vector_fptosi_v8f64_to_v8i64(<8 x double> %a) #0 {
; AVX512VL-32-LABEL: strict_vector_fptosi_v8f64_to_v8i64:
; AVX512VL-32:       # %bb.0:
; AVX512VL-32-NEXT:    pushl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_offset 8
; AVX512VL-32-NEXT:    .cfi_offset %ebp, -8
; AVX512VL-32-NEXT:    movl %esp, %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_register %ebp
; AVX512VL-32-NEXT:    andl $-8, %esp
; AVX512VL-32-NEXT:    subl $64, %esp
; AVX512VL-32-NEXT:    vextractf32x4 $2, %zmm0, %xmm1
; AVX512VL-32-NEXT:    vmovlps %xmm1, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovhps %xmm1, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractf32x4 $3, %zmm0, %xmm1
; AVX512VL-32-NEXT:    vmovlps %xmm1, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovhps %xmm1, (%esp)
; AVX512VL-32-NEXT:    vmovlps %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovhps %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractf128 $1, %ymm0, %xmm0
; AVX512VL-32-NEXT:    vmovlps %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovhps %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl (%esp)
; AVX512VL-32-NEXT:    fisttpll (%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $2, (%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm2 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm2, %ymm1, %ymm1
; AVX512VL-32-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
; AVX512VL-32-NEXT:    movl %ebp, %esp
; AVX512VL-32-NEXT:    popl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa %esp, 4
; AVX512VL-32-NEXT:    retl
;
; AVX512VL-64-LABEL: strict_vector_fptosi_v8f64_to_v8i64:
; AVX512VL-64:       # %bb.0:
; AVX512VL-64-NEXT:    vextractf32x4 $3, %zmm0, %xmm1
; AVX512VL-64-NEXT:    vcvttsd2si %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
; AVX512VL-64-NEXT:    vcvttsd2si %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm1
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX512VL-64-NEXT:    vextractf32x4 $2, %zmm0, %xmm2
; AVX512VL-64-NEXT:    vcvttsd2si %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm2 = xmm2[1,0]
; AVX512VL-64-NEXT:    vcvttsd2si %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
; AVX512VL-64-NEXT:    vextractf128 $1, %ymm0, %xmm2
; AVX512VL-64-NEXT:    vcvttsd2si %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm2 = xmm2[1,0]
; AVX512VL-64-NEXT:    vcvttsd2si %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vcvttsd2si %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
; AVX512VL-64-NEXT:    vcvttsd2si %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm0
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm3[0],xmm0[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
; AVX512VL-64-NEXT:    vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512VL-64-NEXT:    retq
;
; AVX512DQ-LABEL: strict_vector_fptosi_v8f64_to_v8i64:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2qq %zmm0, %zmm0
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i64> @llvm.experimental.constrained.fptosi.v8i64.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i64> %ret
}

define <8 x i64> @strict_vector_fptoui_v8f64_to_v8i64(<8 x double> %a) #0 {
; AVX512VL-32-LABEL: strict_vector_fptoui_v8f64_to_v8i64:
; AVX512VL-32:       # %bb.0:
; AVX512VL-32-NEXT:    pushl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_offset 8
; AVX512VL-32-NEXT:    .cfi_offset %ebp, -8
; AVX512VL-32-NEXT:    movl %esp, %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_register %ebp
; AVX512VL-32-NEXT:    pushl %ebx
; AVX512VL-32-NEXT:    pushl %edi
; AVX512VL-32-NEXT:    pushl %esi
; AVX512VL-32-NEXT:    andl $-8, %esp
; AVX512VL-32-NEXT:    subl $80, %esp
; AVX512VL-32-NEXT:    .cfi_offset %esi, -20
; AVX512VL-32-NEXT:    .cfi_offset %edi, -16
; AVX512VL-32-NEXT:    .cfi_offset %ebx, -12
; AVX512VL-32-NEXT:    vpermilpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512VL-32-NEXT:    vmovsd {{.*#+}} xmm1 = mem[0],zero
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vxorpd %xmm2, %xmm2, %xmm2
; AVX512VL-32-NEXT:    vmovapd %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovsd %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; AVX512VL-32-NEXT:    vextractf128 $1, %ymm0, %xmm3
; AVX512VL-32-NEXT:    vpermilpd {{.*#+}} xmm4 = xmm3[1,0]
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm4
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovapd %xmm1, %xmm5
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm5, %xmm5 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm5, %xmm4, %xmm4
; AVX512VL-32-NEXT:    vmovsd %xmm4, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovapd %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovsd %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; AVX512VL-32-NEXT:    vextractf32x4 $2, %zmm0, %xmm3
; AVX512VL-32-NEXT:    vpermilpd {{.*#+}} xmm4 = xmm3[1,0]
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm4
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovapd %xmm1, %xmm5
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm5, %xmm5 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm5, %xmm4, %xmm4
; AVX512VL-32-NEXT:    vmovsd %xmm4, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, %edi
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovapd %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovsd %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, %esi
; AVX512VL-32-NEXT:    vextractf32x4 $3, %zmm0, %xmm3
; AVX512VL-32-NEXT:    vpermilpd {{.*#+}} xmm4 = xmm3[1,0]
; AVX512VL-32-NEXT:    xorl %edx, %edx
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm4
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovapd %xmm1, %xmm5
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm5, %xmm5 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm5, %xmm4, %xmm4
; AVX512VL-32-NEXT:    vmovsd %xmm4, (%esp)
; AVX512VL-32-NEXT:    fldl (%esp)
; AVX512VL-32-NEXT:    fisttpll (%esp)
; AVX512VL-32-NEXT:    setae %dl
; AVX512VL-32-NEXT:    shll $31, %edx
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %edx
; AVX512VL-32-NEXT:    xorl %eax, %eax
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %cl
; AVX512VL-32-NEXT:    kmovw %ecx, %k1
; AVX512VL-32-NEXT:    vmovapd %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovsd %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    xorl %ecx, %ecx
; AVX512VL-32-NEXT:    vcomisd %xmm1, %xmm0
; AVX512VL-32-NEXT:    setb %bl
; AVX512VL-32-NEXT:    kmovw %ebx, %k1
; AVX512VL-32-NEXT:    vmovsd %xmm2, %xmm1, %xmm1 {%k1}
; AVX512VL-32-NEXT:    vsubsd %xmm1, %xmm0, %xmm0
; AVX512VL-32-NEXT:    vmovsd %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fldl {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, %eax, %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $2, (%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $3, %edx, %xmm0, %xmm0
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, %esi, %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $3, %edi, %xmm1, %xmm1
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm2 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[-0-9]+}}(%e{{[sb]}}p), %xmm2, %xmm2 # 4-byte Folded Reload
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vpinsrd $3, {{[-0-9]+}}(%e{{[sb]}}p), %xmm2, %xmm2 # 4-byte Folded Reload
; AVX512VL-32-NEXT:    setae %cl
; AVX512VL-32-NEXT:    shll $31, %ecx
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %ecx
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm3 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, %ecx, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm3, %xmm3
; AVX512VL-32-NEXT:    vpinsrd $3, {{[-0-9]+}}(%e{{[sb]}}p), %xmm3, %xmm3 # 4-byte Folded Reload
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm2, %ymm3, %ymm1
; AVX512VL-32-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
; AVX512VL-32-NEXT:    leal -12(%ebp), %esp
; AVX512VL-32-NEXT:    popl %esi
; AVX512VL-32-NEXT:    popl %edi
; AVX512VL-32-NEXT:    popl %ebx
; AVX512VL-32-NEXT:    popl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa %esp, 4
; AVX512VL-32-NEXT:    retl
;
; AVX512VL-64-LABEL: strict_vector_fptoui_v8f64_to_v8i64:
; AVX512VL-64:       # %bb.0:
; AVX512VL-64-NEXT:    vextractf32x4 $3, %zmm0, %xmm1
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm1
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX512VL-64-NEXT:    vextractf32x4 $2, %zmm0, %xmm2
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm2 = xmm2[1,0]
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
; AVX512VL-64-NEXT:    vextractf128 $1, %ymm0, %xmm2
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm2 = xmm2[1,0]
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
; AVX512VL-64-NEXT:    vcvttsd2usi %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm0
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm3[0],xmm0[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
; AVX512VL-64-NEXT:    vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512VL-64-NEXT:    retq
;
; AVX512DQ-LABEL: strict_vector_fptoui_v8f64_to_v8i64:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2uqq %zmm0, %zmm0
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i64> @llvm.experimental.constrained.fptoui.v8i64.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i64> %ret
}

define <8 x i64> @strict_vector_fptosi_v8f32_to_v8i64(<8 x float> %a) #0 {
; AVX512VL-32-LABEL: strict_vector_fptosi_v8f32_to_v8i64:
; AVX512VL-32:       # %bb.0:
; AVX512VL-32-NEXT:    pushl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_offset 8
; AVX512VL-32-NEXT:    .cfi_offset %ebp, -8
; AVX512VL-32-NEXT:    movl %esp, %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_register %ebp
; AVX512VL-32-NEXT:    andl $-8, %esp
; AVX512VL-32-NEXT:    subl $64, %esp
; AVX512VL-32-NEXT:    vextracti128 $1, %ymm0, %xmm1
; AVX512VL-32-NEXT:    vmovd %xmm1, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractps $1, %xmm1, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractps $2, %xmm1, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractps $3, %xmm1, (%esp)
; AVX512VL-32-NEXT:    vmovd %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractps $1, %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractps $2, %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vextractps $3, %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds (%esp)
; AVX512VL-32-NEXT:    fisttpll (%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $2, (%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm2 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vpinsrd $3, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm2, %ymm1, %ymm1
; AVX512VL-32-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
; AVX512VL-32-NEXT:    movl %ebp, %esp
; AVX512VL-32-NEXT:    popl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa %esp, 4
; AVX512VL-32-NEXT:    retl
;
; AVX512VL-64-LABEL: strict_vector_fptosi_v8f32_to_v8i64:
; AVX512VL-64:       # %bb.0:
; AVX512VL-64-NEXT:    vextractf128 $1, %ymm0, %xmm1
; AVX512VL-64-NEXT:    vpermilps {{.*#+}} xmm2 = xmm1[3,1,2,3]
; AVX512VL-64-NEXT:    vcvttss2si %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm3 = xmm1[1,0]
; AVX512VL-64-NEXT:    vcvttss2si %xmm3, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vcvttss2si %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm1[1,1,3,3]
; AVX512VL-64-NEXT:    vcvttss2si %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm1
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm1[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm2, %ymm1, %ymm1
; AVX512VL-64-NEXT:    vpermilps {{.*#+}} xmm2 = xmm0[3,1,2,3]
; AVX512VL-64-NEXT:    vcvttss2si %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512VL-64-NEXT:    vcvttss2si %xmm3, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vcvttss2si %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vmovshdup {{.*#+}} xmm0 = xmm0[1,1,3,3]
; AVX512VL-64-NEXT:    vcvttss2si %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm0
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm3[0],xmm0[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
; AVX512VL-64-NEXT:    vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512VL-64-NEXT:    retq
;
; AVX512DQ-LABEL: strict_vector_fptosi_v8f32_to_v8i64:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttps2qq %ymm0, %zmm0
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i64> @llvm.experimental.constrained.fptosi.v8i64.v8f32(<8 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i64> %ret
}

define <8 x i64> @strict_vector_fptoui_v8f32_to_v8i64(<8 x float> %a) #0 {
; AVX512VL-32-LABEL: strict_vector_fptoui_v8f32_to_v8i64:
; AVX512VL-32:       # %bb.0:
; AVX512VL-32-NEXT:    pushl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_offset 8
; AVX512VL-32-NEXT:    .cfi_offset %ebp, -8
; AVX512VL-32-NEXT:    movl %esp, %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa_register %ebp
; AVX512VL-32-NEXT:    pushl %ebx
; AVX512VL-32-NEXT:    pushl %edi
; AVX512VL-32-NEXT:    pushl %esi
; AVX512VL-32-NEXT:    andl $-8, %esp
; AVX512VL-32-NEXT:    subl $80, %esp
; AVX512VL-32-NEXT:    .cfi_offset %esi, -20
; AVX512VL-32-NEXT:    .cfi_offset %edi, -16
; AVX512VL-32-NEXT:    .cfi_offset %ebx, -12
; AVX512VL-32-NEXT:    vmovshdup {{.*#+}} xmm3 = xmm0[1,1,3,3]
; AVX512VL-32-NEXT:    vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vxorps %xmm2, %xmm2, %xmm2
; AVX512VL-32-NEXT:    vmovaps %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovss %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; AVX512VL-32-NEXT:    vpermilps {{.*#+}} xmm3 = xmm0[3,1,2,3]
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovaps %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovss %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; AVX512VL-32-NEXT:    vpermilpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovaps %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovss %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; AVX512VL-32-NEXT:    vextractf128 $1, %ymm0, %xmm3
; AVX512VL-32-NEXT:    vmovshdup {{.*#+}} xmm4 = xmm3[1,1,3,3]
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm4
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovaps %xmm1, %xmm5
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm5, %xmm5 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm5, %xmm4, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm4, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, %edi
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovaps %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm4, %xmm3, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm4, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    movl $0, %eax
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    movl %eax, %esi
; AVX512VL-32-NEXT:    vpermilps {{.*#+}} xmm4 = xmm3[3,1,2,3]
; AVX512VL-32-NEXT:    xorl %edx, %edx
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm4
; AVX512VL-32-NEXT:    setb %al
; AVX512VL-32-NEXT:    kmovw %eax, %k1
; AVX512VL-32-NEXT:    vmovaps %xmm1, %xmm5
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm5, %xmm5 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm5, %xmm4, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm4, (%esp)
; AVX512VL-32-NEXT:    flds (%esp)
; AVX512VL-32-NEXT:    fisttpll (%esp)
; AVX512VL-32-NEXT:    setae %dl
; AVX512VL-32-NEXT:    shll $31, %edx
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %edx
; AVX512VL-32-NEXT:    vpermilpd {{.*#+}} xmm3 = xmm3[1,0]
; AVX512VL-32-NEXT:    xorl %eax, %eax
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm3
; AVX512VL-32-NEXT:    setb %cl
; AVX512VL-32-NEXT:    kmovw %ecx, %k1
; AVX512VL-32-NEXT:    vmovaps %xmm1, %xmm4
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm4, %xmm4 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm4, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vmovss %xmm3, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    setae %al
; AVX512VL-32-NEXT:    shll $31, %eax
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %eax
; AVX512VL-32-NEXT:    xorl %ecx, %ecx
; AVX512VL-32-NEXT:    vcomiss %xmm1, %xmm0
; AVX512VL-32-NEXT:    setb %bl
; AVX512VL-32-NEXT:    kmovw %ebx, %k1
; AVX512VL-32-NEXT:    vmovss %xmm2, %xmm1, %xmm1 {%k1}
; AVX512VL-32-NEXT:    vsubss %xmm1, %xmm0, %xmm0
; AVX512VL-32-NEXT:    vmovss %xmm0, {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    flds {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    fisttpll {{[0-9]+}}(%esp)
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, %eax, %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $2, (%esp), %xmm0, %xmm0
; AVX512VL-32-NEXT:    vpinsrd $3, %edx, %xmm0, %xmm0
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, %esi, %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm1, %xmm1
; AVX512VL-32-NEXT:    vpinsrd $3, %edi, %xmm1, %xmm1
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm2 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, {{[-0-9]+}}(%e{{[sb]}}p), %xmm2, %xmm2 # 4-byte Folded Reload
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm2, %xmm2
; AVX512VL-32-NEXT:    vpinsrd $3, {{[-0-9]+}}(%e{{[sb]}}p), %xmm2, %xmm2 # 4-byte Folded Reload
; AVX512VL-32-NEXT:    setae %cl
; AVX512VL-32-NEXT:    shll $31, %ecx
; AVX512VL-32-NEXT:    xorl {{[0-9]+}}(%esp), %ecx
; AVX512VL-32-NEXT:    vmovd {{.*#+}} xmm3 = mem[0],zero,zero,zero
; AVX512VL-32-NEXT:    vpinsrd $1, %ecx, %xmm3, %xmm3
; AVX512VL-32-NEXT:    vpinsrd $2, {{[0-9]+}}(%esp), %xmm3, %xmm3
; AVX512VL-32-NEXT:    vpinsrd $3, {{[-0-9]+}}(%e{{[sb]}}p), %xmm3, %xmm3 # 4-byte Folded Reload
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
; AVX512VL-32-NEXT:    vinserti128 $1, %xmm2, %ymm3, %ymm1
; AVX512VL-32-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
; AVX512VL-32-NEXT:    leal -12(%ebp), %esp
; AVX512VL-32-NEXT:    popl %esi
; AVX512VL-32-NEXT:    popl %edi
; AVX512VL-32-NEXT:    popl %ebx
; AVX512VL-32-NEXT:    popl %ebp
; AVX512VL-32-NEXT:    .cfi_def_cfa %esp, 4
; AVX512VL-32-NEXT:    retl
;
; AVX512VL-64-LABEL: strict_vector_fptoui_v8f32_to_v8i64:
; AVX512VL-64:       # %bb.0:
; AVX512VL-64-NEXT:    vextractf128 $1, %ymm0, %xmm1
; AVX512VL-64-NEXT:    vpermilps {{.*#+}} xmm2 = xmm1[3,1,2,3]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm3 = xmm1[1,0]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm3, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm1[1,1,3,3]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm1, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm1
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm1[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm2, %ymm1, %ymm1
; AVX512VL-64-NEXT:    vpermilps {{.*#+}} xmm2 = xmm0[3,1,2,3]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm2, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm2
; AVX512VL-64-NEXT:    vpermilpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm3, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm3
; AVX512VL-64-NEXT:    vmovshdup {{.*#+}} xmm0 = xmm0[1,1,3,3]
; AVX512VL-64-NEXT:    vcvttss2usi %xmm0, %rax
; AVX512VL-64-NEXT:    vmovq %rax, %xmm0
; AVX512VL-64-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm3[0],xmm0[0]
; AVX512VL-64-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
; AVX512VL-64-NEXT:    vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512VL-64-NEXT:    retq
;
; AVX512DQ-LABEL: strict_vector_fptoui_v8f32_to_v8i64:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttps2uqq %ymm0, %zmm0
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i64> @llvm.experimental.constrained.fptoui.v8i64.v8f32(<8 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i64> %ret
}

define <8 x i32> @strict_vector_fptosi_v8f64_to_v8i32(<8 x double> %a) #0 {
; CHECK-LABEL: strict_vector_fptosi_v8f64_to_v8i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttpd2dq %zmm0, %ymm0
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i32> @llvm.experimental.constrained.fptosi.v8i32.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i32> %ret
}

define <8 x i32> @strict_vector_fptoui_v8f64_to_v8i32(<8 x double> %a) #0 {
; CHECK-LABEL: strict_vector_fptoui_v8f64_to_v8i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttpd2udq %zmm0, %ymm0
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i32> @llvm.experimental.constrained.fptoui.v8i32.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i32> %ret
}

define <8 x i16> @strict_vector_fptosi_v8f64_to_v8i16(<8 x double> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptosi_v8f64_to_v8i16:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512VL-NEXT:    vpmovdw %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptosi_v8f64_to_v8i16:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512DQ-NEXT:    vpmovdw %zmm0, %ymm0
; AVX512DQ-NEXT:    # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i16> @llvm.experimental.constrained.fptosi.v8i16.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i16> %ret
}

define <8 x i16> @strict_vector_fptoui_v8f64_to_v8i16(<8 x double> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptoui_v8f64_to_v8i16:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512VL-NEXT:    vpmovdw %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptoui_v8f64_to_v8i16:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512DQ-NEXT:    vpmovdw %zmm0, %ymm0
; AVX512DQ-NEXT:    # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i16> @llvm.experimental.constrained.fptoui.v8i16.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i16> %ret
}

define <8 x i8> @strict_vector_fptosi_v8f64_to_v8i8(<8 x double> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptosi_v8f64_to_v8i8:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512VL-NEXT:    vpmovdb %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptosi_v8f64_to_v8i8:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512DQ-NEXT:    vpmovdb %zmm0, %xmm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i8> @llvm.experimental.constrained.fptosi.v8i8.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i8> %ret
}

define <8 x i8> @strict_vector_fptoui_v8f64_to_v8i8(<8 x double> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptoui_v8f64_to_v8i8:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512VL-NEXT:    vpmovdb %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptoui_v8f64_to_v8i8:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512DQ-NEXT:    vpmovdb %zmm0, %xmm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i8> @llvm.experimental.constrained.fptoui.v8i8.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i8> %ret
}

define <8 x i1> @strict_vector_fptosi_v8f64_to_v8i1(<8 x double> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptosi_v8f64_to_v8i1:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512VL-NEXT:    vptestmd %ymm0, %ymm0, %k1
; AVX512VL-NEXT:    vpcmpeqd %ymm0, %ymm0, %ymm0
; AVX512VL-NEXT:    vmovdqa32 %ymm0, %ymm0 {%k1} {z}
; AVX512VL-NEXT:    vpmovdw %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptosi_v8f64_to_v8i1:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
; AVX512DQ-NEXT:    vpmovm2d %k0, %zmm0
; AVX512DQ-NEXT:    vpmovdw %zmm0, %ymm0
; AVX512DQ-NEXT:    # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i1> @llvm.experimental.constrained.fptosi.v8i1.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i1> %ret
}

define <8 x i1> @strict_vector_fptoui_v8f64_to_v8i1(<8 x double> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptoui_v8f64_to_v8i1:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512VL-NEXT:    vpslld $31, %ymm0, %ymm0
; AVX512VL-NEXT:    vptestmd %ymm0, %ymm0, %k1
; AVX512VL-NEXT:    vpcmpeqd %ymm0, %ymm0, %ymm0
; AVX512VL-NEXT:    vmovdqa32 %ymm0, %ymm0 {%k1} {z}
; AVX512VL-NEXT:    vpmovdw %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptoui_v8f64_to_v8i1:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttpd2dq %zmm0, %ymm0
; AVX512DQ-NEXT:    vpslld $31, %ymm0, %ymm0
; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
; AVX512DQ-NEXT:    vpmovm2d %k0, %zmm0
; AVX512DQ-NEXT:    vpmovdw %zmm0, %ymm0
; AVX512DQ-NEXT:    # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <8 x i1> @llvm.experimental.constrained.fptoui.v8i1.v8f64(<8 x double> %a,
                                              metadata !"fpexcept.strict")
  ret <8 x i1> %ret
}

define <16 x i32> @strict_vector_fptosi_v16f32_to_v16i32(<16 x float> %a) #0 {
; CHECK-LABEL: strict_vector_fptosi_v16f32_to_v16i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttps2dq %zmm0, %zmm0
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i32> @llvm.experimental.constrained.fptosi.v16i32.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i32> %ret
}

define <16 x i32> @strict_vector_fptoui_v16f32_to_v16i32(<16 x float> %a) #0 {
; CHECK-LABEL: strict_vector_fptoui_v16f32_to_v16i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttps2udq %zmm0, %zmm0
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i32> @llvm.experimental.constrained.fptoui.v16i32.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i32> %ret
}

define <16 x i16> @strict_vector_fptosi_v16f32_to_v16i16(<16 x float> %a) #0 {
; CHECK-LABEL: strict_vector_fptosi_v16f32_to_v16i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttps2dq %zmm0, %zmm0
; CHECK-NEXT:    vpmovdw %zmm0, %ymm0
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i16> @llvm.experimental.constrained.fptosi.v16i16.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i16> %ret
}

define <16 x i16> @strict_vector_fptoui_v16f32_to_v16i16(<16 x float> %a) #0 {
; CHECK-LABEL: strict_vector_fptoui_v16f32_to_v16i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttps2dq %zmm0, %zmm0
; CHECK-NEXT:    vpmovdw %zmm0, %ymm0
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i16> @llvm.experimental.constrained.fptoui.v16i16.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i16> %ret
}

define <16 x i8> @strict_vector_fptosi_v16f32_to_v16i8(<16 x float> %a) #0 {
; CHECK-LABEL: strict_vector_fptosi_v16f32_to_v16i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttps2dq %zmm0, %zmm0
; CHECK-NEXT:    vpmovdb %zmm0, %xmm0
; CHECK-NEXT:    vzeroupper
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i8> @llvm.experimental.constrained.fptosi.v16i8.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i8> %ret
}

define <16 x i8> @strict_vector_fptoui_v16f32_to_v16i8(<16 x float> %a) #0 {
; CHECK-LABEL: strict_vector_fptoui_v16f32_to_v16i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vcvttps2dq %zmm0, %zmm0
; CHECK-NEXT:    vpmovdb %zmm0, %xmm0
; CHECK-NEXT:    vzeroupper
; CHECK-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i8> @llvm.experimental.constrained.fptoui.v16i8.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i8> %ret
}

define <16 x i1> @strict_vector_fptosi_v16f32_to_v16i1(<16 x float> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptosi_v16f32_to_v16i1:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttps2dq %zmm0, %zmm0
; AVX512VL-NEXT:    vptestmd %zmm0, %zmm0, %k1
; AVX512VL-NEXT:    vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
; AVX512VL-NEXT:    vpmovdb %zmm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptosi_v16f32_to_v16i1:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttps2dq %zmm0, %zmm0
; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
; AVX512DQ-NEXT:    vpmovm2d %k0, %zmm0
; AVX512DQ-NEXT:    vpmovdb %zmm0, %xmm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i1> @llvm.experimental.constrained.fptosi.v16i1.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i1> %ret
}

define <16 x i1> @strict_vector_fptoui_v16f32_to_v16i1(<16 x float> %a) #0 {
; AVX512VL-LABEL: strict_vector_fptoui_v16f32_to_v16i1:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vcvttps2dq %zmm0, %zmm0
; AVX512VL-NEXT:    vpslld $31, %zmm0, %zmm0
; AVX512VL-NEXT:    vptestmd %zmm0, %zmm0, %k1
; AVX512VL-NEXT:    vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
; AVX512VL-NEXT:    vpmovdb %zmm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    ret{{[l|q]}}
;
; AVX512DQ-LABEL: strict_vector_fptoui_v16f32_to_v16i1:
; AVX512DQ:       # %bb.0:
; AVX512DQ-NEXT:    vcvttps2dq %zmm0, %zmm0
; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
; AVX512DQ-NEXT:    vpmovm2d %k0, %zmm0
; AVX512DQ-NEXT:    vpmovdb %zmm0, %xmm0
; AVX512DQ-NEXT:    vzeroupper
; AVX512DQ-NEXT:    ret{{[l|q]}}
  %ret = call <16 x i1> @llvm.experimental.constrained.fptoui.v16i1.v16f32(<16 x float> %a,
                                              metadata !"fpexcept.strict")
  ret <16 x i1> %ret
}


attributes #0 = { strictfp }