llvm/test/Transforms/SLPVectorizer/X86/arith-div.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -mtriple=x86_64-unknown -passes=slp-vectorizer -S | FileCheck %s --check-prefix=SSE
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=slm -passes=slp-vectorizer -S | FileCheck %s --check-prefix=SLM
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -mattr=-prefer-128-bit -passes=slp-vectorizer -S | FileCheck %s --check-prefix=AVX
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -mattr=+prefer-128-bit -passes=slp-vectorizer -S | FileCheck %s --check-prefix=SSE
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -mattr=-prefer-128-bit -passes=slp-vectorizer -S | FileCheck %s --check-prefix=AVX
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -mattr=+prefer-128-bit -passes=slp-vectorizer -S | FileCheck %s --check-prefix=SSE
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=knl -passes=slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=-prefer-256-bit -passes=slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=+prefer-256-bit -passes=slp-vectorizer -S | FileCheck %s --check-prefix=AVX

@a64 = common global [8 x i64] zeroinitializer, align 64
@b64 = common global [8 x i64] zeroinitializer, align 64
@c64 = common global [8 x i64] zeroinitializer, align 64
@a32 = common global [16 x i32] zeroinitializer, align 64
@b32 = common global [16 x i32] zeroinitializer, align 64
@c32 = common global [16 x i32] zeroinitializer, align 64
@a16 = common global [32 x i16] zeroinitializer, align 64
@b16 = common global [32 x i16] zeroinitializer, align 64
@c16 = common global [32 x i16] zeroinitializer, align 64
@a8  = common global [64 x i8] zeroinitializer, align 64
@b8  = common global [64 x i8] zeroinitializer, align 64
@c8  = common global [64 x i8] zeroinitializer, align 64

define void @sdiv_v16i32_uniformconst() {
; SSE-LABEL: @sdiv_v16i32_uniformconst(
; SSE-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SSE-NEXT:    [[TMP2:%.*]] = sdiv <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SSE-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP4:%.*]] = sdiv <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP6:%.*]] = sdiv <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SSE-NEXT:    [[TMP8:%.*]] = sdiv <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SSE-NEXT:    ret void
;
; SLM-LABEL: @sdiv_v16i32_uniformconst(
; SLM-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SLM-NEXT:    [[TMP2:%.*]] = sdiv <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SLM-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP4:%.*]] = sdiv <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP6:%.*]] = sdiv <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SLM-NEXT:    [[TMP8:%.*]] = sdiv <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SLM-NEXT:    ret void
;
; AVX-LABEL: @sdiv_v16i32_uniformconst(
; AVX-NEXT:    [[TMP1:%.*]] = load <8 x i32>, ptr @a32, align 4
; AVX-NEXT:    [[TMP2:%.*]] = sdiv <8 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP2]], ptr @c32, align 4
; AVX-NEXT:    [[TMP3:%.*]] = load <8 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; AVX-NEXT:    [[TMP4:%.*]] = sdiv <8 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; AVX-NEXT:    ret void
;
; AVX512-LABEL: @sdiv_v16i32_uniformconst(
; AVX512-NEXT:    [[TMP1:%.*]] = load <16 x i32>, ptr @a32, align 4
; AVX512-NEXT:    [[TMP2:%.*]] = sdiv <16 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX512-NEXT:    store <16 x i32> [[TMP2]], ptr @c32, align 4
; AVX512-NEXT:    ret void
;
  %a0  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 0 ), align 4
  %a1  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 1 ), align 4
  %a2  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 2 ), align 4
  %a3  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 3 ), align 4
  %a4  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4 ), align 4
  %a5  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 5 ), align 4
  %a6  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 6 ), align 4
  %a7  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 7 ), align 4
  %a8  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8 ), align 4
  %a9  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 9 ), align 4
  %a10 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 10), align 4
  %a11 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 11), align 4
  %a12 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
  %a13 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 13), align 4
  %a14 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 14), align 4
  %a15 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 15), align 4
  %r0  = sdiv i32 %a0 , 5
  %r1  = sdiv i32 %a1 , 5
  %r2  = sdiv i32 %a2 , 5
  %r3  = sdiv i32 %a3 , 5
  %r4  = sdiv i32 %a4 , 5
  %r5  = sdiv i32 %a5 , 5
  %r6  = sdiv i32 %a6 , 5
  %r7  = sdiv i32 %a7 , 5
  %r8  = sdiv i32 %a8 , 5
  %r9  = sdiv i32 %a9 , 5
  %r10 = sdiv i32 %a10, 5
  %r11 = sdiv i32 %a11, 5
  %r12 = sdiv i32 %a12, 5
  %r13 = sdiv i32 %a13, 5
  %r14 = sdiv i32 %a14, 5
  %r15 = sdiv i32 %a15, 5
  store i32 %r0 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 0 ), align 4
  store i32 %r1 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 1 ), align 4
  store i32 %r2 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 2 ), align 4
  store i32 %r3 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 3 ), align 4
  store i32 %r4 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4 ), align 4
  store i32 %r5 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 5 ), align 4
  store i32 %r6 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 6 ), align 4
  store i32 %r7 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 7 ), align 4
  store i32 %r8 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8 ), align 4
  store i32 %r9 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 9 ), align 4
  store i32 %r10, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 10), align 4
  store i32 %r11, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 11), align 4
  store i32 %r12, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
  store i32 %r13, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 13), align 4
  store i32 %r14, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 14), align 4
  store i32 %r15, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 15), align 4
  ret void
}

define void @srem_v16i32_uniformconst() {
; SSE-LABEL: @srem_v16i32_uniformconst(
; SSE-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SSE-NEXT:    [[TMP2:%.*]] = srem <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SSE-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP4:%.*]] = srem <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP6:%.*]] = srem <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SSE-NEXT:    [[TMP8:%.*]] = srem <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SSE-NEXT:    ret void
;
; SLM-LABEL: @srem_v16i32_uniformconst(
; SLM-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SLM-NEXT:    [[TMP2:%.*]] = srem <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SLM-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP4:%.*]] = srem <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP6:%.*]] = srem <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SLM-NEXT:    [[TMP8:%.*]] = srem <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SLM-NEXT:    ret void
;
; AVX-LABEL: @srem_v16i32_uniformconst(
; AVX-NEXT:    [[TMP1:%.*]] = load <8 x i32>, ptr @a32, align 4
; AVX-NEXT:    [[TMP2:%.*]] = srem <8 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP2]], ptr @c32, align 4
; AVX-NEXT:    [[TMP3:%.*]] = load <8 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; AVX-NEXT:    [[TMP4:%.*]] = srem <8 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; AVX-NEXT:    ret void
;
; AVX512-LABEL: @srem_v16i32_uniformconst(
; AVX512-NEXT:    [[TMP1:%.*]] = load <16 x i32>, ptr @a32, align 4
; AVX512-NEXT:    [[TMP2:%.*]] = srem <16 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX512-NEXT:    store <16 x i32> [[TMP2]], ptr @c32, align 4
; AVX512-NEXT:    ret void
;
  %a0  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 0 ), align 4
  %a1  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 1 ), align 4
  %a2  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 2 ), align 4
  %a3  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 3 ), align 4
  %a4  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4 ), align 4
  %a5  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 5 ), align 4
  %a6  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 6 ), align 4
  %a7  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 7 ), align 4
  %a8  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8 ), align 4
  %a9  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 9 ), align 4
  %a10 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 10), align 4
  %a11 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 11), align 4
  %a12 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
  %a13 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 13), align 4
  %a14 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 14), align 4
  %a15 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 15), align 4
  %r0  = srem i32 %a0 , 5
  %r1  = srem i32 %a1 , 5
  %r2  = srem i32 %a2 , 5
  %r3  = srem i32 %a3 , 5
  %r4  = srem i32 %a4 , 5
  %r5  = srem i32 %a5 , 5
  %r6  = srem i32 %a6 , 5
  %r7  = srem i32 %a7 , 5
  %r8  = srem i32 %a8 , 5
  %r9  = srem i32 %a9 , 5
  %r10 = srem i32 %a10, 5
  %r11 = srem i32 %a11, 5
  %r12 = srem i32 %a12, 5
  %r13 = srem i32 %a13, 5
  %r14 = srem i32 %a14, 5
  %r15 = srem i32 %a15, 5
  store i32 %r0 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 0 ), align 4
  store i32 %r1 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 1 ), align 4
  store i32 %r2 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 2 ), align 4
  store i32 %r3 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 3 ), align 4
  store i32 %r4 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4 ), align 4
  store i32 %r5 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 5 ), align 4
  store i32 %r6 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 6 ), align 4
  store i32 %r7 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 7 ), align 4
  store i32 %r8 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8 ), align 4
  store i32 %r9 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 9 ), align 4
  store i32 %r10, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 10), align 4
  store i32 %r11, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 11), align 4
  store i32 %r12, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
  store i32 %r13, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 13), align 4
  store i32 %r14, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 14), align 4
  store i32 %r15, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 15), align 4
  ret void
}

define void @udiv_v16i32_uniformconst() {
; SSE-LABEL: @udiv_v16i32_uniformconst(
; SSE-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SSE-NEXT:    [[TMP2:%.*]] = udiv <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SSE-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP4:%.*]] = udiv <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP6:%.*]] = udiv <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SSE-NEXT:    [[TMP8:%.*]] = udiv <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SSE-NEXT:    ret void
;
; SLM-LABEL: @udiv_v16i32_uniformconst(
; SLM-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SLM-NEXT:    [[TMP2:%.*]] = udiv <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SLM-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP4:%.*]] = udiv <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP6:%.*]] = udiv <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SLM-NEXT:    [[TMP8:%.*]] = udiv <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SLM-NEXT:    ret void
;
; AVX-LABEL: @udiv_v16i32_uniformconst(
; AVX-NEXT:    [[TMP1:%.*]] = load <8 x i32>, ptr @a32, align 4
; AVX-NEXT:    [[TMP2:%.*]] = udiv <8 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP2]], ptr @c32, align 4
; AVX-NEXT:    [[TMP3:%.*]] = load <8 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; AVX-NEXT:    [[TMP4:%.*]] = udiv <8 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; AVX-NEXT:    ret void
;
; AVX512-LABEL: @udiv_v16i32_uniformconst(
; AVX512-NEXT:    [[TMP1:%.*]] = load <16 x i32>, ptr @a32, align 4
; AVX512-NEXT:    [[TMP2:%.*]] = udiv <16 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX512-NEXT:    store <16 x i32> [[TMP2]], ptr @c32, align 4
; AVX512-NEXT:    ret void
;
  %a0  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 0 ), align 4
  %a1  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 1 ), align 4
  %a2  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 2 ), align 4
  %a3  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 3 ), align 4
  %a4  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4 ), align 4
  %a5  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 5 ), align 4
  %a6  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 6 ), align 4
  %a7  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 7 ), align 4
  %a8  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8 ), align 4
  %a9  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 9 ), align 4
  %a10 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 10), align 4
  %a11 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 11), align 4
  %a12 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
  %a13 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 13), align 4
  %a14 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 14), align 4
  %a15 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 15), align 4
  %r0  = udiv i32 %a0 , 5
  %r1  = udiv i32 %a1 , 5
  %r2  = udiv i32 %a2 , 5
  %r3  = udiv i32 %a3 , 5
  %r4  = udiv i32 %a4 , 5
  %r5  = udiv i32 %a5 , 5
  %r6  = udiv i32 %a6 , 5
  %r7  = udiv i32 %a7 , 5
  %r8  = udiv i32 %a8 , 5
  %r9  = udiv i32 %a9 , 5
  %r10 = udiv i32 %a10, 5
  %r11 = udiv i32 %a11, 5
  %r12 = udiv i32 %a12, 5
  %r13 = udiv i32 %a13, 5
  %r14 = udiv i32 %a14, 5
  %r15 = udiv i32 %a15, 5
  store i32 %r0 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 0 ), align 4
  store i32 %r1 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 1 ), align 4
  store i32 %r2 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 2 ), align 4
  store i32 %r3 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 3 ), align 4
  store i32 %r4 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4 ), align 4
  store i32 %r5 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 5 ), align 4
  store i32 %r6 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 6 ), align 4
  store i32 %r7 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 7 ), align 4
  store i32 %r8 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8 ), align 4
  store i32 %r9 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 9 ), align 4
  store i32 %r10, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 10), align 4
  store i32 %r11, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 11), align 4
  store i32 %r12, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
  store i32 %r13, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 13), align 4
  store i32 %r14, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 14), align 4
  store i32 %r15, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 15), align 4
  ret void
}

define void @urem_v16i32_uniformconst() {
; SSE-LABEL: @urem_v16i32_uniformconst(
; SSE-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SSE-NEXT:    [[TMP2:%.*]] = urem <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SSE-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP4:%.*]] = urem <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SSE-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP6:%.*]] = urem <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SSE-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SSE-NEXT:    [[TMP8:%.*]] = urem <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SSE-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SSE-NEXT:    ret void
;
; SLM-LABEL: @urem_v16i32_uniformconst(
; SLM-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @a32, align 4
; SLM-NEXT:    [[TMP2:%.*]] = urem <4 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP2]], ptr @c32, align 4
; SLM-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP4:%.*]] = urem <4 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4), align 4
; SLM-NEXT:    [[TMP5:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP6:%.*]] = urem <4 x i32> [[TMP5]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP6]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; SLM-NEXT:    [[TMP7:%.*]] = load <4 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
; SLM-NEXT:    [[TMP8:%.*]] = urem <4 x i32> [[TMP7]], <i32 5, i32 5, i32 5, i32 5>
; SLM-NEXT:    store <4 x i32> [[TMP8]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
; SLM-NEXT:    ret void
;
; AVX-LABEL: @urem_v16i32_uniformconst(
; AVX-NEXT:    [[TMP1:%.*]] = load <8 x i32>, ptr @a32, align 4
; AVX-NEXT:    [[TMP2:%.*]] = urem <8 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP2]], ptr @c32, align 4
; AVX-NEXT:    [[TMP3:%.*]] = load <8 x i32>, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8), align 4
; AVX-NEXT:    [[TMP4:%.*]] = urem <8 x i32> [[TMP3]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX-NEXT:    store <8 x i32> [[TMP4]], ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8), align 4
; AVX-NEXT:    ret void
;
; AVX512-LABEL: @urem_v16i32_uniformconst(
; AVX512-NEXT:    [[TMP1:%.*]] = load <16 x i32>, ptr @a32, align 4
; AVX512-NEXT:    [[TMP2:%.*]] = urem <16 x i32> [[TMP1]], <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
; AVX512-NEXT:    store <16 x i32> [[TMP2]], ptr @c32, align 4
; AVX512-NEXT:    ret void
;
  %a0  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 0 ), align 4
  %a1  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 1 ), align 4
  %a2  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 2 ), align 4
  %a3  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 3 ), align 4
  %a4  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 4 ), align 4
  %a5  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 5 ), align 4
  %a6  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 6 ), align 4
  %a7  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 7 ), align 4
  %a8  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 8 ), align 4
  %a9  = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 9 ), align 4
  %a10 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 10), align 4
  %a11 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 11), align 4
  %a12 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 12), align 4
  %a13 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 13), align 4
  %a14 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 14), align 4
  %a15 = load i32, ptr getelementptr inbounds ([16 x i32], ptr @a32, i32 0, i64 15), align 4
  %r0  = urem i32 %a0 , 5
  %r1  = urem i32 %a1 , 5
  %r2  = urem i32 %a2 , 5
  %r3  = urem i32 %a3 , 5
  %r4  = urem i32 %a4 , 5
  %r5  = urem i32 %a5 , 5
  %r6  = urem i32 %a6 , 5
  %r7  = urem i32 %a7 , 5
  %r8  = urem i32 %a8 , 5
  %r9  = urem i32 %a9 , 5
  %r10 = urem i32 %a10, 5
  %r11 = urem i32 %a11, 5
  %r12 = urem i32 %a12, 5
  %r13 = urem i32 %a13, 5
  %r14 = urem i32 %a14, 5
  %r15 = urem i32 %a15, 5
  store i32 %r0 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 0 ), align 4
  store i32 %r1 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 1 ), align 4
  store i32 %r2 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 2 ), align 4
  store i32 %r3 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 3 ), align 4
  store i32 %r4 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 4 ), align 4
  store i32 %r5 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 5 ), align 4
  store i32 %r6 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 6 ), align 4
  store i32 %r7 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 7 ), align 4
  store i32 %r8 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 8 ), align 4
  store i32 %r9 , ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 9 ), align 4
  store i32 %r10, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 10), align 4
  store i32 %r11, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 11), align 4
  store i32 %r12, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 12), align 4
  store i32 %r13, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 13), align 4
  store i32 %r14, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 14), align 4
  store i32 %r15, ptr getelementptr inbounds ([16 x i32], ptr @c32, i32 0, i64 15), align 4
  ret void
}