derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
1a182e2dac58879f215c0fad0f5f78c76dcafe56
TheRock/compiler/amd-llvm/clang/test/OpenMP/xteam_red_split_codegen.cpp
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
Details
CodeQL / Analyze (python) (push) Canceled after 0s
Details
pre-commit / pre-commit (push) Canceled after 0s
Details
Multi-Arch CI / setup (push) Canceled after 0s
Details
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Details
Setup (Multi-Arch) / setup (push) Canceled after 0s
Details
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Details
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
Details
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Details
Multi-Arch CI / CI Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Details
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Details
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

1328 lines
93 KiB
C++
Raw Blame History

// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --function-signature --include-generated-funcs --replace-value-regex "__omp_offloading_[0-9a-z]+_[0-9a-z]+" "reduction_size[.].+[.]" "pl_cond[.].+[.|,]" --prefix-filecheck-ir-name _
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-unknown-unknown -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -emit-llvm-bc %s -o %t-ppc-host.bc
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s
// expected-no-diagnostics
int main()
{
int N = 1000;
double a[N];
for (int i=0; i<N; i++)
a[i]=i;
double sum1, sum2, sum3, sum4;
sum1 = sum2 = sum3 = sum4 = 0;
#pragma omp target teams map(tofrom:sum1)
#pragma omp distribute parallel for reduction(+:sum1)
{
for (int k = 0; k< N; k++) {
sum1 += a[k];
}
}
#pragma omp target teams map(tofrom:sum1) thread_limit(64)
#pragma omp distribute parallel for reduction(+:sum1)
{
for (int k = 0; k< N; k++) {
sum1 += a[k];
}
}
#pragma omp target map(tofrom:sum1)
#pragma omp teams
#pragma omp distribute parallel for reduction(+:sum1)
{
{
for (int k = 0; k< N; k++) {
sum1 += a[k];
}
}
}
#pragma omp target map(tofrom:sum1)
#pragma omp teams
#pragma omp distribute parallel for reduction(+:sum1) num_threads(128)
{
{
for (int k = 0; k< N; k++) {
sum1 += a[k];
}
}
}
#pragma omp target map(tofrom:sum1)
#pragma omp teams
#pragma omp distribute parallel for reduction(+:sum1)
for (int k = 0; k< N; k++) {
sum1 += a[k];
}
#pragma omp target map(tofrom:sum1)
#pragma omp teams
#pragma omp distribute parallel for reduction(+:sum1) num_threads(128)
for (int k = 0; k< N; k++) {
sum1 += a[k];
}
#pragma omp target map(tofrom:sum1)
#pragma omp teams distribute parallel for reduction(+:sum1)
for (int j = 0; j< N; j=j+1)
sum1 += a[j];
#pragma omp target map(tofrom:sum1)
#pragma omp teams distribute parallel for reduction(+:sum1) thread_limit(512) num_threads(128)
for (int j = 0; j< N; j=j+1)
sum1 += a[j];
#pragma omp target map(tofrom:sum1)
{
#pragma omp teams distribute parallel for reduction(+:sum1)
for (int j = 0; j< N; j=j+1)
sum1 += a[j];
}
#pragma omp target map(tofrom:sum1)
{
#pragma omp teams distribute parallel for reduction(+:sum1) thread_limit(512) num_threads(256)
for (int j = 0; j< N; j=j+1)
sum1 += a[j];
}
#pragma omp target teams distribute parallel for map(tofrom:sum2) reduction(+:sum2)
for (int j = 0; j< N; j=j+2)
sum2 += a[j];
#pragma omp target teams distribute parallel for map(tofrom:sum2) reduction(+:sum2) thread_limit(512) num_threads(128)
for (int j = 0; j< N; j=j+2)
sum2 += a[j];
}
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l18
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[K:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[K_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[K]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16:![0-9]+]], !align [[META17:![0-9]+]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP18:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l26
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR3:[0-9]+]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[K:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[K_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[K]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP20:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l34
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[K:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[K_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[K]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP21:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l45
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR4:[0-9]+]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[K:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[K_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[K]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP22:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l56
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[K:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[K_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[K]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP23:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l63
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR4]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[K:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[K_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[K]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[K_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP24:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l70
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[J:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[J_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[J]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP25:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l75
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR4]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[J:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[J_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[J]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP26:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l80
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[J:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[J_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[J]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP27:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l87
// CHECK-SAME: (ptr noundef nonnull align 8 dereferenceable(8) [[SUM1:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR5:[0-9]+]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SUM1_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[J:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[SUM1_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM1_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[J_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[J]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store ptr [[SUM1]], ptr [[SUM1_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM1_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub nsw i32 [[TMP7]], 0
// CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUB]], 1
// CHECK-NEXT: [[SUB3:%.*]] = sub nsw i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[CMP:%.*]] = icmp sle i32 [[TMP18]], [[TMP19]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul nsw i32 [[TMP20]], 1
// CHECK-NEXT: [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD]], ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS4:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS4]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP28:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l94
// CHECK-SAME: (i64 noundef [[N:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[SUM2:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[SUM2_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[J:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[SUM2_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM2_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[J_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[J]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[SUM2]], ptr [[SUM2_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM2_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub i32 [[TMP7]], -1
// CHECK-NEXT: [[DIV:%.*]] = udiv i32 [[SUB]], 2
// CHECK-NEXT: [[SUB3:%.*]] = sub i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[ADD:%.*]] = add i32 [[TMP19]], 1
// CHECK-NEXT: [[CMP:%.*]] = icmp ult i32 [[TMP18]], [[ADD]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul i32 [[TMP20]], 2
// CHECK-NEXT: [[ADD4:%.*]] = add i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD4]], ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS5:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS5]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP29:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_main_l98
// CHECK-SAME: (i64 noundef [[N:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[SUM2:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 8 dereferenceable(8) [[A:%.*]], ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]], ptr noalias noundef [[DYN_PTR:%.*]]) #[[ATTR4]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[SUM2_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DOTADDR1:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[J:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[SUM2_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM2_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
// CHECK-NEXT: [[DOTADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR]] to ptr
// CHECK-NEXT: [[DOTADDR1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTADDR1]] to ptr
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[J_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[J]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_]] to ptr
// CHECK-NEXT: [[DOTCAPTURE_EXPR_2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTCAPTURE_EXPR_2]] to ptr
// CHECK-NEXT: [[DOTOMP_LB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_LB]] to ptr
// CHECK-NEXT: [[DOTOMP_UB_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_UB]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[SUM2]], ptr [[SUM2_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[A]], ptr [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP0]], ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[TMP1]], ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[SUM2_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP4:%.*]] = load ptr, ptr [[A_ADDR_ASCAST]], align 8, !nonnull [[META16]], !align [[META17]]
// CHECK-NEXT: call void @__kmpc_specialized_kernel_init()
// CHECK-NEXT: [[TMP5:%.*]] = alloca double, align 8, addrspace(5)
// CHECK-NEXT: store double 0.000000e+00, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[N_ADDR_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP6]], ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR__ASCAST]], align 4
// CHECK-NEXT: [[SUB:%.*]] = sub i32 [[TMP7]], -1
// CHECK-NEXT: [[DIV:%.*]] = udiv i32 [[SUB]], 2
// CHECK-NEXT: [[SUB3:%.*]] = sub i32 [[DIV]], 1
// CHECK-NEXT: store i32 [[SUB3]], ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: store i32 0, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[DOTCAPTURE_EXPR_2_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP8]], ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB_ASCAST]], align 4
// CHECK-NEXT: store i32 [[TMP9]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_get_hardware_thread_id_in_block()
// CHECK-NEXT: [[NVPTX_NUM_THREADS:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[GPU_BLOCK_ID:%.*]] = call i32 @llvm.amdgcn.workgroup.id.x()
// CHECK-NEXT: [[TMP11:%.*]] = mul i32 [[GPU_BLOCK_ID]], [[NVPTX_NUM_THREADS]]
// CHECK-NEXT: [[TMP12:%.*]] = add i32 [[TMP11]], [[TMP10]]
// CHECK-NEXT: [[TMP13:%.*]] = mul i32 [[TMP12]], 1
// CHECK-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP15:%.*]] = add i32 [[TMP13]], [[TMP14]]
// CHECK-NEXT: [[TMP16:%.*]] = call i32 @__kmpc_get_hardware_num_blocks()
// CHECK-NEXT: [[TMP17:%.*]] = zext i32 [[TMP15]] to i64
// CHECK-NEXT: store i32 [[TMP15]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
// CHECK-NEXT: [[TMP18:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP19:%.*]] = load i32, ptr [[DOTOMP_UB_ASCAST]], align 4
// CHECK-NEXT: [[ADD:%.*]] = add i32 [[TMP19]], 1
// CHECK-NEXT: [[CMP:%.*]] = icmp ult i32 [[TMP18]], [[ADD]]
// CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY:%.*]], label [[FOR_END:%.*]]
// CHECK: for.body:
// CHECK-NEXT: [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[MUL:%.*]] = mul i32 [[TMP20]], 2
// CHECK-NEXT: [[ADD4:%.*]] = add i32 0, [[MUL]]
// CHECK-NEXT: store i32 [[ADD4]], ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[TMP21:%.*]] = load i32, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP21]] to i64
// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[TMP4]], i64 [[IDXPROM]]
// CHECK-NEXT: [[TMP22:%.*]] = load double, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = fadd double [[TMP23]], [[TMP22]]
// CHECK-NEXT: store double [[TMP24]], ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: br label [[FOR_INC:%.*]]
// CHECK: for.inc:
// CHECK-NEXT: [[NVPTX_NUM_THREADS5:%.*]] = call i32 @__kmpc_get_hardware_num_threads_in_block()
// CHECK-NEXT: [[TMP25:%.*]] = mul i32 [[NVPTX_NUM_THREADS5]], [[TMP16]]
// CHECK-NEXT: [[TMP26:%.*]] = mul i32 [[TMP25]], 1
// CHECK-NEXT: [[TMP27:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: [[TMP28:%.*]] = add i32 [[TMP26]], [[TMP27]]
// CHECK-NEXT: store i32 [[TMP28]], ptr [[DOTOMP_IV_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND]], !llvm.loop [[LOOP30:![0-9]+]]
// CHECK: for.end:
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTADDR1_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load double, ptr addrspace(5) [[TMP5]], align 8
// CHECK-NEXT: call void @__kmpc_xteamr_d_16x64(double [[TMP31]], ptr [[TMP2]], ptr [[TMP29]], ptr [[TMP30]], ptr @__kmpc_rfun_sum_d, ptr @__kmpc_rfun_sum_lds_d, double 0.000000e+00, i64 [[TMP17]], i32 [[TMP16]], i32 1)
// CHECK-NEXT: ret void
//
Reference in new issue View Git Blame Copy Permalink