b52d33e6de
I used a script to reuse existing check lines rather than creating new ones. There are more opportunities to reduce the line count but the "check generated functions" logic makes that somewhat tricky. FWIW, we really should redo the update script with all these use cases in mind... Differential Revision: https://reviews.llvm.org/D128686
3106 lines
219 KiB
C++
3106 lines
219 KiB
C++
// 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 _
|
|
// Test host codegen.
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK1
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK1
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK3
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK3
|
|
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
|
|
// Test target codegen - host bc file has to be created first.
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --check-prefix=CHECK9
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK9
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck %s --check-prefix=CHECK11
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK11
|
|
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
|
|
// Test host codegen.
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK1
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK1
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK3
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK3
|
|
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
|
|
// Test target codegen - host bc file has to be created first.
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --check-prefix=CHECK9
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK9
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck %s --check-prefix=CHECK11
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK11
|
|
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc
|
|
// RUN: %clang_cc1 -no-opaque-pointers -verify -fopenmp-simd -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o %t %s
|
|
// RUN: %clang_cc1 -no-opaque-pointers -fopenmp-simd -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
|
|
|
|
// expected-no-diagnostics
|
|
#ifndef HEADER
|
|
#define HEADER
|
|
|
|
|
|
|
|
|
|
// We have 8 target regions, but only 6 that actually will generate offloading
|
|
// code and have mapped arguments, and only 4 have all-constant map sizes.
|
|
|
|
|
|
|
|
// Check target registration is registered as a Ctor.
|
|
|
|
|
|
template<typename tx, typename ty>
|
|
struct TT{
|
|
tx X;
|
|
ty Y;
|
|
};
|
|
|
|
int foo(int n) {
|
|
int a = 0;
|
|
short aa = 0;
|
|
float b[10];
|
|
float bn[n];
|
|
double c[5][10];
|
|
double cn[5][n];
|
|
TT<long long, char> d;
|
|
|
|
#pragma omp target parallel nowait
|
|
{
|
|
}
|
|
|
|
#pragma omp target parallel if(target: 0)
|
|
{
|
|
a += 1;
|
|
}
|
|
|
|
|
|
#pragma omp target parallel if(target: 1)
|
|
{
|
|
aa += 1;
|
|
#pragma omp cancel parallel
|
|
}
|
|
|
|
|
|
|
|
|
|
#pragma omp target parallel if(target: n>10)
|
|
{
|
|
a += 1;
|
|
aa += 1;
|
|
}
|
|
|
|
// We capture 3 VLA sizes in this target region
|
|
|
|
|
|
|
|
|
|
|
|
// The names below are not necessarily consistent with the names used for the
|
|
// addresses above as some are repeated.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
#pragma omp target parallel if(target: n>20)
|
|
{
|
|
a += 1;
|
|
b[2] += 1.0;
|
|
bn[3] += 1.0;
|
|
c[1][2] += 1.0;
|
|
cn[1][3] += 1.0;
|
|
d.X += 1;
|
|
d.Y += 1;
|
|
}
|
|
|
|
return a;
|
|
}
|
|
|
|
// Check that the offloading functions are emitted and that the arguments are
|
|
// correct and loaded correctly for the target regions in foo().
|
|
|
|
|
|
|
|
// Create stack storage and store argument in there.
|
|
|
|
// Create stack storage and store argument in there.
|
|
|
|
// Create stack storage and store argument in there.
|
|
|
|
// Create local storage for each capture.
|
|
|
|
|
|
|
|
// To reduce complexity, we're only going as far as validating the signature of the outlined parallel function.
|
|
|
|
template<typename tx>
|
|
tx ftemplate(int n) {
|
|
tx a = 0;
|
|
short aa = 0;
|
|
tx b[10];
|
|
|
|
#pragma omp target parallel if(target: n>40)
|
|
{
|
|
a += 1;
|
|
aa += 1;
|
|
b[2] += 1;
|
|
}
|
|
|
|
return a;
|
|
}
|
|
|
|
static
|
|
int fstatic(int n) {
|
|
int a = 0;
|
|
short aa = 0;
|
|
char aaa = 0;
|
|
int b[10];
|
|
|
|
#pragma omp target parallel if(target: n>50)
|
|
{
|
|
a += 1;
|
|
aa += 1;
|
|
aaa += 1;
|
|
b[2] += 1;
|
|
}
|
|
|
|
return a;
|
|
}
|
|
|
|
struct S1 {
|
|
double a;
|
|
|
|
int r1(int n){
|
|
int b = n+1;
|
|
short int c[2][n];
|
|
|
|
#pragma omp target parallel if(target: n>60)
|
|
{
|
|
this->a = (double)b + 1.5;
|
|
c[1][1] = ++a;
|
|
}
|
|
|
|
return c[1][1] + (int)b;
|
|
}
|
|
};
|
|
|
|
int bar(int n){
|
|
int a = 0;
|
|
|
|
a += foo(n);
|
|
|
|
S1 S;
|
|
a += S.r1(n);
|
|
|
|
a += fstatic(n);
|
|
|
|
a += ftemplate<int>(n);
|
|
|
|
return a;
|
|
}
|
|
|
|
|
|
|
|
// We capture 2 VLA sizes in this target region
|
|
|
|
|
|
// The names below are not necessarily consistent with the names used for the
|
|
// addresses above as some are repeated.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
// Check that the offloading functions are emitted and that the arguments are
|
|
// correct and loaded correctly for the target regions of the callees of bar().
|
|
|
|
// Create local storage for each capture.
|
|
// Store captures in the context.
|
|
|
|
|
|
// To reduce complexity, we're only going as far as validating the signature of the outlined parallel function.
|
|
|
|
|
|
// Create local storage for each capture.
|
|
// Store captures in the context.
|
|
|
|
|
|
|
|
|
|
// To reduce complexity, we're only going as far as validating the signature of the outlined parallel function.
|
|
|
|
// Create local storage for each capture.
|
|
// Store captures in the context.
|
|
|
|
|
|
|
|
// To reduce complexity, we're only going as far as validating the signature of the outlined parallel function.
|
|
|
|
|
|
#endif
|
|
// CHECK1-LABEL: define {{[^@]+}}@_Z3fooi
|
|
// CHECK1-SAME: (i32 noundef signext [[N:%.*]]) #[[ATTR0:[0-9]+]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[AA:%.*]] = alloca i16, align 2
|
|
// CHECK1-NEXT: [[B:%.*]] = alloca [10 x float], align 4
|
|
// CHECK1-NEXT: [[SAVED_STACK:%.*]] = alloca i8*, align 8
|
|
// CHECK1-NEXT: [[__VLA_EXPR0:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[C:%.*]] = alloca [5 x [10 x double]], align 8
|
|
// CHECK1-NEXT: [[__VLA_EXPR1:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[D:%.*]] = alloca [[STRUCT_TT:%.*]], align 8
|
|
// CHECK1-NEXT: [[AGG_CAPTURED:%.*]] = alloca [[STRUCT_ANON:%.*]], align 1
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [1 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [1 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [1 x i8*], align 8
|
|
// CHECK1-NEXT: [[A_CASTED3:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED5:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_BASEPTRS7:%.*]] = alloca [2 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_PTRS8:%.*]] = alloca [2 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_MAPPERS9:%.*]] = alloca [2 x i8*], align 8
|
|
// CHECK1-NEXT: [[A_CASTED12:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_BASEPTRS16:%.*]] = alloca [9 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_PTRS17:%.*]] = alloca [9 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_MAPPERS18:%.*]] = alloca [9 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_SIZES:%.*]] = alloca [9 x i64], align 8
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]])
|
|
// CHECK1-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK1-NEXT: store i16 0, i16* [[AA]], align 2
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = call i8* @llvm.stacksave()
|
|
// CHECK1-NEXT: store i8* [[TMP3]], i8** [[SAVED_STACK]], align 8
|
|
// CHECK1-NEXT: [[VLA:%.*]] = alloca float, i64 [[TMP2]], align 4
|
|
// CHECK1-NEXT: store i64 [[TMP2]], i64* [[__VLA_EXPR0]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = zext i32 [[TMP4]] to i64
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = mul nuw i64 5, [[TMP5]]
|
|
// CHECK1-NEXT: [[VLA1:%.*]] = alloca double, i64 [[TMP6]], align 8
|
|
// CHECK1-NEXT: store i64 [[TMP5]], i64* [[__VLA_EXPR1]], align 8
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = call i8* @__kmpc_omp_target_task_alloc(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]], i32 1, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, %struct.kmp_task_t_with_privates*)* @.omp_task_entry. to i32 (i32, i8*)*), i64 -1)
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = bitcast i8* [[TMP7]] to %struct.kmp_task_t_with_privates*
|
|
// CHECK1-NEXT: [[TMP9:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T_WITH_PRIVATES:%.*]], %struct.kmp_task_t_with_privates* [[TMP8]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP10:%.*]] = call i32 @__kmpc_omp_task(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]], i8* [[TMP7]])
|
|
// CHECK1-NEXT: [[TMP11:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP11]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[TMP12:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l104(i64 [[TMP12]]) #[[ATTR4:[0-9]+]]
|
|
// CHECK1-NEXT: [[TMP13:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP13]], i16* [[CONV2]], align 2
|
|
// CHECK1-NEXT: [[TMP14:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP15:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP16:%.*]] = bitcast i8** [[TMP15]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP14]], i64* [[TMP16]], align 8
|
|
// CHECK1-NEXT: [[TMP17:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP18:%.*]] = bitcast i8** [[TMP17]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP14]], i64* [[TMP18]], align 8
|
|
// CHECK1-NEXT: [[TMP19:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 0
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP19]], align 8
|
|
// CHECK1-NEXT: [[TMP20:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP21:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP22:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110.region_id, i32 1, i8** [[TMP20]], i8** [[TMP21]], i64* getelementptr inbounds ([1 x i64], [1 x i64]* @.offload_sizes, i32 0, i32 0), i64* getelementptr inbounds ([1 x i64], [1 x i64]* @.offload_maptypes, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK1-NEXT: [[TMP23:%.*]] = icmp ne i32 [[TMP22]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP23]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK1: omp_offload.failed:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110(i64 [[TMP14]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK1: omp_offload.cont:
|
|
// CHECK1-NEXT: [[TMP24:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[CONV4:%.*]] = bitcast i64* [[A_CASTED3]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP24]], i32* [[CONV4]], align 4
|
|
// CHECK1-NEXT: [[TMP25:%.*]] = load i64, i64* [[A_CASTED3]], align 8
|
|
// CHECK1-NEXT: [[TMP26:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK1-NEXT: [[CONV6:%.*]] = bitcast i64* [[AA_CASTED5]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP26]], i16* [[CONV6]], align 2
|
|
// CHECK1-NEXT: [[TMP27:%.*]] = load i64, i64* [[AA_CASTED5]], align 8
|
|
// CHECK1-NEXT: [[TMP28:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP28]], 10
|
|
// CHECK1-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK1: omp_if.then:
|
|
// CHECK1-NEXT: [[TMP29:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_BASEPTRS7]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP30:%.*]] = bitcast i8** [[TMP29]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP25]], i64* [[TMP30]], align 8
|
|
// CHECK1-NEXT: [[TMP31:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_PTRS8]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP32:%.*]] = bitcast i8** [[TMP31]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP25]], i64* [[TMP32]], align 8
|
|
// CHECK1-NEXT: [[TMP33:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_MAPPERS9]], i64 0, i64 0
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP33]], align 8
|
|
// CHECK1-NEXT: [[TMP34:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_BASEPTRS7]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP35:%.*]] = bitcast i8** [[TMP34]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP27]], i64* [[TMP35]], align 8
|
|
// CHECK1-NEXT: [[TMP36:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_PTRS8]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP37:%.*]] = bitcast i8** [[TMP36]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP27]], i64* [[TMP37]], align 8
|
|
// CHECK1-NEXT: [[TMP38:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_MAPPERS9]], i64 0, i64 1
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP38]], align 8
|
|
// CHECK1-NEXT: [[TMP39:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_BASEPTRS7]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP40:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_PTRS8]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP41:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119.region_id, i32 2, i8** [[TMP39]], i8** [[TMP40]], i64* getelementptr inbounds ([2 x i64], [2 x i64]* @.offload_sizes.5, i32 0, i32 0), i64* getelementptr inbounds ([2 x i64], [2 x i64]* @.offload_maptypes.6, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK1-NEXT: [[TMP42:%.*]] = icmp ne i32 [[TMP41]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP42]], label [[OMP_OFFLOAD_FAILED10:%.*]], label [[OMP_OFFLOAD_CONT11:%.*]]
|
|
// CHECK1: omp_offload.failed10:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119(i64 [[TMP25]], i64 [[TMP27]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT11]]
|
|
// CHECK1: omp_offload.cont11:
|
|
// CHECK1-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK1: omp_if.else:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119(i64 [[TMP25]], i64 [[TMP27]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK1: omp_if.end:
|
|
// CHECK1-NEXT: [[TMP43:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[CONV13:%.*]] = bitcast i64* [[A_CASTED12]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP43]], i32* [[CONV13]], align 4
|
|
// CHECK1-NEXT: [[TMP44:%.*]] = load i64, i64* [[A_CASTED12]], align 8
|
|
// CHECK1-NEXT: [[TMP45:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CMP14:%.*]] = icmp sgt i32 [[TMP45]], 20
|
|
// CHECK1-NEXT: br i1 [[CMP14]], label [[OMP_IF_THEN15:%.*]], label [[OMP_IF_ELSE21:%.*]]
|
|
// CHECK1: omp_if.then15:
|
|
// CHECK1-NEXT: [[TMP46:%.*]] = mul nuw i64 [[TMP2]], 4
|
|
// CHECK1-NEXT: [[TMP47:%.*]] = mul nuw i64 5, [[TMP5]]
|
|
// CHECK1-NEXT: [[TMP48:%.*]] = mul nuw i64 [[TMP47]], 8
|
|
// CHECK1-NEXT: [[TMP49:%.*]] = bitcast [9 x i64]* [[DOTOFFLOAD_SIZES]] to i8*
|
|
// CHECK1-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP49]], i8* align 8 bitcast ([9 x i64]* @.offload_sizes.8 to i8*), i64 72, i1 false)
|
|
// CHECK1-NEXT: [[TMP50:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP51:%.*]] = bitcast i8** [[TMP50]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP44]], i64* [[TMP51]], align 8
|
|
// CHECK1-NEXT: [[TMP52:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP53:%.*]] = bitcast i8** [[TMP52]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP44]], i64* [[TMP53]], align 8
|
|
// CHECK1-NEXT: [[TMP54:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 0
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP54]], align 8
|
|
// CHECK1-NEXT: [[TMP55:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP56:%.*]] = bitcast i8** [[TMP55]] to [10 x float]**
|
|
// CHECK1-NEXT: store [10 x float]* [[B]], [10 x float]** [[TMP56]], align 8
|
|
// CHECK1-NEXT: [[TMP57:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP58:%.*]] = bitcast i8** [[TMP57]] to [10 x float]**
|
|
// CHECK1-NEXT: store [10 x float]* [[B]], [10 x float]** [[TMP58]], align 8
|
|
// CHECK1-NEXT: [[TMP59:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 1
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP59]], align 8
|
|
// CHECK1-NEXT: [[TMP60:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP61:%.*]] = bitcast i8** [[TMP60]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP2]], i64* [[TMP61]], align 8
|
|
// CHECK1-NEXT: [[TMP62:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP63:%.*]] = bitcast i8** [[TMP62]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP2]], i64* [[TMP63]], align 8
|
|
// CHECK1-NEXT: [[TMP64:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 2
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP64]], align 8
|
|
// CHECK1-NEXT: [[TMP65:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 3
|
|
// CHECK1-NEXT: [[TMP66:%.*]] = bitcast i8** [[TMP65]] to float**
|
|
// CHECK1-NEXT: store float* [[VLA]], float** [[TMP66]], align 8
|
|
// CHECK1-NEXT: [[TMP67:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 3
|
|
// CHECK1-NEXT: [[TMP68:%.*]] = bitcast i8** [[TMP67]] to float**
|
|
// CHECK1-NEXT: store float* [[VLA]], float** [[TMP68]], align 8
|
|
// CHECK1-NEXT: [[TMP69:%.*]] = getelementptr inbounds [9 x i64], [9 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 3
|
|
// CHECK1-NEXT: store i64 [[TMP46]], i64* [[TMP69]], align 8
|
|
// CHECK1-NEXT: [[TMP70:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 3
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP70]], align 8
|
|
// CHECK1-NEXT: [[TMP71:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 4
|
|
// CHECK1-NEXT: [[TMP72:%.*]] = bitcast i8** [[TMP71]] to [5 x [10 x double]]**
|
|
// CHECK1-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[TMP72]], align 8
|
|
// CHECK1-NEXT: [[TMP73:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 4
|
|
// CHECK1-NEXT: [[TMP74:%.*]] = bitcast i8** [[TMP73]] to [5 x [10 x double]]**
|
|
// CHECK1-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[TMP74]], align 8
|
|
// CHECK1-NEXT: [[TMP75:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 4
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP75]], align 8
|
|
// CHECK1-NEXT: [[TMP76:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 5
|
|
// CHECK1-NEXT: [[TMP77:%.*]] = bitcast i8** [[TMP76]] to i64*
|
|
// CHECK1-NEXT: store i64 5, i64* [[TMP77]], align 8
|
|
// CHECK1-NEXT: [[TMP78:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 5
|
|
// CHECK1-NEXT: [[TMP79:%.*]] = bitcast i8** [[TMP78]] to i64*
|
|
// CHECK1-NEXT: store i64 5, i64* [[TMP79]], align 8
|
|
// CHECK1-NEXT: [[TMP80:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 5
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP80]], align 8
|
|
// CHECK1-NEXT: [[TMP81:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 6
|
|
// CHECK1-NEXT: [[TMP82:%.*]] = bitcast i8** [[TMP81]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP5]], i64* [[TMP82]], align 8
|
|
// CHECK1-NEXT: [[TMP83:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 6
|
|
// CHECK1-NEXT: [[TMP84:%.*]] = bitcast i8** [[TMP83]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP5]], i64* [[TMP84]], align 8
|
|
// CHECK1-NEXT: [[TMP85:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 6
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP85]], align 8
|
|
// CHECK1-NEXT: [[TMP86:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 7
|
|
// CHECK1-NEXT: [[TMP87:%.*]] = bitcast i8** [[TMP86]] to double**
|
|
// CHECK1-NEXT: store double* [[VLA1]], double** [[TMP87]], align 8
|
|
// CHECK1-NEXT: [[TMP88:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 7
|
|
// CHECK1-NEXT: [[TMP89:%.*]] = bitcast i8** [[TMP88]] to double**
|
|
// CHECK1-NEXT: store double* [[VLA1]], double** [[TMP89]], align 8
|
|
// CHECK1-NEXT: [[TMP90:%.*]] = getelementptr inbounds [9 x i64], [9 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 7
|
|
// CHECK1-NEXT: store i64 [[TMP48]], i64* [[TMP90]], align 8
|
|
// CHECK1-NEXT: [[TMP91:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 7
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP91]], align 8
|
|
// CHECK1-NEXT: [[TMP92:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 8
|
|
// CHECK1-NEXT: [[TMP93:%.*]] = bitcast i8** [[TMP92]] to %struct.TT**
|
|
// CHECK1-NEXT: store %struct.TT* [[D]], %struct.TT** [[TMP93]], align 8
|
|
// CHECK1-NEXT: [[TMP94:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 8
|
|
// CHECK1-NEXT: [[TMP95:%.*]] = bitcast i8** [[TMP94]] to %struct.TT**
|
|
// CHECK1-NEXT: store %struct.TT* [[D]], %struct.TT** [[TMP95]], align 8
|
|
// CHECK1-NEXT: [[TMP96:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS18]], i64 0, i64 8
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP96]], align 8
|
|
// CHECK1-NEXT: [[TMP97:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS16]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP98:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS17]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP99:%.*]] = getelementptr inbounds [9 x i64], [9 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP100:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144.region_id, i32 9, i8** [[TMP97]], i8** [[TMP98]], i64* [[TMP99]], i64* getelementptr inbounds ([9 x i64], [9 x i64]* @.offload_maptypes.9, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK1-NEXT: [[TMP101:%.*]] = icmp ne i32 [[TMP100]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP101]], label [[OMP_OFFLOAD_FAILED19:%.*]], label [[OMP_OFFLOAD_CONT20:%.*]]
|
|
// CHECK1: omp_offload.failed19:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144(i64 [[TMP44]], [10 x float]* [[B]], i64 [[TMP2]], float* [[VLA]], [5 x [10 x double]]* [[C]], i64 5, i64 [[TMP5]], double* [[VLA1]], %struct.TT* [[D]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT20]]
|
|
// CHECK1: omp_offload.cont20:
|
|
// CHECK1-NEXT: br label [[OMP_IF_END22:%.*]]
|
|
// CHECK1: omp_if.else21:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144(i64 [[TMP44]], [10 x float]* [[B]], i64 [[TMP2]], float* [[VLA]], [5 x [10 x double]]* [[C]], i64 5, i64 [[TMP5]], double* [[VLA1]], %struct.TT* [[D]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_IF_END22]]
|
|
// CHECK1: omp_if.end22:
|
|
// CHECK1-NEXT: [[TMP102:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[TMP103:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8
|
|
// CHECK1-NEXT: call void @llvm.stackrestore(i8* [[TMP103]])
|
|
// CHECK1-NEXT: ret i32 [[TMP102]]
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100
|
|
// CHECK1-SAME: () #[[ATTR2:[0-9]+]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @.omp_outlined. to void (i32*, i32*, ...)*))
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined.
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]]) #[[ATTR3:[0-9]+]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_task_entry.
|
|
// CHECK1-SAME: (i32 noundef signext [[TMP0:%.*]], %struct.kmp_task_t_with_privates* noalias noundef [[TMP1:%.*]]) #[[ATTR5:[0-9]+]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR_I:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[DOTPART_ID__ADDR_I:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTPRIVATES__ADDR_I:%.*]] = alloca i8*, align 8
|
|
// CHECK1-NEXT: [[DOTCOPY_FN__ADDR_I:%.*]] = alloca void (i8*, ...)*, align 8
|
|
// CHECK1-NEXT: [[DOTTASK_T__ADDR_I:%.*]] = alloca i8*, align 8
|
|
// CHECK1-NEXT: [[__CONTEXT_ADDR_I:%.*]] = alloca %struct.anon*, align 8
|
|
// CHECK1-NEXT: [[DOTADDR:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[DOTADDR1:%.*]] = alloca %struct.kmp_task_t_with_privates*, align 8
|
|
// CHECK1-NEXT: store i32 [[TMP0]], i32* [[DOTADDR]], align 4
|
|
// CHECK1-NEXT: store %struct.kmp_task_t_with_privates* [[TMP1]], %struct.kmp_task_t_with_privates** [[DOTADDR1]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i32, i32* [[DOTADDR]], align 4
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load %struct.kmp_task_t_with_privates*, %struct.kmp_task_t_with_privates** [[DOTADDR1]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T_WITH_PRIVATES:%.*]], %struct.kmp_task_t_with_privates* [[TMP3]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T:%.*]], %struct.kmp_task_t* [[TMP4]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T]], %struct.kmp_task_t* [[TMP4]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = load i8*, i8** [[TMP6]], align 8
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = bitcast i8* [[TMP7]] to %struct.anon*
|
|
// CHECK1-NEXT: [[TMP9:%.*]] = bitcast %struct.kmp_task_t_with_privates* [[TMP3]] to i8*
|
|
// CHECK1-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META12:![0-9]+]])
|
|
// CHECK1-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META15:![0-9]+]])
|
|
// CHECK1-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META17:![0-9]+]])
|
|
// CHECK1-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META19:![0-9]+]])
|
|
// CHECK1-NEXT: store i32 [[TMP2]], i32* [[DOTGLOBAL_TID__ADDR_I]], align 4, !noalias !21
|
|
// CHECK1-NEXT: store i32* [[TMP5]], i32** [[DOTPART_ID__ADDR_I]], align 8, !noalias !21
|
|
// CHECK1-NEXT: store i8* null, i8** [[DOTPRIVATES__ADDR_I]], align 8, !noalias !21
|
|
// CHECK1-NEXT: store void (i8*, ...)* null, void (i8*, ...)** [[DOTCOPY_FN__ADDR_I]], align 8, !noalias !21
|
|
// CHECK1-NEXT: store i8* [[TMP9]], i8** [[DOTTASK_T__ADDR_I]], align 8, !noalias !21
|
|
// CHECK1-NEXT: store %struct.anon* [[TMP8]], %struct.anon** [[__CONTEXT_ADDR_I]], align 8, !noalias !21
|
|
// CHECK1-NEXT: [[TMP10:%.*]] = load %struct.anon*, %struct.anon** [[__CONTEXT_ADDR_I]], align 8, !noalias !21
|
|
// CHECK1-NEXT: [[TMP11:%.*]] = call i32 @__tgt_target_teams_nowait_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100.region_id, i32 0, i8** null, i8** null, i64* null, i64* null, i8** null, i8** null, i32 1, i32 0, i32 0, i8* null, i32 0, i8* null) #[[ATTR4]]
|
|
// CHECK1-NEXT: [[TMP12:%.*]] = icmp ne i32 [[TMP11]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP12]], label [[OMP_OFFLOAD_FAILED_I:%.*]], label [[DOTOMP_OUTLINED__1_EXIT:%.*]]
|
|
// CHECK1: omp_offload.failed.i:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100() #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[DOTOMP_OUTLINED__1_EXIT]]
|
|
// CHECK1: .omp_outlined..1.exit:
|
|
// CHECK1-NEXT: ret i32 0
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l104
|
|
// CHECK1-SAME: (i64 noundef [[A:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP0]], i32* [[CONV1]], align 4
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64)* @.omp_outlined..2 to void (i32*, i32*, ...)*), i64 [[TMP1]])
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..2
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110
|
|
// CHECK1-SAME: (i64 noundef [[AA:%.*]]) #[[ATTR2]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP0]], i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64)* @.omp_outlined..3 to void (i32*, i32*, ...)*), i64 [[TMP1]])
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..3
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[AA:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = sext i16 [[TMP0]] to i32
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = trunc i32 [[ADD]] to i16
|
|
// CHECK1-NEXT: store i16 [[CONV2]], i16* [[CONV]], align 2
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i32, i32* [[TMP1]], align 4
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = call i32 @__kmpc_cancel(%struct.ident_t* @[[GLOB1]], i32 [[TMP2]], i32 1)
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = icmp ne i32 [[TMP3]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP4]], label [[DOTCANCEL_EXIT:%.*]], label [[DOTCANCEL_CONTINUE:%.*]]
|
|
// CHECK1: .cancel.exit:
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = call i32 @__kmpc_cancel_barrier(%struct.ident_t* @[[GLOB2:[0-9]+]], i32 [[TMP2]])
|
|
// CHECK1-NEXT: br label [[DOTCANCEL_CONTINUE]]
|
|
// CHECK1: .cancel.continue:
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119
|
|
// CHECK1-SAME: (i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]]) #[[ATTR2]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP0]], i32* [[CONV2]], align 4
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[CONV3:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP2]], i16* [[CONV3]], align 2
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i64)* @.omp_outlined..4 to void (i32*, i32*, ...)*), i64 [[TMP1]], i64 [[TMP3]])
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..4
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = sext i16 [[TMP1]] to i32
|
|
// CHECK1-NEXT: [[ADD3:%.*]] = add nsw i32 [[CONV2]], 1
|
|
// CHECK1-NEXT: [[CONV4:%.*]] = trunc i32 [[ADD3]] to i16
|
|
// CHECK1-NEXT: store i16 [[CONV4]], i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144
|
|
// CHECK1-SAME: (i64 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i64 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 8 dereferenceable(400) [[C:%.*]], i64 noundef [[VLA1:%.*]], i64 noundef [[VLA3:%.*]], double* noundef nonnull align 8 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 8 dereferenceable(16) [[D:%.*]]) #[[ATTR2]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 8
|
|
// CHECK1-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR4:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 8
|
|
// CHECK1-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 8
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 8
|
|
// CHECK1-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA3]], i64* [[VLA_ADDR4]], align 8
|
|
// CHECK1-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 8
|
|
// CHECK1-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load i64, i64* [[VLA_ADDR4]], align 8
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[CONV5:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP8]], i32* [[CONV5]], align 4
|
|
// CHECK1-NEXT: [[TMP9:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 9, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, [10 x float]*, i64, float*, [5 x [10 x double]]*, i64, i64, double*, %struct.TT*)* @.omp_outlined..7 to void (i32*, i32*, ...)*), i64 [[TMP9]], [10 x float]* [[TMP0]], i64 [[TMP1]], float* [[TMP2]], [5 x [10 x double]]* [[TMP3]], i64 [[TMP4]], i64 [[TMP5]], double* [[TMP6]], %struct.TT* [[TMP7]])
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..7
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i64 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 8 dereferenceable(400) [[C:%.*]], i64 noundef [[VLA1:%.*]], i64 noundef [[VLA3:%.*]], double* noundef nonnull align 8 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 8 dereferenceable(16) [[D:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 8
|
|
// CHECK1-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR4:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 8
|
|
// CHECK1-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 8
|
|
// CHECK1-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA3]], i64* [[VLA_ADDR4]], align 8
|
|
// CHECK1-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 8
|
|
// CHECK1-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load i64, i64* [[VLA_ADDR4]], align 8
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP8]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x float], [10 x float]* [[TMP0]], i64 0, i64 2
|
|
// CHECK1-NEXT: [[TMP9:%.*]] = load float, float* [[ARRAYIDX]], align 4
|
|
// CHECK1-NEXT: [[CONV5:%.*]] = fpext float [[TMP9]] to double
|
|
// CHECK1-NEXT: [[ADD6:%.*]] = fadd double [[CONV5]], 1.000000e+00
|
|
// CHECK1-NEXT: [[CONV7:%.*]] = fptrunc double [[ADD6]] to float
|
|
// CHECK1-NEXT: store float [[CONV7]], float* [[ARRAYIDX]], align 4
|
|
// CHECK1-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds float, float* [[TMP2]], i64 3
|
|
// CHECK1-NEXT: [[TMP10:%.*]] = load float, float* [[ARRAYIDX8]], align 4
|
|
// CHECK1-NEXT: [[CONV9:%.*]] = fpext float [[TMP10]] to double
|
|
// CHECK1-NEXT: [[ADD10:%.*]] = fadd double [[CONV9]], 1.000000e+00
|
|
// CHECK1-NEXT: [[CONV11:%.*]] = fptrunc double [[ADD10]] to float
|
|
// CHECK1-NEXT: store float [[CONV11]], float* [[ARRAYIDX8]], align 4
|
|
// CHECK1-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds [5 x [10 x double]], [5 x [10 x double]]* [[TMP3]], i64 0, i64 1
|
|
// CHECK1-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds [10 x double], [10 x double]* [[ARRAYIDX12]], i64 0, i64 2
|
|
// CHECK1-NEXT: [[TMP11:%.*]] = load double, double* [[ARRAYIDX13]], align 8
|
|
// CHECK1-NEXT: [[ADD14:%.*]] = fadd double [[TMP11]], 1.000000e+00
|
|
// CHECK1-NEXT: store double [[ADD14]], double* [[ARRAYIDX13]], align 8
|
|
// CHECK1-NEXT: [[TMP12:%.*]] = mul nsw i64 1, [[TMP5]]
|
|
// CHECK1-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[TMP6]], i64 [[TMP12]]
|
|
// CHECK1-NEXT: [[ARRAYIDX16:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX15]], i64 3
|
|
// CHECK1-NEXT: [[TMP13:%.*]] = load double, double* [[ARRAYIDX16]], align 8
|
|
// CHECK1-NEXT: [[ADD17:%.*]] = fadd double [[TMP13]], 1.000000e+00
|
|
// CHECK1-NEXT: store double [[ADD17]], double* [[ARRAYIDX16]], align 8
|
|
// CHECK1-NEXT: [[X:%.*]] = getelementptr inbounds [[STRUCT_TT:%.*]], %struct.TT* [[TMP7]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP14:%.*]] = load i64, i64* [[X]], align 8
|
|
// CHECK1-NEXT: [[ADD18:%.*]] = add nsw i64 [[TMP14]], 1
|
|
// CHECK1-NEXT: store i64 [[ADD18]], i64* [[X]], align 8
|
|
// CHECK1-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_TT]], %struct.TT* [[TMP7]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP15:%.*]] = load i8, i8* [[Y]], align 8
|
|
// CHECK1-NEXT: [[CONV19:%.*]] = sext i8 [[TMP15]] to i32
|
|
// CHECK1-NEXT: [[ADD20:%.*]] = add nsw i32 [[CONV19]], 1
|
|
// CHECK1-NEXT: [[CONV21:%.*]] = trunc i32 [[ADD20]] to i8
|
|
// CHECK1-NEXT: store i8 [[CONV21]], i8* [[Y]], align 8
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@_Z3bari
|
|
// CHECK1-SAME: (i32 noundef signext [[N:%.*]]) #[[ATTR0]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[S:%.*]] = alloca [[STRUCT_S1:%.*]], align 8
|
|
// CHECK1-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CALL:%.*]] = call noundef signext i32 @_Z3fooi(i32 noundef signext [[TMP0]])
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], [[CALL]]
|
|
// CHECK1-NEXT: store i32 [[ADD]], i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CALL1:%.*]] = call noundef signext i32 @_ZN2S12r1Ei(%struct.S1* noundef nonnull align 8 dereferenceable(8) [[S]], i32 noundef signext [[TMP2]])
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[ADD2:%.*]] = add nsw i32 [[TMP3]], [[CALL1]]
|
|
// CHECK1-NEXT: store i32 [[ADD2]], i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CALL3:%.*]] = call noundef signext i32 @_ZL7fstatici(i32 noundef signext [[TMP4]])
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[ADD4:%.*]] = add nsw i32 [[TMP5]], [[CALL3]]
|
|
// CHECK1-NEXT: store i32 [[ADD4]], i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CALL5:%.*]] = call noundef signext i32 @_Z9ftemplateIiET_i(i32 noundef signext [[TMP6]])
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[ADD6:%.*]] = add nsw i32 [[TMP7]], [[CALL5]]
|
|
// CHECK1-NEXT: store i32 [[ADD6]], i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: ret i32 [[TMP8]]
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@_ZN2S12r1Ei
|
|
// CHECK1-SAME: (%struct.S1* noundef nonnull align 8 dereferenceable(8) [[THIS:%.*]], i32 noundef signext [[N:%.*]]) #[[ATTR0]] comdat align 2 {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 8
|
|
// CHECK1-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[B:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[SAVED_STACK:%.*]] = alloca i8*, align 8
|
|
// CHECK1-NEXT: [[__VLA_EXPR0:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[B_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [5 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [5 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [5 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_SIZES:%.*]] = alloca [5 x i64], align 8
|
|
// CHECK1-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK1-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[THIS1:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD]], i32* [[B]], align 4
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = call i8* @llvm.stacksave()
|
|
// CHECK1-NEXT: store i8* [[TMP3]], i8** [[SAVED_STACK]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = mul nuw i64 2, [[TMP2]]
|
|
// CHECK1-NEXT: [[VLA:%.*]] = alloca i16, i64 [[TMP4]], align 2
|
|
// CHECK1-NEXT: store i64 [[TMP2]], i64* [[__VLA_EXPR0]], align 8
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load i32, i32* [[B]], align 4
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[B_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP5]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = load i64, i64* [[B_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP7]], 60
|
|
// CHECK1-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK1: omp_if.then:
|
|
// CHECK1-NEXT: [[A:%.*]] = getelementptr inbounds [[STRUCT_S1:%.*]], %struct.S1* [[THIS1]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = mul nuw i64 2, [[TMP2]]
|
|
// CHECK1-NEXT: [[TMP9:%.*]] = mul nuw i64 [[TMP8]], 2
|
|
// CHECK1-NEXT: [[TMP10:%.*]] = bitcast [5 x i64]* [[DOTOFFLOAD_SIZES]] to i8*
|
|
// CHECK1-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP10]], i8* align 8 bitcast ([5 x i64]* @.offload_sizes.11 to i8*), i64 40, i1 false)
|
|
// CHECK1-NEXT: [[TMP11:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP12:%.*]] = bitcast i8** [[TMP11]] to %struct.S1**
|
|
// CHECK1-NEXT: store %struct.S1* [[THIS1]], %struct.S1** [[TMP12]], align 8
|
|
// CHECK1-NEXT: [[TMP13:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP14:%.*]] = bitcast i8** [[TMP13]] to double**
|
|
// CHECK1-NEXT: store double* [[A]], double** [[TMP14]], align 8
|
|
// CHECK1-NEXT: [[TMP15:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 0
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP15]], align 8
|
|
// CHECK1-NEXT: [[TMP16:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP17:%.*]] = bitcast i8** [[TMP16]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP6]], i64* [[TMP17]], align 8
|
|
// CHECK1-NEXT: [[TMP18:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP19:%.*]] = bitcast i8** [[TMP18]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP6]], i64* [[TMP19]], align 8
|
|
// CHECK1-NEXT: [[TMP20:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 1
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP20]], align 8
|
|
// CHECK1-NEXT: [[TMP21:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP22:%.*]] = bitcast i8** [[TMP21]] to i64*
|
|
// CHECK1-NEXT: store i64 2, i64* [[TMP22]], align 8
|
|
// CHECK1-NEXT: [[TMP23:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP24:%.*]] = bitcast i8** [[TMP23]] to i64*
|
|
// CHECK1-NEXT: store i64 2, i64* [[TMP24]], align 8
|
|
// CHECK1-NEXT: [[TMP25:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 2
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP25]], align 8
|
|
// CHECK1-NEXT: [[TMP26:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 3
|
|
// CHECK1-NEXT: [[TMP27:%.*]] = bitcast i8** [[TMP26]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP2]], i64* [[TMP27]], align 8
|
|
// CHECK1-NEXT: [[TMP28:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 3
|
|
// CHECK1-NEXT: [[TMP29:%.*]] = bitcast i8** [[TMP28]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP2]], i64* [[TMP29]], align 8
|
|
// CHECK1-NEXT: [[TMP30:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 3
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP30]], align 8
|
|
// CHECK1-NEXT: [[TMP31:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 4
|
|
// CHECK1-NEXT: [[TMP32:%.*]] = bitcast i8** [[TMP31]] to i16**
|
|
// CHECK1-NEXT: store i16* [[VLA]], i16** [[TMP32]], align 8
|
|
// CHECK1-NEXT: [[TMP33:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 4
|
|
// CHECK1-NEXT: [[TMP34:%.*]] = bitcast i8** [[TMP33]] to i16**
|
|
// CHECK1-NEXT: store i16* [[VLA]], i16** [[TMP34]], align 8
|
|
// CHECK1-NEXT: [[TMP35:%.*]] = getelementptr inbounds [5 x i64], [5 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 4
|
|
// CHECK1-NEXT: store i64 [[TMP9]], i64* [[TMP35]], align 8
|
|
// CHECK1-NEXT: [[TMP36:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 4
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP36]], align 8
|
|
// CHECK1-NEXT: [[TMP37:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP38:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP39:%.*]] = getelementptr inbounds [5 x i64], [5 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP40:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216.region_id, i32 5, i8** [[TMP37]], i8** [[TMP38]], i64* [[TMP39]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @.offload_maptypes.12, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK1-NEXT: [[TMP41:%.*]] = icmp ne i32 [[TMP40]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP41]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK1: omp_offload.failed:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216(%struct.S1* [[THIS1]], i64 [[TMP6]], i64 2, i64 [[TMP2]], i16* [[VLA]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK1: omp_offload.cont:
|
|
// CHECK1-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK1: omp_if.else:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216(%struct.S1* [[THIS1]], i64 [[TMP6]], i64 2, i64 [[TMP2]], i16* [[VLA]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK1: omp_if.end:
|
|
// CHECK1-NEXT: [[TMP42:%.*]] = mul nsw i64 1, [[TMP2]]
|
|
// CHECK1-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i16, i16* [[VLA]], i64 [[TMP42]]
|
|
// CHECK1-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i16, i16* [[ARRAYIDX]], i64 1
|
|
// CHECK1-NEXT: [[TMP43:%.*]] = load i16, i16* [[ARRAYIDX2]], align 2
|
|
// CHECK1-NEXT: [[CONV3:%.*]] = sext i16 [[TMP43]] to i32
|
|
// CHECK1-NEXT: [[TMP44:%.*]] = load i32, i32* [[B]], align 4
|
|
// CHECK1-NEXT: [[ADD4:%.*]] = add nsw i32 [[CONV3]], [[TMP44]]
|
|
// CHECK1-NEXT: [[TMP45:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8
|
|
// CHECK1-NEXT: call void @llvm.stackrestore(i8* [[TMP45]])
|
|
// CHECK1-NEXT: ret i32 [[ADD4]]
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@_ZL7fstatici
|
|
// CHECK1-SAME: (i32 noundef signext [[N:%.*]]) #[[ATTR0]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[AA:%.*]] = alloca i16, align 2
|
|
// CHECK1-NEXT: [[AAA:%.*]] = alloca i8, align 1
|
|
// CHECK1-NEXT: [[B:%.*]] = alloca [10 x i32], align 4
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AAA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [4 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [4 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [4 x i8*], align 8
|
|
// CHECK1-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK1-NEXT: store i16 0, i16* [[AA]], align 2
|
|
// CHECK1-NEXT: store i8 0, i8* [[AAA]], align 1
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP0]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP2]], i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i8, i8* [[AAA]], align 1
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = bitcast i64* [[AAA_CASTED]] to i8*
|
|
// CHECK1-NEXT: store i8 [[TMP4]], i8* [[CONV2]], align 1
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load i64, i64* [[AAA_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP6]], 50
|
|
// CHECK1-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK1: omp_if.then:
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = bitcast i8** [[TMP7]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP1]], i64* [[TMP8]], align 8
|
|
// CHECK1-NEXT: [[TMP9:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP10:%.*]] = bitcast i8** [[TMP9]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP1]], i64* [[TMP10]], align 8
|
|
// CHECK1-NEXT: [[TMP11:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 0
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP11]], align 8
|
|
// CHECK1-NEXT: [[TMP12:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP13:%.*]] = bitcast i8** [[TMP12]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP3]], i64* [[TMP13]], align 8
|
|
// CHECK1-NEXT: [[TMP14:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP15:%.*]] = bitcast i8** [[TMP14]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP3]], i64* [[TMP15]], align 8
|
|
// CHECK1-NEXT: [[TMP16:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 1
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP16]], align 8
|
|
// CHECK1-NEXT: [[TMP17:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP18:%.*]] = bitcast i8** [[TMP17]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP5]], i64* [[TMP18]], align 8
|
|
// CHECK1-NEXT: [[TMP19:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP20:%.*]] = bitcast i8** [[TMP19]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP5]], i64* [[TMP20]], align 8
|
|
// CHECK1-NEXT: [[TMP21:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 2
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP21]], align 8
|
|
// CHECK1-NEXT: [[TMP22:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 3
|
|
// CHECK1-NEXT: [[TMP23:%.*]] = bitcast i8** [[TMP22]] to [10 x i32]**
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP23]], align 8
|
|
// CHECK1-NEXT: [[TMP24:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 3
|
|
// CHECK1-NEXT: [[TMP25:%.*]] = bitcast i8** [[TMP24]] to [10 x i32]**
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP25]], align 8
|
|
// CHECK1-NEXT: [[TMP26:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 3
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP26]], align 8
|
|
// CHECK1-NEXT: [[TMP27:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP28:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP29:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198.region_id, i32 4, i8** [[TMP27]], i8** [[TMP28]], i64* getelementptr inbounds ([4 x i64], [4 x i64]* @.offload_sizes.14, i32 0, i32 0), i64* getelementptr inbounds ([4 x i64], [4 x i64]* @.offload_maptypes.15, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK1-NEXT: [[TMP30:%.*]] = icmp ne i32 [[TMP29]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP30]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK1: omp_offload.failed:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198(i64 [[TMP1]], i64 [[TMP3]], i64 [[TMP5]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK1: omp_offload.cont:
|
|
// CHECK1-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK1: omp_if.else:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198(i64 [[TMP1]], i64 [[TMP3]], i64 [[TMP5]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK1: omp_if.end:
|
|
// CHECK1-NEXT: [[TMP31:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: ret i32 [[TMP31]]
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@_Z9ftemplateIiET_i
|
|
// CHECK1-SAME: (i32 noundef signext [[N:%.*]]) #[[ATTR0]] comdat {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK1-NEXT: [[AA:%.*]] = alloca i16, align 2
|
|
// CHECK1-NEXT: [[B:%.*]] = alloca [10 x i32], align 4
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [3 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [3 x i8*], align 8
|
|
// CHECK1-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [3 x i8*], align 8
|
|
// CHECK1-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK1-NEXT: store i16 0, i16* [[AA]], align 2
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP0]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP2]], i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK1-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP4]], 40
|
|
// CHECK1-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK1: omp_if.then:
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = bitcast i8** [[TMP5]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP1]], i64* [[TMP6]], align 8
|
|
// CHECK1-NEXT: [[TMP7:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP8:%.*]] = bitcast i8** [[TMP7]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP1]], i64* [[TMP8]], align 8
|
|
// CHECK1-NEXT: [[TMP9:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 0
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP9]], align 8
|
|
// CHECK1-NEXT: [[TMP10:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP11:%.*]] = bitcast i8** [[TMP10]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP3]], i64* [[TMP11]], align 8
|
|
// CHECK1-NEXT: [[TMP12:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 1
|
|
// CHECK1-NEXT: [[TMP13:%.*]] = bitcast i8** [[TMP12]] to i64*
|
|
// CHECK1-NEXT: store i64 [[TMP3]], i64* [[TMP13]], align 8
|
|
// CHECK1-NEXT: [[TMP14:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 1
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP14]], align 8
|
|
// CHECK1-NEXT: [[TMP15:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP16:%.*]] = bitcast i8** [[TMP15]] to [10 x i32]**
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP16]], align 8
|
|
// CHECK1-NEXT: [[TMP17:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 2
|
|
// CHECK1-NEXT: [[TMP18:%.*]] = bitcast i8** [[TMP17]] to [10 x i32]**
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP18]], align 8
|
|
// CHECK1-NEXT: [[TMP19:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_MAPPERS]], i64 0, i64 2
|
|
// CHECK1-NEXT: store i8* null, i8** [[TMP19]], align 8
|
|
// CHECK1-NEXT: [[TMP20:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP21:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP22:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181.region_id, i32 3, i8** [[TMP20]], i8** [[TMP21]], i64* getelementptr inbounds ([3 x i64], [3 x i64]* @.offload_sizes.17, i32 0, i32 0), i64* getelementptr inbounds ([3 x i64], [3 x i64]* @.offload_maptypes.18, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK1-NEXT: [[TMP23:%.*]] = icmp ne i32 [[TMP22]], 0
|
|
// CHECK1-NEXT: br i1 [[TMP23]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK1: omp_offload.failed:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181(i64 [[TMP1]], i64 [[TMP3]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK1: omp_offload.cont:
|
|
// CHECK1-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK1: omp_if.else:
|
|
// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181(i64 [[TMP1]], i64 [[TMP3]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK1-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK1: omp_if.end:
|
|
// CHECK1-NEXT: [[TMP24:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK1-NEXT: ret i32 [[TMP24]]
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216
|
|
// CHECK1-SAME: (%struct.S1* noundef [[THIS:%.*]], i64 noundef [[B:%.*]], i64 noundef [[VLA:%.*]], i64 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR2]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 8
|
|
// CHECK1-NEXT: [[B_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[B]], i64* [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[B_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[CONV3:%.*]] = bitcast i64* [[B_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP4]], i32* [[CONV3]], align 4
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load i64, i64* [[B_CASTED]], align 8
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.S1*, i64, i64, i64, i16*)* @.omp_outlined..10 to void (i32*, i32*, ...)*), %struct.S1* [[TMP0]], i64 [[TMP5]], i64 [[TMP1]], i64 [[TMP2]], i16* [[TMP3]])
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..10
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], %struct.S1* noundef [[THIS:%.*]], i64 noundef [[B:%.*]], i64 noundef [[VLA:%.*]], i64 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[B]], i64* [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[B_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[CONV3:%.*]] = sitofp i32 [[TMP4]] to double
|
|
// CHECK1-NEXT: [[ADD:%.*]] = fadd double [[CONV3]], 1.500000e+00
|
|
// CHECK1-NEXT: [[A:%.*]] = getelementptr inbounds [[STRUCT_S1:%.*]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK1-NEXT: store double [[ADD]], double* [[A]], align 8
|
|
// CHECK1-NEXT: [[A4:%.*]] = getelementptr inbounds [[STRUCT_S1]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load double, double* [[A4]], align 8
|
|
// CHECK1-NEXT: [[INC:%.*]] = fadd double [[TMP5]], 1.000000e+00
|
|
// CHECK1-NEXT: store double [[INC]], double* [[A4]], align 8
|
|
// CHECK1-NEXT: [[CONV5:%.*]] = fptosi double [[INC]] to i16
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = mul nsw i64 1, [[TMP2]]
|
|
// CHECK1-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i16, i16* [[TMP3]], i64 [[TMP6]]
|
|
// CHECK1-NEXT: [[ARRAYIDX6:%.*]] = getelementptr inbounds i16, i16* [[ARRAYIDX]], i64 1
|
|
// CHECK1-NEXT: store i16 [[CONV5]], i16* [[ARRAYIDX6]], align 2
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198
|
|
// CHECK1-SAME: (i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], i64 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR2]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AAA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AAA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AAA]], i64* [[AAA_ADDR]], align 8
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = bitcast i64* [[AAA_ADDR]] to i8*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[CONV3:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP1]], i32* [[CONV3]], align 4
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[CONV4:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP3]], i16* [[CONV4]], align 2
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP5:%.*]] = load i8, i8* [[CONV2]], align 1
|
|
// CHECK1-NEXT: [[CONV5:%.*]] = bitcast i64* [[AAA_CASTED]] to i8*
|
|
// CHECK1-NEXT: store i8 [[TMP5]], i8* [[CONV5]], align 1
|
|
// CHECK1-NEXT: [[TMP6:%.*]] = load i64, i64* [[AAA_CASTED]], align 8
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i64, i64, [10 x i32]*)* @.omp_outlined..13 to void (i32*, i32*, ...)*), i64 [[TMP2]], i64 [[TMP4]], i64 [[TMP6]], [10 x i32]* [[TMP0]])
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..13
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], i64 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AAA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AAA]], i64* [[AAA_ADDR]], align 8
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = bitcast i64* [[AAA_ADDR]] to i8*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[CONV3:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK1-NEXT: [[ADD4:%.*]] = add nsw i32 [[CONV3]], 1
|
|
// CHECK1-NEXT: [[CONV5:%.*]] = trunc i32 [[ADD4]] to i16
|
|
// CHECK1-NEXT: store i16 [[CONV5]], i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i8, i8* [[CONV2]], align 1
|
|
// CHECK1-NEXT: [[CONV6:%.*]] = sext i8 [[TMP3]] to i32
|
|
// CHECK1-NEXT: [[ADD7:%.*]] = add nsw i32 [[CONV6]], 1
|
|
// CHECK1-NEXT: [[CONV8:%.*]] = trunc i32 [[ADD7]] to i8
|
|
// CHECK1-NEXT: store i8 [[CONV8]], i8* [[CONV2]], align 1
|
|
// CHECK1-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i64 0, i64 2
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK1-NEXT: [[ADD9:%.*]] = add nsw i32 [[TMP4]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD9]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181
|
|
// CHECK1-SAME: (i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR2]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK1-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK1-NEXT: store i32 [[TMP1]], i32* [[CONV2]], align 4
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[CONV3:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK1-NEXT: store i16 [[TMP3]], i16* [[CONV3]], align 2
|
|
// CHECK1-NEXT: [[TMP4:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK1-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i64, [10 x i32]*)* @.omp_outlined..16 to void (i32*, i32*, ...)*), i64 [[TMP2]], i64 [[TMP4]], [10 x i32]* [[TMP0]])
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..16
|
|
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR3]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK1-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK1-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK1-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK1-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK1-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[CONV2:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK1-NEXT: [[ADD3:%.*]] = add nsw i32 [[CONV2]], 1
|
|
// CHECK1-NEXT: [[CONV4:%.*]] = trunc i32 [[ADD3]] to i16
|
|
// CHECK1-NEXT: store i16 [[CONV4]], i16* [[CONV1]], align 2
|
|
// CHECK1-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i64 0, i64 2
|
|
// CHECK1-NEXT: [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK1-NEXT: [[ADD5:%.*]] = add nsw i32 [[TMP3]], 1
|
|
// CHECK1-NEXT: store i32 [[ADD5]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK1-LABEL: define {{[^@]+}}@.omp_offloading.requires_reg
|
|
// CHECK1-SAME: () #[[ATTR8:[0-9]+]] {
|
|
// CHECK1-NEXT: entry:
|
|
// CHECK1-NEXT: call void @__tgt_register_requires(i64 1)
|
|
// CHECK1-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@_Z3fooi
|
|
// CHECK3-SAME: (i32 noundef [[N:%.*]]) #[[ATTR0:[0-9]+]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA:%.*]] = alloca i16, align 2
|
|
// CHECK3-NEXT: [[B:%.*]] = alloca [10 x float], align 4
|
|
// CHECK3-NEXT: [[SAVED_STACK:%.*]] = alloca i8*, align 4
|
|
// CHECK3-NEXT: [[__VLA_EXPR0:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[C:%.*]] = alloca [5 x [10 x double]], align 8
|
|
// CHECK3-NEXT: [[__VLA_EXPR1:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[D:%.*]] = alloca [[STRUCT_TT:%.*]], align 4
|
|
// CHECK3-NEXT: [[AGG_CAPTURED:%.*]] = alloca [[STRUCT_ANON:%.*]], align 1
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [1 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [1 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [1 x i8*], align 4
|
|
// CHECK3-NEXT: [[A_CASTED2:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED3:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_BASEPTRS5:%.*]] = alloca [2 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_PTRS6:%.*]] = alloca [2 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_MAPPERS7:%.*]] = alloca [2 x i8*], align 4
|
|
// CHECK3-NEXT: [[A_CASTED10:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_BASEPTRS13:%.*]] = alloca [9 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_PTRS14:%.*]] = alloca [9 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_MAPPERS15:%.*]] = alloca [9 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_SIZES:%.*]] = alloca [9 x i64], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]])
|
|
// CHECK3-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i16 0, i16* [[AA]], align 2
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = call i8* @llvm.stacksave()
|
|
// CHECK3-NEXT: store i8* [[TMP2]], i8** [[SAVED_STACK]], align 4
|
|
// CHECK3-NEXT: [[VLA:%.*]] = alloca float, i32 [[TMP1]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[__VLA_EXPR0]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = mul nuw i32 5, [[TMP3]]
|
|
// CHECK3-NEXT: [[VLA1:%.*]] = alloca double, i32 [[TMP4]], align 8
|
|
// CHECK3-NEXT: store i32 [[TMP3]], i32* [[__VLA_EXPR1]], align 4
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = call i8* @__kmpc_omp_target_task_alloc(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]], i32 1, i32 20, i32 1, i32 (i32, i8*)* bitcast (i32 (i32, %struct.kmp_task_t_with_privates*)* @.omp_task_entry. to i32 (i32, i8*)*), i64 -1)
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = bitcast i8* [[TMP5]] to %struct.kmp_task_t_with_privates*
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T_WITH_PRIVATES:%.*]], %struct.kmp_task_t_with_privates* [[TMP6]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = call i32 @__kmpc_omp_task(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]], i8* [[TMP5]])
|
|
// CHECK3-NEXT: [[TMP9:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP9]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP10:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l104(i32 [[TMP10]]) #[[ATTR4:[0-9]+]]
|
|
// CHECK3-NEXT: [[TMP11:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP11]], i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[TMP12:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP13:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP14:%.*]] = bitcast i8** [[TMP13]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP12]], i32* [[TMP14]], align 4
|
|
// CHECK3-NEXT: [[TMP15:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP16:%.*]] = bitcast i8** [[TMP15]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP12]], i32* [[TMP16]], align 4
|
|
// CHECK3-NEXT: [[TMP17:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 0
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP17]], align 4
|
|
// CHECK3-NEXT: [[TMP18:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP19:%.*]] = getelementptr inbounds [1 x i8*], [1 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP20:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110.region_id, i32 1, i8** [[TMP18]], i8** [[TMP19]], i64* getelementptr inbounds ([1 x i64], [1 x i64]* @.offload_sizes, i32 0, i32 0), i64* getelementptr inbounds ([1 x i64], [1 x i64]* @.offload_maptypes, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK3-NEXT: [[TMP21:%.*]] = icmp ne i32 [[TMP20]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP21]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK3: omp_offload.failed:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110(i32 [[TMP12]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK3: omp_offload.cont:
|
|
// CHECK3-NEXT: [[TMP22:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP22]], i32* [[A_CASTED2]], align 4
|
|
// CHECK3-NEXT: [[TMP23:%.*]] = load i32, i32* [[A_CASTED2]], align 4
|
|
// CHECK3-NEXT: [[TMP24:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK3-NEXT: [[CONV4:%.*]] = bitcast i32* [[AA_CASTED3]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP24]], i16* [[CONV4]], align 2
|
|
// CHECK3-NEXT: [[TMP25:%.*]] = load i32, i32* [[AA_CASTED3]], align 4
|
|
// CHECK3-NEXT: [[TMP26:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP26]], 10
|
|
// CHECK3-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK3: omp_if.then:
|
|
// CHECK3-NEXT: [[TMP27:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_BASEPTRS5]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP28:%.*]] = bitcast i8** [[TMP27]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP23]], i32* [[TMP28]], align 4
|
|
// CHECK3-NEXT: [[TMP29:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_PTRS6]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP30:%.*]] = bitcast i8** [[TMP29]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP23]], i32* [[TMP30]], align 4
|
|
// CHECK3-NEXT: [[TMP31:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_MAPPERS7]], i32 0, i32 0
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP31]], align 4
|
|
// CHECK3-NEXT: [[TMP32:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_BASEPTRS5]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP33:%.*]] = bitcast i8** [[TMP32]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP25]], i32* [[TMP33]], align 4
|
|
// CHECK3-NEXT: [[TMP34:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_PTRS6]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP35:%.*]] = bitcast i8** [[TMP34]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP25]], i32* [[TMP35]], align 4
|
|
// CHECK3-NEXT: [[TMP36:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_MAPPERS7]], i32 0, i32 1
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP36]], align 4
|
|
// CHECK3-NEXT: [[TMP37:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_BASEPTRS5]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP38:%.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[DOTOFFLOAD_PTRS6]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP39:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119.region_id, i32 2, i8** [[TMP37]], i8** [[TMP38]], i64* getelementptr inbounds ([2 x i64], [2 x i64]* @.offload_sizes.5, i32 0, i32 0), i64* getelementptr inbounds ([2 x i64], [2 x i64]* @.offload_maptypes.6, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK3-NEXT: [[TMP40:%.*]] = icmp ne i32 [[TMP39]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP40]], label [[OMP_OFFLOAD_FAILED8:%.*]], label [[OMP_OFFLOAD_CONT9:%.*]]
|
|
// CHECK3: omp_offload.failed8:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119(i32 [[TMP23]], i32 [[TMP25]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_OFFLOAD_CONT9]]
|
|
// CHECK3: omp_offload.cont9:
|
|
// CHECK3-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK3: omp_if.else:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119(i32 [[TMP23]], i32 [[TMP25]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK3: omp_if.end:
|
|
// CHECK3-NEXT: [[TMP41:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP41]], i32* [[A_CASTED10]], align 4
|
|
// CHECK3-NEXT: [[TMP42:%.*]] = load i32, i32* [[A_CASTED10]], align 4
|
|
// CHECK3-NEXT: [[TMP43:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CMP11:%.*]] = icmp sgt i32 [[TMP43]], 20
|
|
// CHECK3-NEXT: br i1 [[CMP11]], label [[OMP_IF_THEN12:%.*]], label [[OMP_IF_ELSE18:%.*]]
|
|
// CHECK3: omp_if.then12:
|
|
// CHECK3-NEXT: [[TMP44:%.*]] = mul nuw i32 [[TMP1]], 4
|
|
// CHECK3-NEXT: [[TMP45:%.*]] = sext i32 [[TMP44]] to i64
|
|
// CHECK3-NEXT: [[TMP46:%.*]] = mul nuw i32 5, [[TMP3]]
|
|
// CHECK3-NEXT: [[TMP47:%.*]] = mul nuw i32 [[TMP46]], 8
|
|
// CHECK3-NEXT: [[TMP48:%.*]] = sext i32 [[TMP47]] to i64
|
|
// CHECK3-NEXT: [[TMP49:%.*]] = bitcast [9 x i64]* [[DOTOFFLOAD_SIZES]] to i8*
|
|
// CHECK3-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 [[TMP49]], i8* align 4 bitcast ([9 x i64]* @.offload_sizes.8 to i8*), i32 72, i1 false)
|
|
// CHECK3-NEXT: [[TMP50:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP51:%.*]] = bitcast i8** [[TMP50]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP42]], i32* [[TMP51]], align 4
|
|
// CHECK3-NEXT: [[TMP52:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP53:%.*]] = bitcast i8** [[TMP52]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP42]], i32* [[TMP53]], align 4
|
|
// CHECK3-NEXT: [[TMP54:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 0
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP54]], align 4
|
|
// CHECK3-NEXT: [[TMP55:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP56:%.*]] = bitcast i8** [[TMP55]] to [10 x float]**
|
|
// CHECK3-NEXT: store [10 x float]* [[B]], [10 x float]** [[TMP56]], align 4
|
|
// CHECK3-NEXT: [[TMP57:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP58:%.*]] = bitcast i8** [[TMP57]] to [10 x float]**
|
|
// CHECK3-NEXT: store [10 x float]* [[B]], [10 x float]** [[TMP58]], align 4
|
|
// CHECK3-NEXT: [[TMP59:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 1
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP59]], align 4
|
|
// CHECK3-NEXT: [[TMP60:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP61:%.*]] = bitcast i8** [[TMP60]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP61]], align 4
|
|
// CHECK3-NEXT: [[TMP62:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP63:%.*]] = bitcast i8** [[TMP62]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP63]], align 4
|
|
// CHECK3-NEXT: [[TMP64:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 2
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP64]], align 4
|
|
// CHECK3-NEXT: [[TMP65:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 3
|
|
// CHECK3-NEXT: [[TMP66:%.*]] = bitcast i8** [[TMP65]] to float**
|
|
// CHECK3-NEXT: store float* [[VLA]], float** [[TMP66]], align 4
|
|
// CHECK3-NEXT: [[TMP67:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 3
|
|
// CHECK3-NEXT: [[TMP68:%.*]] = bitcast i8** [[TMP67]] to float**
|
|
// CHECK3-NEXT: store float* [[VLA]], float** [[TMP68]], align 4
|
|
// CHECK3-NEXT: [[TMP69:%.*]] = getelementptr inbounds [9 x i64], [9 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 3
|
|
// CHECK3-NEXT: store i64 [[TMP45]], i64* [[TMP69]], align 4
|
|
// CHECK3-NEXT: [[TMP70:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 3
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP70]], align 4
|
|
// CHECK3-NEXT: [[TMP71:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 4
|
|
// CHECK3-NEXT: [[TMP72:%.*]] = bitcast i8** [[TMP71]] to [5 x [10 x double]]**
|
|
// CHECK3-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[TMP72]], align 4
|
|
// CHECK3-NEXT: [[TMP73:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 4
|
|
// CHECK3-NEXT: [[TMP74:%.*]] = bitcast i8** [[TMP73]] to [5 x [10 x double]]**
|
|
// CHECK3-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[TMP74]], align 4
|
|
// CHECK3-NEXT: [[TMP75:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 4
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP75]], align 4
|
|
// CHECK3-NEXT: [[TMP76:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 5
|
|
// CHECK3-NEXT: [[TMP77:%.*]] = bitcast i8** [[TMP76]] to i32*
|
|
// CHECK3-NEXT: store i32 5, i32* [[TMP77]], align 4
|
|
// CHECK3-NEXT: [[TMP78:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 5
|
|
// CHECK3-NEXT: [[TMP79:%.*]] = bitcast i8** [[TMP78]] to i32*
|
|
// CHECK3-NEXT: store i32 5, i32* [[TMP79]], align 4
|
|
// CHECK3-NEXT: [[TMP80:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 5
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP80]], align 4
|
|
// CHECK3-NEXT: [[TMP81:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 6
|
|
// CHECK3-NEXT: [[TMP82:%.*]] = bitcast i8** [[TMP81]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP3]], i32* [[TMP82]], align 4
|
|
// CHECK3-NEXT: [[TMP83:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 6
|
|
// CHECK3-NEXT: [[TMP84:%.*]] = bitcast i8** [[TMP83]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP3]], i32* [[TMP84]], align 4
|
|
// CHECK3-NEXT: [[TMP85:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 6
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP85]], align 4
|
|
// CHECK3-NEXT: [[TMP86:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 7
|
|
// CHECK3-NEXT: [[TMP87:%.*]] = bitcast i8** [[TMP86]] to double**
|
|
// CHECK3-NEXT: store double* [[VLA1]], double** [[TMP87]], align 4
|
|
// CHECK3-NEXT: [[TMP88:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 7
|
|
// CHECK3-NEXT: [[TMP89:%.*]] = bitcast i8** [[TMP88]] to double**
|
|
// CHECK3-NEXT: store double* [[VLA1]], double** [[TMP89]], align 4
|
|
// CHECK3-NEXT: [[TMP90:%.*]] = getelementptr inbounds [9 x i64], [9 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 7
|
|
// CHECK3-NEXT: store i64 [[TMP48]], i64* [[TMP90]], align 4
|
|
// CHECK3-NEXT: [[TMP91:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 7
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP91]], align 4
|
|
// CHECK3-NEXT: [[TMP92:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 8
|
|
// CHECK3-NEXT: [[TMP93:%.*]] = bitcast i8** [[TMP92]] to %struct.TT**
|
|
// CHECK3-NEXT: store %struct.TT* [[D]], %struct.TT** [[TMP93]], align 4
|
|
// CHECK3-NEXT: [[TMP94:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 8
|
|
// CHECK3-NEXT: [[TMP95:%.*]] = bitcast i8** [[TMP94]] to %struct.TT**
|
|
// CHECK3-NEXT: store %struct.TT* [[D]], %struct.TT** [[TMP95]], align 4
|
|
// CHECK3-NEXT: [[TMP96:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_MAPPERS15]], i32 0, i32 8
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP96]], align 4
|
|
// CHECK3-NEXT: [[TMP97:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_BASEPTRS13]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP98:%.*]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[DOTOFFLOAD_PTRS14]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP99:%.*]] = getelementptr inbounds [9 x i64], [9 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP100:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144.region_id, i32 9, i8** [[TMP97]], i8** [[TMP98]], i64* [[TMP99]], i64* getelementptr inbounds ([9 x i64], [9 x i64]* @.offload_maptypes.9, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK3-NEXT: [[TMP101:%.*]] = icmp ne i32 [[TMP100]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP101]], label [[OMP_OFFLOAD_FAILED16:%.*]], label [[OMP_OFFLOAD_CONT17:%.*]]
|
|
// CHECK3: omp_offload.failed16:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144(i32 [[TMP42]], [10 x float]* [[B]], i32 [[TMP1]], float* [[VLA]], [5 x [10 x double]]* [[C]], i32 5, i32 [[TMP3]], double* [[VLA1]], %struct.TT* [[D]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_OFFLOAD_CONT17]]
|
|
// CHECK3: omp_offload.cont17:
|
|
// CHECK3-NEXT: br label [[OMP_IF_END19:%.*]]
|
|
// CHECK3: omp_if.else18:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144(i32 [[TMP42]], [10 x float]* [[B]], i32 [[TMP1]], float* [[VLA]], [5 x [10 x double]]* [[C]], i32 5, i32 [[TMP3]], double* [[VLA1]], %struct.TT* [[D]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_IF_END19]]
|
|
// CHECK3: omp_if.end19:
|
|
// CHECK3-NEXT: [[TMP102:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[TMP103:%.*]] = load i8*, i8** [[SAVED_STACK]], align 4
|
|
// CHECK3-NEXT: call void @llvm.stackrestore(i8* [[TMP103]])
|
|
// CHECK3-NEXT: ret i32 [[TMP102]]
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100
|
|
// CHECK3-SAME: () #[[ATTR2:[0-9]+]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @.omp_outlined. to void (i32*, i32*, ...)*))
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined.
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]]) #[[ATTR3:[0-9]+]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_task_entry.
|
|
// CHECK3-SAME: (i32 noundef [[TMP0:%.*]], %struct.kmp_task_t_with_privates* noalias noundef [[TMP1:%.*]]) #[[ATTR5:[0-9]+]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR_I:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTPART_ID__ADDR_I:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTPRIVATES__ADDR_I:%.*]] = alloca i8*, align 4
|
|
// CHECK3-NEXT: [[DOTCOPY_FN__ADDR_I:%.*]] = alloca void (i8*, ...)*, align 4
|
|
// CHECK3-NEXT: [[DOTTASK_T__ADDR_I:%.*]] = alloca i8*, align 4
|
|
// CHECK3-NEXT: [[__CONTEXT_ADDR_I:%.*]] = alloca %struct.anon*, align 4
|
|
// CHECK3-NEXT: [[DOTADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTADDR1:%.*]] = alloca %struct.kmp_task_t_with_privates*, align 4
|
|
// CHECK3-NEXT: store i32 [[TMP0]], i32* [[DOTADDR]], align 4
|
|
// CHECK3-NEXT: store %struct.kmp_task_t_with_privates* [[TMP1]], %struct.kmp_task_t_with_privates** [[DOTADDR1]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i32, i32* [[DOTADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load %struct.kmp_task_t_with_privates*, %struct.kmp_task_t_with_privates** [[DOTADDR1]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T_WITH_PRIVATES:%.*]], %struct.kmp_task_t_with_privates* [[TMP3]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T:%.*]], %struct.kmp_task_t* [[TMP4]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = getelementptr inbounds [[STRUCT_KMP_TASK_T]], %struct.kmp_task_t* [[TMP4]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = load i8*, i8** [[TMP6]], align 4
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = bitcast i8* [[TMP7]] to %struct.anon*
|
|
// CHECK3-NEXT: [[TMP9:%.*]] = bitcast %struct.kmp_task_t_with_privates* [[TMP3]] to i8*
|
|
// CHECK3-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META13:![0-9]+]])
|
|
// CHECK3-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META16:![0-9]+]])
|
|
// CHECK3-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META18:![0-9]+]])
|
|
// CHECK3-NEXT: call void @llvm.experimental.noalias.scope.decl(metadata [[META20:![0-9]+]])
|
|
// CHECK3-NEXT: store i32 [[TMP2]], i32* [[DOTGLOBAL_TID__ADDR_I]], align 4, !noalias !22
|
|
// CHECK3-NEXT: store i32* [[TMP5]], i32** [[DOTPART_ID__ADDR_I]], align 4, !noalias !22
|
|
// CHECK3-NEXT: store i8* null, i8** [[DOTPRIVATES__ADDR_I]], align 4, !noalias !22
|
|
// CHECK3-NEXT: store void (i8*, ...)* null, void (i8*, ...)** [[DOTCOPY_FN__ADDR_I]], align 4, !noalias !22
|
|
// CHECK3-NEXT: store i8* [[TMP9]], i8** [[DOTTASK_T__ADDR_I]], align 4, !noalias !22
|
|
// CHECK3-NEXT: store %struct.anon* [[TMP8]], %struct.anon** [[__CONTEXT_ADDR_I]], align 4, !noalias !22
|
|
// CHECK3-NEXT: [[TMP10:%.*]] = load %struct.anon*, %struct.anon** [[__CONTEXT_ADDR_I]], align 4, !noalias !22
|
|
// CHECK3-NEXT: [[TMP11:%.*]] = call i32 @__tgt_target_teams_nowait_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100.region_id, i32 0, i8** null, i8** null, i64* null, i64* null, i8** null, i8** null, i32 1, i32 0, i32 0, i8* null, i32 0, i8* null) #[[ATTR4]]
|
|
// CHECK3-NEXT: [[TMP12:%.*]] = icmp ne i32 [[TMP11]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP12]], label [[OMP_OFFLOAD_FAILED_I:%.*]], label [[DOTOMP_OUTLINED__1_EXIT:%.*]]
|
|
// CHECK3: omp_offload.failed.i:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100() #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[DOTOMP_OUTLINED__1_EXIT]]
|
|
// CHECK3: .omp_outlined..1.exit:
|
|
// CHECK3-NEXT: ret i32 0
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l104
|
|
// CHECK3-SAME: (i32 noundef [[A:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP0]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32)* @.omp_outlined..2 to void (i32*, i32*, ...)*), i32 [[TMP1]])
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined..2
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110
|
|
// CHECK3-SAME: (i32 noundef [[AA:%.*]]) #[[ATTR2]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP0]], i16* [[CONV1]], align 2
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32)* @.omp_outlined..3 to void (i32*, i32*, ...)*), i32 [[TMP1]])
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined..3
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[AA:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = sext i16 [[TMP0]] to i32
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK3-NEXT: [[CONV2:%.*]] = trunc i32 [[ADD]] to i16
|
|
// CHECK3-NEXT: store i16 [[CONV2]], i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i32, i32* [[TMP1]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = call i32 @__kmpc_cancel(%struct.ident_t* @[[GLOB1]], i32 [[TMP2]], i32 1)
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = icmp ne i32 [[TMP3]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP4]], label [[DOTCANCEL_EXIT:%.*]], label [[DOTCANCEL_CONTINUE:%.*]]
|
|
// CHECK3: .cancel.exit:
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = call i32 @__kmpc_cancel_barrier(%struct.ident_t* @[[GLOB2:[0-9]+]], i32 [[TMP2]])
|
|
// CHECK3-NEXT: br label [[DOTCANCEL_CONTINUE]]
|
|
// CHECK3: .cancel.continue:
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119
|
|
// CHECK3-SAME: (i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]]) #[[ATTR2]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP0]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP2]], i16* [[CONV1]], align 2
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, i32)* @.omp_outlined..4 to void (i32*, i32*, ...)*), i32 [[TMP1]], i32 [[TMP3]])
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined..4
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = sext i16 [[TMP1]] to i32
|
|
// CHECK3-NEXT: [[ADD2:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK3-NEXT: [[CONV3:%.*]] = trunc i32 [[ADD2]] to i16
|
|
// CHECK3-NEXT: store i16 [[CONV3]], i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144
|
|
// CHECK3-SAME: (i32 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i32 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 4 dereferenceable(400) [[C:%.*]], i32 noundef [[VLA1:%.*]], i32 noundef [[VLA3:%.*]], double* noundef nonnull align 4 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 4 dereferenceable(12) [[D:%.*]]) #[[ATTR2]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 4
|
|
// CHECK3-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR4:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 4
|
|
// CHECK3-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 4
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 4
|
|
// CHECK3-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA3]], i32* [[VLA_ADDR4]], align 4
|
|
// CHECK3-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 4
|
|
// CHECK3-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load i32, i32* [[VLA_ADDR4]], align 4
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP8]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP9:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 9, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, [10 x float]*, i32, float*, [5 x [10 x double]]*, i32, i32, double*, %struct.TT*)* @.omp_outlined..7 to void (i32*, i32*, ...)*), i32 [[TMP9]], [10 x float]* [[TMP0]], i32 [[TMP1]], float* [[TMP2]], [5 x [10 x double]]* [[TMP3]], i32 [[TMP4]], i32 [[TMP5]], double* [[TMP6]], %struct.TT* [[TMP7]])
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined..7
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i32 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 4 dereferenceable(400) [[C:%.*]], i32 noundef [[VLA1:%.*]], i32 noundef [[VLA3:%.*]], double* noundef nonnull align 4 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 4 dereferenceable(12) [[D:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 4
|
|
// CHECK3-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR4:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 4
|
|
// CHECK3-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 4
|
|
// CHECK3-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA3]], i32* [[VLA_ADDR4]], align 4
|
|
// CHECK3-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 4
|
|
// CHECK3-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load i32, i32* [[VLA_ADDR4]], align 4
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP8]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x float], [10 x float]* [[TMP0]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP9:%.*]] = load float, float* [[ARRAYIDX]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = fpext float [[TMP9]] to double
|
|
// CHECK3-NEXT: [[ADD5:%.*]] = fadd double [[CONV]], 1.000000e+00
|
|
// CHECK3-NEXT: [[CONV6:%.*]] = fptrunc double [[ADD5]] to float
|
|
// CHECK3-NEXT: store float [[CONV6]], float* [[ARRAYIDX]], align 4
|
|
// CHECK3-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds float, float* [[TMP2]], i32 3
|
|
// CHECK3-NEXT: [[TMP10:%.*]] = load float, float* [[ARRAYIDX7]], align 4
|
|
// CHECK3-NEXT: [[CONV8:%.*]] = fpext float [[TMP10]] to double
|
|
// CHECK3-NEXT: [[ADD9:%.*]] = fadd double [[CONV8]], 1.000000e+00
|
|
// CHECK3-NEXT: [[CONV10:%.*]] = fptrunc double [[ADD9]] to float
|
|
// CHECK3-NEXT: store float [[CONV10]], float* [[ARRAYIDX7]], align 4
|
|
// CHECK3-NEXT: [[ARRAYIDX11:%.*]] = getelementptr inbounds [5 x [10 x double]], [5 x [10 x double]]* [[TMP3]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds [10 x double], [10 x double]* [[ARRAYIDX11]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP11:%.*]] = load double, double* [[ARRAYIDX12]], align 8
|
|
// CHECK3-NEXT: [[ADD13:%.*]] = fadd double [[TMP11]], 1.000000e+00
|
|
// CHECK3-NEXT: store double [[ADD13]], double* [[ARRAYIDX12]], align 8
|
|
// CHECK3-NEXT: [[TMP12:%.*]] = mul nsw i32 1, [[TMP5]]
|
|
// CHECK3-NEXT: [[ARRAYIDX14:%.*]] = getelementptr inbounds double, double* [[TMP6]], i32 [[TMP12]]
|
|
// CHECK3-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX14]], i32 3
|
|
// CHECK3-NEXT: [[TMP13:%.*]] = load double, double* [[ARRAYIDX15]], align 8
|
|
// CHECK3-NEXT: [[ADD16:%.*]] = fadd double [[TMP13]], 1.000000e+00
|
|
// CHECK3-NEXT: store double [[ADD16]], double* [[ARRAYIDX15]], align 8
|
|
// CHECK3-NEXT: [[X:%.*]] = getelementptr inbounds [[STRUCT_TT:%.*]], %struct.TT* [[TMP7]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP14:%.*]] = load i64, i64* [[X]], align 4
|
|
// CHECK3-NEXT: [[ADD17:%.*]] = add nsw i64 [[TMP14]], 1
|
|
// CHECK3-NEXT: store i64 [[ADD17]], i64* [[X]], align 4
|
|
// CHECK3-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_TT]], %struct.TT* [[TMP7]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP15:%.*]] = load i8, i8* [[Y]], align 4
|
|
// CHECK3-NEXT: [[CONV18:%.*]] = sext i8 [[TMP15]] to i32
|
|
// CHECK3-NEXT: [[ADD19:%.*]] = add nsw i32 [[CONV18]], 1
|
|
// CHECK3-NEXT: [[CONV20:%.*]] = trunc i32 [[ADD19]] to i8
|
|
// CHECK3-NEXT: store i8 [[CONV20]], i8* [[Y]], align 4
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@_Z3bari
|
|
// CHECK3-SAME: (i32 noundef [[N:%.*]]) #[[ATTR0]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[S:%.*]] = alloca [[STRUCT_S1:%.*]], align 4
|
|
// CHECK3-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CALL:%.*]] = call noundef i32 @_Z3fooi(i32 noundef [[TMP0]])
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], [[CALL]]
|
|
// CHECK3-NEXT: store i32 [[ADD]], i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CALL1:%.*]] = call noundef i32 @_ZN2S12r1Ei(%struct.S1* noundef nonnull align 4 dereferenceable(8) [[S]], i32 noundef [[TMP2]])
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[ADD2:%.*]] = add nsw i32 [[TMP3]], [[CALL1]]
|
|
// CHECK3-NEXT: store i32 [[ADD2]], i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CALL3:%.*]] = call noundef i32 @_ZL7fstatici(i32 noundef [[TMP4]])
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[ADD4:%.*]] = add nsw i32 [[TMP5]], [[CALL3]]
|
|
// CHECK3-NEXT: store i32 [[ADD4]], i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CALL5:%.*]] = call noundef i32 @_Z9ftemplateIiET_i(i32 noundef [[TMP6]])
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[ADD6:%.*]] = add nsw i32 [[TMP7]], [[CALL5]]
|
|
// CHECK3-NEXT: store i32 [[ADD6]], i32* [[A]], align 4
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: ret i32 [[TMP8]]
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@_ZN2S12r1Ei
|
|
// CHECK3-SAME: (%struct.S1* noundef nonnull align 4 dereferenceable(8) [[THIS:%.*]], i32 noundef [[N:%.*]]) #[[ATTR0]] comdat align 2 {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 4
|
|
// CHECK3-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[SAVED_STACK:%.*]] = alloca i8*, align 4
|
|
// CHECK3-NEXT: [[__VLA_EXPR0:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [5 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [5 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [5 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_SIZES:%.*]] = alloca [5 x i64], align 4
|
|
// CHECK3-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[THIS1:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD]], i32* [[B]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = call i8* @llvm.stacksave()
|
|
// CHECK3-NEXT: store i8* [[TMP2]], i8** [[SAVED_STACK]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = mul nuw i32 2, [[TMP1]]
|
|
// CHECK3-NEXT: [[VLA:%.*]] = alloca i16, i32 [[TMP3]], align 2
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[__VLA_EXPR0]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[B]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP4]], i32* [[B_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load i32, i32* [[B_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP6]], 60
|
|
// CHECK3-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK3: omp_if.then:
|
|
// CHECK3-NEXT: [[A:%.*]] = getelementptr inbounds [[STRUCT_S1:%.*]], %struct.S1* [[THIS1]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = mul nuw i32 2, [[TMP1]]
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = mul nuw i32 [[TMP7]], 2
|
|
// CHECK3-NEXT: [[TMP9:%.*]] = sext i32 [[TMP8]] to i64
|
|
// CHECK3-NEXT: [[TMP10:%.*]] = bitcast [5 x i64]* [[DOTOFFLOAD_SIZES]] to i8*
|
|
// CHECK3-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 [[TMP10]], i8* align 4 bitcast ([5 x i64]* @.offload_sizes.11 to i8*), i32 40, i1 false)
|
|
// CHECK3-NEXT: [[TMP11:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP12:%.*]] = bitcast i8** [[TMP11]] to %struct.S1**
|
|
// CHECK3-NEXT: store %struct.S1* [[THIS1]], %struct.S1** [[TMP12]], align 4
|
|
// CHECK3-NEXT: [[TMP13:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP14:%.*]] = bitcast i8** [[TMP13]] to double**
|
|
// CHECK3-NEXT: store double* [[A]], double** [[TMP14]], align 4
|
|
// CHECK3-NEXT: [[TMP15:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 0
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP15]], align 4
|
|
// CHECK3-NEXT: [[TMP16:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP17:%.*]] = bitcast i8** [[TMP16]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP5]], i32* [[TMP17]], align 4
|
|
// CHECK3-NEXT: [[TMP18:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP19:%.*]] = bitcast i8** [[TMP18]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP5]], i32* [[TMP19]], align 4
|
|
// CHECK3-NEXT: [[TMP20:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 1
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP20]], align 4
|
|
// CHECK3-NEXT: [[TMP21:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP22:%.*]] = bitcast i8** [[TMP21]] to i32*
|
|
// CHECK3-NEXT: store i32 2, i32* [[TMP22]], align 4
|
|
// CHECK3-NEXT: [[TMP23:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP24:%.*]] = bitcast i8** [[TMP23]] to i32*
|
|
// CHECK3-NEXT: store i32 2, i32* [[TMP24]], align 4
|
|
// CHECK3-NEXT: [[TMP25:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 2
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP25]], align 4
|
|
// CHECK3-NEXT: [[TMP26:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 3
|
|
// CHECK3-NEXT: [[TMP27:%.*]] = bitcast i8** [[TMP26]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP27]], align 4
|
|
// CHECK3-NEXT: [[TMP28:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 3
|
|
// CHECK3-NEXT: [[TMP29:%.*]] = bitcast i8** [[TMP28]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP29]], align 4
|
|
// CHECK3-NEXT: [[TMP30:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 3
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP30]], align 4
|
|
// CHECK3-NEXT: [[TMP31:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 4
|
|
// CHECK3-NEXT: [[TMP32:%.*]] = bitcast i8** [[TMP31]] to i16**
|
|
// CHECK3-NEXT: store i16* [[VLA]], i16** [[TMP32]], align 4
|
|
// CHECK3-NEXT: [[TMP33:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 4
|
|
// CHECK3-NEXT: [[TMP34:%.*]] = bitcast i8** [[TMP33]] to i16**
|
|
// CHECK3-NEXT: store i16* [[VLA]], i16** [[TMP34]], align 4
|
|
// CHECK3-NEXT: [[TMP35:%.*]] = getelementptr inbounds [5 x i64], [5 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 4
|
|
// CHECK3-NEXT: store i64 [[TMP9]], i64* [[TMP35]], align 4
|
|
// CHECK3-NEXT: [[TMP36:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 4
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP36]], align 4
|
|
// CHECK3-NEXT: [[TMP37:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP38:%.*]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP39:%.*]] = getelementptr inbounds [5 x i64], [5 x i64]* [[DOTOFFLOAD_SIZES]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP40:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216.region_id, i32 5, i8** [[TMP37]], i8** [[TMP38]], i64* [[TMP39]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @.offload_maptypes.12, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK3-NEXT: [[TMP41:%.*]] = icmp ne i32 [[TMP40]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP41]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK3: omp_offload.failed:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216(%struct.S1* [[THIS1]], i32 [[TMP5]], i32 2, i32 [[TMP1]], i16* [[VLA]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK3: omp_offload.cont:
|
|
// CHECK3-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK3: omp_if.else:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216(%struct.S1* [[THIS1]], i32 [[TMP5]], i32 2, i32 [[TMP1]], i16* [[VLA]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK3: omp_if.end:
|
|
// CHECK3-NEXT: [[TMP42:%.*]] = mul nsw i32 1, [[TMP1]]
|
|
// CHECK3-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i16, i16* [[VLA]], i32 [[TMP42]]
|
|
// CHECK3-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i16, i16* [[ARRAYIDX]], i32 1
|
|
// CHECK3-NEXT: [[TMP43:%.*]] = load i16, i16* [[ARRAYIDX2]], align 2
|
|
// CHECK3-NEXT: [[CONV:%.*]] = sext i16 [[TMP43]] to i32
|
|
// CHECK3-NEXT: [[TMP44:%.*]] = load i32, i32* [[B]], align 4
|
|
// CHECK3-NEXT: [[ADD3:%.*]] = add nsw i32 [[CONV]], [[TMP44]]
|
|
// CHECK3-NEXT: [[TMP45:%.*]] = load i8*, i8** [[SAVED_STACK]], align 4
|
|
// CHECK3-NEXT: call void @llvm.stackrestore(i8* [[TMP45]])
|
|
// CHECK3-NEXT: ret i32 [[ADD3]]
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@_ZL7fstatici
|
|
// CHECK3-SAME: (i32 noundef [[N:%.*]]) #[[ATTR0]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA:%.*]] = alloca i16, align 2
|
|
// CHECK3-NEXT: [[AAA:%.*]] = alloca i8, align 1
|
|
// CHECK3-NEXT: [[B:%.*]] = alloca [10 x i32], align 4
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AAA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [4 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [4 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [4 x i8*], align 4
|
|
// CHECK3-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i16 0, i16* [[AA]], align 2
|
|
// CHECK3-NEXT: store i8 0, i8* [[AAA]], align 1
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP0]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP2]], i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i8, i8* [[AAA]], align 1
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = bitcast i32* [[AAA_CASTED]] to i8*
|
|
// CHECK3-NEXT: store i8 [[TMP4]], i8* [[CONV1]], align 1
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load i32, i32* [[AAA_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP6]], 50
|
|
// CHECK3-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK3: omp_if.then:
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = bitcast i8** [[TMP7]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP8]], align 4
|
|
// CHECK3-NEXT: [[TMP9:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP10:%.*]] = bitcast i8** [[TMP9]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP10]], align 4
|
|
// CHECK3-NEXT: [[TMP11:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 0
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP11]], align 4
|
|
// CHECK3-NEXT: [[TMP12:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP13:%.*]] = bitcast i8** [[TMP12]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP3]], i32* [[TMP13]], align 4
|
|
// CHECK3-NEXT: [[TMP14:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP15:%.*]] = bitcast i8** [[TMP14]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP3]], i32* [[TMP15]], align 4
|
|
// CHECK3-NEXT: [[TMP16:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 1
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP16]], align 4
|
|
// CHECK3-NEXT: [[TMP17:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP18:%.*]] = bitcast i8** [[TMP17]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP5]], i32* [[TMP18]], align 4
|
|
// CHECK3-NEXT: [[TMP19:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP20:%.*]] = bitcast i8** [[TMP19]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP5]], i32* [[TMP20]], align 4
|
|
// CHECK3-NEXT: [[TMP21:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 2
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP21]], align 4
|
|
// CHECK3-NEXT: [[TMP22:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 3
|
|
// CHECK3-NEXT: [[TMP23:%.*]] = bitcast i8** [[TMP22]] to [10 x i32]**
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP23]], align 4
|
|
// CHECK3-NEXT: [[TMP24:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 3
|
|
// CHECK3-NEXT: [[TMP25:%.*]] = bitcast i8** [[TMP24]] to [10 x i32]**
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP25]], align 4
|
|
// CHECK3-NEXT: [[TMP26:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 3
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP26]], align 4
|
|
// CHECK3-NEXT: [[TMP27:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP28:%.*]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP29:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198.region_id, i32 4, i8** [[TMP27]], i8** [[TMP28]], i64* getelementptr inbounds ([4 x i64], [4 x i64]* @.offload_sizes.14, i32 0, i32 0), i64* getelementptr inbounds ([4 x i64], [4 x i64]* @.offload_maptypes.15, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK3-NEXT: [[TMP30:%.*]] = icmp ne i32 [[TMP29]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP30]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK3: omp_offload.failed:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198(i32 [[TMP1]], i32 [[TMP3]], i32 [[TMP5]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK3: omp_offload.cont:
|
|
// CHECK3-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK3: omp_if.else:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198(i32 [[TMP1]], i32 [[TMP3]], i32 [[TMP5]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK3: omp_if.end:
|
|
// CHECK3-NEXT: [[TMP31:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: ret i32 [[TMP31]]
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@_Z9ftemplateIiET_i
|
|
// CHECK3-SAME: (i32 noundef [[N:%.*]]) #[[ATTR0]] comdat {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[A:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA:%.*]] = alloca i16, align 2
|
|
// CHECK3-NEXT: [[B:%.*]] = alloca [10 x i32], align 4
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [3 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [3 x i8*], align 4
|
|
// CHECK3-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [3 x i8*], align 4
|
|
// CHECK3-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 0, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i16 0, i16* [[AA]], align 2
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP0]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i16, i16* [[AA]], align 2
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP2]], i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[N_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP4]], 40
|
|
// CHECK3-NEXT: br i1 [[CMP]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]]
|
|
// CHECK3: omp_if.then:
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = bitcast i8** [[TMP5]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP6]], align 4
|
|
// CHECK3-NEXT: [[TMP7:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP8:%.*]] = bitcast i8** [[TMP7]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[TMP8]], align 4
|
|
// CHECK3-NEXT: [[TMP9:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 0
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP9]], align 4
|
|
// CHECK3-NEXT: [[TMP10:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP11:%.*]] = bitcast i8** [[TMP10]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP3]], i32* [[TMP11]], align 4
|
|
// CHECK3-NEXT: [[TMP12:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 1
|
|
// CHECK3-NEXT: [[TMP13:%.*]] = bitcast i8** [[TMP12]] to i32*
|
|
// CHECK3-NEXT: store i32 [[TMP3]], i32* [[TMP13]], align 4
|
|
// CHECK3-NEXT: [[TMP14:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 1
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP14]], align 4
|
|
// CHECK3-NEXT: [[TMP15:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP16:%.*]] = bitcast i8** [[TMP15]] to [10 x i32]**
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP16]], align 4
|
|
// CHECK3-NEXT: [[TMP17:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP18:%.*]] = bitcast i8** [[TMP17]] to [10 x i32]**
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[TMP18]], align 4
|
|
// CHECK3-NEXT: [[TMP19:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_MAPPERS]], i32 0, i32 2
|
|
// CHECK3-NEXT: store i8* null, i8** [[TMP19]], align 4
|
|
// CHECK3-NEXT: [[TMP20:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP21:%.*]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[DOTOFFLOAD_PTRS]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP22:%.*]] = call i32 @__tgt_target_teams_mapper(%struct.ident_t* @[[GLOB1]], i64 -1, i8* @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181.region_id, i32 3, i8** [[TMP20]], i8** [[TMP21]], i64* getelementptr inbounds ([3 x i64], [3 x i64]* @.offload_sizes.17, i32 0, i32 0), i64* getelementptr inbounds ([3 x i64], [3 x i64]* @.offload_maptypes.18, i32 0, i32 0), i8** null, i8** null, i32 1, i32 0)
|
|
// CHECK3-NEXT: [[TMP23:%.*]] = icmp ne i32 [[TMP22]], 0
|
|
// CHECK3-NEXT: br i1 [[TMP23]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
|
|
// CHECK3: omp_offload.failed:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181(i32 [[TMP1]], i32 [[TMP3]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_OFFLOAD_CONT]]
|
|
// CHECK3: omp_offload.cont:
|
|
// CHECK3-NEXT: br label [[OMP_IF_END:%.*]]
|
|
// CHECK3: omp_if.else:
|
|
// CHECK3-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181(i32 [[TMP1]], i32 [[TMP3]], [10 x i32]* [[B]]) #[[ATTR4]]
|
|
// CHECK3-NEXT: br label [[OMP_IF_END]]
|
|
// CHECK3: omp_if.end:
|
|
// CHECK3-NEXT: [[TMP24:%.*]] = load i32, i32* [[A]], align 4
|
|
// CHECK3-NEXT: ret i32 [[TMP24]]
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216
|
|
// CHECK3-SAME: (%struct.S1* noundef [[THIS:%.*]], i32 noundef [[B:%.*]], i32 noundef [[VLA:%.*]], i32 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR2]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 4
|
|
// CHECK3-NEXT: [[B_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[B]], i32* [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP4]], i32* [[B_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load i32, i32* [[B_CASTED]], align 4
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.S1*, i32, i32, i32, i16*)* @.omp_outlined..10 to void (i32*, i32*, ...)*), %struct.S1* [[TMP0]], i32 [[TMP5]], i32 [[TMP1]], i32 [[TMP2]], i16* [[TMP3]])
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined..10
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], %struct.S1* noundef [[THIS:%.*]], i32 noundef [[B:%.*]], i32 noundef [[VLA:%.*]], i32 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[B]], i32* [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = sitofp i32 [[TMP4]] to double
|
|
// CHECK3-NEXT: [[ADD:%.*]] = fadd double [[CONV]], 1.500000e+00
|
|
// CHECK3-NEXT: [[A:%.*]] = getelementptr inbounds [[STRUCT_S1:%.*]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK3-NEXT: store double [[ADD]], double* [[A]], align 4
|
|
// CHECK3-NEXT: [[A3:%.*]] = getelementptr inbounds [[STRUCT_S1]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load double, double* [[A3]], align 4
|
|
// CHECK3-NEXT: [[INC:%.*]] = fadd double [[TMP5]], 1.000000e+00
|
|
// CHECK3-NEXT: store double [[INC]], double* [[A3]], align 4
|
|
// CHECK3-NEXT: [[CONV4:%.*]] = fptosi double [[INC]] to i16
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = mul nsw i32 1, [[TMP2]]
|
|
// CHECK3-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i16, i16* [[TMP3]], i32 [[TMP6]]
|
|
// CHECK3-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds i16, i16* [[ARRAYIDX]], i32 1
|
|
// CHECK3-NEXT: store i16 [[CONV4]], i16* [[ARRAYIDX5]], align 2
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198
|
|
// CHECK3-SAME: (i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], i32 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR2]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AAA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AAA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AAA]], i32* [[AAA_ADDR]], align 4
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = bitcast i32* [[AAA_ADDR]] to i8*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV2:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP3]], i16* [[CONV2]], align 2
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP5:%.*]] = load i8, i8* [[CONV1]], align 1
|
|
// CHECK3-NEXT: [[CONV3:%.*]] = bitcast i32* [[AAA_CASTED]] to i8*
|
|
// CHECK3-NEXT: store i8 [[TMP5]], i8* [[CONV3]], align 1
|
|
// CHECK3-NEXT: [[TMP6:%.*]] = load i32, i32* [[AAA_CASTED]], align 4
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, i32, i32, [10 x i32]*)* @.omp_outlined..13 to void (i32*, i32*, ...)*), i32 [[TMP2]], i32 [[TMP4]], i32 [[TMP6]], [10 x i32]* [[TMP0]])
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined..13
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], i32 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AAA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AAA]], i32* [[AAA_ADDR]], align 4
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = bitcast i32* [[AAA_ADDR]] to i8*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV2:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK3-NEXT: [[ADD3:%.*]] = add nsw i32 [[CONV2]], 1
|
|
// CHECK3-NEXT: [[CONV4:%.*]] = trunc i32 [[ADD3]] to i16
|
|
// CHECK3-NEXT: store i16 [[CONV4]], i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i8, i8* [[CONV1]], align 1
|
|
// CHECK3-NEXT: [[CONV5:%.*]] = sext i8 [[TMP3]] to i32
|
|
// CHECK3-NEXT: [[ADD6:%.*]] = add nsw i32 [[CONV5]], 1
|
|
// CHECK3-NEXT: [[CONV7:%.*]] = trunc i32 [[ADD6]] to i8
|
|
// CHECK3-NEXT: store i8 [[CONV7]], i8* [[CONV1]], align 1
|
|
// CHECK3-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK3-NEXT: [[ADD8:%.*]] = add nsw i32 [[TMP4]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD8]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181
|
|
// CHECK3-SAME: (i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR2]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK3-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[TMP1]], i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK3-NEXT: store i16 [[TMP3]], i16* [[CONV1]], align 2
|
|
// CHECK3-NEXT: [[TMP4:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK3-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, i32, [10 x i32]*)* @.omp_outlined..16 to void (i32*, i32*, ...)*), i32 [[TMP2]], i32 [[TMP4]], [10 x i32]* [[TMP0]])
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_outlined..16
|
|
// CHECK3-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR3]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK3-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK3-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK3-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK3-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK3-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK3-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[CONV1:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK3-NEXT: [[ADD2:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK3-NEXT: [[CONV3:%.*]] = trunc i32 [[ADD2]] to i16
|
|
// CHECK3-NEXT: store i16 [[CONV3]], i16* [[CONV]], align 2
|
|
// CHECK3-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i32 0, i32 2
|
|
// CHECK3-NEXT: [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK3-NEXT: [[ADD4:%.*]] = add nsw i32 [[TMP3]], 1
|
|
// CHECK3-NEXT: store i32 [[ADD4]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK3-LABEL: define {{[^@]+}}@.omp_offloading.requires_reg
|
|
// CHECK3-SAME: () #[[ATTR8:[0-9]+]] {
|
|
// CHECK3-NEXT: entry:
|
|
// CHECK3-NEXT: call void @__tgt_register_requires(i64 1)
|
|
// CHECK3-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100
|
|
// CHECK9-SAME: () #[[ATTR0:[0-9]+]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1:[0-9]+]], i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @.omp_outlined. to void (i32*, i32*, ...)*))
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@.omp_outlined.
|
|
// CHECK9-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]]) #[[ATTR1:[0-9]+]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110
|
|
// CHECK9-SAME: (i64 noundef [[AA:%.*]]) #[[ATTR0]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK9-NEXT: store i16 [[TMP0]], i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK9-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64)* @.omp_outlined..1 to void (i32*, i32*, ...)*), i64 [[TMP1]])
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@.omp_outlined..1
|
|
// CHECK9-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[AA:%.*]]) #[[ATTR1]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = sext i16 [[TMP0]] to i32
|
|
// CHECK9-NEXT: [[ADD:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK9-NEXT: [[CONV2:%.*]] = trunc i32 [[ADD]] to i16
|
|
// CHECK9-NEXT: store i16 [[CONV2]], i16* [[CONV]], align 2
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i32, i32* [[TMP1]], align 4
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = call i32 @__kmpc_cancel(%struct.ident_t* @[[GLOB1]], i32 [[TMP2]], i32 1)
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = icmp ne i32 [[TMP3]], 0
|
|
// CHECK9-NEXT: br i1 [[TMP4]], label [[DOTCANCEL_EXIT:%.*]], label [[DOTCANCEL_CONTINUE:%.*]]
|
|
// CHECK9: .cancel.exit:
|
|
// CHECK9-NEXT: [[TMP5:%.*]] = call i32 @__kmpc_cancel_barrier(%struct.ident_t* @[[GLOB2:[0-9]+]], i32 [[TMP2]])
|
|
// CHECK9-NEXT: br label [[DOTCANCEL_CONTINUE]]
|
|
// CHECK9: .cancel.continue:
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119
|
|
// CHECK9-SAME: (i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]]) #[[ATTR0]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[CONV2:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK9-NEXT: store i32 [[TMP0]], i32* [[CONV2]], align 4
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[CONV3:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK9-NEXT: store i16 [[TMP2]], i16* [[CONV3]], align 2
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK9-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i64)* @.omp_outlined..2 to void (i32*, i32*, ...)*), i64 [[TMP1]], i64 [[TMP3]])
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@.omp_outlined..2
|
|
// CHECK9-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]]) #[[ATTR1]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK9-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[CONV2:%.*]] = sext i16 [[TMP1]] to i32
|
|
// CHECK9-NEXT: [[ADD3:%.*]] = add nsw i32 [[CONV2]], 1
|
|
// CHECK9-NEXT: [[CONV4:%.*]] = trunc i32 [[ADD3]] to i16
|
|
// CHECK9-NEXT: store i16 [[CONV4]], i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144
|
|
// CHECK9-SAME: (i64 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i64 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 8 dereferenceable(400) [[C:%.*]], i64 noundef [[VLA1:%.*]], i64 noundef [[VLA3:%.*]], double* noundef nonnull align 8 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 8 dereferenceable(16) [[D:%.*]]) #[[ATTR0]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 8
|
|
// CHECK9-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR4:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 8
|
|
// CHECK9-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 8
|
|
// CHECK9-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 8
|
|
// CHECK9-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA3]], i64* [[VLA_ADDR4]], align 8
|
|
// CHECK9-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 8
|
|
// CHECK9-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: [[TMP5:%.*]] = load i64, i64* [[VLA_ADDR4]], align 8
|
|
// CHECK9-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP8:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[CONV5:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK9-NEXT: store i32 [[TMP8]], i32* [[CONV5]], align 4
|
|
// CHECK9-NEXT: [[TMP9:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK9-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 9, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, [10 x float]*, i64, float*, [5 x [10 x double]]*, i64, i64, double*, %struct.TT*)* @.omp_outlined..3 to void (i32*, i32*, ...)*), i64 [[TMP9]], [10 x float]* [[TMP0]], i64 [[TMP1]], float* [[TMP2]], [5 x [10 x double]]* [[TMP3]], i64 [[TMP4]], i64 [[TMP5]], double* [[TMP6]], %struct.TT* [[TMP7]])
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@.omp_outlined..3
|
|
// CHECK9-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i64 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 8 dereferenceable(400) [[C:%.*]], i64 noundef [[VLA1:%.*]], i64 noundef [[VLA3:%.*]], double* noundef nonnull align 8 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 8 dereferenceable(16) [[D:%.*]]) #[[ATTR1]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 8
|
|
// CHECK9-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR4:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 8
|
|
// CHECK9-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 8
|
|
// CHECK9-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 8
|
|
// CHECK9-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA3]], i64* [[VLA_ADDR4]], align 8
|
|
// CHECK9-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 8
|
|
// CHECK9-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: [[TMP5:%.*]] = load i64, i64* [[VLA_ADDR4]], align 8
|
|
// CHECK9-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP8:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP8]], 1
|
|
// CHECK9-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x float], [10 x float]* [[TMP0]], i64 0, i64 2
|
|
// CHECK9-NEXT: [[TMP9:%.*]] = load float, float* [[ARRAYIDX]], align 4
|
|
// CHECK9-NEXT: [[CONV5:%.*]] = fpext float [[TMP9]] to double
|
|
// CHECK9-NEXT: [[ADD6:%.*]] = fadd double [[CONV5]], 1.000000e+00
|
|
// CHECK9-NEXT: [[CONV7:%.*]] = fptrunc double [[ADD6]] to float
|
|
// CHECK9-NEXT: store float [[CONV7]], float* [[ARRAYIDX]], align 4
|
|
// CHECK9-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds float, float* [[TMP2]], i64 3
|
|
// CHECK9-NEXT: [[TMP10:%.*]] = load float, float* [[ARRAYIDX8]], align 4
|
|
// CHECK9-NEXT: [[CONV9:%.*]] = fpext float [[TMP10]] to double
|
|
// CHECK9-NEXT: [[ADD10:%.*]] = fadd double [[CONV9]], 1.000000e+00
|
|
// CHECK9-NEXT: [[CONV11:%.*]] = fptrunc double [[ADD10]] to float
|
|
// CHECK9-NEXT: store float [[CONV11]], float* [[ARRAYIDX8]], align 4
|
|
// CHECK9-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds [5 x [10 x double]], [5 x [10 x double]]* [[TMP3]], i64 0, i64 1
|
|
// CHECK9-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds [10 x double], [10 x double]* [[ARRAYIDX12]], i64 0, i64 2
|
|
// CHECK9-NEXT: [[TMP11:%.*]] = load double, double* [[ARRAYIDX13]], align 8
|
|
// CHECK9-NEXT: [[ADD14:%.*]] = fadd double [[TMP11]], 1.000000e+00
|
|
// CHECK9-NEXT: store double [[ADD14]], double* [[ARRAYIDX13]], align 8
|
|
// CHECK9-NEXT: [[TMP12:%.*]] = mul nsw i64 1, [[TMP5]]
|
|
// CHECK9-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[TMP6]], i64 [[TMP12]]
|
|
// CHECK9-NEXT: [[ARRAYIDX16:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX15]], i64 3
|
|
// CHECK9-NEXT: [[TMP13:%.*]] = load double, double* [[ARRAYIDX16]], align 8
|
|
// CHECK9-NEXT: [[ADD17:%.*]] = fadd double [[TMP13]], 1.000000e+00
|
|
// CHECK9-NEXT: store double [[ADD17]], double* [[ARRAYIDX16]], align 8
|
|
// CHECK9-NEXT: [[X:%.*]] = getelementptr inbounds [[STRUCT_TT:%.*]], %struct.TT* [[TMP7]], i32 0, i32 0
|
|
// CHECK9-NEXT: [[TMP14:%.*]] = load i64, i64* [[X]], align 8
|
|
// CHECK9-NEXT: [[ADD18:%.*]] = add nsw i64 [[TMP14]], 1
|
|
// CHECK9-NEXT: store i64 [[ADD18]], i64* [[X]], align 8
|
|
// CHECK9-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_TT]], %struct.TT* [[TMP7]], i32 0, i32 1
|
|
// CHECK9-NEXT: [[TMP15:%.*]] = load i8, i8* [[Y]], align 8
|
|
// CHECK9-NEXT: [[CONV19:%.*]] = sext i8 [[TMP15]] to i32
|
|
// CHECK9-NEXT: [[ADD20:%.*]] = add nsw i32 [[CONV19]], 1
|
|
// CHECK9-NEXT: [[CONV21:%.*]] = trunc i32 [[ADD20]] to i8
|
|
// CHECK9-NEXT: store i8 [[CONV21]], i8* [[Y]], align 8
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198
|
|
// CHECK9-SAME: (i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], i64 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR0]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AAA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK9-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AAA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AAA]], i64* [[AAA_ADDR]], align 8
|
|
// CHECK9-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[CONV2:%.*]] = bitcast i64* [[AAA_ADDR]] to i8*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[CONV3:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK9-NEXT: store i32 [[TMP1]], i32* [[CONV3]], align 4
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[CONV4:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK9-NEXT: store i16 [[TMP3]], i16* [[CONV4]], align 2
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK9-NEXT: [[TMP5:%.*]] = load i8, i8* [[CONV2]], align 1
|
|
// CHECK9-NEXT: [[CONV5:%.*]] = bitcast i64* [[AAA_CASTED]] to i8*
|
|
// CHECK9-NEXT: store i8 [[TMP5]], i8* [[CONV5]], align 1
|
|
// CHECK9-NEXT: [[TMP6:%.*]] = load i64, i64* [[AAA_CASTED]], align 8
|
|
// CHECK9-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i64, i64, [10 x i32]*)* @.omp_outlined..4 to void (i32*, i32*, ...)*), i64 [[TMP2]], i64 [[TMP4]], i64 [[TMP6]], [10 x i32]* [[TMP0]])
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@.omp_outlined..4
|
|
// CHECK9-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], i64 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR1]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AAA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK9-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AAA]], i64* [[AAA_ADDR]], align 8
|
|
// CHECK9-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[CONV2:%.*]] = bitcast i64* [[AAA_ADDR]] to i8*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK9-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[CONV3:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK9-NEXT: [[ADD4:%.*]] = add nsw i32 [[CONV3]], 1
|
|
// CHECK9-NEXT: [[CONV5:%.*]] = trunc i32 [[ADD4]] to i16
|
|
// CHECK9-NEXT: store i16 [[CONV5]], i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load i8, i8* [[CONV2]], align 1
|
|
// CHECK9-NEXT: [[CONV6:%.*]] = sext i8 [[TMP3]] to i32
|
|
// CHECK9-NEXT: [[ADD7:%.*]] = add nsw i32 [[CONV6]], 1
|
|
// CHECK9-NEXT: [[CONV8:%.*]] = trunc i32 [[ADD7]] to i8
|
|
// CHECK9-NEXT: store i8 [[CONV8]], i8* [[CONV2]], align 1
|
|
// CHECK9-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i64 0, i64 2
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK9-NEXT: [[ADD9:%.*]] = add nsw i32 [[TMP4]], 1
|
|
// CHECK9-NEXT: store i32 [[ADD9]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216
|
|
// CHECK9-SAME: (%struct.S1* noundef [[THIS:%.*]], i64 noundef [[B:%.*]], i64 noundef [[VLA:%.*]], i64 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR0]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 8
|
|
// CHECK9-NEXT: [[B_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[B]], i64* [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[B_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[CONV3:%.*]] = bitcast i64* [[B_CASTED]] to i32*
|
|
// CHECK9-NEXT: store i32 [[TMP4]], i32* [[CONV3]], align 4
|
|
// CHECK9-NEXT: [[TMP5:%.*]] = load i64, i64* [[B_CASTED]], align 8
|
|
// CHECK9-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.S1*, i64, i64, i64, i16*)* @.omp_outlined..5 to void (i32*, i32*, ...)*), %struct.S1* [[TMP0]], i64 [[TMP5]], i64 [[TMP1]], i64 [[TMP2]], i16* [[TMP3]])
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@.omp_outlined..5
|
|
// CHECK9-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], %struct.S1* noundef [[THIS:%.*]], i64 noundef [[B:%.*]], i64 noundef [[VLA:%.*]], i64 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR1]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 8
|
|
// CHECK9-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[B]], i64* [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[B_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[CONV3:%.*]] = sitofp i32 [[TMP4]] to double
|
|
// CHECK9-NEXT: [[ADD:%.*]] = fadd double [[CONV3]], 1.500000e+00
|
|
// CHECK9-NEXT: [[A:%.*]] = getelementptr inbounds [[STRUCT_S1:%.*]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK9-NEXT: store double [[ADD]], double* [[A]], align 8
|
|
// CHECK9-NEXT: [[A4:%.*]] = getelementptr inbounds [[STRUCT_S1]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK9-NEXT: [[TMP5:%.*]] = load double, double* [[A4]], align 8
|
|
// CHECK9-NEXT: [[INC:%.*]] = fadd double [[TMP5]], 1.000000e+00
|
|
// CHECK9-NEXT: store double [[INC]], double* [[A4]], align 8
|
|
// CHECK9-NEXT: [[CONV5:%.*]] = fptosi double [[INC]] to i16
|
|
// CHECK9-NEXT: [[TMP6:%.*]] = mul nsw i64 1, [[TMP2]]
|
|
// CHECK9-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i16, i16* [[TMP3]], i64 [[TMP6]]
|
|
// CHECK9-NEXT: [[ARRAYIDX6:%.*]] = getelementptr inbounds i16, i16* [[ARRAYIDX]], i64 1
|
|
// CHECK9-NEXT: store i16 [[CONV5]], i16* [[ARRAYIDX6]], align 2
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181
|
|
// CHECK9-SAME: (i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR0]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK9-NEXT: [[A_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_CASTED:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[CONV2:%.*]] = bitcast i64* [[A_CASTED]] to i32*
|
|
// CHECK9-NEXT: store i32 [[TMP1]], i32* [[CONV2]], align 4
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i64, i64* [[A_CASTED]], align 8
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[CONV3:%.*]] = bitcast i64* [[AA_CASTED]] to i16*
|
|
// CHECK9-NEXT: store i16 [[TMP3]], i16* [[CONV3]], align 2
|
|
// CHECK9-NEXT: [[TMP4:%.*]] = load i64, i64* [[AA_CASTED]], align 8
|
|
// CHECK9-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i64, [10 x i32]*)* @.omp_outlined..6 to void (i32*, i32*, ...)*), i64 [[TMP2]], i64 [[TMP4]], [10 x i32]* [[TMP0]])
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK9-LABEL: define {{[^@]+}}@.omp_outlined..6
|
|
// CHECK9-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[A:%.*]], i64 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR1]] {
|
|
// CHECK9-NEXT: entry:
|
|
// CHECK9-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
|
|
// CHECK9-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[AA_ADDR:%.*]] = alloca i64, align 8
|
|
// CHECK9-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 8
|
|
// CHECK9-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[A]], i64* [[A_ADDR]], align 8
|
|
// CHECK9-NEXT: store i64 [[AA]], i64* [[AA_ADDR]], align 8
|
|
// CHECK9-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
|
|
// CHECK9-NEXT: [[CONV1:%.*]] = bitcast i64* [[AA_ADDR]] to i16*
|
|
// CHECK9-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 8
|
|
// CHECK9-NEXT: [[TMP1:%.*]] = load i32, i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK9-NEXT: store i32 [[ADD]], i32* [[CONV]], align 4
|
|
// CHECK9-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[CONV2:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK9-NEXT: [[ADD3:%.*]] = add nsw i32 [[CONV2]], 1
|
|
// CHECK9-NEXT: [[CONV4:%.*]] = trunc i32 [[ADD3]] to i16
|
|
// CHECK9-NEXT: store i16 [[CONV4]], i16* [[CONV1]], align 2
|
|
// CHECK9-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i64 0, i64 2
|
|
// CHECK9-NEXT: [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK9-NEXT: [[ADD5:%.*]] = add nsw i32 [[TMP3]], 1
|
|
// CHECK9-NEXT: store i32 [[ADD5]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK9-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l100
|
|
// CHECK11-SAME: () #[[ATTR0:[0-9]+]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1:[0-9]+]], i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @.omp_outlined. to void (i32*, i32*, ...)*))
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@.omp_outlined.
|
|
// CHECK11-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]]) #[[ATTR1:[0-9]+]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l110
|
|
// CHECK11-SAME: (i32 noundef [[AA:%.*]]) #[[ATTR0]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK11-NEXT: store i16 [[TMP0]], i16* [[CONV1]], align 2
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK11-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32)* @.omp_outlined..1 to void (i32*, i32*, ...)*), i32 [[TMP1]])
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@.omp_outlined..1
|
|
// CHECK11-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[AA:%.*]]) #[[ATTR1]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = sext i16 [[TMP0]] to i32
|
|
// CHECK11-NEXT: [[ADD:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK11-NEXT: [[CONV2:%.*]] = trunc i32 [[ADD]] to i16
|
|
// CHECK11-NEXT: store i16 [[CONV2]], i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i32, i32* [[TMP1]], align 4
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = call i32 @__kmpc_cancel(%struct.ident_t* @[[GLOB1]], i32 [[TMP2]], i32 1)
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = icmp ne i32 [[TMP3]], 0
|
|
// CHECK11-NEXT: br i1 [[TMP4]], label [[DOTCANCEL_EXIT:%.*]], label [[DOTCANCEL_CONTINUE:%.*]]
|
|
// CHECK11: .cancel.exit:
|
|
// CHECK11-NEXT: [[TMP5:%.*]] = call i32 @__kmpc_cancel_barrier(%struct.ident_t* @[[GLOB2:[0-9]+]], i32 [[TMP2]])
|
|
// CHECK11-NEXT: br label [[DOTCANCEL_CONTINUE]]
|
|
// CHECK11: .cancel.continue:
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l119
|
|
// CHECK11-SAME: (i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]]) #[[ATTR0]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[TMP0]], i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK11-NEXT: store i16 [[TMP2]], i16* [[CONV1]], align 2
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK11-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, i32)* @.omp_outlined..2 to void (i32*, i32*, ...)*), i32 [[TMP1]], i32 [[TMP3]])
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@.omp_outlined..2
|
|
// CHECK11-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]]) #[[ATTR1]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP0]], 1
|
|
// CHECK11-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = sext i16 [[TMP1]] to i32
|
|
// CHECK11-NEXT: [[ADD2:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK11-NEXT: [[CONV3:%.*]] = trunc i32 [[ADD2]] to i16
|
|
// CHECK11-NEXT: store i16 [[CONV3]], i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z3fooi_l144
|
|
// CHECK11-SAME: (i32 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i32 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 4 dereferenceable(400) [[C:%.*]], i32 noundef [[VLA1:%.*]], i32 noundef [[VLA3:%.*]], double* noundef nonnull align 4 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 4 dereferenceable(12) [[D:%.*]]) #[[ATTR0]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 4
|
|
// CHECK11-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR4:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 4
|
|
// CHECK11-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 4
|
|
// CHECK11-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 4
|
|
// CHECK11-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA3]], i32* [[VLA_ADDR4]], align 4
|
|
// CHECK11-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 4
|
|
// CHECK11-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: [[TMP5:%.*]] = load i32, i32* [[VLA_ADDR4]], align 4
|
|
// CHECK11-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP8:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[TMP8]], i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP9:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 9, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, [10 x float]*, i32, float*, [5 x [10 x double]]*, i32, i32, double*, %struct.TT*)* @.omp_outlined..3 to void (i32*, i32*, ...)*), i32 [[TMP9]], [10 x float]* [[TMP0]], i32 [[TMP1]], float* [[TMP2]], [5 x [10 x double]]* [[TMP3]], i32 [[TMP4]], i32 [[TMP5]], double* [[TMP6]], %struct.TT* [[TMP7]])
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@.omp_outlined..3
|
|
// CHECK11-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], [10 x float]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]], i32 noundef [[VLA:%.*]], float* noundef nonnull align 4 dereferenceable(4) [[BN:%.*]], [5 x [10 x double]]* noundef nonnull align 4 dereferenceable(400) [[C:%.*]], i32 noundef [[VLA1:%.*]], i32 noundef [[VLA3:%.*]], double* noundef nonnull align 4 dereferenceable(8) [[CN:%.*]], %struct.TT* noundef nonnull align 4 dereferenceable(12) [[D:%.*]]) #[[ATTR1]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca [10 x float]*, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[BN_ADDR:%.*]] = alloca float*, align 4
|
|
// CHECK11-NEXT: [[C_ADDR:%.*]] = alloca [5 x [10 x double]]*, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR4:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[CN_ADDR:%.*]] = alloca double*, align 4
|
|
// CHECK11-NEXT: [[D_ADDR:%.*]] = alloca %struct.TT*, align 4
|
|
// CHECK11-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store [10 x float]* [[B]], [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: store float* [[BN]], float** [[BN_ADDR]], align 4
|
|
// CHECK11-NEXT: store [5 x [10 x double]]* [[C]], [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA3]], i32* [[VLA_ADDR4]], align 4
|
|
// CHECK11-NEXT: store double* [[CN]], double** [[CN_ADDR]], align 4
|
|
// CHECK11-NEXT: store %struct.TT* [[D]], %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load [10 x float]*, [10 x float]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load float*, float** [[BN_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: [[TMP5:%.*]] = load i32, i32* [[VLA_ADDR4]], align 4
|
|
// CHECK11-NEXT: [[TMP6:%.*]] = load double*, double** [[CN_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP7:%.*]] = load %struct.TT*, %struct.TT** [[D_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP8:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP8]], 1
|
|
// CHECK11-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x float], [10 x float]* [[TMP0]], i32 0, i32 2
|
|
// CHECK11-NEXT: [[TMP9:%.*]] = load float, float* [[ARRAYIDX]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = fpext float [[TMP9]] to double
|
|
// CHECK11-NEXT: [[ADD5:%.*]] = fadd double [[CONV]], 1.000000e+00
|
|
// CHECK11-NEXT: [[CONV6:%.*]] = fptrunc double [[ADD5]] to float
|
|
// CHECK11-NEXT: store float [[CONV6]], float* [[ARRAYIDX]], align 4
|
|
// CHECK11-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds float, float* [[TMP2]], i32 3
|
|
// CHECK11-NEXT: [[TMP10:%.*]] = load float, float* [[ARRAYIDX7]], align 4
|
|
// CHECK11-NEXT: [[CONV8:%.*]] = fpext float [[TMP10]] to double
|
|
// CHECK11-NEXT: [[ADD9:%.*]] = fadd double [[CONV8]], 1.000000e+00
|
|
// CHECK11-NEXT: [[CONV10:%.*]] = fptrunc double [[ADD9]] to float
|
|
// CHECK11-NEXT: store float [[CONV10]], float* [[ARRAYIDX7]], align 4
|
|
// CHECK11-NEXT: [[ARRAYIDX11:%.*]] = getelementptr inbounds [5 x [10 x double]], [5 x [10 x double]]* [[TMP3]], i32 0, i32 1
|
|
// CHECK11-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds [10 x double], [10 x double]* [[ARRAYIDX11]], i32 0, i32 2
|
|
// CHECK11-NEXT: [[TMP11:%.*]] = load double, double* [[ARRAYIDX12]], align 8
|
|
// CHECK11-NEXT: [[ADD13:%.*]] = fadd double [[TMP11]], 1.000000e+00
|
|
// CHECK11-NEXT: store double [[ADD13]], double* [[ARRAYIDX12]], align 8
|
|
// CHECK11-NEXT: [[TMP12:%.*]] = mul nsw i32 1, [[TMP5]]
|
|
// CHECK11-NEXT: [[ARRAYIDX14:%.*]] = getelementptr inbounds double, double* [[TMP6]], i32 [[TMP12]]
|
|
// CHECK11-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX14]], i32 3
|
|
// CHECK11-NEXT: [[TMP13:%.*]] = load double, double* [[ARRAYIDX15]], align 8
|
|
// CHECK11-NEXT: [[ADD16:%.*]] = fadd double [[TMP13]], 1.000000e+00
|
|
// CHECK11-NEXT: store double [[ADD16]], double* [[ARRAYIDX15]], align 8
|
|
// CHECK11-NEXT: [[X:%.*]] = getelementptr inbounds [[STRUCT_TT:%.*]], %struct.TT* [[TMP7]], i32 0, i32 0
|
|
// CHECK11-NEXT: [[TMP14:%.*]] = load i64, i64* [[X]], align 4
|
|
// CHECK11-NEXT: [[ADD17:%.*]] = add nsw i64 [[TMP14]], 1
|
|
// CHECK11-NEXT: store i64 [[ADD17]], i64* [[X]], align 4
|
|
// CHECK11-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_TT]], %struct.TT* [[TMP7]], i32 0, i32 1
|
|
// CHECK11-NEXT: [[TMP15:%.*]] = load i8, i8* [[Y]], align 4
|
|
// CHECK11-NEXT: [[CONV18:%.*]] = sext i8 [[TMP15]] to i32
|
|
// CHECK11-NEXT: [[ADD19:%.*]] = add nsw i32 [[CONV18]], 1
|
|
// CHECK11-NEXT: [[CONV20:%.*]] = trunc i32 [[ADD19]] to i8
|
|
// CHECK11-NEXT: store i8 [[CONV20]], i8* [[Y]], align 4
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZL7fstatici_l198
|
|
// CHECK11-SAME: (i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], i32 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR0]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AAA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK11-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AAA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AAA]], i32* [[AAA_ADDR]], align 4
|
|
// CHECK11-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = bitcast i32* [[AAA_ADDR]] to i8*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[TMP1]], i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV2:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK11-NEXT: store i16 [[TMP3]], i16* [[CONV2]], align 2
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP5:%.*]] = load i8, i8* [[CONV1]], align 1
|
|
// CHECK11-NEXT: [[CONV3:%.*]] = bitcast i32* [[AAA_CASTED]] to i8*
|
|
// CHECK11-NEXT: store i8 [[TMP5]], i8* [[CONV3]], align 1
|
|
// CHECK11-NEXT: [[TMP6:%.*]] = load i32, i32* [[AAA_CASTED]], align 4
|
|
// CHECK11-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, i32, i32, [10 x i32]*)* @.omp_outlined..4 to void (i32*, i32*, ...)*), i32 [[TMP2]], i32 [[TMP4]], i32 [[TMP6]], [10 x i32]* [[TMP0]])
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@.omp_outlined..4
|
|
// CHECK11-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], i32 noundef [[AAA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR1]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AAA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK11-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AAA]], i32* [[AAA_ADDR]], align 4
|
|
// CHECK11-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = bitcast i32* [[AAA_ADDR]] to i8*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK11-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV2:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK11-NEXT: [[ADD3:%.*]] = add nsw i32 [[CONV2]], 1
|
|
// CHECK11-NEXT: [[CONV4:%.*]] = trunc i32 [[ADD3]] to i16
|
|
// CHECK11-NEXT: store i16 [[CONV4]], i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load i8, i8* [[CONV1]], align 1
|
|
// CHECK11-NEXT: [[CONV5:%.*]] = sext i8 [[TMP3]] to i32
|
|
// CHECK11-NEXT: [[ADD6:%.*]] = add nsw i32 [[CONV5]], 1
|
|
// CHECK11-NEXT: [[CONV7:%.*]] = trunc i32 [[ADD6]] to i8
|
|
// CHECK11-NEXT: store i8 [[CONV7]], i8* [[CONV1]], align 1
|
|
// CHECK11-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i32 0, i32 2
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK11-NEXT: [[ADD8:%.*]] = add nsw i32 [[TMP4]], 1
|
|
// CHECK11-NEXT: store i32 [[ADD8]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__ZN2S12r1Ei_l216
|
|
// CHECK11-SAME: (%struct.S1* noundef [[THIS:%.*]], i32 noundef [[B:%.*]], i32 noundef [[VLA:%.*]], i32 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR0]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 4
|
|
// CHECK11-NEXT: [[B_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[B]], i32* [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = load i32, i32* [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[TMP4]], i32* [[B_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP5:%.*]] = load i32, i32* [[B_CASTED]], align 4
|
|
// CHECK11-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.S1*, i32, i32, i32, i16*)* @.omp_outlined..5 to void (i32*, i32*, ...)*), %struct.S1* [[TMP0]], i32 [[TMP5]], i32 [[TMP1]], i32 [[TMP2]], i16* [[TMP3]])
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@.omp_outlined..5
|
|
// CHECK11-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], %struct.S1* noundef [[THIS:%.*]], i32 noundef [[B:%.*]], i32 noundef [[VLA:%.*]], i32 noundef [[VLA1:%.*]], i16* noundef nonnull align 2 dereferenceable(2) [[C:%.*]]) #[[ATTR1]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[THIS_ADDR:%.*]] = alloca %struct.S1*, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[VLA_ADDR2:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[C_ADDR:%.*]] = alloca i16*, align 4
|
|
// CHECK11-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store %struct.S1* [[THIS]], %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[B]], i32* [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA]], i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[VLA1]], i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: store i16* [[C]], i16** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load %struct.S1*, %struct.S1** [[THIS_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[VLA_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i32, i32* [[VLA_ADDR2]], align 4
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load i16*, i16** [[C_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = load i32, i32* [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = sitofp i32 [[TMP4]] to double
|
|
// CHECK11-NEXT: [[ADD:%.*]] = fadd double [[CONV]], 1.500000e+00
|
|
// CHECK11-NEXT: [[A:%.*]] = getelementptr inbounds [[STRUCT_S1:%.*]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK11-NEXT: store double [[ADD]], double* [[A]], align 4
|
|
// CHECK11-NEXT: [[A3:%.*]] = getelementptr inbounds [[STRUCT_S1]], %struct.S1* [[TMP0]], i32 0, i32 0
|
|
// CHECK11-NEXT: [[TMP5:%.*]] = load double, double* [[A3]], align 4
|
|
// CHECK11-NEXT: [[INC:%.*]] = fadd double [[TMP5]], 1.000000e+00
|
|
// CHECK11-NEXT: store double [[INC]], double* [[A3]], align 4
|
|
// CHECK11-NEXT: [[CONV4:%.*]] = fptosi double [[INC]] to i16
|
|
// CHECK11-NEXT: [[TMP6:%.*]] = mul nsw i32 1, [[TMP2]]
|
|
// CHECK11-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i16, i16* [[TMP3]], i32 [[TMP6]]
|
|
// CHECK11-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds i16, i16* [[ARRAYIDX]], i32 1
|
|
// CHECK11-NEXT: store i16 [[CONV4]], i16* [[ARRAYIDX5]], align 2
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9ftemplateIiET_i_l181
|
|
// CHECK11-SAME: (i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR0]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK11-NEXT: [[A_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_CASTED:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[TMP1]], i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i32, i32* [[A_CASTED]], align 4
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = bitcast i32* [[AA_CASTED]] to i16*
|
|
// CHECK11-NEXT: store i16 [[TMP3]], i16* [[CONV1]], align 2
|
|
// CHECK11-NEXT: [[TMP4:%.*]] = load i32, i32* [[AA_CASTED]], align 4
|
|
// CHECK11-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32, i32, [10 x i32]*)* @.omp_outlined..6 to void (i32*, i32*, ...)*), i32 [[TMP2]], i32 [[TMP4]], [10 x i32]* [[TMP0]])
|
|
// CHECK11-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK11-LABEL: define {{[^@]+}}@.omp_outlined..6
|
|
// CHECK11-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i32 noundef [[A:%.*]], i32 noundef [[AA:%.*]], [10 x i32]* noundef nonnull align 4 dereferenceable(40) [[B:%.*]]) #[[ATTR1]] {
|
|
// CHECK11-NEXT: entry:
|
|
// CHECK11-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 4
|
|
// CHECK11-NEXT: [[A_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[AA_ADDR:%.*]] = alloca i32, align 4
|
|
// CHECK11-NEXT: [[B_ADDR:%.*]] = alloca [10 x i32]*, align 4
|
|
// CHECK11-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[A]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: store i32 [[AA]], i32* [[AA_ADDR]], align 4
|
|
// CHECK11-NEXT: store [10 x i32]* [[B]], [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[CONV:%.*]] = bitcast i32* [[AA_ADDR]] to i16*
|
|
// CHECK11-NEXT: [[TMP0:%.*]] = load [10 x i32]*, [10 x i32]** [[B_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP1:%.*]] = load i32, i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], 1
|
|
// CHECK11-NEXT: store i32 [[ADD]], i32* [[A_ADDR]], align 4
|
|
// CHECK11-NEXT: [[TMP2:%.*]] = load i16, i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[CONV1:%.*]] = sext i16 [[TMP2]] to i32
|
|
// CHECK11-NEXT: [[ADD2:%.*]] = add nsw i32 [[CONV1]], 1
|
|
// CHECK11-NEXT: [[CONV3:%.*]] = trunc i32 [[ADD2]] to i16
|
|
// CHECK11-NEXT: store i16 [[CONV3]], i16* [[CONV]], align 2
|
|
// CHECK11-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x i32], [10 x i32]* [[TMP0]], i32 0, i32 2
|
|
// CHECK11-NEXT: [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
|
|
// CHECK11-NEXT: [[ADD4:%.*]] = add nsw i32 [[TMP3]], 1
|
|
// CHECK11-NEXT: store i32 [[ADD4]], i32* [[ARRAYIDX]], align 4
|
|
// CHECK11-NEXT: ret void
|
|
//
|