330d8983d2
In a nutshell, this moves our libomptarget code to populate the offload subproject. With this commit, users need to enable the new LLVM/Offload subproject as a runtime in their cmake configuration. No further changes are expected for downstream code. Tests and other components still depend on OpenMP and have also not been renamed. The results below are for a build in which OpenMP and Offload are enabled runtimes. In addition to the pure `git mv`, we needed to adjust some CMake files. Nothing is intended to change semantics. ``` ninja check-offload ``` Works with the X86 and AMDGPU offload tests ``` ninja check-openmp ``` Still works but doesn't build offload tests anymore. ``` ls install/lib ``` Shows all expected libraries, incl. - `libomptarget.devicertl.a` - `libomptarget-nvptx-sm_90.bc` - `libomptarget.rtl.amdgpu.so` -> `libomptarget.rtl.amdgpu.so.18git` - `libomptarget.so` -> `libomptarget.so.18git` Fixes: https://github.com/llvm/llvm-project/issues/75124 --------- Co-authored-by: Saiyedul Islam <Saiyedul.Islam@amd.com>
74 lines
2.3 KiB
C++
74 lines
2.3 KiB
C++
// RUN: %libomptarget-compilexx-and-run-generic
|
|
// RUN: %libomptarget-compileoptxx-and-run-generic
|
|
|
|
// FIXME: This is a bug in host offload, this should run fine.
|
|
// UNSUPPORTED: aarch64-unknown-linux-gnu
|
|
// UNSUPPORTED: aarch64-unknown-linux-gnu-LTO
|
|
// UNSUPPORTED: x86_64-pc-linux-gnu
|
|
// UNSUPPORTED: x86_64-pc-linux-gnu-LTO
|
|
|
|
// This test validates that the OpenMP target reductions to find a minimum work
|
|
// as indended for a few common data types.
|
|
|
|
#include <algorithm>
|
|
#include <cassert>
|
|
#include <limits>
|
|
#include <vector>
|
|
|
|
template <class Tp> void test_min_idx_reduction() {
|
|
const Tp length = 1000;
|
|
const Tp nminimas = 8;
|
|
std::vector<float> a(length, 3.0f);
|
|
const Tp step = length / nminimas;
|
|
for (Tp i = 0; i < nminimas; i++) {
|
|
a[i * step] -= 1.0f;
|
|
}
|
|
for (Tp i = 0; i < nminimas; i++) {
|
|
Tp idx = a.size();
|
|
float *b = a.data();
|
|
#pragma omp target teams distribute parallel for reduction(min : idx) \
|
|
map(always, to : b[0 : length])
|
|
for (Tp j = 0; j < length - 1; j++) {
|
|
if (b[j] < b[j + 1]) {
|
|
idx = std::min(idx, j);
|
|
}
|
|
}
|
|
assert(idx == i * step &&
|
|
"#pragma omp target teams distribute parallel for "
|
|
"reduction(min:<identifier list>) does not work as intended.");
|
|
a[idx] += 1.0f;
|
|
}
|
|
}
|
|
|
|
template <class Tp> void test_min_val_reduction() {
|
|
const int length = 1000;
|
|
const int half = length / 2;
|
|
std::vector<Tp> a(length, (Tp)3);
|
|
a[half] -= (Tp)1;
|
|
Tp min_val = std::numeric_limits<Tp>::max();
|
|
Tp *b = a.data();
|
|
#pragma omp target teams distribute parallel for reduction(min : min_val) \
|
|
map(always, to : b[0 : length])
|
|
for (int i = 0; i < length; i++) {
|
|
min_val = std::min(min_val, b[i]);
|
|
}
|
|
assert(std::abs(((double)a[half + 1]) - ((double)min_val) - 1.0) < 1e-6 &&
|
|
"#pragma omp target teams distribute parallel for "
|
|
"reduction(min:<identifier list>) does not work as intended.");
|
|
}
|
|
|
|
int main() {
|
|
// Reducing over indices
|
|
test_min_idx_reduction<int>();
|
|
test_min_idx_reduction<unsigned int>();
|
|
test_min_idx_reduction<long>();
|
|
|
|
// Reducing over values
|
|
test_min_val_reduction<int>();
|
|
test_min_val_reduction<unsigned int>();
|
|
test_min_val_reduction<long>();
|
|
test_min_val_reduction<float>();
|
|
test_min_val_reduction<double>();
|
|
return 0;
|
|
}
|