feat(rocthrust): allow grafting thrust onto benchmarks

[Naraenda]

Motivation

Allows compilation of rocThrust benchmark with non-rocThrust libraries by adding some required compatibility layer and including said non-rocThrust header-only library.

Required by #8499.

JIRA ID : AIPRIMS-208

Technical Details

I've added a CMakePresets.json because putting the required flags in CMakeLists.txt is not the way to go. This also isn't user-facing so I don't want to include this in user facing documentation.

Test Plan

It's a secret.

Test Result

Works on my machine.

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

[github-actions]

This pull request has been inactive for 25 days and will be marked as stale.

If you would like to keep this PR open, please:

• Add new commits
• Add a comment explaining why it should remain open

This PR will be automatically closed in 5 days if no further activity occurs.

[therock-pr-bot]

❌ PR Check — Action Required

Check	Status	Details
🌿 Branch Name	✅ Pass	—
📝 PR Title/Description	✅ Pass	—
⛔ Forbidden Files	✅ Pass	—
🧪 Unit Test	❌ Fail	Error: Source/code files changed without an accompanying unit test. Expected: add at least one test file named like test_<name>.py / test_<name>.cpp (or <name>_test.*). Current: code file(s) changed: projects/rocthrust/benchmark/bench_utils/bench_utils.hpp, projects/rocthrust/benchmark/bench_utils/generation_utils.hpp, projects/rocthrust/benchmark/bench_utils/thrust_compat.hpp, projects/rocthrust/testing/unittest/meta.h; no test file found
🔎 pre-commit	✅ Pass	—
🚫 Draft PR	🔜 To Be Enabled	—
🚩 Feature Flag	🔜 To Be Enabled	—
📊 Code Coverage	🔜 To Be Enabled	—

⚠️ 1 policy check(s) failed. Please address the issues above before this PR can be Reviewed.

🚫 Please fix the failed policies

• ❌ Unit Test

The Not ready to Review label was added to this PR. Once all policies pass, the label is removed automatically.

📖 Need help? See the Policy FAQ for details on every check and how to fix failures.

[therock-pr-bot]

🚫 Please fix the failed policies before requesting reviews.

The following policy checks failed:

• ❌ Unit Test

The Not ready to Review label has been added to this PR.
Once all policies pass, the label will be removed automatically.

[MyNameIsTrez]

All the code changes look correct to me, thanks.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #8538      +/-   ##
===========================================
- Coverage    64.20%   64.20%   -0.00%     
===========================================
  Files         2761     2761              
  Lines       450031   450035       +4     
  Branches     66215    66215              
===========================================
- Hits        288929   288927       -2     
- Misses      140236   140240       +4     
- Partials     20866    20868       +2     

Flag	Coverage Δ		*Carryforward flag
TensileLite	34.35% <ø> (ø)		Carriedforward from 2dcef14
TensileLite-CPP	38.28% <ø> (ø)		Carriedforward from 2dcef14
TensileLite-Unit	35.52% <ø> (ø)		Carriedforward from 2dcef14
hipBLAS	90.62% <ø> (ø)		Carriedforward from 2dcef14
hipBLASLt	34.63% <ø> (ø)		Carriedforward from 2dcef14
hipCUB	82.68% <ø> (ø)		Carriedforward from 2dcef14
hipDNN	86.31% <ø> (ø)		Carriedforward from 2dcef14
hipFFT	50.30% <ø> (ø)		Carriedforward from 2dcef14
hipRAND	76.12% <ø> (ø)		Carriedforward from 2dcef14
hipSOLVER	69.18% <ø> (ø)		Carriedforward from 2dcef14
hipSPARSE	86.27% <ø> (ø)		Carriedforward from 2dcef14
rocBLAS	47.95% <ø> (ø)		Carriedforward from 2dcef14
rocFFT	49.16% <ø> (ø)		Carriedforward from 2dcef14
rocRAND	57.02% <ø> (+0.01%)	⬆️
rocSOLVER	76.84% <ø> (ø)		Carriedforward from 2dcef14
rocSPARSE	72.44% <ø> (ø)		Carriedforward from 2dcef14
rocThrust	91.34% <ø> (-0.01%)	⬇️

*This pull request uses carry forward flags. Click here to find out more.
see 6 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[NguyenNhuDi]

Using -DCODE_COVERAGE=ON -DBUILD_TEST=ON -DBUILD_BENCHMARK=ON while compiling for rocThrust I am getting the following error:

lld: error: undefined symbol: __llvm_profile_instrument_gpu
>>> referenced by /tmp/testframework-gfx1201-84f2d8.o:(dummy_kernel())
>>> referenced by /tmp/testframework-gfx1201-84f2d8.o:(dummy_kernel())
clang++: error: amdgcn-link command failed with exit code 1 (use -v to see invocation)
failed to execute:/opt/rocm/lib/llvm/bin/clang++  --offload-arch=gfx1201  -DTHRUST_DEVICE_SYSTEM=THRUST_DEVICE_SYSTEM_HIP -DTHRUST_HOST_SYSTEM=THRUST_HOST_SYSTEM_CPP -DUSE_LIBCUDACXX -DUSE_LIBHIPCXX -DUSE_PROF_API=1 -D__HIP_PLATFORM_AMD__=1 -I/dockerx/rocm-libraries/projects/rocthrust/build/thrust/include -I/dockerx/rocm-libraries/projects/rocthrust/thrust/.. -isystem /dockerx/rocm-libraries/projects/rocthrust/testing -isystem /dockerx/rocm-libraries/projects/rocthrust/build/_deps/googletest-src/googletest/include -isystem /dockerx/rocm-libraries/projects/rocthrust/build/_deps/googletest-src/googletest -w --offload-compress -Wno-unused-command-line-argument -O0 -fprofile-instr-generate -fcoverage-mapping -O3 -DNDEBUG -std=c++17 -Wall -Wextra -x hip -MD -MT testing/CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o -MF CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o.d -o "CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o" -c /dockerx/rocm-libraries/projects/rocthrust/testing/unittest/hip/testframework.cu
make[2]: *** [testing/CMakeFiles/testframework.dir/build.make:90: testing/CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o] Error 1
make[2]: *** Waiting for unfinished jobs....
lld: error: undefined symbol: __llvm_profile_instrument_gpu
>>> referenced by /tmp/test_unique_ptr_allocate_deallocate-gfx1201-a1b802.o:(void thrust::THRUST_200805_400400_NS::hip_rocprim::__parallel_for::kernel<256u, thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, 1u>(thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, int))
>>> referenced by /tmp/test_unique_ptr_allocate_deallocate-gfx1201-a1b802.o:(void thrust::THRUST_200805_400400_NS::hip_rocprim::__parallel_for::kernel<256u, thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, 1u>(thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, int))
>>> referenced by /tmp/test_unique_ptr_allocate_deallocate-gfx1201-a1b802.o:(void thrust::THRUST_200805_400400_NS::hip_rocprim::__parallel_for::kernel<256u, thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, 1u>(thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, int))
>>> referenced 41 more times

[Naraenda]

Using -DCODE_COVERAGE=ON -DBUILD_TEST=ON -DBUILD_BENCHMARK=ON while compiling for rocThrust I am getting the following error:

lld: error: undefined symbol: __llvm_profile_instrument_gpu
>>> referenced by /tmp/testframework-gfx1201-84f2d8.o:(dummy_kernel())
>>> referenced by /tmp/testframework-gfx1201-84f2d8.o:(dummy_kernel())
clang++: error: amdgcn-link command failed with exit code 1 (use -v to see invocation)
failed to execute:/opt/rocm/lib/llvm/bin/clang++  --offload-arch=gfx1201  -DTHRUST_DEVICE_SYSTEM=THRUST_DEVICE_SYSTEM_HIP -DTHRUST_HOST_SYSTEM=THRUST_HOST_SYSTEM_CPP -DUSE_LIBCUDACXX -DUSE_LIBHIPCXX -DUSE_PROF_API=1 -D__HIP_PLATFORM_AMD__=1 -I/dockerx/rocm-libraries/projects/rocthrust/build/thrust/include -I/dockerx/rocm-libraries/projects/rocthrust/thrust/.. -isystem /dockerx/rocm-libraries/projects/rocthrust/testing -isystem /dockerx/rocm-libraries/projects/rocthrust/build/_deps/googletest-src/googletest/include -isystem /dockerx/rocm-libraries/projects/rocthrust/build/_deps/googletest-src/googletest -w --offload-compress -Wno-unused-command-line-argument -O0 -fprofile-instr-generate -fcoverage-mapping -O3 -DNDEBUG -std=c++17 -Wall -Wextra -x hip -MD -MT testing/CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o -MF CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o.d -o "CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o" -c /dockerx/rocm-libraries/projects/rocthrust/testing/unittest/hip/testframework.cu
make[2]: *** [testing/CMakeFiles/testframework.dir/build.make:90: testing/CMakeFiles/testframework.dir/unittest/hip/testframework.cu.o] Error 1
make[2]: *** Waiting for unfinished jobs....
lld: error: undefined symbol: __llvm_profile_instrument_gpu
>>> referenced by /tmp/test_unique_ptr_allocate_deallocate-gfx1201-a1b802.o:(void thrust::THRUST_200805_400400_NS::hip_rocprim::__parallel_for::kernel<256u, thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, 1u>(thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, int))
>>> referenced by /tmp/test_unique_ptr_allocate_deallocate-gfx1201-a1b802.o:(void thrust::THRUST_200805_400400_NS::hip_rocprim::__parallel_for::kernel<256u, thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, 1u>(thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, int))
>>> referenced by /tmp/test_unique_ptr_allocate_deallocate-gfx1201-a1b802.o:(void thrust::THRUST_200805_400400_NS::hip_rocprim::__parallel_for::kernel<256u, thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, 1u>(thrust::THRUST_200805_400400_NS::hip_rocprim::for_each_f<thrust::THRUST_200805_400400_NS::device_ptr<A>, thrust::THRUST_200805_400400_NS::detail::wrapped_function<thrust::THRUST_200805_400400_NS::default_delete<A, void>::operator()(thrust::THRUST_200805_400400_NS::device_ptr<A>) const::'lambda'(A&), void>>, int, int))
>>> referenced 41 more times

Ah, weird. I think something must have gone wrong during a rebase, or conflict resolution. Should be resolved with b76aa95.