Add build-time flag to enable OpenMP in Lightning gate kernels (#510)

* Add OMP to CNOT and CZ AVX kernels * Add OMP to CNOT and CZ to LM kernels * Add pragma-controlled parallelism to gate kernels for LQ * Add pragma-controlled parallelism to gate kernels for LQ +1 * Add compile-time generation for gate kernel OMP threading in LQ * Rename OMP macros and add SIMD annotations to short loops * Update macros & uses * Remove simd macro from constexpr/consteval * Remove simd from additional constexpr/consteval operations * Update changelog * Auto update version * Fix LM kernels to support changes; remove PI OMP * Fix format and CI test builds * Auto update version * Auto update version * Update pennylane_lightning/core/src/simulators/lightning_qubit/gates/cpu_kernels/GatePragmas.hpp Co-authored-by: Ali Asadi <[email protected]> * Auto update version * Trigger CI * Improve kernel testing with OpenMP * Add missing directory return in CI * Auto update version * Trigger CI * Add my name to the changelog and remove redundant comment * Update pennylane_lightning/core/src/simulators/lightning_qubit/gates/cpu_kernels/GatePragmas.hpp Co-authored-by: Amintor Dusko <[email protected]> --------- Co-authored-by: Dev version update bot <github-actions[bot]@users.noreply.github.com> Co-authored-by: Ali Asadi <[email protected]> Co-authored-by: Amintor Dusko <[email protected]>
PennyLaneAI · Jan 12, 2024 · ac8795d · ac8795d
1 parent f591388
commit ac8795d
Show file tree

Hide file tree

Showing 39 changed files with 264 additions and 152 deletions.
diff --git a/.github/CHANGELOG.md b/.github/CHANGELOG.md
@@ -9,6 +9,9 @@
 
 ### Improvements
 
+* OpenMP acceleration can now be enabled at compile time for all `lightning.qubit` gate kernels using the "-DLQ_ENABLE_KERNEL_OMP=1" CMake argument.
+  [(#510)](https://github.com/PennyLaneAI/pennylane-lightning/pull/510)
+
 ### Documentation
 
 ### Bug fixes
@@ -17,6 +20,7 @@
 
 This release contains contributions from (in alphabetical order):
 
+Lee J. O'Riordan,
 Shuli Shu
 
 ---
@@ -58,7 +62,7 @@ Shuli Shu
   [(#540)] (https://github.com/PennyLaneAI/pennylane-lightning/pull/540)
 
 ### Improvements
-
+ 
 * Ensure aligned memory used for numpy arrays with state-vector without reallocations.
   [(#572)](https://github.com/PennyLaneAI/pennylane-lightning/pull/572)
 

diff --git a/.github/workflows/tests_linux.yml b/.github/workflows/tests_linux.yml
@@ -59,27 +59,46 @@ jobs:
               -DPL_BACKEND=${{ matrix.pl_backend }} \
               -DCMAKE_CXX_COMPILER=$(which g++-$GCC_VERSION) \
               -DENABLE_COVERAGE=ON
+
+            cmake . -BBuildKernelOMP -G Ninja \
+              -DCMAKE_BUILD_TYPE=Debug \
+              -DBUILD_TESTS=ON \
+              -DENABLE_PYTHON=OFF \
+              -DPL_BACKEND=${{ matrix.pl_backend }} \
+              -DCMAKE_CXX_COMPILER=$(which g++-$GCC_VERSION) \
+              -DENABLE_COVERAGE=ON \
+              -DLQ_ENABLE_KERNEL_OMP=ON
+
             cmake --build ./Build
-            cd ./Build
-            mkdir -p ./tests/results
-            for file in *runner ; do ./$file --order lex --reporter junit --out ./tests/results/report_$file.xml; done;
-            lcov --directory . -b ../pennylane_lightning/core/src --capture --output-file coverage.info
-            lcov --remove coverage.info '/usr/*' --output-file coverage.info
+            cmake --build ./BuildKernelOMP
+
+            for d in Build BuildKernelOMP; do
+              cd ./$d
+              mkdir -p ./tests/results
+              for file in *runner ; do ./$file --order lex --reporter junit --out ./tests/results/report_$file.xml; done;
+              lcov --directory . -b ../pennylane_lightning/core/src --capture --output-file coverage.info
+              lcov --remove coverage.info '/usr/*' --output-file coverage.info
+              cd ..
+            done
+            lcov --add-tracefile ./Build/coverage.info -a ./BuildKernelOMP/coverage.info -o coverage.info
             mv coverage.info coverage-${{ github.job }}-${{ matrix.pl_backend }}.info
 
       - name: Upload test results
         uses: actions/upload-artifact@v3
         if: always()
         with:
           name: ubuntu-tests-reports-${{ github.job }}-${{ matrix.pl_backend }}
-          path: ./Build/tests/results/
+          path: |
+            ./Build/tests/results/
+            ./BuildKernelOMP/tests/results/
+
           if-no-files-found: error
 
       - name: Upload code coverage results
         uses: actions/upload-artifact@v3
         with:
           name: ubuntu-codecov-results-cpp
-          path: ./Build/coverage-${{ github.job }}-${{ matrix.pl_backend }}.info
+          path: ./coverage-${{ github.job }}-${{ matrix.pl_backend }}.info
           if-no-files-found: error
 
   pythontests:
@@ -158,7 +177,7 @@ jobs:
       - name: Install backend device
         run: |
           cd main
-          CMAKE_ARGS="-DPL_BACKEND=${{ matrix.pl_backend }} -DENABLE_PYTHON=ON -DCMAKE_CXX_COMPILER=$(which g++-$GCC_VERSION)" \
+          CMAKE_ARGS="-DPL_BACKEND=${{ matrix.pl_backend }} -DLQ_ENABLE_KERNEL_OMP=ON -DENABLE_PYTHON=ON -DCMAKE_CXX_COMPILER=$(which g++-$GCC_VERSION)" \
           python -m pip install -e . -vv
 
       - name: Run PennyLane-Lightning unit tests
@@ -265,7 +284,8 @@ jobs:
               -DBUILD_TESTS=ON \
               -DENABLE_LAPACK=OFF \
               -DCMAKE_CXX_COMPILER=$(which g++-$GCC_VERSION) \
-              -DENABLE_COVERAGE=ON
+              -DENABLE_COVERAGE=ON \
+              -DLQ_ENABLE_KERNEL_OMP=ON
             cmake --build ./Build
             cd ./Build
             mkdir -p ./tests/results

diff --git a/pennylane_lightning/core/_version.py b/pennylane_lightning/core/_version.py
@@ -16,4 +16,4 @@
    Version number (major.minor.patch[-label])
 """
 
-__version__ = "0.35.0-dev0"
+__version__ = "0.35.0-dev1"
diff --git a/pennylane_lightning/core/src/simulators/lightning_qubit/CMakeLists.txt b/pennylane_lightning/core/src/simulators/lightning_qubit/CMakeLists.txt
@@ -20,6 +20,7 @@ add_library(lightning_qubit STATIC ${LQUBIT_FILES})
 
 option(ENABLE_BLAS "Enable BLAS" OFF)
 option(ENABLE_GATE_DISPATCHER "Enable gate kernel dispatching on AVX/AVX2/AVX512" ON)
+option(LQ_ENABLE_KERNEL_OMP "Enable OpenMP pragmas for gate kernels" OFF)
 
 # Inform the compiler that this device is enabled.
 target_compile_options(lightning_compile_options INTERFACE "-D_ENABLE_PLQUBIT=1")
@@ -46,6 +47,10 @@ else()
     message(STATUS "ENABLE_BLAS is OFF.")
 endif()
 
+if(LQ_ENABLE_KERNEL_OMP)
+    add_definitions("-DPL_LQ_KERNEL_OMP")
+endif()
+
 target_link_libraries(lightning_qubit PUBLIC    lightning_compile_options
                                                 lightning_external_libs
                                                 lightning_base

diff --git a/...ng/core/src/simulators/lightning_qubit/gates/cpu_kernels/GateImplementationsAVXCommon.hpp b/...ng/core/src/simulators/lightning_qubit/gates/cpu_kernels/GateImplementationsAVXCommon.hpp
@@ -23,6 +23,7 @@
 #include "Error.hpp"
 #include "GateImplementationsLM.hpp"
 #include "GateOperation.hpp"
+#include "GatePragmas.hpp"
 #include "Gates.hpp"
 #include "KernelType.hpp"
 #include "LinearAlgebra.hpp"