Merge branch 'main' into sum-pooling-bench

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,27 @@
+---
+name: Bug report
+about: Create a report to help us improve
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+**Describe the bug**
+A clear and concise description of what the bug is.
+
+**To Reproduce**
+Steps to reproduce the behavior.
+
+**Expected behavior**
+A clear and concise description of what you expected to happen.
+
+**Screenshots**
+If applicable, add screenshots to help explain your problem.
+
+**Desktop (please complete the following information):**
+ - OS: [e.g. Ubuntu]
+ - Version [e.g. 20.04]
+
+**Additional context**
+Add any other context about the problem here.

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,20 @@
+---
+name: Feature request
+about: Suggest an idea for this project
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+**Is your feature request related to a problem? Please describe.**
+A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
+
+**Describe the solution you'd like**
+A clear and concise description of what you want to happen.
+
+**Describe alternatives you've considered**
+A clear and concise description of any alternative solutions or features you've considered.
+
+**Additional context**
+Add any other context about the feature request here.

CMakeLists.txt

@@ -16,7 +16,7 @@ endif()
 
 project(buddy-benchmark LANGUAGES CXX C)
 
-set(CMAKE_CXX_STANDARD 14)
+set(CMAKE_CXX_STANDARD 17)
 set(CMAKE_CXX_STANDARD_REQUIRED YES)
 set(CMAKE_CXX_FLAGS "-no-pie")
 set(CMAKE_C_FLAGS "-no-pie")
@@ -104,15 +104,19 @@ endif()
 # Find OpenCV
 #-------------------------------------------------------------------------------
 
-find_package(OpenCV REQUIRED CONFIG)
-include_directories(${OpenCV_INCLUDE_DIRS})
+if(DEFINED IMAGE_PROCESSING_BENCHMARKS OR DEEP_LEARNING_BENCHMARKS)
+  find_package(OpenCV REQUIRED CONFIG)
+  include_directories(${OpenCV_INCLUDE_DIRS})
+endif()
 
 #-------------------------------------------------------------------------------
 # Find PNG
 #-------------------------------------------------------------------------------
 
-find_package(PNG REQUIRED)
-include_directories(${PNG_INCLUDE_DIR})
+if(DEFINED IMAGE_PROCESSING_BENCHMARKS OR DEEP_LEARNING_BENCHMARKS)
+  find_package(PNG REQUIRED)
+  include_directories(${PNG_INCLUDE_DIR})
+endif()
 
 #-------------------------------------------------------------------------------
 # Hardware detection

README.md

@@ -94,6 +94,24 @@ Run the DepthwiseConv2DNhwcHwc operation benchmark:
 $ cd <path to build>/bin && ./depthwise-conv-2d-nhwc-hwc-benchmark
 ```
 
+## Audio Processing Benchmark
+
+Currently, the audio processing benchmark includes the following frameworks or optimizers:
+
+- KFR ([link](https://github.com/kfrlib/kfr))
+
+*Note: Please replace the `/PATH/TO/*` with your local path.*
+
+```
+$ cd buddy-benchmark
+$ mkdir build && cd build
+$ cmake -G Ninja .. \
+    -DAUDIO_PROCESSING_BENCHMARKS=ON \
+    -DCMAKE_CXX_COMPILER=clang++ \
+    -DKFR_DIR=/PATH/TO/KFR/SOURCE/CODE \
+$ ninja audio-processing-benchmark
+```
+
 ## Testing
 
 ```

benchmarks/AudioProcessing/CMakeLists.txt

@@ -0,0 +1,20 @@
+#-------------------------------------------------------------------------------
+# Audio Processing Benchmark Target
+#-------------------------------------------------------------------------------
+
+add_subdirectory(${KFR_DIR} ./kfr)
+
+include_directories(${KFR_DIR}/include)
+
+add_executable(audio-processing-benchmark
+  KFRFir.cpp 
+  Main.cpp
+)
+
+target_link_directories(audio-processing-benchmark PRIVATE ${KFR_DIR}/build/)
+target_link_libraries(audio-processing-benchmark
+  PRIVATE
+  kfr_io
+  kfr_dft
+  GoogleBenchmark
+)

benchmarks/AudioProcessing/KFRFir.cpp

@@ -0,0 +1,70 @@
+//===- KFRFir.cpp ---------------------------------------------------------===//
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the benchmark for KFR Fir function.
+//
+//===----------------------------------------------------------------------===//
+
+#include <benchmark/benchmark.h>
+#include <kfr/base.hpp>
+#include <kfr/dft.hpp>
+#include <kfr/dsp.hpp>
+#include <kfr/io.hpp>
+
+using namespace kfr;
+
+univector<fbase, 1023> taps127;
+univector<float, 2000000> aud;
+univector<float> result;
+
+// Initialize univector.
+void initializeKFRFir() {
+  expression_pointer<fbase> kaiser =
+      to_pointer(window_kaiser(taps127.size(), 3.0));
+  fir_lowpass(taps127, 0.2, kaiser, true);
+  audio_reader_wav<float> reader(open_file_for_reading(
+      "../../benchmarks/AudioProcessing/Audios/NASA_Mars.wav"));
+  reader.read(aud.data(), aud.size());
+}
+
+// Benchmarking function.
+static void KFR_FIR(benchmark::State &state) {
+  for (auto _ : state) {
+    for (int i = 0; i < state.range(0); ++i) {
+      result = kfr::fir(aud, taps127);
+    }
+  }
+}
+
+// Register benchmarking function.
+BENCHMARK(KFR_FIR)->Arg(1);
+
+// Generate result wav file.
+void generateResultKFRFir() {
+  univector<float> generateResult = kfr::fir(aud, taps127);
+
+  audio_writer_wav<float> writer(open_file_for_writing("./ResultKFRFir.wav"),
+                                 audio_format{1 /* channel */,
+                                              audio_sample_type::i24,
+                                              100000 /* sample rate */});
+  writer.write(generateResult.data(), generateResult.size());
+  println("Sample Rate  = ", writer.format().samplerate);
+  println("Channels     = ", writer.format().channels);
+  println("Length       = ", writer.format().length);
+  println("Duration (s) = ",
+          writer.format().length / writer.format().samplerate);
+  println("Bit depth    = ", audio_sample_bit_depth(writer.format().type));
+}

benchmarks/AudioProcessing/Main.cpp

@@ -0,0 +1,33 @@
+//===- Main.cpp -----------------------------------------------------------===//
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+//
+//===----------------------------------------------------------------------===//
+//
+// This is the main file of the audio processing benchmark.
+//
+//===----------------------------------------------------------------------===//
+
+#include <benchmark/benchmark.h>
+
+void initializeKFRFir();
+
+void generateResultKFRFir();
+
+int main(int argc, char **argv) {
+  initializeKFRFir();
+  ::benchmark::Initialize(&argc, argv);
+  ::benchmark::RunSpecifiedBenchmarks();
+  generateResultKFRFir();
+  return 0;
+}

benchmarks/CMakeLists.txt

@@ -5,3 +5,7 @@ endif()
 if(DEEP_LEARNING_BENCHMARKS)
   add_subdirectory(DeepLearning)
 endif()
+
+if(AUDIO_PROCESSING_BENCHMARKS)
+  add_subdirectory(AudioProcessing)
+endif()

benchmarks/DeepLearning/Models/Inception-V3/CMakeLists.txt

@@ -18,7 +18,7 @@ add_buddy_model_benchmark(inception-v3-benchmark
     -memref-expand -arith-expand 
     --convert-memref-to-llvm 
     --convert-math-to-llvm 
-    --convert-std-to-llvm='emit-c-wrappers=1'
+    --convert-func-to-llvm='emit-c-wrappers=1'
     --reconcile-unrealized-casts
   SOURCE Main.cpp InceptionBenchmark.cpp
 )

benchmarks/DeepLearning/Models/MobileNet-V2/CMakeLists.txt

@@ -18,7 +18,7 @@ add_buddy_model_benchmark(mobilenet-benchmark
     -memref-expand -arith-expand 
     --convert-memref-to-llvm 
     --convert-math-to-llvm 
-    --convert-std-to-llvm='emit-c-wrappers=1'
+    --convert-func-to-llvm='emit-c-wrappers=1'
     --reconcile-unrealized-casts
   SOURCE Main.cpp MobileNetBenchmark.cpp
 )

benchmarks/DeepLearning/Models/MobileNet-V3/CMakeLists.txt

@@ -18,7 +18,7 @@ add_buddy_model_benchmark(mobilenet-v3-benchmark
     -memref-expand -arith-expand 
     --convert-memref-to-llvm 
     --convert-math-to-llvm 
-    --convert-std-to-llvm='emit-c-wrappers=1'
+    --convert-func-to-llvm='emit-c-wrappers=1'
     --reconcile-unrealized-casts
   SOURCE Main.cpp MobileNetBenchmark.cpp
 )

benchmarks/DeepLearning/Models/ResNet-V2-50/CMakeLists.txt

@@ -18,7 +18,7 @@ add_buddy_model_benchmark(resnet_v2_50-benchmark
     -memref-expand -arith-expand 
     --convert-memref-to-llvm 
     --convert-math-to-llvm 
-    --convert-std-to-llvm='emit-c-wrappers=1'
+    --convert-func-to-llvm='emit-c-wrappers=1'
     --reconcile-unrealized-casts
   SOURCE Main.cpp ResNetBenchmark.cpp
 )

benchmarks/DeepLearning/Ops/Conv2DNchwFchwOp/CMakeLists.txt

@@ -8,7 +8,7 @@ add_buddy_ops_benchmark(conv-2d-nchw-fchw-benchmark
     -convert-linalg-to-llvm 
     -lower-affine 
     --convert-memref-to-llvm 
-    -convert-std-to-llvm='emit-c-wrappers=1' 
+    -convert-func-to-llvm='emit-c-wrappers=1' 
     -reconcile-unrealized-casts
   SOURCE Main.cpp MLIROptBenchmark.cpp
 )

benchmarks/DeepLearning/Ops/Conv2DNhwcHwcfOp/CMakeLists.txt

@@ -8,7 +8,7 @@ add_buddy_ops_benchmark(conv-2d-nhwc-hwcf-benchmark
     -convert-linalg-to-llvm 
     -lower-affine 
     --convert-memref-to-llvm 
-    -convert-std-to-llvm='emit-c-wrappers=1' 
+    -convert-func-to-llvm='emit-c-wrappers=1' 
     -reconcile-unrealized-casts
   SOURCE Main.cpp MLIROptBenchmark.cpp
 )

benchmarks/DeepLearning/Ops/DepthwiseConv2DNhwcHwcOp/CMakeLists.txt

@@ -8,7 +8,7 @@ add_buddy_ops_benchmark(depthwise-conv-2d-nhwc-hwc-benchmark
     -convert-linalg-to-llvm 
     -lower-affine 
     --convert-memref-to-llvm 
-    -convert-std-to-llvm='emit-c-wrappers=1' 
+    -convert-func-to-llvm='emit-c-wrappers=1' 
     -reconcile-unrealized-casts
   SOURCE Main.cpp MLIROptBenchmark.cpp
 )

benchmarks/ImageProcessing/CMakeLists.txt

@@ -25,11 +25,10 @@ message(STATUS "Configuring MLIR Linalg Tile Size: ${MLIR_LINALG_TILE}")
 add_custom_command(OUTPUT mlir-conv2d.o
   COMMAND ${LLVM_MLIR_BINARY_DIR}/mlir-opt
           ${BUDDY_SOURCE_DIR}/benchmarks/ImageProcessing/MLIRConv2D.mlir 
-            -linalg-tile="tile-sizes=${MLIR_LINALG_TILE},${MLIR_LINALG_TILE}"
             -convert-linalg-to-loops
             -convert-scf-to-cf -convert-linalg-to-llvm
             -lower-affine -convert-scf-to-cf --convert-memref-to-llvm 
-            -convert-std-to-llvm='emit-c-wrappers=1'
+            -convert-func-to-llvm='emit-c-wrappers=1'
             -reconcile-unrealized-casts | 
           ${LLVM_MLIR_BINARY_DIR}/mlir-translate --mlir-to-llvmir |
           ${LLVM_MLIR_BINARY_DIR}/llc -mtriple=${BUDDY_OPT_TRIPLE} 
@@ -49,7 +48,7 @@ add_custom_command(OUTPUT buddy-conv2d.o
             -conv-vectorization="strip-mining=${SPLITING_SIZE}" 
             -convert-vector-to-scf 
             -lower-affine -convert-scf-to-cf -convert-vector-to-llvm 
-            -convert-memref-to-llvm -convert-std-to-llvm='emit-c-wrappers=1' 
+            -convert-memref-to-llvm -convert-func-to-llvm='emit-c-wrappers=1' 
             -reconcile-unrealized-casts | 
           ${LLVM_MLIR_BINARY_DIR}/mlir-translate --mlir-to-llvmir |
           ${LLVM_MLIR_BINARY_DIR}/llc -mtriple=${BUDDY_OPT_TRIPLE} 
@@ -59,6 +58,23 @@ add_custom_command(OUTPUT buddy-conv2d.o
 add_library(BuddyConv2D STATIC buddy-conv2d.o)
 set_target_properties(BuddyConv2D PROPERTIES LINKER_LANGUAGE CXX)
 
+# add_custom_command(OUTPUT buddy-conv2d.o
+#   COMMAND ${BUDDY_OPT_BUILD_DIR}/bin/buddy-opt
+#           ${BUDDY_SOURCE_DIR}/benchmarks/ImageProcessing/BuddyConv2D.mlir 
+#             -linalg-tile="tile-sizes=${MLIR_LINALG_TILE},${MLIR_LINALG_TILE}"
+#             -conv-vectorization="tile-sizes=${MLIR_LINALG_TILE},${MLIR_LINALG_TILE}" 
+#             -convert-vector-to-scf 
+#             -lower-affine -convert-scf-to-cf -convert-vector-to-llvm 
+#             -convert-memref-to-llvm -convert-func-to-llvm='emit-c-wrappers=1' 
+#             -reconcile-unrealized-casts | 
+#           ${LLVM_MLIR_BINARY_DIR}/mlir-translate --mlir-to-llvmir |
+#           ${LLVM_MLIR_BINARY_DIR}/llc -mtriple=${BUDDY_OPT_TRIPLE} 
+#             -mattr=${BUDDY_OPT_ATTR} --filetype=obj 
+#             -o ${BUDDY_BINARY_DIR}/../benchmarks/ImageProcessing/buddy-conv2d.o
+# )
+# add_library(BuddyConv2D STATIC buddy-conv2d.o)
+# set_target_properties(BuddyConv2D PROPERTIES LINKER_LANGUAGE CXX)
+
 #-------------------------------------------------------------------------------
 # Buddy DIP Dialect Corr2D Operation + CB Algorithm
 #-------------------------------------------------------------------------------
@@ -68,7 +84,7 @@ add_custom_command(OUTPUT buddy-corr2d.o
           ${BUDDY_SOURCE_DIR}/benchmarks/ImageProcessing/BuddyCorr2D.mlir 
             -lower-dip="DIP-strip-mining=${SPLITING_SIZE}" 
             -lower-affine -convert-scf-to-cf -convert-vector-to-llvm 
-            -convert-memref-to-llvm -convert-std-to-llvm='emit-c-wrappers=1' 
+            -convert-memref-to-llvm -convert-func-to-llvm='emit-c-wrappers=1' 
             -reconcile-unrealized-casts | 
           ${LLVM_MLIR_BINARY_DIR}/mlir-translate --mlir-to-llvmir |
           ${LLVM_MLIR_BINARY_DIR}/llc -mtriple=${BUDDY_OPT_TRIPLE} 

benchmarks/ImageProcessing/Main.cpp

@@ -19,6 +19,7 @@
 //===----------------------------------------------------------------------===//
 
 #include <benchmark/benchmark.h>
+#include <stdexcept>
 
 void initializeMLIRConv2D(char **);
 void initializeBuddyConv2D(char **);