Skip to content

Commit

Permalink
Update openmp installation methods and M1 benchmarks (Tencent#3107)
Browse files Browse the repository at this point in the history 
* Update M1 benchmark

* Update openmp installation and use method

* Fixing typo

* And more info about Xcode
  • Loading branch information
@zhiqwang
zhiqwang authored Jul 25, 2021
1 parent d91cccf commit 225e0fe
Show file tree
Hide file tree
Showing 2 changed files with 71 additions and 101 deletions.
128 changes: 66 additions & 62 deletions benchmark/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1721,74 +1721,78 @@ num_threads = 1
powersave = 0
gpu_device = -1
cooling_down = 0
squeezenet min = 4.79 max = 4.98 avg = 4.84
squeezenet_int8 min = 10.10 max = 10.17 avg = 10.13
mobilenet min = 9.12 max = 9.48 avg = 9.37
mobilenet_int8 min = 13.68 max = 13.71 avg = 13.70
mobilenet_v2 min = 5.64 max = 5.89 avg = 5.73
mobilenet_v3 min = 4.79 max = 4.88 avg = 4.84
shufflenet min = 3.65 max = 3.76 avg = 3.67
shufflenet_v2 min = 3.92 max = 3.93 avg = 3.93
mnasnet min = 5.75 max = 5.92 avg = 5.83
proxylessnasnet min = 7.34 max = 7.54 avg = 7.50
efficientnet_b0 min = 9.25 max = 9.55 avg = 9.40
regnety_400m min = 9.39 max = 9.45 avg = 9.41
blazeface min = 1.89 max = 1.92 avg = 1.91
googlenet min = 25.55 max = 25.91 avg = 25.65
googlenet_int8 min = 39.34 max = 39.80 avg = 39.48
resnet18 min = 16.89 max = 17.18 avg = 16.99
resnet18_int8 min = 30.98 max = 31.92 avg = 31.24
alexnet min = 30.90 max = 31.35 avg = 31.09
vgg16 min = 74.09 max = 76.50 avg = 75.19
vgg16_int8 min = 212.09 max = 219.74 avg = 215.01
resnet50 min = 43.59 max = 44.04 avg = 43.75
resnet50_int8 min = 69.55 max = 71.29 avg = 70.33
squeezenet_ssd min = 20.97 max = 21.41 avg = 21.11
squeezenet_ssd_int8 min = 33.53 max = 34.74 avg = 33.98
mobilenet_ssd min = 22.03 max = 22.79 avg = 22.28
mobilenet_ssd_int8 min = 28.83 max = 29.71 avg = 29.10
mobilenet_yolo min = 40.67 max = 42.66 avg = 41.24
mobilenetv2_yolov3 min = 20.89 max = 22.23 avg = 21.33
yolov4-tiny min = 30.70 max = 32.17 avg = 30.94
MacBook-Pro benchmark % ./benchncnn 10 8 0 0 0
squeezenet min = 4.80 max = 5.05 avg = 4.86
squeezenet_int8 min = 4.02 max = 4.13 avg = 4.04
mobilenet min = 9.09 max = 9.41 avg = 9.22
mobilenet_int8 min = 4.65 max = 4.76 avg = 4.70
mobilenet_v2 min = 5.64 max = 5.83 avg = 5.73
mobilenet_v3 min = 4.64 max = 4.85 avg = 4.76
shufflenet min = 3.48 max = 3.63 avg = 3.56
shufflenet_v2 min = 3.69 max = 3.81 avg = 3.73
mnasnet min = 5.67 max = 5.94 avg = 5.77
proxylessnasnet min = 7.03 max = 7.28 avg = 7.20
efficientnet_b0 min = 9.13 max = 9.53 avg = 9.28
efficientnetv2_b0 min = 17.37 max = 18.47 avg = 17.63
regnety_400m min = 7.64 max = 8.08 avg = 7.72
blazeface min = 1.80 max = 1.89 avg = 1.83
googlenet min = 25.71 max = 25.90 avg = 25.81
googlenet_int8 min = 16.89 max = 17.10 avg = 16.97
resnet18 min = 17.16 max = 17.28 avg = 17.20
resnet18_int8 min = 15.55 max = 15.75 avg = 15.64
alexnet min = 30.60 max = 31.11 avg = 30.69
vgg16 min = 73.41 max = 75.37 avg = 73.91
vgg16_int8 min = 103.81 max = 105.15 avg = 104.19
resnet50 min = 43.47 max = 44.24 avg = 43.68
resnet50_int8 min = 30.37 max = 35.25 avg = 31.61
squeezenet_ssd min = 20.97 max = 21.21 avg = 21.12
squeezenet_ssd_int8 min = 19.34 max = 19.54 avg = 19.42
mobilenet_ssd min = 22.18 max = 22.58 avg = 22.28
mobilenet_ssd_int8 min = 13.27 max = 15.31 avg = 14.05
mobilenet_yolo min = 40.78 max = 41.04 avg = 40.89
mobilenetv2_yolov3 min = 20.87 max = 21.92 avg = 21.02
yolov4-tiny min = 30.73 max = 32.37 avg = 31.29
nanodet_m min = 8.54 max = 8.86 avg = 8.65
MacBook-Pro benchmark % ./benchncnn 10 8 0 0 0
[0 Apple M1] queueC=0[1] queueG=0[1] queueT=0[1]
[0 Apple M1] bugsbn1=0 bugbilz=97 bugcopc=0 bugihfa=0
[0 Apple M1] bugsbn1=0 bugbilz=151 bugcopc=0 bugihfa=0
[0 Apple M1] fp16-p/s/a=1/1/1 int8-p/s/a=1/1/1
[0 Apple M1] subgroup=32 basic=1 vote=1 ballot=1 shuffle=1
loop_count = 10
num_threads = 8
powersave = 0
gpu_device = 0
cooling_down = 0
squeezenet min = 1.90 max = 2.60 avg = 2.08
squeezenet_int8 min = 6.17 max = 10.19 avg = 7.95
mobilenet min = 2.52 max = 3.05 avg = 2.74
mobilenet_int8 min = 7.26 max = 9.46 avg = 8.17
mobilenet_v2 min = 2.84 max = 3.36 avg = 3.03
mobilenet_v3 min = 3.49 max = 4.24 avg = 3.79
shufflenet min = 1.97 max = 2.44 avg = 2.16
shufflenet_v2 min = 2.43 max = 3.00 avg = 2.60
mnasnet min = 2.55 max = 3.42 avg = 2.89
proxylessnasnet min = 3.08 max = 3.70 avg = 3.29
efficientnet_b0 min = 4.78 max = 5.97 avg = 5.18
regnety_400m min = 4.04 max = 4.63 avg = 4.22
blazeface min = 1.06 max = 1.40 avg = 1.20
googlenet min = 4.92 max = 5.72 avg = 5.22
googlenet_int8 min = 20.62 max = 26.34 avg = 23.78
resnet18 min = 3.83 max = 4.46 avg = 4.18
resnet18_int8 min = 15.37 max = 19.79 avg = 16.95
alexnet min = 7.29 max = 7.64 avg = 7.48
vgg16 min = 32.43 max = 33.01 avg = 32.68
vgg16_int8 min = 77.96 max = 84.55 avg = 80.97
resnet50 min = 7.40 max = 8.02 avg = 7.68
resnet50_int8 min = 37.31 max = 44.22 avg = 40.42
squeezenet_ssd min = 7.24 max = 8.37 avg = 7.75
squeezenet_ssd_int8 min = 20.78 max = 26.64 avg = 23.11
mobilenet_ssd min = 4.79 max = 5.96 avg = 5.20
mobilenet_ssd_int8 min = 17.72 max = 21.80 avg = 19.39
mobilenet_yolo min = 6.70 max = 8.50 avg = 7.64
mobilenetv2_yolov3 min = 7.82 max = 9.02 avg = 8.27
yolov4-tiny min = 12.15 max = 13.90 avg = 12.87
squeezenet min = 1.86 max = 2.22 avg = 2.01
squeezenet_int8 min = 2.38 max = 8.40 avg = 5.13
mobilenet min = 2.50 max = 2.91 avg = 2.64
mobilenet_int8 min = 2.29 max = 5.26 avg = 3.54
mobilenet_v2 min = 2.93 max = 3.12 avg = 2.98
mobilenet_v3 min = 3.36 max = 3.61 avg = 3.48
shufflenet min = 1.99 max = 2.54 avg = 2.18
shufflenet_v2 min = 2.35 max = 2.84 avg = 2.52
mnasnet min = 2.81 max = 3.33 avg = 2.92
proxylessnasnet min = 3.21 max = 3.62 avg = 3.36
efficientnet_b0 min = 4.74 max = 5.73 avg = 5.07
efficientnetv2_b0 min = 12.04 max = 13.04 avg = 12.61
regnety_400m min = 3.86 max = 4.04 avg = 3.98
blazeface min = 0.98 max = 1.11 avg = 1.03
googlenet min = 4.86 max = 5.38 avg = 5.02
googlenet_int8 min = 9.43 max = 15.72 avg = 10.44
resnet18 min = 3.92 max = 4.59 avg = 4.24
resnet18_int8 min = 6.83 max = 7.57 avg = 7.35
alexnet min = 7.49 max = 7.87 avg = 7.65
vgg16 min = 34.10 max = 35.29 avg = 34.60
vgg16_int8 min = 40.09 max = 44.66 avg = 41.95
resnet50 min = 7.22 max = 7.83 avg = 7.42
resnet50_int8 min = 14.52 max = 20.56 avg = 15.78
squeezenet_ssd min = 8.52 max = 13.79 avg = 9.98
squeezenet_ssd_int8 min = 12.38 max = 15.44 avg = 13.37
mobilenet_ssd min = 4.83 max = 6.00 avg = 5.31
mobilenet_ssd_int8 min = 7.26 max = 13.12 avg = 9.01
mobilenet_yolo min = 7.22 max = 8.66 avg = 7.99
mobilenetv2_yolov3 min = 7.46 max = 8.06 avg = 7.80
yolov4-tiny min = 12.17 max = 13.95 avg = 12.82
nanodet_m min = 3.54 max = 4.78 avg = 3.86
```
44 changes: 5 additions & 39 deletions docs/how-to-build/how-to-build.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -136,44 +136,12 @@ Note: To speed up compilation process on multi core machines, configuring `cmake
***

### Build for MacOS
Install xcode and protobuf
First install Xcode or Xcode Command Line Tools according to your needs.

```shell
# Install protobuf via homebrew
brew install protobuf
```
Then install `protobuf` and `libomp` via homebrew

Download and install openmp for multithreading inference feature
```shell
wget https://github.com/llvm/llvm-project/releases/download/llvmorg-11.0.0/openmp-11.0.0.src.tar.xz
tar -xf openmp-11.0.0.src.tar.xz
cd openmp-11.0.0.src

# apply some compilation fix
sed -i'' -e '/.size __kmp_unnamed_critical_addr/d' runtime/src/z_Linux_asm.S
sed -i'' -e 's/__kmp_unnamed_critical_addr/___kmp_unnamed_critical_addr/g' runtime/src/z_Linux_asm.S

mkdir -p build-x86_64
cd build-x86_64
cmake -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=install -DCMAKE_OSX_ARCHITECTURES="x86_64" \
-DLIBOMP_ENABLE_SHARED=OFF -DLIBOMP_OMPT_SUPPORT=OFF -DLIBOMP_USE_HWLOC=OFF ..
cmake --build . -j 4
cmake --build . --target install
cd ..

mkdir -p build-arm64
cd build-arm64
cmake -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=install -DCMAKE_OSX_ARCHITECTURES="arm64" \
-DLIBOMP_ENABLE_SHARED=OFF -DLIBOMP_OMPT_SUPPORT=OFF -DLIBOMP_USE_HWLOC=OFF ..
cmake --build . -j 4
cmake --build . --target install
cd ..

lipo -create build-x86_64/install/lib/libomp.a build-arm64/install/lib/libomp.a -o libomp.a

# copy openmp library and header files to xcode toolchain sysroot
sudo cp build-x86_64/install/include/* /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include
sudo cp libomp.a /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/lib
brew install protobuf libomp
```

Download and install Vulkan SDK from https://vulkan.lunarg.com/sdk/home
Expand All @@ -193,17 +161,15 @@ mkdir -p build
cd build

cmake -DCMAKE_OSX_ARCHITECTURES="x86_64;arm64" \
-DOpenMP_C_FLAGS="-Xclang -fopenmp" -DOpenMP_CXX_FLAGS="-Xclang -fopenmp" \
-DOpenMP_C_LIB_NAMES="libomp" -DOpenMP_CXX_LIB_NAMES="libomp" \
-DOpenMP_libomp_LIBRARY="/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/lib/libomp.a" \
-DVulkan_INCLUDE_DIR=`pwd`/../vulkansdk-macos-1.2.162.0/MoltenVK/include \
-DVulkan_LIBRARY=`pwd`/../vulkansdk-macos-1.2.162.0/MoltenVK/dylib/macOS/libMoltenVK.dylib \
-DNCNN_VULKAN=ON ..
-DNCNN_VULKAN=ON -DNCNN_BUILD_EXAMPLES=ON ..

cmake --build . -j 4
cmake --build . --target install
```

*Note: If you encounter `libomp` related errors during installation, you can also check our GitHub Actions at [here](https://github.com/Tencent/ncnn/blob/d91cccf/.github/workflows/macos-x64-gpu.yml#L50-L68) to install and use `openmp`.*
***

### Build for ARM Cortex-A family with cross-compiling
Expand Down

0 comments on commit 225e0fe

Please sign in to comment.