blob: 2528f98fba6397d7e66586a610770fe3e1a6af22 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
|
#!/bin/bash
set -ex
sudo apt -q=2 update
sudo apt-get -q=2 install -y --no-install-recommends git
sudo apt-get -q=2 install -y --no-install-recommends scons
sudo apt-get -q=2 install -y --no-install-recommends gcc-arm-linux-gnueabihf
sudo apt-get -q=2 install -y --no-install-recommends g++-arm-linux-gnueabihf
sudo apt-get -q=2 install -y --no-install-recommends curl
sudo apt-get -q=2 install -y --no-install-recommends autoconf
sudo apt-get -q=2 install -y --no-install-recommends libtool
sudo apt-get -q=2 install -y --no-install-recommends cmake
sudo apt -q=2 install -y --no-install-recommends build-essential cmake libpthread-stubs0-dev
sudo apt -q=2 install -y --no-install-recommends python-pip python3-pip virtualenv python-dev python3-dev xxd
# Set local configuration
git config --global user.email "ci_notify@linaro.org"
git config --global user.name "Linaro CI"
git clone --depth 1 "http://review.mlplatform.org/ml/ComputeLibrary"
git clone https://github.com/Arm-software/armnn
wget https://dl.bintray.com/boostorg/release/1.64.0/source/boost_1_64_0.tar.bz2 && tar xf boost_1_64_0.tar.bz2
git clone --depth 1 -b v3.5.0 https://github.com/google/protobuf.git
git clone --depth 1 https://github.com/tensorflow/tensorflow.git --branch r2.0 --single-branch
wget -O flatbuffers-1.10.0.tar.gz https://github.com/google/flatbuffers/archive/v1.10.0.tar.gz && tar xf flatbuffers-1.10.0.tar.gz
if [ -n "$GERRIT_PROJECT" ] && [ $GERRIT_EVENT_TYPE == "patchset-created" ]; then
cd armnn
GERRIT_URL="http://${GERRIT_HOST}/${GERRIT_PROJECT}"
if git pull ${GERRIT_URL} ${GERRIT_REFSPEC} | grep -q "Automatic merge failed"; then
git reset --hard
echo "Retrying to apply the patch with: git fetch && git checkout."
if ! { git fetch ${GERRIT_URL} ${GERRIT_REFSPEC} | git checkout FETCH_HEAD; }; then
git reset --hard
echo "Error: *** Error patch merge failed"
exit 1
fi
fi
fi
cd ${WORKSPACE}/ComputeLibrary
scons extra_cxx_flags="-fPIC" Werror=0 debug=0 asserts=0 neon=1 opencl=0 os=linux arch=armv7a examples=1
cd ${WORKSPACE}/boost_1_64_0
./bootstrap.sh
rm project-config.jam || true
wget --no-check-certificate http://people.linaro.org/~theodore.grey/project-config.jam
./b2 \
--build-dir=${WORKSPACE}/boost_1_64_0/build toolset=gcc link=static cxxflags=-fPIC \
--with-filesystem \
--with-test \
--with-log \
--with-program_options install --prefix=${WORKSPACE}/boost
cd $WORKSPACE/protobuf
git submodule update --init --recursive
./autogen.sh
./configure --prefix=$WORKSPACE/protobuf-host
make -j$(nproc)
make install
make clean
./autogen.sh
./configure --prefix=$WORKSPACE/protobuf-arm32 --host=arm-linux CC=arm-linux-gnueabihf-gcc CXX=arm-linux-gnueabihf-g++ --with-protoc=$WORKSPACE/protobuf-host/bin/protoc
make -j$(nproc)
make install
cd $WORKSPACE/tensorflow
../armnn/scripts/generate_tensorflow_protobuf.sh ../tensorflow-protobuf ../protobuf-host
cd $WORKSPACE/flatbuffers-1.10.0
mkdir build && cd build
cmake .. \
-DFLATBUFFERS_BUILD_FLATC=1 \
-DCMAKE_INSTALL_PREFIX:PATH=$WORKSPACE/flatbuffers
make all install
cd $WORKSPACE/flatbuffers-1.10.0
mkdir build-arm32 && cd build-arm32
cmake .. -DCMAKE_C_COMPILER=arm-linux-gnueabihf-gcc \
-DCMAKE_CXX_COMPILER=arm-linux-gnueabihf-g++ \
-DFLATBUFFERS_BUILD_FLATC=1 \
-DCMAKE_INSTALL_PREFIX:PATH=$WORKSPACE/flatbuffers-arm32 \
-DFLATBUFFERS_BUILD_TESTS=0
make all install
cd $WORKSPACE
mkdir tflite
cd tflite
cp $WORKSPACE/tensorflow/tensorflow/lite/schema/schema.fbs .
$WORKSPACE/flatbuffers-1.10.0/build/flatc -c --gen-object-api --reflect-types --reflect-names schema.fbs
cd $WORKSPACE/armnn
mkdir build
cd build
cmake .. -DCMAKE_LINKER=/usr/bin/arm-linux-gnueabihf-ld \
-DCMAKE_C_COMPILER=/usr/bin/arm-linux-gnueabihf-gcc \
-DCMAKE_CXX_COMPILER=/usr/bin/arm-linux-gnueabihf-g++ \
-DCMAKE_C_COMPILER_FLAGS=-fPIC \
-DCMAKE_CXX_FLAGS=-mfpu=neon \
-DARMCOMPUTE_ROOT=$WORKSPACE/ComputeLibrary \
-DARMCOMPUTE_BUILD_DIR=$WORKSPACE/ComputeLibrary/build \
-DBOOST_ROOT=$WORKSPACE/boost \
-DTF_GENERATED_SOURCES=$WORKSPACE/tensorflow-protobuf \
-DBUILD_TF_PARSER=1 \
-DBUILD_TF_LITE_PARSER=1 \
-DTF_LITE_GENERATED_PATH=$WORKSPACE/tflite \
-DFLATBUFFERS_ROOT=$WORKSPACE/flatbuffers-arm32 \
-DFLATC_DIR=$WORKSPACE/flatbuffers-1.10.0/build \
-DPROTOBUF_ROOT=$WORKSPACE/protobuf-arm32 \
-DARMCOMPUTENEON=1 \
-DARMNNREF=1
make -j$(nproc)
export XZ_DEFAULTS="-T 0"
cd ${WORKSPACE}
rm -rf boost_*.tar.bz2 boost_* protobuf tensorflow
find ${WORKSPACE} -type f -name *.o -delete
tar -cJf /tmp/armnn-full-32.tar.xz ${WORKSPACE}
mv armnn/build .
mv protobuf-arm32/lib/libprotobuf.so.15.0.0 build
rm -rf boost armnn ComputeLibrary flatbuffers protobuf-host tensorflow-protobuf builders.sh
tar -cJf /tmp/armnn-32.tar.xz ${WORKSPACE}
mkdir ${WORKSPACE}/out
mv /tmp/armnn-32.tar.xz ${WORKSPACE}/out
mv /tmp/armnn-full-32.tar.xz ${WORKSPACE}/out
cd ${WORKSPACE}/out && sha256sum > SHA256SUMS.txt
|
