Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Test update toolchain for unix gpu #53

Closed
wants to merge 52 commits into from
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
52 commits
Select commit Hold shift + click to select a range
d0804b1
update nvidiadocker command & remove cuda compat
ChaiBapchya Apr 27, 2020
715078a
replace cu101 with cuda since compat is no longer to be used
ChaiBapchya Apr 28, 2020
9ad16b7
skip flaky tests
ChaiBapchya Apr 28, 2020
efbf764
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya Apr 30, 2020
2dc639d
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya Apr 30, 2020
b083818
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya Apr 30, 2020
e4212d8
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya Apr 30, 2020
f4ab730
get rid of ubuntu_build_cuda and point ubuntu_cu101 to base gpu inste…
ChaiBapchya Apr 28, 2020
68deb2a
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya Apr 30, 2020
7fc7282
Revert "get rid of ubuntu_build_cuda and point ubuntu_cu101 to base g…
ChaiBapchya May 1, 2020
1098a22
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 1, 2020
3685e1f
Merge branch 'update_gpu_toolchain' of github.com:ChaiBapchya/incubat…
ChaiBapchya May 1, 2020
62578d3
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 1, 2020
01bfc4b
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 1, 2020
6f7fcbf
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 1, 2020
08c1898
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 1, 2020
65ee27e
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 1, 2020
d8c0d9f
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 1, 2020
e9a8d5f
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 3, 2020
1f69890
add linux gpu g4 node to all tests using g3 in unix-gpu pipeline
ChaiBapchya May 3, 2020
dffcde0
add linux gpu g4 node to all tests using g3 in unix-gpu pipeline
ChaiBapchya May 3, 2020
f4c34fb
Merge branch 'update_gpu_toolchain' of github.com:ChaiBapchya/incubat…
ChaiBapchya May 3, 2020
2ec0365
resolve remnant code due to simultaneous pushes from cron job and me
ChaiBapchya May 3, 2020
359e03f
compile int64gpu on g4
ChaiBapchya May 3, 2020
471c3ae
update p3 tests to run on g4
ChaiBapchya May 3, 2020
e0ace5d
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 4, 2020
2ff3f94
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 4, 2020
f348f73
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 4, 2020
429eec5
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 4, 2020
58d9122
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 4, 2020
d7d5589
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 6, 2020
e138e12
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 6, 2020
1f25c7f
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 6, 2020
940d43e
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 6, 2020
633290c
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 6, 2020
9da9b4f
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 6, 2020
3c9b1e9
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 6, 2020
578c784
Revert "update p3 tests to run on g4"
ChaiBapchya May 7, 2020
896b7ed
Revert "skip flaky tests"
ChaiBapchya May 7, 2020
8c4a47a
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 7, 2020
d93ce60
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 7, 2020
10fc2ef
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 7, 2020
fb860bb
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 8, 2020
8ef28e3
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 9, 2020
152d2f5
remove gpuwithcompatenv
ChaiBapchya May 9, 2020
e068021
Revert "replace cu101 with cuda since compat is no longer to be used"
ChaiBapchya May 9, 2020
25207d6
fix indent and incorrect docker target
ChaiBapchya May 9, 2020
068580e
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 10, 2020
b4d3f9b
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 10, 2020
c750c5a
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 11, 2020
39a9afd
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 11, 2020
9d60f50
Merge branch 'master' into update_gpu_toolchain
ChaiBapchya May 11, 2020
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions ci/Jenkinsfile_utils.groovy
Original file line number Diff line number Diff line change
Expand Up @@ -257,6 +257,7 @@ def assign_node_labels(args) {
// knowing about the limitations.
NODE_LINUX_CPU = args.linux_cpu
NODE_LINUX_GPU = args.linux_gpu
NODE_LINUX_GPU_G4 = args.linux_gpu_g4
NODE_LINUX_GPU_P3 = args.linux_gpu_p3
NODE_WINDOWS_CPU = args.windows_cpu
NODE_WINDOWS_GPU = args.windows_gpu
Expand Down
3 changes: 2 additions & 1 deletion ci/build.py
Original file line number Diff line number Diff line change
Expand Up @@ -227,8 +227,9 @@ def container_run(docker_client: SafeDockerClient,

# Equivalent command
docker_cmd_list = [
"nvidia-docker" if nvidia_runtime else "docker",
"docker",
'run',
"--gpus all" if nvidia_runtime else "",
"--cap-add",
"SYS_PTRACE", # Required by ASAN
'--rm',
Expand Down
6 changes: 0 additions & 6 deletions ci/docker/Dockerfile.build.ubuntu
Original file line number Diff line number Diff line change
Expand Up @@ -164,9 +164,3 @@ RUN export DEBIAN_FRONTEND=noninteractive && \
apt install -y --no-install-recommends \
cuda-10-1 && \
rm -rf /var/lib/apt/lists/*


FROM gpu as gpuwithcompatenv
# TVMOP requires /usr/local/cuda/compat is no LD_LIBRARY_PATH.
# This should be fixed and deleted.
ENV LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/cuda/compat
2 changes: 1 addition & 1 deletion ci/docker/docker-compose.yml
Original file line number Diff line number Diff line change
Expand Up @@ -103,7 +103,7 @@ services:
build:
context: .
dockerfile: Dockerfile.build.ubuntu
target: gpuwithcompatenv
target: gpu
args:
BASE_IMAGE: nvidia/cuda:10.1-cudnn7-devel-ubuntu18.04
cache_from:
Expand Down
26 changes: 13 additions & 13 deletions ci/jenkins/Jenkins_steps.groovy
Original file line number Diff line number Diff line change
Expand Up @@ -151,7 +151,7 @@ def compile_unix_int64_cpu(lib_name) {

def compile_unix_int64_gpu(lib_name) {
return ['GPU: USE_INT64_TENSOR_SIZE': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/build-gpu-int64') {
timeout(time: max_time, unit: 'MINUTES') {
utils.init_git()
Expand Down Expand Up @@ -816,7 +816,7 @@ def test_unix_python3_mkl_cpu(lib_name) {

def test_unix_python3_gpu(lib_name) {
return ['Python3: GPU': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/ut-python3-gpu') {
try {
utils.unpack_and_init(lib_name, mx_lib_cython)
Expand Down Expand Up @@ -916,7 +916,7 @@ def test_unix_python3_mkldnn_mkl_cpu(lib_name) {

def test_unix_python3_mkldnn_gpu(lib_name) {
return ['Python3: MKLDNN-GPU': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/ut-python3-mkldnn-gpu') {
try {
utils.unpack_and_init(lib_name, mx_mkldnn_lib)
Expand All @@ -932,7 +932,7 @@ def test_unix_python3_mkldnn_gpu(lib_name) {

def test_unix_python3_mkldnn_nocudnn_gpu(lib_name) {
return ['Python3: MKLDNN-GPU-NOCUDNN': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/ut-python3-mkldnn-gpu-nocudnn') {
try {
utils.unpack_and_init(lib_name, mx_mkldnn_lib)
Expand Down Expand Up @@ -966,7 +966,7 @@ def test_unix_python3_tensorrt_gpu(lib_name) {

def test_unix_python3_integration_gpu(lib_name) {
return ['Python Integration GPU': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/it-python-gpu') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_lib)
Expand All @@ -980,7 +980,7 @@ def test_unix_python3_integration_gpu(lib_name) {

def test_unix_cpp_package_gpu(lib_name) {
return ['cpp-package GPU Makefile': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/it-cpp-package') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_lib_cpp_examples_make)
Expand All @@ -994,7 +994,7 @@ def test_unix_cpp_package_gpu(lib_name) {

def test_unix_capi_cpp_package(lib_name) {
return ['capi-cpp-package GPU Makefile': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/it-capi-cpp-package') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_lib_cpp_capi_make)
Expand Down Expand Up @@ -1036,7 +1036,7 @@ def test_unix_scala_mkldnn_cpu(lib_name){

def test_unix_scala_gpu(lib_name) {
return ['Scala: GPU Makefile': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/ut-scala-gpu') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_lib_make)
Expand Down Expand Up @@ -1119,7 +1119,7 @@ def test_unix_perl_cpu(lib_name) {

def test_unix_cpp_gpu(lib_name) {
return ['Cpp: GPU': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/ut-cpp-gpu') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_cmake_lib)
Expand Down Expand Up @@ -1147,7 +1147,7 @@ def test_unix_cpp_cpu(lib_name) {

def test_unix_perl_gpu(lib_name) {
return ['Perl: GPU Makefile': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/ut-perl-gpu') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_lib_make)
Expand All @@ -1161,11 +1161,11 @@ def test_unix_perl_gpu(lib_name) {

def test_unix_r_gpu(lib_name) {
return ['R: GPU': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/ut-r-gpu') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_lib)
utils.docker_run('ubuntu_gpu_cu101', 'unittest_ubuntu_gpu_R', true)
utils.docker_run('ubuntu_gpu_cu101', 'unittest_ubuntu_gpu_R', true)
utils.publish_test_coverage()
}
}
Expand Down Expand Up @@ -1229,7 +1229,7 @@ def test_unix_distributed_kvstore_cpu(lib_name) {

def test_unix_distributed_kvstore_gpu(lib_name) {
return ['dist-kvstore tests GPU': {
node(NODE_LINUX_GPU) {
node(NODE_LINUX_GPU_G4) {
ws('workspace/it-dist-kvstore') {
timeout(time: max_time, unit: 'MINUTES') {
utils.unpack_and_init(lib_name, mx_lib)
Expand Down
2 changes: 1 addition & 1 deletion ci/jenkins/Jenkinsfile_unix_gpu
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,7 @@ node('utility') {
utils = load('ci/Jenkinsfile_utils.groovy')
custom_steps = load('ci/jenkins/Jenkins_steps.groovy')
}
utils.assign_node_labels(utility: 'utility', linux_cpu: 'mxnetlinux-cpu', linux_gpu: 'mxnetlinux-gpu', linux_gpu_p3: 'mxnetlinux-gpu-p3')
utils.assign_node_labels(utility: 'utility', linux_cpu: 'mxnetlinux-cpu', linux_gpu: 'mxnetlinux-gpu', linux_gpu_p3: 'mxnetlinux-gpu-p3',linux_gpu_g4: 'mxnetlinux-gpu-g4')

utils.main_wrapper(
core_logic: {
Expand Down