Skip to content

Fix to execute efa tests on mainline #1083

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.

Sign up for GitHub

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

Merged
merged 21 commits into from
Apr 30, 2021
Merged

Fix to execute efa tests on mainline #1083

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
21 commits
Select commit Hold shift + click to select a range
1b8b92c
fix to execute efa tests
jeet4320 Apr 30, 2021
f602a87
efa flag true
jeet4320 Apr 30, 2021
b2ffff7
revert code
jeet4320 Apr 30, 2021
cfe4d85
revert code
jeet4320 Apr 30, 2021
16b25e5
new changes
jeet4320 Apr 30, 2021
248b340
run eks and fix efa
jeet4320 Apr 30, 2021
b63d910
remove code
jeet4320 Apr 30, 2021
70d081e
add efa marker to inference
jeet4320 Apr 30, 2021
fecd4f1
add efa marker to inference
jeet4320 Apr 30, 2021
bef800f
testtfs conftest
jeet4320 Apr 30, 2021
bd7d929
testtfs conftest
jeet4320 Apr 30, 2021
c18b01c
use efa for training
jeet4320 Apr 30, 2021
d847879
use efa for training
jeet4320 Apr 30, 2021
2ace3c0
new changes
jeet4320 Apr 30, 2021
1463031
new changes
jeet4320 Apr 30, 2021
2a69536
new changes
jeet4320 Apr 30, 2021
80dca54
new changes
jeet4320 Apr 30, 2021
944d740
new changes
jeet4320 Apr 30, 2021
2480dc0
not efa
jeet4320 Apr 30, 2021
390efcb
revert
jeet4320 Apr 30, 2021
d1daee1
Merge branch 'master' into add-efa-flag
jeet4320 Apr 30, 2021
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
2 changes: 2 additions & 0 deletions test/dlc_tests/eks/eks_manifest_templates/tensorflow/training/multi_node_gpu_training.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,8 @@ spec:
- -x
- LD_LIBRARY_PATH
- -x
- RDMAV_FORK_SAFE=1
- -x
- PATH
- -x
- NCCL_SOCKET_IFNAME=eth0
Expand Down
13 changes: 1 addition & 12 deletions test/sagemaker_tests/mxnet/training/conftest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,18 +46,7 @@ def pytest_addoption(parser):
parser.addoption('--tag', default=None)
parser.addoption('--generate-coverage-doc', default=False, action='store_true',
help='use this option to generate test coverage doc')
parser.addoption(
Copy link
Contributor Author

@jeet4320 jeet4320 Apr 30, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this was added for local SM tests, but we dont have EFA for local SM tests

"--efa", action="store_true", default=False, help="Run only efa tests",
)

def pytest_configure(config):
config.addinivalue_line("markers", "efa(): explicitly mark to run efa tests")

def pytest_runtest_setup(item):
if item.config.getoption("--efa"):
efa_tests = [mark for mark in item.iter_markers(name="efa")]
if not efa_tests:
pytest.skip("Skipping non-efa tests")


def pytest_collection_modifyitems(session, config, items):
if config.getoption("--generate-coverage-doc"):
Expand Down
14 changes: 0 additions & 14 deletions test/sagemaker_tests/pytorch/training/conftest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,20 +105,6 @@ def pytest_addoption(parser):
parser.addoption('--tag', default=None)
parser.addoption('--generate-coverage-doc', default=False, action='store_true',
help='use this option to generate test coverage doc')
parser.addoption(
"--efa", action="store_true", default=False, help="Run only efa tests",
)


def pytest_configure(config):
config.addinivalue_line("markers", "efa(): explicitly mark to run efa tests")


def pytest_runtest_setup(item):
if item.config.getoption("--efa"):
efa_tests = [mark for mark in item.iter_markers(name="efa")]
if not efa_tests:
pytest.skip("Skipping non-efa tests")


def pytest_collection_modifyitems(session, config, items):
Expand Down
11 changes: 0 additions & 11 deletions test/sagemaker_tests/tensorflow/tensorflow1_training/integration/conftest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,18 +42,7 @@ def pytest_addoption(parser):
parser.addoption('--instance-type', default=None)
parser.addoption('--generate-coverage-doc', default=False, action='store_true',
help='use this option to generate test coverage doc')
parser.addoption(
"--efa", action="store_true", default=False, help="Run only efa tests",
)

def pytest_configure(config):
config.addinivalue_line("markers", "efa(): explicitly mark to run efa tests")

def pytest_runtest_setup(item):
if item.config.getoption("--efa"):
efa_tests = [mark for mark in item.iter_markers(name="efa")]
if not efa_tests:
pytest.skip("Skipping non-efa tests")

def pytest_collection_modifyitems(session, config, items):
if config.getoption("--generate-coverage-doc"):
Expand Down
11 changes: 0 additions & 11 deletions test/sagemaker_tests/tensorflow/tensorflow2_training/integration/conftest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,18 +42,7 @@ def pytest_addoption(parser):
parser.addoption('--instance-type', default=None)
parser.addoption('--generate-coverage-doc', default=False, action='store_true',
help='use this option to generate test coverage doc')
parser.addoption(
"--efa", action="store_true", default=False, help="Run only efa tests",
)

def pytest_configure(config):
config.addinivalue_line("markers", "efa(): explicitly mark to run efa tests")

def pytest_runtest_setup(item):
if item.config.getoption("--efa"):
efa_tests = [mark for mark in item.iter_markers(name="efa")]
if not efa_tests:
pytest.skip("Skipping non-efa tests")

def pytest_collection_modifyitems(session, config, items):
if config.getoption("--generate-coverage-doc"):
Expand Down
7 changes: 6 additions & 1 deletion test/test_utils/sagemaker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -139,10 +139,15 @@ def generate_sagemaker_pytest_cmd(image, sagemaker_test_type):
local_test_report = os.path.join(UBUNTU_HOME_DIR, "test", f"{job_type}_{tag}_sm_local.xml")
is_py3 = " python3 -m "

efa_flag = ""
if job_type == "training" and (framework_major_version == "tensorflow" or framework == "pytorch"):
efa_dedicated = os.getenv("EFA_DEDICATED", "False").lower() == "true"
efa_flag = '--efa' if efa_dedicated else '-m not efa'

remote_pytest_cmd = (
f"pytest -rA {integration_path} --region {region} --processor {processor} {docker_base_arg} "
f"{sm_remote_docker_base_name} --tag {tag} {framework_version_arg} {framework_version} "
f"{aws_id_arg} {account_id} {instance_type_arg} {instance_type} --junitxml {test_report}"
f"{aws_id_arg} {account_id} {instance_type_arg} {instance_type} {efa_flag} --junitxml {test_report}"
Copy link
Contributor Author

@jeet4320 jeet4320 Apr 30, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

efa is just needed for SM training remote tests

)

if processor == "eia" :
Expand Down