Skip to content

Fix: Properly Respect CUDA_VISIBLE_DEVICES Setting #40

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

Open
LeonGuertler wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix: Properly Respect CUDA_VISIBLE_DEVICES Setting #40

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
30 changes: 20 additions & 10 deletions oat/interface.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,8 @@
# See the License for the specific language governing permissions and
# limitations under the License.
"""Defining how components interface with each other."""
import os

import logging
from typing import Type

Expand Down Expand Up @@ -79,9 +81,11 @@ def get_program(
label=label,
)
)
local_resources[label] = local_multi_processing.PythonProcess(
env={"CUDA_VISIBLE_DEVICES": str(i)}
)

env = os.environ.copy()
if "CUDA_VISIBLE_DEVICES" not in env:
env["CUDA_VISIBLE_DEVICES"] = str(i)
local_resources[label] = local_multi_processing.PythonProcess(env=env)

# Learner.
master_addr = "0.0.0.0"
Expand All @@ -101,9 +105,13 @@ def get_program(
ipc_server,
)
program.add_node(master_learner, label=label)
local_resources[label] = local_multi_processing.PythonProcess(
env={"CUDA_VISIBLE_DEVICES": str(learner_gpus[0])}
)

# check if CUDA_VISIBLE_DEVICES it is already set
env = os.environ.copy()
if "CUDA_VISIBLE_DEVICES" not in env:
env["CUDA_VISIBLE_DEVICES"] = str(learner_gpus[0])
local_resources[label] = local_multi_processing.PythonProcess(env=env)

for i in range(1, len(learner_gpus)):
label = f"learner_{i}"
worker_learner = lp.PyClassNode(
Expand All @@ -119,8 +127,10 @@ def get_program(
ipc_server,
)
program.add_node(worker_learner, label=label)
local_resources[label] = local_multi_processing.PythonProcess(
env={"CUDA_VISIBLE_DEVICES": str(learner_gpus[i])}
)

return program, local_resources
env = os.environ.copy()
if "CUDA_VISIBLE_DEVICES" not in env:
env["CUDA_VISIBLE_DEVICES"] = str(learner_gpus[i])
local_resources[label] = local_multi_processing.PythonProcess(env=env)

return program, local_resources