Skip to content

feat: Added --build_variant flag for cpu only build. #8329

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

feat: Added --build_variant flag for cpu only build. #8329

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
23 changes: 19 additions & 4 deletions build.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1239,10 +1239,10 @@ def create_dockerfile_linux(

WORKDIR /opt/tritonserver
COPY --chown=1000:1000 NVIDIA_Deep_Learning_Container_License.pdf .
RUN find /opt/tritonserver/python -maxdepth 1 -type f -name \\
"tritonserver-*.whl" | xargs -I {} pip install --upgrade {}[all] && \\
find /opt/tritonserver/python -maxdepth 1 -type f -name \\
"tritonfrontend-*.whl" | xargs -I {} pip install --upgrade {}[all]
RUN find /opt/tritonserver/python -maxdepth 1 -type f -name \
"tritonserver-*.whl" | xargs -I {} pip install --upgrade {}[$VARIANT] && \
find /opt/tritonserver/python -maxdepth 1 -type f -name \
"tritonfrontend-*.whl" | xargs -I {} pip install --upgrade {}[$VARIANT];

RUN pip3 install -r python/openai/requirements.txt

Expand Down Expand Up @@ -1295,6 +1295,7 @@ def dockerfile_prepare_container_linux(argmap, backends, enable_gpu, target_mach
df = """
ARG TRITON_VERSION
ARG TRITON_CONTAINER_VERSION
ARG VARIANT=all

ENV TRITON_SERVER_VERSION ${TRITON_VERSION}
ENV NVIDIA_TRITON_SERVER_VERSION ${TRITON_CONTAINER_VERSION}
Expand Down Expand Up @@ -1898,6 +1899,10 @@ def create_docker_build_script(script_name, container_install_dir, container_ci_
f"--secret id=NVPL_SLIM_URL",
f"--build-arg BUILD_PUBLIC_VLLM={build_public_vllm}",
]
if FLAGS.build_variant:
finalargs += [
"--build-arg VARIANT="+(FLAGS.build_variant),
]
finalargs += [
"-t",
"tritonserver",
Expand Down Expand Up @@ -2763,6 +2768,13 @@ def enable_all():
default=DEFAULT_TRITON_VERSION_MAP["rhel_py_version"],
help="This flag sets the Python version for RHEL platform of Triton Inference Server to be built. Default: the latest supported version.",
)
parser.add_argument(
"--build_variant",
required=False,
type=str,
default="all",
help="Can be set to all or cpu,Default value is all."
)
parser.add_argument(
"--build-secret",
action="append",
Expand Down Expand Up @@ -2802,6 +2814,9 @@ def enable_all():
FLAGS.extra_backend_cmake_arg = []
if FLAGS.build_secret is None:
FLAGS.build_secret = []
if hasattr(FLAGS, 'build_variant') and FLAGS.build_variant not in ["all", "cpu"]:
raise ValueError(f"Invalid build_variant value: {FLAGS.build_variant}. Expected 'all' or 'cpu'.")


# if --enable-all is specified, then update FLAGS to enable all
# settings, backends, repo-agents, caches, file systems, endpoints, etc.
Expand Down