Skip to content

Support google-dns for nvidia gpt docker #361

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 4 commits into from
Apr 13, 2025
Merged

Support google-dns for nvidia gpt docker #361

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
9c05469
Add an option to kill the busiest process group
amd-arsuresh Apr 10, 2025
4ff39b5
[Automated Commit] Format Codebase [skip ci]
github-actions[bot] Apr 10, 2025
a2857e4
Allow google-dns for nvidia gptj docker
arjunsuresh Apr 13, 2025
62f762e
[Automated Commit] Format Codebase [skip ci]
github-actions[bot] Apr 13, 2025
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
5 changes: 5 additions & 0 deletions script/get-ml-model-gptj/customize.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,11 @@ def preprocess(i):
env['MLC_CUDA_DEVICE_PROP_GPU_COMPUTE_CAPABILITY']) *
10)
env['MLC_GPU_ARCH'] = gpu_arch
env['DOCKER_RUN_ARGS'] = f" -v {env['MLC_NVIDIA_MLPERF_SCRATCH_PATH']}:/mnt"

if is_true(env.get('MLC_DOCKER_USE_GOOGLE_DNS', '')):
env['DOCKER_RUN_ARGS'] += ' --dns 8.8.8.8 --dns 8.8.4.4 '

env['MLC_TMP_REQUIRE_DOWNLOAD'] = 'no'

else:
Expand Down
3 changes: 2 additions & 1 deletion script/get-ml-model-gptj/run-nvidia.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,8 @@ make -C docker build
test $? -eq 0 || exit $?

export RUN_CMD="bash -c 'python3 scripts/build_wheel.py -a=${MLC_GPU_ARCH} --clean --install --trt_root /usr/local/tensorrt/ && python examples/quantization/quantize.py --dtype=float16 --output_dir=/mnt/models/GPTJ-6B/fp8-quantized-ammo/GPTJ-FP8-quantized --model_dir=/mnt/models/GPTJ-6B/checkpoint-final --qformat=fp8 --kv_cache_dtype=fp8 '"
export DOCKER_RUN_ARGS=" -v ${MLC_NVIDIA_MLPERF_SCRATCH_PATH}:/mnt"


make -C docker run LOCAL_USER=1
test $? -eq 0 || exit $?

Expand Down
3 changes: 3 additions & 0 deletions script/kill-process/customize.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,9 @@ def generate_kill_command(env):
# all in that group
kill_cmd = f"pkill -g $(pgrep -o {process_name} | xargs ps -o pgid= -p | tr -d ' ')"

elif is_true(env.get("MLC_KILL_BUSIEST_PROCESS_GROUP")):
kill_cmd = r"busy_pgid=\$(ps -eo pgid,pcpu --sort=-pcpu | head -n 2 | tail -n 1 | awk '{print $1}') && kill -- -\$busy_pgid"

else:
if env.get("MLC_KILL_PROCESS_ID"):
process_id = env["MLC_KILL_PROCESS_ID"]
Expand Down
1 change: 1 addition & 0 deletions script/kill-process/meta.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,5 +18,6 @@ input_mapping:
group: MLC_KILL_PROCESS_GROUP
pid: MLC_KILL_PROCESS_ID
pname: MLC_KILL_PROCESS_NAME
busy_group: MLC_KILL_BUSIEST_PROCESS_GROUP


Loading