Skip to content

Fixed autoparallel integration tests on ROCm. #2321

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
wenchenvincent wants to merge 1 commit into
base: main
Choose a base branch
from
+8 −1
Open

Fixed autoparallel integration tests on ROCm. #2321

Show file tree
Hide file tree
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
8 changes: 8 additions & 0 deletions .github/workflows/integration_test_8gpu_autoparallel.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,4 +74,12 @@ jobs:
sudo mkdir -p "$RUNNER_TEMP/artifacts-to-be-uploaded"
sudo chown -R $(id -u):$(id -g) "$RUNNER_TEMP/artifacts-to-be-uploaded"

# HIP runtime relies on the file /opt/amdgpu/share/libdrm/amdgpu.ids to look up the product name of AMDGPU.
# The docker image only has /usr/share/libdrm/amdgpu.ids and it is out of date. So this is the workaround.
if [[ "${{ matrix.gpu-arch-type }}" == "rocm" ]]; then
sudo curl -L -o /usr/share/libdrm/amdgpu.ids https://gitlab.freedesktop.org/mesa/libdrm/-/raw/main/data/amdgpu.ids
sudo mkdir -p /opt/amdgpu/share/libdrm
sudo ln -sf /usr/share/libdrm/amdgpu.ids /opt/amdgpu/share/libdrm/amdgpu.ids
fi

python -m torchtitan.experiments.autoparallel.tests.integration_tests --gpu_arch_type ${{ matrix.gpu-arch-type }} $RUNNER_TEMP/artifacts-to-be-uploaded --ngpu 4
1 change: 0 additions & 1 deletion torchtitan/experiments/autoparallel/tests/integration_tests.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,6 @@ def build_autoparallel_test_list() -> list[OverrideDefinitions]:
"llama3 AutoParallel FSDP+TP",
"llama3_autoparallel_fsdp_tp",
ngpu=4,
skip_rocm_test=True,
),
# TODO: Re-enable this once we fix the test
# deepseek_v3 tests
Expand Down
Loading