8000 [Monitoring] enable rocm monitoring for trunk and general tests by yangw-dev · Pull Request #153455 · pytorch/pytorch · GitHub
[go: up one dir, main page]
Skip to content

[Monitoring] enable rocm monitoring for trunk and general tests #153455

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

Draft
wants to merge 2 commits into
base: main
Choose a base branch
from
Draft

[Monitoring] enable rocm monitoring for trunk and general tests #153455

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fae5c17
add model test1
yangw-dev May 13, 2025
a2ed4b1
add model test1
yangw-dev May 14, 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
1 change: 1 addition & 0 deletions .github/workflows/rocm-mi300.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,4 +70,5 @@ jobs:
build-environment: linux-jammy-rocm-py3.10
docker-image: ${{ needs.linux-jammy-rocm-py3_10-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-jammy-rocm-py3_10-build.outputs.test-matrix }}
disable-monitor: false
secrets: inherit
1 change: 1 addition & 0 deletions .github/workflows/rocm.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,4 +58,5 @@ jobs:
build-environment: linux-jammy-rocm-py3.10
docker-image: ${{ needs.linux-jammy-rocm-py3_10-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-jammy-rocm-py3_10-build.outputs.test-matrix }}
disable-monitor: false
secrets: inherit
1 change: 1 addition & 0 deletions .github/workflows/trunk.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -182,6 +182,7 @@ jobs:
build-environment: linux-jammy-rocm-py3.10
docker-image: ${{ needs.linux-jammy-rocm-py3_10-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-jammy-rocm-py3_10-build.outputs.test-matrix }}
disable-monitor: false
tests-to-include: "test_nn test_torch test_cuda test_ops test_unary_ufuncs test_binary_ufuncs test_autograd inductor/test_torchinductor distributed/test_c10d_common distributed/test_c10d_nccl"
secrets: inherit

Expand Down
14 changes: 9 additions & 5 deletions tools/stats/monitor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -154,7 +154,7 @@ def add_data(self, data: UsageData) -> None:
with self._lock:
self._data_list.append(data)

def add_error(self, error: Exception) -> None:
def add_error(self, error: Exception | str) -> None:
with self._lock:
self._data_errors.append(str(error))

Expand Down Expand Up @@ -380,13 +380,17 @@ def _collect_gpu_data(self) -> list[GpuData]:
)
)
elif self._has_amdsmi:
# Iterate over the available GPUs
gpu_data_list = []
for handle in self._gpu_handles:
# see https://rocm.docs.amd.com/projects/amdsmi/en/docs-5.7.0/py-interface_readme_link.html
engine_usage = amdsmi.amdsmi_get_gpu_activity(handle)
usage = f"check engine_usage: {engine_usage}"
self.shared_resource.add_error(usage)

gpu_uuid = amdsmi.amdsmi_get_gpu_device_uuid(handle)
gpu_utilization = engine_usage["gfx_activity"]
gpu_mem_utilization = gpu_utilization["umc_activity"]

gpu_utilization = engine_usage.get("gfx_activity", 0)
gpu_mem_utilization = engine_usage.get("umc_activity", 0)

gpu_data_list.append(
GpuData(
uuid=gpu_uuid,
Expand Down
Loading
0