[CI][XPU] enable unit test for XPU device by DiweiSun · Pull Request #2814 · pytorch/ao

DiweiSun · 2025-08-20T01:57:39Z

Enabling CI testing for the torchao project on the Intel XPU (GPU) platform to ensure functional correctness, performance consistency, and long-term compatibility as both torchao and XPU support evolve.

This reverts commit a736c41.

pytorch-bot · 2025-08-20T01:57:43Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2814

📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure, 10 Cancelled Jobs

As of commit 030121f with merge base 2db4c76 ():

NEW FAILURE - The following job has failed:

xpu-test / test (gh)
Error response from daemon: page not found

CANCELLED JOBS - The following jobs were cancelled. Please retry:

Run 1xH100 Tests / test (H100, linux.aws.h100, --pre torch torchvision torchaudio fbgemm-gpu-genai --index-url https... / linux-job (gh)
##[error]The operation was canceled.
Run 1xL4 Tests / test (SM-89, linux.g6.4xlarge.experimental.nvidia.gpu, --pre torch --index-url https://download.p... / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test (CPU 2.6, linux.4xlarge, torch==2.6.0 --index-url https://download.pytorch.org/whl/cpu, cpu) / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test (CPU 2.7, linux.4xlarge, torch==2.7.0 --index-url https://download.pytorch.org/whl/cpu, cpu) / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test (CPU 2.8, linux.4xlarge, torch==2.8.0 --index-url https://download.pytorch.org/whl/cpu, cpu) / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test (CUDA 2.6, linux.g5.12xlarge.nvidia.gpu, torch==2.6.0, cuda, 12.6) / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test (CUDA 2.7, linux.g5.12xlarge.nvidia.gpu, torch==2.7.0, cuda, 12.6) / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test (CUDA 2.8, linux.g5.12xlarge.nvidia.gpu, torch==2.8.0, cuda, 12.6) / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test-nightly (CPU Nightly, linux.4xlarge, --pre torch --index-url https://download.pytorch.org/wh... / linux-job (gh)
##[error]The operation was canceled.
Run Regression Tests / test-nightly (CUDA Nightly, linux.g5.12xlarge.nvidia.gpu, --pre torch --index-url https://downloa... / linux-job (gh)
##[error]The operation was canceled.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

.github/workflows/pr-test-xpu.yml

chuanqi129 · 2025-08-22T05:37:59Z

.github/workflows/pr-test-xpu.yml

+      - name: Clean all stopped docker containers
+        if: always()
+        shell: bash
+        run: |
+          # Prune all stopped containers.
+          # If other runner is pruning on this node, will skip.
+          nprune=$(ps -ef | grep -c "docker container prune")
+          if [[ $nprune -eq 1 ]]; then
+            docker container prune -f
+          fi
+
+      - name: Runner health check GPU count
+        if: always()
+        shell: bash
+        run: |
+          ngpu=$(timeout 30 clinfo -l | grep -c -E 'Device' || true)
+          msg="Please file an issue on pytorch/ao reporting the faulty runner. Include a link to the runner logs so the runner can be identified"
+          if [[ $ngpu -eq 0 ]]; then
+            echo "Error: Failed to detect any GPUs on the runner"
+            echo "$msg"
+            exit 1
+          fi
+
+      - name: Use following to pull public copy of the image
+        id: print-ghcr-mirror
+        shell: bash
+        run: |
+          echo "docker pull ${DOCKER_IMAGE}"
+          docker pull ${DOCKER_IMAGE}


I think all those steps we also need in here https://github.com/pytorch/pytorch/blob/main/.github/workflows/_xpu-test.yml#L79-L114

Ported done. Please kindly help review.

chuanqi129 · 2025-08-22T05:38:42Z

.github/workflows/pr-test-xpu.yml

+          if-no-files-found: ignore
+          path: ./**/core.[1-9]*
+
+      - name: Teardown XPU


Can reuse the action in pytorch directly

yes, this is literally ported from pytorch

Co-authored-by: Wang, Chuanqi <chuanqi.wang@intel.com>

This reverts commit d9ab09e.

This reverts commit 3085c2b.

.github/workflows/pr-test-xpu.yml

Co-authored-by: Wang, Chuanqi <chuanqi.wang@intel.com>

DiweiSun · 2025-09-08T02:37:01Z

.github/workflows/pr-test-xpu.yml

+      - ciflow/xpu/*
+  pull_request:
+    branches:
+      - main


remove pull-request after review

chuanqi129 · 2025-09-10T06:30:53Z

@pytorchbot label "ciflow/xpu"

pytorch-bot · 2025-09-10T06:30:58Z

Didn't find following labels among repository labels: ciflow/xpu

liangan1 · 2025-09-15T07:42:17Z

@pytorchbot label "ciflow/xpu"

pytorch-bot · 2025-09-15T07:42:25Z

Unknown label ciflow/xpu.
Currently recognized labels are

ciflow/benchmark
ciflow/tutorials
ciflow/rocm
ciflow/4xh100

liangan1 · 2025-09-16T23:40:43Z

@pytorchbot label "ciflow/xpu"

pytorch-bot · 2025-09-16T23:40:49Z

Warning: Unknown label ciflow/xpu.
Currently recognized labels are

ciflow/benchmark
ciflow/tutorials
ciflow/rocm
ciflow/4xh100

Please add the new label to .github/pytorch-probot.yml

pytorch-bot · 2025-09-17T02:35:25Z

Unknown label ciflow/xpu.
Currently recognized labels are

ciflow/benchmark
ciflow/tutorials
ciflow/rocm
ciflow/4xh100

DiweiSun · 2025-09-17T03:18:33Z

@pytorchbot label "ciflow/xpu"

liangan1 · 2025-09-17T03:19:59Z

@pytorchbot label "ciflow/xpu"

liangan1 · 2025-09-17T03:21:17Z

@pytorchbot label "ciflow/xpu"

pytorch-bot · 2025-09-17T03:21:25Z

Unknown label ciflow/xpu.
Currently recognized labels are

ciflow/benchmark
ciflow/tutorials
ciflow/rocm
ciflow/4xh100

DiweiSun added 3 commits August 19, 2025 08:28

enable xpu ci test

a736c41

Revert "enable xpu ci test"

7c96ad4

This reverts commit a736c41.

enable ci test for xpu

d1122fc

meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Aug 20, 2025

Create ci_test_xpu.sh

4593d95

chuanqi129 suggested changes Aug 22, 2025

View reviewed changes

DiweiSun and others added 2 commits August 22, 2025 14:30

Update .github/workflows/pr-test-xpu.yml

7d90b8c

Co-authored-by: Wang, Chuanqi <chuanqi.wang@intel.com>

Update .github/workflows/pr-test-xpu.yml

e6bc407

Co-authored-by: Wang, Chuanqi <chuanqi.wang@intel.com>

liangan1 mentioned this pull request Sep 2, 2025

Intel XPU CI Enabling #2917

Open

9 tasks

DiweiSun added 4 commits September 2, 2025 10:00

fix for trigger scenarios

c34601f

port from pytorch repo

3085c2b

Rename action.yml to xpu-action.yml

d9ab09e

update to align with pytorch

f87892a

liangan1 added topic: for developers Use this tag if this PR is mainly developer facing ci labels Sep 4, 2025

DiweiSun added 2 commits September 3, 2025 19:40

Revert "Rename action.yml to xpu-action.yml"

c6f07b5

This reverts commit d9ab09e.

Revert "port from pytorch repo"

544593a

This reverts commit 3085c2b.

chuanqi129 reviewed Sep 4, 2025

View reviewed changes

.github/workflows/pr-test-xpu.yml Show resolved Hide resolved

DiweiSun and others added 2 commits September 4, 2025 14:43

Update .github/workflows/pr-test-xpu.yml

2e1dc50

Co-authored-by: Wang, Chuanqi <chuanqi.wang@intel.com>

debug for runner

188a0f8

DiweiSun changed the title ~~Molly/enable xpu ci~~ [CI][XPU] enable unit test for XPU device Sep 8, 2025

DiweiSun commented Sep 8, 2025

View reviewed changes

DiweiSun added 7 commits September 8, 2025 00:58

lint format fix

421d02c

format fix

6f6cd17

format fix

bae7000

format fix

7bd3d29

format fix

4fd2909

format fix

4a7d9af

format fix

c3f4384

format fix

e8936cb

trigger by tag only

50e56ec

pytorch-bot bot added the ciflow/xpu label used to trigger xpu CI jobs label Sep 15, 2025

add xpu label for xpuci

5a46341

pytorch-bot bot removed the ciflow/xpu label used to trigger xpu CI jobs label Sep 16, 2025

pytorch-bot bot added the ciflow/xpu label used to trigger xpu CI jobs label Sep 16, 2025

fix docker path

030121f

pytorch-bot bot removed the ciflow/xpu label used to trigger xpu CI jobs label Sep 17, 2025

liangan1 added the ciflow/xpu label used to trigger xpu CI jobs label Sep 17, 2025

liangan1 removed the ciflow/xpu label used to trigger xpu CI jobs label Sep 17, 2025

pytorch-bot bot added the ciflow/xpu label used to trigger xpu CI jobs label Sep 17, 2025

DiweiSun closed this Sep 24, 2025

Conversation

DiweiSun commented Aug 20, 2025

Uh oh!

pytorch-bot bot commented Aug 20, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2814

❌ 1 New Failure, 10 Cancelled Jobs

Uh oh!

Uh oh!

Uh oh!

Uh oh!

chuanqi129 Aug 22, 2025

Choose a reason for hiding this comment

Uh oh!

DiweiSun Sep 2, 2025

Choose a reason for hiding this comment

Uh oh!

chuanqi129 Aug 22, 2025

Choose a reason for hiding this comment

Uh oh!

DiweiSun Sep 2, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

DiweiSun Sep 8, 2025

Choose a reason for hiding this comment

Uh oh!

chuanqi129 commented Sep 10, 2025

Uh oh!

pytorch-bot bot commented Sep 10, 2025

Uh oh!

liangan1 commented Sep 15, 2025

Uh oh!

pytorch-bot bot commented Sep 15, 2025

Uh oh!

liangan1 commented Sep 16, 2025

Uh oh!

pytorch-bot bot commented Sep 16, 2025

Uh oh!

pytorch-bot bot commented Sep 17, 2025

Uh oh!

DiweiSun commented Sep 17, 2025

Uh oh!

liangan1 commented Sep 17, 2025

Uh oh!

liangan1 commented Sep 17, 2025

Uh oh!

pytorch-bot bot commented Sep 17, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

pytorch-bot bot commented Aug 20, 2025 •

edited

Loading