From 20ce7cc1eaf9350050a8ace0c7771bad8af60012 Mon Sep 17 00:00:00 2001 From: Michael Hudgins Date: Mon, 24 Jun 2024 16:19:46 -0400 Subject: [PATCH 01/37] Create cloud-tpu-presubmit --- .github/workflows/cloud-tpu-presubmit | 90 +++++++++++++++++++++++++++ 1 file changed, 90 insertions(+) create mode 100644 .github/workflows/cloud-tpu-presubmit diff --git a/.github/workflows/cloud-tpu-presubmit b/.github/workflows/cloud-tpu-presubmit new file mode 100644 index 000000000000..f15f51271666 --- /dev/null +++ b/.github/workflows/cloud-tpu-presubmit @@ -0,0 +1,90 @@ +# Cloud TPU CI +name: Cloud TPU Presubmit +# Run on pull_request that is labeled as "optional_ci_tpu" or workflow dispatch +on: + pull_request: + branches: + - main + types: [labeled, synchronize] + workflow_dispatch: +# Cancel any previous iterations if a new commit is pushed +concurrency: + group: ${{ github.workflow }}-${{ github.ref }} + cancel-in-progress: true +jobs: + cloud-tpu-test: + # TODO: confirm final naming for optional label + if: contains(github.event.pull_request.labels.*.name, 'optional_ci_tpu') + name: "TPU v5e x 8 Presubmit" + env: + ENABLE_PJRT_COMPATIBILITY: 1 + # TODO: Needs final runs-on value + runs-on: arc-linux-x86-ct5lp-224-8tpu + container: + # TODO: Needs newer, light weight image + image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 + timeout-minutes: 45 + defaults: + run: + shell: bash -ex {0} + steps: + - uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # ratchet:actions/checkout@v4 + - name: Install JAX test requirements + run: | + pip install -U -r build/test-requirements.txt + - name: DEBUG HALT + run: | + echo "Halting" + sleep 180m + # TODO: build jax should be done on a step prior or we should just bazel test + - name: Build JAX + run: | + pip uninstall -y jaxlib + python3 build/build.py --use_clang + pip install -e . + ls -la dist/*.whl + pip install dist/*.whl + # Note the version it installs! Should be today's date + pip install -U --no-index --pre libtpu-nightly -f https://storage.googleapis.com/jax-releases/libtpu_releases.html + # Everything being built in this step downgrades numpy, reupgrade it + pip install "numpy>=2.0.0" + python3 -c 'import sys; print("python version:", sys.version)' + python3 -c 'import jax; print("jax version:", jax.__version__)' + python3 -c 'import jaxlib; print("jaxlib version:", jaxlib.__version__)' + # strings $HOME/.local/lib/python3.10/site-packages/libtpu/libtpu.so | grep 'Built on' + python3 -c 'import jax; print("libtpu version:", + jax.lib.xla_bridge.get_backend().platform_version)' + - name: Run tests + env: + JAX_PLATFORMS: tpu,cpu + PY_COLORS: 1 + NUM_TESTS: 8 + JAX_NUM_GENERATED_CASES: 25 + run: | + # Run single-accelerator tests in parallel + mkdir results + JAX_ENABLE_TPU_XDIST=true python3 -m pytest -n=$NUM_TESTS --tb=short \ + --junitxml=results/singlejunit.xml --maxfail=20 -m "not multiaccelerator" tests examples + # Run multi-accelerator across all chips + python3 -m pytest --tb=short --junitxml=results/multijunit.xml \ + --maxfail=20 -m "multiaccelerator" tests + - name: 'Upload Artifact' + if: success() || failure() + uses: actions/upload-artifact@65462800fd760344b1a7b4382951275a0abb4808 # ratchet:actions/upload-artifact@v4 + with: + name: junit + path: | + results/singlejunit.xml + results/multijunit.xml + retention-days: 1 + + + + + + + + + + + From 7c9506cca30e945976aade4a08d9e6788e2b42b0 Mon Sep 17 00:00:00 2001 From: Michael Hudgins Date: Mon, 24 Jun 2024 16:21:14 -0400 Subject: [PATCH 02/37] Rename cloud-tpu-presubmit to cloud-tpu-presubmit.yml --- .../workflows/{cloud-tpu-presubmit => cloud-tpu-presubmit.yml} | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename .github/workflows/{cloud-tpu-presubmit => cloud-tpu-presubmit.yml} (100%) diff --git a/.github/workflows/cloud-tpu-presubmit b/.github/workflows/cloud-tpu-presubmit.yml similarity index 100% rename from .github/workflows/cloud-tpu-presubmit rename to .github/workflows/cloud-tpu-presubmit.yml From 882973098e81f5a45c969ced8cafab76c73ca67c Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Wed, 26 Jun 2024 22:10:34 +0000 Subject: [PATCH 03/37] Add a tpu presubmit workflow --- .github/workflows/cloud-tpu-presubmit.yml | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index f15f51271666..28832d466307 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -16,6 +16,10 @@ jobs: # TODO: confirm final naming for optional label if: contains(github.event.pull_request.labels.*.name, 'optional_ci_tpu') name: "TPU v5e x 8 Presubmit" + strategy: + fail-fast: false # don't cancel all jobs on failure + matrix: + instances: ["one", "two"] env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value From c594236580a03bd0cc135a2a49c4c7b26e939c68 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Thu, 27 Jun 2024 16:25:12 -0700 Subject: [PATCH 04/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 28832d466307..4bc00edae716 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -18,8 +18,8 @@ jobs: name: "TPU v5e x 8 Presubmit" strategy: fail-fast: false # don't cancel all jobs on failure - matrix: - instances: ["one", "two"] +# matrix: +# instances: ["one", "two"] env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value From 067e211bc226e40b341890772d6e4a7352d17262 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Thu, 27 Jun 2024 21:05:04 -0700 Subject: [PATCH 05/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 4bc00edae716..28832d466307 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -18,8 +18,8 @@ jobs: name: "TPU v5e x 8 Presubmit" strategy: fail-fast: false # don't cancel all jobs on failure -# matrix: -# instances: ["one", "two"] + matrix: + instances: ["one", "two"] env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value From 9b4b45e4360fe79a780dcc2d1d49b002789b9697 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 8 Jul 2024 17:28:32 -0700 Subject: [PATCH 06/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 28832d466307..d8b0daf98a6f 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -19,7 +19,7 @@ jobs: strategy: fail-fast: false # don't cancel all jobs on failure matrix: - instances: ["one", "two"] + instances: ["one", "two", "three", "four"] env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value From 6fa2f66f985044ff7ec95d5f73049a69624085bc Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Fri, 2 Aug 2024 09:19:37 -0700 Subject: [PATCH 07/37] Create gcs_access.yml --- .github/workflows/gcs_access.yml | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) create mode 100644 .github/workflows/gcs_access.yml diff --git a/.github/workflows/gcs_access.yml b/.github/workflows/gcs_access.yml new file mode 100644 index 000000000000..c5707f84db44 --- /dev/null +++ b/.github/workflows/gcs_access.yml @@ -0,0 +1,27 @@ +# WorkFlow dispatch to stress test the runner system +name: GCS Access Test +on: + pull_request: + branches: + - main + workflow_dispatch: +# Cancel any previous iterations if a new commit is pushed +concurrency: + group: ${{ github.workflow }}-${{ github.ref }} + cancel-in-progress: true +jobs: + cpu-stress: + strategy: + matrix: + runner: ["arc-linux-x86-n2-64"] + instances: ["1"] + runs-on: ${{ matrix.runner }} + timeout-minutes: 45 + container: + # TODO: Needs newer, light weight image + image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 + steps: + - uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # ratchet:actions/checkout@v4 + - name: Gcloud + run: | + curl -X GET -H "Authorization: Bearer $(gcloud auth print-access-token)" "https://storage.googleapis.com/storage/v1/b/quoct-tf-test/o" From 6181511ee781457264b402285f6acdd9b7b1e900 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Fri, 2 Aug 2024 09:48:03 -0700 Subject: [PATCH 08/37] Update gcs_access.yml --- .github/workflows/gcs_access.yml | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/.github/workflows/gcs_access.yml b/.github/workflows/gcs_access.yml index c5707f84db44..1fd16c9ce172 100644 --- a/.github/workflows/gcs_access.yml +++ b/.github/workflows/gcs_access.yml @@ -25,3 +25,7 @@ jobs: - name: Gcloud run: | curl -X GET -H "Authorization: Bearer $(gcloud auth print-access-token)" "https://storage.googleapis.com/storage/v1/b/quoct-tf-test/o" + - name: Install JAX + run: | + pip install -U -r build/test-requirements.txt + pip install -U --pre jax==0.4.31.dev20240625 -f https://storage.googleapis.com/jax-releases/jax_nightly_releases.html From b7d6a1a143af07ecca4291899b743decdf84085c Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 5 Aug 2024 11:29:05 -0700 Subject: [PATCH 09/37] Update gcs_access.yml --- .github/workflows/gcs_access.yml | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/.github/workflows/gcs_access.yml b/.github/workflows/gcs_access.yml index 1fd16c9ce172..ff7775a7e530 100644 --- a/.github/workflows/gcs_access.yml +++ b/.github/workflows/gcs_access.yml @@ -22,6 +22,10 @@ jobs: image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 steps: - uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # ratchet:actions/checkout@v4 + - name: 'Set up Cloud SDK' + uses: 'google-github-actions/setup-gcloud@v2' + with: + version: '>= 363.0.0' - name: Gcloud run: | curl -X GET -H "Authorization: Bearer $(gcloud auth print-access-token)" "https://storage.googleapis.com/storage/v1/b/quoct-tf-test/o" From b08fb4ae41fc08c1defd4519ae606a2293ae8532 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 13 Aug 2024 16:47:42 -0700 Subject: [PATCH 10/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 43 +---------------------- 1 file changed, 1 insertion(+), 42 deletions(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index d8b0daf98a6f..91de7d81075b 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: arc-linux-x86-ct5lp-224-8tpu + runs-on: arc-linux-x86-n2-32-dev container: # TODO: Needs newer, light weight image image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 @@ -40,47 +40,6 @@ jobs: run: | echo "Halting" sleep 180m - # TODO: build jax should be done on a step prior or we should just bazel test - - name: Build JAX - run: | - pip uninstall -y jaxlib - python3 build/build.py --use_clang - pip install -e . - ls -la dist/*.whl - pip install dist/*.whl - # Note the version it installs! Should be today's date - pip install -U --no-index --pre libtpu-nightly -f https://storage.googleapis.com/jax-releases/libtpu_releases.html - # Everything being built in this step downgrades numpy, reupgrade it - pip install "numpy>=2.0.0" - python3 -c 'import sys; print("python version:", sys.version)' - python3 -c 'import jax; print("jax version:", jax.__version__)' - python3 -c 'import jaxlib; print("jaxlib version:", jaxlib.__version__)' - # strings $HOME/.local/lib/python3.10/site-packages/libtpu/libtpu.so | grep 'Built on' - python3 -c 'import jax; print("libtpu version:", - jax.lib.xla_bridge.get_backend().platform_version)' - - name: Run tests - env: - JAX_PLATFORMS: tpu,cpu - PY_COLORS: 1 - NUM_TESTS: 8 - JAX_NUM_GENERATED_CASES: 25 - run: | - # Run single-accelerator tests in parallel - mkdir results - JAX_ENABLE_TPU_XDIST=true python3 -m pytest -n=$NUM_TESTS --tb=short \ - --junitxml=results/singlejunit.xml --maxfail=20 -m "not multiaccelerator" tests examples - # Run multi-accelerator across all chips - python3 -m pytest --tb=short --junitxml=results/multijunit.xml \ - --maxfail=20 -m "multiaccelerator" tests - - name: 'Upload Artifact' - if: success() || failure() - uses: actions/upload-artifact@65462800fd760344b1a7b4382951275a0abb4808 # ratchet:actions/upload-artifact@v4 - with: - name: junit - path: | - results/singlejunit.xml - results/multijunit.xml - retention-days: 1 From ad8965dee05a60e84711345cc5850c7122596c99 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 13 Aug 2024 16:48:16 -0700 Subject: [PATCH 11/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 91de7d81075b..73b6b7dea8f3 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -19,7 +19,7 @@ jobs: strategy: fail-fast: false # don't cancel all jobs on failure matrix: - instances: ["one", "two", "three", "four"] + instances: ["one"] env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value From f6d55dc9f952183245c2758c42d4044136a4d072 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 13 Aug 2024 16:48:27 -0700 Subject: [PATCH 12/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 73b6b7dea8f3..ad16a44314ad 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -19,7 +19,7 @@ jobs: strategy: fail-fast: false # don't cancel all jobs on failure matrix: - instances: ["one"] + instances: ["one", "two"] env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value From 755a5416c83a2174608a3f98fe1ef23461216ada Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 13 Aug 2024 16:58:50 -0700 Subject: [PATCH 13/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index ad16a44314ad..ffd163c98014 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: arc-linux-x86-n2-32-dev + runs-on: arc-linux-x86-n2-64-dev container: # TODO: Needs newer, light weight image image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 From ad5105bea250d003b0618897c120d83afad25109 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 13 Aug 2024 17:47:52 -0700 Subject: [PATCH 14/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index ffd163c98014..ce83d3792f43 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -39,7 +39,7 @@ jobs: - name: DEBUG HALT run: | echo "Halting" - sleep 180m + sleep 30m From 7da0a6a2ea76835f7af914dc4dcf425bff5342bb Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Wed, 14 Aug 2024 15:27:42 -0700 Subject: [PATCH 15/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index ce83d3792f43..d43ddea5c31a 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: arc-linux-x86-n2-64-dev + runs-on: arc-linux-x86-n2-64-test container: # TODO: Needs newer, light weight image image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 From 4388e50aab5ee3d5f3f1353d2fad4e068c071d5b Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 20 Aug 2024 11:34:08 -0700 Subject: [PATCH 16/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index d43ddea5c31a..0aaa5a9d6676 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: arc-linux-x86-n2-64-test + runs-on: t2a-standard-48-dev container: # TODO: Needs newer, light weight image image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 From 4a000df885d512c42465629c6a139c411fd794fd Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 20 Aug 2024 12:52:58 -0700 Subject: [PATCH 17/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 0aaa5a9d6676..54fbfd332a87 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -26,7 +26,7 @@ jobs: runs-on: t2a-standard-48-dev container: # TODO: Needs newer, light weight image - image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 + image: us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/build-arm64:jax-latest-multi-python timeout-minutes: 45 defaults: run: From 22c801636c60b386de8024d3e7b97692d02e4e9a Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Fri, 23 Aug 2024 15:00:18 -0700 Subject: [PATCH 18/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 54fbfd332a87..8b2aa9e030f4 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: t2a-standard-48-dev + runs-on: arc-linux-x86-n2-64-test container: # TODO: Needs newer, light weight image image: us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/build-arm64:jax-latest-multi-python From d42fbee5e1bc2ed5f06be554a5ed561995a2bb85 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Fri, 23 Aug 2024 15:07:58 -0700 Subject: [PATCH 19/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 8b2aa9e030f4..d43ddea5c31a 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -26,7 +26,7 @@ jobs: runs-on: arc-linux-x86-n2-64-test container: # TODO: Needs newer, light weight image - image: us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/build-arm64:jax-latest-multi-python + image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 timeout-minutes: 45 defaults: run: From 7d0afdd1df9586ee2b7d204dd81c5543d8276b8f Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 8 Oct 2024 14:14:54 -0700 Subject: [PATCH 20/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index d43ddea5c31a..218b7e93cef4 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: arc-linux-x86-n2-64-test + runs-on: linux-x86-n2-16-dev container: # TODO: Needs newer, light weight image image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 From 7cd8fe0e227d4d0f6e19453ccf4a38b5f5a1f772 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 8 Oct 2024 14:16:51 -0700 Subject: [PATCH 21/37] Delete .github/workflows/gcs_access.yml --- .github/workflows/gcs_access.yml | 35 -------------------------------- 1 file changed, 35 deletions(-) delete mode 100644 .github/workflows/gcs_access.yml diff --git a/.github/workflows/gcs_access.yml b/.github/workflows/gcs_access.yml deleted file mode 100644 index ff7775a7e530..000000000000 --- a/.github/workflows/gcs_access.yml +++ /dev/null @@ -1,35 +0,0 @@ -# WorkFlow dispatch to stress test the runner system -name: GCS Access Test -on: - pull_request: - branches: - - main - workflow_dispatch: -# Cancel any previous iterations if a new commit is pushed -concurrency: - group: ${{ github.workflow }}-${{ github.ref }} - cancel-in-progress: true -jobs: - cpu-stress: - strategy: - matrix: - runner: ["arc-linux-x86-n2-64"] - instances: ["1"] - runs-on: ${{ matrix.runner }} - timeout-minutes: 45 - container: - # TODO: Needs newer, light weight image - image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 - steps: - - uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # ratchet:actions/checkout@v4 - - name: 'Set up Cloud SDK' - uses: 'google-github-actions/setup-gcloud@v2' - with: - version: '>= 363.0.0' - - name: Gcloud - run: | - curl -X GET -H "Authorization: Bearer $(gcloud auth print-access-token)" "https://storage.googleapis.com/storage/v1/b/quoct-tf-test/o" - - name: Install JAX - run: | - pip install -U -r build/test-requirements.txt - pip install -U --pre jax==0.4.31.dev20240625 -f https://storage.googleapis.com/jax-releases/jax_nightly_releases.html From aa6f6181ad14127794e80613613d6d25c47855a4 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Wed, 9 Oct 2024 08:55:31 -0700 Subject: [PATCH 22/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 218b7e93cef4..9abbd7c74534 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: linux-x86-n2-16-dev + runs-on: linux-x86-n2-16-quoct container: # TODO: Needs newer, light weight image image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 From e6db8e201f2fe4eb8bf800d5d347bf1cf22ba988 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Thu, 10 Oct 2024 13:27:57 -0700 Subject: [PATCH 23/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 9abbd7c74534..4d5fd3df6c84 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -23,7 +23,7 @@ jobs: env: ENABLE_PJRT_COMPATIBILITY: 1 # TODO: Needs final runs-on value - runs-on: linux-x86-n2-16-quoct + runs-on: linux-x86-n2-16 container: # TODO: Needs newer, light weight image image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 From b85462d8be89aac9b46349c2c73b33555f17defd Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 15:25:21 -0700 Subject: [PATCH 24/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 95 +++++++++++++---------- 1 file changed, 53 insertions(+), 42 deletions(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 4d5fd3df6c84..d87a7518bea2 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -1,53 +1,64 @@ -# Cloud TPU CI -name: Cloud TPU Presubmit -# Run on pull_request that is labeled as "optional_ci_tpu" or workflow dispatch +name: Build JAX Artifacts + on: pull_request: branches: - main - types: [labeled, synchronize] - workflow_dispatch: -# Cancel any previous iterations if a new commit is pushed -concurrency: - group: ${{ github.workflow }}-${{ github.ref }} - cancel-in-progress: true + workflow_call: + jobs: - cloud-tpu-test: - # TODO: confirm final naming for optional label - if: contains(github.event.pull_request.labels.*.name, 'optional_ci_tpu') - name: "TPU v5e x 8 Presubmit" - strategy: - fail-fast: false # don't cancel all jobs on failure - matrix: - instances: ["one", "two"] - env: - ENABLE_PJRT_COMPATIBILITY: 1 - # TODO: Needs final runs-on value - runs-on: linux-x86-n2-16 - container: - # TODO: Needs newer, light weight image - image: index.docker.io/tensorflow/build@sha256:7fb38f0319bda36393cad7f40670aa22352b44421bb906f5cf34d543acd8e1d2 # ratchet:tensorflow/build:latest-python3.11 - timeout-minutes: 45 + build: + continue-on-error: true defaults: run: - shell: bash -ex {0} - steps: - - uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # ratchet:actions/checkout@v4 - - name: Install JAX test requirements - run: | - pip install -U -r build/test-requirements.txt - - name: DEBUG HALT - run: | - echo "Halting" - sleep 30m - - - - - - - + # Explicitly set the shell to bash to override the default Windows environment, i.e, cmd. + shell: bash + strategy: + matrix: + runner: ["windows-x86-n2-64-dev"] + artifact: ["jax", "jaxlib", "jax-cuda-pjrt", "jax-cuda-plugin"] + python: ["3.10", "3.11", "3.12"] + # jax-cuda-pjrt and jax are pure Python packages so they do not need to be built for each + # Python version. + exclude: + # Pure Python packages do not need to be built for each Python version. + - artifact: "jax-cuda-pjrt" + python: "3.10" + - artifact: "jax-cuda-pjrt" + python: "3.11" + - artifact: "jax" + python: "3.10" + - artifact: "jax" + python: "3.11" + # jax is a pure Python package so it does not need to be built on multiple platforms. + - artifact: "jax" + runner: "windows-x86-n2-64" + - artifact: "jax" + runner: "linux-arm64-t2a-48" + # jax-cuda-plugin and jax-cuda-pjrt are not supported on Windows. + - artifact: "jax-cuda-plugin" + runner: "windows-x86-n2-64" + - artifact: "jax-cuda-pjrt" + runner: "windows-x86-n2-64" + runs-on: ${{ matrix.runner }} + container: ${{ (contains(matrix.runner, 'linux-x86') && 'us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/ml-build:latest') || + (contains(matrix.runner, 'linux-arm64') && 'us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/linux-arm64-arc-container:latest') || + (contains(matrix.runner, 'windows-x86') && null) }} + env: + # Do not run Docker container for Linux runners. Linux runners already run in a Docker container. + JAXCI_RUN_DOCKER_CONTAINER: 0 + # Use RBE to build the artifacts where possibl (Linux x86 and Windows). + JAXCI_BUILD_ARTIFACT_WITH_RBE: 1 + steps: + - uses: actions/checkout@v3 + # Halt for testing + - name: Wait For Connection + uses: ./actions/ci_connection/ + - name: Build ${{ matrix.artifact }} + env: + JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" + run: ./ci/build_artifacts.sh "ci/envs/build_artifacts/${{ matrix.artifact }}" From d3d0729b8306490b9d134ed7cfa60546819f0b7a Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 15:37:12 -0700 Subject: [PATCH 25/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index d87a7518bea2..954a12d4b555 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,4 +61,5 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: ./ci/build_artifacts.sh "ci/envs/build_artifacts/${{ matrix.artifact }}" + run: choco install llvm --version=18.1.4 --yes --no-progress --allow-downgrade + From 5493f2dba8287ee81b878219e19e351c8d58e821 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 15:48:04 -0700 Subject: [PATCH 26/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 954a12d4b555..df492721bb50 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,5 +61,5 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: choco install llvm --version=18.1.4 --yes --no-progress --allow-downgrade + run: ls From 2590d18c923380b55410b6cfdfde4a8d5134d136 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 15:56:04 -0700 Subject: [PATCH 27/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index df492721bb50..bf46d0b94c6e 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,5 +61,5 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: ls + run: ls && timeout 30 && ls From ce2a59b15af60380316499fe1066724454d1bdc6 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 16:17:09 -0700 Subject: [PATCH 28/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index bf46d0b94c6e..82f0883343d8 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,5 +61,5 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: ls && timeout 30 && ls + run: ls; timeout 30; ls From 0123a73e9bab4b71871bed49b685bed9054eb24d Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 16:19:13 -0700 Subject: [PATCH 29/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 82f0883343d8..9c87aecdd638 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,5 +61,5 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: ls; timeout 30; ls + run: timeout 100 From 81ffebe50d7d50d61ee177bfb909672b95135646 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 16:20:31 -0700 Subject: [PATCH 30/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 9c87aecdd638..90787cb318ec 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,5 +61,4 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: timeout 100 - + run: ls From 12f0281c25f762df469449cb86c57bb7f9404b0e Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 16:22:19 -0700 Subject: [PATCH 31/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 90787cb318ec..16934708b8b4 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,4 +61,4 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: ls + run: sleep 100 From 5a226f232e813a4055dcef23aacaba681e801c9d Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 28 Oct 2024 16:39:33 -0700 Subject: [PATCH 32/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 16934708b8b4..36651ba97179 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -61,4 +61,4 @@ jobs: - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: sleep 100 + run: sleep 400 From e2b9c92d6736f24d3dd31601352fda25d1a39163 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 3 Dec 2024 19:09:17 -0800 Subject: [PATCH 33/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 36651ba97179..d853d3de313f 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -15,9 +15,9 @@ jobs: shell: bash strategy: matrix: - runner: ["windows-x86-n2-64-dev"] + runner: ["linux-x86-n2-64"] artifact: ["jax", "jaxlib", "jax-cuda-pjrt", "jax-cuda-plugin"] - python: ["3.10", "3.11", "3.12"] + python: ["3.11"] # jax-cuda-pjrt and jax are pure Python packages so they do not need to be built for each # Python version. exclude: @@ -43,9 +43,7 @@ jobs: runs-on: ${{ matrix.runner }} - container: ${{ (contains(matrix.runner, 'linux-x86') && 'us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/ml-build:latest') || - (contains(matrix.runner, 'linux-arm64') && 'us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/linux-arm64-arc-container:latest') || - (contains(matrix.runner, 'windows-x86') && null) }} + container: "us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/ml-build:702526224" env: # Do not run Docker container for Linux runners. Linux runners already run in a Docker container. @@ -55,10 +53,11 @@ jobs: steps: - uses: actions/checkout@v3 - # Halt for testing - - name: Wait For Connection - uses: ./actions/ci_connection/ - name: Build ${{ matrix.artifact }} env: JAXCI_HERMETIC_PYTHON_VERSION: "${{ matrix.python }}" - run: sleep 400 + run: echo "Testing quoct" + - name: Another Test + run: ls + - name: Another Test + run: sleep 500 From f2a74c2cfce0c2b9fb4aae5d49227ab04dd98033 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Tue, 3 Dec 2024 19:45:05 -0800 Subject: [PATCH 34/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index d853d3de313f..0991d74fbd43 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -15,7 +15,7 @@ jobs: shell: bash strategy: matrix: - runner: ["linux-x86-n2-64"] + runner: ["linux-x86-n2-64-quoct"] artifact: ["jax", "jaxlib", "jax-cuda-pjrt", "jax-cuda-plugin"] python: ["3.11"] # jax-cuda-pjrt and jax are pure Python packages so they do not need to be built for each From e70fd114b1ab3fc43ef36e610a2f9a7b5ff770f4 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Thu, 5 Dec 2024 14:32:35 -0800 Subject: [PATCH 35/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 0991d74fbd43..e94ca3535446 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -15,7 +15,7 @@ jobs: shell: bash strategy: matrix: - runner: ["linux-x86-n2-64-quoct"] + runner: ["linux-arm64-c4a-16"] artifact: ["jax", "jaxlib", "jax-cuda-pjrt", "jax-cuda-plugin"] python: ["3.11"] # jax-cuda-pjrt and jax are pure Python packages so they do not need to be built for each From d721eac27b3897ef366a6e166adf607838c0fc41 Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 16 Dec 2024 11:22:47 -0800 Subject: [PATCH 36/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index e94ca3535446..97f22038bd71 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -60,4 +60,4 @@ jobs: - name: Another Test run: ls - name: Another Test - run: sleep 500 + run: sleep 10000 From af1467a3ab9c700289bbca8d3ba1bd344ec1a36c Mon Sep 17 00:00:00 2001 From: Quoc Truong Date: Mon, 16 Dec 2024 13:06:39 -0800 Subject: [PATCH 37/37] Update cloud-tpu-presubmit.yml --- .github/workflows/cloud-tpu-presubmit.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/cloud-tpu-presubmit.yml b/.github/workflows/cloud-tpu-presubmit.yml index 97f22038bd71..9365639d4916 100644 --- a/.github/workflows/cloud-tpu-presubmit.yml +++ b/.github/workflows/cloud-tpu-presubmit.yml @@ -43,7 +43,7 @@ jobs: runs-on: ${{ matrix.runner }} - container: "us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/ml-build:702526224" + container: "us-central1-docker.pkg.dev/tensorflow-sigs/tensorflow/ml-build:latest" env: # Do not run Docker container for Linux runners. Linux runners already run in a Docker container.