Merge branch 'main' into npetrovic/ttl-sharded-sweep-coregrid

tenstorrent · Jun 7, 2024 · 496d973 · 496d973
2 parents db3992c + 5ee76a6
commit 496d973
Show file tree

Hide file tree

Showing 884 changed files with 31,076 additions and 13,485 deletions.
diff --git a/.github/actions/install-metal-deps/dependencies.json b/.github/actions/install-metal-deps/dependencies.json
@@ -5,7 +5,6 @@
     "python3.8-venv=3.8.10-0ubuntu1~20.04.9",
     "libgoogle-glog-dev=0.4.0-1build1",
     "libyaml-cpp-dev=0.6.2-4ubuntu1",
-    "libboost-all-dev=1.71.0.0ubuntu2",
     "libsndfile1=1.0.28-7ubuntu0.2",
     "libhwloc-dev",
     "graphviz",

diff --git a/.github/actions/install-python-deps/action.yml b/.github/actions/install-python-deps/action.yml
@@ -17,5 +17,6 @@ runs:
         venv-dir: ${{ github.workspace }}/python_env
         cache-dependency-path: |
           tt_metal/python_env/requirements-dev.txt
+          docs/requirements-docs.txt
           pyproject.toml
         install-cmd: ./create_venv.sh
diff --git a/.github/workflows/build-artifact.yaml b/.github/workflows/build-artifact.yaml
@@ -32,11 +32,11 @@ jobs:
           git submodule update --init --recursive
       - name: Build tt-metal and libs
         run: |
-          cmake -B build -G Ninja -DCMAKE_CXX_COMPILER=clang++-17
+          cmake -B build -G Ninja
           cmake --build build --target tests
           cmake --build build --target install
       - name: 'Tar files'
-        run: tar -cvf ttm_${{ matrix.arch }}.tar build/hw build/lib tt_eager/tt_lib/*.so ttnn/ttnn/*.so build/programming_examples build/test build/tools
+        run: tar -cvf ttm_${{ matrix.arch }}.tar build/hw build/lib tt_eager/tt_lib/*.so ttnn/ttnn/*.so build/programming_examples build/test build/tools runtime
       - name: 'Upload Artifact'
         uses: actions/upload-artifact@v4
         with:

diff --git a/.github/workflows/build.yaml b/.github/workflows/build.yaml
@@ -34,7 +34,7 @@ jobs:
           echo "TT_METAL_HOME=$(pwd)" >> $GITHUB_ENV
       - name: Build tt-metal libraries
         run: |
-          cmake -B build -G Ninja -DCMAKE_CXX_COMPILER=clang++-17
+          cmake -B build -G Ninja
           cmake --build build
       - name: Build tt-metal C++ tests
         run: |

diff --git a/.github/workflows/eager-package-main.yaml b/.github/workflows/eager-package-main.yaml
@@ -78,12 +78,6 @@ jobs:
           source env/bin/activate
           python3 -m tt_metal.scripts.get_home_dir --short
           echo "TT_METAL_HOME=$(python3 -m tt_metal.scripts.get_home_dir --short)" >> $GITHUB_ENV
-      - name: Set up kernel builds
-        working-directory: tests/end_to_end_tests
-        run: |
-          echo $TT_METAL_HOME
-          source env/bin/activate
-          python3 -m tt_metal.scripts.set_up_kernels prepare
       - name: Activate env and run release tests - silicon
         timeout-minutes: 2
         shell: bash

diff --git a/.github/workflows/fast-dispatch-full-regressions-and-models.yaml b/.github/workflows/fast-dispatch-full-regressions-and-models.yaml
@@ -19,13 +19,14 @@ jobs:
       matrix:
         test-group:
           [
-            { name: "Common models GS", arch: grayskull, cmd: tests/scripts/nightly/run_common_models.sh, timeout: 40 },
-            { name: "Common models N300 WH B0", arch: wormhole_b0, cmd: tests/scripts/nightly/run_common_models.sh, timeout: 40 },
-            { name: "GS-only ttnn nightly", arch: grayskull, cmd: tests/scripts/nightly/run_ttnn.sh, timeout: 40 },
-            { name: "GS-only models", arch: grayskull, cmd: tests/scripts/nightly/run_gs_only.sh, timeout: 40 },
-            { name: "N300 WH-only models", arch: wormhole_b0, cmd: tests/scripts/nightly/run_wh_b0_only.sh, timeout: 60 },
+            { name: "Common models GS", arch: grayskull, cmd: tests/scripts/single_card/nightly/run_common_models.sh, timeout: 40 },
+            { name: "Common models N300 WH B0", arch: wormhole_b0, cmd: tests/scripts/single_card/nightly/run_common_models.sh, timeout: 40 },
+            { name: "GS-only ttnn nightly", arch: grayskull, cmd: tests/scripts/single_card/nightly/run_ttnn.sh, timeout: 40 },
+            { name: "GS-only models", arch: grayskull, cmd: tests/scripts/single_card/nightly/run_gs_only.sh, timeout: 40 },
+            { name: "N300 WH-only models", arch: wormhole_b0, cmd: tests/scripts/single_card/nightly/run_wh_b0_only.sh, timeout: 30 },
             { name: "API tests GS", arch: grayskull, cmd: ./tests/scripts/run_tests.sh --tt-arch grayskull --pipeline-type frequent_api --dispatch-mode fast, timeout: 40 },
             { name: "API tests N300 WH B0", arch: wormhole_b0, cmd: ./tests/scripts/run_tests.sh --tt-arch wormhole_b0 --pipeline-type frequent_api --dispatch-mode fast, timeout: 40 },
+            { name: "[Unstable] N300 models", arch: wormhole_b0, cmd: tests/scripts/single_card/nightly/run_wh_b0_unstable.sh, timeout: 35 },
           ]
     name: FD ${{ matrix.test-group.name }} ${{ matrix.test-group.arch }}
     env:

diff --git a/.github/workflows/single-card-demo-tests.yaml b/.github/workflows/single-card-demo-tests.yaml
@@ -0,0 +1,64 @@
+name: "[Single-card] Demo tests"
+
+on:
+  workflow_dispatch:
+  schedule:
+    - cron: "0 0 * * 1,2,3,4,5"
+    - cron: "0 */4 * * 0,6"
+
+jobs:
+  build-artifact:
+    uses: ./.github/workflows/build-artifact.yaml
+    with:
+      arch: '["wormhole_b0"]'
+    secrets: inherit
+  t3000-demo-tests:
+    needs: build-artifact
+    strategy:
+      fail-fast: false
+      matrix:
+        test-group: [
+          {
+            name: "N150",
+            arch: wormhole_b0,
+            runs-on: ["wormhole_b0", "multi-chip-num-pcie-1", "multi-chip-num-chips-1"],
+            cmd: './tests/scripts/run_tests.sh --tt-arch wormhole_b0 --pipeline-type demos_single_card_n150 --dispatch-mode ""'
+          },
+          {
+            name: "N300",
+            arch: wormhole_b0,
+            runs-on: ["wormhole_b0", "multi-chip-num-pcie-1", "multi-chip-num-chips-2"],
+            cmd: './tests/scripts/run_tests.sh --tt-arch wormhole_b0 --pipeline-type demos_single_card_n300 --dispatch-mode ""'
+          }
+        ]
+    name: ${{ matrix.test-group.name }}
+    env:
+      TT_METAL_ENV: ${{ vars.TT_METAL_ENV }}
+      ARCH_NAME: ${{ matrix.test-group.arch }}
+      LOGURU_LEVEL: INFO
+      LD_LIBRARY_PATH: ${{ github.workspace }}/build/lib
+    environment: dev
+    runs-on: ${{ matrix.test-group.runs-on }}
+    steps:
+      - uses: tenstorrent-metal/metal-workflows/.github/actions/[email protected]
+      - name: Ensure weka mount is active
+        run: |
+          sudo systemctl restart mnt-MLPerf.mount
+          sudo /etc/rc.local
+          ls -al /mnt/MLPerf/bit_error_tests
+      - name: Set up dynamic env vars for build
+        run: |
+          echo "TT_METAL_HOME=$(pwd)" >> $GITHUB_ENV
+      - uses: actions/download-artifact@v4
+        with:
+          name: TTMetal_build_${{ matrix.test-group.arch }}
+      - name: Extract files
+        run: tar -xvf ttm_${{ matrix.test-group.arch }}.tar
+      - uses: ./.github/actions/install-python-deps
+      - name: Run demo regression tests
+        timeout-minutes: 150
+        run: |
+          source ${{ github.workspace }}/python_env/bin/activate
+          cd $TT_METAL_HOME
+          export PYTHONPATH=$TT_METAL_HOME
+          ${{ matrix.test-group.cmd }}
diff --git a/.github/workflows/t3000-demo-tests.yaml b/.github/workflows/t3000-demo-tests.yaml
@@ -3,7 +3,7 @@ name: "[T3K] T3000 demo tests"
 on:
   workflow_dispatch:
   schedule:
-    - cron: '0 0 * * 6' # This cron schedule runs the workflow every Saturday at 12am UTC
+    - cron: '0 0 * * 1,3,5' # This cron schedule runs the workflow every Monday/Wednesday/Friday at 12am UTC
 
 jobs:
   build-artifact:
@@ -53,11 +53,6 @@ jobs:
           cd $TT_METAL_HOME
           export PYTHONPATH=$TT_METAL_HOME
           ${{ matrix.test-group.cmd }}
-      - uses: ./.github/actions/slack-report
-        if: ${{ failure() }}
-        with:
-          slack_webhook_url: ${{ secrets.SLACK_WEBHOOK_URL }}
-          owner: U03P1KMAZFE # Tapasvi Patel
       - name: Disable performance mode
         if: always()
         run: |

diff --git a/.github/workflows/t3000-frequent-tests.yaml b/.github/workflows/t3000-frequent-tests.yaml
@@ -50,8 +50,3 @@ jobs:
           cd $TT_METAL_HOME
           export PYTHONPATH=$TT_METAL_HOME
           ${{ matrix.test-group.cmd }}
-      - uses: ./.github/actions/slack-report
-        if: ${{ failure() }}
-        with:
-          slack_webhook_url: ${{ secrets.SLACK_WEBHOOK_URL }}
-          owner: U03P1KMAZFE # Tapasvi Patel
diff --git a/.github/workflows/t3000-model-perf-tests.yaml b/.github/workflows/t3000-model-perf-tests.yaml
@@ -67,11 +67,6 @@ jobs:
           cd $TT_METAL_HOME
           export PYTHONPATH=$TT_METAL_HOME
           ${{ matrix.test-group.cmd }}
-      - uses: ./.github/actions/slack-report
-        if: ${{ failure() }}
-        with:
-          slack_webhook_url: ${{ secrets.SLACK_WEBHOOK_URL }}
-          owner: U03P1KMAZFE # Tapasvi Patel
       - name: Check perf report exists
         id: check-perf-report
         if: ${{ !cancelled() }}

diff --git a/.github/workflows/t3000-profiler-tests.yaml b/.github/workflows/t3000-profiler-tests.yaml
@@ -39,8 +39,3 @@ jobs:
         timeout-minutes: 30
         run: |
           ./tests/scripts/run_profiler_regressions.sh
-      - uses: ./.github/actions/slack-report
-        if: ${{ failure() }}
-        with:
-          slack_webhook_url: ${{ secrets.SLACK_WEBHOOK_URL }}
-          owner: U03P1KMAZFE # Tapasvi Patel
diff --git a/.github/workflows/t3000-unit-tests.yaml b/.github/workflows/t3000-unit-tests.yaml
@@ -50,8 +50,3 @@ jobs:
           cd $TT_METAL_HOME
           export PYTHONPATH=$TT_METAL_HOME
           ${{ matrix.test-group.cmd }}
-      - uses: ./.github/actions/slack-report
-        if: ${{ failure() }}
-        with:
-          slack_webhook_url: ${{ secrets.SLACK_WEBHOOK_URL }}
-          owner: U03P1KMAZFE # Tapasvi Patel
diff --git a/.github/workflows/ttnn-post-commit.yaml b/.github/workflows/ttnn-post-commit.yaml
@@ -25,8 +25,7 @@ jobs:
             cmd: pytest $TT_METAL_HOME/tests/ttnn/unit_tests -xv --splits 2 --group 2 -m "not disable_fast_runtime_mode"
           - name: ttnn group 3
             cmd: pytest $TT_METAL_HOME/tests/ttnn/unit_tests -xv -m requires_fast_runtime_mode_off
-            env:
-              TTNN_CONFIG_OVERRIDES: '{"enable_fast_runtime_mode": false}'
+            fast_runtime_mode_off: true
           - name: ttnn cpp tests
             cmd: ./build/test/ttnn/unit_tests_ttnn
     name: ${{ matrix.test-group.name }} ${{ matrix.runner-info.arch }} ${{ matrix.runner-info.name }}
@@ -47,6 +46,10 @@ jobs:
       - name: Extract files
         run: tar -xvf ttm_${{ matrix.runner-info.arch }}.tar
       - uses: ./.github/actions/install-python-deps
+      - name: Set ttnn fast runtime if exists in config
+        if: ${{ matrix.test-group.fast_runtime_mode_off }}
+        run: |
+          echo "TTNN_CONFIG_OVERRIDES={\"enable_fast_runtime_mode\": false}" >> $GITHUB_ENV
       - name: ${{ matrix.test-group.name }} tests
         timeout-minutes: 45
         run: |

diff --git a/.gitignore b/.gitignore
@@ -1,6 +1,7 @@
 .cache
 .vscode
 .idea
+runtime
 *.log
 *.csv
 *.xlsx
@@ -9,7 +10,7 @@ test_hlk_args_init_gen
 tt_build
 tt_debug
 build
-python_env/
+/python_env/
 
 /llk_out/
 /out/
@@ -118,3 +119,6 @@ compile_commands.json
 # rpath_check
 tt_eager/tt_lib/.rpath_checked*
 ttnn/ttnn/.rpath_checked
+
+# exclude packages brough in from CPM
+.cpmcache