From 0eeedbe36714bf1292ee488caebc7062e8104222 Mon Sep 17 00:00:00 2001 From: Pauline Bailly-Masson <155966238+paulinebm@users.noreply.github.com> Date: Tue, 7 Jan 2025 15:16:45 +0100 Subject: [PATCH] Create build-trtllm2 --- .github/workflows/build-trtllm2 | 51 +++++++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+) create mode 100644 .github/workflows/build-trtllm2 diff --git a/.github/workflows/build-trtllm2 b/.github/workflows/build-trtllm2 new file mode 100644 index 00000000..34d00ee2 --- /dev/null +++ b/.github/workflows/build-trtllm2 @@ -0,0 +1,51 @@ +name: Build TensorRT-LLM2 +on: + workflow_call: + inputs: + runs-on: + type: string + description: "Which instance type to use to run the workflow" + required: true + + outputs: + docker_image: + description: "Reference to the Docker Image build by this workflow" + value: ${{ jobs.build-and-push.outputs.docker_image }} + label: + description: "Label generated for this build" + value: ${{ jobs.build-and-push.outputs.label }} + +permissions: + contents: read # Required to check out repository. + id-token: write # Required to authenticate via OIDC. + +jobs: + build-and-push: + concurrency: + group: ${{ github.workflow }}-${{ github.job }}-${{ github.head_ref || github.run_id }} + outputs: + docker_image: ${{ steps.final.outputs.docker_image }} + label: ${ steps.final.outputs.label }} + runs-on: + group: ${{ inputs.runs-on }} + steps: + - uses: actions/checkout@v4 + + - name: Inject slug/short variables + uses: rlespinasse/github-slug-action@v4.4.1 + + - name: Extract TensorRT-LLM version + run: | + echo "TENSORRT_LLM_VERSION=$(grep -oP '([a-z,0-9]{40})' $GITHUB_WORKSPACE/backends/trtllm/cmake/trtllm.cmake)" >> $GITHUB_ENV + echo "TensorRT-LLM version: ${{ env.TENSORRT_LLM_VERSION }}" + - name: Set LABEL + run: | + echo "LABEL=trtllm" >> $GITHUB_ENV + - name: "Configure AWS Credentials" + id: aws-creds + uses: aws-actions/configure-aws-credentials@v4 + with: + aws-region: us-east-1 + role-to-assume: ${{ secrets.AWS_ROLE_GITHUB_TGI_TEST }} + role-duration-seconds: 7200 + output-credentials: true