Migrate workflows to linux_job_v2 (#6348)

1. Migrate all the workflows to liux_job_v2 since its uses Manylinux
2.28 : RFC: pytorch/pytorch#123649
2. Remove stronghold workflow, since not used
3. Use get_stable_cuda_version to validate repacked binary size whls

Author: atalman

.github/workflows/stronghold.yml

@@ -8,7 +8,7 @@ on:
 
 jobs:
   test-linux:
-    uses: ./.github/workflows/linux_job.yml
+    uses: ./.github/workflows/linux_job_v2.yml
     strategy:
       fail-fast: false
       matrix:

.github/workflows/test-export-matrix-variables.yml

@@ -24,7 +24,7 @@ jobs:
   test-tools:
     name: Test tools
     if: ${{ github.repository == 'pytorch/test-infra' }}
-    uses: ./.github/workflows/linux_job.yml
+    uses: ./.github/workflows/linux_job_v2.yml
     with:
       docker-image: python:3.11.0-slim-bullseye
       runner: linux.large
@@ -42,7 +42,7 @@ jobs:
   test-github-scripts:
     name: Test github scripts
     if: ${{ github.repository == 'pytorch/test-infra' }}
-    uses: ./.github/workflows/linux_job.yml
+    uses: ./.github/workflows/linux_job_v2.yml
     with:
       docker-image: python:3.11.0-slim-bullseye
       runner: linux.large

.github/workflows/tests.yml

@@ -104,7 +104,7 @@ jobs:
     strategy:
       matrix: ${{ fromJson(needs.generate-aarch64-linux-matrix.outputs.matrix) }}
       fail-fast: false
-    uses: pytorch/test-infra/.github/workflows/linux_job.yml@main
+    uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
     name: ${{ matrix.build_name }}
     with:
       runner: ${{ matrix.validation_runner }}

.github/workflows/validate-aarch64-linux-binaries.yml

@@ -55,7 +55,7 @@ jobs:
     strategy:
       matrix: ${{ fromJson(needs.generate-matrix.outputs.matrix) }}
       fail-fast: false
-    uses: ./.github/workflows/linux_job.yml
+    uses: ./.github/workflows/linux_job_v2.yml
     with:
       runner: ${{ matrix.validation_runner }}
       repository: "pytorch/pytorch"

.github/workflows/validate-docker-images.yml

@@ -127,7 +127,7 @@ jobs:
     strategy:
       matrix: ${{ fromJson(needs.generate-linux-matrix.outputs.matrix) }}
       fail-fast: false
-    uses: ./.github/workflows/linux_job.yml
+    uses: ./.github/workflows/linux_job_v2.yml
     name: ${{ matrix.build_name }}
     with:
       runner: ${{ matrix.validation_runner }}
@@ -172,7 +172,7 @@ jobs:
         source ../../test-infra/.github/scripts/validate_binaries.sh
 
   linux-amazon-2023:
-    uses: ./.github/workflows/linux_job.yml
+    uses: ./.github/workflows/linux_job_v2.yml
     name: amazon-linux-2023-test
     with:
       runner: "linux.g5.4xlarge.nvidia.gpu"

.github/workflows/validate-linux-binaries.yml

@@ -33,7 +33,7 @@ jobs:
     strategy:
       matrix: ${{ fromJson(needs.generate-linux-matrix.outputs.matrix) }}
       fail-fast: false
-    uses: ./.github/workflows/linux_job.yml
+    uses: ./.github/workflows/linux_job_v2.yml
     name: ${{ matrix.build_name }}
     with:
       runner: ${{ matrix.validation_runner }}
@@ -45,8 +45,13 @@ jobs:
       script: |
         set -ex
 
+        # shellcheck disable=SC2086
+        CUDA_VERSION=$(python3 ../../test-infra/tools/scripts/get_stable_cuda_version.py --channel ${MATRIX_CHANNEL})
+        # shellcheck disable=SC2086
+        CUDA_VERSION_NODOT=$(echo ${CUDA_VERSION} | tr -d '.')
+
         # skip testing on cu126
-        if [[ ${MATRIX_DESIRED_CUDA} == "cu126" ]]; then
+        if [[ ${MATRIX_DESIRED_CUDA} != "cu${CUDA_VERSION_NODOT}" ]]; then
           exit 0
         fi