Files
TheRock/.github/workflows/multi_arch_build_portable_linux.yml
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

322 lines
14 KiB
YAML

# Copyright Advanced Micro Devices, Inc.
# SPDX-License-Identifier: MIT
# Multi-Arch Build - Sharded Pipeline for Linux
#
# This workflow builds TheRock in stages:
# 1. compiler-runtime (generic) - sysdeps, base, compiler, runtimes, profiler-core
# 2. runtime-tests (generic) - hip-tests, rocrtst (parallel to math-libs)
# 3. wsl-rocdxg (generic) - WSL-only ROCDXG bridge library (after compiler-runtime)
# 4. math-libs (per-arch) - BLAS, FFT, etc.
# 5. comm-libs (per-arch) - RCCL, rocshmem (parallel to math-libs)
# 6. storage-libs (generic) - hipfile (parallel to math-libs)
# 7. debug-tools (generic) - amd-dbgapi, rocr-debug-agent, rocgdb (parallel to math-libs)
# 8. dctools-core (generic) - RDC (parallel to math-libs)
# 9. profiler-apps (generic) - rocprofiler-systems (parallel to math-libs)
# 10. media-libs (generic) - sysdeps-amd-mesa, rocdecode, rocjpeg
name: Multi-Arch Build (Linux)
on:
workflow_call:
inputs:
artifact_group:
type: string
matrix_per_family_json:
type: string
description: "JSON array of {amdgpu_family, test-runs-on} objects for per-arch stages"
dist_amdgpu_families:
type: string
description: "Semicolon-separated list of all GPU families for dist targets"
build_variant_label:
type: string
build_variant_cmake_preset:
type: string
build_variant_suffix:
type: string
rocm_package_version:
type: string
prebuilt_stages:
type: string
default: ""
description: "Comma-separated build stages to skip (artifacts already copied by the orchestrator)"
build_runs_on:
type: string
default: "aws-linux-scale-rocm-prod-rc"
description: "Build runner label (selected by configure script with weighted distribution)"
release_type:
description: 'Release type: "ci", "dev", "nightly", or "prerelease"'
type: string
default: ci
repository:
description: "Repository to checkout. Defaults to github.repository."
type: string
default: ""
ref:
description: "Branch, tag, or SHA to checkout. Defaults to the triggering ref."
type: string
default: ""
external_repo_config:
description: "JSON config for external repo checkout (repository, ref, checkout_path, source_package, fetch_sources_args)"
type: string
default: ""
permissions:
contents: read
jobs:
# ==========================================================================
# STAGE: compiler-runtime (generic)
# ==========================================================================
compiler-runtime:
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'compiler-runtime') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: compiler-runtime
stage_display_name: "Stage - Compiler Runtime"
timeout_minutes: 480 # 8 hours (compiler is big)
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: runtime-tests (generic, parallel to math-libs)
# ==========================================================================
runtime-tests:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'runtime-tests') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: runtime-tests
stage_display_name: "Stage - Runtime Tests"
timeout_minutes: 120 # 2 hours
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: wsl-rocdxg (generic)
# ==========================================================================
wsl-rocdxg:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'wsl-rocdxg') }}
uses: ./.github/workflows/multi_arch_build_wsl_rocdxg_artifacts.yml
secrets: inherit
with:
stage_name: wsl-rocdxg
stage_display_name: "Stage - WSL ROCDXG"
timeout_minutes: 120 # 2 hours
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: math-libs (per-arch)
# ==========================================================================
math-libs:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'math-libs') }}
strategy:
# `fail-fast: false` lets all families complete even if one fails.
# This is useful for CI (to see all failures at once), but for releases
# this risks publishing incomplete packages. Release pipelines may want
# to either set `fail-fast: true` or add a validation step that blocks
# promotion unless all jobs succeeded.
fail-fast: false
matrix:
family_info: ${{ fromJSON(inputs.matrix_per_family_json) }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: math-libs
stage_display_name: "Stage - Math Libs (${{ matrix.family_info.amdgpu_family }})"
timeout_minutes: 480 # 8 hours
amdgpu_family: ${{ matrix.family_info.amdgpu_family }}
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: comm-libs (generic, single job parallel to math-libs)
# RCCL and rocshmem are both TARGET_NEUTRAL artifacts. RCCL is built for
# Instinct/CDNA targets only (https://github.com/ROCm/TheRock/issues/2130).
# Running as a single generic job avoids redundant per-arch builds.
# ==========================================================================
comm-libs:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'comm-libs') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: comm-libs
stage_display_name: "Stage - Comm Libs"
timeout_minutes: 240 # 4 hours
amdgpu_family: ""
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: storage-libs (generic, single job parallel to math-libs)
# hipFile is a TARGET_NEUTRAL artifact.
# ==========================================================================
storage-libs:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'storage-libs') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: storage-libs
stage_display_name: "Stage - Storage Libs"
timeout_minutes: 240 # 4 hours
amdgpu_family: ""
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: debug-tools (generic, parallel to math-libs)
# ==========================================================================
debug-tools:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'debug-tools') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: debug-tools
stage_display_name: "Stage - Debug Tools"
timeout_minutes: 180 # 3 hours
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: dctools-core (generic, parallel to math-libs)
# ==========================================================================
dctools-core:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'dctools-core') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: dctools-core
stage_display_name: "Stage - DC Tools Core"
timeout_minutes: 120 # 2 hours
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: profiler-apps (generic, parallel to math-libs)
# ==========================================================================
profiler-apps:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'profiler-apps') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: profiler-apps
stage_display_name: "Stage - Profiler Apps"
timeout_minutes: 180 # 3 hours
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write
# ==========================================================================
# STAGE: media-libs (generic)
# ==========================================================================
media-libs:
needs: compiler-runtime
if: ${{ !cancelled() && !failure() && !contains(inputs.prebuilt_stages, 'media-libs') }}
uses: ./.github/workflows/multi_arch_build_portable_linux_artifacts.yml
secrets: inherit
with:
stage_name: media-libs
stage_display_name: "Stage - Media Libs"
timeout_minutes: 180 # 3 hours
dist_amdgpu_families: ${{ inputs.dist_amdgpu_families }}
rocm_package_version: ${{ inputs.rocm_package_version }}
build_variant_cmake_preset: ${{ inputs.build_variant_cmake_preset }}
build_runs_on: ${{ inputs.build_runs_on }}
release_type: ${{ inputs.release_type }}
repository: ${{ inputs.repository }}
ref: ${{ inputs.ref }}
external_repo_config: ${{ inputs.external_repo_config }}
permissions:
contents: read
id-token: write