[AMD] Add ROCm 10 (gfx942 / gfx950) release images (#36434)
This commit is contained in:
@@ -11,12 +11,13 @@ on:
|
||||
workflow_dispatch:
|
||||
inputs:
|
||||
rocm_version:
|
||||
description: 'ROCm image version ("all" runs every flavor, as the nightly schedule does)'
|
||||
description: 'ROCm image version ("all" runs the scheduled ROCm 7.2 matrix; select rocm10 explicitly)'
|
||||
required: false
|
||||
type: choice
|
||||
default: 'all'
|
||||
options:
|
||||
- 'all'
|
||||
- rocm10
|
||||
- rocm724
|
||||
- rocm720
|
||||
aiter_ref:
|
||||
@@ -110,7 +111,7 @@ on:
|
||||
workflow_call:
|
||||
inputs:
|
||||
rocm_version:
|
||||
description: 'ROCm image version ("all" runs every flavor)'
|
||||
description: 'ROCm image version ("all" runs the scheduled ROCm 7.2 matrix; pass rocm10 explicitly)'
|
||||
required: false
|
||||
type: string
|
||||
# A single flavor, unlike the schedule and the dispatch form: a caller
|
||||
@@ -444,7 +445,7 @@ jobs:
|
||||
|
||||
- name: Install dependencies
|
||||
run: bash scripts/ci/amd/amd_ci_install_dependency.sh
|
||||
- name: Performance Test ROCm 7.2 (2-GPU Text Models)
|
||||
- name: Performance Test (2-GPU Text Models)
|
||||
timeout-minutes: 120
|
||||
run: |
|
||||
> github_summary.md # Clear summary file
|
||||
@@ -1008,7 +1009,7 @@ jobs:
|
||||
echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
|
||||
exit ${TEST_EXIT_CODE:-0}
|
||||
|
||||
- name: Performance Test MI35x ROCm 7.2 (8-GPU DeepSeek-V3.2 Basic)
|
||||
- name: Performance Test MI35x (8-GPU DeepSeek-V3.2 Basic)
|
||||
timeout-minutes: 150
|
||||
continue-on-error: true
|
||||
run: |
|
||||
|
||||
@@ -42,6 +42,7 @@ on:
|
||||
type: choice
|
||||
default: rocm724
|
||||
options:
|
||||
- rocm10
|
||||
- rocm724
|
||||
- rocm720
|
||||
- rocm700
|
||||
|
||||
@@ -22,6 +22,7 @@ on:
|
||||
type: choice
|
||||
default: rocm724
|
||||
options:
|
||||
- rocm10
|
||||
- rocm724
|
||||
- rocm720
|
||||
target_stage_select:
|
||||
|
||||
@@ -0,0 +1,118 @@
|
||||
name: Release Docker Images Nightly ROCm 10 (AMD)
|
||||
on:
|
||||
workflow_dispatch:
|
||||
inputs:
|
||||
job_select:
|
||||
description: 'Select which release job to run'
|
||||
required: false
|
||||
type: choice
|
||||
default: 'all'
|
||||
options:
|
||||
- 'all'
|
||||
- publish
|
||||
gpu_arch:
|
||||
description: 'Select which ROCm 10 GPU arch to build'
|
||||
required: false
|
||||
type: choice
|
||||
default: 'all'
|
||||
options:
|
||||
- 'all'
|
||||
- gfx942-rocm1000
|
||||
- gfx950-rocm1000
|
||||
schedule:
|
||||
- cron: '0 12 * * *'
|
||||
|
||||
concurrency:
|
||||
group: ${{ github.workflow }}-${{ github.event.number || github.sha }}
|
||||
cancel-in-progress: true
|
||||
|
||||
jobs:
|
||||
publish:
|
||||
if: github.repository == 'sgl-project/sglang' && (github.event_name != 'workflow_dispatch' || inputs.job_select == 'all' || inputs.job_select == 'publish')
|
||||
runs-on: amd-docker-scale
|
||||
environment: 'prod'
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
gpu_arch: ${{ fromJson((github.event_name == 'workflow_dispatch' && inputs.gpu_arch != 'all' && inputs.gpu_arch != '') && format('["{0}"]', inputs.gpu_arch) || '["gfx942-rocm1000", "gfx950-rocm1000"]') }}
|
||||
build_type: ['all']
|
||||
steps:
|
||||
- name: Checkout repository
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 0
|
||||
|
||||
- name: Set up Python
|
||||
uses: actions/setup-python@v5
|
||||
with:
|
||||
python-version: '3.10'
|
||||
|
||||
- name: Set date
|
||||
run: echo "DATE=$(date +%Y%m%d)" >> "$GITHUB_ENV"
|
||||
|
||||
- name: Get version from latest tag
|
||||
id: version
|
||||
run: |
|
||||
VERSION=$(python3 scripts/release/get_version_tag.py --tag-only | sed 's/^v//')
|
||||
|
||||
if [ -z "$VERSION" ]; then
|
||||
echo "::error::Could not determine version from git tags"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
COMMIT_HASH=$(git rev-parse --short HEAD)
|
||||
PRETEND_VERSION="${VERSION}.dev${DATE}+g${COMMIT_HASH}"
|
||||
|
||||
echo "version=${VERSION}" >> "$GITHUB_OUTPUT"
|
||||
echo "pretend_version=${PRETEND_VERSION}" >> "$GITHUB_OUTPUT"
|
||||
echo "Detected version: ${VERSION}"
|
||||
echo "Pretend version for pip: ${PRETEND_VERSION}"
|
||||
|
||||
- name: Login to Docker Hub (AMD)
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_AMD_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_AMD_TOKEN }}
|
||||
|
||||
- name: Build and push to rocm/sgl-dev
|
||||
env:
|
||||
BUILD_TYPE: ${{ matrix.build_type }}
|
||||
GPU_ARCH: ${{ matrix.gpu_arch }}
|
||||
PRETEND_VERSION: ${{ steps.version.outputs.pretend_version }}
|
||||
VERSION: ${{ steps.version.outputs.version }}
|
||||
run: |
|
||||
case "$GPU_ARCH" in
|
||||
gfx942-rocm1000) rocm_tag="rocm10-mi30x" ;;
|
||||
gfx950-rocm1000) rocm_tag="rocm10-mi35x" ;;
|
||||
*)
|
||||
echo "Unsupported GPU arch: $GPU_ARCH"
|
||||
exit 1
|
||||
;;
|
||||
esac
|
||||
|
||||
image_tag="v${VERSION}-${rocm_tag}-${DATE}"
|
||||
echo "IMAGE_TAG=${image_tag}" >> "$GITHUB_ENV"
|
||||
|
||||
docker build . \
|
||||
-f docker/rocm.Dockerfile \
|
||||
--build-arg "SGL_BRANCH=${GITHUB_SHA}" \
|
||||
--build-arg "BUILD_TYPE=${BUILD_TYPE}" \
|
||||
--build-arg "GPU_ARCH=${GPU_ARCH}" \
|
||||
--build-arg ENABLE_MORI=1 \
|
||||
--build-arg ENABLE_NIXL=1 \
|
||||
--build-arg "SETUPTOOLS_SCM_PRETEND_VERSION=${PRETEND_VERSION}" \
|
||||
--build-arg UBUNTU_MIRROR=https://archive.ubuntu.com \
|
||||
-t "rocm/sgl-dev:${image_tag}" \
|
||||
--no-cache
|
||||
docker push "rocm/sgl-dev:${image_tag}"
|
||||
|
||||
- name: Login to Docker Hub (lmsys)
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_TOKEN }}
|
||||
|
||||
- name: Push to lmsysorg/sglang-rocm
|
||||
run: |
|
||||
docker tag "rocm/sgl-dev:${IMAGE_TAG}" "lmsysorg/sglang-rocm:${IMAGE_TAG}"
|
||||
docker push "lmsysorg/sglang-rocm:${IMAGE_TAG}"
|
||||
Reference in New Issue
Block a user