[AMD][DI][CI] 2/N Add DSV4 DP8/EP8 and MTP MI355X 1P1D nightly recipes (#29784)

Co-authored-by: bingxche <bingxche@amd.com>
This commit is contained in:
Zhaoyi Li
2026-07-01 23:46:48 +08:00
committed by GitHub
co-authored by bingxche
parent 2a6e5c60fe
commit 9bb7de9258
15 changed files with 970 additions and 3 deletions
@@ -142,8 +142,40 @@ jobs:
# Manual dispatch input wins; otherwise use the latest image resolved
# by the setup job; otherwise the launcher falls back to the recipe default.
IMAGE_OVERRIDE: ${{ inputs.image != '' && inputs.image || needs.setup.outputs.image }}
# Keep the scheduler off mia1-p01-g20: its ionic RDMA driver ABI
# mismatches the container, so MORI reports "no active RDMA device"
# and the disagg server dies on init. Remove once the node is fixed.
SLURM_EXCLUDE: mia1-p01-g20
run: bash scripts/ci/slurm/launch_mi355x.sh
- name: Pack logs
if: always()
run: |
LOG_DIR="$HOME/.mi355x_ci/${MATRIX_CONFIG_NAME}"
LOG_BUNDLE="${GITHUB_WORKSPACE}/${RESULT_FILENAME}_logs.tar.gz"
if [ ! -d "$LOG_DIR" ]; then
echo "WARN: log directory not found: $LOG_DIR"
exit 0
fi
mapfile -d '' LOG_FILES < <(
find "$LOG_DIR" -maxdepth 1 -type f \
\( -name '*.log' -o -name 'server_exit_*' -o -name 'bench_exit' \) \
-printf '%P\0' | sort -z
)
if [ "${#LOG_FILES[@]}" -eq 0 ]; then
echo "WARN: no log files found in $LOG_DIR"
exit 0
fi
STAGING_DIR="$(mktemp -d)"
trap 'rm -rf "$STAGING_DIR"' EXIT
for log_file in "${LOG_FILES[@]}"; do
cp -a "$LOG_DIR/$log_file" "$STAGING_DIR/$log_file" || true
done
tar czf "$LOG_BUNDLE" -C "$STAGING_DIR" .
echo "Packed ${#LOG_FILES[@]} log file(s) -> $LOG_BUNDLE"
- name: Process results
if: always()
run: |
@@ -164,7 +196,9 @@ jobs:
uses: actions/upload-artifact@v4
with:
name: mi355x-${{ matrix.config.name }}-${{ github.run_id }}
path: ${{ github.workspace }}/*.json
path: |
${{ github.workspace }}/*.json
${{ github.workspace }}/*_logs.tar.gz
retention-days: 30
if-no-files-found: warn