1768bb58dd
## ISO build consolidation - Remove separate debug/prod split: overlay-debug/, build-debug.sh, mkimg.bee_debug.sh, genapkovl-bee_debug.sh all deleted - Single overlay: iso/overlay/ (was overlay-debug content) - Single build script: build.sh (SSH, TUI, NVIDIA, vendor tools, bee-release) - Single mkimage profile: bee (with dropbear, dialog, strace, gcompat, etc.) ## NVIDIA fixes - Modules now stored at /usr/local/lib/nvidia/ instead of /lib/modules/<kver>/extra/nvidia/ — modloop squashfs mounts over that path at boot making overlay content there inaccessible - bee-nvidia init: load via insmod (absolute path), not modprobe - bee-nvidia init: create libnvidia-ml.so.1/libcuda.so.1 symlinks in /usr/lib/ - build-nvidia-module.sh: always install linux-lts-dev (not conditional) — stale 6.6.x headers caused wrong-kernel modules that never loaded at runtime - build-nvidia-module.sh: create soname symlinks in cache - KERNEL_VERSION in VERSIONS updated 6.6 → 6.12 - gcompat added to ISO packages (nvidia-smi is a glibc binary on musl Alpine) ## Service ordering - bee-audit: add `after bee-nvidia` so NVIDIA enrichment always succeeds ## New tooling - iso/builder/smoketest.sh: SSH smoke test for post-boot ISO validation - iso/builder/build-gpu-burn.sh: builds gpu_burn vendor binary (CUDA 12.8+) - vendor/gpu_burn included automatically if placed in iso/vendor/ Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
83 lines
2.7 KiB
Bash
83 lines
2.7 KiB
Bash
#!/bin/sh
|
|
# build-gpu-burn.sh — build gpu_burn stress tool and output static-ish binary to DIST_DIR
|
|
#
|
|
# gpu_burn requires nvcc (CUDA toolkit). This script downloads a minimal CUDA toolkit
|
|
# runfile, extracts only nvcc + headers, builds gpu_burn, then cleans up the toolkit.
|
|
#
|
|
# Output: $DIST_DIR/gpu_burn (ready to copy into ISO vendor/)
|
|
#
|
|
# Usage: sh build-gpu-burn.sh <dist-dir>
|
|
|
|
set -e
|
|
|
|
DIST_DIR="$1"
|
|
[ -n "$DIST_DIR" ] || { echo "usage: $0 <dist-dir>"; exit 1; }
|
|
mkdir -p "$DIST_DIR"
|
|
|
|
OUTPUT="$DIST_DIR/gpu_burn"
|
|
if [ -f "$OUTPUT" ] && [ -s "$OUTPUT" ]; then
|
|
echo "=== gpu_burn cached: $OUTPUT ==="
|
|
exit 0
|
|
fi
|
|
|
|
# CUDA toolkit version for building — only nvcc + headers needed, not the full runtime.
|
|
# Must be <= max CUDA version supported by the NVIDIA driver in VERSIONS.
|
|
# Driver 590.48.01 supports up to CUDA 13.1; use 12.6 (stable, widely tested).
|
|
CUDA_VERSION="12.8.1"
|
|
CUDA_BUILD="570.124.06"
|
|
CUDA_RUN="/var/tmp/cuda-${CUDA_VERSION}.run"
|
|
CUDA_DIR="/var/tmp/cuda-toolkit-${CUDA_VERSION}"
|
|
|
|
echo "=== building gpu_burn (CUDA ${CUDA_VERSION}) ==="
|
|
|
|
# Install build dependencies
|
|
apk add --quiet gcc g++ make git wget libxml2
|
|
|
|
# Download CUDA toolkit runfile if not cached
|
|
if [ ! -s "$CUDA_RUN" ]; then
|
|
echo "=== downloading CUDA ${CUDA_VERSION} toolkit ==="
|
|
wget -q --show-progress -O "$CUDA_RUN" \
|
|
"https://developer.download.nvidia.com/compute/cuda/${CUDA_VERSION}/local_installers/cuda_${CUDA_VERSION}_${CUDA_BUILD}_linux.run"
|
|
fi
|
|
|
|
# Extract toolkit (nvcc + headers only — skip driver, samples, docs to save time/space)
|
|
if [ ! -d "$CUDA_DIR/bin/nvcc" ] && [ ! -f "$CUDA_DIR/bin/nvcc" ]; then
|
|
echo "=== extracting CUDA toolkit ==="
|
|
rm -rf "$CUDA_DIR"
|
|
sh "$CUDA_RUN" \
|
|
--silent \
|
|
--toolkit \
|
|
--toolkitpath="$CUDA_DIR" \
|
|
--no-opengl-libs \
|
|
--no-drm \
|
|
--override 2>&1 | tail -5
|
|
fi
|
|
|
|
NVCC="$CUDA_DIR/bin/nvcc"
|
|
[ -f "$NVCC" ] || { echo "ERROR: nvcc not found after extraction: $NVCC"; exit 1; }
|
|
echo "nvcc: $("$NVCC" --version | head -1)"
|
|
|
|
# Clone gpu_burn source
|
|
GPU_BURN_DIR="/var/tmp/gpu-burn-src"
|
|
if [ ! -d "$GPU_BURN_DIR/.git" ]; then
|
|
echo "=== cloning gpu-burn ==="
|
|
git clone --depth=1 https://github.com/wilicc/gpu-burn.git "$GPU_BURN_DIR"
|
|
else
|
|
echo "=== gpu-burn source already cloned ==="
|
|
fi
|
|
|
|
# Build
|
|
echo "=== building gpu_burn ==="
|
|
cd "$GPU_BURN_DIR"
|
|
make clean 2>/dev/null || true
|
|
CUDA_PATH="$CUDA_DIR" make 2>&1
|
|
|
|
[ -f "$GPU_BURN_DIR/gpu_burn" ] || { echo "ERROR: gpu_burn binary not produced"; exit 1; }
|
|
|
|
cp "$GPU_BURN_DIR/gpu_burn" "$OUTPUT"
|
|
cp "$GPU_BURN_DIR/compare.ptx" "$(dirname "$OUTPUT")/compare.ptx" 2>/dev/null || true
|
|
|
|
echo "=== gpu_burn build complete ==="
|
|
ls -lh "$OUTPUT"
|
|
echo "NOTE: compare.ptx must be present in same dir as gpu_burn at runtime"
|