1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
|
#!/bin/bash
echo "🚀 Starting intelligent NVIDIA Docker setup..."
echo "📦 Updating NVIDIA driver to latest..."
sudo apt update
sudo ubuntu-drivers autoinstall
echo "🔄 Rebooting required after driver update. Run this script again after reboot."
# Check if reboot is needed
if [ -f /var/run/reboot-required ]; then
echo "⚠️ System reboot required. Please reboot and run this script again."
echo "After reboot, run: sudo reboot && ./setup-ubuntu.sh"
exit 0
fi
echo "🔍 Detecting NVIDIA driver version..."
DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader,nounits | head -1)
echo "Found NVIDIA driver: $DRIVER_VERSION"
echo "🔍 Determine compatible CUDA version based on driver..."
get_compatible_cuda_version() {
local driver_ver=$1
# Extract major version (e.g., 560 from 560.35.03)
local major_ver=$(echo $driver_ver | cut -d'.' -f1)
# NVIDIA Driver-CUDA Compatibility Matrix
if [ $major_ver -ge 565 ]; then
echo "12.9" # Driver 565+ supports CUDA 12.9
elif [ $major_ver -ge 560 ]; then
echo "12.6" # Driver 560+ supports CUDA 12.6 (your current case)
elif [ $major_ver -ge 555 ]; then
echo "12.5" # Driver 555+ supports CUDA 12.5
elif [ $major_ver -ge 550 ]; then
echo "12.4" # Driver 550+ supports CUDA 12.4
elif [ $major_ver -ge 535 ]; then
echo "12.2" # Driver 535+ supports CUDA 12.2
elif [ $major_ver -ge 525 ]; then
echo "12.0" # Driver 525+ supports CUDA 12.0
else
echo "11.8" # Fallback to CUDA 11.8
fi
}
COMPATIBLE_CUDA=$(get_compatible_cuda_version $DRIVER_VERSION)
echo "✅ Compatible CUDA version: $COMPATIBLE_CUDA"
echo "📦 Installing latest nvidia-container-toolkit..."
curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor --yes -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
&& curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list
sudo apt-get update
sudo apt-get install -y nvidia-container-toolkit
echo "📦 Configure Docker"
sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker
echo "🔍 Finding latest patch version for CUDA $COMPATIBLE_CUDA..."
AVAILABLE_VERSIONS=$(curl -s "https://registry.hub.docker.com/v2/repositories/nvidia/cuda/tags/?page_size=100" | jq -r '.results[].name' | grep -E "^${COMPATIBLE_CUDA}\.[0-9]+-base-ubuntu24\.04$" | head -1)
if [ -n "$AVAILABLE_VERSIONS" ]; then
# Extract full version (e.g., "12.9.1" from "12.9.1-base-ubuntu24.04")
FULL_CUDA_VERSION=$(echo "$AVAILABLE_VERSIONS" | cut -d'-' -f1)
echo "✅ Found CUDA version: $FULL_CUDA_VERSION"
else
echo "❌ No CUDA $COMPATIBLE_CUDA images found for Ubuntu 24.04, trying Ubuntu 22.04..."
AVAILABLE_VERSIONS=$(curl -s "https://registry.hub.docker.com/v2/repositories/nvidia/cuda/tags/?page_size=100" | jq -r '.results[].name' | grep -E "^${COMPATIBLE_CUDA}\.[0-9]+-base-ubuntu22\.04$" | head -1)
if [ -n "$AVAILABLE_VERSIONS" ]; then
FULL_CUDA_VERSION=$(echo "$AVAILABLE_VERSIONS" | cut -d'-' -f1)
UBUNTU_VERSION="22.04"
echo "✅ Found CUDA version: $FULL_CUDA_VERSION for Ubuntu 22.04"
else
echo "❌ No compatible CUDA images found"
exit 1
fi
fi
echo "🧪 Testing with automatically detected compatible CUDA version: $COMPATIBLE_CUDA..."
if docker run --rm --runtime=nvidia nvidia/cuda:${FULL_CUDA_VERSION}-base-ubuntu${UBUNTU_VERSION:-24.04} nvidia-smi; then
echo "✅ GPU support working with CUDA $FULL_CUDA_VERSION!"
# Update Dockerfile with compatible version
if [ -f "Dockerfile" ]; then
sed -i "s/ARG NVIDIA_CUDA_VERSION=.*/ARG NVIDIA_CUDA_VERSION=${FULL_CUDA_VERSION}/" Dockerfile
if [ "${UBUNTU_VERSION:-24.04}" = "22.04" ]; then
sed -i "s/ARG UBUNTU_VERSION=.*/ARG UBUNTU_VERSION=22.04/" Dockerfile
fi
echo "✅ Updated Dockerfile to use CUDA $FULL_CUDA_VERSION"
fi
else
echo "❌ GPU test failed with CUDA $FULL_CUDA_VERSION"
fi
|
