feat: nvidia driver extension 575.57.08

This patch deprecates the NVIDIA toolkit extension and introduces a new
nvidia-driver extension (in production/lts versions and open
source/proprietary flavors). The NVIDIA container toolkit must be
installed independently, via a future Talos extension, the NVIDIA GPU
Operator, or by the cluster administator.

The extension depends on the new glibc extension (siderolabs#473) and participates
in its filesystem subroot by installing all the NVIDIA components in it.

Finally, the extension runs a service that will bind mount this glibc
subroot at `/run/nvidia/driver` and run the `nvidia-persistenced`
daemon.

This careful setup allows the NVIDIA GPU Operator to utilize this
extension as if it were a traditional NVIDIA driver container.

Signed-off-by: Jean-Francois Roy <[email protected]>

Author: jfroy

.kres.yaml

@@ -35,12 +35,12 @@ spec:
     - nebula
     - nfsd
     - nut-client
-    - nvidia-container-toolkit-lts
-    - nvidia-container-toolkit-production
     - nvidia-fabricmanager-lts
     - nvidia-fabricmanager-production
-    - nvidia-open-gpu-kernel-modules-lts
-    - nvidia-open-gpu-kernel-modules-production
+    - nvidia-driver-lts
+    - nvidia-driver-production
+    - nvidia-driver-proprietary-lts
+    - nvidia-driver-proprietary-production
     - nvme-cli
     - panfrost
     - qemu-guest-agent

Makefile

@@ -90,12 +90,12 @@ TARGETS += metal-agent
 TARGETS += nebula
 TARGETS += nfsd
 TARGETS += nut-client
-TARGETS += nvidia-container-toolkit-lts
-TARGETS += nvidia-container-toolkit-production
 TARGETS += nvidia-fabricmanager-lts
 TARGETS += nvidia-fabricmanager-production
-TARGETS += nvidia-open-gpu-kernel-modules-lts
-TARGETS += nvidia-open-gpu-kernel-modules-production
+TARGETS += nvidia-driver-lts
+TARGETS += nvidia-driver-production
+TARGETS += nvidia-driver-proprietary-lts
+TARGETS += nvidia-driver-proprietary-production
 TARGETS += nvme-cli
 TARGETS += panfrost
 TARGETS += qemu-guest-agent

go.work

@@ -2,6 +2,5 @@ go 1.23.0
 
 use (
 	./examples/hello-world-service/src
-	./nvidia-gpu/nvidia-container-toolkit/nvidia-container-runtime-wrapper
-	./nvidia-gpu/nvidia-container-toolkit/nvidia-persistenced-wrapper
+	./nvidia-gpu/nvidia-driver/service
 )