Allow running ramalama without a GPU

docs/ramalama-bench.1.md

@@ -36,6 +36,11 @@ write, and m for mknod(2).
 
 Example: --device=/dev/dri/renderD128:/dev/xvdc:rwm
 
+The device specification is passed directly to the underlying container engine.  See documentation of the supported container engine for more information.
+
+Pass '--device=none' explicitly add no device to the container, eg for
+running a CPU-only performance comparison.
+
 #### **--env**=
 
 Set environment variables inside of the container.

docs/ramalama-run.1.md

@@ -49,6 +49,9 @@ Example: --device=/dev/dri/renderD128:/dev/xvdc:rwm
 
 The device specification is passed directly to the underlying container engine.  See documentation of the supported container engine for more information.
 
+Pass '--device=none' explicitly add no device to the container, eg for
+running a CPU-only performance comparison.
+
 #### **--env**=
 
 Set environment variables inside of the container.

docs/ramalama-serve.1.md

@@ -75,6 +75,9 @@ Example: --device=/dev/dri/renderD128:/dev/xvdc:rwm
 
 The device specification is passed directly to the underlying container engine. See documentation of the supported container engine for more information.
 
+Pass '--device=none' explicitly add no device to the container, eg for
+running a CPU-only performance comparison.
+
 #### **--dri**=*on* | *off*
 Enable or disable mounting `/dev/dri` into the container when running with `--api=llama-stack` (enabled by default). Use to prevent access to the host device when not required, or avoid errors in environments where `/dev/dri` is not available.
 

ramalama/cli.py

@@ -775,7 +775,11 @@ def runtime_options(parser, command):
             "-d", "--detach", action="store_true", dest="detach", help="run the container in detached mode"
         )
     parser.add_argument(
-        "--device", dest="device", action='append', type=str, help="device to leak in to the running container"
+        "--device",
+        dest="device",
+        action='append',
+        type=str,
+        help="device to leak in to the running container (or 'none' to pass no device)",
     )
     parser.add_argument(
         "--env",

ramalama/engine.py

@@ -118,6 +118,10 @@ def add_port_option(self):
             self.exec_args += ["-p", f"{host}{self.args.port}:{self.args.port}"]
 
     def add_device_options(self):
+        request_no_device = getattr(self.args, "device", None) == ['none']
+        if request_no_device:
+            return
+
         if getattr(self.args, "device", None):
             for device_arg in self.args.device:
                 self.exec_args += ["--device", device_arg]