vmm: Support direct device assignment

With the VFIO crate, we can now support directly assigned PCI devices
into cloud-hypervisor guests.

We support assigning multiple host devices, through the --device command
line parameter. This parameter takes the host device sysfs path.

Fixes: #60

Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

vmm/src/config.rs

@@ -69,6 +69,7 @@ pub struct VmParams<'a> {
     pub pmem: Option<Vec<&'a str>>,
     pub serial: &'a str,
     pub console: &'a str,
+    pub devices: Option<Vec<&'a str>>,
 }
 
 fn parse_size(size: &str) -> Result<u64> {
@@ -379,6 +380,19 @@ impl<'a> ConsoleConfig<'a> {
     }
 }
 
+#[derive(Debug)]
+pub struct DeviceConfig<'a> {
+    pub path: &'a Path,
+}
+
+impl<'a> DeviceConfig<'a> {
+    pub fn parse(device: &'a str) -> Result<Self> {
+        Ok(DeviceConfig {
+            path: Path::new(device),
+        })
+    }
+}
+
 pub struct VmConfig<'a> {
     pub cpus: CpusConfig,
     pub memory: MemoryConfig<'a>,
@@ -391,6 +405,7 @@ pub struct VmConfig<'a> {
     pub pmem: Option<Vec<PmemConfig<'a>>>,
     pub serial: ConsoleConfig<'a>,
     pub console: ConsoleConfig<'a>,
+    pub devices: Option<Vec<DeviceConfig<'a>>>,
 }
 
 impl<'a> VmConfig<'a> {
@@ -437,6 +452,15 @@ impl<'a> VmConfig<'a> {
             return Err(Error::ParseTTYParam);
         }
 
+        let mut devices: Option<Vec<DeviceConfig>> = None;
+        if let Some(device_list) = &vm_params.devices {
+            let mut device_config_list = Vec::new();
+            for item in device_list.iter() {
+                device_config_list.push(DeviceConfig::parse(item)?);
+            }
+            devices = Some(device_config_list);
+        }
+
         Ok(VmConfig {
             cpus: CpusConfig::parse(vm_params.cpus)?,
             memory: MemoryConfig::parse(vm_params.memory)?,
@@ -449,6 +473,7 @@ impl<'a> VmConfig<'a> {
             pmem,
             serial,
             console,
+            devices,
         })
     }
 }