specs-go/config: Define RDMA cgroup

Linux kernel 4.11 adds support for RDMA cgroup resource controller.
This allows limiting maximum number of open hca_handle and maximum number of
hca_objects which can be created by processes.

config-linux: Add documentation for Linux RDMA cgroup

Add documentation, example and link to kernel documentation for
Linux RDMA cgroup.

additionalProperties is defined for the JSON Schema draft-04 in [1]
with clearer documentation in draft-07 [2]. It is supportd by
gojsonschema since xeipuuv/gojsonschema@0572d9d (added
additionalProperties with inner schema, 2013-06-21).

[1]: https://tools.ietf.org/html/draft-fge-json-schema-validation-00#section-5.4.4
[2]: https://tools.ietf.org/html/draft-handrews-json-schema-validation-00#section-6.5.6

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: W. Trevor King <wking@tremily.us>

Author: paravmellanox

config-linux.md

@@ -169,7 +169,7 @@ In addition to any devices configured with this setting, the runtime MUST also s
 ## <a name="configLinuxControlGroups" />Control groups
 
 Also known as cgroups, they are used to restrict resource usage for a container and handle device access.
-cgroups provide controls (through controllers) to restrict cpu, memory, IO, pids and network for the container.
+cgroups provide controls (through controllers) to restrict cpu, memory, IO, pids, network and RDMA resources for the container.
 For more information, see the [kernel cgroups documentation][cgroup-v1].
 
 ### <a name="configLinuxCgroupsPath" />Cgroups Path
@@ -455,6 +455,36 @@ The following parameters can be specified to set up the controller:
    }
 ```
 
+### <a name="configLinuxRDMA" />RDMA
+
+**`rdma`** (object, OPTIONAL) represents the cgroup subsystem `rdma`.
+For more information, see the kernel cgroups documentation about [rdma][cgroup-v1-rdma].
+
+The name of the device to limit is the entry key.
+Entry values are objects with the following properties:
+
+* **`hcaHandles`** *(uint32, OPTIONAL)* - specifies the maximum number of hca_handles in the cgroup
+* **`hcaObjects`** *(uint32, OPTIONAL)* - specifies the maximum number of hca_objects in the cgroup
+
+You MUST specify at least one of the `hcaHandles` or `hcaObjects` in a given entry, and MAY specify both.
+
+#### Example
+
+```json
+"rdma": {
+    "mlx5_1": {
+        "hcaHandles": 3,
+        "hcaObjects": 10000
+    },
+    "mlx4_0": {
+        "hcaObjects": 1000
+    },
+    "rxe3": {
+        "hcaObjects": 10000
+    }
+}
+```
+
 ## <a name="configLinuxIntelRdt" />IntelRdt
 
 **`intelRdt`** (object, OPTIONAL) represents the [Intel Resource Director Technology][intel-rdt-cat-kernel-interface].
@@ -647,6 +677,7 @@ The following parameters can be specified to set up seccomp:
 [cgroup-v1-net-cls]: https://www.kernel.org/doc/Documentation/cgroup-v1/net_cls.txt
 [cgroup-v1-net-prio]: https://www.kernel.org/doc/Documentation/cgroup-v1/net_prio.txt
 [cgroup-v1-pids]: https://www.kernel.org/doc/Documentation/cgroup-v1/pids.txt
+[cgroup-v1-rdma]: https://www.kernel.org/doc/Documentation/cgroup-v1/rdma.txt
 [cgroup-v2]: https://www.kernel.org/doc/Documentation/cgroup-v2.txt
 [devices]: https://www.kernel.org/doc/Documentation/admin-guide/devices.txt
 [devpts]: https://www.kernel.org/doc/Documentation/filesystems/devpts.txt

schema/config-linux.json

@@ -175,6 +175,12 @@
                                 }
                             }
                         }
+                    },
+                    "rdma": {
+                        "type": "object",
+                        "additionalProperties": {
+                            "$ref": "defs-linux.json#/definitions/Rdma"
+                        }
                     }
                 }
             },

schema/defs-linux.json

@@ -240,6 +240,17 @@
                 "priority"
             ]
         },
+        "Rdma": {
+            "type": "object",
+            "properties": {
+                "hcaHandles": {
+                    "$ref": "defs.json#/definitions/uint32"
+                },
+                "hcaObjects": {
+                    "$ref": "defs.json#/definitions/uint32"
+                }
+            }
+        },
         "NamespaceType": {
             "type": "string",
             "enum": [

schema/test/config/bad/linux-rdma.json

@@ -0,0 +1,15 @@
+{
+    "ociVersion": "1.0.0",
+    "root": {
+        "path": "rootfs"
+    },
+    "linux": {
+        "resources": {
+            "rdma": {
+                "mlx5_1": {
+                    "hcaHandles": "not a uint32"
+                }
+            }
+        }
+    }
+}

schema/test/config/good/linux-rdma.json

@@ -0,0 +1,22 @@
+{
+    "ociVersion": "1.0.0",
+    "root": {
+        "path": "rootfs"
+    },
+    "linux": {
+        "resources": {
+            "rdma": {
+                "mlx5_1": {
+                    "hcaHandles": 3,
+                    "hcaObjects": 10000
+                },
+                "mlx4_0": {
+                    "hcaObjects": 1000
+                },
+                "rxe3": {
+                    "hcaObjects": 10000
+                }
+            }
+        }
+    }
+}

specs-go/config.go

@@ -320,6 +320,14 @@ type LinuxNetwork struct {
 	Priorities []LinuxInterfacePriority `json:"priorities,omitempty"`
 }
 
+// LinuxRdma for Linux cgroup 'rdma' resource management (Linux 4.11)
+type LinuxRdma struct {
+	// Maximum number of HCA handles that can be opened. Default is "no limit".
+	HcaHandles *uint32 `json:"hcaHandles,omitempty"`
+	// Maximum number of HCA objects that can be created. Default is "no limit".
+	HcaObjects *uint32 `json:"hcaObjects,omitempty"`
+}
+
 // LinuxResources has container runtime resource constraints
 type LinuxResources struct {
 	// Devices configures the device whitelist.
@@ -336,6 +344,10 @@ type LinuxResources struct {
 	HugepageLimits []LinuxHugepageLimit `json:"hugepageLimits,omitempty"`
 	// Network restriction configuration
 	Network *LinuxNetwork `json:"network,omitempty"`
+	// Rdma resource restriction configuration.
+	// Limits are a set of key value pairs that define RDMA resource limits,
+	// where the key is device name and value is resource limits.
+	Rdma map[string]LinuxRdma `json:"rdma,omitempty"`
 }
 
 // LinuxDevice represents the mknod information for a Linux special device file