IP Feature Subgraph Ordering. More...

Include dependency graph for ip_feature_registration.c:

Macros
#define	foreach_af_cast

Functions
static int	comma_split (u8 s, u8 a, u8 *b)

clib_error_t *	vnet_feature_arc_init (vlib_main_t vm, vnet_config_main_t vcm, char *feature_start_nodes, int num_feature_start_nodes, vnet_ip_feature_registration_t first_reg, char ***in_feature_nodes)
	Initialize a feature graph arc. More...

static clib_error_t *	show_ip_features_command_fn (vlib_main_t vm, unformat_input_t input, vlib_cli_command_t *cmd)
	Display the set of available ip features. More...

void	ip_interface_features_show (vlib_main_t vm, const char pname, ip_config_main_t *cm, u32 sw_if_index)
	Display the set of IP features configured on a specific interface. More...

static clib_error_t *	show_ip_interface_features_command_fn (vlib_main_t vm, unformat_input_t input, vlib_cli_command_t *cmd)

Variables
static const char *	vnet_cast_names [] = VNET_CAST_NAMES

static vlib_cli_command_t	show_ip_features_command
	(constructor) VLIB_CLI_COMMAND (show_ip_features_command) More...

static vlib_cli_command_t	show_ip_interface_features_command
	(constructor) VLIB_CLI_COMMAND (show_ip_interface_features_command) More...

Detailed Description

IP Feature Subgraph Ordering.

Dynamically compute IP feature subgraph ordering by performing a topological sort across a set of "feature A before feature B" and "feature C after feature B" constraints.

Use the topological sort result to set up vnet_config_main_t's for use at runtime.

Feature subgraph arcs are simple enough. They start at specific fixed nodes, and end at specific fixed nodes. In between, a per-interface current feature configuration dictates which additional nodes each packet visits. Each so-called feature node can [of course] drop any specific packet.

See ip4_forward.c, ip6_forward.c in this directory to see the current rx-unicast, rx-multicast, and tx feature subgraph arc definitions.

Let's say that we wish to add a new feature to the ip4 unicast feature subgraph arc, which needs to run before ip4-lookup. In either base code or a plugin,

 #include <vnet/ip/ip_feature_registration.h>

and add the new feature as shown:

 VNET_IP4_UNICAST_FEATURE_INIT (ip4_lookup, static) =
 {
   .node_name = "my-ip4-unicast-feature",
   .runs_before = ORDER_CONSTRAINTS {"ip4-lookup", 0}
   .feature_index = &my_feature_index,
 };

Here's the standard coding pattern to enable / disable my-ip4-unicast-feature on an interface:

 ip4_main_t *im = &ip4_main;
 ip_lookup_main_t *lm = &im->lookup_main;
 ip_config_main_t *rx_cm =
     &lm->feature_config_mains[VNET_IP_RX_UNICAST_FEAT];

 sw_if_index = <interface-handle>
 ci = rx_cm->config_index_by_sw_if_index[sw_if_index];
 ci = (is_add
       ? vnet_config_add_feature
       : vnet_config_del_feature)
   (vm, &rx_cm->config_main,
    ci,
    my_feature_index,
    0 / * &config struct if feature uses private config data * /,
    0 / * sizeof config struct if feature uses private config data * /);
 rx_cm->config_index_by_sw_if_index[sw_if_index] = ci;

For tx features, add this line after setting

 tx_cm->config_index_by_sw_if_index = ci.

This maintains a per-interface "at least one TX feature enabled" bitmap:

 vnet_config_update_tx_feature_count (lm, tx_cm, sw_if_index, is_add);

Here's how to obtain the correct next node index in packet processing code, aka in the implementation of my-ip4-unicast-feature:

 ip_lookup_main_t * lm = sm->ip4_lookup_main;
 ip_config_main_t * cm = &lm->feature_config_mains[VNET_IP_RX_UNICAST_FEAT];

 Call vnet_get_config_data to set next0, and to advance
 b0->current_config_index:

 config_data0 = vnet_get_config_data (&cm->config_main,
                                      &b0->current_config_index,
                                      &next0,
                                      0 / * sizeof config data * /);

Nodes are free to drop or otherwise redirect packets. Packets which "pass" should be enqueued via the next0 arc computed by vnet_get_config_data.

Definition in file ip_feature_registration.c.

Macro Definition Documentation

#define foreach_af_cast

Value:

_(4, VNET_IP_RX_UNICAST_FEAT, "ip4 unicast")            \
_(4, VNET_IP_RX_MULTICAST_FEAT, "ip4 multicast")        \
_(4, VNET_IP_TX_FEAT, "ip4 output")                     \
_(6, VNET_IP_RX_UNICAST_FEAT, "ip6 unicast")            \
_(6, VNET_IP_RX_MULTICAST_FEAT, "ip6 multicast")        \
_(6, VNET_IP_TX_FEAT, "ip6 output")

Definition at line 326 of file ip_feature_registration.c.

Function Documentation

static int comma_split	(	u8 *	s,
		u8 **	a,
		u8 **	b
	)

static

Definition at line 117 of file ip_feature_registration.c.

Here is the caller graph for this function:

void ip_interface_features_show	(	vlib_main_t *	vm,
		const char *	pname,
		ip_config_main_t *	cm,
		u32	sw_if_index
	)

Display the set of IP features configured on a specific interface.

Definition at line 415 of file ip_feature_registration.c.

Here is the call graph for this function:

Here is the caller graph for this function:

static clib_error_t* show_ip_features_command_fn	(	vlib_main_t *	vm,
		unformat_input_t *	input,
		vlib_cli_command_t *	cmd
	)

static

Display the set of available ip features.

Useful for verifying that expected features are present

Definition at line 339 of file ip_feature_registration.c.

Here is the call graph for this function:

static clib_error_t* show_ip_interface_features_command_fn	(	vlib_main_t *	vm,
		unformat_input_t *	input,
		vlib_cli_command_t *	cmd
	)

static

Definition at line 465 of file ip_feature_registration.c.

Here is the call graph for this function:

clib_error_t* vnet_feature_arc_init	(	vlib_main_t *	vm,
		vnet_config_main_t *	vcm,
		char **	feature_start_nodes,
		int	num_feature_start_nodes,
		vnet_ip_feature_registration_t *	first_reg,
		char ***	in_feature_nodes
	)

Initialize a feature graph arc.

Parameters

	vm	vlib main structure pointer
	vcm	vnet config main structure pointer
	feature_start_nodes	names of start-nodes which use this feature graph arc
	num_feature_start_nodes	number of start-nodes
	first_reg	first element in [an attribute((constructor)) function built, or otherwise created] singly-linked list of feature registrations
[out]	in_feature_nodes	returned vector of topologically-sorted feature node names, for use in show commands

Returns: 0 on success, otherwise an error message. Errors are fatal since they invariably involve mistyped node-names, or genuinely missing node-names

Definition at line 151 of file ip_feature_registration.c.

Here is the call graph for this function:

Here is the caller graph for this function:

Variable Documentation

vlib_cli_command_t show_ip_features_command

static

Initial value:

= {
  .path = "show ip features",
  .short_help = "show ip features",
  .function = show_ip_features_command_fn,
}

(constructor) VLIB_CLI_COMMAND (show_ip_features_command)

Definition at line 404 of file ip_feature_registration.c.

vlib_cli_command_t show_ip_interface_features_command

static

Initial value:

= {
  .path = "show ip interface features",
  .short_help = "show ip interface features <interface>",
  .function = show_ip_interface_features_command_fn,
}

(constructor) VLIB_CLI_COMMAND (show_ip_interface_features_command)

Definition at line 521 of file ip_feature_registration.c.

const char* vnet_cast_names[] = VNET_CAST_NAMES

static

Definition at line 114 of file ip_feature_registration.c.

Macros

Functions

Variables

Detailed Description

Macro Definition Documentation

Function Documentation

Variable Documentation