NETISR(9) BSD Kernel Developer's Manual NETISR(9)NAME
netisr — Kernel network dispatch service
SYNOPSIS
#include <net/netisr.h>
void
netisr_register(const struct netisr_handler *nhp);
void
netisr_unregister(const struct netisr_handler *nhp);
int
netisr_dispatch(u_int proto, struct mbuf *m);
int
netisr_dispatch_src(u_int proto, uintptr_t source, struct mbuf *m);
int
netisr_queue(u_int proto, struct mbuf *m);
int
netisr_queue_src(u_int proto, uintptr_t source, struct mbuf *m);
void
netisr_clearqdrops(const struct netisr_handler *nhp);
void
netisr_getqdrops(const struct netisr_handler *nhp, u_int64_t *qdropsp);
void
netisr_getqlimit(const struct netisr_handler *nhp, u_int *qlimitp);
int
netisr_setqlimit(const struct netisr_handler *nhp, u_int qlimit);
u_int
netisr_default_flow2cpu(u_int flowid);
u_int
netisr_get_cpucount(void);
u_int
netisr_get_cpuid(u_int cpunumber);
DESCRIPTION
The netisr kernel interface suite allows device drivers (and other packet
sources) to direct packets to protocols for directly dispatched or
deferred processing.
Protocol registration
Protocols register and unregister handlers using netisr_register() and
netisr_unregister(), and may also manage queue limits and statistics
using the netisr_clearqdrops(), netisr_getqdrops(), netisr_getqlimit(),
and netisr_setqlimit.()
netisr supports multi-processor execution of handlers, and relies on a
combination of source ordering and protocol-specific ordering and work-
placement policies to decide how do distribute work across one or more
worker threads. Registering protocols will declare one of three poli‐
cies:
NETISR_POLICY_SOURCE netisr should maintain source ordering without
advice from the protocol. netisr will ignore any
flow IDs present on mbuf headers for the purposes
of work placement.
NETISR_POLICY_FLOW netisr should maintain flow ordering as defined by
the mbuf header flow ID field. If the protocol
implements nh_m2flow, then netisr will query the
protocol in the evet that the mbuf doesn't have a
flow ID, falling back on source ordering.
NETISR_POLICY_CPU netisr will entirely delegate all work placement
decisions to the protocol, querying nh_m2cpuid for
each packet.
Registration is declared using struct netisr_handler, whose fields are
defined as follows:
const char * nh_name Unique character string name of the proto‐
col, which may be included in 2(sysctl) MIB
names, so should not contain whitespace.
netisr_handler_t nh_handler Protocol handler function that will be
invoked on each packet received for the pro‐
tocol.
netisr_m2flow_t nh_m2flow Optional protocol function to generate a
flow ID and set M_FLOWID for packets that do
not enter netisr with M_FLOWID defined.
Will be used only with NETISR_POLICY_FLOW.
netisr_m2cpuid_t nh_m2cpuid Protocol function to determine what CPU a
packet should be processed on. Will be used
only with NETISR_POLICY_CPU.
netisr_drainedcpu_t nh_drainedcpu
Optional callback function that will be
invoked when a per-CPU queue was drained.
It will never fire for directly dispatched
packets. Unless fully understood, this spe‐
cial-purpose function should not be used.
u_int nh_proto Protocol number used by both protocols to
identify themselves to netisr, and by packet
sources to select what handler will be used
to process packets. A table of supported
protocol numbers appears below. For imple‐
mentation reasons, protocol numbers great
than 15 are currently unsupported.
u_int nh_qlimit The maximum per-CPU queue depth for the pro‐
tocol; due to internal implementation
details, the effective queue depth may be as
much as twice this number.
u_int nh_policy The ordering and work placement policy for
the protocol, as described earlier.
Packet source interface
Packet sources, such as network interfaces, may request protocol process‐
ing using the netisr_dispatch() and netisr_queue() interfaces. Both
accept a protocol number and mbuf argument, but while netisr_queue() will
always execute the protocol handler asynchonously in a deferred context,
netisr_dispatch() will optionally direct dispatch if permitted by global
and per-protocol policy.
In order to provide additional load balancing and flow information,
packet sources may also specify an opaque source identifier, which in
practice might be a network interface number or socket pointer, using the
netisr_dispatch_src() and netisr_queue_src() variants.
Protocol number constants
The follow protocol numbers are currently defined:
NETISR_IP IPv4
NETISR_IGMP IGMPv3 loopback
NETISR_ROUTE Routing socket loopback
NETISR_AARP Appletalk AARP
NETISR_ATALK1 Appletalk phase 1
NETISR_ATALK2 Appletalk phase 2
NETISR_ARP ARP
NETISR_IPX IPX/SPX
NETISR_IPV6 IPv6
NETISR_NATM ATM
NETISR_EPAIR epair(4)AUTHORS
This manual page and the netisr implementation were written by Robert N.
M. Watson.
BSD July 26, 2009 BSD