freebsd-dev/sys/net/if_var.h

511 lines
17 KiB
C
Raw Normal View History

/*
* Copyright (c) 1982, 1986, 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* From: @(#)if.h 8.1 (Berkeley) 6/10/93
1999-08-28 01:08:13 +00:00
* $FreeBSD$
*/
#ifndef _NET_IF_VAR_H_
#define _NET_IF_VAR_H_
/*
* Structures defining a network interface, providing a packet
* transport mechanism (ala level 0 of the PUP protocols).
*
* Each interface accepts output datagrams of a specified maximum
* length, and provides higher level routines with input datagrams
* received from its medium.
*
* Output occurs when the routine if_output is called, with three parameters:
* (*ifp->if_output)(ifp, m, dst, rt)
* Here m is the mbuf chain to be sent and dst is the destination address.
* The output routine encapsulates the supplied datagram if necessary,
* and then transmits it on its medium.
*
* On input, each interface unwraps the data received by it, and either
* places it on the input queue of an internetwork datagram routine
* and posts the associated software interrupt, or passes the datagram to a raw
* packet input routine.
*
* Routines exist for locating interfaces by their addresses
* or for locating an interface on a certain network, as well as more general
* routing and gateway routines maintaining information used to locate
* interfaces. These routines live in the files if.c and route.c
*/
#ifdef __STDC__
/*
* Forward structure declarations for function prototypes [sic].
*/
struct mbuf;
struct thread;
struct rtentry;
struct rt_addrinfo;
struct socket;
struct ether_header;
#endif
#include <sys/queue.h> /* get TAILQ macros */
#ifdef _KERNEL
#include <sys/mbuf.h>
#include <sys/eventhandler.h>
#endif /* _KERNEL */
#include <sys/lock.h> /* XXX */
#include <sys/mutex.h> /* XXX */
#include <sys/event.h> /* XXX */
#define IF_DUNIT_NONE -1
TAILQ_HEAD(ifnethead, ifnet); /* we use TAILQs so that the order of */
TAILQ_HEAD(ifaddrhead, ifaddr); /* instantiation is preserved in the list */
TAILQ_HEAD(ifprefixhead, ifprefix);
TAILQ_HEAD(ifmultihead, ifmultiaddr);
/*
* Structure defining a queue for a network interface.
*/
struct ifqueue {
struct mbuf *ifq_head;
struct mbuf *ifq_tail;
int ifq_len;
int ifq_maxlen;
int ifq_drops;
struct mtx ifq_mtx;
};
/*
* Structure defining a network interface.
*
* (Would like to call this struct ``if'', but C isn't PL/1.)
*/
/*
* NB: For FreeBSD, it is assumed that each NIC driver's softc starts with
* one of these structures, typically held within an arpcom structure.
*
* struct <foo>_softc {
* struct arpcom {
* struct ifnet ac_if;
* ...
* } <arpcom> ;
* ...
* };
*
* The assumption is used in a number of places, including many
* files in sys/net, device drivers, and sys/dev/mii.c:miibus_attach().
*
* Unfortunately devices' softc are opaque, so we depend on this layout
* to locate the struct ifnet from the softc in the generic code.
*
*/
struct ifnet {
void *if_softc; /* pointer to driver state */
TAILQ_ENTRY(ifnet) if_link; /* all struct ifnets are chained */
char if_xname[IFNAMSIZ]; /* external name (name + unit) */
const char *if_dname; /* driver name */
int if_dunit; /* unit or IF_DUNIT_NONE */
struct ifaddrhead if_addrhead; /* linked list of addresses per if */
/*
* if_addrhead is the list of all addresses associated to
* an interface.
* Some code in the kernel assumes that first element
* of the list has type AF_LINK, and contains sockaddr_dl
* addresses which store the link-level address and the name
* of the interface.
* However, access to the AF_LINK address through this
* field is deprecated. Use ifaddr_byindex() instead.
*/
struct klist if_klist; /* events attached to this if */
int if_pcount; /* number of promiscuous listeners */
struct bpf_if *if_bpf; /* packet filter structure */
u_short if_index; /* numeric abbreviation for this if */
short if_timer; /* time 'til if_watchdog called */
u_short if_nvlans; /* number of active vlans */
int if_flags; /* up/down, broadcast, etc. */
int if_capabilities; /* interface capabilities */
int if_capenable; /* enabled features */
void *if_linkmib; /* link-type-specific MIB data */
size_t if_linkmiblen; /* length of above data */
struct if_data if_data;
struct ifmultihead if_multiaddrs; /* multicast addresses configured */
int if_amcount; /* number of all-multicast requests */
/* procedure handles */
int (*if_output) /* output routine (enqueue) */
2002-03-19 21:54:18 +00:00
(struct ifnet *, struct mbuf *, struct sockaddr *,
struct rtentry *);
void (*if_input) /* input routine (from h/w driver) */
(struct ifnet *, struct mbuf *);
void (*if_start) /* initiate output routine */
2002-03-19 21:54:18 +00:00
(struct ifnet *);
int (*if_ioctl) /* ioctl routine */
2002-03-19 21:54:18 +00:00
(struct ifnet *, u_long, caddr_t);
void (*if_watchdog) /* timer routine */
2002-03-19 21:54:18 +00:00
(struct ifnet *);
void (*if_init) /* Init routine */
2002-03-19 21:54:18 +00:00
(void *);
int (*if_resolvemulti) /* validate/resolve multicast */
2002-03-19 21:54:18 +00:00
(struct ifnet *, struct sockaddr **, struct sockaddr *);
struct ifqueue if_snd; /* output queue */
const u_int8_t *if_broadcastaddr; /* linklevel broadcast bytestring */
struct lltable *lltables; /* list of L3-L2 resolution tables */
Modify the MAC Framework so that instead of embedding a (struct label) in various kernel objects to represent security data, we embed a (struct label *) pointer, which now references labels allocated using a UMA zone (mac_label.c). This allows the size and shape of struct label to be varied without changing the size and shape of these kernel objects, which become part of the frozen ABI with 5-STABLE. This opens the door for boot-time selection of the number of label slots, and hence changes to the bound on the number of simultaneous labeled policies at boot-time instead of compile-time. This also makes it easier to embed label references in new objects as required for locking/caching with fine-grained network stack locking, such as inpcb structures. This change also moves us further in the direction of hiding the structure of kernel objects from MAC policy modules, not to mention dramatically reducing the number of '&' symbols appearing in both the MAC Framework and MAC policy modules, and improving readability. While this results in minimal performance change with MAC enabled, it will observably shrink the size of a number of critical kernel data structures for the !MAC case, and should have a small (but measurable) performance benefit (i.e., struct vnode, struct socket) do to memory conservation and reduced cost of zeroing memory. NOTE: Users of MAC must recompile their kernel and all MAC modules as a result of this change. Because this is an API change, third party MAC modules will also need to be updated to make less use of the '&' symbol. Suggestions from: bmilekic Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
struct label *if_label; /* interface MAC label */
/* these are only used by IPv6 */
struct ifprefixhead if_prefixhead; /* list of prefixes per if */
void *if_afdata[AF_MAX];
int if_afdata_initialized;
struct mtx if_afdata_mtx;
};
2002-03-19 21:54:18 +00:00
typedef void if_init_f_t(void *);
/*
* XXX These aliases are terribly dangerous because they could apply
* to anything.
*/
#define if_mtu if_data.ifi_mtu
#define if_type if_data.ifi_type
#define if_physical if_data.ifi_physical
#define if_addrlen if_data.ifi_addrlen
#define if_hdrlen if_data.ifi_hdrlen
#define if_metric if_data.ifi_metric
#define if_link_state if_data.ifi_link_state
#define if_baudrate if_data.ifi_baudrate
#define if_hwassist if_data.ifi_hwassist
#define if_ipackets if_data.ifi_ipackets
#define if_ierrors if_data.ifi_ierrors
#define if_opackets if_data.ifi_opackets
#define if_oerrors if_data.ifi_oerrors
#define if_collisions if_data.ifi_collisions
#define if_ibytes if_data.ifi_ibytes
#define if_obytes if_data.ifi_obytes
#define if_imcasts if_data.ifi_imcasts
#define if_omcasts if_data.ifi_omcasts
#define if_iqdrops if_data.ifi_iqdrops
#define if_noproto if_data.ifi_noproto
#define if_lastchange if_data.ifi_lastchange
#define if_recvquota if_data.ifi_recvquota
#define if_xmitquota if_data.ifi_xmitquota
#define if_rawoutput(if, m, sa) if_output(if, m, sa, (struct rtentry *)0)
/* for compatibility with other BSDs */
#define if_addrlist if_addrhead
#define if_list if_link
/*
* Output queues (ifp->if_snd) and slow device input queues (*ifp->if_slowq)
* are queues of messages stored on ifqueue structures
* (defined above). Entries are added to and deleted from these structures
* by these macros, which should be called with ipl raised to splimp().
*/
Change and clean the mutex lock interface. mtx_enter(lock, type) becomes: mtx_lock(lock) for sleep locks (MTX_DEF-initialized locks) mtx_lock_spin(lock) for spin locks (MTX_SPIN-initialized) similarily, for releasing a lock, we now have: mtx_unlock(lock) for MTX_DEF and mtx_unlock_spin(lock) for MTX_SPIN. We change the caller interface for the two different types of locks because the semantics are entirely different for each case, and this makes it explicitly clear and, at the same time, it rids us of the extra `type' argument. The enter->lock and exit->unlock change has been made with the idea that we're "locking data" and not "entering locked code" in mind. Further, remove all additional "flags" previously passed to the lock acquire/release routines with the exception of two: MTX_QUIET and MTX_NOSWITCH The functionality of these flags is preserved and they can be passed to the lock/unlock routines by calling the corresponding wrappers: mtx_{lock, unlock}_flags(lock, flag(s)) and mtx_{lock, unlock}_spin_flags(lock, flag(s)) for MTX_DEF and MTX_SPIN locks, respectively. Re-inline some lock acq/rel code; in the sleep lock case, we only inline the _obtain_lock()s in order to ensure that the inlined code fits into a cache line. In the spin lock case, we inline recursion and actually only perform a function call if we need to spin. This change has been made with the idea that we generally tend to avoid spin locks and that also the spin locks that we do have and are heavily used (i.e. sched_lock) do recurse, and therefore in an effort to reduce function call overhead for some architectures (such as alpha), we inline recursion for this case. Create a new malloc type for the witness code and retire from using the M_DEV type. The new type is called M_WITNESS and is only declared if WITNESS is enabled. Begin cleaning up some machdep/mutex.h code - specifically updated the "optimized" inlined code in alpha/mutex.h and wrote MTX_LOCK_SPIN and MTX_UNLOCK_SPIN asm macros for the i386/mutex.h as we presently need those. Finally, caught up to the interface changes in all sys code. Contributors: jake, jhb, jasone (in no particular order)
2001-02-09 06:11:45 +00:00
#define IF_LOCK(ifq) mtx_lock(&(ifq)->ifq_mtx)
#define IF_UNLOCK(ifq) mtx_unlock(&(ifq)->ifq_mtx)
#define _IF_QFULL(ifq) ((ifq)->ifq_len >= (ifq)->ifq_maxlen)
#define _IF_DROP(ifq) ((ifq)->ifq_drops++)
#define _IF_QLEN(ifq) ((ifq)->ifq_len)
#define _IF_ENQUEUE(ifq, m) do { \
(m)->m_nextpkt = NULL; \
if ((ifq)->ifq_tail == NULL) \
(ifq)->ifq_head = m; \
else \
(ifq)->ifq_tail->m_nextpkt = m; \
(ifq)->ifq_tail = m; \
(ifq)->ifq_len++; \
} while (0)
#define IF_ENQUEUE(ifq, m) do { \
IF_LOCK(ifq); \
_IF_ENQUEUE(ifq, m); \
IF_UNLOCK(ifq); \
} while (0)
#define _IF_PREPEND(ifq, m) do { \
(m)->m_nextpkt = (ifq)->ifq_head; \
if ((ifq)->ifq_tail == NULL) \
(ifq)->ifq_tail = (m); \
(ifq)->ifq_head = (m); \
(ifq)->ifq_len++; \
} while (0)
#define IF_PREPEND(ifq, m) do { \
IF_LOCK(ifq); \
_IF_PREPEND(ifq, m); \
IF_UNLOCK(ifq); \
} while (0)
#define _IF_DEQUEUE(ifq, m) do { \
(m) = (ifq)->ifq_head; \
if (m) { \
if (((ifq)->ifq_head = (m)->m_nextpkt) == 0) \
(ifq)->ifq_tail = NULL; \
(m)->m_nextpkt = NULL; \
(ifq)->ifq_len--; \
} \
} while (0)
#define IF_DEQUEUE(ifq, m) do { \
IF_LOCK(ifq); \
_IF_DEQUEUE(ifq, m); \
IF_UNLOCK(ifq); \
} while (0)
#define IF_DRAIN(ifq) do { \
struct mbuf *m; \
IF_LOCK(ifq); \
for (;;) { \
_IF_DEQUEUE(ifq, m); \
if (m == NULL) \
break; \
m_freem(m); \
} \
IF_UNLOCK(ifq); \
} while (0)
#ifdef _KERNEL
/* interface address change event */
typedef void (*ifaddr_event_handler_t)(void *, struct ifnet *);
EVENTHANDLER_DECLARE(ifaddr_event, ifaddr_event_handler_t);
/* new interface arrival event */
typedef void (*ifnet_arrival_event_handler_t)(void *, struct ifnet *);
EVENTHANDLER_DECLARE(ifnet_arrival_event, ifnet_arrival_event_handler_t);
/* interface departure event */
typedef void (*ifnet_departure_event_handler_t)(void *, struct ifnet *);
EVENTHANDLER_DECLARE(ifnet_departure_event, ifnet_departure_event_handler_t);
/* interface clone event */
typedef void (*if_clone_event_handler_t)(void *, struct if_clone *);
EVENTHANDLER_DECLARE(if_clone_event, if_clone_event_handler_t);
#define IF_AFDATA_LOCK_INIT(ifp) \
mtx_init(&(ifp)->if_afdata_mtx, "if_afdata", NULL, MTX_DEF)
#define IF_AFDATA_LOCK(ifp) mtx_lock(&(ifp)->if_afdata_mtx)
#define IF_AFDATA_TRYLOCK(ifp) mtx_trylock(&(ifp)->if_afdata_mtx)
#define IF_AFDATA_UNLOCK(ifp) mtx_unlock(&(ifp)->if_afdata_mtx)
#define IF_AFDATA_DESTROY(ifp) mtx_destroy(&(ifp)->if_afdata_mtx)
#define IF_HANDOFF(ifq, m, ifp) if_handoff(ifq, m, ifp, 0)
#define IF_HANDOFF_ADJ(ifq, m, ifp, adj) if_handoff(ifq, m, ifp, adj)
1998-04-15 17:47:40 +00:00
static __inline int
if_handoff(struct ifqueue *ifq, struct mbuf *m, struct ifnet *ifp, int adjust)
{
int active = 0;
IF_LOCK(ifq);
if (_IF_QFULL(ifq)) {
_IF_DROP(ifq);
IF_UNLOCK(ifq);
m_freem(m);
return (0);
}
if (ifp != NULL) {
ifp->if_obytes += m->m_pkthdr.len + adjust;
if (m->m_flags & (M_BCAST|M_MCAST))
ifp->if_omcasts++;
active = ifp->if_flags & IFF_OACTIVE;
}
_IF_ENQUEUE(ifq, m);
IF_UNLOCK(ifq);
if (ifp != NULL && !active)
2002-05-07 18:11:55 +00:00
(*ifp->if_start)(ifp);
return (1);
}
/*
* 72 was chosen below because it is the size of a TCP/IP
* header (40) + the minimum mss (32).
*/
#define IF_MINMTU 72
#define IF_MAXMTU 65535
#endif /* _KERNEL */
/*
* The ifaddr structure contains information about one address
* of an interface. They are maintained by the different address families,
* are allocated and attached when an address is set, and are linked
* together so all addresses for an interface can be located.
*
* NOTE: a 'struct ifaddr' is always at the beginning of a larger
* chunk of malloc'ed memory, where we store the three addresses
* (ifa_addr, ifa_dstaddr and ifa_netmask) referenced here.
*/
struct ifaddr {
struct sockaddr *ifa_addr; /* address of interface */
struct sockaddr *ifa_dstaddr; /* other end of p-to-p link */
#define ifa_broadaddr ifa_dstaddr /* broadcast address interface */
struct sockaddr *ifa_netmask; /* used to determine subnet */
struct if_data if_data; /* not all members are meaningful */
struct ifnet *ifa_ifp; /* back-pointer to interface */
TAILQ_ENTRY(ifaddr) ifa_link; /* queue macro glue */
void (*ifa_rtrequest) /* check or clean routes (+ or -)'d */
2002-03-19 21:54:18 +00:00
(int, struct rtentry *, struct rt_addrinfo *);
u_short ifa_flags; /* mostly rt_flags for cloning */
u_int ifa_refcnt; /* references to this structure */
int ifa_metric; /* cost of going out this interface */
int (*ifa_claim_addr) /* check if an addr goes to this if */
2002-03-19 21:54:18 +00:00
(struct ifaddr *, struct sockaddr *);
2002-12-18 11:46:59 +00:00
struct mtx ifa_mtx;
};
#define IFA_ROUTE RTF_UP /* route installed */
/* for compatibility with other BSDs */
#define ifa_list ifa_link
2002-12-18 11:46:59 +00:00
#define IFA_LOCK_INIT(ifa) \
mtx_init(&(ifa)->ifa_mtx, "ifaddr", NULL, MTX_DEF)
#define IFA_LOCK(ifa) mtx_lock(&(ifa)->ifa_mtx)
#define IFA_UNLOCK(ifa) mtx_unlock(&(ifa)->ifa_mtx)
#define IFA_DESTROY(ifa) mtx_destroy(&(ifa)->ifa_mtx)
/*
* The prefix structure contains information about one prefix
* of an interface. They are maintained by the different address families,
* are allocated and attached when a prefix or an address is set,
* and are linked together so all prefixes for an interface can be located.
*/
struct ifprefix {
struct sockaddr *ifpr_prefix; /* prefix of interface */
struct ifnet *ifpr_ifp; /* back-pointer to interface */
TAILQ_ENTRY(ifprefix) ifpr_list; /* queue macro glue */
u_char ifpr_plen; /* prefix length in bits */
u_char ifpr_type; /* protocol dependent prefix type */
};
/*
* Multicast address structure. This is analogous to the ifaddr
* structure except that it keeps track of multicast addresses.
* Also, the reference count here is a count of requests for this
* address, not a count of pointers to this structure.
*/
struct ifmultiaddr {
TAILQ_ENTRY(ifmultiaddr) ifma_link; /* queue macro glue */
struct sockaddr *ifma_addr; /* address this membership is for */
struct sockaddr *ifma_lladdr; /* link-layer translation, if any */
struct ifnet *ifma_ifp; /* back-pointer to interface */
u_int ifma_refcount; /* reference count */
void *ifma_protospec; /* protocol-specific state, if any */
};
#ifdef _KERNEL
#define IFAFREE(ifa) \
do { \
IFA_LOCK(ifa); \
KASSERT((ifa)->ifa_refcnt > 0, \
("ifa %p !(ifa_refcnt > 0)", ifa)); \
if (--(ifa)->ifa_refcnt == 0) { \
IFA_DESTROY(ifa); \
free(ifa, M_IFADDR); \
} else \
IFA_UNLOCK(ifa); \
2002-12-18 11:46:59 +00:00
} while (0)
#define IFAREF(ifa) \
do { \
IFA_LOCK(ifa); \
++(ifa)->ifa_refcnt; \
IFA_UNLOCK(ifa); \
} while (0)
2002-12-22 05:35:03 +00:00
extern struct mtx ifnet_lock;
#define IFNET_LOCK_INIT() \
mtx_init(&ifnet_lock, "ifnet", NULL, MTX_DEF | MTX_RECURSE)
2002-12-22 05:35:03 +00:00
#define IFNET_WLOCK() mtx_lock(&ifnet_lock)
#define IFNET_WUNLOCK() mtx_unlock(&ifnet_lock)
#define IFNET_RLOCK() IFNET_WLOCK()
#define IFNET_RUNLOCK() IFNET_WUNLOCK()
struct ifindex_entry {
2001-12-14 19:32:47 +00:00
struct ifnet *ife_ifnet;
struct ifaddr *ife_ifnet_addr;
dev_t ife_dev;
};
#define ifnet_byindex(idx) ifindex_table[(idx)].ife_ifnet
/*
* Given the index, ifaddr_byindex() returns the one and only
* link-level ifaddr for the interface. You are not supposed to use
* it to traverse the list of addresses associated to the interface.
*/
#define ifaddr_byindex(idx) ifindex_table[(idx)].ife_ifnet_addr
#define ifdev_byindex(idx) ifindex_table[(idx)].ife_dev
extern struct ifnethead ifnet;
extern struct ifindex_entry *ifindex_table;
extern int ifqmaxlen;
extern struct ifnet *loif; /* first loopback interface */
extern int if_index;
2002-03-19 21:54:18 +00:00
int if_addmulti(struct ifnet *, struct sockaddr *, struct ifmultiaddr **);
int if_allmulti(struct ifnet *, int);
void if_attach(struct ifnet *);
int if_delmulti(struct ifnet *, struct sockaddr *);
void if_detach(struct ifnet *);
void if_down(struct ifnet *);
void if_initname(struct ifnet *, const char *, int);
int if_printf(struct ifnet *, const char *, ...) __printflike(2, 3);
2002-03-19 21:54:18 +00:00
int if_setlladdr(struct ifnet *, const u_char *, int);
void if_up(struct ifnet *);
/*void ifinit(void);*/ /* declared in systm.h for main() */
int ifioctl(struct socket *, u_long, caddr_t, struct thread *);
int ifpromisc(struct ifnet *, int);
struct ifnet *ifunit(const char *);
struct ifaddr *ifa_ifwithaddr(struct sockaddr *);
struct ifaddr *ifa_ifwithdstaddr(struct sockaddr *);
struct ifaddr *ifa_ifwithnet(struct sockaddr *);
struct ifaddr *ifa_ifwithroute(int, struct sockaddr *, struct sockaddr *);
struct ifaddr *ifaof_ifpforaddr(struct sockaddr *, struct ifnet *);
struct ifmultiaddr *ifmaof_ifpforaddr(struct sockaddr *, struct ifnet *);
int if_simloop(struct ifnet *ifp, struct mbuf *m, int af, int hlen);
void if_clone_attach(struct if_clone *);
void if_clone_detach(struct if_clone *);
int if_clone_create(char *, int);
int if_clone_destroy(const char *);
#define IF_LLADDR(ifp) \
LLADDR((struct sockaddr_dl *) ifaddr_byindex((ifp)->if_index)->ifa_addr)
Device Polling code for -current. Non-SMP, i386-only, no polling in the idle loop at the moment. To use this code you must compile a kernel with options DEVICE_POLLING and at runtime enable polling with sysctl kern.polling.enable=1 The percentage of CPU reserved to userland can be set with sysctl kern.polling.user_frac=NN (default is 50) while the remainder is used by polling device drivers and netisr's. These are the only two variables that you should need to touch. There are a few more parameters in kern.polling but the default values are adequate for all purposes. See the code in kern_poll.c for more details on them. Polling in the idle loop will be implemented shortly by introducing a kernel thread which does the job. Until then, the amount of CPU dedicated to polling will never exceed (100-user_frac). The equivalent (actually, better) code for -stable is at http://info.iet.unipi.it/~luigi/polling/ and also supports polling in the idle loop. NOTE to Alpha developers: There is really nothing in this code that is i386-specific. If you move the 2 lines supporting the new option from sys/conf/{files,options}.i386 to sys/conf/{files,options} I am pretty sure that this should work on the Alpha as well, just that I do not have a suitable test box to try it. If someone feels like trying it, I would appreciate it. NOTE to other developers: sure some things could be done better, and as always I am open to constructive criticism, which a few of you have already given and I greatly appreciated. However, before proposing radical architectural changes, please take some time to possibly try out this code, or at the very least read the comments in kern_poll.c, especially re. the reason why I am using a soft netisr and cannot (I believe) replace it with a simple timeout. Quick description of files touched by this commit: sys/conf/files.i386 new file kern/kern_poll.c sys/conf/options.i386 new option sys/i386/i386/trap.c poll in trap (disabled by default) sys/kern/kern_clock.c initialization and hardclock hooks. sys/kern/kern_intr.c minor swi_net changes sys/kern/kern_poll.c the bulk of the code. sys/net/if.h new flag sys/net/if_var.h declaration for functions used in device drivers. sys/net/netisr.h NETISR_POLL sys/dev/fxp/if_fxp.c sys/dev/fxp/if_fxpvar.h sys/pci/if_dc.c sys/pci/if_dcreg.h sys/pci/if_sis.c sys/pci/if_sisreg.h device driver modifications
2001-12-14 17:56:12 +00:00
#ifdef DEVICE_POLLING
enum poll_cmd { POLL_ONLY, POLL_AND_CHECK_STATUS, POLL_DEREGISTER };
2002-03-19 21:54:18 +00:00
typedef void poll_handler_t(struct ifnet *ifp, enum poll_cmd cmd, int count);
int ether_poll_register(poll_handler_t *h, struct ifnet *ifp);
int ether_poll_deregister(struct ifnet *ifp);
Device Polling code for -current. Non-SMP, i386-only, no polling in the idle loop at the moment. To use this code you must compile a kernel with options DEVICE_POLLING and at runtime enable polling with sysctl kern.polling.enable=1 The percentage of CPU reserved to userland can be set with sysctl kern.polling.user_frac=NN (default is 50) while the remainder is used by polling device drivers and netisr's. These are the only two variables that you should need to touch. There are a few more parameters in kern.polling but the default values are adequate for all purposes. See the code in kern_poll.c for more details on them. Polling in the idle loop will be implemented shortly by introducing a kernel thread which does the job. Until then, the amount of CPU dedicated to polling will never exceed (100-user_frac). The equivalent (actually, better) code for -stable is at http://info.iet.unipi.it/~luigi/polling/ and also supports polling in the idle loop. NOTE to Alpha developers: There is really nothing in this code that is i386-specific. If you move the 2 lines supporting the new option from sys/conf/{files,options}.i386 to sys/conf/{files,options} I am pretty sure that this should work on the Alpha as well, just that I do not have a suitable test box to try it. If someone feels like trying it, I would appreciate it. NOTE to other developers: sure some things could be done better, and as always I am open to constructive criticism, which a few of you have already given and I greatly appreciated. However, before proposing radical architectural changes, please take some time to possibly try out this code, or at the very least read the comments in kern_poll.c, especially re. the reason why I am using a soft netisr and cannot (I believe) replace it with a simple timeout. Quick description of files touched by this commit: sys/conf/files.i386 new file kern/kern_poll.c sys/conf/options.i386 new option sys/i386/i386/trap.c poll in trap (disabled by default) sys/kern/kern_clock.c initialization and hardclock hooks. sys/kern/kern_intr.c minor swi_net changes sys/kern/kern_poll.c the bulk of the code. sys/net/if.h new flag sys/net/if_var.h declaration for functions used in device drivers. sys/net/netisr.h NETISR_POLL sys/dev/fxp/if_fxp.c sys/dev/fxp/if_fxpvar.h sys/pci/if_dc.c sys/pci/if_dcreg.h sys/pci/if_sis.c sys/pci/if_sisreg.h device driver modifications
2001-12-14 17:56:12 +00:00
#endif /* DEVICE_POLLING */
#endif /* _KERNEL */
#endif /* !_NET_IF_VAR_H_ */