[net-next-2.6.git] / include / linux / skbuff.h

/*
 *	Definitions for the 'struct sk_buff' memory handlers.
 *
 *	Authors:
 *		Alan Cox, <gw4pts@gw4pts.ampr.org>
 *		Florian La Roche, <rzsfl@rz.uni-sb.de>
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License
 *	as published by the Free Software Foundation; either version
 *	2 of the License, or (at your option) any later version.
 */

#ifndef _LINUX_SKBUFF_H
#define _LINUX_SKBUFF_H

#include <linux/kernel.h>
#include <linux/compiler.h>
#include <linux/time.h>
#include <linux/cache.h>

#include <asm/atomic.h>
#include <asm/types.h>
#include <linux/spinlock.h>
#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/poll.h>
#include <linux/net.h>
#include <linux/textsearch.h>
#include <net/checksum.h>
#include <linux/dmaengine.h>

#define HAVE_ALLOC_SKB		/* For the drivers to know */
#define HAVE_ALIGNABLE_SKB	/* Ditto 8)		   */

#define CHECKSUM_NONE 0
#define CHECKSUM_HW 1
#define CHECKSUM_UNNECESSARY 2

#define SKB_DATA_ALIGN(X)	(((X) + (SMP_CACHE_BYTES - 1)) & \
				 ~(SMP_CACHE_BYTES - 1))
#define SKB_MAX_ORDER(X, ORDER)	(((PAGE_SIZE << (ORDER)) - (X) - \
				  sizeof(struct skb_shared_info)) & \
				  ~(SMP_CACHE_BYTES - 1))
#define SKB_MAX_HEAD(X)		(SKB_MAX_ORDER((X), 0))
#define SKB_MAX_ALLOC		(SKB_MAX_ORDER(0, 2))

/* A. Checksumming of received packets by device.
 *
 *	NONE: device failed to checksum this packet.
 *		skb->csum is undefined.
 *
 *	UNNECESSARY: device parsed packet and wouldbe verified checksum.
 *		skb->csum is undefined.
 *	      It is bad option, but, unfortunately, many of vendors do this.
 *	      Apparently with secret goal to sell you new device, when you
 *	      will add new protocol to your host. F.e. IPv6. 8)
 *
 *	HW: the most generic way. Device supplied checksum of _all_
 *	    the packet as seen by netif_rx in skb->csum.
 *	    NOTE: Even if device supports only some protocols, but
 *	    is able to produce some skb->csum, it MUST use HW,
 *	    not UNNECESSARY.
 *
 * B. Checksumming on output.
 *
 *	NONE: skb is checksummed by protocol or csum is not required.
 *
 *	HW: device is required to csum packet as seen by hard_start_xmit
 *	from skb->h.raw to the end and to record the checksum
 *	at skb->h.raw+skb->csum.
 *
 *	Device must show its capabilities in dev->features, set
 *	at device setup time.
 *	NETIF_F_HW_CSUM	- it is clever device, it is able to checksum
 *			  everything.
 *	NETIF_F_NO_CSUM - loopback or reliable single hop media.
 *	NETIF_F_IP_CSUM - device is dumb. It is able to csum only
 *			  TCP/UDP over IPv4. Sigh. Vendors like this
 *			  way by an unknown reason. Though, see comment above
 *			  about CHECKSUM_UNNECESSARY. 8)
 *
 *	Any questions? No questions, good. 		--ANK
 */

struct net_device;

#ifdef CONFIG_NETFILTER
struct nf_conntrack {
	atomic_t use;
	void (*destroy)(struct nf_conntrack *);
};

#ifdef CONFIG_BRIDGE_NETFILTER
struct nf_bridge_info {
	atomic_t use;
	struct net_device *physindev;
	struct net_device *physoutdev;
#if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
	struct net_device *netoutdev;
#endif
	unsigned int mask;
	unsigned long data[32 / sizeof(unsigned long)];
};
#endif

#endif

struct sk_buff_head {
	/* These two members must be first. */
	struct sk_buff	*next;
	struct sk_buff	*prev;

	__u32		qlen;
	spinlock_t	lock;
};

struct sk_buff;

/* To allow 64K frame to be packed as single skb without frag_list */
#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 2)

typedef struct skb_frag_struct skb_frag_t;

struct skb_frag_struct {
	struct page *page;
	__u16 page_offset;
	__u16 size;
};

/* This data is invariant across clones and lives at
 * the end of the header data, ie. at skb->end.
 */
struct skb_shared_info {
	atomic_t	dataref;
	unsigned short	nr_frags;
	unsigned short	gso_size;
	/* Warning: this field is not always filled in (UFO)! */
	unsigned short	gso_segs;
	unsigned short  gso_type;
	unsigned int    ip6_frag_id;
	struct sk_buff	*frag_list;
	skb_frag_t	frags[MAX_SKB_FRAGS];
};

/* We divide dataref into two halves.  The higher 16 bits hold references
 * to the payload part of skb->data.  The lower 16 bits hold references to
 * the entire skb->data.  It is up to the users of the skb to agree on
 * where the payload starts.
 *
 * All users must obey the rule that the skb->data reference count must be
 * greater than or equal to the payload reference count.
 *
 * Holding a reference to the payload part means that the user does not
 * care about modifications to the header part of skb->data.
 */
#define SKB_DATAREF_SHIFT 16
#define SKB_DATAREF_MASK ((1 << SKB_DATAREF_SHIFT) - 1)

struct skb_timeval {
	u32	off_sec;
	u32	off_usec;
};


enum {
	SKB_FCLONE_UNAVAILABLE,
	SKB_FCLONE_ORIG,
	SKB_FCLONE_CLONE,
};

enum {
	SKB_GSO_TCPV4 = 1 << 0,
	SKB_GSO_UDPV4 = 1 << 1,

	/* This indicates the skb is from an untrusted source. */
	SKB_GSO_DODGY = 1 << 2,

	/* This indicates the tcp segment has CWR set. */
	SKB_GSO_TCPV4_ECN = 1 << 3,
};

/** 
 *	struct sk_buff - socket buffer
 *	@next: Next buffer in list
 *	@prev: Previous buffer in list
 *	@sk: Socket we are owned by
 *	@tstamp: Time we arrived
 *	@dev: Device we arrived on/are leaving by
 *	@input_dev: Device we arrived on
 *	@h: Transport layer header
 *	@nh: Network layer header
 *	@mac: Link layer header
 *	@dst: destination entry
 *	@sp: the security path, used for xfrm
 *	@cb: Control buffer. Free for use by every layer. Put private vars here
 *	@len: Length of actual data
 *	@data_len: Data length
 *	@mac_len: Length of link layer header
 *	@csum: Checksum
 *	@local_df: allow local fragmentation
 *	@cloned: Head may be cloned (check refcnt to be sure)
 *	@nohdr: Payload reference only, must not modify header
 *	@pkt_type: Packet class
 *	@fclone: skbuff clone status
 *	@ip_summed: Driver fed us an IP checksum
 *	@priority: Packet queueing priority
 *	@users: User count - see {datagram,tcp}.c
 *	@protocol: Packet protocol from driver
 *	@truesize: Buffer size 
 *	@head: Head of buffer
 *	@data: Data head pointer
 *	@tail: Tail pointer
 *	@end: End pointer
 *	@destructor: Destruct function
 *	@nfmark: Can be used for communication between hooks
 *	@nfct: Associated connection, if any
 *	@ipvs_property: skbuff is owned by ipvs
 *	@nfctinfo: Relationship of this skb to the connection
 *	@nfct_reasm: netfilter conntrack re-assembly pointer
 *	@nf_bridge: Saved data about a bridged frame - see br_netfilter.c
 *	@tc_index: Traffic control index
 *	@tc_verd: traffic control verdict
 *	@dma_cookie: a cookie to one of several possible DMA operations
 *		done by skb DMA functions
 *	@secmark: security marking
 */

struct sk_buff {
	/* These two members must be first. */
	struct sk_buff		*next;
	struct sk_buff		*prev;

	struct sock		*sk;
	struct skb_timeval	tstamp;
	struct net_device	*dev;
	struct net_device	*input_dev;

	union {
		struct tcphdr	*th;
		struct udphdr	*uh;
		struct icmphdr	*icmph;
		struct igmphdr	*igmph;
		struct iphdr	*ipiph;
		struct ipv6hdr	*ipv6h;
		unsigned char	*raw;
	} h;

	union {
		struct iphdr	*iph;
		struct ipv6hdr	*ipv6h;
		struct arphdr	*arph;
		unsigned char	*raw;
	} nh;

	union {
	  	unsigned char 	*raw;
	} mac;

	struct  dst_entry	*dst;
	struct	sec_path	*sp;

	/*
	 * This is the control buffer. It is free to use for every
	 * layer. Please put your private variables there. If you
	 * want to keep them across layers you have to do a skb_clone()
	 * first. This is owned by whoever has the skb queued ATM.
	 */
	char			cb[48];

	unsigned int		len,
				data_len,
				mac_len,
				csum;
	__u32			priority;
	__u8			local_df:1,
				cloned:1,
				ip_summed:2,
				nohdr:1,
				nfctinfo:3;
	__u8			pkt_type:3,
				fclone:2,
				ipvs_property:1;
	__be16			protocol;

	void			(*destructor)(struct sk_buff *skb);
#ifdef CONFIG_NETFILTER
	struct nf_conntrack	*nfct;
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
	struct sk_buff		*nfct_reasm;
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
	struct nf_bridge_info	*nf_bridge;
#endif
	__u32			nfmark;
#endif /* CONFIG_NETFILTER */
#ifdef CONFIG_NET_SCHED
	__u16			tc_index;	/* traffic control index */
#ifdef CONFIG_NET_CLS_ACT
	__u16			tc_verd;	/* traffic control verdict */
#endif
#endif
#ifdef CONFIG_NET_DMA
	dma_cookie_t		dma_cookie;
#endif
#ifdef CONFIG_NETWORK_SECMARK
	__u32			secmark;
#endif


	/* These elements must be at the end, see alloc_skb() for details.  */
	unsigned int		truesize;
	atomic_t		users;
	unsigned char		*head,
				*data,
				*tail,
				*end;
};

#ifdef __KERNEL__
/*
 *	Handling routines are only of interest to the kernel
 */
#include <linux/slab.h>

#include <asm/system.h>
Commit	Line	Data
1da177e4 LT	1	/*
	2	* Definitions for the 'struct sk_buff' memory handlers.
	3	*
	4	* Authors:
	5	* Alan Cox, <gw4pts@gw4pts.ampr.org>
	6	* Florian La Roche, <rzsfl@rz.uni-sb.de>
	7	*
	8	* This program is free software; you can redistribute it and/or
	9	* modify it under the terms of the GNU General Public License
	10	* as published by the Free Software Foundation; either version
	11	* 2 of the License, or (at your option) any later version.
	12	*/
	13
	14	#ifndef _LINUX_SKBUFF_H
	15	#define _LINUX_SKBUFF_H
	16
1da177e4 LT	17	#include <linux/kernel.h>
	18	#include <linux/compiler.h>
	19	#include <linux/time.h>
	20	#include <linux/cache.h>
	21
	22	#include <asm/atomic.h>
	23	#include <asm/types.h>
	24	#include <linux/spinlock.h>
	25	#include <linux/mm.h>
	26	#include <linux/highmem.h>
	27	#include <linux/poll.h>
	28	#include <linux/net.h>
3fc7e8a6	29	#include <linux/textsearch.h>
1da177e4	30	#include <net/checksum.h>
97fc2f08	31	#include <linux/dmaengine.h>
1da177e4 LT	32
	33	#define HAVE_ALLOC_SKB /* For the drivers to know */
	34	#define HAVE_ALIGNABLE_SKB /* Ditto 8) */
1da177e4 LT	35
	36	#define CHECKSUM_NONE 0
	37	#define CHECKSUM_HW 1
	38	#define CHECKSUM_UNNECESSARY 2
	39
	40	#define SKB_DATA_ALIGN(X) (((X) + (SMP_CACHE_BYTES - 1)) & \
	41	~(SMP_CACHE_BYTES - 1))
	42	#define SKB_MAX_ORDER(X, ORDER) (((PAGE_SIZE << (ORDER)) - (X) - \
	43	sizeof(struct skb_shared_info)) & \
	44	~(SMP_CACHE_BYTES - 1))
	45	#define SKB_MAX_HEAD(X) (SKB_MAX_ORDER((X), 0))
	46	#define SKB_MAX_ALLOC (SKB_MAX_ORDER(0, 2))
	47
	48	/* A. Checksumming of received packets by device.
	49	*
	50	* NONE: device failed to checksum this packet.
	51	* skb->csum is undefined.
	52	*
	53	* UNNECESSARY: device parsed packet and wouldbe verified checksum.
	54	* skb->csum is undefined.
	55	* It is bad option, but, unfortunately, many of vendors do this.
	56	* Apparently with secret goal to sell you new device, when you
	57	* will add new protocol to your host. F.e. IPv6. 8)
	58	*
	59	* HW: the most generic way. Device supplied checksum of _all_
	60	* the packet as seen by netif_rx in skb->csum.
	61	* NOTE: Even if device supports only some protocols, but
	62	* is able to produce some skb->csum, it MUST use HW,
	63	* not UNNECESSARY.
	64	*
	65	* B. Checksumming on output.
	66	*
	67	* NONE: skb is checksummed by protocol or csum is not required.
	68	*
	69	* HW: device is required to csum packet as seen by hard_start_xmit
	70	* from skb->h.raw to the end and to record the checksum
	71	* at skb->h.raw+skb->csum.
	72	*
	73	* Device must show its capabilities in dev->features, set
	74	* at device setup time.
	75	* NETIF_F_HW_CSUM - it is clever device, it is able to checksum
	76	* everything.
	77	* NETIF_F_NO_CSUM - loopback or reliable single hop media.
	78	* NETIF_F_IP_CSUM - device is dumb. It is able to csum only
	79	* TCP/UDP over IPv4. Sigh. Vendors like this
	80	* way by an unknown reason. Though, see comment above
	81	* about CHECKSUM_UNNECESSARY. 8)
	82	*
	83	* Any questions? No questions, good. --ANK
	84	*/
	85
1da177e4 LT	86	struct net_device;
	87
	88	#ifdef CONFIG_NETFILTER
	89	struct nf_conntrack {
	90	atomic_t use;
	91	void (destroy)(struct nf_conntrack );
	92	};
	93
	94	#ifdef CONFIG_BRIDGE_NETFILTER
	95	struct nf_bridge_info {
	96	atomic_t use;
	97	struct net_device *physindev;
	98	struct net_device *physoutdev;
	99	#if defined(CONFIG_VLAN_8021Q) \|\| defined(CONFIG_VLAN_8021Q_MODULE)
	100	struct net_device *netoutdev;
	101	#endif
	102	unsigned int mask;
	103	unsigned long data[32 / sizeof(unsigned long)];
	104	};
	105	#endif
	106
	107	#endif
	108
	109	struct sk_buff_head {
	110	/* These two members must be first. */
	111	struct sk_buff *next;
	112	struct sk_buff *prev;
	113
	114	__u32 qlen;
	115	spinlock_t lock;
	116	};
	117
	118	struct sk_buff;
	119
	120	/* To allow 64K frame to be packed as single skb without frag_list */
	121	#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 2)
	122
	123	typedef struct skb_frag_struct skb_frag_t;
	124
	125	struct skb_frag_struct {
	126	struct page *page;
	127	__u16 page_offset;
	128	__u16 size;
	129	};
	130
	131	/* This data is invariant across clones and lives at
	132	* the end of the header data, ie. at skb->end.
	133	*/
	134	struct skb_shared_info {
	135	atomic_t dataref;
4947d3ef	136	unsigned short nr_frags;
7967168c HX	137	unsigned short gso_size;
	138	/* Warning: this field is not always filled in (UFO)! */
	139	unsigned short gso_segs;
	140	unsigned short gso_type;
e89e9cf5	141	unsigned int ip6_frag_id;
1da177e4 LT	142	struct sk_buff *frag_list;
	143	skb_frag_t frags[MAX_SKB_FRAGS];
	144	};
	145
	146	/* We divide dataref into two halves. The higher 16 bits hold references
	147	* to the payload part of skb->data. The lower 16 bits hold references to
	148	* the entire skb->data. It is up to the users of the skb to agree on
	149	* where the payload starts.
	150	*
	151	* All users must obey the rule that the skb->data reference count must be
	152	* greater than or equal to the payload reference count.
	153	*
	154	* Holding a reference to the payload part means that the user does not
	155	* care about modifications to the header part of skb->data.
	156	*/
	157	#define SKB_DATAREF_SHIFT 16
	158	#define SKB_DATAREF_MASK ((1 << SKB_DATAREF_SHIFT) - 1)
	159
a61bbcf2 PM	160	struct skb_timeval {
	161	u32 off_sec;
	162	u32 off_usec;
	163	};
	164
d179cd12 DM	165
	166	enum {
	167	SKB_FCLONE_UNAVAILABLE,
	168	SKB_FCLONE_ORIG,
	169	SKB_FCLONE_CLONE,
	170	};
	171
7967168c HX	172	enum {
	173	SKB_GSO_TCPV4 = 1 << 0,
	174	SKB_GSO_UDPV4 = 1 << 1,
576a30eb HX	175
	176	/* This indicates the skb is from an untrusted source. */
	177	SKB_GSO_DODGY = 1 << 2,
b0da8537 MC	178
	179	/* This indicates the tcp segment has CWR set. */
	180	SKB_GSO_TCPV4_ECN = 1 << 3,
7967168c HX	181	};
7967168c HX	182
1da177e4 LT	183	/**
	184	* struct sk_buff - socket buffer
	185	* @next: Next buffer in list
	186	* @prev: Previous buffer in list
1da177e4	187	* @sk: Socket we are owned by
325ed823	188	* @tstamp: Time we arrived
1da177e4 LT	189	* @dev: Device we arrived on/are leaving by
1da177e4 LT	190	* @input_dev: Device we arrived on
1da177e4 LT	191	* @h: Transport layer header
	192	* @nh: Network layer header
	193	* @mac: Link layer header
67be2dd1 MW	194	* @dst: destination entry
67be2dd1 MW	195	* @sp: the security path, used for xfrm
1da177e4 LT	196	* @cb: Control buffer. Free for use by every layer. Put private vars here
	197	* @len: Length of actual data
	198	* @data_len: Data length
	199	* @mac_len: Length of link layer header
	200	* @csum: Checksum
67be2dd1	201	* @local_df: allow local fragmentation
1da177e4 LT	202	* @cloned: Head may be cloned (check refcnt to be sure)
	203	* @nohdr: Payload reference only, must not modify header
	204	* @pkt_type: Packet class
c83c2486	205	* @fclone: skbuff clone status
1da177e4 LT	206	* @ip_summed: Driver fed us an IP checksum
	207	* @priority: Packet queueing priority
	208	* @users: User count - see {datagram,tcp}.c
	209	* @protocol: Packet protocol from driver
1da177e4 LT	210	* @truesize: Buffer size
	211	* @head: Head of buffer
	212	* @data: Data head pointer
	213	* @tail: Tail pointer
	214	* @end: End pointer
	215	* @destructor: Destruct function
	216	* @nfmark: Can be used for communication between hooks
1da177e4	217	* @nfct: Associated connection, if any
c83c2486	218	* @ipvs_property: skbuff is owned by ipvs
1da177e4	219	* @nfctinfo: Relationship of this skb to the connection
461ddf3b	220	* @nfct_reasm: netfilter conntrack re-assembly pointer
1da177e4	221	* @nf_bridge: Saved data about a bridged frame - see br_netfilter.c
1da177e4 LT	222	* @tc_index: Traffic control index
1da177e4 LT	223	* @tc_verd: traffic control verdict
f4b8ea78 RD	224	* @dma_cookie: a cookie to one of several possible DMA operations
f4b8ea78 RD	225	* done by skb DMA functions
984bc16c	226	* @secmark: security marking
1da177e4 LT	227	*/
	228
	229	struct sk_buff {
	230	/* These two members must be first. */
	231	struct sk_buff *next;
	232	struct sk_buff *prev;
	233
1da177e4	234	struct sock *sk;
a61bbcf2	235	struct skb_timeval tstamp;
1da177e4 LT	236	struct net_device *dev;
1da177e4 LT	237	struct net_device *input_dev;
1da177e4 LT	238
	239	union {
	240	struct tcphdr *th;
	241	struct udphdr *uh;
	242	struct icmphdr *icmph;
	243	struct igmphdr *igmph;
	244	struct iphdr *ipiph;
	245	struct ipv6hdr *ipv6h;
	246	unsigned char *raw;
	247	} h;
	248
	249	union {
	250	struct iphdr *iph;
	251	struct ipv6hdr *ipv6h;
	252	struct arphdr *arph;
	253	unsigned char *raw;
	254	} nh;
	255
	256	union {
	257	unsigned char *raw;
	258	} mac;
	259
	260	struct dst_entry *dst;
	261	struct sec_path *sp;
	262
	263	/*
	264	* This is the control buffer. It is free to use for every
	265	* layer. Please put your private variables there. If you
	266	* want to keep them across layers you have to do a skb_clone()
	267	* first. This is owned by whoever has the skb queued ATM.
	268	*/
3e3850e9	269	char cb[48];
1da177e4 LT	270
	271	unsigned int len,
	272	data_len,
	273	mac_len,
	274	csum;
1da177e4	275	__u32 priority;
1cbb3380 TG	276	__u8 local_df:1,
	277	cloned:1,
	278	ip_summed:2,
6869c4d8 HW	279	nohdr:1,
6869c4d8 HW	280	nfctinfo:3;
d179cd12	281	__u8 pkt_type:3,
b84f4cc9 PM	282	fclone:2,
b84f4cc9 PM	283	ipvs_property:1;
a0d3bea3	284	__be16 protocol;
1da177e4 LT	285
	286	void (destructor)(struct sk_buff skb);
	287	#ifdef CONFIG_NETFILTER
1da177e4	288	struct nf_conntrack *nfct;
9fb9cbb1 YK	289	#if defined(CONFIG_NF_CONNTRACK) \|\| defined(CONFIG_NF_CONNTRACK_MODULE)
	290	struct sk_buff *nfct_reasm;
	291	#endif
1da177e4 LT	292	#ifdef CONFIG_BRIDGE_NETFILTER
	293	struct nf_bridge_info *nf_bridge;
	294	#endif
77d2ca35	295	__u32 nfmark;
1da177e4	296	#endif /* CONFIG_NETFILTER */
1da177e4	297	#ifdef CONFIG_NET_SCHED
b6b99eb5	298	__u16 tc_index; /* traffic control index */
1da177e4	299	#ifdef CONFIG_NET_CLS_ACT
b6b99eb5	300	__u16 tc_verd; /* traffic control verdict */
1da177e4	301	#endif
1da177e4	302	#endif
97fc2f08 CL	303	#ifdef CONFIG_NET_DMA
	304	dma_cookie_t dma_cookie;
	305	#endif
984bc16c JM	306	#ifdef CONFIG_NETWORK_SECMARK
	307	__u32 secmark;
	308	#endif
1da177e4 LT	309
	310
	311	/* These elements must be at the end, see alloc_skb() for details. */
	312	unsigned int truesize;
	313	atomic_t users;
	314	unsigned char *head,
	315	*data,
	316	*tail,
	317	*end;
	318	};
	319
	320	#ifdef __KERNEL__
	321	/*
	322	* Handling routines are only of interest to the kernel
	323	*/
	324	#include <linux/slab.h>
	325
	326	#include <asm/system.h>
	327