[people/arne_f/kernel.git] / net / smc / smc_core.h

/*
 * Shared Memory Communications over RDMA (SMC-R) and RoCE
 *
 *  Definitions for SMC Connections, Link Groups and Links
 *
 *  Copyright IBM Corp. 2016
 *
 *  Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
 */

#ifndef _SMC_CORE_H
#define _SMC_CORE_H

#include <linux/atomic.h>
#include <rdma/ib_verbs.h>

#include "smc.h"
#include "smc_ib.h"

#define SMC_RMBS_PER_LGR_MAX	255	/* max. # of RMBs per link group */

struct smc_lgr_list {			/* list of link group definition */
	struct list_head	list;
	spinlock_t		lock;	/* protects list of link groups */
};

extern struct smc_lgr_list	smc_lgr_list; /* list of link groups */

enum smc_lgr_role {		/* possible roles of a link group */
	SMC_CLNT,	/* client */
	SMC_SERV	/* server */
};

#define SMC_WR_BUF_SIZE		48	/* size of work request buffer */

struct smc_wr_buf {
	u8	raw[SMC_WR_BUF_SIZE];
};

struct smc_link {
	struct smc_ib_device	*smcibdev;	/* ib-device */
	u8			ibport;		/* port - values 1 | 2 */
	struct ib_pd		*roce_pd;	/* IB protection domain,
						 * unique for every RoCE QP
						 */
	struct ib_qp		*roce_qp;	/* IB queue pair */
	struct ib_qp_attr	qp_attr;	/* IB queue pair attributes */

	struct smc_wr_buf	*wr_tx_bufs;	/* WR send payload buffers */
	struct ib_send_wr	*wr_tx_ibs;	/* WR send meta data */
	struct ib_sge		*wr_tx_sges;	/* WR send gather meta data */
	struct smc_wr_tx_pend	*wr_tx_pends;	/* WR send waiting for CQE */
	/* above four vectors have wr_tx_cnt elements and use the same index */
	dma_addr_t		wr_tx_dma_addr;	/* DMA address of wr_tx_bufs */
	atomic_long_t		wr_tx_id;	/* seq # of last sent WR */
	unsigned long		*wr_tx_mask;	/* bit mask of used indexes */
	u32			wr_tx_cnt;	/* number of WR send buffers */
	wait_queue_head_t	wr_tx_wait;	/* wait for free WR send buf */

	struct smc_wr_buf	*wr_rx_bufs;	/* WR recv payload buffers */
	struct ib_recv_wr	*wr_rx_ibs;	/* WR recv meta data */
	struct ib_sge		*wr_rx_sges;	/* WR recv scatter meta data */
	/* above three vectors have wr_rx_cnt elements and use the same index */
	dma_addr_t		wr_rx_dma_addr;	/* DMA address of wr_rx_bufs */
	u64			wr_rx_id;	/* seq # of last recv WR */
	u32			wr_rx_cnt;	/* number of WR recv buffers */

	union ib_gid		gid;		/* gid matching used vlan id */
	u32			peer_qpn;	/* QP number of peer */
	enum ib_mtu		path_mtu;	/* used mtu */
	enum ib_mtu		peer_mtu;	/* mtu size of peer */
	u32			psn_initial;	/* QP tx initial packet seqno */
	u32			peer_psn;	/* QP rx initial packet seqno */
	u8			peer_mac[ETH_ALEN];	/* = gid[8:10||13:15] */
	u8			peer_gid[sizeof(union ib_gid)];	/* gid of peer*/
	u8			link_id;	/* unique # within link group */
	struct completion	llc_confirm;	/* wait for rx of conf link */
	struct completion	llc_confirm_resp; /* wait 4 rx of cnf lnk rsp */
};

/* For now we just allow one parallel link per link group. The SMC protocol
 * allows more (up to 8).
 */
#define SMC_LINKS_PER_LGR_MAX	1
#define SMC_SINGLE_LINK		0

#define SMC_FIRST_CONTACT	1		/* first contact to a peer */
#define SMC_REUSE_CONTACT	0		/* follow-on contact to a peer*/

/* tx/rx buffer list element for sndbufs list and rmbs list of a lgr */
struct smc_buf_desc {
	struct list_head	list;
	u64			dma_addr[SMC_LINKS_PER_LGR_MAX];
						/* mapped address of buffer */
	void			*cpu_addr;	/* virtual address of buffer */
	u32			rkey[SMC_LINKS_PER_LGR_MAX];
						/* for rmb only:
						 * rkey provided to peer
						 */
	u32			used;		/* currently used / unused */
};

struct smc_rtoken {				/* address/key of remote RMB */
	u64			dma_addr;
	u32			rkey;
};

#define SMC_LGR_ID_SIZE		4

struct smc_link_group {
	struct list_head	list;
	enum smc_lgr_role	role;		/* client or server */
	__be32			daddr;		/* destination ip address */
	struct smc_link		lnk[SMC_LINKS_PER_LGR_MAX];	/* smc link */
	char			peer_systemid[SMC_SYSTEMID_LEN];
						/* unique system_id of peer */
	struct rb_root		conns_all;	/* connection tree */
	rwlock_t		conns_lock;	/* protects conns_all */
	unsigned int		conns_num;	/* current # of connections */
	unsigned short		vlan_id;	/* vlan id of link group */

	struct list_head	sndbufs[SMC_RMBE_SIZES];/* tx buffers */
	rwlock_t		sndbufs_lock;	/* protects tx buffers */
	struct list_head	rmbs[SMC_RMBE_SIZES];	/* rx buffers */
	rwlock_t		rmbs_lock;	/* protects rx buffers */
	struct smc_rtoken	rtokens[SMC_RMBS_PER_LGR_MAX]
				       [SMC_LINKS_PER_LGR_MAX];
						/* remote addr/key pairs */
	unsigned long		rtokens_used_mask[BITS_TO_LONGS(
							SMC_RMBS_PER_LGR_MAX)];
						/* used rtoken elements */

	u8			id[SMC_LGR_ID_SIZE];	/* unique lgr id */
	struct delayed_work	free_work;	/* delayed freeing of an lgr */
	bool			sync_err;	/* lgr no longer fits to peer */
};

/* Find the connection associated with the given alert token in the link group.
 * To use rbtrees we have to implement our own search core.
 * Requires @conns_lock
 * @token	alert token to search for
 * @lgr		 link group to search in
 * Returns connection associated with token if found, NULL otherwise.
 */
static inline struct smc_connection *smc_lgr_find_conn(
	u32 token, struct smc_link_group *lgr)
{
	struct smc_connection *res = NULL;
	struct rb_node *node;

	node = lgr->conns_all.rb_node;
	while (node) {
		struct smc_connection *cur = rb_entry(node,
					struct smc_connection, alert_node);

		if (cur->alert_token_local > token) {
			node = node->rb_left;
		} else {
			if (cur->alert_token_local < token) {
				node = node->rb_right;
			} else {
				res = cur;
				break;
			}
		}
	}

	return res;
}

struct smc_sock;
struct smc_clc_msg_accept_confirm;

void smc_lgr_free(struct smc_link_group *lgr);
void smc_lgr_terminate(struct smc_link_group *lgr);
int smc_sndbuf_create(struct smc_sock *smc);
int smc_rmb_create(struct smc_sock *smc);
int smc_rmb_rtoken_handling(struct smc_connection *conn,
			    struct smc_clc_msg_accept_confirm *clc);

#endif
Commit	Line	Data
0cfdd8f9 UB	1	/*
	2	* Shared Memory Communications over RDMA (SMC-R) and RoCE
	3	*
	4	* Definitions for SMC Connections, Link Groups and Links
	5	*
	6	* Copyright IBM Corp. 2016
	7	*
	8	* Author(s): Ursula Braun <ubraun@linux.vnet.ibm.com>
	9	*/
	10
	11	#ifndef _SMC_CORE_H
	12	#define _SMC_CORE_H
	13
f38ba179	14	#include <linux/atomic.h>
0cfdd8f9 UB	15	#include <rdma/ib_verbs.h>
	16
	17	#include "smc.h"
	18	#include "smc_ib.h"
	19
cd6851f3 UB	20	#define SMC_RMBS_PER_LGR_MAX 255 /* max. # of RMBs per link group */
cd6851f3 UB	21
0cfdd8f9 UB	22	struct smc_lgr_list { /* list of link group definition */
	23	struct list_head list;
	24	spinlock_t lock; /* protects list of link groups */
	25	};
	26
	27	extern struct smc_lgr_list smc_lgr_list; /* list of link groups */
	28
	29	enum smc_lgr_role { /* possible roles of a link group */
	30	SMC_CLNT, /* client */
	31	SMC_SERV /* server */
	32	};
	33
f38ba179 UB	34	#define SMC_WR_BUF_SIZE 48 /* size of work request buffer */
	35
	36	struct smc_wr_buf {
	37	u8 raw[SMC_WR_BUF_SIZE];
	38	};
	39
0cfdd8f9 UB	40	struct smc_link {
	41	struct smc_ib_device smcibdev; / ib-device */
	42	u8 ibport; /* port - values 1 \| 2 */
f38ba179 UB	43	struct ib_pd roce_pd; / IB protection domain,
	44	* unique for every RoCE QP
	45	*/
0cfdd8f9 UB	46	struct ib_qp roce_qp; / IB queue pair */
0cfdd8f9 UB	47	struct ib_qp_attr qp_attr; /* IB queue pair attributes */
f38ba179 UB	48
	49	struct smc_wr_buf wr_tx_bufs; / WR send payload buffers */
	50	struct ib_send_wr wr_tx_ibs; / WR send meta data */
	51	struct ib_sge wr_tx_sges; / WR send gather meta data */
	52	struct smc_wr_tx_pend wr_tx_pends; / WR send waiting for CQE */
	53	/* above four vectors have wr_tx_cnt elements and use the same index */
	54	dma_addr_t wr_tx_dma_addr; /* DMA address of wr_tx_bufs */
	55	atomic_long_t wr_tx_id; /* seq # of last sent WR */
	56	unsigned long wr_tx_mask; / bit mask of used indexes */
	57	u32 wr_tx_cnt; /* number of WR send buffers */
	58	wait_queue_head_t wr_tx_wait; /* wait for free WR send buf */
	59
	60	struct smc_wr_buf wr_rx_bufs; / WR recv payload buffers */
	61	struct ib_recv_wr wr_rx_ibs; / WR recv meta data */
	62	struct ib_sge wr_rx_sges; / WR recv scatter meta data */
	63	/* above three vectors have wr_rx_cnt elements and use the same index */
	64	dma_addr_t wr_rx_dma_addr; /* DMA address of wr_rx_bufs */
	65	u64 wr_rx_id; /* seq # of last recv WR */
	66	u32 wr_rx_cnt; /* number of WR recv buffers */
	67
0cfdd8f9 UB	68	union ib_gid gid; /* gid matching used vlan id */
	69	u32 peer_qpn; /* QP number of peer */
	70	enum ib_mtu path_mtu; /* used mtu */
	71	enum ib_mtu peer_mtu; /* mtu size of peer */
	72	u32 psn_initial; /* QP tx initial packet seqno */
	73	u32 peer_psn; /* QP rx initial packet seqno */
	74	u8 peer_mac[ETH_ALEN]; /* = gid[8:10\|\|13:15] */
	75	u8 peer_gid[sizeof(union ib_gid)]; /* gid of peer*/
9bf9abea UB	76	u8 link_id; /* unique # within link group */
	77	struct completion llc_confirm; /* wait for rx of conf link */
	78	struct completion llc_confirm_resp; /* wait 4 rx of cnf lnk rsp */
0cfdd8f9 UB	79	};
	80
	81	/* For now we just allow one parallel link per link group. The SMC protocol
	82	* allows more (up to 8).
	83	*/
	84	#define SMC_LINKS_PER_LGR_MAX 1
	85	#define SMC_SINGLE_LINK 0
	86
	87	#define SMC_FIRST_CONTACT 1 /* first contact to a peer */
	88	#define SMC_REUSE_CONTACT 0 /* follow-on contact to a peer*/
	89
cd6851f3 UB	90	/* tx/rx buffer list element for sndbufs list and rmbs list of a lgr */
	91	struct smc_buf_desc {
	92	struct list_head list;
	93	u64 dma_addr[SMC_LINKS_PER_LGR_MAX];
	94	/* mapped address of buffer */
	95	void cpu_addr; / virtual address of buffer */
263eec9b	96	u32 rkey[SMC_LINKS_PER_LGR_MAX];
bd4ad577 UB	97	/* for rmb only:
	98	* rkey provided to peer
	99	*/
cd6851f3 UB	100	u32 used; /* currently used / unused */
	101	};
	102
bd4ad577 UB	103	struct smc_rtoken { /* address/key of remote RMB */
	104	u64 dma_addr;
	105	u32 rkey;
	106	};
	107
9bf9abea UB	108	#define SMC_LGR_ID_SIZE 4
9bf9abea UB	109
0cfdd8f9 UB	110	struct smc_link_group {
	111	struct list_head list;
	112	enum smc_lgr_role role; /* client or server */
	113	__be32 daddr; /* destination ip address */
	114	struct smc_link lnk[SMC_LINKS_PER_LGR_MAX]; /* smc link */
	115	char peer_systemid[SMC_SYSTEMID_LEN];
	116	/* unique system_id of peer */
	117	struct rb_root conns_all; /* connection tree */
	118	rwlock_t conns_lock; /* protects conns_all */
	119	unsigned int conns_num; /* current # of connections */
	120	unsigned short vlan_id; /* vlan id of link group */
cd6851f3 UB	121
	122	struct list_head sndbufs[SMC_RMBE_SIZES];/* tx buffers */
	123	rwlock_t sndbufs_lock; /* protects tx buffers */
	124	struct list_head rmbs[SMC_RMBE_SIZES]; /* rx buffers */
	125	rwlock_t rmbs_lock; /* protects rx buffers */
bd4ad577 UB	126	struct smc_rtoken rtokens[SMC_RMBS_PER_LGR_MAX]
	127	[SMC_LINKS_PER_LGR_MAX];
	128	/* remote addr/key pairs */
	129	unsigned long rtokens_used_mask[BITS_TO_LONGS(
	130	SMC_RMBS_PER_LGR_MAX)];
	131	/* used rtoken elements */
	132
9bf9abea	133	u8 id[SMC_LGR_ID_SIZE]; /* unique lgr id */
0cfdd8f9 UB	134	struct delayed_work free_work; /* delayed freeing of an lgr */
	135	bool sync_err; /* lgr no longer fits to peer */
	136	};
	137
	138	/* Find the connection associated with the given alert token in the link group.
	139	* To use rbtrees we have to implement our own search core.
	140	* Requires @conns_lock
	141	* @token alert token to search for
	142	* @lgr link group to search in
	143	* Returns connection associated with token if found, NULL otherwise.
	144	*/
	145	static inline struct smc_connection *smc_lgr_find_conn(
	146	u32 token, struct smc_link_group *lgr)
	147	{
	148	struct smc_connection *res = NULL;
	149	struct rb_node *node;
	150
	151	node = lgr->conns_all.rb_node;
	152	while (node) {
	153	struct smc_connection *cur = rb_entry(node,
	154	struct smc_connection, alert_node);
	155
	156	if (cur->alert_token_local > token) {
	157	node = node->rb_left;
	158	} else {
	159	if (cur->alert_token_local < token) {
	160	node = node->rb_right;
	161	} else {
	162	res = cur;
	163	break;
	164	}
	165	}
	166	}
	167
	168	return res;
	169	}
	170
cd6851f3 UB	171	struct smc_sock;
	172	struct smc_clc_msg_accept_confirm;
	173
0cfdd8f9 UB	174	void smc_lgr_free(struct smc_link_group *lgr);
0cfdd8f9 UB	175	void smc_lgr_terminate(struct smc_link_group *lgr);
cd6851f3 UB	176	int smc_sndbuf_create(struct smc_sock *smc);
cd6851f3 UB	177	int smc_rmb_create(struct smc_sock *smc);
bd4ad577 UB	178	int smc_rmb_rtoken_handling(struct smc_connection *conn,
bd4ad577 UB	179	struct smc_clc_msg_accept_confirm *clc);
0cfdd8f9 UB	180
0cfdd8f9 UB	181	#endif