*/
/*
* This file is part of Lustre, http://www.lustre.org/
- * Lustre is a trademark of Sun Microsystems, Inc.
*
* lnet/include/lnet/lib-types.h
*
#include <linux/uio.h>
#include <linux/semaphore.h>
#include <linux/types.h>
+#include <linux/kref.h>
+#include <net/genetlink.h>
+#include <uapi/linux/lnet/lnet-nl.h>
#include <uapi/linux/lnet/lnet-dlc.h>
#include <uapi/linux/lnet/lnetctl.h>
#include <uapi/linux/lnet/nidstr.h>
/* Max payload size */
#define LNET_MAX_PAYLOAD LNET_MTU
-#define LNET_MAX_IOV (LNET_MAX_PAYLOAD >> PAGE_SHIFT)
+/** limit on the number of fragments in discontiguous MDs */
+#define LNET_MAX_IOV 256
/*
* This is the maximum health value.
* All local and peer NIs created have their health default to this value.
*/
#define LNET_MAX_HEALTH_VALUE 1000
+#define LNET_MAX_SELECTION_PRIORITY UINT_MAX
/* forward refs */
struct lnet_libmd;
/* cpt to lock */
int rspt_cpt;
/* nid of next hop */
- lnet_nid_t rspt_next_hop_nid;
+ struct lnet_nid rspt_next_hop_nid;
/* deadline of the REPLY/ACK */
ktime_t rspt_deadline;
/* parent MD */
enum lnet_msg_hstatus msg_health_status;
/* This is a recovery message */
bool msg_recovery;
+ /* force an RDMA even if the message size is < 4K */
+ bool msg_rdma_force;
/* the number of times a transmission has been retried */
int msg_retry_count;
/* flag to indicate that we do not want to resend this message */
unsigned int msg_wanted;
unsigned int msg_offset;
unsigned int msg_niov;
- struct kvec *msg_iov;
struct bio_vec *msg_kiov;
struct lnet_event msg_ev;
struct lnet_rsp_tracker *md_rspt_ptr;
lnet_handler_t md_handler;
struct lnet_handle_md md_bulk_handle;
- union {
- struct kvec iov[LNET_MAX_IOV];
- struct bio_vec kiov[LNET_MAX_IOV];
- } md_iov;
+ struct bio_vec md_kiov[LNET_MAX_IOV];
};
-#define LNET_MD_FLAG_ZOMBIE (1 << 0)
-#define LNET_MD_FLAG_AUTO_UNLINK (1 << 1)
-#define LNET_MD_FLAG_ABORTED (1 << 2)
+#define LNET_MD_FLAG_ZOMBIE BIT(0)
+#define LNET_MD_FLAG_AUTO_UNLINK BIT(1)
+#define LNET_MD_FLAG_ABORTED BIT(2)
+/* LNET_MD_FLAG_HANDLING is set when a non-unlink event handler
+ * is being called for an event relating to the md.
+ * It ensures only one such handler runs at a time.
+ * The final "unlink" event is only called once the
+ * md_refcount has reached zero, and this flag has been cleared,
+ * ensuring that it doesn't race with any other event handler
+ * call.
+ */
+#define LNET_MD_FLAG_HANDLING BIT(3)
+#define LNET_MD_FLAG_DISCARD BIT(4)
struct lnet_test_peer {
/* info about peers we are trying to fail */
struct list_head tp_list; /* ln_test_peers */
- lnet_nid_t tp_nid; /* matching nid */
+ struct lnet_nid tp_nid; /* matching nid */
unsigned int tp_threshold; /* # failures to simulate */
};
#define LNET_COOKIE_TYPE_BITS 2
#define LNET_COOKIE_MASK ((1ULL << LNET_COOKIE_TYPE_BITS) - 1ULL)
+struct netstrfns {
+ u32 nf_type;
+ char *nf_name;
+ char *nf_modname;
+ void (*nf_addr2str)(u32 addr, char *str, size_t size);
+ void (*nf_addr2str_size)(const __be32 *addr, size_t asize,
+ char *str, size_t size);
+ int (*nf_str2addr)(const char *str, int nob, u32 *addr);
+ int (*nf_str2addr_size)(const char *str, int nob,
+ __be32 *addr, size_t *asize);
+ int (*nf_parse_addrlist)(char *str, int len,
+ struct list_head *list);
+ int (*nf_print_addrlist)(char *buffer, int count,
+ struct list_head *list);
+ int (*nf_match_addr)(u32 addr, struct list_head *list);
+ int (*nf_min_max)(struct list_head *nidlist, u32 *min_nid,
+ u32 *max_nid);
+};
+
struct lnet_ni; /* forward ref */
struct socket;
int (*lnd_ctl)(struct lnet_ni *ni, unsigned int cmd, void *arg);
/* In data movement APIs below, payload buffers are described as a set
- * of 'niov' fragments which are...
- * EITHER
- * in virtual memory (struct kvec *iov != NULL)
- * OR
- * in pages (kernel only: plt_kiov_t *kiov != NULL).
+ * of 'niov' fragments which are in pages.
* The LND may NOT overwrite these fragment descriptors.
* An 'offset' and may specify a byte offset within the set of
* fragments to start from
* credit if the LND does flow control. */
int (*lnd_recv)(struct lnet_ni *ni, void *private, struct lnet_msg *msg,
int delayed, unsigned int niov,
- struct kvec *iov, struct bio_vec *kiov,
+ struct bio_vec *kiov,
unsigned int offset, unsigned int mlen, unsigned int rlen);
/* lnet_parse() has had to delay processing of this message
/* accept a new connection */
int (*lnd_accept)(struct lnet_ni *ni, struct socket *sock);
+
+ /* get dma_dev priority */
+ unsigned int (*lnd_get_dev_prio)(struct lnet_ni *ni,
+ unsigned int dev_idx);
};
struct lnet_tx_queue {
* lnet/include/lnet/nidstr.h */
__u32 net_id;
- /* priority of the network */
- __u32 net_prio;
+ /* round robin selection */
+ __u32 net_seq;
/* total number of CPTs in the array */
__u32 net_ncpts;
/* cumulative CPTs of all NIs in this net */
__u32 *net_cpts;
+ /* relative net selection priority */
+ __u32 net_sel_priority;
+
/* network tunables */
struct lnet_ioctl_config_lnd_cmn_tunables net_tunables;
/* protects access to net_last_alive */
spinlock_t net_lock;
+
+ /* list of router nids preferred for this network */
+ struct list_head net_rtr_pref_nids;
};
struct lnet_ni {
__u32 *ni_cpts;
/* interface's NID */
- lnet_nid_t ni_nid;
+ struct lnet_nid ni_nid;
/* instance-specific data */
void *ni_data;
/* Recovery state. Protected by lnet_ni_lock() */
__u32 ni_recovery_state;
+ /* When to send the next recovery ping */
+ time64_t ni_next_ping;
+ /* How many pings sent during current recovery period did not receive
+ * a reply. NB: reset whenever _any_ message arrives on this NI
+ */
+ unsigned int ni_ping_count;
+
/* per NI LND tunables */
struct lnet_lnd_tunables ni_lnd_tunables;
*/
atomic_t ni_fatal_error_on;
+ /* the relative selection priority of this NI */
+ __u32 ni_sel_priority;
+
/*
- * equivalent interfaces to use
- * This is an array because socklnd bonding can still be configured
+ * equivalent interface to use
*/
- char *ni_interfaces[LNET_INTERFACES_NUM];
+ char *ni_interface;
struct net *ni_net_ns; /* original net namespace */
};
#define LNET_PING_INFO_TO_BUFFER(PINFO) \
container_of((PINFO), struct lnet_ping_buffer, pb_info)
+struct lnet_nid_list {
+ struct list_head nl_list;
+ lnet_nid_t nl_nid;
+};
+
struct lnet_peer_ni {
/* chain on lpn_peer_nis */
struct list_head lpni_peer_nis;
/* network peer is on */
struct lnet_net *lpni_net;
/* peer's NID */
- lnet_nid_t lpni_nid;
+ struct lnet_nid lpni_nid;
/* # refs */
- atomic_t lpni_refcount;
+ struct kref lpni_kref;
/* health value for the peer */
atomic_t lpni_healthv;
/* recovery ping mdh */
struct lnet_handle_md lpni_recovery_ping_mdh;
+ /* When to send the next recovery ping */
+ time64_t lpni_next_ping;
+ /* How many pings sent during current recovery period did not receive
+ * a reply. NB: reset whenever _any_ message arrives from this peer NI
+ */
+ unsigned int lpni_ping_count;
/* CPT this peer attached on */
int lpni_cpt;
/* state flags -- protected by lpni_lock */
/* preferred local nids: if only one, use lpni_pref.nid */
union lpni_pref {
lnet_nid_t nid;
- lnet_nid_t *nids;
+ struct list_head nids;
} lpni_pref;
+ /* list of router nids preferred for this peer NI */
+ struct list_head lpni_rtr_pref_nids;
+ /* The relative selection priority of this peer NI */
+ __u32 lpni_sel_priority;
/* number of preferred NIDs in lnpi_pref_nids */
__u32 lpni_pref_nnids;
};
/* Preferred path added due to traffic on non-MR peer_ni */
-#define LNET_PEER_NI_NON_MR_PREF (1 << 0)
+#define LNET_PEER_NI_NON_MR_PREF BIT(0)
/* peer is being recovered. */
-#define LNET_PEER_NI_RECOVERY_PENDING (1 << 1)
+#define LNET_PEER_NI_RECOVERY_PENDING BIT(1)
/* recovery ping failed */
-#define LNET_PEER_NI_RECOVERY_FAILED (1 << 2)
+#define LNET_PEER_NI_RECOVERY_FAILED BIT(2)
/* peer is being deleted */
-#define LNET_PEER_NI_DELETING (1 << 3)
+#define LNET_PEER_NI_DELETING BIT(3)
struct lnet_peer {
/* chain on pt_peer_list */
struct list_head lp_rtr_list;
/* primary NID of the peer */
- lnet_nid_t lp_primary_nid;
+ struct lnet_nid lp_primary_nid;
+
+ /* source NID to use during discovery */
+ struct lnet_nid lp_disc_src_nid;
+ /* destination NID to use during discovery */
+ struct lnet_nid lp_disc_dst_nid;
/* net to perform discovery on */
__u32 lp_disc_net_id;
*
* A peer is marked ROUTER if it indicates so in the feature bit.
*/
-#define LNET_PEER_MULTI_RAIL (1 << 0) /* Multi-rail aware */
-#define LNET_PEER_NO_DISCOVERY (1 << 1) /* Peer disabled discovery */
-#define LNET_PEER_ROUTER_ENABLED (1 << 2) /* router feature enabled */
+#define LNET_PEER_MULTI_RAIL BIT(0) /* Multi-rail aware */
+#define LNET_PEER_NO_DISCOVERY BIT(1) /* Peer disabled discovery */
+#define LNET_PEER_ROUTER_ENABLED BIT(2) /* router feature enabled */
/*
* A peer is marked CONFIGURED if it was configured by DLC.
* A peer that was created as the result of inbound traffic will not
* be marked at all.
*/
-#define LNET_PEER_CONFIGURED (1 << 3) /* Configured via DLC */
-#define LNET_PEER_DISCOVERED (1 << 4) /* Peer was discovered */
-#define LNET_PEER_REDISCOVER (1 << 5) /* Discovery was disabled */
+#define LNET_PEER_CONFIGURED BIT(3) /* Configured via DLC */
+#define LNET_PEER_DISCOVERED BIT(4) /* Peer was discovered */
+#define LNET_PEER_REDISCOVER BIT(5) /* Discovery was disabled */
/*
* A peer is marked DISCOVERING when discovery is in progress.
* The other flags below correspond to stages of discovery.
*/
-#define LNET_PEER_DISCOVERING (1 << 6) /* Discovering */
-#define LNET_PEER_DATA_PRESENT (1 << 7) /* Remote peer data present */
-#define LNET_PEER_NIDS_UPTODATE (1 << 8) /* Remote peer info uptodate */
-#define LNET_PEER_PING_SENT (1 << 9) /* Waiting for REPLY to Ping */
-#define LNET_PEER_PUSH_SENT (1 << 10) /* Waiting for ACK of Push */
-#define LNET_PEER_PING_FAILED (1 << 11) /* Ping send failure */
-#define LNET_PEER_PUSH_FAILED (1 << 12) /* Push send failure */
+#define LNET_PEER_DISCOVERING BIT(6) /* Discovering */
+#define LNET_PEER_DATA_PRESENT BIT(7) /* Remote peer data present */
+#define LNET_PEER_NIDS_UPTODATE BIT(8) /* Remote peer info uptodate */
+#define LNET_PEER_PING_SENT BIT(9) /* Waiting for REPLY to Ping */
+#define LNET_PEER_PUSH_SENT BIT(10) /* Waiting for ACK of Push */
+#define LNET_PEER_PING_FAILED BIT(11) /* Ping send failure */
+#define LNET_PEER_PUSH_FAILED BIT(12) /* Push send failure */
/*
* A ping can be forced as a way to fix up state, or as a manual
* intervention by an admin.
* A push can be forced in circumstances that would normally not
* allow for one to happen.
*/
-#define LNET_PEER_FORCE_PING (1 << 13) /* Forced Ping */
-#define LNET_PEER_FORCE_PUSH (1 << 14) /* Forced Push */
+#define LNET_PEER_FORCE_PING BIT(13) /* Forced Ping */
+#define LNET_PEER_FORCE_PUSH BIT(14) /* Forced Push */
/* force delete even if router */
-#define LNET_PEER_RTR_NI_FORCE_DEL (1 << 15)
+#define LNET_PEER_RTR_NI_FORCE_DEL BIT(15)
/* gw undergoing alive discovery */
-#define LNET_PEER_RTR_DISCOVERY (1 << 16)
+#define LNET_PEER_RTR_DISCOVERY BIT(16)
/* gw has undergone discovery (does not indicate success or failure) */
-#define LNET_PEER_RTR_DISCOVERED (1 << 17)
+#define LNET_PEER_RTR_DISCOVERED BIT(17)
/* peer is marked for deletion */
-#define LNET_PEER_MARK_DELETION (1 << 18)
+#define LNET_PEER_MARK_DELETION BIT(18)
+/* lnet_peer_del()/lnet_peer_del_locked() has been called on the peer */
+#define LNET_PEER_MARK_DELETED BIT(19)
+/* lock primary NID to what's requested by ULP */
+#define LNET_PEER_LOCK_PRIMARY BIT(20)
+/* this is for informational purposes only. It is set if a peer gets
+ * configured from Lustre with a primary NID which belongs to another peer
+ * which is also configured by Lustre as the primary NID.
+ */
+#define LNET_PEER_BAD_CONFIG BIT(21)
struct lnet_peer_net {
/* chain on lp_peer_nets */
/* peer net health */
int lpn_healthv;
- /* time of last router net check attempt */
- time64_t lpn_rtrcheck_timestamp;
+ /* time of next router ping on this net */
+ time64_t lpn_next_ping;
/* selection sequence number */
__u32 lpn_seq;
+ /* relative peer net selection priority */
+ __u32 lpn_sel_priority;
+
/* reference count */
atomic_t lpn_refcount;
};
struct list_head lr_list; /* chain on net */
struct list_head lr_gwlist; /* chain on gateway */
struct lnet_peer *lr_gateway; /* router node */
- lnet_nid_t lr_nid; /* NID used to add route */
+ struct lnet_nid lr_nid; /* NID used to add route */
__u32 lr_net; /* remote network number */
__u32 lr_lnet; /* local network number */
int lr_seq; /* sequence for round-robin */
__u32 lr_hops; /* how far I am */
unsigned int lr_priority; /* route priority */
- bool lr_alive; /* cached route aliveness */
+ atomic_t lr_alive; /* cached route aliveness */
bool lr_single_hop; /* this route is single-hop */
};
enum lnet_match_flags {
/* Didn't match anything */
- LNET_MATCHMD_NONE = (1 << 0),
+ LNET_MATCHMD_NONE = BIT(0),
/* Matched OK */
- LNET_MATCHMD_OK = (1 << 1),
+ LNET_MATCHMD_OK = BIT(1),
/* Must be discarded */
- LNET_MATCHMD_DROP = (1 << 2),
+ LNET_MATCHMD_DROP = BIT(2),
/* match and buffer is exhausted */
- LNET_MATCHMD_EXHAUSTED = (1 << 3),
+ LNET_MATCHMD_EXHAUSTED = BIT(3),
/* match or drop */
LNET_MATCHMD_FINISH = (LNET_MATCHMD_OK | LNET_MATCHMD_DROP),
};
/* Options for struct lnet_portal::ptl_options */
-#define LNET_PTL_LAZY (1 << 0)
-#define LNET_PTL_MATCH_UNIQUE (1 << 1) /* unique match, for RDMA */
-#define LNET_PTL_MATCH_WILDCARD (1 << 2) /* wildcard match, request portal */
+#define LNET_PTL_LAZY BIT(0)
+#define LNET_PTL_MATCH_UNIQUE BIT(1) /* unique match, for RDMA */
+#define LNET_PTL_MATCH_WILDCARD BIT(2) /* wildcard match, request portal */
/* parameter for matching operations (GET, PUT) */
struct lnet_match_info {
void **msc_resenders;
};
+/* This UDSP structures need to match the user space liblnetconfig structures
+ * in order for the marshall and unmarshall functions to be common.
+ */
+
+/* Net is described as a
+ * 1. net type
+ * 2. num range
+ */
+struct lnet_ud_net_descr {
+ __u32 udn_net_type;
+ struct list_head udn_net_num_range;
+};
+
+/* each NID range is defined as
+ * 1. net descriptor
+ * 2. address range descriptor
+ */
+struct lnet_ud_nid_descr {
+ struct lnet_ud_net_descr ud_net_id;
+ struct list_head ud_addr_range;
+ __u32 ud_mem_size;
+};
+
+/* a UDSP rule can have up to three user defined NID descriptors
+ * - src: defines the local NID range for the rule
+ * - dst: defines the peer NID range for the rule
+ * - rte: defines the router NID range for the rule
+ *
+ * An action union defines the action to take when the rule
+ * is matched
+ */
+struct lnet_udsp {
+ struct list_head udsp_on_list;
+ __u32 udsp_idx;
+ struct lnet_ud_nid_descr udsp_src;
+ struct lnet_ud_nid_descr udsp_dst;
+ struct lnet_ud_nid_descr udsp_rte;
+ enum lnet_udsp_action_type udsp_action_type;
+ union {
+ __u32 udsp_priority;
+ } udsp_action;
+};
+
/* Peer Discovery states */
#define LNET_DC_STATE_SHUTDOWN 0 /* not started */
#define LNET_DC_STATE_RUNNING 1 /* started up OK */
* work loops
*/
struct completion ln_started;
+ /* UDSP list */
+ struct list_head ln_udsp_list;
};
+static const struct nla_policy scalar_attr_policy[LN_SCALAR_CNT + 1] = {
+ [LN_SCALAR_ATTR_LIST] = { .type = NLA_NESTED },
+ [LN_SCALAR_ATTR_LIST_SIZE] = { .type = NLA_U16 },
+ [LN_SCALAR_ATTR_INDEX] = { .type = NLA_U16 },
+ [LN_SCALAR_ATTR_NLA_TYPE] = { .type = NLA_U16 },
+ [LN_SCALAR_ATTR_VALUE] = { .type = NLA_STRING },
+ [LN_SCALAR_ATTR_KEY_FORMAT] = { .type = NLA_U16 },
+};
+
+int lnet_genl_send_scalar_list(struct sk_buff *msg, u32 portid, u32 seq,
+ const struct genl_family *family, int flags,
+ u8 cmd, const struct ln_key_list *data[]);
+
+/* Special workaround for pre-4.19 kernels to send error messages
+ * from dumpit routines. Newer kernels will send message with
+ * NL_SET_ERR_MSG information by default if NETLINK_EXT_ACK is set.
+ */
+static inline int lnet_nl_send_error(struct sk_buff *msg, int portid, int seq,
+ int error)
+{
+#ifndef HAVE_NL_DUMP_WITH_EXT_ACK
+ struct nlmsghdr *nlh;
+
+ if (!error)
+ return 0;
+
+ nlh = nlmsg_put(msg, portid, seq, NLMSG_ERROR, sizeof(error), 0);
+ if (!nlh)
+ return -ENOMEM;
+#ifdef HAVE_NL_PARSE_WITH_EXT_ACK
+ netlink_ack(msg, nlh, error, NULL);
+#else
+ netlink_ack(msg, nlh, error);
+#endif
+ return nlmsg_len(nlh);
+#else
+ return error;
+#endif
+}
+
#endif