Viewing: o2iblnd.h
/* SPDX-License-Identifier: GPL-2.0 */
/* Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
* Copyright (c) 2011, 2017, Intel Corporation.
*/
/* This file is part of Lustre, http://www.lustre.org/
*
* Author: Eric Barton <eric@bartonsoftware.com>
*/
#include <linux/module.h>
#include <linux/kernel.h>
#if defined(EXTERNAL_OFED_BUILD) && !defined(HAVE_OFED_IB_DMA_MAP_SG_SANE)
#undef CONFIG_INFINIBAND_VIRT_DMA
#endif
#ifdef HAVE_OFED_COMPAT_RDMA
#include <linux/compat-2.6.h>
#endif
#include <linux/kthread.h>
#include <linux/mm.h>
#include <linux/string.h>
#include <linux/stat.h>
#include <linux/errno.h>
#include <linux/unistd.h>
#include <linux/uio.h>
#include <asm/uaccess.h>
#include <asm/io.h>
#include <linux/init.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/stat.h>
#include <linux/list.h>
#include <linux/kmod.h>
#include <linux/sysctl.h>
#include <linux/pci.h>
#include <net/sock.h>
#include <linux/in.h>
#include <rdma/rdma_cm.h>
#include <rdma/ib_cm.h>
#include <lustre_compat/rdma/ib_verbs.h>
#ifdef HAVE_OFED_FMR_POOL_API
#include <rdma/ib_fmr_pool.h>
#endif
#define DEBUG_SUBSYSTEM S_LND
#include <linux/libcfs/libcfs.h>
#include <linux/lnet/lib-lnet.h>
#include <linux/lnet/lnet_rdma.h>
#include "o2iblnd-idl.h"
enum kiblnd_ni_lnd_tunables_attr {
LNET_NET_O2IBLND_TUNABLES_ATTR_UNSPEC = 0,
LNET_NET_O2IBLND_TUNABLES_ATTR_HIW_PEER_CREDITS,
LNET_NET_O2IBLND_TUNABLES_ATTR_CONCURRENT_SENDS,
LNET_NET_O2IBLND_TUNABLES_ATTR_MAP_ON_DEMAND,
LNET_NET_O2IBLND_TUNABLES_ATTR_FMR_POOL_SIZE,
LNET_NET_O2IBLND_TUNABLES_ATTR_FMR_FLUSH_TRIGGER,
LNET_NET_O2IBLND_TUNABLES_ATTR_FMR_CACHE,
LNET_NET_O2IBLND_TUNABLES_ATTR_NTX,
LNET_NET_O2IBLND_TUNABLES_ATTR_CONNS_PER_PEER,
LNET_NET_O2IBLND_TUNABLES_ATTR_LND_TIMEOUT,
LNET_NET_O2IBLND_TUNABLES_ATTR_LND_TOS,
__LNET_NET_O2IBLND_TUNABLES_ATTR_MAX_PLUS_ONE,
};
#define LNET_NET_O2IBLND_TUNABLES_ATTR_MAX (__LNET_NET_O2IBLND_TUNABLES_ATTR_MAX_PLUS_ONE - 1)
#define IBLND_PEER_HASH_BITS 7 /* log2 of # peer_ni lists */
#define IBLND_N_SCHED 2
#define IBLND_N_SCHED_HIGH 4
struct kib_tunables {
int *kib_dev_failover; /* HCA failover */
unsigned int *kib_service; /* IB service number */
int *kib_cksum; /* checksum struct kib_msg? */
int *kib_timeout; /* comms timeout (seconds) */
int *kib_keepalive; /* keepalive timeout (seconds) */
char **kib_default_ipif; /* default IPoIB interface */
int *kib_retry_count;
int *kib_rnr_retry_count;
int *kib_ib_mtu; /* IB MTU */
int *kib_require_priv_port;/* accept only privileged ports */
int *kib_use_priv_port; /* use privileged port for active connect */
/* # threads on each CPT */
int *kib_nscheds;
int *kib_wrq_sge; /* # sg elements per wrq */
int *kib_use_fastreg_gaps; /* enable discontiguous fastreg fragment support */
};
extern struct kib_tunables kiblnd_tunables;
extern struct lnet_ioctl_config_o2iblnd_tunables kib_default_tunables;
#define IBLND_MSG_QUEUE_SIZE_V1 8 /* V1 only : # messages/RDMAs in-flight */
#define IBLND_CREDIT_HIGHWATER_V1 7 /* V1 only : when eagerly to return credits */
#define IBLND_CREDITS_DEFAULT 8 /* default # of peer_ni credits */
#define IBLND_CREDITS_MAX ((typeof(((struct kib_msg *) 0)->ibm_credits)) - 1) /* Max # of peer_ni credits */
#define IBLND_TIMEOUT_DEFAULT 50 /* Default o2iblnd timeout in seconds */
/* 2 OOB shall suffice for 1 keepalive and 1 returning credits */
#define IBLND_OOB_CAPABLE(v) ((v) != IBLND_MSG_VERSION_1)
#define IBLND_OOB_MSGS(v) (IBLND_OOB_CAPABLE(v) ? 2 : 0)
/* max size of queued messages (inc hdr) */
#define IBLND_MSG_SIZE (4<<10)
/* max # of fragments supported. + 1 for unaligned case */
#define IBLND_MAX_RDMA_FRAGS (LNET_MAX_IOV + 1)
/************************/
/* derived constants... */
/* Pools (shared by connections on each CPT) */
/* These pools can grow at runtime, so don't need give a very large value */
#define IBLND_TX_POOL 256
#define IBLND_FMR_POOL 256
#define IBLND_FMR_POOL_FLUSH 192
/* RX messages (per connection) */
#define IBLND_RX_MSGS(c) \
((c->ibc_queue_depth) * 2 + IBLND_OOB_MSGS(c->ibc_version))
#define IBLND_RX_MSG_BYTES(c) (IBLND_RX_MSGS(c) * IBLND_MSG_SIZE)
#define IBLND_RX_MSG_PAGES(c) \
((IBLND_RX_MSG_BYTES(c) + PAGE_SIZE - 1) / PAGE_SIZE)
/* WRs and CQEs (per connection) */
#define IBLND_RECV_WRS(c) IBLND_RX_MSGS(c)
/* 2 = LNet msg + Transfer chain */
#define IBLND_CQ_ENTRIES(c) (IBLND_RECV_WRS(c) + kiblnd_send_wrs(c))
struct kib_hca_dev;
enum kib_dev_caps {
IBLND_DEV_CAPS_FASTREG_ENABLED = BIT(0),
IBLND_DEV_CAPS_FASTREG_GAPS_SUPPORT = BIT(1),
#ifdef HAVE_OFED_FMR_POOL_API
IBLND_DEV_CAPS_FMR_ENABLED = BIT(2),
#endif
};
#define IS_FAST_REG_DEV(dev) \
((dev)->ibd_dev_caps & IBLND_DEV_CAPS_FASTREG_ENABLED)
struct kib_dev {
struct list_head ibd_list; /* chain on kib_devs */
struct list_head ibd_fail_list; /* chain on kib_failed_devs */
struct sockaddr_storage ibd_addr; /* Interface network address */
/** IPoIB interface name */
char ibd_ifname[IFALIASZ];
int ibd_nnets; /* # nets extant */
time64_t ibd_next_failover;
/* # failover failures */
int ibd_failed_failover;
/* failover in progress */
unsigned int ibd_failover;
/* IPoIB interface is a bonding master */
unsigned int ibd_can_failover;
struct list_head ibd_nets;
struct kib_hca_dev *ibd_hdev;
enum kib_dev_caps ibd_dev_caps;
};
struct kib_hca_dev {
struct rdma_cm_id *ibh_cmid; /* listener cmid */
struct ib_device *ibh_ibdev; /* IB device */
int ibh_page_shift; /* page shift of current HCA */
int ibh_page_size; /* page size of current HCA */
__u64 ibh_page_mask; /* page mask of current HCA */
__u64 ibh_mr_size; /* size of MR */
int ibh_max_qp_wr; /* maximum work requests size */
struct ib_pd *ibh_pd; /* PD */
u8 ibh_port; /* port number */
struct ib_event_handler
ibh_event_handler; /* IB event handler */
int ibh_state; /* device status */
#define IBLND_DEV_PORT_DOWN 0
#define IBLND_DEV_PORT_ACTIVE 1
#define IBLND_DEV_FATAL 2
struct kib_dev *ibh_dev; /* owner */
atomic_t ibh_ref; /* refcount */
};
/** # of seconds to keep pool alive */
#define IBLND_POOL_DEADLINE 300
/** # of seconds to retry if allocation failed */
#define IBLND_POOL_RETRY 1
struct kib_pages {
int ibp_npages; /* # pages */
struct page *ibp_pages[]; /* page array */
};
struct kib_pool;
struct kib_poolset;
typedef int (*kib_ps_pool_create_t)(struct kib_poolset *ps,
int inc, struct kib_pool **pp_po);
typedef void (*kib_ps_pool_destroy_t)(struct kib_pool *po);
typedef void (*kib_ps_node_init_t)(struct kib_pool *po, struct list_head *node);
typedef void (*kib_ps_node_fini_t)(struct kib_pool *po, struct list_head *node);
struct kib_net;
#define IBLND_POOL_NAME_LEN 32
struct kib_poolset {
/* serialize */
spinlock_t ps_lock;
/* network it belongs to */
struct kib_net *ps_net;
/* pool set name */
char ps_name[IBLND_POOL_NAME_LEN];
/* list of pools */
struct list_head ps_pool_list;
/* failed pool list */
struct list_head ps_failed_pool_list;
/* time stamp for retry if failed to allocate */
time64_t ps_next_retry;
/* is allocating new pool */
int ps_increasing;
/* new pool size */
int ps_pool_size;
/* CPT id */
int ps_cpt;
/* create a new pool */
kib_ps_pool_create_t ps_pool_create;
/* destroy a pool */
kib_ps_pool_destroy_t ps_pool_destroy;
/* initialize new allocated node */
kib_ps_node_init_t ps_node_init;
/* finalize node */
kib_ps_node_fini_t ps_node_fini;
};
struct kib_pool {
/* chain on pool list */
struct list_head po_list;
/* pre-allocated node */
struct list_head po_free_list;
/* pool_set of this pool */
struct kib_poolset *po_owner;
/* deadline of this pool */
time64_t po_deadline;
/* # of elements in use */
int po_allocated;
/* pool is created on failed HCA */
int po_failed;
/* # of pre-allocated elements */
int po_size;
};
struct kib_tx_poolset {
struct kib_poolset tps_poolset; /* pool-set */
__u64 tps_next_tx_cookie; /* cookie of TX */
};
struct kib_tx_pool {
struct kib_pool tpo_pool; /* pool */
struct kib_hca_dev *tpo_hdev; /* device for this pool */
struct kib_tx *tpo_tx_descs; /* all the tx descriptors */
struct kib_pages *tpo_tx_pages; /* premapped tx msg pages */
};
struct kib_fmr_poolset {
spinlock_t fps_lock; /* serialize */
struct kib_net *fps_net; /* IB network */
struct list_head fps_pool_list; /* FMR pool list */
struct list_head fps_failed_pool_list; /* FMR pool list */
__u64 fps_version; /* validity stamp */
int fps_cpt; /* CPT id */
int fps_pool_size;
int fps_flush_trigger;
int fps_cache;
/* is allocating new pool */
int fps_increasing;
/* time stamp for retry if failed to allocate */
time64_t fps_next_retry;
};
struct kib_fast_reg_descriptor { /* For fast registration */
struct list_head frd_list;
struct ib_rdma_wr frd_inv_wr;
struct ib_reg_wr frd_fastreg_wr;
struct ib_mr *frd_mr;
bool frd_valid;
bool frd_posted;
};
struct kib_fmr_pool {
struct list_head fpo_list; /* chain on pool list */
struct kib_hca_dev *fpo_hdev; /* device for this pool */
struct kib_fmr_poolset *fpo_owner; /* owner of this pool */
#ifdef HAVE_OFED_FMR_POOL_API
union {
struct {
struct ib_fmr_pool *fpo_fmr_pool; /* IB FMR pool */
} fmr;
#endif
struct { /* For fast registration */
struct list_head fpo_pool_list;
int fpo_pool_size;
} fast_reg;
#ifdef HAVE_OFED_FMR_POOL_API
};
bool fpo_is_fmr; /* True if FMR pools allocated */
#endif
time64_t fpo_deadline; /* deadline of this pool */
int fpo_failed; /* fmr pool is failed */
int fpo_map_count; /* # of mapped FMR */
};
struct kib_fmr {
struct kib_fmr_pool *fmr_pool; /* pool of FMR */
#ifdef HAVE_OFED_FMR_POOL_API
struct ib_pool_fmr *fmr_pfmr; /* IB pool fmr */
#endif /* HAVE_OFED_FMR_POOL_API */
struct kib_fast_reg_descriptor *fmr_frd;
u32 fmr_key;
};
struct kib_net {
/* chain on struct kib_dev::ibd_nets */
struct list_head ibn_list;
__u64 ibn_incarnation;/* my epoch */
int ibn_init; /* initialisation state */
int ibn_shutdown; /* shutting down? */
atomic_t ibn_npeers; /* # peers extant */
atomic_t ibn_nconns; /* # connections extant */
struct kib_tx_poolset **ibn_tx_ps; /* tx pool-set */
struct kib_fmr_poolset **ibn_fmr_ps; /* fmr pool-set */
struct kib_dev *ibn_dev; /* underlying IB device */
struct lnet_ni *ibn_ni; /* LNet interface */
};
#define KIB_THREAD_SHIFT 16
#define KIB_THREAD_ID(cpt, tid) ((cpt) << KIB_THREAD_SHIFT | (tid))
#define KIB_THREAD_CPT(id) ((id) >> KIB_THREAD_SHIFT)
#define KIB_THREAD_TID(id) ((id) & ((1UL << KIB_THREAD_SHIFT) - 1))
struct kib_sched_info {
/* serialise */
spinlock_t ibs_lock;
/* schedulers sleep here */
wait_queue_head_t ibs_waitq;
/* conns to check for rx completions */
struct list_head ibs_conns;
/* number of scheduler threads */
int ibs_nthreads;
/* max allowed scheduler threads */
int ibs_nthreads_max;
int ibs_cpt; /* CPT id */
};
struct kib_data {
int kib_init; /* initialisation state */
int kib_shutdown; /* shut down? */
struct list_head kib_devs; /* IB devices extant */
/* list head of failed devices */
struct list_head kib_failed_devs;
/* schedulers sleep here */
wait_queue_head_t kib_failover_waitq;
atomic_t kib_nthreads; /* # live threads */
/* stabilize net/dev/peer_ni/conn ops */
rwlock_t kib_global_lock;
/* hash table of all my known peers */
DECLARE_HASHTABLE(kib_peers, IBLND_PEER_HASH_BITS);
/* the connd task (serialisation assertions) */
void *kib_connd;
/* connections to setup/teardown */
struct list_head kib_connd_conns;
/* connections with zero refcount */
struct list_head kib_connd_zombies;
/* connections to reconnect */
struct list_head kib_reconn_list;
/* peers wait for reconnection */
struct list_head kib_reconn_wait;
/*
* The second that peers are pulled out from \a kib_reconn_wait
* for reconnection.
*/
time64_t kib_reconn_sec;
/* connection daemon sleeps here */
wait_queue_head_t kib_connd_waitq;
spinlock_t kib_connd_lock; /* serialise */
struct ib_qp_attr kib_error_qpa; /* QP->ERROR */
/* percpt data for schedulers */
struct kib_sched_info **kib_scheds;
};
#define IBLND_INIT_NOTHING 0
#define IBLND_INIT_DATA 1
#define IBLND_INIT_ALL 2
struct kib_rx { /* receive message */
/* queue for attention */
struct list_head rx_list;
/* owning conn */
struct kib_conn *rx_conn;
/* # bytes received (-1 while posted) */
int rx_nob;
/* message buffer (host vaddr) */
struct kib_msg *rx_msg;
/* message buffer (I/O addr) */
__u64 rx_msgaddr;
/* for dma_unmap_single() */
DEFINE_DMA_UNMAP_ADDR(rx_msgunmap);
};
#define IBLND_POSTRX_DONT_POST 0 /* don't post */
#define IBLND_POSTRX_NO_CREDIT 1 /* post: no credits */
#define IBLND_POSTRX_PEER_CREDIT 2 /* post: give peer_ni back 1 credit */
#define IBLND_POSTRX_RSRVD_CREDIT 3 /* post: give myself back 1 reserved credit */
struct kib_tx { /* transmit message */
/* queue on idle_txs ibc_tx_queue etc. */
struct list_head tx_list;
/* pool I'm from */
struct kib_tx_pool *tx_pool;
/* owning conn */
struct kib_conn *tx_conn;
/* # tx callbacks outstanding */
short tx_sending;
/* queued for sending */
unsigned long tx_queued:1,
/* waiting for peer_ni */
tx_waiting:1,
/* force RDMA */
tx_gpu:1;
/* LNET completion status */
int tx_status;
/* health status of the transmit */
enum lnet_msg_hstatus tx_hstatus;
/* completion deadline */
ktime_t tx_deadline;
/* completion cookie */
__u64 tx_cookie;
/* lnet msgs to finalize on completion */
struct lnet_msg *tx_lntmsg[2];
/* message buffer (host vaddr) */
struct kib_msg *tx_msg;
/* message buffer (I/O addr) */
__u64 tx_msgaddr;
/* for dma_unmap_single() */
DEFINE_DMA_UNMAP_ADDR(tx_msgunmap);
/* # send work items */
int tx_nwrq;
/* # used scatter/gather elements */
int tx_nsge;
/* send work items... */
struct ib_rdma_wr *tx_wrq;
/* ...and their memory */
struct ib_sge *tx_sge;
/* rdma descriptor */
struct kib_rdma_desc *tx_rd;
/* # entries in... */
int tx_nfrags;
/* dma_map_sg descriptor */
struct scatterlist *tx_frags;
/* rdma phys page addrs */
__u64 *tx_pages;
/* gaps in fragments */
bool tx_gaps;
/* FMR */
struct kib_fmr tx_fmr;
/* dma direction */
int tx_dmadir;
};
struct kib_connvars {
/* connection-in-progress variables */
struct kib_msg cv_msg;
};
struct kib_conn {
/* scheduler information */
struct kib_sched_info *ibc_sched;
/* owning peer_ni */
struct kib_peer_ni *ibc_peer;
/* HCA bound on */
struct kib_hca_dev *ibc_hdev;
/* stash on peer_ni's conn list */
struct list_head ibc_list;
/* schedule for attention */
struct list_head ibc_sched_list;
/* version of connection */
__u16 ibc_version;
/* reconnect later */
__u16 ibc_reconnect:1;
/* which instance of the peer */
__u64 ibc_incarnation;
/* # users */
atomic_t ibc_refcount;
/* what's happening */
int ibc_state;
/* # uncompleted sends */
int ibc_nsends_posted;
/* # uncompleted NOOPs */
int ibc_noops_posted;
/* # credits I have */
int ibc_credits;
/* # credits to return */
int ibc_outstanding_credits;
/* # ACK/DONE msg credits */
int ibc_reserved_credits;
/* set on comms error */
int ibc_comms_error;
/* connections queue depth */
__u16 ibc_queue_depth;
/* connections max frags */
__u16 ibc_max_frags;
/* receive buffers owned */
unsigned int ibc_nrx:16;
/* scheduled for attention */
unsigned int ibc_scheduled:1;
/* CQ callback fired */
unsigned int ibc_ready:1;
/* time of last send */
ktime_t ibc_last_send;
/** link chain for kiblnd_check_conns only */
struct list_head ibc_connd_list;
/** rxs completed before ESTABLISHED */
struct list_head ibc_early_rxs;
/** IBLND_MSG_NOOPs for IBLND_MSG_VERSION_1 */
struct list_head ibc_tx_noops;
/* sends that need a credit */
struct list_head ibc_tx_queue;
/* sends that don't need a credit */
struct list_head ibc_tx_queue_nocred;
/* sends that need to reserve an ACK/DONE msg */
struct list_head ibc_tx_queue_rsrvd;
/* active tx awaiting completion */
struct list_head ibc_active_txs;
/* zombie tx awaiting done */
struct list_head ibc_zombie_txs;
/* serialise */
spinlock_t ibc_lock;
/* the rx descs */
struct kib_rx *ibc_rxs;
/* premapped rx msg pages */
struct kib_pages *ibc_rx_pages;
/* CM id */
struct rdma_cm_id *ibc_cmid;
/* completion queue */
struct ib_cq *ibc_cq;
/* in-progress connection state */
struct kib_connvars *ibc_connvars;
};
#define IBLND_CONN_INIT 0 /* being initialised */
#define IBLND_CONN_ACTIVE_CONNECT 1 /* active sending req */
#define IBLND_CONN_PASSIVE_WAIT 2 /* passive waiting for rtu */
#define IBLND_CONN_ESTABLISHED 3 /* connection established */
#define IBLND_CONN_CLOSING 4 /* being closed */
#define IBLND_CONN_DISCONNECTED 5 /* disconnected */
struct kib_peer_ni {
/* on peer_ni hash chain */
struct hlist_node ibp_list;
/* who's on the other end(s) */
struct lnet_nid ibp_nid;
/* LNet interface */
struct lnet_ni *ibp_ni;
/* all active connections */
struct list_head ibp_conns;
/* connections with an inflight active connect request */
struct list_head ibp_connreqs;
/* next connection to send on for round robin */
struct kib_conn *ibp_next_conn;
/* msgs waiting for a conn */
struct list_head ibp_tx_queue;
/* incarnation of peer_ni */
__u64 ibp_incarnation;
/* when (in seconds) I was last alive */
time64_t ibp_last_alive;
/* # users */
struct kref ibp_kref;
/* version of peer_ni */
__u16 ibp_version;
/* current passive connection attempts */
unsigned short ibp_accepting;
/* current active connection attempts */
unsigned short ibp_connecting;
/* reconnect this peer_ni later */
unsigned char ibp_reconnecting;
/* counter of how many times we triggered a conn race */
unsigned char ibp_races;
/* # consecutive reconnection attempts to this peer */
unsigned int ibp_reconnected;
/* errno on closing this peer_ni */
int ibp_error;
/* max map_on_demand */
__u16 ibp_max_frags;
/* max_peer_credits */
__u16 ibp_queue_depth;
/* reduced value which allows conn to be created if max fails */
__u16 ibp_queue_depth_mod;
/* Number of connections allocated. */
atomic_t ibp_nconns;
};
extern struct kib_data kiblnd_data;
extern void kiblnd_hdev_destroy(struct kib_hca_dev *hdev);
int kiblnd_msg_queue_size(int version, struct lnet_ni *ni);
#define RDMA_RESOLVE_TIMEOUT (5 * MSEC_PER_SEC) /* 5 seconds */
static inline int kiblnd_timeout(void)
{
return *kiblnd_tunables.kib_timeout ?: lnet_get_lnd_timeout();
}
/* lnd_connreq_timeout = lnd_timeout / 4 */
static inline int kiblnd_connreq_timeout_ms(void)
{
return max(RDMA_RESOLVE_TIMEOUT, kiblnd_timeout() * MSEC_PER_SEC / 4);
}
static inline int
kiblnd_concurrent_sends(int version, struct lnet_ni *ni)
{
struct lnet_ioctl_config_o2iblnd_tunables *tunables;
int concurrent_sends;
tunables = &ni->ni_lnd_tunables.lnd_tun_u.lnd_o2ib;
concurrent_sends = tunables->lnd_concurrent_sends;
if (version == IBLND_MSG_VERSION_1) {
if (concurrent_sends > IBLND_MSG_QUEUE_SIZE_V1 * 2)
return IBLND_MSG_QUEUE_SIZE_V1 * 2;
if (concurrent_sends < IBLND_MSG_QUEUE_SIZE_V1 / 2)
return IBLND_MSG_QUEUE_SIZE_V1 / 2;
}
return concurrent_sends;
}
static inline void
kiblnd_hdev_addref_locked(struct kib_hca_dev *hdev)
{
LASSERT(atomic_read(&hdev->ibh_ref) > 0);
atomic_inc(&hdev->ibh_ref);
}
static inline void
kiblnd_hdev_decref(struct kib_hca_dev *hdev)
{
LASSERT(atomic_read(&hdev->ibh_ref) > 0);
if (atomic_dec_and_test(&hdev->ibh_ref))
kiblnd_hdev_destroy(hdev);
}
static inline int
kiblnd_dev_can_failover(struct kib_dev *dev)
{
if (!list_empty(&dev->ibd_fail_list)) /* already scheduled */
return 0;
if (*kiblnd_tunables.kib_dev_failover == 0) /* disabled */
return 0;
if (*kiblnd_tunables.kib_dev_failover > 1) /* force failover */
return 1;
return dev->ibd_can_failover;
}
static inline void kiblnd_conn_addref(struct kib_conn *conn)
{
#ifdef O2IBLND_CONN_REFCOUNT_DEBUG
CDEBUG(D_NET, "conn[%p] (%d)++\n",
(conn), atomic_read(&(conn)->ibc_refcount));
#endif
atomic_inc(&(conn)->ibc_refcount);
}
static inline void kiblnd_conn_decref(struct kib_conn *conn)
{
unsigned long flags;
#ifdef O2IBLND_CONN_REFCOUNT_DEBUG
CDEBUG(D_NET, "conn[%p] (%d)--\n",
(conn), atomic_read(&(conn)->ibc_refcount));
#endif
LASSERT(atomic_read(&(conn)->ibc_refcount) > 0);
if (atomic_dec_and_test(&(conn)->ibc_refcount)) {
spin_lock_irqsave(&kiblnd_data.kib_connd_lock, flags);
list_add_tail(&(conn)->ibc_list,
&kiblnd_data.kib_connd_zombies);
wake_up(&kiblnd_data.kib_connd_waitq);
spin_unlock_irqrestore(&kiblnd_data.kib_connd_lock, flags);
}
}
void kiblnd_destroy_peer(struct kref *kref);
static inline void kiblnd_peer_addref(struct kib_peer_ni *peer_ni)
{
CDEBUG(D_NET, "peer_ni[%p] -> %s (%d)++\n",
peer_ni, libcfs_nidstr(&peer_ni->ibp_nid),
kref_read(&peer_ni->ibp_kref));
kref_get(&(peer_ni)->ibp_kref);
}
static inline void kiblnd_peer_decref(struct kib_peer_ni *peer_ni)
{
CDEBUG(D_NET, "peer_ni[%p] -> %s (%d)--\n",
peer_ni, libcfs_nidstr(&peer_ni->ibp_nid),
kref_read(&peer_ni->ibp_kref));
kref_put(&peer_ni->ibp_kref, kiblnd_destroy_peer);
}
static inline bool
kiblnd_peer_connecting(struct kib_peer_ni *peer_ni)
{
return peer_ni->ibp_connecting != 0 ||
peer_ni->ibp_reconnecting != 0 ||
peer_ni->ibp_accepting != 0;
}
static inline bool
kiblnd_peer_idle(struct kib_peer_ni *peer_ni)
{
return !kiblnd_peer_connecting(peer_ni) && list_empty(&peer_ni->ibp_conns);
}
static inline int
kiblnd_peer_active(struct kib_peer_ni *peer_ni)
{
/* Am I in the peer_ni hash table? */
return !hlist_unhashed(&peer_ni->ibp_list);
}
static inline struct kib_conn *
kiblnd_get_conn_locked(struct kib_peer_ni *peer_ni)
{
struct list_head *next;
LASSERT(!list_empty(&peer_ni->ibp_conns));
/* Advance to next connection, be sure to skip the head node */
if (!peer_ni->ibp_next_conn ||
peer_ni->ibp_next_conn->ibc_list.next == &peer_ni->ibp_conns)
next = peer_ni->ibp_conns.next;
else
next = peer_ni->ibp_next_conn->ibc_list.next;
peer_ni->ibp_next_conn = list_entry(next, struct kib_conn, ibc_list);
return peer_ni->ibp_next_conn;
}
static inline int
kiblnd_send_keepalive(struct kib_conn *conn)
{
s64 keepalive_ns = *kiblnd_tunables.kib_keepalive * NSEC_PER_SEC;
return (*kiblnd_tunables.kib_keepalive > 0) &&
ktime_after(ktime_get(),
ktime_add_ns(conn->ibc_last_send, keepalive_ns));
}
/* when to return credits eagerly */
static inline int
kiblnd_credits_highwater(struct lnet_ioctl_config_o2iblnd_tunables *t,
struct lnet_ioctl_config_lnd_cmn_tunables *nt,
struct kib_conn *conn)
{
int credits_hiw = IBLND_CREDIT_HIGHWATER_V1;
if ((conn->ibc_version) == IBLND_MSG_VERSION_1)
return credits_hiw;
/* if queue depth is negotiated down, calculate hiw proportionally */
credits_hiw = (conn->ibc_queue_depth * t->lnd_peercredits_hiw) /
nt->lct_peer_tx_credits;
return credits_hiw;
}
static inline int
kiblnd_need_noop(struct kib_conn *conn)
{
struct lnet_ni *ni = conn->ibc_peer->ibp_ni;
struct lnet_ioctl_config_o2iblnd_tunables *tunables;
struct lnet_ioctl_config_lnd_cmn_tunables *net_tunables;
LASSERT(conn->ibc_state >= IBLND_CONN_ESTABLISHED);
tunables = &ni->ni_lnd_tunables.lnd_tun_u.lnd_o2ib;
net_tunables = &ni->ni_net->net_tunables;
if (conn->ibc_outstanding_credits <
kiblnd_credits_highwater(tunables, net_tunables, conn) &&
!kiblnd_send_keepalive(conn))
return 0; /* No need to send NOOP */
if (IBLND_OOB_CAPABLE(conn->ibc_version)) {
if (!list_empty(&conn->ibc_tx_queue_nocred))
return 0; /* NOOP can be piggybacked */
/* No tx to piggyback NOOP onto or no credit to send a tx */
return (list_empty(&conn->ibc_tx_queue) ||
conn->ibc_credits == 0);
}
if (!list_empty(&conn->ibc_tx_noops) || /* NOOP is queued */
!list_empty(&conn->ibc_tx_queue_nocred) || /* piggyback NOOP */
conn->ibc_credits == 0) /* no credit */
return 0;
if (conn->ibc_credits == 1 && /* last credit reserved for */
conn->ibc_outstanding_credits == 0) /* giving back credits */
return 0;
/* No tx to piggyback NOOP onto or no credit to send a tx */
return (list_empty(&conn->ibc_tx_queue) || conn->ibc_credits == 1);
}
static inline void
kiblnd_abort_receives(struct kib_conn *conn)
{
ib_modify_qp(conn->ibc_cmid->qp,
&kiblnd_data.kib_error_qpa, IB_QP_STATE);
}
static inline const char *
kiblnd_queue2str(struct kib_conn *conn, struct list_head *q)
{
if (q == &conn->ibc_tx_queue)
return "tx_queue";
if (q == &conn->ibc_tx_queue_rsrvd)
return "tx_queue_rsrvd";
if (q == &conn->ibc_tx_queue_nocred)
return "tx_queue_nocred";
if (q == &conn->ibc_active_txs)
return "active_txs";
LBUG();
return NULL;
}
/* CAVEAT EMPTOR: We rely on descriptor alignment to allow us to use the
* lowest bits of the work request id to stash the work item type. */
#define IBLND_WID_INVAL 0
#define IBLND_WID_TX 1
#define IBLND_WID_RX 2
#define IBLND_WID_RDMA 3
#define IBLND_WID_MR 4
#define IBLND_WID_MASK 7UL
static inline __u64
kiblnd_ptr2wreqid (void *ptr, int type)
{
unsigned long lptr = (unsigned long)ptr;
LASSERT((lptr & IBLND_WID_MASK) == 0);
LASSERT((type & ~IBLND_WID_MASK) == 0);
return (__u64)(lptr | type);
}
static inline void *
kiblnd_wreqid2ptr (__u64 wreqid)
{
return (void *)(((unsigned long)wreqid) & ~IBLND_WID_MASK);
}
static inline int
kiblnd_wreqid2type (__u64 wreqid)
{
return (wreqid & IBLND_WID_MASK);
}
static inline void
kiblnd_set_conn_state(struct kib_conn *conn, int state)
{
conn->ibc_state = state;
smp_mb();
}
static inline void
kiblnd_init_msg(struct kib_msg *msg, int type, int body_nob)
{
msg->ibm_type = type;
msg->ibm_nob = offsetof(struct kib_msg, ibm_u) + body_nob;
}
static inline int
kiblnd_rd_size(struct kib_rdma_desc *rd)
{
int size;
int i;
for (i = size = 0; i < rd->rd_nfrags; i++)
size += rd->rd_frags[i].rf_nob;
return size;
}
static inline __u64
kiblnd_rd_frag_addr(struct kib_rdma_desc *rd, int index)
{
return rd->rd_frags[index].rf_addr;
}
static inline int
kiblnd_rd_frag_size(struct kib_rdma_desc *rd, int index)
{
return rd->rd_frags[index].rf_nob;
}
static inline __u32
kiblnd_rd_frag_key(struct kib_rdma_desc *rd, int index)
{
return rd->rd_key;
}
static inline int
kiblnd_rd_consume_frag(struct kib_rdma_desc *rd, int index, __u32 nob)
{
if (nob < rd->rd_frags[index].rf_nob) {
rd->rd_frags[index].rf_addr += nob;
rd->rd_frags[index].rf_nob -= nob;
} else {
index++;
}
return index;
}
static inline int
kiblnd_rd_msg_size(struct kib_rdma_desc *rd, int msgtype, int n)
{
return msgtype == IBLND_MSG_GET_REQ ?
offsetof(struct kib_get_msg, ibgm_rd.rd_frags[n]) :
offsetof(struct kib_putack_msg, ibpam_rd.rd_frags[n]);
}
static inline __u64
kiblnd_dma_mapping_error(struct ib_device *dev, u64 dma_addr)
{
return ib_dma_mapping_error(dev, dma_addr);
}
static inline __u64 kiblnd_dma_map_single(struct ib_device *dev,
void *msg, size_t size,
enum dma_data_direction direction)
{
return ib_dma_map_single(dev, msg, size, direction);
}
static inline void kiblnd_dma_unmap_single(struct ib_device *dev,
__u64 addr, size_t size,
enum dma_data_direction direction)
{
ib_dma_unmap_single(dev, addr, size, direction);
}
#define KIBLND_UNMAP_ADDR_SET(p, m, a) do {} while (0)
#define KIBLND_UNMAP_ADDR(p, m, a) (a)
static inline
int kiblnd_dma_map_sg(struct kib_hca_dev *hdev, struct kib_tx *tx)
{
struct scatterlist *sg = tx->tx_frags;
int nents = tx->tx_nfrags;
enum dma_data_direction direction = tx->tx_dmadir;
if (tx->tx_gpu)
return lnet_rdma_map_sg_attrs(hdev->ibh_ibdev->dma_device,
sg, nents, direction);
return ib_dma_map_sg(hdev->ibh_ibdev, sg, nents, direction);
}
static inline
void kiblnd_dma_unmap_sg(struct kib_hca_dev *hdev, struct kib_tx *tx)
{
struct scatterlist *sg = tx->tx_frags;
int nents = tx->tx_nfrags;
enum dma_data_direction direction = tx->tx_dmadir;
if (tx->tx_gpu)
lnet_rdma_unmap_sg(hdev->ibh_ibdev->dma_device,
sg, nents, direction);
else
ib_dma_unmap_sg(hdev->ibh_ibdev, sg, nents, direction);
}
#ifndef HAVE_OFED_IB_SG_DMA_ADDRESS
#include <linux/scatterlist.h>
#define ib_sg_dma_address(dev, sg) sg_dma_address(sg)
#define ib_sg_dma_len(dev, sg) sg_dma_len(sg)
#endif
static inline __u64 kiblnd_sg_dma_address(struct ib_device *dev,
struct scatterlist *sg)
{
return ib_sg_dma_address(dev, sg);
}
static inline unsigned int kiblnd_sg_dma_len(struct ib_device *dev,
struct scatterlist *sg)
{
return ib_sg_dma_len(dev, sg);
}
#ifndef HAVE_OFED_RDMA_CONNECT_LOCKED
#define rdma_connect_locked(cmid, cpp) rdma_connect(cmid, cpp)
#endif
/* XXX We use KIBLND_CONN_PARAM(e) as writable buffer, it's not strictly
* right because OFED1.2 defines it as const, to use it we have to add
* (void *) cast to overcome "const" */
#define KIBLND_CONN_PARAM(e) ((e)->param.conn.private_data)
#define KIBLND_CONN_PARAM_LEN(e) ((e)->param.conn.private_data_len)
void kiblnd_abort_txs(struct kib_conn *conn, struct list_head *txs);
void kiblnd_map_rx_descs(struct kib_conn *conn);
void kiblnd_unmap_rx_descs(struct kib_conn *conn);
void kiblnd_pool_free_node(struct kib_pool *pool, struct list_head *node);
struct list_head *kiblnd_pool_alloc_node(struct kib_poolset *ps);
int kiblnd_fmr_pool_map(struct kib_fmr_poolset *fps, struct kib_tx *tx,
struct kib_rdma_desc *rd, u32 nob, u64 iov,
struct kib_fmr *fmr);
void kiblnd_fmr_pool_unmap(struct kib_fmr *fmr, int status);
int kiblnd_tunables_setup(struct lnet_lnd_tunables *lnd_tunables,
struct lnet_ioctl_config_lnd_cmn_tunables *net_tunables);
int kiblnd_tunables_init(void);
int kiblnd_connd(void *arg);
int kiblnd_scheduler(void *arg);
#define kiblnd_thread_start(fn, data, namefmt, arg...) \
({ \
struct task_struct *__task = kthread_run(fn, data, \
namefmt, ##arg); \
if (!IS_ERR(__task)) \
atomic_inc(&kiblnd_data.kib_nthreads); \
PTR_ERR_OR_ZERO(__task); \
})
int kiblnd_failover_thread(void *arg);
int kiblnd_alloc_pages(struct kib_pages **pp, int cpt, int npages);
int kiblnd_cm_callback(struct rdma_cm_id *cmid, struct rdma_cm_event *event);
int kiblnd_translate_mtu(int value);
int kiblnd_dev_failover(struct kib_dev *dev, struct net *ns);
int kiblnd_create_peer(struct lnet_ni *ni, struct kib_peer_ni **peerp,
struct lnet_nid *nid);
bool kiblnd_reconnect_peer(struct kib_peer_ni *peer);
void kiblnd_destroy_dev(struct kib_dev *dev);
void kiblnd_unlink_peer_locked(struct kib_peer_ni *peer_ni);
struct kib_peer_ni *kiblnd_find_peer_locked(struct lnet_ni *ni,
struct lnet_nid *nid);
int kiblnd_close_stale_conns_locked(struct kib_peer_ni *peer_ni,
int version, u64 incarnation);
int kiblnd_close_peer_conns_locked(struct kib_peer_ni *peer_ni, int why);
struct kib_conn *kiblnd_create_conn(struct kib_peer_ni *peer_ni,
struct rdma_cm_id *cmid,
int state, int version);
void kiblnd_destroy_conn(struct kib_conn *conn);
void kiblnd_close_conn(struct kib_conn *conn, int error);
void kiblnd_close_conn_locked(struct kib_conn *conn, int error);
void kiblnd_launch_tx(struct lnet_ni *ni, struct kib_tx *tx,
struct lnet_nid *nid);
void kiblnd_txlist_done(struct list_head *txlist, int status,
enum lnet_msg_hstatus hstatus);
void kiblnd_qp_event(struct ib_event *event, void *arg);
void kiblnd_cq_event(struct ib_event *event, void *arg);
void kiblnd_cq_completion(struct ib_cq *cq, void *arg);
void kiblnd_pack_msg(struct lnet_ni *ni, struct kib_msg *msg, int version,
int credits, struct lnet_nid *dstnid, u64 dststamp);
int kiblnd_unpack_msg(struct kib_msg *msg, int nob);
int kiblnd_post_rx(struct kib_rx *rx, int credit);
int kiblnd_send(struct lnet_ni *ni, void *private, struct lnet_msg *lntmsg);
int kiblnd_recv(struct lnet_ni *ni, void *private, struct lnet_msg *lntmsg,
int delayed, struct iov_iter *to, unsigned int rlen);
unsigned int kiblnd_get_dev_prio(struct lnet_ni *ni, unsigned int dev_idx);
#define kiblnd_dump_conn_dbg(conn) \
({ \
if (conn && conn->ibc_cmid) \
CDEBUG(D_NET, "conn %p state %d nposted %d/%d c/o/r %d/%d/%d ce %d : cm_id %p qp_num 0x%x device_name %s\n", \
conn, \
conn->ibc_state, \
conn->ibc_noops_posted, \
conn->ibc_nsends_posted, \
conn->ibc_credits, \
conn->ibc_outstanding_credits, \
conn->ibc_reserved_credits, \
conn->ibc_comms_error, \
conn->ibc_cmid, \
conn->ibc_cmid->qp ? conn->ibc_cmid->qp->qp_num : 0, \
conn->ibc_cmid->qp ? (conn->ibc_cmid->qp->device ? dev_name(&conn->ibc_cmid->qp->device->dev) : "NULL") : "NULL"); \
else if (conn) \
CDEBUG(D_NET, "conn %p state %d nposted %d/%d c/o/r %d/%d/%d ce %d : cm_id NULL\n", \
conn, \
conn->ibc_state, \
conn->ibc_noops_posted, \
conn->ibc_nsends_posted, \
conn->ibc_credits, \
conn->ibc_outstanding_credits, \
conn->ibc_reserved_credits, \
conn->ibc_comms_error \
); \
})
