LU-13972 o2iblnd: Don't retry indefinitely

[fs/lustre-release.git] / lnet / klnds / o2iblnd / o2iblnd.h

diff --git a/lnet/klnds/o2iblnd/o2iblnd.h b/lnet/klnds/o2iblnd/o2iblnd.h
index 86e86a7..8ff3b8c 100644 (file)
--- a/lnet/klnds/o2iblnd/o2iblnd.h
+++ b/lnet/klnds/o2iblnd/o2iblnd.h
@@ -34,6 +34,17 @@
  * Author: Eric Barton <eric@bartonsoftware.com>
  */
 
+#include <linux/module.h>
+#include <linux/kernel.h>
+
+#if defined(NEED_LOCKDEP_IS_HELD_DISCARD_CONST) \
+ && defined(CONFIG_LOCKDEP) \
+ && defined(lockdep_is_held)
+#undef lockdep_is_held
+       #define lockdep_is_held(lock) \
+               lock_is_held((struct lockdep_map *)&(lock)->dep_map)
+#endif
+
 #ifdef HAVE_COMPAT_RDMA
 #include <linux/compat-2.6.h>
 
@@ -41,10 +52,11 @@
 #undef NEED_KTIME_GET_REAL_NS
 #endif
 
+/* MOFED has its own bitmap_alloc backport */
+#define HAVE_BITMAP_ALLOC 1
+
 #endif
 
-#include <linux/module.h>
-#include <linux/kernel.h>
 #include <linux/kthread.h>
 #include <linux/mm.h>
 #include <linux/string.h>
@@ -79,8 +91,6 @@
 #include <lnet/lib-lnet.h>
 
 #define IBLND_PEER_HASH_SIZE           101     /* # peer_ni lists */
-/* # scheduler loops before reschedule */
-#define IBLND_RESCHED                  100
 
 #define IBLND_N_SCHED                  2
 #define IBLND_N_SCHED_HIGH             4
@@ -117,16 +127,15 @@ extern struct kib_tunables  kiblnd_tunables;
                        min(t->lnd_peercredits_hiw, (__u32)conn->ibc_queue_depth - 1))
 
 #ifdef HAVE_RDMA_CREATE_ID_5ARG
-# define kiblnd_rdma_create_id(ns, cb, dev, ps, qpt) rdma_create_id(ns, cb, \
-                                                                   dev, ps, \
-                                                                   qpt)
+# define kiblnd_rdma_create_id(ns, cb, dev, ps, qpt) \
+        rdma_create_id((ns) ? (ns) : &init_net, cb, dev, ps, qpt)
 #else
 # ifdef HAVE_RDMA_CREATE_ID_4ARG
-#  define kiblnd_rdma_create_id(ns, cb, dev, ps, qpt) rdma_create_id(cb, dev, \
-                                                                    ps, qpt)
+#  define kiblnd_rdma_create_id(ns, cb, dev, ps, qpt) \
+         rdma_create_id(cb, dev, ps, qpt)
 # else
-#  define kiblnd_rdma_create_id(ns, cb, dev, ps, qpt) rdma_create_id(cb, dev, \
-                                                                    ps)
+#  define kiblnd_rdma_create_id(ns, cb, dev, ps, qpt) \
+         rdma_create_id(cb, dev, ps)
 # endif
 #endif
 
@@ -205,6 +214,13 @@ struct kib_hca_dev {
        struct ib_mr        *ibh_mrs;           /* global MR */
 #endif
        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 */
 };
@@ -374,6 +390,7 @@ struct kib_net {
        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
@@ -751,6 +768,8 @@ struct kib_peer_ni {
        unsigned char           ibp_races;
        /* # consecutive reconnection attempts to this peer */
        unsigned int            ibp_reconnected;
+       /* number of total active retries */
+       unsigned int            ibp_retries;
        /* errno on closing this peer_ni */
        int                     ibp_error;
        /* max map_on_demand */
@@ -778,6 +797,12 @@ extern void kiblnd_hdev_destroy(struct kib_hca_dev *hdev);
 
 int kiblnd_msg_queue_size(int version, struct lnet_ni *ni);
 
+static inline int kiblnd_timeout(void)
+{
+       return *kiblnd_tunables.kib_timeout ? *kiblnd_tunables.kib_timeout :
+               lnet_get_lnd_timeout();
+}
+
 static inline int
 kiblnd_concurrent_sends(int version, struct lnet_ni *ni)
 {
@@ -1053,7 +1078,7 @@ kiblnd_rd_frag_addr(struct kib_rdma_desc *rd, int index)
         return rd->rd_frags[index].rf_addr;
 }
 
-static inline __u32
+static inline int
 kiblnd_rd_frag_size(struct kib_rdma_desc *rd, int index)
 {
         return rd->rd_frags[index].rf_nob;
@@ -1210,7 +1235,7 @@ 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, unsigned int niov, struct kvec *iov,
-               lnet_kiov_t *kiov, unsigned int offset, unsigned int mlen,
+               int delayed, unsigned int niov,
+               struct bio_vec *kiov, unsigned int offset, unsigned int mlen,
                unsigned int rlen);