Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma

Pull rdma updates from Doug Ledford:
 "First set of updates for 4.11 kernel merge window

   - Add new Broadcom bnxt_re RoCE driver
   - rxe driver updates
   - ioctl cleanups
   - ETH_P_IBOE declaration cleanup
   - IPoIB changes
   - Add port state cache
   - Allow srpt driver to accept guids as port names in config
   - Update to hfi1 driver
   - Update to srp driver
   - Lots of misc minor changes all over"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (114 commits)
  RDMA/bnxt_re: fix for "bnxt_en: Update to firmware interface spec 1.7.0."
  rdma_cm: fail iwarp accepts w/o connection params
  IB/srp: Drain the send queue before destroying a QP
  IB/core: Add support for draining IB_POLL_DIRECT completion queues
  IB/srp: Improve an error path
  IB/srp: Make a diagnostic message more informative
  IB/srp: Document locking conventions
  IB/srp: Fix race conditions related to task management
  IB/srp: Avoid that duplicate responses trigger a kernel bug
  IB/SRP: Avoid using IB_MR_TYPE_SG_GAPS
  RDMA/qedr: Fix some error handling
  RDMA/bnxt_re: add DCB dependency
  IB/hns: include linux/module.h
  IB/vmw_pvrdma: Expose vendor error to ULPs
  vmw_pvrdma: switch to pci_alloc_irq_vectors
  IB/hfi1: use size_t for passing array length
  IB/ipoib: Remove redudant label
  IB/ipoib: remove the unnecessary memory free
  IB/mthca: switch to pci_alloc_irq_vectors
  IB/hfi1: Code reuse with memdup_copy
  ...

7 files changed, 158 insertions, 12 deletions

diff --git a/include/rdma/ib_cache.h b/include/rdma/ib_cache.h
index e30f19bd4a41..385ec88ee9e5 100644
--- a/include/rdma/ib_cache.h
+++ b/include/rdma/ib_cache.h
@@ -165,4 +165,17 @@ int ib_get_cached_lmc(struct ib_device *device,
 		      u8                port_num,
 		      u8                *lmc);
 
+/**
+ * ib_get_cached_port_state - Returns a cached port state table entry
+ * @device: The device to query.
+ * @port_num: The port number of the device to query.
+ * @port_state: port_state for the specified port for that device.
+ *
+ * ib_get_cached_port_state() fetches the specified port_state table entry stored in
+ * the local software cache.
+ */
+int ib_get_cached_port_state(struct ib_device *device,
+			      u8                port_num,
+			      enum ib_port_state *port_active);
+
 #endif /* _IB_CACHE_H */
diff --git a/include/rdma/ib_hdrs.h b/include/rdma/ib_hdrs.h
index 408439fe911e..c755325f0831 100644
--- a/include/rdma/ib_hdrs.h
+++ b/include/rdma/ib_hdrs.h
@@ -75,6 +75,12 @@
 #define IB_GRH_FLOW_SHIFT	0
 #define IB_GRH_NEXT_HDR		0x1B
 
+#define IB_AETH_CREDIT_SHIFT	24
+#define IB_AETH_CREDIT_MASK	0x1F
+#define IB_AETH_CREDIT_INVAL	0x1F
+#define IB_AETH_NAK_SHIFT	29
+#define IB_MSN_MASK		0xFFFFFF
+
 struct ib_reth {
 	__be64 vaddr;        /* potentially unaligned */
 	__be32 rkey;
diff --git a/include/rdma/ib_sa.h b/include/rdma/ib_sa.h
index 5ee7aab95eb8..fd0e53219f93 100644
--- a/include/rdma/ib_sa.h
+++ b/include/rdma/ib_sa.h
@@ -153,12 +153,12 @@ struct ib_sa_path_rec {
 	union ib_gid sgid;
 	__be16       dlid;
 	__be16       slid;
-	int          raw_traffic;
+	u8           raw_traffic;
 	/* reserved */
 	__be32       flow_label;
 	u8           hop_limit;
 	u8           traffic_class;
-	int          reversible;
+	u8           reversible;
 	u8           numb_path;
 	__be16       pkey;
 	__be16       qos_class;
@@ -220,7 +220,7 @@ struct ib_sa_mcmember_rec {
 	u8           hop_limit;
 	u8           scope;
 	u8           join_state;
-	int          proxy_join;
+	u8           proxy_join;
 };
 
 /* Service Record Component Mask Sec 15.2.5.14 Ver 1.1	*/
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index b567e4452a47..8c61532cf521 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -1789,12 +1789,17 @@ enum ib_mad_result {
 
 #define IB_DEVICE_NAME_MAX 64
 
+struct ib_port_cache {
+	struct ib_pkey_cache  *pkey;
+	struct ib_gid_table   *gid;
+	u8                     lmc;
+	enum ib_port_state     port_state;
+};
+
 struct ib_cache {
 	rwlock_t                lock;
 	struct ib_event_handler event_handler;
-	struct ib_pkey_cache  **pkey_cache;
-	struct ib_gid_table   **gid_cache;
-	u8                     *lmc_cache;
+	struct ib_port_cache   *ports;
 };
 
 struct ib_dma_mapping_ops {
@@ -2289,6 +2294,13 @@ static inline u8 rdma_end_port(const struct ib_device *device)
 	return rdma_cap_ib_switch(device) ? 0 : device->phys_port_cnt;
 }
 
+static inline int rdma_is_port_valid(const struct ib_device *device,
+				     unsigned int port)
+{
+	return (port >= rdma_start_port(device) &&
+		port <= rdma_end_port(device));
+}
+
 static inline bool rdma_protocol_ib(const struct ib_device *device, u8 port_num)
 {
 	return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_IB;
diff --git a/include/rdma/rdma_vt.h b/include/rdma/rdma_vt.h
index 861e23eaebda..8fc1ca7b6f23 100644
--- a/include/rdma/rdma_vt.h
+++ b/include/rdma/rdma_vt.h
@@ -164,7 +164,7 @@ struct rvt_driver_params {
 /* Protection domain */
 struct rvt_pd {
 	struct ib_pd ibpd;
-	int user;               /* non-zero if created from user space */
+	bool user;
 };
 
 /* Address handle */
@@ -335,6 +335,8 @@ struct rvt_driver_provided {
 	/* Notify driver a mad agent has been removed */
 	void (*notify_free_mad_agent)(struct rvt_dev_info *rdi, int port_idx);
 
+	/* Notify driver to restart rc */
+	void (*notify_restart_rc)(struct rvt_qp *qp, u32 psn, int wait);
 };
 
 struct rvt_dev_info {
@@ -483,6 +485,23 @@ static inline struct rvt_qp *rvt_lookup_qpn(struct rvt_dev_info *rdi,
 	return qp;
 }
 
+/**
+ * rvt_mod_retry_timer - mod a retry timer
+ * @qp - the QP
+ * Modify a potentially already running retry timer
+ */
+static inline void rvt_mod_retry_timer(struct rvt_qp *qp)
+{
+	struct ib_qp *ibqp = &qp->ibqp;
+	struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
+
+	lockdep_assert_held(&qp->s_lock);
+	qp->s_flags |= RVT_S_TIMER;
+	/* 4.096 usec. * (1 << qp->timeout) */
+	mod_timer(&qp->s_timer, jiffies + qp->timeout_jiffies +
+		  rdi->busy_jiffies);
+}
+
 struct rvt_dev_info *rvt_alloc_device(size_t size, int nports);
 void rvt_dealloc_device(struct rvt_dev_info *rdi);
 int rvt_register_device(struct rvt_dev_info *rvd);
diff --git a/include/rdma/rdmavt_mr.h b/include/rdma/rdmavt_mr.h
index de59de28b6a2..f418bd5571a5 100644
--- a/include/rdma/rdmavt_mr.h
+++ b/include/rdma/rdmavt_mr.h
@@ -52,6 +52,7 @@
  * For Memory Regions. This stuff should probably be moved into rdmavt/mr.h once
  * drivers no longer need access to the MR directly.
  */
+#include <linux/percpu-refcount.h>
 
 /*
  * A segment is a linear region of low physical memory.
@@ -79,11 +80,11 @@ struct rvt_mregion {
 	int access_flags;
 	u32 max_segs;           /* number of rvt_segs in all the arrays */
 	u32 mapsz;              /* size of the map array */
+	atomic_t lkey_invalid;	/* true if current lkey is invalid */
 	u8  page_shift;         /* 0 - non unform/non powerof2 sizes */
 	u8  lkey_published;     /* in global table */
-	atomic_t lkey_invalid;	/* true if current lkey is invalid */
+	struct percpu_ref refcount;
 	struct completion comp; /* complete when refcount goes to zero */
-	atomic_t refcount;
 	struct rvt_segarray *map[0];    /* the segments */
 };
 
@@ -123,13 +124,12 @@ struct rvt_sge_state {
 
 static inline void rvt_put_mr(struct rvt_mregion *mr)
 {
-	if (unlikely(atomic_dec_and_test(&mr->refcount)))
-		complete(&mr->comp);
+	percpu_ref_put(&mr->refcount);
 }
 
 static inline void rvt_get_mr(struct rvt_mregion *mr)
 {
-	atomic_inc(&mr->refcount);
+	percpu_ref_get(&mr->refcount);
 }
 
 static inline void rvt_put_ss(struct rvt_sge_state *ss)
@@ -141,4 +141,54 @@ static inline void rvt_put_ss(struct rvt_sge_state *ss)
 	}
 }
 
+static inline u32 rvt_get_sge_length(struct rvt_sge *sge, u32 length)
+{
+	u32 len = sge->length;
+
+	if (len > length)
+		len = length;
+	if (len > sge->sge_length)
+		len = sge->sge_length;
+
+	return len;
+}
+
+static inline void rvt_update_sge(struct rvt_sge_state *ss, u32 length,
+				  bool release)
+{
+	struct rvt_sge *sge = &ss->sge;
+
+	sge->vaddr += length;
+	sge->length -= length;
+	sge->sge_length -= length;
+	if (sge->sge_length == 0) {
+		if (release)
+			rvt_put_mr(sge->mr);
+		if (--ss->num_sge)
+			*sge = *ss->sg_list++;
+	} else if (sge->length == 0 && sge->mr->lkey) {
+		if (++sge->n >= RVT_SEGSZ) {
+			if (++sge->m >= sge->mr->mapsz)
+				return;
+			sge->n = 0;
+		}
+		sge->vaddr = sge->mr->map[sge->m]->segs[sge->n].vaddr;
+		sge->length = sge->mr->map[sge->m]->segs[sge->n].length;
+	}
+}
+
+static inline void rvt_skip_sge(struct rvt_sge_state *ss, u32 length,
+				bool release)
+{
+	struct rvt_sge *sge = &ss->sge;
+
+	while (length) {
+		u32 len = rvt_get_sge_length(sge, length);
+
+		WARN_ON_ONCE(len == 0);
+		rvt_update_sge(ss, len, release);
+		length -= len;
+	}
+}
+
 #endif          /* DEF_RDMAVT_INCMRH */
diff --git a/include/rdma/rdmavt_qp.h b/include/rdma/rdmavt_qp.h
index f3dbd157ae5c..f3816396c76a 100644
--- a/include/rdma/rdmavt_qp.h
+++ b/include/rdma/rdmavt_qp.h
@@ -144,6 +144,8 @@
 #define RVT_FLUSH_RECV			0x40
 #define RVT_PROCESS_OR_FLUSH_SEND \
 	(RVT_PROCESS_SEND_OK | RVT_FLUSH_SEND)
+#define RVT_SEND_OR_FLUSH_OR_RECV_OK \
+	(RVT_PROCESS_SEND_OK | RVT_FLUSH_SEND | RVT_PROCESS_RECV_OK)
 
 /*
  * Internal send flags
@@ -370,6 +372,7 @@ struct rvt_qp {
 
 	struct rvt_sge_state s_ack_rdma_sge;
 	struct timer_list s_timer;
+	struct hrtimer s_rnr_timer;
 
 	atomic_t local_ops_pending; /* number of fast_reg/local_inv reqs */
 
@@ -467,6 +470,15 @@ static inline struct rvt_rwqe *rvt_get_rwqe_ptr(struct rvt_rq *rq, unsigned n)
 }
 
 /**
+ * rvt_is_user_qp - return if this is user mode QP
+ * @qp - the target QP
+ */
+static inline bool rvt_is_user_qp(struct rvt_qp *qp)
+{
+	return !!qp->pid;
+}
+
+/**
  * rvt_get_qp - get a QP reference
  * @qp - the QP to hold
  */
@@ -582,6 +594,32 @@ static inline void rvt_qp_swqe_complete(
 	}
 }
 
+/*
+ * Compare the lower 24 bits of the msn values.
+ * Returns an integer <, ==, or > than zero.
+ */
+static inline int rvt_cmp_msn(u32 a, u32 b)
+{
+	return (((int)a) - ((int)b)) << 8;
+}
+
+/**
+ * rvt_compute_aeth - compute the AETH (syndrome + MSN)
+ * @qp: the queue pair to compute the AETH for
+ *
+ * Returns the AETH.
+ */
+__be32 rvt_compute_aeth(struct rvt_qp *qp);
+
+/**
+ * rvt_get_credit - flush the send work queue of a QP
+ * @qp: the qp who's send work queue to flush
+ * @aeth: the Acknowledge Extended Transport Header
+ *
+ * The QP s_lock should be held.
+ */
+void rvt_get_credit(struct rvt_qp *qp, u32 aeth);
+
 /**
  * @qp - the qp pair
  * @len - the length
@@ -607,6 +645,14 @@ static inline u32 rvt_div_mtu(struct rvt_qp *qp, u32 len)
 extern const int  ib_rvt_state_ops[];
 
 struct rvt_dev_info;
+void rvt_comm_est(struct rvt_qp *qp);
 int rvt_error_qp(struct rvt_qp *qp, enum ib_wc_status err);
+void rvt_rc_error(struct rvt_qp *qp, enum ib_wc_status err);
+unsigned long rvt_rnr_tbl_to_usec(u32 index);
+enum hrtimer_restart rvt_rc_rnr_retry(struct hrtimer *t);
+void rvt_add_rnr_timer(struct rvt_qp *qp, u32 aeth);
+void rvt_del_timers_sync(struct rvt_qp *qp);
+void rvt_stop_rc_timers(struct rvt_qp *qp);
+void rvt_add_retry_timer(struct rvt_qp *qp);
 
 #endif          /* DEF_RDMAVT_INCQP_H */