Merge branch 'mpi-master' into wip-k-fmlpwip-k-fmlp

Conflicts:
	litmus/sched_cedf.c

10 files changed, 106 insertions, 116 deletions

diff --git a/drivers/infiniband/hw/cxgb3/Kconfig b/drivers/infiniband/hw/cxgb3/Kconfig
index 2acec3fadf69..2b6352b85485 100644
--- a/drivers/infiniband/hw/cxgb3/Kconfig
+++ b/drivers/infiniband/hw/cxgb3/Kconfig
@@ -10,7 +10,7 @@ config INFINIBAND_CXGB3
           our website at <http://www.chelsio.com>.
 
           For customer support, please visit our customer support page at
-          <http://www.chelsio.com/support.htm>.
+          <http://www.chelsio.com/support.html>.
 
           Please send feedback to <linux-bugs@chelsio.com>.
 
diff --git a/drivers/infiniband/hw/cxgb3/Makefile b/drivers/infiniband/hw/cxgb3/Makefile
index 7e7b5a66f042..621619c794e5 100644
--- a/drivers/infiniband/hw/cxgb3/Makefile
+++ b/drivers/infiniband/hw/cxgb3/Makefile
@@ -1,10 +1,8 @@
-EXTRA_CFLAGS += -Idrivers/net/cxgb3
+ccflags-y := -Idrivers/net/cxgb3
 
 obj-$(CONFIG_INFINIBAND_CXGB3) += iw_cxgb3.o
 
 iw_cxgb3-y :=  iwch_cm.o iwch_ev.o iwch_cq.o iwch_qp.o iwch_mem.o \
                iwch_provider.o iwch.o cxio_hal.o cxio_resource.o
 
-ifdef CONFIG_INFINIBAND_CXGB3_DEBUG
-EXTRA_CFLAGS += -DDEBUG
-endif
+ccflags-$(CONFIG_INFINIBAND_CXGB3_DEBUG) += -DDEBUG
diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c
index 005b7b52bc1e..c3f5aca4ef00 100644
--- a/drivers/infiniband/hw/cxgb3/cxio_hal.c
+++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c
@@ -160,6 +160,7 @@ int cxio_create_cq(struct cxio_rdev *rdev_p, struct t3_cq *cq, int kernel)
         struct rdma_cq_setup setup;
         int size = (1UL << (cq->size_log2)) * sizeof(struct t3_cqe);
 
+        size += 1; /* one extra page for storing cq-in-err state */
         cq->cqid = cxio_hal_get_cqid(rdev_p->rscp);
         if (!cq->cqid)
                 return -ENOMEM;
@@ -188,6 +189,7 @@ int cxio_create_cq(struct cxio_rdev *rdev_p, struct t3_cq *cq, int kernel)
         return (rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, RDMA_CQ_SETUP, &setup));
 }
 
+#ifdef notyet
 int cxio_resize_cq(struct cxio_rdev *rdev_p, struct t3_cq *cq)
 {
         struct rdma_cq_setup setup;
@@ -199,6 +201,7 @@ int cxio_resize_cq(struct cxio_rdev *rdev_p, struct t3_cq *cq)
         setup.ovfl_mode = 1;
         return (rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, RDMA_CQ_SETUP, &setup));
 }
+#endif
 
 static u32 get_qpid(struct cxio_rdev *rdev_p, struct cxio_ucontext *uctx)
 {
diff --git a/drivers/infiniband/hw/cxgb3/cxio_wr.h b/drivers/infiniband/hw/cxgb3/cxio_wr.h
index e5ddb63e7d23..83d2e19d31ae 100644
--- a/drivers/infiniband/hw/cxgb3/cxio_wr.h
+++ b/drivers/infiniband/hw/cxgb3/cxio_wr.h
@@ -689,7 +689,7 @@ struct t3_swrq {
  * A T3 WQ implements both the SQ and RQ.
  */
 struct t3_wq {
-        union t3_wr *queue;             /* DMA accessable memory */
+        union t3_wr *queue;             /* DMA accessible memory */
         dma_addr_t dma_addr;            /* DMA address for HW */
         DEFINE_DMA_UNMAP_ADDR(mapping); /* unmap kruft */
         u32 error;                      /* 1 once we go to ERROR */
@@ -728,6 +728,22 @@ struct t3_cq {
 #define CQ_VLD_ENTRY(ptr,size_log2,cqe) (Q_GENBIT(ptr,size_log2) == \
                                          CQE_GENBIT(*cqe))
 
+struct t3_cq_status_page {
+        u32 cq_err;
+};
+
+static inline int cxio_cq_in_error(struct t3_cq *cq)
+{
+        return ((struct t3_cq_status_page *)
+                &cq->queue[1 << cq->size_log2])->cq_err;
+}
+
+static inline void cxio_set_cq_in_error(struct t3_cq *cq)
+{
+        ((struct t3_cq_status_page *)
+         &cq->queue[1 << cq->size_log2])->cq_err = 1;
+}
+
 static inline void cxio_set_wq_in_error(struct t3_wq *wq)
 {
         wq->queue->wq_in_err.err |= 1;
diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c
index 13c88871dc3b..0a5008fbebac 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_cm.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c
@@ -338,23 +338,12 @@ static struct rtable *find_route(struct t3cdev *dev, __be32 local_ip,
                                  __be16 peer_port, u8 tos)
 {
         struct rtable *rt;
-        struct flowi fl = {
-                .oif = 0,
-                .nl_u = {
-                         .ip4_u = {
-                                   .daddr = peer_ip,
-                                   .saddr = local_ip,
-                                   .tos = tos}
-                         },
-                .proto = IPPROTO_TCP,
-                .uli_u = {
-                          .ports = {
-                                    .sport = local_port,
-                                    .dport = peer_port}
-                          }
-        };
-
-        if (ip_route_output_flow(&init_net, &rt, &fl, NULL, 0))
+        struct flowi4 fl4;
+
+        rt = ip_route_output_ports(&init_net, &fl4, NULL, peer_ip, local_ip,
+                                   peer_port, local_port, IPPROTO_TCP,
+                                   tos, 0);
+        if (IS_ERR(rt))
                 return NULL;
         return rt;
 }
@@ -925,7 +914,7 @@ static void process_mpa_reply(struct iwch_ep *ep, struct sk_buff *skb)
                 goto err;
 
         if (peer2peer && iwch_rqes_posted(ep->com.qp) == 0) {
-                iwch_post_zb_read(ep->com.qp);
+                iwch_post_zb_read(ep);
         }
 
         goto out;
@@ -1089,37 +1078,45 @@ static int tx_ack(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
         struct iwch_ep *ep = ctx;
         struct cpl_wr_ack *hdr = cplhdr(skb);
         unsigned int credits = ntohs(hdr->credits);
+        unsigned long flags;
+        int post_zb = 0;
 
         PDBG("%s ep %p credits %u\n", __func__, ep, credits);
 
         if (credits == 0) {
-                PDBG(KERN_ERR "%s 0 credit ack  ep %p state %u\n",
-                        __func__, ep, state_read(&ep->com));
+                PDBG("%s 0 credit ack  ep %p state %u\n",
+                     __func__, ep, state_read(&ep->com));
                 return CPL_RET_BUF_DONE;
         }
 
+        spin_lock_irqsave(&ep->com.lock, flags);
         BUG_ON(credits != 1);
         dst_confirm(ep->dst);
         if (!ep->mpa_skb) {
                 PDBG("%s rdma_init wr_ack ep %p state %u\n",
-                        __func__, ep, state_read(&ep->com));
+                        __func__, ep, ep->com.state);
                 if (ep->mpa_attr.initiator) {
                         PDBG("%s initiator ep %p state %u\n",
-                                __func__, ep, state_read(&ep->com));
-                        if (peer2peer)
-                                iwch_post_zb_read(ep->com.qp);
+                                __func__, ep, ep->com.state);
+                        if (peer2peer && ep->com.state == FPDU_MODE)
+                                post_zb = 1;
                 } else {
                         PDBG("%s responder ep %p state %u\n",
-                                __func__, ep, state_read(&ep->com));
-                        ep->com.rpl_done = 1;
-                        wake_up(&ep->com.waitq);
+                                __func__, ep, ep->com.state);
+                        if (ep->com.state == MPA_REQ_RCVD) {
+                                ep->com.rpl_done = 1;
+                                wake_up(&ep->com.waitq);
+                        }
                 }
         } else {
                 PDBG("%s lsm ack ep %p state %u freeing skb\n",
-                        __func__, ep, state_read(&ep->com));
+                        __func__, ep, ep->com.state);
                 kfree_skb(ep->mpa_skb);
                 ep->mpa_skb = NULL;
         }
+        spin_unlock_irqrestore(&ep->com.lock, flags);
+        if (post_zb)
+                iwch_post_zb_read(ep);
         return CPL_RET_BUF_DONE;
 }
 
diff --git a/drivers/infiniband/hw/cxgb3/iwch_ev.c b/drivers/infiniband/hw/cxgb3/iwch_ev.c
index 6afc89e7572c..71e0d845da3d 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_ev.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_ev.c
@@ -76,6 +76,14 @@ static void post_qp_event(struct iwch_dev *rnicp, struct iwch_cq *chp,
         atomic_inc(&qhp->refcnt);
         spin_unlock(&rnicp->lock);
 
+        if (qhp->attr.state == IWCH_QP_STATE_RTS) {
+                attrs.next_state = IWCH_QP_STATE_TERMINATE;
+                iwch_modify_qp(qhp->rhp, qhp, IWCH_QP_ATTR_NEXT_STATE,
+                               &attrs, 1);
+                if (send_term)
+                        iwch_post_terminate(qhp, rsp_msg);
+        }
+
         event.event = ib_event;
         event.device = chp->ibcq.device;
         if (ib_event == IB_EVENT_CQ_ERR)
@@ -86,13 +94,7 @@ static void post_qp_event(struct iwch_dev *rnicp, struct iwch_cq *chp,
         if (qhp->ibqp.event_handler)
                 (*qhp->ibqp.event_handler)(&event, qhp->ibqp.qp_context);
 
-        if (qhp->attr.state == IWCH_QP_STATE_RTS) {
-                attrs.next_state = IWCH_QP_STATE_TERMINATE;
-                iwch_modify_qp(qhp->rhp, qhp, IWCH_QP_ATTR_NEXT_STATE,
-                               &attrs, 1);
-                if (send_term)
-                        iwch_post_terminate(qhp, rsp_msg);
-        }
+        (*chp->ibcq.comp_handler)(&chp->ibcq, chp->ibcq.cq_context);
 
         if (atomic_dec_and_test(&qhp->refcnt))
                 wake_up(&qhp->wait);
@@ -179,7 +181,6 @@ void iwch_ev_dispatch(struct cxio_rdev *rdev_p, struct sk_buff *skb)
         case TPT_ERR_BOUND:
         case TPT_ERR_INVALIDATE_SHARED_MR:
         case TPT_ERR_INVALIDATE_MR_WITH_MW_BOUND:
-                (*chp->ibcq.comp_handler)(&chp->ibcq, chp->ibcq.cq_context);
                 post_qp_event(rnicp, chp, rsp_msg, IB_EVENT_QP_ACCESS_ERR, 1);
                 break;
 
diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c
index fca0b4b747e4..2e2741307af4 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_provider.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c
@@ -154,6 +154,8 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, int entries, int ve
         struct iwch_create_cq_resp uresp;
         struct iwch_create_cq_req ureq;
         struct iwch_ucontext *ucontext = NULL;
+        static int warned;
+        size_t resplen;
 
         PDBG("%s ib_dev %p entries %d\n", __func__, ibdev, entries);
         rhp = to_iwch_dev(ibdev);
@@ -217,15 +219,26 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, int entries, int ve
                 uresp.key = ucontext->key;
                 ucontext->key += PAGE_SIZE;
                 spin_unlock(&ucontext->mmap_lock);
-                if (ib_copy_to_udata(udata, &uresp, sizeof (uresp))) {
+                mm->key = uresp.key;
+                mm->addr = virt_to_phys(chp->cq.queue);
+                if (udata->outlen < sizeof uresp) {
+                        if (!warned++)
+                                printk(KERN_WARNING MOD "Warning - "
+                                       "downlevel libcxgb3 (non-fatal).\n");
+                        mm->len = PAGE_ALIGN((1UL << uresp.size_log2) *
+                                             sizeof(struct t3_cqe));
+                        resplen = sizeof(struct iwch_create_cq_resp_v0);
+                } else {
+                        mm->len = PAGE_ALIGN(((1UL << uresp.size_log2) + 1) *
+                                             sizeof(struct t3_cqe));
+                        uresp.memsize = mm->len;
+                        resplen = sizeof uresp;
+                }
+                if (ib_copy_to_udata(udata, &uresp, resplen)) {
                         kfree(mm);
                         iwch_destroy_cq(&chp->ibcq);
                         return ERR_PTR(-EFAULT);
                 }
-                mm->key = uresp.key;
-                mm->addr = virt_to_phys(chp->cq.queue);
-                mm->len = PAGE_ALIGN((1UL << uresp.size_log2) *
-                                             sizeof (struct t3_cqe));
                 insert_mmap(ucontext, mm);
         }
         PDBG("created cqid 0x%0x chp %p size 0x%0x, dma_addr 0x%0llx\n",
@@ -1414,6 +1427,7 @@ int iwch_register_device(struct iwch_dev *dev)
         dev->ibdev.post_send = iwch_post_send;
         dev->ibdev.post_recv = iwch_post_receive;
         dev->ibdev.get_protocol_stats = iwch_get_mib;
+        dev->ibdev.uverbs_abi_ver = IWCH_UVERBS_ABI_VERSION;
 
         dev->ibdev.iwcm = kmalloc(sizeof(struct iw_cm_verbs), GFP_KERNEL);
         if (!dev->ibdev.iwcm)
diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.h b/drivers/infiniband/hw/cxgb3/iwch_provider.h
index a237d49bdcc9..9a342c9b220d 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_provider.h
+++ b/drivers/infiniband/hw/cxgb3/iwch_provider.h
@@ -332,11 +332,9 @@ int iwch_bind_mw(struct ib_qp *qp,
                              struct ib_mw_bind *mw_bind);
 int iwch_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc);
 int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg);
-int iwch_post_zb_read(struct iwch_qp *qhp);
+int iwch_post_zb_read(struct iwch_ep *ep);
 int iwch_register_device(struct iwch_dev *dev);
 void iwch_unregister_device(struct iwch_dev *dev);
-int iwch_quiesce_qps(struct iwch_cq *chp);
-int iwch_resume_qps(struct iwch_cq *chp);
 void stop_read_rep_timer(struct iwch_qp *qhp);
 int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php,
                       struct iwch_mr *mhp, int shift);
diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c
index c64d27bf2c15..ecd313f359a4 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_qp.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c
@@ -738,7 +738,7 @@ static inline void build_term_codes(struct respQ_msg_t *rsp_msg,
         }
 }
 
-int iwch_post_zb_read(struct iwch_qp *qhp)
+int iwch_post_zb_read(struct iwch_ep *ep)
 {
         union t3_wr *wqe;
         struct sk_buff *skb;
@@ -761,10 +761,10 @@ int iwch_post_zb_read(struct iwch_qp *qhp)
         wqe->read.local_len = cpu_to_be32(0);
         wqe->read.local_to = cpu_to_be64(1);
         wqe->send.wrh.op_seop_flags = cpu_to_be32(V_FW_RIWR_OP(T3_WR_READ));
-        wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(qhp->ep->hwtid)|
+        wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(ep->hwtid)|
                                                 V_FW_RIWR_LEN(flit_cnt));
         skb->priority = CPL_PRIORITY_DATA;
-        return iwch_cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb);
+        return iwch_cxgb3_ofld_send(ep->com.qp->rhp->rdev.t3cdev_p, skb);
 }
 
 /*
@@ -802,14 +802,12 @@ int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg)
 /*
  * Assumes qhp lock is held.
  */
-static void __flush_qp(struct iwch_qp *qhp, unsigned long *flag)
+static void __flush_qp(struct iwch_qp *qhp, struct iwch_cq *rchp,
+                                struct iwch_cq *schp, unsigned long *flag)
 {
-        struct iwch_cq *rchp, *schp;
         int count;
         int flushed;
 
-        rchp = get_chp(qhp->rhp, qhp->attr.rcq);
-        schp = get_chp(qhp->rhp, qhp->attr.scq);
 
         PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp);
         /* take a ref on the qhp since we must release the lock */
@@ -847,10 +845,23 @@ static void __flush_qp(struct iwch_qp *qhp, unsigned long *flag)
 
 static void flush_qp(struct iwch_qp *qhp, unsigned long *flag)
 {
-        if (qhp->ibqp.uobject)
+        struct iwch_cq *rchp, *schp;
+
+        rchp = get_chp(qhp->rhp, qhp->attr.rcq);
+        schp = get_chp(qhp->rhp, qhp->attr.scq);
+
+        if (qhp->ibqp.uobject) {
                 cxio_set_wq_in_error(&qhp->wq);
-        else
-                __flush_qp(qhp, flag);
+                cxio_set_cq_in_error(&rchp->cq);
+                (*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context);
+                if (schp != rchp) {
+                        cxio_set_cq_in_error(&schp->cq);
+                        (*schp->ibcq.comp_handler)(&schp->ibcq,
+                                                   schp->ibcq.cq_context);
+                }
+                return;
+        }
+        __flush_qp(qhp, rchp, schp, flag);
 }
 
 
@@ -1138,59 +1149,3 @@ out:
         PDBG("%s exit state %d\n", __func__, qhp->attr.state);
         return ret;
 }
-
-static int quiesce_qp(struct iwch_qp *qhp)
-{
-        spin_lock_irq(&qhp->lock);
-        iwch_quiesce_tid(qhp->ep);
-        qhp->flags |= QP_QUIESCED;
-        spin_unlock_irq(&qhp->lock);
-        return 0;
-}
-
-static int resume_qp(struct iwch_qp *qhp)
-{
-        spin_lock_irq(&qhp->lock);
-        iwch_resume_tid(qhp->ep);
-        qhp->flags &= ~QP_QUIESCED;
-        spin_unlock_irq(&qhp->lock);
-        return 0;
-}
-
-int iwch_quiesce_qps(struct iwch_cq *chp)
-{
-        int i;
-        struct iwch_qp *qhp;
-
-        for (i=0; i < T3_MAX_NUM_QP; i++) {
-                qhp = get_qhp(chp->rhp, i);
-                if (!qhp)
-                        continue;
-                if ((qhp->attr.rcq == chp->cq.cqid) && !qp_quiesced(qhp)) {
-                        quiesce_qp(qhp);
-                        continue;
-                }
-                if ((qhp->attr.scq == chp->cq.cqid) && !qp_quiesced(qhp))
-                        quiesce_qp(qhp);
-        }
-        return 0;
-}
-
-int iwch_resume_qps(struct iwch_cq *chp)
-{
-        int i;
-        struct iwch_qp *qhp;
-
-        for (i=0; i < T3_MAX_NUM_QP; i++) {
-                qhp = get_qhp(chp->rhp, i);
-                if (!qhp)
-                        continue;
-                if ((qhp->attr.rcq == chp->cq.cqid) && qp_quiesced(qhp)) {
-                        resume_qp(qhp);
-                        continue;
-                }
-                if ((qhp->attr.scq == chp->cq.cqid) && qp_quiesced(qhp))
-                        resume_qp(qhp);
-        }
-        return 0;
-}
diff --git a/drivers/infiniband/hw/cxgb3/iwch_user.h b/drivers/infiniband/hw/cxgb3/iwch_user.h
index cb7086f558c1..a277c31fcaf7 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_user.h
+++ b/drivers/infiniband/hw/cxgb3/iwch_user.h
@@ -45,10 +45,18 @@ struct iwch_create_cq_req {
         __u64 user_rptr_addr;
 };
 
+struct iwch_create_cq_resp_v0 {
+        __u64 key;
+        __u32 cqid;
+        __u32 size_log2;
+};
+
 struct iwch_create_cq_resp {
         __u64 key;
         __u32 cqid;
         __u32 size_log2;
+        __u32 memsize;
+        __u32 reserved;
 };
 
 struct iwch_create_qp_resp {