RDMA/cxgb3: Handle EEH events

- wrap calls into cxgb3 and fail them if we're in the middle
  of a PCI EEH event.

- correctly unwind and release endpoint and other resources when
  we are in an EEH event.

- dispatch IB_EVENT_DEVICE_FATAL event when cxgb3 notifies iw_cxgb3 of
  a fatal error.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c
index a4a82bff71004a48ce39ce138bac27e683765d21..8d71086f5a1c2b8c0e55bb6168ee0cdb474de520 100644 (file)
--- a/drivers/infiniband/hw/cxgb3/cxio_hal.c
+++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c
@@ -152,7 +152,7 @@ static int cxio_hal_clear_qp_ctx(struct cxio_rdev *rdev_p, u32 qpid)
        sge_cmd = qpid << 8 | 3;
        wqe->sge_cmd = cpu_to_be64(sge_cmd);
        skb->priority = CPL_PRIORITY_CONTROL;
-       return (cxgb3_ofld_send(rdev_p->t3cdev_p, skb));
+       return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb);
 }
 
 int cxio_create_cq(struct cxio_rdev *rdev_p, struct t3_cq *cq)
@@ -571,7 +571,7 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p)
             (unsigned long long) rdev_p->ctrl_qp.dma_addr,
             rdev_p->ctrl_qp.workq, 1 << T3_CTRL_QP_SIZE_LOG2);
        skb->priority = CPL_PRIORITY_CONTROL;
-       return (cxgb3_ofld_send(rdev_p->t3cdev_p, skb));
+       return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb);
 err:
        kfree_skb(skb);
        return err;
@@ -701,7 +701,7 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry,
        u32 stag_idx;
        u32 wptr;
 
-       if (rdev_p->flags)
+       if (cxio_fatal_error(rdev_p))
                return -EIO;
 
        stag_state = stag_state > 0;
@@ -858,7 +858,7 @@ int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr)
        wqe->qp_dma_size = cpu_to_be32(attr->qp_dma_size);
        wqe->irs = cpu_to_be32(attr->irs);
        skb->priority = 0;      /* 0=>ToeQ; 1=>CtrlQ */
-       return (cxgb3_ofld_send(rdev_p->t3cdev_p, skb));
+       return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb);
 }
 
 void cxio_register_ev_cb(cxio_hal_ev_callback_func_t ev_cb)
@@ -1041,9 +1041,9 @@ void cxio_rdev_close(struct cxio_rdev *rdev_p)
                cxio_hal_pblpool_destroy(rdev_p);
                cxio_hal_rqtpool_destroy(rdev_p);
                list_del(&rdev_p->entry);
-               rdev_p->t3cdev_p->ulp = NULL;
                cxio_hal_destroy_ctrl_qp(rdev_p);
                cxio_hal_destroy_resource(rdev_p->rscp);
+               rdev_p->t3cdev_p->ulp = NULL;
        }
 }
 

diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h
index 094a66d1480c1dea4e971178b269f22a8287a85b..bfd03bf8be546d4dd05198687187305c9502df43 100644 (file)
--- a/drivers/infiniband/hw/cxgb3/cxio_hal.h
+++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h
@@ -115,6 +115,11 @@ struct cxio_rdev {
 #define        CXIO_ERROR_FATAL        1
 };
 
+static inline int cxio_fatal_error(struct cxio_rdev *rdev_p)
+{
+       return rdev_p->flags & CXIO_ERROR_FATAL;
+}
+
 static inline int cxio_num_stags(struct cxio_rdev *rdev_p)
 {
        return min((int)T3_MAX_NUM_STAG, (int)((rdev_p->rnic_info.tpt_top - rdev_p->rnic_info.tpt_base) >> 5));
@@ -188,6 +193,7 @@ void cxio_count_scqes(struct t3_cq *cq, struct t3_wq *wq, int *count);
 void cxio_flush_hw_cq(struct t3_cq *cq);
 int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe,
                     u8 *cqe_flushed, u64 *cookie, u32 *credit);
+int iwch_cxgb3_ofld_send(struct t3cdev *tdev, struct sk_buff *skb);
 
 #define MOD "iw_cxgb3: "
 #define PDBG(fmt, args...) pr_debug(MOD fmt, ## args)

diff --git a/drivers/infiniband/hw/cxgb3/iwch.c b/drivers/infiniband/hw/cxgb3/iwch.c
index 37a4fc264a07a10bc34ff8e7f92b34869153e640..26fc0a4eaa749f91477a711be17710e98dc36ff1 100644 (file)
--- a/drivers/infiniband/hw/cxgb3/iwch.c
+++ b/drivers/infiniband/hw/cxgb3/iwch.c
@@ -165,12 +165,19 @@ static void close_rnic_dev(struct t3cdev *tdev)
 static void iwch_err_handler(struct t3cdev *tdev, u32 status, u32 error)
 {
        struct cxio_rdev *rdev = tdev->ulp;
+       struct iwch_dev *rnicp = rdev_to_iwch_dev(rdev);
+       struct ib_event event;
 
-       if (status == OFFLOAD_STATUS_DOWN)
+       if (status == OFFLOAD_STATUS_DOWN) {
                rdev->flags = CXIO_ERROR_FATAL;
 
-       return;
+               event.device = &rnicp->ibdev;
+               event.event  = IB_EVENT_DEVICE_FATAL;
+               event.element.port_num = 0;
+               ib_dispatch_event(&event);
+       }
 
+       return;
 }
 
 static int __init iwch_init_module(void)

diff --git a/drivers/infiniband/hw/cxgb3/iwch.h b/drivers/infiniband/hw/cxgb3/iwch.h
index 3773453b2cf06e02a2cca70fde7c0f13bce710b3..84735506333f3b7b90b392b2fceabdb49d23cdd8 100644 (file)
--- a/drivers/infiniband/hw/cxgb3/iwch.h
+++ b/drivers/infiniband/hw/cxgb3/iwch.h
@@ -117,6 +117,11 @@ static inline struct iwch_dev *to_iwch_dev(struct ib_device *ibdev)
        return container_of(ibdev, struct iwch_dev, ibdev);
 }
 
+static inline struct iwch_dev *rdev_to_iwch_dev(struct cxio_rdev *rdev)
+{
+       return container_of(rdev, struct iwch_dev, rdev);
+}
+
 static inline int t3b_device(const struct iwch_dev *rhp)
 {
        return rhp->rdev.t3cdev_p->type == T3B;

diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c
index 8699947aaf6cdfa036295d0b9dfbb49b54a30625..59e1c5f00785dd1d693a52ccd76b2103d570fd48 100644 (file)
--- a/drivers/infiniband/hw/cxgb3/iwch_cm.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c
@@ -139,6 +139,38 @@ static void stop_ep_timer(struct iwch_ep *ep)
        put_ep(&ep->com);
 }
 
+int iwch_l2t_send(struct t3cdev *tdev, struct sk_buff *skb, struct l2t_entry *l2e)
+{
+       int     error = 0;
+       struct cxio_rdev *rdev;
+
+       rdev = (struct cxio_rdev *)tdev->ulp;
+       if (cxio_fatal_error(rdev)) {
+               kfree_skb(skb);
+               return -EIO;
+       }
+       error = l2t_send(tdev, skb, l2e);
+       if (error)
+               kfree_skb(skb);
+       return error;
+}
+
+int iwch_cxgb3_ofld_send(struct t3cdev *tdev, struct sk_buff *skb)
+{
+       int     error = 0;
+       struct cxio_rdev *rdev;
+
+       rdev = (struct cxio_rdev *)tdev->ulp;
+       if (cxio_fatal_error(rdev)) {
+               kfree_skb(skb);
+               return -EIO;
+       }
+       error = cxgb3_ofld_send(tdev, skb);
+       if (error)
+               kfree_skb(skb);
+       return error;
+}
+
 static void release_tid(struct t3cdev *tdev, u32 hwtid, struct sk_buff *skb)
 {
        struct cpl_tid_release *req;
@@ -150,7 +182,7 @@ static void release_tid(struct t3cdev *tdev, u32 hwtid, struct sk_buff *skb)
        req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
        OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_TID_RELEASE, hwtid));
        skb->priority = CPL_PRIORITY_SETUP;
-       cxgb3_ofld_send(tdev, skb);
+       iwch_cxgb3_ofld_send(tdev, skb);
        return;
 }
 
@@ -172,8 +204,7 @@ int iwch_quiesce_tid(struct iwch_ep *ep)
        req->val = cpu_to_be64(1 << S_TCB_RX_QUIESCE);
 
        skb->priority = CPL_PRIORITY_DATA;
-       cxgb3_ofld_send(ep->com.tdev, skb);
-       return 0;
+       return iwch_cxgb3_ofld_send(ep->com.tdev, skb);
 }
 
 int iwch_resume_tid(struct iwch_ep *ep)
@@ -194,8 +225,7 @@ int iwch_resume_tid(struct iwch_ep *ep)
        req->val = 0;
 
        skb->priority = CPL_PRIORITY_DATA;
-       cxgb3_ofld_send(ep->com.tdev, skb);
-       return 0;
+       return iwch_cxgb3_ofld_send(ep->com.tdev, skb);
 }
 
 static void set_emss(struct iwch_ep *ep, u16 opt)
@@ -382,7 +412,7 @@ static void abort_arp_failure(struct t3cdev *dev, struct sk_buff *skb)
 
        PDBG("%s t3cdev %p\n", __func__, dev);
        req->cmd = CPL_ABORT_NO_RST;
-       cxgb3_ofld_send(dev, skb);
+       iwch_cxgb3_ofld_send(dev, skb);
 }
 
 static int send_halfclose(struct iwch_ep *ep, gfp_t gfp)
@@ -402,8 +432,7 @@ static int send_halfclose(struct iwch_ep *ep, gfp_t gfp)
        req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_CLOSE_CON));
        req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid));
        OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_CON_REQ, ep->hwtid));
-       l2t_send(ep->com.tdev, skb, ep->l2t);
-       return 0;
+       return iwch_l2t_send(ep->com.tdev, skb, ep->l2t);
 }
 
 static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp)
@@ -424,8 +453,7 @@ static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp)
        req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid));
        OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ABORT_REQ, ep->hwtid));
        req->cmd = CPL_ABORT_SEND_RST;
-       l2t_send(ep->com.tdev, skb, ep->l2t);
-       return 0;
+       return iwch_l2t_send(ep->com.tdev, skb, ep->l2t);
 }
 
 static int send_connect(struct iwch_ep *ep)
@@ -469,8 +497,7 @@ static int send_connect(struct iwch_ep *ep)
        req->opt0l = htonl(opt0l);
        req->params = 0;
        req->opt2 = htonl(opt2);
-       l2t_send(ep->com.tdev, skb, ep->l2t);
-       return 0;
+       return iwch_l2t_send(ep->com.tdev, skb, ep->l2t);
 }
 
 static void send_mpa_req(struct iwch_ep *ep, struct sk_buff *skb)
@@ -527,7 +554,7 @@ static void send_mpa_req(struct iwch_ep *ep, struct sk_buff *skb)
        req->sndseq = htonl(ep->snd_seq);
        BUG_ON(ep->mpa_skb);
        ep->mpa_skb = skb;
-       l2t_send(ep->com.tdev, skb, ep->l2t);
+       iwch_l2t_send(ep->com.tdev, skb, ep->l2t);
        start_ep_timer(ep);
        state_set(&ep->com, MPA_REQ_SENT);
        return;
@@ -578,8 +605,7 @@ static int send_mpa_reject(struct iwch_ep *ep, const void *pdata, u8 plen)
        req->sndseq = htonl(ep->snd_seq);
        BUG_ON(ep->mpa_skb);
        ep->mpa_skb = skb;
-       l2t_send(ep->com.tdev, skb, ep->l2t);
-       return 0;
+       return iwch_l2t_send(ep->com.tdev, skb, ep->l2t);
 }
 
 static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen)
@@ -630,8 +656,7 @@ static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen)
        req->sndseq = htonl(ep->snd_seq);
        ep->mpa_skb = skb;
        state_set(&ep->com, MPA_REP_SENT);
-       l2t_send(ep->com.tdev, skb, ep->l2t);
-       return 0;
+       return iwch_l2t_send(ep->com.tdev, skb, ep->l2t);
 }
 
 static int act_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
@@ -795,7 +820,7 @@ static int update_rx_credits(struct iwch_ep *ep, u32 credits)
        OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_RX_DATA_ACK, ep->hwtid));
        req->credit_dack = htonl(V_RX_CREDITS(credits) | V_RX_FORCE_ACK(1));
        skb->priority = CPL_PRIORITY_ACK;
-       cxgb3_ofld_send(ep->com.tdev, skb);
+       iwch_cxgb3_ofld_send(ep->com.tdev, skb);
        return credits;
 }
 
@@ -1203,8 +1228,7 @@ static int listen_start(struct iwch_listen_ep *ep)
        req->opt1 = htonl(V_CONN_POLICY(CPL_CONN_POLICY_ASK));
 
        skb->priority = 1;
-       cxgb3_ofld_send(ep->com.tdev, skb);
-       return 0;
+       return iwch_cxgb3_ofld_send(ep->com.tdev, skb);
 }
 
 static int pass_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
@@ -1237,8 +1261,7 @@ static int listen_stop(struct iwch_listen_ep *ep)
        req->cpu_idx = 0;
        OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_LISTSRV_REQ, ep->stid));
        skb->priority = 1;
-       cxgb3_ofld_send(ep->com.tdev, skb);
-       return 0;
+       return iwch_cxgb3_ofld_send(ep->com.tdev, skb);
 }
 
 static int close_listsrv_rpl(struct t3cdev *tdev, struct sk_buff *skb,
@@ -1286,7 +1309,7 @@ static void accept_cr(struct iwch_ep *ep, __be32 peer_ip, struct sk_buff *skb)
        rpl->opt2 = htonl(opt2);
        rpl->rsvd = rpl->opt2;  /* workaround for HW bug */
        skb->priority = CPL_PRIORITY_SETUP;
-       l2t_send(ep->com.tdev, skb, ep->l2t);
+       iwch_l2t_send(ep->com.tdev, skb, ep->l2t);
 
        return;
 }
@@ -1315,7 +1338,7 @@ static void reject_cr(struct t3cdev *tdev, u32 hwtid, __be32 peer_ip,
                rpl->opt0l_status = htonl(CPL_PASS_OPEN_REJECT);
                rpl->opt2 = 0;
                rpl->rsvd = rpl->opt2;
-               cxgb3_ofld_send(tdev, skb);
+               iwch_cxgb3_ofld_send(tdev, skb);
        }
 }
 
@@ -1613,7 +1636,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
        rpl->wr.wr_lo = htonl(V_WR_TID(ep->hwtid));
        OPCODE_TID(rpl) = htonl(MK_OPCODE_TID(CPL_ABORT_RPL, ep->hwtid));
        rpl->cmd = CPL_ABORT_NO_RST;
-       cxgb3_ofld_send(ep->com.tdev, rpl_skb);
+       iwch_cxgb3_ofld_send(ep->com.tdev, rpl_skb);
 out:
        if (release)
                release_ep_resources(ep);
@@ -2017,8 +2040,11 @@ int iwch_destroy_listen(struct iw_cm_id *cm_id)
        ep->com.rpl_done = 0;
        ep->com.rpl_err = 0;
        err = listen_stop(ep);
+       if (err)
+               goto done;
        wait_event(ep->com.waitq, ep->com.rpl_done);
        cxgb3_free_stid(ep->com.tdev, ep->stid);
+done:
        err = ep->com.rpl_err;
        cm_id->rem_ref(cm_id);
        put_ep(&ep->com);
@@ -2030,12 +2056,22 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp)
        int ret=0;
        unsigned long flags;
        int close = 0;
+       int fatal = 0;
+       struct t3cdev *tdev;
+       struct cxio_rdev *rdev;
 
        spin_lock_irqsave(&ep->com.lock, flags);
 
        PDBG("%s ep %p state %s, abrupt %d\n", __func__, ep,
             states[ep->com.state], abrupt);
 
+       tdev = (struct t3cdev *)ep->com.tdev;
+       rdev = (struct cxio_rdev *)tdev->ulp;
+       if (cxio_fatal_error(rdev)) {
+               fatal = 1;
+               close_complete_upcall(ep);
+               ep->com.state = DEAD;
+       }
        switch (ep->com.state) {
        case MPA_REQ_WAIT:
        case MPA_REQ_SENT:
@@ -2075,7 +2111,11 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp)
                        ret = send_abort(ep, NULL, gfp);
                else
                        ret = send_halfclose(ep, gfp);
+               if (ret)
+                       fatal = 1;
        }
+       if (fatal)
+               release_ep_resources(ep);
        return ret;
 }
 

diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c
index c758fbd5847887b6e22e49ade0d0512a0af5592c..2f546a6253300a5d7cc1d505e404272dc86ddc10 100644 (file)
--- a/drivers/infiniband/hw/cxgb3/iwch_qp.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c
@@ -751,7 +751,7 @@ int iwch_post_zb_read(struct iwch_qp *qhp)
        wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(qhp->ep->hwtid)|
                                                V_FW_RIWR_LEN(flit_cnt));
        skb->priority = CPL_PRIORITY_DATA;
-       return cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb);
+       return iwch_cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb);
 }
 
 /*
@@ -783,7 +783,7 @@ int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg)
                         V_FW_RIWR_FLAGS(T3_COMPLETION_FLAG | T3_NOTIFY_FLAG));
        wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(qhp->ep->hwtid));
        skb->priority = CPL_PRIORITY_DATA;
-       return cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb);
+       return iwch_cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb);
 }
 
 /*