diff options
Diffstat (limited to 'drivers/infiniband/ulp/ipoib/ipoib_cm.c')
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_cm.c | 487 |
1 files changed, 273 insertions, 214 deletions
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 3eceb61e3532..b610d36295bb 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -38,6 +38,8 @@ #include <linux/slab.h> #include <linux/vmalloc.h> #include <linux/moduleparam.h> +#include <linux/sched/signal.h> +#include <linux/sched/mm.h> #include "ipoib.h" @@ -63,6 +65,8 @@ MODULE_PARM_DESC(cm_data_debug_level, #define IPOIB_CM_RX_DELAY (3 * 256 * HZ) #define IPOIB_CM_RX_UPDATE_MASK (0x3) +#define IPOIB_CM_RX_RESERVE (ALIGN(IPOIB_HARD_LEN, 16) - IPOIB_ENCAP_LEN) + static struct ib_qp_attr ipoib_cm_err_attr = { .qp_state = IB_QPS_ERR }; @@ -70,12 +74,11 @@ static struct ib_qp_attr ipoib_cm_err_attr = { #define IPOIB_CM_RX_DRAIN_WRID 0xffffffff static struct ib_send_wr ipoib_cm_rx_drain_wr = { - .wr_id = IPOIB_CM_RX_DRAIN_WRID, .opcode = IB_WR_SEND, }; static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, - struct ib_cm_event *event); + const struct ib_cm_event *event); static void ipoib_cm_dma_unmap_rx(struct ipoib_dev_priv *priv, int frags, u64 mapping[IPOIB_CM_RX_SG]) @@ -90,8 +93,7 @@ static void ipoib_cm_dma_unmap_rx(struct ipoib_dev_priv *priv, int frags, static int ipoib_cm_post_receive_srq(struct net_device *dev, int id) { - struct ipoib_dev_priv *priv = netdev_priv(dev); - struct ib_recv_wr *bad_wr; + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i, ret; priv->cm.rx_wr.wr_id = id | IPOIB_OP_CM | IPOIB_OP_RECV; @@ -99,7 +101,7 @@ static int ipoib_cm_post_receive_srq(struct net_device *dev, int id) for (i = 0; i < priv->cm.num_frags; ++i) priv->cm.rx_sge[i].addr = priv->cm.srq_ring[id].mapping[i]; - ret = ib_post_srq_recv(priv->cm.srq, &priv->cm.rx_wr, &bad_wr); + ret = ib_post_srq_recv(priv->cm.srq, &priv->cm.rx_wr, NULL); if (unlikely(ret)) { ipoib_warn(priv, "post srq failed for buf %d (%d)\n", id, ret); ipoib_cm_dma_unmap_rx(priv, priv->cm.num_frags - 1, @@ -116,8 +118,7 @@ static int ipoib_cm_post_receive_nonsrq(struct net_device *dev, struct ib_recv_wr *wr, struct ib_sge *sge, int id) { - struct ipoib_dev_priv *priv = netdev_priv(dev); - struct ib_recv_wr *bad_wr; + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i, ret; wr->wr_id = id | IPOIB_OP_CM | IPOIB_OP_RECV; @@ -125,7 +126,7 @@ static int ipoib_cm_post_receive_nonsrq(struct net_device *dev, for (i = 0; i < IPOIB_CM_RX_SG; ++i) sge[i].addr = rx->rx_ring[id].mapping[i]; - ret = ib_post_recv(rx->qp, wr, &bad_wr); + ret = ib_post_recv(rx->qp, wr, NULL); if (unlikely(ret)) { ipoib_warn(priv, "post recv failed for buf %d (%d)\n", id, ret); ipoib_cm_dma_unmap_rx(priv, IPOIB_CM_RX_SG - 1, @@ -140,21 +141,22 @@ static int ipoib_cm_post_receive_nonsrq(struct net_device *dev, static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev, struct ipoib_cm_rx_buf *rx_ring, int id, int frags, - u64 mapping[IPOIB_CM_RX_SG]) + u64 mapping[IPOIB_CM_RX_SG], + gfp_t gfp) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct sk_buff *skb; int i; - skb = dev_alloc_skb(IPOIB_CM_HEAD_SIZE + 12); + skb = dev_alloc_skb(ALIGN(IPOIB_CM_HEAD_SIZE + IPOIB_PSEUDO_LEN, 16)); if (unlikely(!skb)) return NULL; /* - * IPoIB adds a 4 byte header. So we need 12 more bytes to align the + * IPoIB adds a IPOIB_ENCAP_LEN byte header, this will align the * IP header to a multiple of 16. */ - skb_reserve(skb, 12); + skb_reserve(skb, IPOIB_CM_RX_RESERVE); mapping[0] = ib_dma_map_single(priv->ca, skb->data, IPOIB_CM_HEAD_SIZE, DMA_FROM_DEVICE); @@ -164,7 +166,7 @@ static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev, } for (i = 0; i < frags; i++) { - struct page *page = alloc_page(GFP_ATOMIC); + struct page *page = alloc_page(gfp); if (!page) goto partial_error; @@ -193,7 +195,7 @@ partial_error: static void ipoib_cm_free_rx_ring(struct net_device *dev, struct ipoib_cm_rx_buf *rx_ring) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i; for (i = 0; i < ipoib_recvq_size; ++i) @@ -208,7 +210,6 @@ static void ipoib_cm_free_rx_ring(struct net_device *dev, static void ipoib_cm_start_rx_drain(struct ipoib_dev_priv *priv) { - struct ib_send_wr *bad_wr; struct ipoib_cm_rx *p; /* We only reserved 1 extra slot in CQ for drain WRs, so @@ -222,7 +223,8 @@ static void ipoib_cm_start_rx_drain(struct ipoib_dev_priv *priv) * error" WC will be immediately generated for each WR we post. */ p = list_entry(priv->cm.rx_flush_list.next, typeof(*p), list); - if (ib_post_send(p->qp, &ipoib_cm_rx_drain_wr, &bad_wr)) + ipoib_cm_rx_drain_wr.wr_id = IPOIB_CM_RX_DRAIN_WRID; + if (ib_post_send(p->qp, &ipoib_cm_rx_drain_wr, NULL)) ipoib_warn(priv, "failed to post drain wr\n"); list_splice_init(&priv->cm.rx_flush_list, &priv->cm.rx_drain_list); @@ -231,7 +233,7 @@ static void ipoib_cm_start_rx_drain(struct ipoib_dev_priv *priv) static void ipoib_cm_rx_event_handler(struct ib_event *event, void *ctx) { struct ipoib_cm_rx *p = ctx; - struct ipoib_dev_priv *priv = netdev_priv(p->dev); + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); unsigned long flags; if (event->event != IB_EVENT_QP_LAST_WQE_REACHED) @@ -247,7 +249,7 @@ static void ipoib_cm_rx_event_handler(struct ib_event *event, void *ctx) static struct ib_qp *ipoib_cm_create_rx_qp(struct net_device *dev, struct ipoib_cm_rx *p) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_init_attr attr = { .event_handler = ipoib_cm_rx_event_handler, .send_cq = priv->recv_cq, /* For drain WR */ @@ -270,9 +272,9 @@ static struct ib_qp *ipoib_cm_create_rx_qp(struct net_device *dev, static int ipoib_cm_modify_rx_qp(struct net_device *dev, struct ib_cm_id *cm_id, struct ib_qp *qp, - unsigned psn) + unsigned int psn) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_attr qp_attr; int qp_attr_mask, ret; @@ -327,11 +329,11 @@ static void ipoib_cm_init_rx_wr(struct net_device *dev, struct ib_recv_wr *wr, struct ib_sge *sge) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i; for (i = 0; i < priv->cm.num_frags; ++i) - sge[i].lkey = priv->mr->lkey; + sge[i].lkey = priv->pd->local_dma_lkey; sge[0].length = IPOIB_CM_HEAD_SIZE; for (i = 1; i < priv->cm.num_frags; ++i) @@ -345,7 +347,7 @@ static void ipoib_cm_init_rx_wr(struct net_device *dev, static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_id, struct ipoib_cm_rx *rx) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct { struct ib_recv_wr wr; struct ib_sge sge[IPOIB_CM_RX_SG]; @@ -353,17 +355,15 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i int ret; int i; - rx->rx_ring = vzalloc(ipoib_recvq_size * sizeof *rx->rx_ring); - if (!rx->rx_ring) { - printk(KERN_WARNING "%s: failed to allocate CM non-SRQ ring (%d entries)\n", - priv->ca->name, ipoib_recvq_size); + rx->rx_ring = vzalloc(array_size(ipoib_recvq_size, + sizeof(*rx->rx_ring))); + if (!rx->rx_ring) return -ENOMEM; - } - t = kmalloc(sizeof *t, GFP_KERNEL); + t = kmalloc(sizeof(*t), GFP_KERNEL); if (!t) { ret = -ENOMEM; - goto err_free; + goto err_free_1; } ipoib_cm_init_rx_wr(dev, &t->wr, t->sge); @@ -382,10 +382,11 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i for (i = 0; i < ipoib_recvq_size; ++i) { if (!ipoib_cm_alloc_rx_skb(dev, rx->rx_ring, i, IPOIB_CM_RX_SG - 1, - rx->rx_ring[i].mapping)) { + rx->rx_ring[i].mapping, + GFP_KERNEL)) { ipoib_warn(priv, "failed to allocate receive buffer %d\n", i); - ret = -ENOMEM; - goto err_count; + ret = -ENOMEM; + goto err_count; } ret = ipoib_cm_post_receive_nonsrq(dev, rx, &t->wr, t->sge, i); if (ret) { @@ -409,16 +410,19 @@ err_count: err_free: kfree(t); + +err_free_1: ipoib_cm_free_rx_ring(dev, rx->rx_ring); return ret; } static int ipoib_cm_send_rep(struct net_device *dev, struct ib_cm_id *cm_id, - struct ib_qp *qp, struct ib_cm_req_event_param *req, - unsigned psn) + struct ib_qp *qp, + const struct ib_cm_req_event_param *req, + unsigned int psn) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_data data = {}; struct ib_cm_rep_param rep = {}; @@ -426,7 +430,7 @@ static int ipoib_cm_send_rep(struct net_device *dev, struct ib_cm_id *cm_id, data.mtu = cpu_to_be32(IPOIB_CM_BUF_SIZE); rep.private_data = &data; - rep.private_data_len = sizeof data; + rep.private_data_len = sizeof(data); rep.flow_control = 0; rep.rnr_retry_count = req->rnr_retry_count; rep.srq = ipoib_cm_has_srq(dev); @@ -435,16 +439,17 @@ static int ipoib_cm_send_rep(struct net_device *dev, struct ib_cm_id *cm_id, return ib_send_cm_rep(cm_id, &rep); } -static int ipoib_cm_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event) +static int ipoib_cm_req_handler(struct ib_cm_id *cm_id, + const struct ib_cm_event *event) { struct net_device *dev = cm_id->context; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx *p; - unsigned psn; + unsigned int psn; int ret; ipoib_dbg(priv, "REQ arrived\n"); - p = kzalloc(sizeof *p, GFP_KERNEL); + p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) return -ENOMEM; p->dev = dev; @@ -460,7 +465,7 @@ static int ipoib_cm_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *even goto err_qp; } - psn = prandom_u32() & 0xffffff; + psn = get_random_u32() & 0xffffff; ret = ipoib_cm_modify_rx_qp(dev, cm_id, p->qp, psn); if (ret) goto err_modify; @@ -472,7 +477,7 @@ static int ipoib_cm_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *even } spin_lock_irq(&priv->lock); - queue_delayed_work(ipoib_workqueue, + queue_delayed_work(priv->wq, &priv->cm.stale_task, IPOIB_CM_RX_DELAY); /* Add this entry to passive ids list head, but do not re-add it * if IB_EVENT_QP_LAST_WQE_REACHED has moved it to flush list. */ @@ -497,7 +502,7 @@ err_qp: } static int ipoib_cm_rx_handler(struct ib_cm_id *cm_id, - struct ib_cm_event *event) + const struct ib_cm_event *event) { struct ipoib_cm_rx *p; struct ipoib_dev_priv *priv; @@ -506,15 +511,14 @@ static int ipoib_cm_rx_handler(struct ib_cm_id *cm_id, case IB_CM_REQ_RECEIVED: return ipoib_cm_req_handler(cm_id, event); case IB_CM_DREQ_RECEIVED: - p = cm_id->context; ib_send_cm_drep(cm_id, NULL, 0); - /* Fall through */ + fallthrough; case IB_CM_REJ_RECEIVED: p = cm_id->context; - priv = netdev_priv(p->dev); + priv = ipoib_priv(p->dev); if (ib_modify_qp(p->qp, &ipoib_cm_err_attr, IB_QP_STATE)) ipoib_warn(priv, "unable to move qp to error state\n"); - /* Fall through */ + fallthrough; default: return 0; } @@ -542,7 +546,7 @@ static void skb_put_frags(struct sk_buff *skb, unsigned int hdr_space, 0, PAGE_SIZE); --skb_shinfo(skb)->nr_frags; } else { - size = min(length, (unsigned) PAGE_SIZE); + size = min_t(unsigned int, length, PAGE_SIZE); skb_frag_size_set(frag, size); skb->data_len += size; @@ -555,7 +559,7 @@ static void skb_put_frags(struct sk_buff *skb, unsigned int hdr_space, void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx_buf *rx_ring; unsigned int wr_id = wc->wr_id & ~(IPOIB_OP_CM | IPOIB_OP_RECV); struct sk_buff *skb, *newskb; @@ -574,7 +578,7 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) spin_lock_irqsave(&priv->lock, flags); list_splice_init(&priv->cm.rx_drain_list, &priv->cm.rx_reap_list); ipoib_cm_start_rx_drain(priv); - queue_work(ipoib_workqueue, &priv->cm.rx_reap_task); + queue_work(priv->wq, &priv->cm.rx_reap_task); spin_unlock_irqrestore(&priv->lock, flags); } else ipoib_warn(priv, "cm recv completion event with wrid %d (> %d)\n", @@ -590,9 +594,9 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) skb = rx_ring[wr_id].skb; if (unlikely(wc->status != IB_WC_SUCCESS)) { - ipoib_dbg(priv, "cm recv error " - "(status=%d, wrid=%d vend_err %x)\n", - wc->status, wr_id, wc->vendor_err); + ipoib_dbg(priv, + "cm recv error (status=%d, wrid=%d vend_err %#x)\n", + wc->status, wr_id, wc->vendor_err); ++dev->stats.rx_dropped; if (has_srq) goto repost; @@ -601,7 +605,7 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) spin_lock_irqsave(&priv->lock, flags); list_move(&p->list, &priv->cm.rx_reap_list); spin_unlock_irqrestore(&priv->lock, flags); - queue_work(ipoib_workqueue, &priv->cm.rx_reap_task); + queue_work(priv->wq, &priv->cm.rx_reap_task); } return; } @@ -622,9 +626,9 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) if (wc->byte_len < IPOIB_CM_COPYBREAK) { int dlen = wc->byte_len; - small_skb = dev_alloc_skb(dlen + 12); + small_skb = dev_alloc_skb(dlen + IPOIB_CM_RX_RESERVE); if (small_skb) { - skb_reserve(small_skb, 12); + skb_reserve(small_skb, IPOIB_CM_RX_RESERVE); ib_dma_sync_single_for_cpu(priv->ca, rx_ring[wr_id].mapping[0], dlen, DMA_FROM_DEVICE); skb_copy_from_linear_data(skb, small_skb->data, dlen); @@ -636,10 +640,12 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) } } - frags = PAGE_ALIGN(wc->byte_len - min(wc->byte_len, - (unsigned)IPOIB_CM_HEAD_SIZE)) / PAGE_SIZE; + frags = PAGE_ALIGN(wc->byte_len - + min_t(u32, wc->byte_len, IPOIB_CM_HEAD_SIZE)) / + PAGE_SIZE; - newskb = ipoib_cm_alloc_rx_skb(dev, rx_ring, wr_id, frags, mapping); + newskb = ipoib_cm_alloc_rx_skb(dev, rx_ring, wr_id, frags, + mapping, GFP_ATOMIC); if (unlikely(!newskb)) { /* * If we can't allocate a new RX buffer, dump @@ -651,7 +657,7 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) } ipoib_cm_dma_unmap_rx(priv, frags, rx_ring[wr_id].mapping); - memcpy(rx_ring[wr_id].mapping, mapping, (frags + 1) * sizeof *mapping); + memcpy(rx_ring[wr_id].mapping, mapping, (frags + 1) * sizeof(*mapping)); ipoib_dbg_data(priv, "received %d bytes, SLID 0x%04x\n", wc->byte_len, wc->slid); @@ -660,8 +666,7 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) copied: skb->protocol = ((struct ipoib_header *) skb->data)->proto; - skb_reset_mac_header(skb); - skb_pull(skb, IPOIB_ENCAP_LEN); + skb_add_pseudo_hdr(skb); ++dev->stats.rx_packets; dev->stats.rx_bytes += skb->len; @@ -691,25 +696,21 @@ repost: static inline int post_send(struct ipoib_dev_priv *priv, struct ipoib_cm_tx *tx, unsigned int wr_id, - u64 addr, int len) + struct ipoib_tx_buf *tx_req) { - struct ib_send_wr *bad_wr; - - priv->tx_sge[0].addr = addr; - priv->tx_sge[0].length = len; + ipoib_build_sge(priv, tx_req); - priv->tx_wr.num_sge = 1; - priv->tx_wr.wr_id = wr_id | IPOIB_OP_CM; + priv->tx_wr.wr.wr_id = wr_id | IPOIB_OP_CM; - return ib_post_send(tx->qp, &priv->tx_wr, &bad_wr); + return ib_post_send(tx->qp, &priv->tx_wr.wr, NULL); } void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_tx *tx) { - struct ipoib_dev_priv *priv = netdev_priv(dev); - struct ipoib_cm_tx_buf *tx_req; - u64 addr; + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct ipoib_tx_buf *tx_req; int rc; + unsigned int usable_sge = tx->max_send_sge - !!skb_headlen(skb); if (unlikely(skb->len > tx->mtu)) { ipoib_warn(priv, "packet len %d (> %d) too long to send, dropping\n", @@ -719,7 +720,23 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_ ipoib_cm_skb_too_long(dev, skb, tx->mtu - IPOIB_ENCAP_LEN); return; } - + if (skb_shinfo(skb)->nr_frags > usable_sge) { + if (skb_linearize(skb) < 0) { + ipoib_warn(priv, "skb could not be linearized\n"); + ++dev->stats.tx_dropped; + ++dev->stats.tx_errors; + dev_kfree_skb_any(skb); + return; + } + /* Does skb_linearize return ok without reducing nr_frags? */ + if (skb_shinfo(skb)->nr_frags > usable_sge) { + ipoib_warn(priv, "too many frags after skb linearize\n"); + ++dev->stats.tx_dropped; + ++dev->stats.tx_errors; + dev_kfree_skb_any(skb); + return; + } + } ipoib_dbg_data(priv, "sending packet: head 0x%x length %d connection 0x%x\n", tx->tx_head, skb->len, tx->qp->qp_num); @@ -732,49 +749,54 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_ */ tx_req = &tx->tx_ring[tx->tx_head & (ipoib_sendq_size - 1)]; tx_req->skb = skb; - addr = ib_dma_map_single(priv->ca, skb->data, skb->len, DMA_TO_DEVICE); - if (unlikely(ib_dma_mapping_error(priv->ca, addr))) { + + if (unlikely(ipoib_dma_map_tx(priv->ca, tx_req))) { ++dev->stats.tx_errors; dev_kfree_skb_any(skb); return; } - tx_req->mapping = addr; + if ((priv->global_tx_head - priv->global_tx_tail) == + ipoib_sendq_size - 1) { + ipoib_dbg(priv, "TX ring 0x%x full, stopping kernel net queue\n", + tx->qp->qp_num); + netif_stop_queue(dev); + } skb_orphan(skb); skb_dst_drop(skb); - rc = post_send(priv, tx, tx->tx_head & (ipoib_sendq_size - 1), - addr, skb->len); + if (netif_queue_stopped(dev)) { + rc = ib_req_notify_cq(priv->send_cq, IB_CQ_NEXT_COMP | + IB_CQ_REPORT_MISSED_EVENTS); + if (unlikely(rc < 0)) + ipoib_warn(priv, "IPoIB/CM:request notify on send CQ failed\n"); + else if (rc) + napi_schedule(&priv->send_napi); + } + + rc = post_send(priv, tx, tx->tx_head & (ipoib_sendq_size - 1), tx_req); if (unlikely(rc)) { - ipoib_warn(priv, "post_send failed, error %d\n", rc); + ipoib_warn(priv, "IPoIB/CM:post_send failed, error %d\n", rc); ++dev->stats.tx_errors; - ib_dma_unmap_single(priv->ca, addr, skb->len, DMA_TO_DEVICE); + ipoib_dma_unmap_tx(priv, tx_req); dev_kfree_skb_any(skb); + + if (netif_queue_stopped(dev)) + netif_wake_queue(dev); } else { - dev->trans_start = jiffies; + netif_trans_update(dev); ++tx->tx_head; - - if (++priv->tx_outstanding == ipoib_sendq_size) { - ipoib_dbg(priv, "TX ring 0x%x full, stopping kernel net queue\n", - tx->qp->qp_num); - netif_stop_queue(dev); - rc = ib_req_notify_cq(priv->send_cq, - IB_CQ_NEXT_COMP | IB_CQ_REPORT_MISSED_EVENTS); - if (rc < 0) - ipoib_warn(priv, "request notify on send CQ failed\n"); - else if (rc) - ipoib_send_comp_handler(priv->send_cq, dev); - } + ++priv->global_tx_head; } } void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_tx *tx = wc->qp->qp_context; unsigned int wr_id = wc->wr_id & ~IPOIB_OP_CM; - struct ipoib_cm_tx_buf *tx_req; + struct ipoib_tx_buf *tx_req; unsigned long flags; ipoib_dbg_data(priv, "cm send completion: id %d, status: %d\n", @@ -788,7 +810,7 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) tx_req = &tx->tx_ring[wr_id]; - ib_dma_unmap_single(priv->ca, tx_req->mapping, tx_req->skb->len, DMA_TO_DEVICE); + ipoib_dma_unmap_tx(priv, tx_req); /* FIXME: is this right? Shouldn't we only increment on success? */ ++dev->stats.tx_packets; @@ -799,25 +821,36 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) netif_tx_lock(dev); ++tx->tx_tail; - if (unlikely(--priv->tx_outstanding == ipoib_sendq_size >> 1) && - netif_queue_stopped(dev) && - test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) + ++priv->global_tx_tail; + + if (unlikely(netif_queue_stopped(dev) && + ((priv->global_tx_head - priv->global_tx_tail) <= + ipoib_sendq_size >> 1) && + test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))) netif_wake_queue(dev); if (wc->status != IB_WC_SUCCESS && wc->status != IB_WC_WR_FLUSH_ERR) { struct ipoib_neigh *neigh; - ipoib_dbg(priv, "failed cm send event " - "(status=%d, wrid=%d vend_err %x)\n", - wc->status, wr_id, wc->vendor_err); + /* IB_WC[_RNR]_RETRY_EXC_ERR error is part of the life cycle, + * so don't make waves. + */ + if (wc->status == IB_WC_RNR_RETRY_EXC_ERR || + wc->status == IB_WC_RETRY_EXC_ERR) + ipoib_dbg(priv, + "%s: failed cm send event (status=%d, wrid=%d vend_err %#x)\n", + __func__, wc->status, wr_id, wc->vendor_err); + else + ipoib_warn(priv, + "%s: failed cm send event (status=%d, wrid=%d vend_err %#x)\n", + __func__, wc->status, wr_id, wc->vendor_err); spin_lock_irqsave(&priv->lock, flags); neigh = tx->neigh; if (neigh) { neigh->cm = NULL; - list_del(&neigh->list); ipoib_neigh_free(neigh); tx->neigh = NULL; @@ -825,7 +858,7 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) { list_move(&tx->list, &priv->cm.reap_list); - queue_work(ipoib_workqueue, &priv->cm.reap_task); + queue_work(priv->wq, &priv->cm.reap_task); } clear_bit(IPOIB_FLAG_OPER_UP, &tx->flags); @@ -838,7 +871,7 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) int ipoib_cm_dev_open(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int ret; if (!IPOIB_CM_SUPPORTED(dev->dev_addr)) @@ -846,16 +879,16 @@ int ipoib_cm_dev_open(struct net_device *dev) priv->cm.id = ib_create_cm_id(priv->ca, ipoib_cm_rx_handler, dev); if (IS_ERR(priv->cm.id)) { - printk(KERN_WARNING "%s: failed to create CM ID\n", priv->ca->name); + pr_warn("%s: failed to create CM ID\n", priv->ca->name); ret = PTR_ERR(priv->cm.id); goto err_cm; } - ret = ib_cm_listen(priv->cm.id, cpu_to_be64(IPOIB_CM_IETF_ID | priv->qp->qp_num), - 0, NULL); + ret = ib_cm_listen(priv->cm.id, + cpu_to_be64(IPOIB_CM_IETF_ID | priv->qp->qp_num)); if (ret) { - printk(KERN_WARNING "%s: failed to listen on ID 0x%llx\n", priv->ca->name, - IPOIB_CM_IETF_ID | priv->qp->qp_num); + pr_warn("%s: failed to listen on ID 0x%llx\n", priv->ca->name, + IPOIB_CM_IETF_ID | priv->qp->qp_num); goto err_listen; } @@ -870,7 +903,7 @@ err_cm: static void ipoib_cm_free_rx_reap_list(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx *rx, *n; LIST_HEAD(list); @@ -893,7 +926,7 @@ static void ipoib_cm_free_rx_reap_list(struct net_device *dev) void ipoib_cm_dev_stop(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx *p; unsigned long begin; int ret; @@ -937,7 +970,7 @@ void ipoib_cm_dev_stop(struct net_device *dev) break; } spin_unlock_irq(&priv->lock); - msleep(1); + usleep_range(1000, 2000); ipoib_drain_cq(dev); spin_lock_irq(&priv->lock); } @@ -949,10 +982,11 @@ void ipoib_cm_dev_stop(struct net_device *dev) cancel_delayed_work(&priv->cm.stale_task); } -static int ipoib_cm_rep_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event) +static int ipoib_cm_rep_handler(struct ib_cm_id *cm_id, + const struct ib_cm_event *event) { struct ipoib_cm_tx *p = cm_id->context; - struct ipoib_dev_priv *priv = netdev_priv(p->dev); + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); struct ipoib_cm_data *data = event->private_data; struct sk_buff_head skqueue; struct ib_qp_attr qp_attr; @@ -995,18 +1029,21 @@ static int ipoib_cm_rep_handler(struct ib_cm_id *cm_id, struct ib_cm_event *even skb_queue_head_init(&skqueue); + netif_tx_lock_bh(p->dev); spin_lock_irq(&priv->lock); set_bit(IPOIB_FLAG_OPER_UP, &p->flags); if (p->neigh) while ((skb = __skb_dequeue(&p->neigh->queue))) __skb_queue_tail(&skqueue, skb); spin_unlock_irq(&priv->lock); + netif_tx_unlock_bh(p->dev); while ((skb = __skb_dequeue(&skqueue))) { skb->dev = p->dev; - if (dev_queue_xmit(skb)) - ipoib_warn(priv, "dev_queue_xmit failed " - "to requeue packet\n"); + ret = dev_queue_xmit(skb); + if (ret) + ipoib_warn(priv, "%s:dev_queue_xmit failed to re-queue packet, ret:%d\n", + __func__, ret); } ret = ib_send_cm_rtu(cm_id, NULL, 0); @@ -1019,27 +1056,35 @@ static int ipoib_cm_rep_handler(struct ib_cm_id *cm_id, struct ib_cm_event *even static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_cm_tx *tx) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_init_attr attr = { - .send_cq = priv->recv_cq, + .send_cq = priv->send_cq, .recv_cq = priv->recv_cq, .srq = priv->cm.srq, .cap.max_send_wr = ipoib_sendq_size, .cap.max_send_sge = 1, .sq_sig_type = IB_SIGNAL_ALL_WR, .qp_type = IB_QPT_RC, - .qp_context = tx + .qp_context = tx, + .create_flags = 0 }; + struct ib_qp *tx_qp; - return ib_create_qp(priv->pd, &attr); + if (dev->features & NETIF_F_SG) + attr.cap.max_send_sge = min_t(u32, priv->ca->attrs.max_send_sge, + MAX_SKB_FRAGS + 1); + + tx_qp = ib_create_qp(priv->pd, &attr); + tx->max_send_sge = attr.cap.max_send_sge; + return tx_qp; } static int ipoib_cm_send_req(struct net_device *dev, struct ib_cm_id *id, struct ib_qp *qp, u32 qpn, - struct ib_sa_path_rec *pathrec) + struct sa_path_rec *pathrec) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_data data = {}; struct ib_cm_req_param req = {}; @@ -1052,7 +1097,7 @@ static int ipoib_cm_send_req(struct net_device *dev, req.qp_num = qp->qp_num; req.qp_type = qp->qp_type; req.private_data = &data; - req.private_data_len = sizeof data; + req.private_data_len = sizeof(data); req.flow_control = 0; req.starting_psn = 0; /* FIXME */ @@ -1074,15 +1119,11 @@ static int ipoib_cm_send_req(struct net_device *dev, static int ipoib_cm_modify_tx_init(struct net_device *dev, struct ib_cm_id *cm_id, struct ib_qp *qp) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_attr qp_attr; int qp_attr_mask, ret; - ret = ib_find_pkey(priv->ca, priv->port, priv->pkey, &qp_attr.pkey_index); - if (ret) { - ipoib_warn(priv, "pkey 0x%x not found: %d\n", priv->pkey, ret); - return ret; - } + qp_attr.pkey_index = priv->pkey_index; qp_attr.qp_state = IB_QPS_INIT; qp_attr.qp_access_flags = IB_ACCESS_LOCAL_WRITE; qp_attr.port_num = priv->port; @@ -1097,22 +1138,25 @@ static int ipoib_cm_modify_tx_init(struct net_device *dev, } static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, - struct ib_sa_path_rec *pathrec) + struct sa_path_rec *pathrec) { - struct ipoib_dev_priv *priv = netdev_priv(p->dev); + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); + unsigned int noio_flag; int ret; - p->tx_ring = vzalloc(ipoib_sendq_size * sizeof *p->tx_ring); + noio_flag = memalloc_noio_save(); + p->tx_ring = vzalloc(array_size(ipoib_sendq_size, sizeof(*p->tx_ring))); if (!p->tx_ring) { - ipoib_warn(priv, "failed to allocate tx ring\n"); + memalloc_noio_restore(noio_flag); ret = -ENOMEM; goto err_tx; } p->qp = ipoib_cm_create_tx_qp(p->dev, p); + memalloc_noio_restore(noio_flag); if (IS_ERR(p->qp)) { ret = PTR_ERR(p->qp); - ipoib_warn(priv, "failed to allocate tx qp: %d\n", ret); + ipoib_warn(priv, "failed to create tx qp: %d\n", ret); goto err_qp; } @@ -1126,13 +1170,13 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, ret = ipoib_cm_modify_tx_init(p->dev, p->id, p->qp); if (ret) { ipoib_warn(priv, "failed to modify tx qp to rtr: %d\n", ret); - goto err_modify; + goto err_modify_send; } ret = ipoib_cm_send_req(p->dev, p->id, p->qp, qpn, pathrec); if (ret) { ipoib_warn(priv, "failed to send cm req: %d\n", ret); - goto err_send_cm; + goto err_modify_send; } ipoib_dbg(priv, "Request connection 0x%x for gid %pI6 qpn 0x%x\n", @@ -1140,8 +1184,7 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, return 0; -err_send_cm: -err_modify: +err_modify_send: ib_destroy_cm_id(p->id); err_id: p->id = NULL; @@ -1155,8 +1198,8 @@ err_tx: static void ipoib_cm_tx_destroy(struct ipoib_cm_tx *p) { - struct ipoib_dev_priv *priv = netdev_priv(p->dev); - struct ipoib_cm_tx_buf *tx_req; + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); + struct ipoib_tx_buf *tx_req; unsigned long begin; ipoib_dbg(priv, "Destroy active connection 0x%x head 0x%x tail 0x%x\n", @@ -1175,7 +1218,7 @@ static void ipoib_cm_tx_destroy(struct ipoib_cm_tx *p) goto timeout; } - msleep(1); + usleep_range(1000, 2000); } } @@ -1183,12 +1226,13 @@ timeout: while ((int) p->tx_tail - (int) p->tx_head < 0) { tx_req = &p->tx_ring[p->tx_tail & (ipoib_sendq_size - 1)]; - ib_dma_unmap_single(priv->ca, tx_req->mapping, tx_req->skb->len, - DMA_TO_DEVICE); + ipoib_dma_unmap_tx(priv, tx_req); dev_kfree_skb_any(tx_req->skb); - ++p->tx_tail; netif_tx_lock_bh(p->dev); - if (unlikely(--priv->tx_outstanding == ipoib_sendq_size >> 1) && + ++p->tx_tail; + ++priv->global_tx_tail; + if (unlikely((priv->global_tx_head - priv->global_tx_tail) <= + ipoib_sendq_size >> 1) && netif_queue_stopped(p->dev) && test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) netif_wake_queue(p->dev); @@ -1203,10 +1247,10 @@ timeout: } static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, - struct ib_cm_event *event) + const struct ib_cm_event *event) { struct ipoib_cm_tx *tx = cm_id->context; - struct ipoib_dev_priv *priv = netdev_priv(tx->dev); + struct ipoib_dev_priv *priv = ipoib_priv(tx->dev); struct net_device *dev = priv->dev; struct ipoib_neigh *neigh; unsigned long flags; @@ -1234,7 +1278,6 @@ static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, if (neigh) { neigh->cm = NULL; - list_del(&neigh->list); ipoib_neigh_free(neigh); tx->neigh = NULL; @@ -1242,7 +1285,7 @@ static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) { list_move(&tx->list, &priv->cm.reap_list); - queue_work(ipoib_workqueue, &priv->cm.reap_task); + queue_work(priv->wq, &priv->cm.reap_task); } spin_unlock_irqrestore(&priv->lock, flags); @@ -1258,31 +1301,30 @@ static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, struct ipoib_cm_tx *ipoib_cm_create_tx(struct net_device *dev, struct ipoib_path *path, struct ipoib_neigh *neigh) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_tx *tx; - tx = kzalloc(sizeof *tx, GFP_ATOMIC); + tx = kzalloc(sizeof(*tx), GFP_ATOMIC); if (!tx) return NULL; neigh->cm = tx; tx->neigh = neigh; - tx->path = path; tx->dev = dev; list_add(&tx->list, &priv->cm.start_list); set_bit(IPOIB_FLAG_INITIALIZED, &tx->flags); - queue_work(ipoib_workqueue, &priv->cm.start_task); + queue_work(priv->wq, &priv->cm.start_task); return tx; } void ipoib_cm_destroy_tx(struct ipoib_cm_tx *tx) { - struct ipoib_dev_priv *priv = netdev_priv(tx->dev); + struct ipoib_dev_priv *priv = ipoib_priv(tx->dev); unsigned long flags; if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) { spin_lock_irqsave(&priv->lock, flags); list_move(&tx->list, &priv->cm.reap_list); - queue_work(ipoib_workqueue, &priv->cm.reap_task); + queue_work(priv->wq, &priv->cm.reap_task); ipoib_dbg(priv, "Reap connection for gid %pI6\n", tx->neigh->daddr + 4); tx->neigh = NULL; @@ -1290,6 +1332,8 @@ void ipoib_cm_destroy_tx(struct ipoib_cm_tx *tx) } } +#define QPN_AND_OPTIONS_OFFSET 4 + static void ipoib_cm_tx_start(struct work_struct *work) { struct ipoib_dev_priv *priv = container_of(work, struct ipoib_dev_priv, @@ -1298,9 +1342,10 @@ static void ipoib_cm_tx_start(struct work_struct *work) struct ipoib_neigh *neigh; struct ipoib_cm_tx *p; unsigned long flags; + struct ipoib_path *path; int ret; - struct ib_sa_path_rec pathrec; + struct sa_path_rec pathrec; u32 qpn; netif_tx_lock_bh(dev); @@ -1310,8 +1355,20 @@ static void ipoib_cm_tx_start(struct work_struct *work) p = list_entry(priv->cm.start_list.next, typeof(*p), list); list_del_init(&p->list); neigh = p->neigh; + qpn = IPOIB_QPN(neigh->daddr); - memcpy(&pathrec, &p->path->pathrec, sizeof pathrec); + /* + * As long as the search is with these 2 locks, + * path existence indicates its validity. + */ + path = __path_find(dev, neigh->daddr + QPN_AND_OPTIONS_OFFSET); + if (!path) { + pr_info("%s ignore not valid path %pI6\n", + __func__, + neigh->daddr + QPN_AND_OPTIONS_OFFSET); + goto free_neigh; + } + memcpy(&pathrec, &path->pathrec, sizeof(pathrec)); spin_unlock_irqrestore(&priv->lock, flags); netif_tx_unlock_bh(dev); @@ -1322,10 +1379,10 @@ static void ipoib_cm_tx_start(struct work_struct *work) spin_lock_irqsave(&priv->lock, flags); if (ret) { +free_neigh: neigh = p->neigh; if (neigh) { neigh->cm = NULL; - list_del(&neigh->list); ipoib_neigh_free(neigh); } list_del(&p->list); @@ -1350,7 +1407,7 @@ static void ipoib_cm_tx_reap(struct work_struct *work) while (!list_empty(&priv->cm.reap_list)) { p = list_entry(priv->cm.reap_list.next, typeof(*p), list); - list_del(&p->list); + list_del_init(&p->list); spin_unlock_irqrestore(&priv->lock, flags); netif_tx_unlock_bh(dev); ipoib_cm_tx_destroy(p); @@ -1369,7 +1426,7 @@ static void ipoib_cm_skb_reap(struct work_struct *work) struct net_device *dev = priv->dev; struct sk_buff *skb; unsigned long flags; - unsigned mtu = priv->mcast_mtu; + unsigned int mtu = priv->mcast_mtu; netif_tx_lock_bh(dev); spin_lock_irqsave(&priv->lock, flags); @@ -1378,11 +1435,15 @@ static void ipoib_cm_skb_reap(struct work_struct *work) spin_unlock_irqrestore(&priv->lock, flags); netif_tx_unlock_bh(dev); - if (skb->protocol == htons(ETH_P_IP)) + if (skb->protocol == htons(ETH_P_IP)) { + memset(IPCB(skb), 0, sizeof(*IPCB(skb))); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); + } #if IS_ENABLED(CONFIG_IPV6) - else if (skb->protocol == htons(ETH_P_IPV6)) + else if (skb->protocol == htons(ETH_P_IPV6)) { + memset(IP6CB(skb), 0, sizeof(*IP6CB(skb))); icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); + } #endif dev_kfree_skb_any(skb); @@ -1397,15 +1458,14 @@ static void ipoib_cm_skb_reap(struct work_struct *work) void ipoib_cm_skb_too_long(struct net_device *dev, struct sk_buff *skb, unsigned int mtu) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int e = skb_queue_empty(&priv->cm.skb_queue); - if (skb_dst(skb)) - skb_dst(skb)->ops->update_pmtu(skb_dst(skb), NULL, skb, mtu); + skb_dst_update_pmtu(skb, mtu); skb_queue_tail(&priv->cm.skb_queue, skb); if (e) - queue_work(ipoib_workqueue, &priv->cm.skb_task); + queue_work(priv->wq, &priv->cm.skb_task); } static void ipoib_cm_rx_reap(struct work_struct *work) @@ -1438,43 +1498,51 @@ static void ipoib_cm_stale_task(struct work_struct *work) } if (!list_empty(&priv->cm.passive_ids)) - queue_delayed_work(ipoib_workqueue, + queue_delayed_work(priv->wq, &priv->cm.stale_task, IPOIB_CM_RX_DELAY); spin_unlock_irq(&priv->lock); } - -static ssize_t show_mode(struct device *d, struct device_attribute *attr, +static ssize_t mode_show(struct device *d, struct device_attribute *attr, char *buf) { - struct ipoib_dev_priv *priv = netdev_priv(to_net_dev(d)); + struct net_device *dev = to_net_dev(d); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags)) - return sprintf(buf, "connected\n"); + return sysfs_emit(buf, "connected\n"); else - return sprintf(buf, "datagram\n"); + return sysfs_emit(buf, "datagram\n"); } -static ssize_t set_mode(struct device *d, struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t mode_store(struct device *d, struct device_attribute *attr, + const char *buf, size_t count) { struct net_device *dev = to_net_dev(d); int ret; - if (!rtnl_trylock()) + if (!rtnl_trylock()) { return restart_syscall(); + } - ret = ipoib_set_mode(dev, buf); + if (dev->reg_state != NETREG_REGISTERED) { + rtnl_unlock(); + return -EPERM; + } - rtnl_unlock(); + ret = ipoib_set_mode(dev, buf); - if (!ret) - return count; + /* The assumption is that the function ipoib_set_mode returned + * with the rtnl held by it, if not the value -EBUSY returned, + * then no need to rtnl_unlock + */ + if (ret != -EBUSY) + rtnl_unlock(); - return ret; + return (!ret || ret == -EBUSY) ? count : ret; } -static DEVICE_ATTR(mode, S_IWUSR | S_IRUGO, show_mode, set_mode); +static DEVICE_ATTR_RW(mode); int ipoib_cm_add_mode_attr(struct net_device *dev) { @@ -1483,7 +1551,7 @@ int ipoib_cm_add_mode_attr(struct net_device *dev) static void ipoib_cm_create_srq(struct net_device *dev, int max_sge) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_srq_init_attr srq_init_attr = { .srq_type = IB_SRQT_BASIC, .attr = { @@ -1494,17 +1562,16 @@ static void ipoib_cm_create_srq(struct net_device *dev, int max_sge) priv->cm.srq = ib_create_srq(priv->pd, &srq_init_attr); if (IS_ERR(priv->cm.srq)) { - if (PTR_ERR(priv->cm.srq) != -ENOSYS) - printk(KERN_WARNING "%s: failed to allocate SRQ, error %ld\n", + if (PTR_ERR(priv->cm.srq) != -EOPNOTSUPP) + pr_warn("%s: failed to allocate SRQ, error %ld\n", priv->ca->name, PTR_ERR(priv->cm.srq)); priv->cm.srq = NULL; return; } - priv->cm.srq_ring = vzalloc(ipoib_recvq_size * sizeof *priv->cm.srq_ring); + priv->cm.srq_ring = vzalloc(array_size(ipoib_recvq_size, + sizeof(*priv->cm.srq_ring))); if (!priv->cm.srq_ring) { - printk(KERN_WARNING "%s: failed to allocate CM SRQ ring (%d entries)\n", - priv->ca->name, ipoib_recvq_size); ib_destroy_srq(priv->cm.srq); priv->cm.srq = NULL; return; @@ -1514,9 +1581,9 @@ static void ipoib_cm_create_srq(struct net_device *dev, int max_sge) int ipoib_cm_dev_init(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); - int i, ret; - struct ib_device_attr attr; + struct ipoib_dev_priv *priv = ipoib_priv(dev); + int max_srq_sge, i; + u8 addr; INIT_LIST_HEAD(&priv->cm.passive_ids); INIT_LIST_HEAD(&priv->cm.reap_list); @@ -1533,19 +1600,13 @@ int ipoib_cm_dev_init(struct net_device *dev) skb_queue_head_init(&priv->cm.skb_queue); - ret = ib_query_device(priv->ca, &attr); - if (ret) { - printk(KERN_WARNING "ib_query_device() failed with %d\n", ret); - return ret; - } - - ipoib_dbg(priv, "max_srq_sge=%d\n", attr.max_srq_sge); + ipoib_dbg(priv, "max_srq_sge=%d\n", priv->ca->attrs.max_srq_sge); - attr.max_srq_sge = min_t(int, IPOIB_CM_RX_SG, attr.max_srq_sge); - ipoib_cm_create_srq(dev, attr.max_srq_sge); + max_srq_sge = min_t(int, IPOIB_CM_RX_SG, priv->ca->attrs.max_srq_sge); + ipoib_cm_create_srq(dev, max_srq_sge); if (ipoib_cm_has_srq(dev)) { - priv->cm.max_cm_mtu = attr.max_srq_sge * PAGE_SIZE - 0x10; - priv->cm.num_frags = attr.max_srq_sge; + priv->cm.max_cm_mtu = max_srq_sge * PAGE_SIZE - 0x10; + priv->cm.num_frags = max_srq_sge; ipoib_dbg(priv, "max_cm_mtu = 0x%x, num_frags=%d\n", priv->cm.max_cm_mtu, priv->cm.num_frags); } else { @@ -1559,7 +1620,8 @@ int ipoib_cm_dev_init(struct net_device *dev) for (i = 0; i < ipoib_recvq_size; ++i) { if (!ipoib_cm_alloc_rx_skb(dev, priv->cm.srq_ring, i, priv->cm.num_frags - 1, - priv->cm.srq_ring[i].mapping)) { + priv->cm.srq_ring[i].mapping, + GFP_KERNEL)) { ipoib_warn(priv, "failed to allocate " "receive buffer %d\n", i); ipoib_cm_dev_cleanup(dev); @@ -1575,24 +1637,21 @@ int ipoib_cm_dev_init(struct net_device *dev) } } - priv->dev->dev_addr[0] = IPOIB_FLAGS_RC; + addr = IPOIB_FLAGS_RC; + dev_addr_mod(dev, 0, &addr, 1); return 0; } void ipoib_cm_dev_cleanup(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); - int ret; + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (!priv->cm.srq) return; ipoib_dbg(priv, "Cleanup ipoib connected mode.\n"); - ret = ib_destroy_srq(priv->cm.srq); - if (ret) - ipoib_warn(priv, "ib_destroy_srq failed: %d\n", ret); - + ib_destroy_srq(priv->cm.srq); priv->cm.srq = NULL; if (!priv->cm.srq_ring) return; |