return;
}
+ /*
+ * Drop packets that this interface sent, ie multicast packets
+ * that the HCA has replicated.
+ */
+ if (wc->slid == priv->local_lid && wc->src_qp == priv->qp->qp_num)
+ goto repost;
+
/*
* If we can't allocate a new RX buffer, dump
* this packet and reuse the old buffer.
*/
if (unlikely(ipoib_alloc_rx_skb(dev, wr_id))) {
- ++priv->stats.rx_dropped;
+ ++dev->stats.rx_dropped;
goto repost;
}
skb_put(skb, wc->byte_len);
skb_pull(skb, IB_GRH_BYTES);
- if (wc->slid != priv->local_lid ||
- wc->src_qp != priv->qp->qp_num) {
- skb->protocol = ((struct ipoib_header *) skb->data)->proto;
- skb_reset_mac_header(skb);
- skb_pull(skb, IPOIB_ENCAP_LEN);
+ skb->protocol = ((struct ipoib_header *) skb->data)->proto;
+ skb_reset_mac_header(skb);
+ skb_pull(skb, IPOIB_ENCAP_LEN);
- dev->last_rx = jiffies;
- ++priv->stats.rx_packets;
- priv->stats.rx_bytes += skb->len;
+ dev->last_rx = jiffies;
+ ++dev->stats.rx_packets;
+ dev->stats.rx_bytes += skb->len;
- skb->dev = dev;
- /* XXX get correct PACKET_ type here */
- skb->pkt_type = PACKET_HOST;
- netif_rx_ni(skb);
- } else {
- ipoib_dbg_data(priv, "dropping loopback packet\n");
- dev_kfree_skb_any(skb);
- }
+ skb->dev = dev;
+ /* XXX get correct PACKET_ type here */
+ skb->pkt_type = PACKET_HOST;
+ netif_receive_skb(skb);
repost:
if (unlikely(ipoib_ib_post_receive(dev, wr_id)))
"for buf %d\n", wr_id);
}
+static int ipoib_dma_map_tx(struct ib_device *ca,
+ struct ipoib_tx_buf *tx_req)
+{
+ struct sk_buff *skb = tx_req->skb;
+ u64 *mapping = tx_req->mapping;
+ int i;
+
+ mapping[0] = ib_dma_map_single(ca, skb->data, skb_headlen(skb),
+ DMA_TO_DEVICE);
+ if (unlikely(ib_dma_mapping_error(ca, mapping[0])))
+ return -EIO;
+
+ for (i = 0; i < skb_shinfo(skb)->nr_frags; ++i) {
+ skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+ mapping[i + 1] = ib_dma_map_page(ca, frag->page,
+ frag->page_offset, frag->size,
+ DMA_TO_DEVICE);
+ if (unlikely(ib_dma_mapping_error(ca, mapping[i + 1])))
+ goto partial_error;
+ }
+ return 0;
+
+partial_error:
+ ib_dma_unmap_single(ca, mapping[0], skb_headlen(skb), DMA_TO_DEVICE);
+
+ for (; i > 0; --i) {
+ skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];
+ ib_dma_unmap_page(ca, mapping[i], frag->size, DMA_TO_DEVICE);
+ }
+ return -EIO;
+}
+
+static void ipoib_dma_unmap_tx(struct ib_device *ca,
+ struct ipoib_tx_buf *tx_req)
+{
+ struct sk_buff *skb = tx_req->skb;
+ u64 *mapping = tx_req->mapping;
+ int i;
+
+ ib_dma_unmap_single(ca, mapping[0], skb_headlen(skb), DMA_TO_DEVICE);
+
+ for (i = 0; i < skb_shinfo(skb)->nr_frags; ++i) {
+ skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+ ib_dma_unmap_page(ca, mapping[i + 1], frag->size,
+ DMA_TO_DEVICE);
+ }
+}
+
static void ipoib_ib_handle_tx_wc(struct net_device *dev, struct ib_wc *wc)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
tx_req = &priv->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->ca, tx_req);
- ++priv->stats.tx_packets;
- priv->stats.tx_bytes += tx_req->skb->len;
+ ++dev->stats.tx_packets;
+ dev->stats.tx_bytes += tx_req->skb->len;
dev_kfree_skb_any(tx_req->skb);
spin_lock_irqsave(&priv->tx_lock, flags);
++priv->tx_tail;
- if (unlikely(test_bit(IPOIB_FLAG_NETIF_STOPPED, &priv->flags)) &&
- priv->tx_head - priv->tx_tail <= ipoib_sendq_size >> 1) {
- clear_bit(IPOIB_FLAG_NETIF_STOPPED, &priv->flags);
+ if (unlikely(--priv->tx_outstanding == ipoib_sendq_size >> 1) &&
+ netif_queue_stopped(dev) &&
+ test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
netif_wake_queue(dev);
- }
spin_unlock_irqrestore(&priv->tx_lock, flags);
if (wc->status != IB_WC_SUCCESS &&
wc->status, wr_id, wc->vendor_err);
}
-static void ipoib_ib_handle_wc(struct net_device *dev, struct ib_wc *wc)
+int ipoib_poll(struct napi_struct *napi, int budget)
{
- if (wc->wr_id & IPOIB_CM_OP_SRQ)
- ipoib_cm_handle_rx_wc(dev, wc);
- else if (wc->wr_id & IPOIB_OP_RECV)
- ipoib_ib_handle_rx_wc(dev, wc);
- else
- ipoib_ib_handle_tx_wc(dev, wc);
+ struct ipoib_dev_priv *priv = container_of(napi, struct ipoib_dev_priv, napi);
+ struct net_device *dev = priv->dev;
+ int done;
+ int t;
+ int n, i;
+
+ done = 0;
+
+poll_more:
+ while (done < budget) {
+ int max = (budget - done);
+
+ t = min(IPOIB_NUM_WC, max);
+ n = ib_poll_cq(priv->cq, t, priv->ibwc);
+
+ for (i = 0; i < n; i++) {
+ struct ib_wc *wc = priv->ibwc + i;
+
+ if (wc->wr_id & IPOIB_OP_RECV) {
+ ++done;
+ if (wc->wr_id & IPOIB_OP_CM)
+ ipoib_cm_handle_rx_wc(dev, wc);
+ else
+ ipoib_ib_handle_rx_wc(dev, wc);
+ } else {
+ if (wc->wr_id & IPOIB_OP_CM)
+ ipoib_cm_handle_tx_wc(dev, wc);
+ else
+ ipoib_ib_handle_tx_wc(dev, wc);
+ }
+ }
+
+ if (n != t)
+ break;
+ }
+
+ if (done < budget) {
+ netif_rx_complete(dev, napi);
+ if (unlikely(ib_req_notify_cq(priv->cq,
+ IB_CQ_NEXT_COMP |
+ IB_CQ_REPORT_MISSED_EVENTS)) &&
+ netif_rx_reschedule(dev, napi))
+ goto poll_more;
+ }
+
+ return done;
}
void ipoib_ib_completion(struct ib_cq *cq, void *dev_ptr)
{
- struct net_device *dev = (struct net_device *) dev_ptr;
+ struct net_device *dev = dev_ptr;
struct ipoib_dev_priv *priv = netdev_priv(dev);
- int n, i;
- ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
- do {
- n = ib_poll_cq(cq, IPOIB_NUM_WC, priv->ibwc);
- for (i = 0; i < n; ++i)
- ipoib_ib_handle_wc(dev, priv->ibwc + i);
- } while (n == IPOIB_NUM_WC);
+ netif_rx_schedule(dev, &priv->napi);
}
static inline int post_send(struct ipoib_dev_priv *priv,
unsigned int wr_id,
struct ib_ah *address, u32 qpn,
- u64 addr, int len)
+ u64 *mapping, int headlen,
+ skb_frag_t *frags,
+ int nr_frags)
{
struct ib_send_wr *bad_wr;
+ int i;
- priv->tx_sge.addr = addr;
- priv->tx_sge.length = len;
-
- priv->tx_wr.wr_id = wr_id;
- priv->tx_wr.wr.ud.remote_qpn = qpn;
- priv->tx_wr.wr.ud.ah = address;
+ priv->tx_sge[0].addr = mapping[0];
+ priv->tx_sge[0].length = headlen;
+ for (i = 0; i < nr_frags; ++i) {
+ priv->tx_sge[i + 1].addr = mapping[i + 1];
+ priv->tx_sge[i + 1].length = frags[i].size;
+ }
+ priv->tx_wr.num_sge = nr_frags + 1;
+ priv->tx_wr.wr_id = wr_id;
+ priv->tx_wr.wr.ud.remote_qpn = qpn;
+ priv->tx_wr.wr.ud.ah = address;
return ib_post_send(priv->qp, &priv->tx_wr, &bad_wr);
}
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct ipoib_tx_buf *tx_req;
- u64 addr;
if (unlikely(skb->len > priv->mcast_mtu + IPOIB_ENCAP_LEN)) {
ipoib_warn(priv, "packet len %d (> %d) too long to send, dropping\n",
skb->len, priv->mcast_mtu + IPOIB_ENCAP_LEN);
- ++priv->stats.tx_dropped;
- ++priv->stats.tx_errors;
+ ++dev->stats.tx_dropped;
+ ++dev->stats.tx_errors;
ipoib_cm_skb_too_long(dev, skb, priv->mcast_mtu);
return;
}
*/
tx_req = &priv->tx_ring[priv->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))) {
- ++priv->stats.tx_errors;
+ 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 (unlikely(post_send(priv, priv->tx_head & (ipoib_sendq_size - 1),
- address->ah, qpn, addr, skb->len))) {
+ address->ah, qpn,
+ tx_req->mapping, skb_headlen(skb),
+ skb_shinfo(skb)->frags, skb_shinfo(skb)->nr_frags))) {
ipoib_warn(priv, "post_send failed\n");
- ++priv->stats.tx_errors;
- ib_dma_unmap_single(priv->ca, addr, skb->len, DMA_TO_DEVICE);
+ ++dev->stats.tx_errors;
+ ipoib_dma_unmap_tx(priv->ca, tx_req);
dev_kfree_skb_any(skb);
} else {
dev->trans_start = jiffies;
address->last_send = priv->tx_head;
++priv->tx_head;
- if (priv->tx_head - priv->tx_tail == ipoib_sendq_size) {
+ if (++priv->tx_outstanding == ipoib_sendq_size) {
ipoib_dbg(priv, "TX ring full, stopping kernel net queue\n");
netif_stop_queue(dev);
- set_bit(IPOIB_FLAG_NETIF_STOPPED, &priv->flags);
}
}
}
__ipoib_reap_ah(dev);
if (!test_bit(IPOIB_STOP_REAPER, &priv->flags))
- queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task, HZ);
+ queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task,
+ round_jiffies_relative(HZ));
}
int ipoib_ib_dev_open(struct net_device *dev)
struct ipoib_dev_priv *priv = netdev_priv(dev);
int ret;
+ if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &priv->pkey_index)) {
+ ipoib_warn(priv, "P_Key 0x%04x not found\n", priv->pkey);
+ clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
+ return -1;
+ }
+ set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
+
ret = ipoib_init_qp(dev);
if (ret) {
ipoib_warn(priv, "ipoib_init_qp returned %d\n", ret);
ret = ipoib_ib_post_receives(dev);
if (ret) {
ipoib_warn(priv, "ipoib_ib_post_receives returned %d\n", ret);
- ipoib_ib_dev_stop(dev);
+ ipoib_ib_dev_stop(dev, 1);
return -1;
}
ret = ipoib_cm_dev_open(dev);
if (ret) {
- ipoib_warn(priv, "ipoib_ib_post_receives returned %d\n", ret);
- ipoib_ib_dev_stop(dev);
+ ipoib_warn(priv, "ipoib_cm_dev_open returned %d\n", ret);
+ ipoib_ib_dev_stop(dev, 1);
return -1;
}
clear_bit(IPOIB_STOP_REAPER, &priv->flags);
- queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task, HZ);
+ queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task,
+ round_jiffies_relative(HZ));
set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
mutex_lock(&pkey_mutex);
set_bit(IPOIB_PKEY_STOP, &priv->flags);
- cancel_delayed_work(&priv->pkey_task);
+ cancel_delayed_work(&priv->pkey_poll_task);
mutex_unlock(&pkey_mutex);
if (flush)
flush_workqueue(ipoib_workqueue);
return pending;
}
-int ipoib_ib_dev_stop(struct net_device *dev)
+void ipoib_drain_cq(struct net_device *dev)
+{
+ struct ipoib_dev_priv *priv = netdev_priv(dev);
+ int i, n;
+ do {
+ n = ib_poll_cq(priv->cq, IPOIB_NUM_WC, priv->ibwc);
+ for (i = 0; i < n; ++i) {
+ /*
+ * Convert any successful completions to flush
+ * errors to avoid passing packets up the
+ * stack after bringing the device down.
+ */
+ if (priv->ibwc[i].status == IB_WC_SUCCESS)
+ priv->ibwc[i].status = IB_WC_WR_FLUSH_ERR;
+
+ if (priv->ibwc[i].wr_id & IPOIB_OP_RECV) {
+ if (priv->ibwc[i].wr_id & IPOIB_OP_CM)
+ ipoib_cm_handle_rx_wc(dev, priv->ibwc + i);
+ else
+ ipoib_ib_handle_rx_wc(dev, priv->ibwc + i);
+ } else {
+ if (priv->ibwc[i].wr_id & IPOIB_OP_CM)
+ ipoib_cm_handle_tx_wc(dev, priv->ibwc + i);
+ else
+ ipoib_ib_handle_tx_wc(dev, priv->ibwc + i);
+ }
+ }
+ } while (n == IPOIB_NUM_WC);
+}
+
+int ipoib_ib_dev_stop(struct net_device *dev, int flush)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct ib_qp_attr qp_attr;
while ((int) priv->tx_tail - (int) priv->tx_head < 0) {
tx_req = &priv->tx_ring[priv->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->ca, tx_req);
dev_kfree_skb_any(tx_req->skb);
++priv->tx_tail;
+ --priv->tx_outstanding;
}
for (i = 0; i < ipoib_recvq_size; ++i) {
goto timeout;
}
+ ipoib_drain_cq(dev);
+
msleep(1);
}
/* Wait for all AHs to be reaped */
set_bit(IPOIB_STOP_REAPER, &priv->flags);
cancel_delayed_work(&priv->ah_reap_task);
- flush_workqueue(ipoib_workqueue);
+ if (flush)
+ flush_workqueue(ipoib_workqueue);
begin = jiffies;
msleep(1);
}
+ ib_req_notify_cq(priv->cq, IB_CQ_NEXT_COMP);
+
return 0;
}
return 0;
}
-void ipoib_ib_dev_flush(struct work_struct *work)
+static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event)
{
- struct ipoib_dev_priv *cpriv, *priv =
- container_of(work, struct ipoib_dev_priv, flush_task);
+ struct ipoib_dev_priv *cpriv;
struct net_device *dev = priv->dev;
+ u16 new_index;
+
+ mutex_lock(&priv->vlan_mutex);
- if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags) ) {
+ /*
+ * Flush any child interfaces too -- they might be up even if
+ * the parent is down.
+ */
+ list_for_each_entry(cpriv, &priv->child_intfs, list)
+ __ipoib_ib_dev_flush(cpriv, pkey_event);
+
+ mutex_unlock(&priv->vlan_mutex);
+
+ if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) {
ipoib_dbg(priv, "Not flushing - IPOIB_FLAG_INITIALIZED not set.\n");
return;
}
return;
}
+ if (pkey_event) {
+ if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) {
+ clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
+ ipoib_ib_dev_down(dev, 0);
+ ipoib_ib_dev_stop(dev, 0);
+ ipoib_pkey_dev_delay_open(dev);
+ return;
+ }
+ set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
+
+ /* restart QP only if P_Key index is changed */
+ if (new_index == priv->pkey_index) {
+ ipoib_dbg(priv, "Not flushing - P_Key index not changed.\n");
+ return;
+ }
+ priv->pkey_index = new_index;
+ }
+
ipoib_dbg(priv, "flushing\n");
ipoib_ib_dev_down(dev, 0);
+ if (pkey_event) {
+ ipoib_ib_dev_stop(dev, 0);
+ ipoib_ib_dev_open(dev);
+ }
+
/*
* The device could have been brought down between the start and when
* we get here, don't bring it back up if it's not configured up
ipoib_ib_dev_up(dev);
ipoib_mcast_restart_task(&priv->restart_task);
}
+}
- mutex_lock(&priv->vlan_mutex);
+void ipoib_ib_dev_flush(struct work_struct *work)
+{
+ struct ipoib_dev_priv *priv =
+ container_of(work, struct ipoib_dev_priv, flush_task);
- /* Flush any child interfaces too */
- list_for_each_entry(cpriv, &priv->child_intfs, list)
- ipoib_ib_dev_flush(&cpriv->flush_task);
+ ipoib_dbg(priv, "Flushing %s\n", priv->dev->name);
+ __ipoib_ib_dev_flush(priv, 0);
+}
- mutex_unlock(&priv->vlan_mutex);
+void ipoib_pkey_event(struct work_struct *work)
+{
+ struct ipoib_dev_priv *priv =
+ container_of(work, struct ipoib_dev_priv, pkey_event_task);
+
+ ipoib_dbg(priv, "Flushing %s and restarting its QP\n", priv->dev->name);
+ __ipoib_ib_dev_flush(priv, 1);
}
void ipoib_ib_dev_cleanup(struct net_device *dev)
void ipoib_pkey_poll(struct work_struct *work)
{
struct ipoib_dev_priv *priv =
- container_of(work, struct ipoib_dev_priv, pkey_task.work);
+ container_of(work, struct ipoib_dev_priv, pkey_poll_task.work);
struct net_device *dev = priv->dev;
ipoib_pkey_dev_check_presence(dev);
mutex_lock(&pkey_mutex);
if (!test_bit(IPOIB_PKEY_STOP, &priv->flags))
queue_delayed_work(ipoib_workqueue,
- &priv->pkey_task,
+ &priv->pkey_poll_task,
HZ);
mutex_unlock(&pkey_mutex);
}
mutex_lock(&pkey_mutex);
clear_bit(IPOIB_PKEY_STOP, &priv->flags);
queue_delayed_work(ipoib_workqueue,
- &priv->pkey_task,
+ &priv->pkey_poll_task,
HZ);
mutex_unlock(&pkey_mutex);
return 1;