#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
};
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);
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)
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);
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;
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, &bad_wr);
}
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_tx_buf *tx_req;
int rc;
if (unlikely(skb->len > tx->mtu)) {
*/
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;
-
skb_orphan(skb);
skb_dst_drop(skb);
- rc = post_send(priv, tx, tx->tx_head & (ipoib_sendq_size - 1),
- addr, skb->len);
+ 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);
++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);
} else {
dev->trans_start = jiffies;
struct ipoib_dev_priv *priv = netdev_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",
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;
}
ret = ib_cm_listen(priv->cm.id, cpu_to_be64(IPOIB_CM_IETF_ID | priv->qp->qp_num),
- 0, NULL);
+ 0);
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);
struct ib_qp *tx_qp;
+ if (dev->features & NETIF_F_SG)
+ attr.cap.max_send_sge = MAX_SKB_FRAGS + 1;
+
tx_qp = ib_create_qp(priv->pd, &attr);
if (PTR_ERR(tx_qp) == -EINVAL) {
- ipoib_warn(priv, "can't use GFP_NOIO for QPs on device %s, using GFP_KERNEL\n",
- priv->ca->name);
attr.create_flags &= ~IB_QP_CREATE_USE_GFP_NOIO;
tx_qp = ib_create_qp(priv->pd, &attr);
}
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_tx_buf *tx_req;
unsigned long begin;
ipoib_dbg(priv, "Destroy active connection 0x%x head 0x%x tail 0x%x\n",
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);
}
}
+#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,
struct ipoib_neigh *neigh;
struct ipoib_cm_tx *p;
unsigned long flags;
+ struct ipoib_path *path;
int ret;
struct ib_sa_path_rec pathrec;
p = list_entry(priv->cm.start_list.next, typeof(*p), list);
list_del_init(&p->list);
neigh = p->neigh;
+
qpn = IPOIB_QPN(neigh->daddr);
+ /*
+ * 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, &p->path->pathrec, sizeof pathrec);
spin_unlock_irqrestore(&priv->lock, flags);
spin_lock_irqsave(&priv->lock, flags);
if (ret) {
+free_neigh:
neigh = p->neigh;
if (neigh) {
neigh->cm = NULL;
spin_unlock_irq(&priv->lock);
}
-
static ssize_t show_mode(struct device *d, struct device_attribute *attr,
char *buf)
{